ORDES
            Volume II-B
Preliminary Technology Assessment Report

         University of Kentucky
         University of Louisville
            May 15, 1977
             PHASE
   OHIO RIVER DASIK ENERGY STUDY

-------
            OHIO RIVER BASIN ENERGY STUDY


                     VOLUME II-B


      PRELIMINARY TECHNOLOGY ASSESSMENT REPORT


UNIVERSITY OF KENTUCKY AND UNIVERSITY OF LOUISVILLE



                    May 15, 1977
                                   Prepared for
                                   Office of Energy,  Minerals,  and Industry
                                   Office of Research and Development
                                   U.  S.  Environmental  Protection Agency
                                   Washington,  D.  C.
                                   Grant  Number R804816-01-0 (U of L)
                                   Grant  Number R804817-01-0 (UK)

-------
                               CONTENTS


                                                               Page

TABLES	II-B-vii

FIGURES	n-B-xi

APPENDICES 	

1.  INTRODUCTION .  .  .  .  ,	II-B-1-1

    1,1.  DEFINITION  AND BRIEF HISTORY OF TECHNOLOGY
          ASSESSMENT  	  	  II-B-1-12

    1,2,  THE REGIONAL  TECHNOLOGY CONFIGURATIONS 	  II-B-1-15

          1.2,1.  BOM HIGH 80/20 MIX	II-B-1-16
          1,2.2,  BOM HIGH 50/50 MIX	II-B-1-16
          1.2.3.  FORD  TECH FIX 100% COAL	II-B-1-25
          1.2.4.  FORD  TECH FIX 100% NUCLEAR 	  ....  II-B-1-25

    1,3.  LOCATION OF EXISTING AND PLANNED FACILITIES  ....  II-B-1-25

    1<4V..JCQMROSLUON OF,*UN1TED,-STATES COALS	II-B-1-25

          1,4.1.  FUEL  VALUE	,	II-B-1-27
          1.4,2.  NON-METALS 	  II-B-1-27
          1.4.3.  METALS 	  II-B-1-30
          1,4,4,  RADIONUCLIDES IN UNITED STATES COALS  .  .  .  ._II-B-1-30

    1,5.  OHIO RIVER  SOCIOLOGICAL STATUS	  II-B-1-37

    1.6,  PHYSICAL AND  BIOLOGICAL DESCRIPTION OF THE OHIO
          RIVER BASIN	,  .  .  II-B-1-49

          1.6,1.  GEOGRAPHIC	.II-B-1-49
          1.6.2.  TOPOGRAPHIC  	  II-B-1-49
          1,6.3.  GEOLOGIC 	  	  II-B-1-50
          1.6,4.  CLIMATOLOGIC 	  II-B-1-51
          1.6,5.  FLOW  CHARACTERISTICS OF THE OHIO RIVER  .  .  .  II-B-1-52
          1.6.6.  MINERAL RESOURCES IN ORBES 	  II-B-1-58
          1.6,7.  BIOLOGICAL RESOURCES IN ORBES  .	  II-B-1-58

2,  PRELIMINARY REVIEW  OF THE RTC'S  	  II-B-2-1

    2.1.  THE SEQUENTIAL MODIFIED DELPHI  PROCESS 	  II-B-2-2

-------
                                                           page

      2.1.1.  FINAL DELPHI PROCESS PROJECTIONS 	 II-B-2-4

              2.1,1.1.  DEMOGRAPHIC ASSUMPTIONS AND
                        PROJECTIONS	 II-B-2-4
              2,1,1.2,  SOCIOECONOMIC ASSUMPTIONS  .... II-B-2-5
              2.1.1,3,  POLITICAL ASSUMPTIONS  	 II-B-2-6
              2,1.1.4,  TECHNOLOGICAL ASSUMPTIONS  .... II-B-2-7

2.2,  POPULATION PROJECTIONS 	 II-B-2-10

2.3.  ANALYSIS OF SCIENTIFIC METHODOLOGIES USED IN
      TECHNOLOGY ASSESSMENT - PRELIMINARY REVIEW OF
      RTC'S	 . II-B-2-30

      2,3.1.  ASSESSMENT APPROACHES AND METHODOLOGIES  . . II-B-2-32
      2,3,2.  METHODOLOGICAL ANALYSIS OF THE ASSESSMENT
              APPROACH IN GENERAL  	 II-B-2-32
      2,3,3,  EVALUATION OF THE DELPHI METHOD IN GENERAL . II-B-2-33
      2,3.4.  EVALUATION OF OTHER ASSUMPTIONS OF THE
              APPROACH IN GENERAL:  PROBLEMS WITH ALL
              FOUR SCENARIOS 	 II-B-2-35

              2,3,4.1.  SCIENTIFIC ASSUMPTIONS 	 II-B-2-35
              2.3,4.-2.  :ETttIQAL'-AS-SUMPTIONS 'BUILT INTO
                        THE ENTIRE ORBES STUDY 	 II-B-2-45

      2.3,5.  METHODOLOGICAL ANALYSIS OF THE ASSESSMENT
              APPROACH IN PARTICULAR 	 II-B-2-47
      2,3,6,  METHODOLOGICAL PROBLEMS WITH THE BOM
              SCENARIOS (ONE AND TWO)	II-B-2-47

              2.3,6,1.  LOGICAL PROBLEMS WITH THE BOM
                        SCENARIOS  	 II-B-2-47
            .  2.3.6.2.  SCIENTIFIC PROBLEMS WITH THE BOM
                        SCENARIOS  	 II-B-2-48
              2,3,6,3,  ETHICAL PROBLEMS WITH THE BOM
                        SCENARIOS  	 II-B-2-49

      2,3.7,  METHODOLOGICAL PROBLEMS WITH THE FTF
              SCENARIOS (THREE AND FOUR) 	 II-B-2-50

              2.3,7.1.  LOGICAL PROBLEMS WITH THE FTF
                        SCENARIOS  	 II-B-2-50
              2.3,7.2.  SCIENTIFIC PROBLEMS WITH THE FTF
                        SCENARIOS  	 II-B-2-50
              2.3.7.3.  ETHICAL PROBLEMS WITH THE FTF
                        SCENARIOS  	 II-B-2-50

-------
                                                               page

          2.3.8.  METHODOLOGICAL PROBLEMS WITH THE NUCLEAR
                  SCENARIOS (ONE, TWO AND FOUR)  ....... II-B-2-51

                  2.3,8,1.  LOGICAL PROBLEMS WITH THE NUCLEAR
                            SCENARIOS  	 II-B-2-51
                  2.3.8.2.  SCIENTIFIC PROBLEMS WITH THE
                            NUCLEAR SCENARIOS  	 II-B-2-51
                  2,3.8.3,  ETHICAL PROBLEMS WITH THE NUCLEAR
                            SCENARIOS  	 II-B-2-53

          2,3,9,  METHODOLOGICAL PROBLEMS WITH THE COAL
                  SCENARIOS  	 	 II-B-2-57

                  2.3.9.1,  LOGICAL PROBLEMS WITH THE COAL
                            SCENARIOS  	 II-B-2-57
                  2,3,9.2,  SCIENTIFIC PROBLEMS WITH THE COAL
                            SCENARIOS  	 II-B-2-57
                  2,3.9.3,  ETHICAL PROBLEMS WITH THE COAL
                            SCENARIOS	 II-B-2-58

          2,3.10, SUMMARY AND CONCLUSIONS  .,....'.... II-B-2-58

3.  PHYSICAL AND CHEMICAL IMPACTS  	 II-B-3-1

    '3.1.  ATR IMPACTS  . . , ,	II-B-3-2

          3.1.1.  ASSESSMENT PERSPECTIVE 	 II-B-3-2

                  3.1.1.1.  INTRODUCTION 	 II-B-3-2
                  3,1.1,2.  BASELINE DATA  	 II-B-3-3
                  3.1.1,3.  REGULATIONS  	 II-B-3-8
                  3,1,1,4.  LIMITATIONS OF ANALYSIS  	 II-B-3-16

          3.1,2.  OVERVIEW OF IMPACTS BY ENERGY CYCLE  .... 11-^-3-17

                  3.1.2.1.  NUCLEAR-ELECTRIC CYCLES  ..... II-B-3-17
                  3.1.2.2.  FOSSIL FUEL ELECTRIC CYCLES  . .  . II-B-3-18
                  3.1.2.3.  SYNTHETIC FUELS FROM COAL  .... II-B-3-18

          3,1.3.  ALTERATION OF BASIC ATMOSPHERIC PROPERTIES . II-B-3-20

                  3.1.3,1.  INSOLATION  	 II-B-3-20
                  3.1,3.2.  VISIBILITY  	 II-B-3-20
                  3,1.3.3,  AIRFLOW PATTERNS AND TURBULENCE
                            PATTERNS 	 II-B-3-21
                  3,1.3,4,  NOISE  	 II-B-3-21
                  3,1.3,5.  HEAT AND MOISTURE  	 II-B-3-21

          3.1,4.  COMPARATIVE IMPACTS OF SCENARIOS 	 II-B-3-23

-------
                                                               page

    3.2.  WATER IMPACTS  ................... II-B-3-47

          3,2,1.  CHANGES IN SURFACE WATER QUANTITY  ..... II-B-3-47

                  3,2.1.1.   DRAINAGE NETWORK ......... II-B-3-47
                  3.2.1.2.   HYDROLOGIC FACTORS ........ II-B-3-54
                  3.2.1,3.   WATER USE FACTORS  ........ II-B-3-55

          3.2.2.  CHANGES IN GROUND WATER QUANTITY ...... II-B-3-58

                  3.2,2.1.   HYDROLOGIC FACTORS ........ II-B-3-58
                  3,2,2.2.   WATER USE FACTORS  ........ II-B-3-59

4.  ECOLOGICAL IMPACTS ..... . .............. II-B-4-1

    4,1,  AIR POLLUTANTS FROM NUCLEAR AMD FOSSIL FUELD POWER
          PLANTS AND THEIR EFFECTS ON THE BIOTA  ....... II-B-4-1

          4,1,1.  AIR EMISSIONS FROM NUCLEAR PLANTS  ..... II-B-4-1
        -  4,1.2.  AIR EMISSIONS FROM FOSSIL FUEL POWER
                  PLANTS ................... II-B-4-6

    4,2.  EFFECTS OF POWER PLANT OPERATION ON AQUATIC
                     'RESOURCES ................ II-B-4-8
          4,2.1,  THERMAL STRESS ON AQUATIC SYSTEMS  ..... II-B-4-8
          4,2.2.  EFFECTS OF IMPINGEMENT BY CONDENSER SCREENS
                  ON AQUATIC ORGANISMS ............ II-B-4-9
          4,2.3.  EFFECTS OF CONDENSER PASSAGE ON AQUATIC
                  ORGANISMS  ................. II-B-4-9
          4,2.4.  EFFECTS OF CHEMICAL DISCHARGES ON AQUATIC
                  ORGANISMS  .  .  .  ...... . ....... II-B-4-1 0
          4.2.5.  SEDIMENT QUALITY  AND BENTHIC BIOTA ..... II-B-4-10
          4.2.6.  EFFECTS OF COAL MINING ON AQUATIC RESOURCES. II-B-4-10
          4,2.7.  COOLING PONDS  ............... II-B-4-11 '

    4.3.  EFFECTS OF POWER DEVELOPMENT ON TERRESTRIAL HABITATS
          IN THE ORBES REGION  ................ II-B-4-12

          4.3.1.  EFFECTS OF COAL MINING ON TERRESTRIAL
                  RESOURCES  ................. II-B-4-12
          4,3.2.  EFFECTS OF THE TRANSPORTATION ON THE
                  TERRESTRIAL RESOURCES OF THE ORBES REGION  . II-B-4-1 3

    4.4,  TERRESTRIAL SYSTEMS  ................ II-B-4-1 3

          4.4.1,  FORESTED AREAS ............... II-B-4-13
          4,4,2.  RIPARIAN SYSTEMS  .............. II-B-4-1 5
          4,4.3.  OPEN AREAS ................. II-B-4-16

-------
                                                               page

    4.5.  AQUATIC ECOLOGICAL RESOURCES OF THE OHIO RIVER
          BASIN	II-B-4-16

          4.5,1,  AQUATIC RESOURCES  	 II-B-4-16
5,  ECONOMIC IMPACTS

    5.1,  INTRODUCTION   	, . , . II-B-5-1

    5,2.  REGIONAL EMPLOYMENT AND INCOME .  	 II-B-5-2

          5.2.1.  BASLINE CONDITIONS 	 II-B-5-2
          5.2.2,  INCOME PROJECTIONS: SCENARIOS I AND II ... II-B-5-3
          5.2.3.  INCOME PROJECTIONS: SCENARIOS III. AND IV . . II-B-5-4
          5.2.4,  EMPLOYMENT PROJECTIONS 	 , 	 II-B-5-4

    5.3,  CAPITAL REQUIREMENTS FOR ENERGY DEVELOPMENT  .... II-B-5-5

    5.4.  DIFFERENTIAL IMPACTS WITHIN ORBES REGION , '	II-B-5-6

          5.4,1,  BASELINE CONDITIONS  ...,,,, ,,.,.. .. , .II-B-5-6

    5,5.  QUESTIONS TO BE'ADDRESSED IN FUTURE-PHASES'OF THE
          STUDY	II-B-5-21

    5,6.  INTRODUCTION TO IMPACTS ON HOUSING 	 II-B-5-24

          5.6,1.  HOUSING MARKETS  	 II-B-5-24
          5,6.2,  HOUSING AND GROWTH DUE TO ENERGY
                  DEVELOPMENT	.  .  . .	II-B-5-25

    5.7,  ENERGY PLANTS AND GROWTH \H THE ORBES REGION .... II-B-5-26

          5.7.1,  HOUSING IN THE IMPACTED COUNTIES 	 II-B-5-26

                  5.7.1.1.   KENTUCKY 	 II-B-5-33
                  5.7,1,2.   INDIANA  	 II-B-5-34
                  5,7.1.3.   ILLINOIS 	 II-B-5-34
                  5,7,1,4,   OHIO 	 II-B-b-35
                  5,7.1.5.   GROWTH DUE TO ENERGY DEVELOPMENT . II-B-5-35

    5,8,  HOUSING IMPACT OF GROWTH 	 II-B-5-36

    5,9,  DIFFERENCES BETWEEN SCENARIOS  	 II-B-5-38

-------
                                                               page

6.  SOCIOLOGICAL IMPACT OF PRESENT ENERGY SYSTEMS	II-B-6-1

    6,1.  RTC CRITIQUE	,	II-B-6-1

    6,2,  POLICY RECOMMENDATIONS . .  ,  	 II-B-6-2

    6.3.  PRELIMINARY ASSESSMENT OF SOCIAL IMPACTS 	 II-B-6-3

          6,3,1.  INTRODUCTION AND OVERVIEW OF THE IMPACT
                  ANALYSIS .  , . ,	II-B-6-3

                  6.3,1.1.  KENTUCKY COUNTIES  	 II-B-6-5
                  6,3,1,2.  ILLINOIS COUNTIES  	 II-B-6-9
                 ' 6.3.1.3.  INDIANA COUNTIES 	 II-B-6-12
                  6,3.1,4,  OHIO COUNTIES  	 II-B-6-16

7.  IMPACTS ON PUBLIC HEALTH

    7.1,  HEALTH EFFECTS DUE TO RADIATION  	 II-B-7-1

          7,1.1,  RADIATION PROTECTION  GUIDELINES  	 II-B-7-1
          7,1,2,  THE DOUBLING DOSE FOR CANCER 	 II-B-7-6

    7.2.  'PUBLIC HEALTH ASPECTS OF TO'SSIL FUEL COMBUSTION  . . II-B-7-11

          7,2.1,  HEALTH EFFECTS ?F SULFUR DIOXIDE EMISSIONS . II-B-7-11
          7.2,2.  HEALTH EFFECTS OF NITROUS OXIDE EMISSIONS  . II-B-7-14
          7.2.3,  HEALTH EFFECTS DUE TO PARTICULATE EMISSIONS. II-B-7-14
          7,2,4.  RADIOACTIVE EMISSIONS FROM COAL-FIRED
                  PLANTS 	 II-B-7-15

    7,3,  IMPACTS OF THE FOUR SCENARIOS	II-B-7-18

-------
                                 TABLES

tables                                                             page

II-B-1-1        % Kwh GENERATED IN 1975 BY TYPE OF PRIME MOVER .    II-B-1-3

II-B-1-2        % Kwh GENERATED IN 1975 BY TYPE OF FUEL ....     II-B-1-3

II-B-1-3        % Kwh GENERATED IN 1975 BY TYPE OF PRIMER MOVER  .   II-B-1-4

II-B-1-4        ELECTRIC UTILITY COMPANIES SERVING IN THE ORBES
                REGION .	     II-B-1-6

II-B-1-5        INTERCONNECTIONS OF PRIVATE UTILITIES SERVING
                ORBES	     II-B-1-8

II-B-1-6        ILLINOIS CONVERSION FACILITIES .	     II-B-1-17

U-B-1-7        KENTUCKY CONVERSION FACILITIES 	     II-B-1-18

II-B-1-8        OHIO CONVERSION FACILITIES ...	     II-B-1-19

II-B-1-9        INDIANA CONVERSION FACILITIES	      II-B-1-20

II-B-1-10       REGIONAL TECHNOLOGY CONFIGURATION SITINGS AND
                PLANT CHARACTERISTICS FOR KENTUCKY 	        II-B-1-21

II-B-1-11       REGIONAL TECHNOLOGY CONFIGURATION SITINGS AND
                PLANT CHARACTERISTICS FOR ILLINOIS 	        II-B-1-22

II-B-1-12       REGIONAL TECHNOLOGY CONFIGURATION SITINGS AND
                PLANT CHARACTERISTICS FOR OHIO 	        II-B-1-23

II-B-1-13       REGIONAL TECHNOLOGY CONFIGURATION SITINGS AND
                PLANT CHARACTERISTICS FOR INDIANA 	      II-B-1-24

II-B-1-14       CLASSIFICATION OF COALS BY RANK 	      II-B-1-28

II-B-1-15       TYPICAL ANALYSES OF COALS OF VARIOUS RANKS .  .     II-B-1-29

II-B-1-16       S02 EMISSIONS STANDARD REQUIREMENTS 	      II-B-1-27

II-B-1-17       CURRENT FEDERAL REGULATIONS FOR STATIONARY
                SOURCES OF NOX . .  .  . •	     II-B-1-31

II-B-1-18       TRACE ELEMENT CONTENT OF AMERICAN COALS .  .  .      II-B-1-32

II-B-1-19       DISTRIBUTION OF ENVIRONMENTALLY HAZARDOUS TRACE
                ELEMENTS 	        II-B-1-33
                                II-B-vii

-------
                           TABLES (continued)





tables                                                             page
II-B-1-20


II-B-1-21
II-B-1-22

II-B-1-23

II-B-1-24

II-B-1-25

II-B-1-26

II-B-1-27

II-B-1-28
II-B-1-29

II-B-1-30

II-B-1-31
II-B-1-32
II-B-2-1

II-B-2-2

II-B-2-3

II-B-2-4

SUMMARY OF ADDITIONAL ANALYTICAL VALUES FOR
25 SAMPLES OF HERRIN (No. 6) COAL FROM THE
ILLINOIS BASIN (ppm) 	
THORIUM-232 AND URANIUM-239 DECAY SERIES ...
LOCATION QUOTIENTS BY INDUSTRIAL SECTORS,
ORBES STATES 	 	
INDUSTRIAL DISTRIBUTIONS IN KENTUCKY AND ORBES
TARGET COUNTIES 	
INDUSTRIAL DISTRIBUTIONS" IN INDIANA AND ORBtS
TARGET COUNTIES 	
INDUSTRIAL DISTRIBUTIONS IN OHIO AND ORBES
TARGET COUNTIES 	
INDUSTRIAL DISTRIBUTIONS IN ILLINOIS AND ORBES
' TARBET1 'COUNTIES 	 	
AVERAGE MONTHLY AIR TEMPERATURES AND PRECIPITATION
AMOUNTS FOR SELECTED STATIONS ON THE OHIO RIVER .
STEADY FLWO DISCHARGE AT THE McALPINE DAM ...
PERIODS OF EXCEPTIONAL LOW FLOW AT THE McALPINE
DAM 	 	
VALUE AND COMPOSITION OF THE MINERALS INDUSTRY
IN THE ORBES STATES 	
PROVEN FUEL RESERVES 	
AVERAGE COMPOSITION OF COALS IN THE ORBES REGION .
POPULATION ESTIMATES AND PROJECTIONS FOR THE ORBES
REGION 1975, 1085, and 2000 	
POPULATION ESTIMATES AND PROJECTIONS FOR ILLINOIS
COUNTIES (ORBES REGION) 	
POPULATION ESTIMATES AND PROJECTIONS FOR INDIANA
COUNTIES (ORBES REGION) 	
POPULATION ESTIMATES AND PROJECTIONS FOR KENTUCKY
COUNTIES (ORBES REGION) 	


II-B-1-34
II-B-1-36

II-B-1-39

II-B-1-40

II-B-1-42

II-B-1-44

II-'B-T-46

II-B-1-53
II-B-1-54

II-B-1-55

II-B-1-59
II-B-1-60
II-B-1-61

II-B-2-13

II-B-2-14

II-B-2-18

II-B-2-22
                               II-B-viii

-------
                           TABLES (continued)
tables                                                             page

II-B-2-5        POPULATION ESTIMATES AND PROJECTIONS FOR OHIO
                COUNTIES (ORBES REGION) 	     II-B-2-27

II-B-3.1-3      COMPARISON OF POWER PLANT EMISSIONS DATA ....    II-B-3-4

II-B-3.1-4      COUNTIES DESIGNATED AQMA's IN WHICH ORBES SCENARIO
                COAL-FIRED POWER FACILITIES HAVE BEEN LOCATED .  .   II-B-3-7

II-B-3.1-5      NEW SOURCE PERFORMANCE STANDARDS FOR COAL FIRED
                POWER PLANTS WITH INPUTS OF 250 x 1Q6 BTU or
                GREATER 	   II-B-3-11

II-B-3.1-6      ALLOWABLE PSD INCREMENTS 	    II-B-3-14

II-B-3.1-7      COMPARISON OF NAAQ AND ESTIMATED MAXIMUM CONCEN-
                TRATIONS FROM ORBES COAL UTILIZATION FACILITIES  .   II-B-3-14

II-B-3.1-8      HEAT REJECTION OF FOSSIL AND NUCLEAR POWER
                PLANTS	    -II-B-3-17

II-B-3.1-9      ANNUAL POLLUTANT LOADINGS FROM COAL-ELECTRIC
                PLANTS AND COAL GASIFICATION PLANTS 	     II-B-3-19

II-B-3.2.0.-1   COAL ENERGY PRODUCTION ADVERSE IMPACT LEVEL .  .     II-B-3-48

II-B-3.2.0.-2   NUCLEAR ENERGY PRODUCTION ADVERSE IMPACT LEVEL .    II-B-3-49

II-B-3.2.1.3.2  ESTIMATED WATER WITHDRAWAL AND PROJECTED REQUIRE-
                MENTS BY PURPOSE, UNITED STATES IN BILLION GALLONS
                DAILY, (%}	_	    II-B-3-57

II-B-4-1        PRINCIPAL RADIONUCLIDES RELEASED IN GASEOUS
                EMISSIONS FROM A 3WR	    II-B-4-2
                  \
II-B-4-2        BIOLOGICAL HALF-LIVES OF ELEMENTS IN MAJOR
                GROUPS OF ORGANISMS 	  	     II-B-4-4

II-B-5-1        POPULATION, EMPLOYMENT AND GNP, 1974 	      II-B-5-2

II-B-5-2        EMPLOYMENT BY SECTOR (TOTALS FOR THE STATES
                ILLINOIS, INDIANA, KENTUCKY, AND OHIO)	     II-B-5-3

II-B-5-3        CAPTIAL REQUIREMENTS AND GNP (1985-2000) .  .  .      II-B-5-5

II-B-5-4        NEW  SITES PLANNED WITHIN SMSA's (1985-2000) .  .     II-B-5-16
                                II-B-ix

-------
                           TABLES (continued)


tables

II-B-5-5        NEW SITES PLANNED OUTSIDE SMSA's BUT WITHIN
                COMMUTING DISTANCE OF SMSA OR CITY OF 20,000
                POPULATION		       II-B-5-17

II-B-5-6        NEW SITES PLANNED OUTSIDE SMSA's BUT WITHIN
                COMMUTING DISTANCE OF SMSA OR CITY OF 20,000
                POPULATION 	       II-B-5-18

II-B-5-7        ILLINOIS COUNTIES 	       II-B-5-27
                KENTUCKY COUNTIES 	       II-B-5-28
                OHIO COUNTIES 	       II-B-5-29
                INDIANA COUNTIES	      II-B-5-30

II-B-5-8        SURPLUS CONSTRUCTION IN LABOR SHEDS 	       II-B-5-37

II-B-7-1        PERMISSIBLE AREA RADIATION DOSE 	       II-B-7-3

II-B-7-2        EFFECTS OF EXTERNAL RADIATION 	       II-B-7-5

II-B-7-3       ?BEST"ESTIMATES 'OF DOUBLING DOSE OR -RADIATION
                FOR HUMAN CANCERS 	       II-B-7-7

II-B-7-4        ASSIGNED DOUBLING DOSE AND INCIDENCE RATES . .      II-B-7-9

II-B-7-5        ESTIMATED DOUBLING DOSES FOR  CRITICAL PRE-DEATH
                YEARS	       II-B-7-10

II-B-7-6        HEALTH IMPACTS OF SULFATE AEROSOL 	       II-B-7-13

II-B-7-7        AIRBORNE RADIOACTIVITY CONCENTRATIONS AND DOSE
                RATES AT THE WIDOWS CREEK PLANT ON 5/13/69 . .      II-B-7-16

II-B-7-8        ANALYSES OF RADIOACTIVITY IN  COAL AND OIL FLY
                ASH	    II-B-7-17
                                II-B-x

-------
 FIGURES
figures
II-B-1-1

II-B-1-2
II-B-1-3
II-B-3.1-1

II-B-3.1.2

II-B-3.1-3

II-B-3.1-4

II-B-3.1-5
II-B-3.1-6

II-B-3.1-7

II-B-3.1-8

II-B-3.1-9

II-B-3.1-10
II-B-3.1-11
II-B-3.1-12
•
II-B-3.1-13

II-B-3.1-14

II-B-3.1-15


ELECTRIC UTILITY SERVICE AREA IN AND AROUND THE
ORBES REGION 	
OHIO RIVER BASIN ENERGY STUDY REGION 	
COAL FIELDS OF THE UNITED STATES 	
AIR QUALITY MAINTENANCE AREAS FOR SULFUR DIOXIDE
AND PARTICULATES 	
GEOGRAPHICAL DISTRIBUTION OF TYPICAL URBAN SUL-
FATE LEVELS IN THE UNITED STATES 	
GEOGRAPHICAL DISTRIBUTION OF. TYPICAL NONURBAN
SULFATE LEVELS IN THE UNITED STATES. .....
NATIONWIDE GEOGRAPHIC VARIATION IN S02 EMISSION
DENSITY 	 , 	 	
NATIONAL VISIBILITY - ISOPLETHS FOR 27 July 1974
S02 EMISSIONS FROM KENTUCKY POWER PLANTS BY
SCENARIO 	 	 	
S02 EMISSIONS FROM ILLINOIS POWER PLANTS BY
SCENARIO . . 	
S02 EMISSIONS FROM INDIANA POWER PLANTS (STATE
ESTIMATE) 	
S02 EMISSIONS FROM INDIANA POWER PLANTS
(EPA ESTIMATE) 	
S02 EMISSIONS FROM OHIO POWER PLANTS BY SCENARIO
S02 EMISSIONS - ORBES TOTAL .... 	
PARTICULATE EMISSIONS FROM KENTUCKY POWER PLANTS
BY SCENARIO 	
PARTICULATE EMISSIONS FROM ILLINOIS POWER PLANTS
BY SCENARIO 	
PARTICULATE EMISSIONS FROM INDIANA POWER PLANTS
BY SCENARIO 	
PARTICULATE EMISSIONS FROM OHIO POWER PLANTS BY
SCENARIO 	
page

II-B-1-10
II-B-1-14
II-B-1-26

II-B-3-9

II-B-3-10

II-B-3-10

II-B-3-10
II-B-3-22

II-B-3-25

II-B-3-25

II-B-3-26

II-B-3-26
II-B-3-27
II-B-3-27

.II-B-3-28

II-B-3-28

II-B-3-29

II-B-3-29
II-B-xi

-------
                           FIGURES (continued)


figures                                                            page

II-B-3.1-16     PARTICULATE EMISSIONS FROM TOTAL ORBES POWER
                PLANTS BY SCENARIO 	 ...       II-B-3-30

II-B-3.2.1.1.-1 WATER DIVERSION 	         II-B-3-50

II-B-3.2.1.1.-2 CROSS SECTION OF DIVERSION DUTCH APPLICATION .      II-B-3-51

II-B-3.2.1.1.-3 BOX-CUT MINING 	       II-B-3-52

II-B-3.2.1.1.-4 CROSS SECTION OF TYPICAL HEAD-OF-HOLLOW FILL .      II-B-3-53

II-B-3.2.1.3.-1 CONSUMPTIVE FLOW BY SCENARIO	      II-B-3-56

II-B-5-1        PERCENT EMPLOYED IN AGRICULTURE AND MINING . .      II-B-5-7

II-B-5-2        PERCENT OF COUNTY PERSONAL INCOME FROM
                AGRICULTURE AND MINING 	II-B-5-8

M-B-5-3       -TROX'IMITY OF'ORBES'COUTNTES TO  SMSA''s  ....       -II-B-5-9-

II-B-5-4        1975 GENERATING CAPACITY 	      II-B-5-10

II-B-5-5        NUMBER OF ELECTRIC GENERATING UNITS BY TYPE  OF
                FUEL:  BOM SCENARIO I (80/20)  	       II-B-5-11

II-B-5-6        NUMBER OF ELECTRIC GENERATING UNITS BY TYPE  OF
                FUEL in 2000 BOM SCENARIO II (50/50)  ....       II-B-5-12

II-B-5-7  .      NUMBER OF ELECTRIC GENERATING UNITS BY TYPE  OF
                FUEL in 2000 FORD TECH FIX SCENARIO III (100%
                COAL)	       II-B-5-13

II-B-5-8  .      NUMBER OF ELECTRIC GENERATING UNITS BY TYPE  OF
                FUEL in 2000 FORD TECH FIX SCENARIO IV (100%
                NUCLEAR) 	       II-B-5-14
                                II-B-xii

-------
                           1.  INTRODUCTION
           In order to reach man's never-ending quest for higher and
higher standards of living, it has required a corresponding ever-
increasing input of energy resources.  With each technological advance,
there has been not only a tremendous energy input into the production of
goods and services, but the goods and services themselves require
additional inputs of energy for their operation.  The automobile is a
classic example, along with the countless electrical applicances which
serve the home and our vocational livelihoods.  Indeed, man has evolved
into a truly hybrid animal; one who probably couldn't survive without
the inanimate energy devices and resources that are now at his service.

           This, in turn, raises both philosophical and factual questions
concerning our future energy requirements.  First, on the philosopical
side, where is the "break-even" point between energy resources that are
necessary and those that are a luxury?  In other words, have we permitted
ourselves to become slaves of our own masterful inventions?  And* if we
have, can we reverse the trend, or if we have not, can we rationally and
truthfully predict what our ultimate future energy reqiurements will be?

           On the factual side, the problems are much more easily defined
even though the solutions to the problems remain elusive.  First, the
world as a whole and most of its geographic subcomponents have enjoyed
plentiful, cheap energy.  Because it was cheap, there was very little
incentive to use it "in a conservative manner.  'Secondly, western civiliza-
tion for the past five or six centuries has been rooted in a tradition
which encouraged growth in population, the production of goods and services
and the accumulation of capital wealth.  None of these events could have
occurred without an even more rapid rate of growth in the consumption of
energy resources.  A fallacious legacy of this tradition is that continu-
ous and infinite growth will occur on our finite planet, Earth.

           Thirdly, our heritage of production, including energy resources,
has ignored the ecologic frailty of our planet.  The interdependences of
the physical  and biological components of the planet, Earth, have been  .
substantially modified in some instances and the future results of these
modifications could hold severe consequences.  It is only recently that we
have begun to understand the importance of some of these complex inter-
dependencies and why it is essential they should not be disturbed.

           This preliminary background sets the stage for an initial look
at only one energy resource -- electricity.  Even though the ORBES project
will eventually consider the production and consumption patterns for
other types of energy within the study area, there appears to be an over-
riding importance attached to the production, distribution and consumption
of electrical energy.   To begin with, the 'production of electricity is
essentially a ubiquitous function throughout the United States.  Every
state in the U. S.  produces and distributes electric power, and when a
particular area demands additional  electric energy, it is generally pro-
duced nearby.  Such is not the case with the production of other forms of
energy.  Fossil fuels and geothermal energy resources are found within
                             II-B-1-1

-------
specific geologic regions; water power and solar energy are controlled by
meteorological and topographic conditions.  Nuclear fuel is manufactured
or processed into usable energy forms and transported to its ultimate
production site just like any other commodity.  In other words, electric-
ity is essentially produced near its points of consumption.  In fact, a
recent EPRI report stated that the average distance between production
centers and consumption centers for electric power in the United States
Is only 50 miles (1).

           The fuels used to generate electricity in the four-state ORBES
region strongly reflect the mineral resource base -- coal.   Tables II-B-
1-1 and II-B-1-2 show that currently, about 75% of the nation's electric-
ity is produced by conventional (fossil fuel) steam, and just over half
of that energy is provided by coal.  However, in the four-state region,
over 90% of the electric generation is produced by conventional steam,
and nearly 90% of the steam is produced in coal-fire boilers.  Although
nuclear steam generation is presently confined to the northern half of
Illinois, future nuclear generating units are already under construction
in Ohio and in the advanced licensing stage in Indiana.  At the present
time, nuclear generating units are still in the discussion stage for
Kentucky.

           In 1975, the four-state region, with about 15% of the nation's
population, generated nearly 16% of the nation's electricity.  Of this
total, 87.6% was generated by the investor-owned electric utilities,
with the remainder being supplied from municipal, cooperative, industrial
and governmental'generating units.  In fact, 'the study region receives
approximately 10% more electrical energy through locally based investor-
own?d electric utilities than does the nation as a whole (See Table II-
B-l-3).

           The investor-owned electric utilities (hereafter called
utility companies or utilities) are publicly-regulated monopoly businesses.
They are required to provide electric power service to customers located
within their respective service areas at rates which are set by the
Public Service Commissions within each state.  And on the other hand, as
an investor-owned business, they are responsible to, and their profits
are divided among, their stockholders.

           At this point in the preliminary analysis, an important
digression is in order.  One of the major functions of the ORBES Task I
group was to determine (a) how much electric energy the ORBES region
would require through the year 2000, (b) what fuels would be used to
generate it, and (c) where would the most likely sites (counties) for
generation be located.   These considerations, among others, were
addressed by the Task I group with the assistance of an advisory com-
mittee, and judgments were handed down without fully considering who
would actually produce the electricity and how it would be consumed.
Although it might be argued that these considerations should be evaluated
by other task groups in the study, the fact that the sites  were
                             II-B-1-2

-------
                          TABLE II-B-1-1





          % Kwh Generated in 1975 by Type of Prime Mover



                  TOTAL ELECTRIC UTILITY  INDUSTRY

Hydro
Conv. Steam
Nuc. Steam
Int. Combust.
U.S.
15.65
75.08
8.97
.29
Ohio, 111.,
Ind., Ky.
1.31
91.19
7.30
.19
*Source:  EEI Stat Yearbook,  1975.   Table 13 S.,.p.  21
                          TABLE II-B-1-2





             % Kwh Generated in 1975 by Type of Fuel



                  TOTAL ELECTRIC UTILITY INDUSTRY

Coal
Fuel Oil
Gas
Nuc.
*Ibid., Table 14 S, p. 23
U.S.
52.87
17.90
18.58
10.66

Ohio, 111.,
Ind. , Ky.
88.56
2.58
1.46
7.40

                        II-B-1-3

-------
                           TABLE II-B-1-3



           % Kwh Generated in 1975 by Type of Primer Mover



                  INVESTOR-OWNED ELECTRIC UTILITIES

Hydro
Conv. Steam
Nuc. Steam
Int. Combust.
U.S.
5.47
84.05
10.43
0.04
Ohio, 111.,
Ind., Ky.
0.36
91.28
8.34
0.03
*Source:  EEI Stat Yearbook, 1975.   Table 15 .S,  p.  24
                        II-B-1-4

-------
 selected by the Task  I team imposed rigid constraints upon the subsequent
 assessment.  The vital information which has escaped the Task I team,
 collectively,  is essentially the structural nature of the electric
 utility industry, the integrated systems which serve specific geographic
 areas, and the increasing complexity of future generation, transmission
 and distribution agreements.

           To begin with, the ORBES region receives electricity from 25
 or 30 corporate utilities, depending upon how one considers multi-unit
 holding companies such as American Electric Power, the Allegheny Power
 System, Electric Energy, Inc., Ohio Valley Electric Corporation, or the
 Tennessee Valley Authority (See Table II-B-1-4).  The service areas of
 these companies, which provide electric power to some part of the 'defined
 region, extend from Central Pennsylvania and Virginia in the east to
 Mississippi in the south, and well into Missouri and Iowa on the west.
 Figure II-B-1-1 provides an approximation of these service areas.  Some
 of these utilities generate electricity within the ORBES region, largely
 for their own system use outside of the region, such as Union Electric
 and TVA, while others generate electricity outside of the region for
 use within, such as the Allegheny Power System and American Electric Power.

           While each utility is responsible for providing electric service
 to its own service area, the methods by which this is accomplised vary
 among the corporate entities.  Some utilities have formed power pools,
 such as CAPCO, KIP, MI10, etc. where interchange agreements have been
 formulated, whereas other utilities have elected to remain more autonomous
 in generating, 'transmitt'irig "and distributing electric-power.  However,
 all utilities that serve the study area are interconnected with adjoining
 utilities (See Table II-B-1-5) for emergency purposes.  In order to
 facilitate the efficient interchange of electric power, a series of 9
 regional reliable councils were established throughout the United States.
 These regional councils report regularly to a national reliability council
whose principal function is planning and coordination of electrical system
 services such that sufficient electricl  energy can be reliably delivered
 to various sections of the country and that crippling power failures such
as the.one which struck the eastern U.  S.  in the 1960's can be averted.
The regional  councils which serve the ORBES region are ECAR (Ohio,
 Indiana and most of Kentucky), MAIN (Illinois), and SERC (the TVA service
areas in southern Kentucky).

           Virtually all  of the utilities  generation, transmission and
 distribution functions are continuously monitored systemwide at a control
center.  The basic procedure followed at a control center is first to
estimate the next day's megawatt load for the entire system and set,
through remote controls,  the generating units at a capacity level  to meet
the expected demand.  Then, by observing continuous recording devices
which meter the inflows or outflows at each of the system's interconnections,
the accuracy of the load estimate can be verified.  If power is flowing
 into the system from the interconnected utilities, then there is in-
 sufficient electricity being generated and the output of a specified
                             II-B-1-5

-------
                       Table  II-B-1-4

   ELECTRIC UTILITY COMPANIES SERVING  IN THE ORBES REGION
Identifying
1975 Owned
Capacity
Code*
AEP
APPC
INMC
KEPC
OHPC
APS .
MOPC
POEC
WEPP
CLEI
TOEC
DULC
OHEC
PEPC
COSO
CIGE
DAPO
KEUC
PSIN
INPL
LOGE
TVA
SOIG
Company Name
American Electric Power Co.
Appalachian Power Co.
Indiana Michigan Power Co.
Kentucky Power Co.
Ohio Power Co.
Allegheny Power System
Monongahela Power Co.
Potomac Edison Co.
West Penn Power Co.
Cleveland Electric Illuminating Co.
Toledo Edison Co.
Duquesne Light Co.
Ohio Edison Co.
Pennsylvania Power Co.
Columbus and Southern Ohio Electric Co.
Cincinnati Gas and Electric Co.
Dayton Power and Light Co.
Kentucky Utilities Co.
Public Service Indiana
Indianapolis Power and Light Co.
Louisville Gas and Electric Co.
Tennessee Valley Authority
Southern Indiana Gas and Electric Co.
in MW
17,212




6,429



3,894
1,062
2,465
3,752

2,105
3,695
2,365
1,706
3,825
2,013
2,177
22,674
750
                           II-B-1-6

-------
                        Table  II-B-1-4 (cont.)
Identifying
1975 Owned
Capacity
Code*
NOIP
COEC
ILPC
CEIL
CEIP
IOGE
UNEC
OVEC**
ELEN**
BUPC
EAKP
BIRI
SOIP
HOOE
* Subsidiary
** Owned by a
Company Name
Northern Indiana Public Service Co.
Commonwealth Edison Co.
Illinois Power Co.
Central Illinois Light Co.
Central Illinois Public Service Co.
Iowa-Illinois Gas and Electric Co.
Union Electric Co.
Ohio Valley Electric Corporation
Electric Energy Incorporated
Buckeye Power Cooperative
.East Kentucky .Power Cooperative
Big Rivers Power Cooperative
Soughern Illinois Power Cooperative
Hoosier Electric Power Cooperative
or member companies idented.
consortium of invested-owned utilities
in MW
1,885
15,337
3,394
1,069
1,632
931
5,595
2,390
1,100
N.A.
N.A.
N.A.
N.A.
N.A.


SOURCE: Electric Light and Power. Aug. 1976, p. 36
                           II-B-1-7

-------
                                                     Table II-B-1-5


                                  INTERCONNECTIONS OF PRIVATE UTILITIES SERVING ORBES*
CO
 i
CO

CLEI
TOEC
OHEC
COSO
OHPC
CIGE
DAPO
MOPC
APPC
KEPC
KEUC
LOGE
TVA
SOIG
PSIN
CTOCOCDMAKKLTSPINCIC
LOHOHIAOPEEOVOSNOOLE
EE.ESPGPPPPUGAIIPIEPI
ICCOCEOCCCCE GNLPCCL
X X
XXX
XX XX
X XX
XX XX XXX
XX XXX
X X
X
X XXX
X XX
X X X X X
X X X X X
X X X X X .X
X X
/
X X X X X X
C I I U
E N 0 N
I M G E
P C E C




X
X


X
X


X

X

-------
I
oo
i
                                                     Table II-B-1-5 (cont'd.)



                                  INTERCONNECTIONS OF PRIVATE UTILITIES SERVING ORBES*

INPL
NOIP
COEC
ILPC
CEIL
CEIP
INMC
IOGE
UNEC
DULC
PEPC
CTOCOCDMAKKLTSPINC
LOHOHIAOPEEOVOSNOO
EEESPGPPPPUGAIIPIE
ICCOCEO^CCCGE GNLPC
X XX
X
X
X X
X
XXX X
X XX X X X X-
X
X
XX X
X X
I C
L E
P I
C L


X X
X
X
X
X
X
X


C I I U
E N 0 N
I M G E
P C E C
X
X
X
X X

X X


X


          *  See Table II-B-1-4 for identifying utility codes.



          SOURCE:  Moody's Public Utility Manual, 1976.

-------
                                                 /r.?*^^\/^^'f?^^s^^\/>  j\x'—VK /"-«••"' /..--ry?C i Q-----.:V'^^..-„
                                                 ifrtf^^^&x^Ct1^*?^ fc*2%^hr!^^&Z£-$&
                                                 )..,& \H/y^"^ ^::^\':^-:q:^%:^?NV^^L---'" r	/^ -^^^^V^^::^^^^-
                                                 '  ^''^-f  •-j:t-S'v—iV-'-'i  -' •.1-i~"if '.•-.£•. V V---N  f >—-^%  -/-•'  < p""//'^   ':;^--^-^'.---.^-~-- •?•::
                                                 -'^•^^-^v^?^-!	'c'>.-XC^s\.i>.- V'I^V:' c-AA.-^X"'-'/-V^M-i^<:-' C<-'"r-:' -c V^.'^^^ /-:>^-
                                                 d^^fe4i*SM^!5^^^^^
                                                                                     Figure   II-B-1-1
                                                                                          Electric Utility Service Area
                                                                                          in and around the ORBES region
                                                                                          Source:  Cahners Publishing
                                                                                          Company Map, Cambridge, Mass..,
                                                                                          196
co
i
 i
H-*
o

-------
                        Legend

                    Figure  II-B-1-1
APPC       Appalachian Power Company
CEIL       Central  Illinois Light Company
CEIP       Central  Illinois Public Service Company
CIGE       Cincinnati Gas and  Electric Company
CLEI       Cleveland Electric  Illuminating Company
COEC       Commonweath Edison
COSO       Columbus and Southern Ohio
DAPO       Dayton Power and Light Company
DULC       Duquesne Light Company
ILPC       Illinois Power Company
INME       Indiana-Michigan Electric Company
INPL       Indianapolis Power  and Light
IOGE       Iowa-Illinois Gas and Electric Company
KEPC       Kentucky nower Company
KEUC       Kentucky Utilities
LOGE       Louisville Gas and  Electric Company
MOPC       Monongahela Power Company
NOIP       Northern Indiana Public Service Company
OHEC       Ohio Edison Company
OHPC       Ohio Power Company
PEPC       Pennsylvania Power  Company
'POEC      'Potomac Edison"Company
PSIN       Public Service Indiana
SOIG       Southern Indiana Gas and Electric Company
TOEC       Toledo Edison Company
TVA        Tennessee Valley Authority
UNEC       Union Electric Company
WEPP       West Penn Power
                   II-B-1-11

-------
generator(s) is raised.  Gonversely, the generating capacity is lowered
if electricity is flowing out of the system through the interconnections.
When an equilibrium is reached, the (instantaneous) system load can be
determined, and the contribution that each piece of generating equipment
is making to provide that load.  System loads, however, are constantly
changing each hour of the day as well as seasonally.  But through con-
tinuous monitoring of the interconnections, the total  generating capacity
can be raised or lowered to meet the system load.

          As increasing demands on the system load approach the margin of
safety, it becomes necessary to install additional generating capacity.
Long before this point is reached, however, a wide range of inter-
dependent decisions must be reached.  What type and size unit(s) should
be added to the system?  Where is the most economically favorable location
that will (a) serve the system load, (b) maintain equity within the trans-
mission system, (c) provide the necessary operating requirements, such
as cooling water, accessibility and minimize additions to the transmission
system?  Should the new unit be a joint effort with other utilities or
should it be the responsibility of a single company?  These are only a
few of the questions that would have to be answered regarding the addi-
tion of a new unit(s).                            '

           The more fundamental and underlying issue here is that if a
utility's service area continues to demand increasing quantities of
electric power, then that particular utility will provide it.  If the
future demand.do.es not grow, then there is no economically sound reason
to expand the utility system.  Of couse, utilities, just like Chambers
of Commce and other development organizations, have a justified and .
vested interest in promoting the movement of new customers into their
service area and for existing customers to use more of their product.

           Applying this issue to the ORBES region, it would seem more
logical to begin with each utility's service area as basic geographic
area (county level data would still generally apply),  analyze the amount
and type of growth that would occur within the area, and finally, allo-
cate new generating plants and transmission systems to sites whic!:
would most effectively serve that growth.   The sum of the growth within
all of the service areas may or may not correspond to final energy
demands of two scenarios now under consideration, but the approach
would appear to be more defensible than one of proportional national
shares.  In any event, such an approach to the question of size and
location of additional generating facilities within the ORBES region
would appear to provide a more realistic base from which to begin an
environmental, social  and economic assessment.

1.1.   DEFINITION AND BRIEF HISTORY OF TECHNOLOGY ASSESSMENT

           On May 10,  1876 President Ulysses S. Grant and Emperor Dom
Pedro of Brazil turned the levers of a 700 ton Corliss steam engine at
Philadelphia, Pennsylvania setting into motion some 8,000 machines
                             II-B-1-12

-------
spread over 13 acres.  The engine, and its mechanical dependents, were
heralded as a great achievement of the time, as indeed they were.  Yet
then as now there were some who questioned the validity of these claims.
Time Inc. Daniel J. Boorstin in his bicentennial essay, "Tomorrow:
The Republic of Technology" quotes English biologist, Thomas Henry
Huxley, as stating, following a visit to the 1876 Philadelphia exhibit,
that he was not "in the slightest degree impressed" (2).  "Size," said
Huxley, "is not grandeur, and territory does not make a nation.  The
great issue, about which hangs a true sublimity, and the terror of over-
hanging fate, is what are you going to do with all these things?"
Today, as we enter the era of technology assessment, we add the question,
"And what are these things going to do to us?"

           Technology assessment is not new to the twentieth century nor
is it in any sense unique to this decade.  The film "The City" produced
in 1937 with a musical score by Aaron Copeland and commentary by Lewis
Mumford, author of "Myth of the Machi ne," is a masterful and sensitive
study of the impact of industrialization on human existence (3, 4).  The
power of the film is not in tha4^ it addressed the major issues of public
health related to living conditions, accident rates and loss of identity
by the individual as a living being having singular significance as a
part of society.  These things you expect to hear.  What the authors and
producers of "The City" did was to give the film credibility by showing
that man can and at times does control his technology and make it one with
nature rather than be controlled and driven by it.  And even though the
model community pictured in the film has long since disappeared beneath
the pall -of Wa^h'tng'ton-to Baltimore -urban "sprawl, the quest for a better
plan persists.  Now, as a result of the Technology Assessment Act of
1972, we are required by law to assess the impact of new technological
developments on mankind (5).

           A team of scientists and engineers from the Universities of
Kentucky and Louisville have subsequently been given the responsibility
by the U.  S. Environmental Protection Agency of developing technology.
assessments for each of four energy production scenarios encompanning
extreme high and low projections for national energy needs.   Ostensibly,
these assessment are to be for the impact of energy production and
utilization in the lower Ohio River Basin - an area taking in all of
Kentucky and most of Ohio, Indiana and Illinois (Figure II-B-1-2).
There is some question, however, of this or any other team being able to
assess just the regional  impact.  Are we not, in terms of our national
power nets, one web?  What, for example, are we to say in our assess-
ments about the article reprinted below from the December 7, 1976 issue
of The Louisville Times?

           "^Electrical golden spike" will link U.  S. power grids

           (AP) Thornton, Colo.   A switch at a power plant here will serve
           as an "electrical  golden spike," linking U. S.  power systems
           just as the transcontinental  railroad joined the country's
           rail systems a century ago.  David Hamil, administrator of
                             II-B-1-13

-------
                                  FIGURE  II-B-1-2

OHIO   RIVER  BASIN  ENERGY  STUDY REGION
                           "X^vii-^'-'-'^i  o-' •'"'"': .O.K k::c*x
                           SSSSSV- | y,:-*/U^lMi::;



                                  i  <"' /RK—
                                       /  /  ,-k/
                         H-QTTTh^ i *xT7l
                                               ^0X^^:'
                              OHIO RIVER DRAINAGE BASIN
                              COUNTIES NOT IN REGION
                              0

                              I
                                            100
I   I
                          200
                                        Scale in Miles




                               SOURCE:  U.S. Bureau of the Census


                               Prepared by Cartographic  Laboratory and

                                         Energy Resources Center,  U I C C


                              II-B-1-14

-------
           the federal Rural Electric Administration, was to throw the
           switch today, joining the nation's east and west power grids
           at Stegall, Neb., near Scottsbluff.
           Uniting the two national power grids is considered necessary
           to prevent regional blackouts and power interruptions.   (

1.2.  THE REGIONAL TECHNOLOGY CONFIGURATIONS

           A number of projections of U. S. energy production have been
made in recent years.  The Task 1 study group, after a review of these
reports, chose to bracket the team's technology assessment between
analyses of impacts of two extremes.  The highest projection selected
was taken from the Bureau of Mines Report, "Energy Through the Year 2000,"
which predicted a demand for 163 quadrillon BTU's by the year 2000 ( ).
Its counterpart was taken from a low energy growth scenario of the Ford
Foundation's Energy Policy Project which projected a national demand for
117 quadrillon BTU's by the year 2000 (6).  Conversion of these figures
into numbers of nuclear, coal-fired and synthetic natural gas power
plants is discussed below.

           It is not the purpose of this study to evaluate the validity
of these projections.  Our task instead is to evaluate the environmental
and socio-economic impacts that would result from making such projections
realities by viture of their becoming national policy.  The reader
interested in projection evaluation is directed to the review by Allen L.
Hammond (Science, Volume 195, January 14, 1977) of Alvin Weinberg's three
volume tome,-^Economic -and Environmental_'Implications ;of_ a U. S. Nuclear
Moratorium," published by the Institute for Energy Analysis, Oak Ridge
Associated Universities, Oak Ridg*, Tennessee, in 1976.
                 •

           Development of the first four scenarios describing possible
energy configurations in the year 2000 for the ORBES region followed
the formate below.

                  Total U. S. Energy  (163-117 Quad's)
                                 *
     Percentage of total U. S. energy supplied by the ORBES region
     (assumed to be 16% for both high and low projections).
                                 4
     Mix of electric energy conversion facilities to supply energy
     in the ORBES region (only coal and nuclear to be considered).
                                •*
     Location of electric energy conversion facilities by county.

           Implicit in the above outline are several features which need
to be delineated.  These are:

     1.  The assumption was made that uranium and coal would be the only
         fuels available in the year 2000 in sufficient quantity to be
         of importance.  Solar, wind, and hydroelectric power, although
                             II-B-1-15

-------
         always available, would be so limited in quantity as to become
         inconsequential; and, in addition to this, it was further assumed
         that the world's oil would be depleted and/or too expensive for
         the utilities to buy by the year 2000.  These assumptions are
         probably valid for the 163 Quad scenarios but almost certainly
         invalid for those based on the 117 Quad figure.

     2.  It was also assumed that electrical power produced in ORBES will
         not be transported in significant quantity outside of the region.
         The validity of this assumption is the converse of the above.
         The low energy scenarios will surely develop into purely regional
         studies.  However, the high energy scenarios will be regional
         only is an increase in population is assumed to occur correspond-
         ing to the increased production of power - an assumption which
         appears now to be in some doubt (See Chapter 2.0, Section 2.2).

           Thus it would appear from the outset that four of the scenarios,
which the reader is reminded of were purposefully selected as extreme
cases, are indeed unlikely.  They will be evaluated nonetheless using the
population statistics compiled by the Task 1 group with the hope that'their
assessments will actually bracket the real situation.

           For the purpose of all four scenarios however, it was agreed
that the energy configurations-, in 1985 have already become fixed and are
subject to very little change.  Because the utilities have planned what
additional energy conversion facilities they will install in the next ten
years, ft Ms'-a 'safe assumption that what they plan is what will indeed
be the situation in 1985.

           The distributions of energy conversion facilities and their
characteristics for the four scenarios are given in Tables II-B-1-5 to
II-B-1-12.

1.2.1.  BOM HIGH 80/20 MIX.  The present mix of conversion facilities in
the ORBES region is shown in Tables II-B-1-1 and II-B-1-2.  It can be seen
from these data that coal currently dominates the fuel mix at 88.56% with
nuclear energy being the second most popular source at 7.40%.   Thus 96%
of the electricity currently being generated in the basin comes from these
two sources and the assumption of their being 100% of the source by the
year 2000 is probably sound.   It is evident that this scenario also pre-
dicts a slight rise in nuclear plants at the expense of coal-fired systems.

1.2.2.  ROM HIGH 50/50 MIX.  An alternate means of developing the basin's
share of i63 Quad's would be to go with nuclear power as at least an
equal to coal.  The impact of this scenario on the ORBES region would be
entirely different from the 80/20 mix, and an assessment of the two
should provide an interesting comparison.  The 80/20 mix scenario pre-
sents the system which one would expect to develop if we chose to simply
increase the region's output dramatically, but with a mix of conversion
facilities similar to the present.  In other words, business as usual but
                             II-B-1-16

-------
                                                        TABLE H-B-1-6

                                             ILLINOIS CONVERSION  FACILITIES
Facilities To Be Added Between 1985 And The Year 2000a
163 Quadb



VCoal
^Nuclear
ron
^j Hydro
Gas
Syng-Lc
Syng-H
Other
Present
Facilities
1975
25
1
35
2
3
0

17
Projected
Facilities
1985
36
6
38
2
4
0

17

Difference
1985-1975
11
5
3
0
1
0

0
Bureau of Mines
80% coal
20% nuclear
25
6
0
0
0
1
1
0
Projections
50% coal
50% nuclear
16
15
0
0
0
1
1
0
117 Quad
Ford Foundation

100% coal 1
2
0
0
0
0


0
Projection

00% nuclear
0
1
0
0
0


0
a .- All plants are 1000 MWE except for coal  conversion units under Ford Foundation scenario which are 600 MWE.

b - Quadrillon BTU's.

c - Synthetic natural  gas conversion units,  one producing low BTU gas and liquids and the other high BTU gas and
    liquids.

-------
                                                        TABLE  H-B-1-7

                                             KENTUCKY  CONVERSION  FACILITIES
Facilities To Be
163 Quadb



>—<
£coal
' Nuclear
>LOil
o° Hydro
Gas
Syng-Lc
Syng-H
Other
Present
Facilities
1975
19
0
4
7
4
0
0
1
Projected
Facilities
1985
29
0
5
7
4
1
0
5

Difference
1985-1975
10
0
1
0
0
1
0
4
Bureau of Mines
80% coal
20% nuclear
31
3
0
0
0
1
1
0
Projections
50% coal
50% nuclear
17
17
0
0
0
1
1
0
Added Between 1985 And The Year 2000a
117 Quad
Ford Foundation

100% coal 1
5
0
0
0
0
0
0
0
Projection

00% nuclear
0
3
0
0
0
0
0
0
a - All  plants are 1000 MWE except for coal  conversion  units  under Ford  Foundation  scenario which  are  600 MWE.

b - Quadrillon BTU's.

c - Synthetic natural  gas conversion units,  one producing low BTU gas and  liquids and  the  other  high BTU gas  and
    liquids.

-------
                                                       TABLE  .II-B-1-8

                                               OHIO  CONVERSION  FACILITIES
Facilities To Be
163 Quadb



TCoal
T'Nuclear
ron
•S Hydro
Gas
Syng-Lc
Syng-H
Other
Present
Facilities
1975
39
0
29
1
2
0
0
16
Projected
Facil ities
1985
48
2
26
2
0
0
0
16
Bureau of Mines Projections
Difference
1985-1975
9
2
-3
1
-2
0
0
0
80% coal
20% nuclear
48
12
0
0
0
1
1
0
50% coal
50% nuclear
30
30
0
.0
0
1
1
0
Added Between 1985 And The Year 2000a
117 Quad
Ford Foundation Projection

100% coal
15
0
0
0
0
0
0
0

100% nuclear
0
9
0
0
0
0
0
0
a - All  plants are 1000 MWE except for coal  conversion  units  under  Ford  Foundation  scenario which  are  600  MWE.

b - Quadrillon BTU's.

c - Synthetic natural  gas conversion units,  one producing low BTU gas  and  liquids and the other high BTU gas and
    liquids.

-------
                                                       TABLE  .H-B-1-9

                                              INDIANA  CONVERSION  FACILITIES
Facilities To Be Added Between 1985 And The Year 2000a




>— »
£coal
^Nuclear
r^Oil
0 Hydro
Gas
Syng-Lc
Syng-H
Other

Present
Facilities
1975
30
0
16
5
1
0
0
3

Projected
Facilities
1985
46
2
16
5
1
0
0
3


Difference
1985-1975
16
2
0
0
0
0
0
0
163
Bureau of M
80% coal
20% nuclear
27
7
0
0
0
1
1
0
Quadb
ines Projections
50% coal
50% nuclear
17
17
0
0
0
1
1
0
117 Quad
Ford Foundation

100% coal 1
5
0
0
0
0
0
0
0

Projection
'
00% nuclear
0
3
0
0
0
0
0
0
a - All plants are 1000 MWE except for coal  conversion  units  under  Ford  Foundation  scenario which are 600 MWE.

b - Quadrillon BTU's.

c - Synthetic natural  gas conversion units,  one producing low BTU gas  and  liquids and the other  high BTU gas and
    liquids.

-------
                                      REG


No. of Plants
BOM 80%-20%
s- "cu
re 3
OJ Li-







nn
co
i
i— •
i
ro
H-*









County
Ballard '
Bracken
Breckinridge
Butler
Carlisle
Gal latin
Rreenup
Henderson
Lewis
Livingston
Marshall
Mason
McLean
Meade
Owen
Russell
Scott
Trigg
Trimble
Union
Webster
Totals

o
o
2
2
2
1
2
3
2
0
0
2
2
0
1
2
1
0
0
2
3
2
2
31

13
0
0
C
0
0
0
0
0
1
0
0
0
0
0
0
2
0
0
0
0
0
3
c
LO
0
0
0
0
0
0
0 ,
1
0
0
0
0
0
0
0
0
le
0
0
0
0
1
No. of Plants
BOM 50%- 50%
i. "cu
10 3
CU U,
1 — 1 — 1
re
o
o
0
0
2
0
2
0
0
0
0
2
2
0
0
2
1
0
0
2
0
2
2
17
o
13
0
3
0
0
0
0
2
0
3
0
0
3
0
0
0
3
0
0
3
0
0
17
c
£
0
0
0
0
0
0
0
1^
0
0
0
0
0
0
0
0
le
0
0
0
0
1
No. of No. of
Plants FTF Plants FTF Cooling
re
o
c re
re i.
-C XI 3
a c +->
100% Coal 100% Nuclear s. £ £
X
X
X
X
X
x
2 . X

X
2.5 X
X
X
x
2.5 X
X
1 x
A A
Y
A
X
X
X
5 3
Transportation
• .£1 -r-
•<-> re
13 D:
o -~^
E cu
CU O)
C i-
•r- re
z: ca
X

X
X
x
x
X
x

X
x

x
X
x
V
A
X
X
X

Source0
c
re
•r- U C
o re i-
e .— cu
•r- 10 *J
t — O. CO
. — Q. CU
— 
x
x
X
X
X
x
. X


x
x

x
X
X


X
X
X

Type Mining
Q.
l/l
0
cu ^r
s z •*.
•£ z S
Q. 0
O. -r-
CU S- *«
CU +J -0
Q iTI U3
x
x
X
X
x
X
X
X

X
x

x
X
X
y
A
X
X
X

a - All  coal  fired plants will  have stack heights of 600 feet regardless of other SO;? control  methods.   An efficiency  of 98%  is  assumed  for  electrostatic
    precipitators.  Nuclear plants will  operate with an efficiency of 31% and coal  fired  plants  at 37%  for the  period  1985-2000.   All  plants,  coal  and
    nuclear,  will  operate at a  capacity factor of 51%.   A unit of conversion for all  nuclear plants is  equal  to 1000 MWe.   Coal jfired  plants have 1000 MWe
    as a unit of conversion for the BOM scenarios and 600 MWe for the Ford Tech Fix scenarios.
b - Elkhorn No. 3 is taken as a typical Appalachian Basin coal  and No.  9 is taken as typical  of the Illinois Basin.   The  composition  of these  coals  is  as
    follows:
                                                     Elkhorn No.  3            •         No.  9                                  ,
                 Percent Ash                                3.9                          10.5
                 Percent Sulfur                             0.9                           3.15
                 BTU/lb, Moisture Free                 14,200                        12,940

c - Compliance coal meeting the specification:  percent sulfur = (6 x 10~^)(BTU/lb).  Scrubbers are lime/limestone  non-regenerative type.

d - High BTU synthetic fuel plant producing 250 million cubic feet of pipe line quality gas per day at 950 BTU per  cubic  feet.

e - Low BTU synthetic fuel plant producing 1400million cubic feet of gas per day at 150 BTU per cubic feet.

-------
             TABLE II-B-1-11
REGIONAL TECHNOLOGY CONFIGURATION SITINGS
  AND PLANT CHARACTERISTICS FOR ILLINOIS

            Illinois Counties



County
Brown
Cass
Clark
Greene
Hamilton
Hancock
Henderson
Iroquois
Jersey
Lawrence
Livingston
Marshall
Mercer
Perry
Pulaski
St. Clair
Schuyler
Scott
Washington
White
Totals
*H1gh BTU
bLow BTU

No.
BOM

of plants
80% - 20%



Coal/Nuclear/Syn-Fuel
2
0
2
2
2
0
0
0
2
2
0
2
0
1
2
0
2
2
2
2
25


0
2
0
0
0
0
0
0
0
0
0
2
2
0
0
0
0
0
0
0
6


0
0
0
0
0
0
0
0
0
0
0
0
0
la
0.
lb
0
0
0
0
2



No.
BOM

of plants
50% - 50%



Coal/Nuclear/Syn-Fuel
2
0
2
2
2
0
0
. 0
2
2
0
0
0
0
2
0
0
0
0
0
16


0
2
0
2
0
2
2
1
0
0
2
2
2
0
0
0
0
0
0
0
15


0
0
0
0 •
0
0
0
0
0
0
0
0
0
la
°K
lb
0
0
0
0
2


No. of No. of
plants plants
FTF FTF 100%
100% Coal Nuclear



2







1








2 1


            II-B-1-22

-------
                              TABLE II-B-1-12
County

Athens
Belmont
Brown
Butler
Clark
Clermont
Franklin
Gallia
Hamilton
Lawrence
Mahoning
Meigs
Miami
Monroe
Montgomery
Morgan
Muskingam
Pickway
Pike
Ross
Scioto
Warren
Washington

  Totals
                  REGIONAL TECHNOLOGY CONFIGURATION SITINGS
                      AND PLANT CHARACTERISTICS FOR OHIO

                              Ohio Counties
No. of plants
BOM 80% - 20%
Coal/Nuclear/Syn-Fuel
3
0
3
3
2
3
2
3
2
3
3
3
2
0
2
3
0
2
0
3
0
3
3
0
1
1
0
0
0
0
1
0
1
. 0
1
0
1
0
1
1
0
1
1
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
No. of
No. of plants plants
BOM 50% - 50% FTF
Coal/Nuclear/Syn-Fuel 100% Coal
2
0
2
2
2
2
2
1
1
1
2
2
2
0
2
2
0
1
0
2
0
0
2
0
4
4
0
0
0
0
4
0
2
0
4
0
4
0
2
1
0
1
2.
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
I

2
1
1
1
1

1

1
1
1

1
1



1

1

No. of
plants
FTF 100%
Nuclear

1
1




1

1

1

1

1
1





1
 48    12
30    30
15
                             II-B-1-23

-------
                            TABLE II-B-1-13
               REGIONAL TECHNOLOGY CONFIGURATION SITINGS
                 AND PLANT CHARACTERISTICS FOR INDIANA

                            Indiana Counties



County
Clark
Crawford
Daviess
Dearborn
Dubois
Fountain
Gibson
Greene
Harrison
Jackson
Jefferson
Knox
Lawrence
Martin
Ohio
Perry
Pike
Posey
Spencer
Sullivan
Switzerland
Tippecanoe
Vermill ion
Warren
Warrick

No.
BOM

of plants
80% - 20%



Coal/Nuclear/Syn-Fuel
1
1
2
1
1
1
1
1
2
1
0
1
1
1
1
2
1
1
1
1
1
1
1
1
1
0
0
1
0
0
1
0
1
1
0
0
0
0
0
0
1
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0

No.
BOM

of plants
50% - 50%



Coal/Nuclear/Syn-Fuel
1
1
1
0
0
1
1
0
2
0
0
0
0
1
0
2
1
1
1
1
1
0
1
1
0
1
1
1
1
0
1
0
1
2
0
1
0
0
0
1
2
0
0
1
1
1
1
1
0
0
0
0
0
0
0
o •
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
No. of No. of
plants plants
FTF FTF 100%
100% Coal Nuclear

1
1


1


1




1

1

1





1

Totals
27
17    17
                           II-B-1-24

-------
just a  lot more of it.  The 50/50 mix scenario addresses the same dramatic
increase  in output but utilizes a greater proportion of nuclear energy,
in the  place of coal.  One would expect both possibilities to have a
dramatic  effect on the mining of coal in the region, but for different
reasons and with quite different results.

1.2.3.  FORD TECH FIX 100% COAL.  The two low growth scenarios address
the possibility of merely replacing those plants which wear out between
1985 and  2000 - and nothing more.  In one instance those plants will be
replaced  with 600 MWE coal-fired units and in the other with 1000 MWE
nuclear plants.  Both scenes assume that the plants on line in 1985 will
be capable of handling the region's needs in the year 2000.  The 100%
coal scenario is essentially the same as a moritorium on nuclear power
in the  region for the years 1985-2000.

1.2.4.  FORD TECH FIX 100% NUCLEAR.  Replacement of worn-out coal, oil
and gas-fired units with centrally-located 1000 MWE nuclear plants was
also considered a possibility for the low growth scenario.  This would
.tend to conserve fossil fuels and at the same time reduce sonra forms of
air and water pollution.

           The two extreme situations (all coal or all nuclear) for the
low growth scenarios were selected purposefully for the sake of compari-
son.  In  general, the Task 1 group felt that the different impacts of
coal and  nuclear conversion facilities would not become evident with a
50/50 mix involving a small number of plants.

1.3.  LOCATION OF EXISTING AND PLANNED FACILITIES

           The Energy Resources Center at the University of Illinois
Chicago Circle Campus has developed as part of the Task 1 effort a
tabulation of existing and planned conversion facilities for the ORBES
region.   These data, expressed in terms of generation capability,
are reproduced in the ORBES Task 1 report, "Development of Plausible
Future Regional Technology Configurations," October 18, 1976.

1.4.  COMPOSITION OF UNITED STATES COALS

           The composition of United States coals is as varied as their
geographic and geologic histories.  In the continental United States five
geologic  basins predominate as sources of steam coal.  These are the
Appalachian and Illinois Basins in the east and the Powder River, Denver
and Colorado Plateau associations in the west (Figure II-B-1-3).  Two-
thirds of the U. S.  coal reserves lie in the western basins which are
becoming  increasingly more important as sources of coal for the east and
midwest.  Estimates of the proportions of western to eastern coal burned
in Illinois, for example, are currently running 40 percent western - 60
percent eastern with the majority of the western coal being shipped to
the Chicago area.  Advantages in the mining of western coal seams, some
of which  are several hundred feet thick, are clearly beginning to outweigh
the cost  of transportation across the continent by unit train.
                             II-B-1-25

-------
 Pacific Coast Region
                                 Western Region
                                                                      Interior Region
                                                                                              Eastern Region
EXPLANATION
Fwag A-.thracite and
*""""' semianthracite
I^JS^ Low-volatile bituminous coal
p^pa Medium- and high-volatile
'"'•fr™ bituminous coal
lJ—Fr-1 Subbituminous coal
rVvt3g Lignite
       Figure II-B-1-3.   Coal  fields  of  the  United  States,
                                            II-B-1-26

-------
1.4.1.  FUEL VALUE

           Since the beginning of the twentieth century, coals have been
classified or ranked according to (1) fixed carbon, on a moisture-free
and mineral-matter-free basis, and (2) heating value, on a moist, mineral-
matter-free basis.  This classification is given in Table II-B-1-14 and
the distribution of coals according to rank is shown in Figure II-B-1-3.
Typical analyses for coals of various ranks are given in Table II-B-1-15.
Anthracite, subbituminous and bituminous coals predominate in the
eastern basins.   The subbituminous and bituminous coals are mined
principally as commercial or steam coal.  Anthracite, which is of
limited reserve, is mined as a coking coal and as a source of carb.on
for mixed media filters.  Fuel values for commerical eastern coals
range from 10,000 to 14,000 BTU per pound.

           Lignites and brown coals are common to the western basins
along with subbituminous and bituminous coals-.  Anthracite is rare in
the west if found at all.  Fuel values for lignite and low grade sub-
bituminous coals range from 7,000 to 9,000 BTU per pound.  Thuse coals
are mined only where the circumstances of mining thick seams with
shallow overburden offset the additional cost of handling and shipping
additional bulk."                       "  .

1.4.2.  NON-METALS

           The principal non-metals in coal (other than carbon, hydrogen
and oxygen) which have an "impact onthe environment upon burning are   .
nitrogen, sulfur  and the halogens.   The halogens enter the environment
as stable water-sojuble salts in both the stack gasses and residue.
Nitrogen and sulfur, on the other hand, leave the plant as gaseous
oxides^ which react readily with atmospheric ozone to give stable end
products; namely, nitric and sulfuric acid.

           In order to meet the new stationary source emission standards
for sulfur dioxide the following relationship between BTU content and
percent sulfur by weight of coal are required.

                           TABLE II-B-1-16

                   S02 EMISSION STANDARD REQUIREMENTS

            BTU  per pound                  Percent Total  Sulfur

               14,000                              0.84
               12,000                              0.72
               10,000                              0.60
                8,000                              0.48
                6,000                              0.36

For the range given in Table II-B-1-15 these data fit the equation:
           percent total sulfur = (6 x 10-5)(BTU per pound).
                             II-B-1-27

-------
                              TABLE  II-B-1-14

                    CLASSIFICATION OF COALS BY RANK
            Rank
Fixed carbon,
  per cent
(dry, mineral -
 matter-free)
Heating value,
BTU/lb (moist,
mineral-matter-
free)
Meta-anthracite
Anthracite
Semianthracite13
Low-volatile bituminous coal
Medium-volatilve bituminous
coal
High-volatile A bituminous
coal
High-volatile B bituminous
coal
High- volatile C bituminous
coalc
Subbituminous A coalc
Subbituminous B coal
Subbituminous C coal
Lignite
a - No ASTM limits.
b - Must be nonagglomerating.
bituminous coal. See note
98 or more
92 to 97.9
86 to 91.9
78 to 85.9

69 to 77.9

Less than 69

Less than 69

Less than 69
Less than 69
Less than 69
Less than 69
Less than 69

If agglomerating,
c for meaning of
a
a
a
a

a

14,000 or more

. 13,000 to 13,999.

11,000 to 12,999
11,000 to 12,999
9,500 to 10,999
8,300 to 9,499
Less than 8,300

classify as low-volatile
"agglomerating."
c - High-volatile C bituminous coal  must be either agglomerating or non-
    weathering, whereas Subbituminous A coal  is both weathering and non-
    agglomerating.   "Agglomerating"  is the tendency of a coal  in the volatile
    matter determination to produce  a button  that is either definitely co-
    herent or that  shows swelling or cell  structure (ASTM D388-38).  "Wea
    thering" is the tendency of a coal to  disintegrate or slack when alter-
    nately dried and wetted in the presence of air (ASTM D388-38).
    (Gas Engineering Handbook, 1969).
                            II-B-1-28

-------
                                        TABLE II-B-1-15


                     TYPICAL  ANALYSES  OF COALS  OF  VARIOUS  RANKSa

                                  ("as received"  basis)
Proximate analysis, per cent

State and
Rank county
Anthracite Pa., Schuyikill
Semianthracite Va., Montgomery
Low-volatile Ark., Sebastian
bituminous coal Pa., Somerset
W. Va., Raleigh
Medium-volatile Ala., Jefferson
bituminous coal Pa., Clearfield
W.Va., McDowell
High-volatile A Ala., Walker
bituminous coal Ky., Pike
Pa.,'Fayette
Pa., Washington
W. Va., Logan
W. Va., Marion
High-volatile B Ky., Hopkins
bituminous coal Utah, Carbon
High-volatile C Colo., Routt
. bituminous coal III., Franklin
Subbituminous Colo., Weld .
coal Wyo., Sheridan
Lignite N. D.. Mercer
Cannel coal Ky., Floyd

' ~ These analyses are based on data in
345. Utah; 347, Ala.; 416, Ark.; 484, Wyo.;

Mois-
t'ire
4.4
2.2
2.5
2.6
3.1
2.4
2.4
1.5
1.4
2.8
1.8
3.9
2.8
1.9
7.6
3.7
9.7
10.2 ,
23.7
23.8
36.3
1.8

Vola-
tile
mat-
ter
3.4
12.4
16.8
17.0
17.4
25.9
24.3
23.6
35.2
33.8
33.0
37.7
35.8
39.2
37.8
43.1
36.4
33.8
28.9
32.4
26.2
44.5

Fixed
car-
bon
83.1
67.4
72.8
73.7
74.8
66.8
65.6
69.5
56.3
58.8
-58.2
52.4
56.3
52.0
45.6
47.8
: v.48.5
m?
43.1
40.5
31.7
46.8

thousands of published coal
411 Wash.
• / Coals (Tech.
: 641,'
III.; 652,
Ash-
soften-
ing
temp.
°F
2790
2800
2070
2740
2830
2770
2520
2360
2440
2830
2660
2380
2880
2130
2160
2180
2?50
2080
2060-
2280
2450
2760

papers
Ky .; 656, Va .; 659,
Penna. anthracite; 671, Tenn.; 700, Mich., No. Dak., So. Dak.. Tex.; and others) Washington, 1925-48.  Reference should also be made
to Fieldner, A. C. and others, Typical Analyses of Coals of the Uniterl States (Bur. of Mines Bull. 446) Washington, 1942.
                                     II-B-1-29

-------
           Coal-fired steam utility generators rank at the top of the
two dirtiest two dozen industrial operations based on NO  emissions (7 ).
Collectively, utility fossil fuel conversion systems produce 48% of the
stationary emission source NOX in the U. S.  with coal being the major
producer at 30.8%.  These conversion systems produce 24% of the total NOX
emissions, automotive sources produce 50%, and the remainder is produced
by other industrial stationary sources.  Nitrogen in the coal, which
averages about 1.5% by weight, is not the principal source of NOX emissions
from coal-fired plants although it does contribute to the total.  The
combustion of coal, oil and gas carried out in air produces NOX compounds
as a result of reactions between atmospheric nitrogen and oxygen at high
temperatures in much the same wasy as NO  compounds are produced by
internal combustion engines.

           Current-federal regulations set forth for stationary NOX
sources are outlined in Table II-B-1-17.

1.4.3.   METALS

           The distribution of trace elements in American coals by
geologic basin is given in Tables II-B-1-18 and II-B-1-19 (10).  These
data, like the data on uraniferous coals which follows, will  serve the
purpose of this study, but by no means should the reader consider the
tabulation to be complete.  The current methodologies for sampling and
analysis of trace elements 'in American coals has improved greatly in the
past few years, but the data still apply only in a very general sense.
Tremendous variations are observed in trace element content in going
from point to point in the same seam, much less in going from seam to
seam or basin to basin.  A good example of this phenomenon is observed
in the data by Gluskoter, et. al_.  (8 ) for 25 samples of Herrin (No.  6)
shown in Table II-B-1-20.

           The question of what is the fate of a given trace element
following combustion is addressed in the impact analysis which follows
Chapter 2.                   '            .

T.4.4.   RADIONUCLIDES IN UNITED STATES COALS

           The chemical affinity of heavy metals for carbonaceous material
has resulted, after many millions of years of leaching from surface ores,
in the deposition of significant amounts of uranium in the coal seams
below.   This phenomenon occurs wherever uranium-bearing strata occur and
                             II-B-1-30

-------
                               TABLE  II-B-1-17

        CURRENT- FEDERAL REGULATIONS FOR STATIONARY SOURCES OF NOV
        Combustion Sources
           Status
 1. Coal-fired steam boilers,
      >250 x 106 Btu/h

 2. Oil-fired steam boilers,
      >250 x 106 Btu/h.

 3. Gas-fired steam boilers,
      >250 x 106 Btu/h.

 4. Lignite-fired stea'Ti boilers,
      >250 x 106 Btu/h.

 5. Coal-fired steam boilers,
      >250 x 106 Btu/h (revision
      of 1971 standard)
 6. Coal-, oil-, gas-fired steam:
      boilers, 0.3 x 10^ to
      250 x 106 Btu.

 7. Stationary internal-combustion
      engines (diesel and gasoline),

 8. Stationary gas turbines
Promulgated Dec. 1971. Sets 0.7 lb/
  106-Btu limit.

Promulgated Dec. 1971. Sets limits at
  0.3 lb/106 Btu.

Promulgated Dec. 1971. Limit is 0.2 lb/
  106 Btu.'

Proposal issued in Dec. 1976 to set
  standard at 0.6 lb/106 Btu.

Technical support document completed;
  regulatory package being prepared.
  A 0.6-lb/!06-Btu standard should be
  written this summer.

Technical study underway. Proposals
  .should issue in about one year.
Proposal being prepared but termed "a
  long way off".

Technical document essentially complete;
  regulation now being written. Proposal
  due this July.
        Process Sources

 9.  Nitric acid plants
10.  Adipic acid manufacture

11.  Dimethyl  terephthalate/
      terephthalic acid plants

12.  Explosives manufacture

13.  Glass fibers for textile
      operations.
Promulgated Dec. 1971. Sets limit at 3 lb
  N02/ton of 100% acid.

Screening study completed.

Screening study initiated to document need
  for new-source performance standard.

Screening study initiated.

Screening study initiated.
 Sources: "NOX Emission Trends and Federal  Regulation," a paper delivered in
 November, 1976 at the AIChE's 69th Annual  Meeting held in Chicago, IL, and
 CE interviews with OAQPS personnel.  Reported in Chemical Engineering,
 February 14, 1977.
                             II-B-1-31

-------
                             TABLE II-B-1-18


                 TRACE-ELEMENT CONTENT OF AMERICAN COALS

                           (Parts per million)
Element
Be
B
Ti
V
Cr
Co
Ni
Cu
Zn
Ga
Ge
Mo
Sn
Y
La
Northern
Great Plains
1.5
116
591
16
'7
2.7
7.2
15 '
59
5.5
1.6
1.7
.9
13
9.5
Western interior
region
1.1
33
250
18
13
4.6
14
11
108
2.0
5.9
3.1
1.3
7.4
6.5
Eastern interior
reqion
2.5
96
450
35
20
3.8
15
11
44
4.1
13
4.3
1.5
7.7
5.1
Appalachian
reqion
2.5
25
340
21
13
5.1
14
15
7.6
4.9
5.8
3.5
.4
14
9.4
SOURCE:  Gluskoter,  H.J.,  Cahil,  R.A.,  Miller,W.G.,  Ruch,  R.R.,  and
         Shimp, N.F.,  "An  Investigation of Trace Elements  in  Coal,"
         Symposium Proceedings:  Environmental Aspects  of Fuel  Conversion
         Technology, II.   EPA-600/02-76-149 (June,  1976).
                            II-B-1-32

-------
                             TABLE II-B-1-19
                DISTRIBUTION OF ENVIRONMENTALLY HAZARDOUS
                              TRACE ELEMENTS

                        (Parts per million in coal)
El ement
Sb
As
Be
Cd
Hg
Pb
Se
Zn
Great Plains
0.67
3
.7
2.1
.1
7.2
.73
33
Western interior
region
3.5
16
2
20
.13
—
5.7
— ••
Eastern interior
region
1.3
14
1.8
2.3
.19
34
2.5
250
Appalachian
region
1.2
18
2.0
.2
.16
12
5.1
13
SOURCE:  Same as Table II-B-1-18.
                            II-B-1-33

-------
                        TABLE II-B-1-20
SUMMARY OF ADDITIONAL ANALYTICAL VALUES FOR 25 SAMPLES OF HERRIN
           (NO. 6) COAL FROM THE ILLINOIS BASIN (ppm)
Element
Ag
Au
Ba
Ce
Cs
Dy
Eu
Hf
I
In
La
Lu
Rb
Sc
Srn
Sr
Ta
Tb
Th
U
W
Yb
Arithmetic
mean
0.19
0.01
130.
11.
1.0
0.99
0.23
0.45
2.0
0.02
.6.9
0.07
14.
2.4
1.1
37.
0.15
0.15
2.01
1.6
0.74
0.51
Standard
deviation
0.24
0.01
150.
4.3
0.26
0.30
0.07
0.14
1.2
0.02
2. .2
0.02
4.0
0.57
0.62
20.
0.05
0.06
0.47
1.1
0.56
0.13
Range
Minimum
0.03
0.0004
33.
4.4
0.49
0.70
0.10
0.24
1.0
0.008
3.3
0.041
7.4
1.4
0.4
19. .
0.10
0.04
1.2
0.5
0.04
0.31
of values
Maximum
0.80
0.032
750.
24.
1.5
1.81
0.40
0.81
5.8
0.09
12.
0.11
20.
3.6
3.8
130.
0.30
0.24
3.3
4.5
2.1
0.77
SOURCE:  Zubovic, P.,"Minor Element Content of Coal  from Illinois
           Beds 5 and 6 and their Correlatives in Indiana and
           Western Kentucky," U.  S. Geological Survey, Open File
           Report, p. 79, 1960.
                       II-B-1-34

-------
thus is common to the coal beds in the American West, particularly in the
Paleocene Fort Union formation in the Dakotas and Montana (.9).  Ore
grade uraniferous lignite containing 0.1% U-^Og or more occurs in this
formation which is the principal coal producing unit of the Dakotas.
Ore grade material, however, is generally confined to thin coal seams
near the surface.

           The topic of coal-uranium geochemistry was discussed at the
International Atomic Energy Agency symposium "Formation of Uranium Ore
Deposits," held in Vienna in 1974 (10).  The data presented at that
symposium shows a wide range of uranium concnjetration in western coals
going from 0.001 to 16.5% by weight.  The reader is cuationed against
assuming that this data is representative of all western coals.  Urani-
ferous coal deposits are highly localized and western coals are generally
considered to be free of radionuclides (11).  Eastern coals appear to
contain negligible radioactivity, although trace amounts of urandium are
present in virtually all subsurface geologic deposits having significant
carbon content.  Herrin or No. 6 Illinois Basin coal, for example, con-
tains an average of 0.0001% uranium 238 (12).

           The reader is also cuationed against believing that uraniferous
coals contain only uranium.   The scheme for the uranium-radium family
shown in Table II-B-1-21 gives the decay daughters for the uranium 238 to
lead 206 sequence.  This sequence lists a total of 15 decay products,
some of which are absolutely deadly to ingest.   The product which is
considered most dangerous in £§don__222. which is an unstable, dense,
radioactive gas having ;a .halflife of 3.82 days.  A person breathing this
material will trap the dense radon molecules in the alveoli of the lungs
from which it will likely not be exhaled.  The  molecule will subsequently
decuy down through the rest of the uranium-radium sequence leaving the
person with an imbedded molecule of lead, which is toxic enough by itself,
and certain damage due to radiation as well.
                             II-B-1-35

-------
                                                        TABLE II-B-1-21
                                            THORIUM-232 AND URANIUM-239 DECAY SERIES
                     Thorium-232 decay series
Uranium-238 decay series9
Isotope
Thorium-232
Radium- 228
Actinium-228
~ Thorium-228
CO J,
^ Radium- 224
w |.
Half-life
1.41 x lO^Oyears
6.7 years
6.13 hrs
1.910 years

3.64 days

Radon-220 (Thoron) 55.3 sec
Polonium- 21 6
I
Lead-212
Bismuth-212 — i
-Polonium-212
64% Thallium-208f-
} Lead-208





.0.145 sec

10.64 hrs
60.60 min
36% 3.04 x 10'7sec
3.10 min
Stable



•

Principal
emission
a
•
B
6
a

a

a
a

8
a (Thallium)
8 (Polonium)
a
8






Isotope
Uranium-238
Thorium-234
Protacinium-234 .
Uranium-234
v
Thorium-230
J'
Radium-226
Radon-222
•I
Polonium-218
Lead-214
Bismuth-214
Polonium-214
Lead-210
i
Bismuth-210
Polonium-210
Lead-206
a
Half -life
g
4.51 x 10 years
24.10 days
6.75 hrs
2.47 x 105years

7.5 x 10 years

1602 years
3.82 days

3.05 min
26.8 min
19.9 min
1.64 x 10'4sec
22 years

5. 01 3. days
138.40 days
Stable

Principal
emission
a
6
8
a

a

a
a

a
8
6
a
8

8
a

C J. l» —. J.AM.*
                                                                  are omitted.
SOURCE:  Remote Sensing for Environmental Sciences, ed. by Erwin Schanda, Springer-Verlag, 1976.

-------
1.5.  OHIO RIVER SOCIOLOGICAL STATUS

        Necessary to the Ohio River Basin Energy Study (ORBES) has been
the collection and analysis of data describing the current sociological
status of the Basin, each state in the Basin, as well as each county in
which a power conversion facility might be located.  In this chapter the
results of this effort as applied to current employment by broad industrial
sector are presented.

        This chapter is divided into two major sections.  In the first
section, employment statistics for each state in the ORBES region will
be analyzed.  The second section will present employment data for each
of the target counties within the region.

        Before analyzing the state-wide data the reader should be made
aware of several conventions followed in this chapter.

        The first point is that in analyzing employment data for the states
of Ohio, Indiana, and Illinois we have used employment statistics for the
entire state, rather than aggregating together just those counties comprising
the ORBES region.  The rationale underlying this choice is based on the
consideration that while each county in a state may be described as having
a distinct labor pool, this thinking is actually an artifact of our data
collection procedures.  In reality workers are quite mobile, and depending
on the location of their residence may commute across both county and state
lines to their place of employment.   Given this fluidity in the boundaries
of labor -markets and pools, 'it seemed unwise to limit this analysis by ex-
cluding the northernmost counties of Indiana, Ohio, and Illinois, since
workers in these areas would be involved in any major labor-intensive con-
struction projects in these states.

        The second point is that in describing employment we have chosen
to show distributions by broad industrial sector, although data are available
which permit the construction of distributions by occupational category.
Generally the industrial distributions are preferred since they give the
reader a better perspective on the types of outputs being produced.   Since
the industrial categories used here are rather lengthy, the following
abbreviations have been used in the tables in this chapter.

        Agriculture, forestry, and fisheries is abbreviated as A.F.F.;
mining is not abbreviated; construction is Const.; manufacturing is Manuf.;
transportation, communication, and other public utilities is abbreviated
as T.C.PU.; wholesale and retail trade as W.R.; finance, insurance, and
real estate is abbreviated as F.I.RE.; business and repair services as B.RS.;
personal services as PS.; entertainment and recreational services as E.RS.;
professional  and related services as P.RS.; and public  administration as PA.

        In Table II-B-1-22 location quotients are listed for each state
and industrial sector.   The location quotient is defined as:
                             II-B-1-37

-------
where the subscript i denotes an industrial sector, s denotes a state,
and n denotes the nation.  The location quotient is thus the ratio
of an industry's share of the economic activity of the economy being
studied (in this case, a state economy) to that industry's share of
another economy (the nation).  If the resultant value is greater than
one it is assumed that the state exports the products of that industry.

        Data in Table II-B-1-22 indicate several important differences
in the economies of the four states comprising the ORBES region.  Ohio's
labor force, for example, is primarily concentrated in manufacturing,-
as is Indiana's.  Illinois, however, has location quotients above unity
in five industrial sectors, including manufacturing; transportation,
communication and public utilities; wholesale and retail trade; finance,
insurance, and real  estate; and business and repair services.  Illinois
is the only one of the four states that exports in the last three of
these industrial sectors.  Kentucky, as well as Illinois, is specialized
in five industrial sectors, although the degree of overlap is minimal.
Kentucky is the only one of the four states with high industrial con-
centrations in agriculture, forestry, and fisheries; mining; construction;
and personal services.  The very high location quotients in these first
two industrial, sectors is noteworthy.

        The four ORBES states also differ in respect to the size of their
employed labor forces.  Kentucky has the smallest labor force at just over
one million workers  (1,088,758), followed by Indiana at just over two
million .workers (2.016.,365).   Ohio has slightly over four million workers
(4,063,780), while Illinois has the largest labor force at about four
and one-half million (4,419,915).

        In Table II-B-1-23 industrial distributions (in percentage form)
are listed for Kentucky, as well as for each target county in the state.

        One important fact to note about the Kentucky target counties
is the relatively small  size of their labor forces.  The 21 target counties
have an average labor force size of only 4,350 workers, with the smallest
labor force (Gallatin County) at only 1,254,-and the largest (Henderson
County) at 13,097.

        In Table II-B-1-24 industrial distributions are presented for the
state of Indiana and its 25 target counties.  The average of the labor
forces in.these counties is 10,229, about 2.5 times greater than in Kentucky.
The Indiana target counties also have a much greater size range, varying
between 1,525 workers in Ohio County to 44,854 workers in Tippecanoe County.

        Table II-B-1-25 lists the  industrial distributions in Ohio and its
23 target counties.   These counties have an average labor force size of
64,600 - almost 15 times the average for Kentucky and over 6 times the
average in Indiana.   The labor force's in the Ohio target counties range
in size from 4,591 (Monroe County) to 353,757 (Hamilton County).

        Data for Illinois are found in Table II-B-1-26.   Its 20 target
counties have an average labor force size of 10,379, only slightly greater
than in Indiana, and well below the average in Ohio.   Labor force sizes
                             II-B-1-38

-------
INDUSTRY
                           Table  n-B-l-22
           LOCATION  QUOTIENTS BY INDUSTRIAL SECTORS,
                         ORBES STATES
OHIO
ILLINOIS
INDIANA
KENTUCKY
A.F.F.
Mining
Const.
Manuf .
T.C.PU.
W.R.
F . I . RE .
B.RS.
PS.
E.RS.
P.RS.
PA.
0.55
0.63
0.84
1.37
0.91
0.96
0.79
0.83
0.81
0.83
0.91
0.77
0.72
0.58
0.85
1.17
1.10
1.01
1.06
1.04
0.79
0.79
0.94
0.80
0.91
0.44
0.89
1.38
0.87
0.95
0.82
0.67
0.83
0.65
0.89
0.68
1.82
3.09
1.18
0.99
1.02
0.93
0.73
0.74
1.06
0.82
0.92
0.86
                          II-B-1-39

-------
               Table  H-B-1-23
INDUSTRIAL DISTRIBUTIONS IN KENTUCKY
       AND ORBES TARGET COUNTIES
INDUSTRY
A.F.F.
Mining
Const.
Hanuf .
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
INDUSTRY
A.F.F.
Mining
Const.
Manuf .
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
KENTUCKY
6.8%
2.5
7.0
25.6
6.9
18.7
3.7
2.3
4.9
0.7
16.2
4.7
1,088,758
GREENUP
2.2
0.7
5.7
29.9
24.5
17.8
2.4
1.4
3.1
0.4
9.3
2.5
9,866
BALLARD
16.2
0.1
15.9
24.2
6.7
13.7
1.3
1.6
4.7
1.0
11.4
3.1
.2,686
HENDERSON
4.9
2.8
6.4
33.0
7.4
18.8
3.1
2.1
5.2
0.5
12.7
3.1
13,097
BRACKEN
27.2
2.1
6.6
27.1
4.3
17.1
2.1
2.2
2.6
0.7
9.2
3.1
2,428
LEWIS
16.2
0.1
8.0
35.6
7.8
13.1
2.2
1.3
2.2
0.3
11.5
1.7
3,750
BRECKINRIDGE
23.4
0.6
8.3
21.2
5.5
14.3
1.4
1.4
4.6
0.1
11.0
8.1
4,843
LIVINGSTON
9.3
4.7
11.2
26.7
11.6
14.5
1.8
1.3
5.4
0.2
8.5
4.8
2,426
BUTLER
14.1-
3.6
10.1
35.3
6.2
11.4
0.9
2.3
3.4
0.0
9.5
3.2
2,948
MCLEAN
15.4
4.5
9.9
28.1
5.9
15.1
2.4
2.0
3.8
0.0
10.6
2.4
3,077
CARLISLE
13.0
0.3
10.3
3 3'. 9
7.0
16.1
1.2
1.3
5.1
0.0
8.4
3.5
1,797
MARSHALL
3.9
0.2
12.2
36.5
6.0
15.6
3.1
2\9
4.8
1.0
9.4
4.5
7,143
GALLATIN
15.0
0.0
10.2
25.6
6.5
20.8
5.3
0.6
3.6
0.0
8.7
2.8
1,254
MASON
15.6
0.1
5.2
27.2
5.2
19.7
3.2
1.9
5.8
0.4
12.8
2.8
6,249
            II-B-1-40

-------
Table  H-B-1-23 (continued)
INDUSTRY
A.F.F.
Mining
Const.
Manuf .
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
INDUSTRY
A.F.F.
Mining
Const.
Manuf .
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
MEADE OWEN RUSSELL
7.5% 30.7 17.1
1.3 0.3 0.0
7.8 7.9 11.0
22.8 15.0 25.3
8.6 5.3 4.0
16.9 13.0 15.0
1.8 2.0 2.0
2.3 1.3 3.4
4.8 5.7 4.3
0.6 0.2 0.1
14.6 9.3 12.3
11.2 9.3 5.7
3,999 2,531 2,852
WEBSTER
11.1
9.1
8.7
24.4
3.9
18.8
1.8
1.4
4.4
0.2
11.5
4.7
3,866
SCOTT TRIGG TRIMBLE UNION
16.8 19.7 22.3 12.5
0.1 1.7 0.8 11.7
5.4 10.2 6.7 8.6
25.6 22.8 28.8 13.5
5.0 5.1 4.8 4.3
13.8 14.6 12.3 16.4
3.0 2.1 2.2 2.3
2.5 0.6 2.2 2.2
5.9 6.0 3.1 5.8
0.2 3.6 0.3 0.2
17.4 10.9 9.4 18.6
9.3 5.9 7.2 3.9
6,901 3,060 1,826 4,813














                                        II-B-1-41

-------
          Table  H-B-1-24
   INDUSTRIAL DISTRIBUTIONS IN
INDIANA AND ORBES TARGET COUNTIES
INDUSTRY
A.F.F.
Mining
Const.
Manuf .
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
INDUSTRY
A.F.F.
Mining
Const.
Manuf.
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
INDIANA
3.4%
0.4
5.3
35.9
5.9
19.1
4.1
2.1
3.9
0.5
15.7
3.7
2,016,365
GIBSON
7.8
3.7
4.9
33.6
5.5
17.7
1.9
1.4
4.7
0.3
15.8
2.7
11,109
CLARK
2.2
0.3
4.7
41.0
6.8
18.2
4.2
1.7
3.4
0.4
12.0
5.0
30,757
GREEN
6.4
4.3
8.4
22.0
6.4
15.2
2.8
1.5
4.2
0.2
13.1
15.3
9,677
CRAWFORD
8.7
2.4
.11.7
35.4
3.9
13.5
2.7
1.1 , ..
3.7
0.0
10.0
6.8
2,594
HARRISON
9.7
1.1
6.5
35.1
6.5
17.9
3.5
1.2
4.3
0.2
9.6
4.4
7,189
DAVIESS
11.8
0.2
6.3
23.0
7.6
16.1
2.5
.1.6
4.8
0.6
13.3
12.0
9,750 .
JACKSON
5.8
0.2
5.5
45.0
4.7
17.3
2.0
1.3
4.1
0.4
10.9
2.8
12,923
DEARBORN
3.3
0.1
7.2
42.9
9.2
15.2
2.8
1.4
2.8
0.4
11.7
2.8
10,333
JEFFERSON
4.4
0.2
3.7
31.2
6.8
18.9
2.1
1.2
3.4
0.5
19.4
8.1
9,902
BUBOIS
7.9
0.7
6.4
41.6
4.5
15.6
2.2
1.6
4.4
0.5
11.9
2.6
11,653
KNOX
10.0
1.0
7.8
17.6
6.0
25.5
2.4
2.3
4.9
0.5
17.5
4.4
15,364
FOUNTAIN
9.1
0.4
6.2
40.4
4.9
15.7
.2.6
1.2
5.8 .
0.4
10.9 .
2.3
6,890
LAWRENCE
3.0
1.3
5.9
37.2
5.1
15.8
2.8
1.5
4.4
0.4
10.1
12.4
14,156
         II-B-1-42

-------
            Table  II-B-1-25
  INDUSTRIAL DISTRIBUTIONS  IN
OHIO AND ORBES TARGET COUNTIES
INDUSTRY
A.F.F.
Mining
Const.
Manuf .
T.C.PU.
W.R. .
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
INDUSTRY
A.F.F.
Mining
Const.
Manuf.
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS
PA.
TOTAL
OHIO
2.1%
0.5
5.0
35.6
6.2
19.2
4.0
2.6
3.7
0.7
16.1
4.2
4.063,780
FRANKLIN
0.8
0.2
5.5
22.9
6.6
21.5
6.9
3.0
4.0
0.7
21.4
6.5
336,132
ATHENS
2.2
0.8
6.4
13.2
6.5
18.9
2.1
1.6
4.5
0.7
40.1
3.0
18,295
GALLIA
7.0
1.1
8.0
15.0
8.1
17.7
2.3
1.8
5.8
0.5
30.1
2.8
7,546
BELMONT
2.9
11.4
5.3
28.2
7.8
19.6
3.1
1.5
3.5
0.7
13.7
2.2
28,159
HAMILTON
0.5
0.1
4.6
32.1
6.1
20.5
5.1
3.2.
4.4
0.8
18.1
4.6
353,757
BROWN
11.2
0.1
7.9
34.1
4.1
16.0
2.7
2.4
4.0
0.3
13.5
3.7
8,847
LAWRENCE
2.1
1.2
6.4
35.8
10.0
19.5
3.3
1.9
4.1
0.6
11.8
3.3
17,593
BULTER
1.4
0.2
5.4
41.3
4.6
16. ••
4.0
2.1
3.7
0.6
17.1
2.6
83,800
HAHONING
1.0
0.2
5.3
37.9
5.7
21.3
3.5
2.2
3.6
0.7
15.4
3.1
111,150
CLARK
2.3
0.2
4.4
35.7
4.9
16.7
3.7
2.4
4.5
0.6
16.1
8.6
58,603
KEIGS
6.9
4.1
11.9
18.2
11.7
22.0
2.1
0.8
4.3
0.4
14.1
3.6
5,710
CLERMONT
2.1
0.2
7.1
39.6
6.8
19.6
3.3
2.9
3.2
0.7
11.6
2.8
34,769
MIAMI
2.8
0.6
4.8
45.0
4.8
17.9
3.0
1.8
3.5
0.5
11.4
3.8
33,300
          II-B-1-44

-------
Table  II-B-1-25  (continued)
INDUSTRY •'
k.F.F.
Mining
Const.
Manuf .
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
INDUSTRY
A.F.F.
'Mining
Const.
Manuf .
T.C.PU.
W.R.
F.I. RE.
B;RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
'• MONROE
7.7
6.1
8.3
36.5
5.9
13.2
1.7
1.5
4.0
0.3
11.2
3.6
4,591
SCIOTO
2.3
0.4
7.2
29.6
10.3
21.7
2.7
1.5
3.8
0.4
15.9
4.1
23,112
MONTGOMERY
0.7
0.1
4.1
38.0
4.5
18.6
3.5
3.2
3.8
0.6
15.6
7.3
239,831
WARREN
2.9
0.2
6.1
46.8
4.9
16.4
2.3
2.1
3.7
C 6
11.1
2.9
30,983
MORGAN
7.8
6.6
8.4
28.7
8.5
15.9
2.3
0.6
3.0
0.5
11.9
5.9
3,746
WASHINGTON
3.0
1.3
9.2
29.6
6.0
20.6
2.7
1.6
5.4
0.5
16.4
3.7
19,643
MUSKINGAN PICKWAY PIKE ROSS
2.6 8.1 5.3 4.8
2.0 0.3 0.5 0.4
5.0 7.8 9.6 6.4
33.3 30.0 28.9 32.9
7.5 6.0 6.7 5.4
21.5 18.3 18.5 18.8
2.8 3.8 2.6 1.9
1.8 2.1 1.8 1.7
4.4 4.0 5.5 4.7
0.6 0.6 0.2 0.7
15.6 14.9 16.3 18.0
2.9 4.2 4.2 4.5
• 27,940 13,269 4,879 20,211














                                      II-B-1-45

-------
            Table II-B-1-26
   INDUSTRIAL DISTRIBUTIONS IN
ILLINOIS AND ORBES TARGET COUNTIES
INDUSTRY
A.F.F.
Mining
Const.
Manuf .
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
..TOTAL
INDUSTRY
A.F.F.
Mining
Const.
Manuf.
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS.
PA.
TOTAL
ILLINOIS
2.7
0.5
5.1
30.3
7.-1
20.2
5.3
3.3
3.7
0.7
16.5
4.4
4, ,.419, 915
HENDERSON
20.1
0.3
5.2
27.3
5.0
19.3
3.5
1.2
3.6
0.0
10.5
4.1
3,209
BROWN
25.3
0.7
9.3
12.6
5.1
21.7
3.7
1.0
5.7
0.5
11.0
3.5
2,021
IROQUOIS
17.3
0.1
5.7
22.2
5.9
22.6
2.9
1.8
4.3
0.1
13.9
3.1
12,438
CASS
12.5
0.2
6.7
20.7
9.5
19.0
3.7
1.9
4.6 .
0.4
15.0
5.9
5,313
JERSEY
10.6
0.4
5.6
28.6
4.5
16.5
2.6
1.1
5.1
0.6
20.9
3.6
6,448
CLARK
12.3
2.4
7.6
19.2
8.4
21.6
3.1
2.5
4.1
0.5
15.5
2.8
5,725
LAWRENCE
8.8
7.3
7.8
18.6
4.2
19.0
4.5
1.4
4.9
0.8
18.7
3.9
6,144
GREEN
20.3
0.3
5.9
24.0
6.0
18.0
2.3
1.3
3.5
0.1 •
15.0
3.1
5,959
LIVINGSTON
12.8
0.8
6.1
26.3
5.2
18.6
2.9
2.0
3.5
0.6
16.6
4.6
15,088
HAMILTON
18.0
5.1
5.9
19.5
3.9
14.8
2.7
2.1
4.6
0.2
18.0
5.2
2,865
MARSHALL
16.0
0.6
4.8
30.2
4.3
21.1
3.1
1.4
3.6
0.3
12.1
2.4
4,923
HANCOCK
16.4
0.6
6.3
21.7
5.1
20.9
2.1
2.0
5.0
0.3
16.8
2.7
8,723
MERCER
17.7
0.4
6.2
22.4
5.5
18.1
' 3.7
1.8
4.4
0.1
15.7
3.9
6.3G4-
           II-B-1-46

-------
Table  II-B-1-26 (continued)
INDUSTRY     PERRY     PULASKI    ST.  CLAIR     SCHUYLER     SCOTT    WASHINGTON    WHITE
A.F.F.
Mining
Const.
Manuf .
T.C.PU.
W.R.
F.I. RE.
B.RS.
PS.
E.RS.
P.RS
PA.
TOTAL
5.4
10.8
6.0
28.6
6.0
15.9
2.7
0.8
3.9
0.6
15.9
3.2
6,978
12.1
2.7
5.3
17.4
8.9
17.2
2.7
3.8
6.3
0.2
19.0
4.4
2,563
1.8
0.7
5.2
24.1
11.4
20.4
5.3
2.9
4.1
0.8
16.1
7.2
96,020
22.3
0.7
7.9
14.7
4.5
20.8
2.8
2.0
4.5
0.6
15.8
3.6
3,069
21.1
0.8
4.5
14.3
10.0
20.7
2.9
0.9
4.1
0.4
16.8
3.6
2,34*4
23.2
1.4
4.7
17.5
5.9
17.5
2.0
1.5
4.0
0.3
17.6
4.3
5,092
12.4
7.8
6.1
15.6
5.4
22.1
2.5
2.0
6.1
0.3
16.1
3.7
6,303
                                       II-B-1-47

-------
range between a low of 2,021 in Brown County to a high of 96,020 in
St. Clair County.

        While this review has been quite superficial,  the data  presented
in Tables II-B-1-22 through II-B-1-26 strongly indicate two important
facts.  First, that the labor forces in the four ORBES states differ sig-
nificantly in both size and areas  of industrial  specialization;  and second,
that the target counties selected  for the possible construction  of energy
conversion facilities also differ  in many important respects, including
both size and industrial concentration.
                            II-B-1-48

-------
1.6.  PHYSICAL AND BIOLOGICAL DESCRIPTION OF THE OHIO RIVER BASIN
1.6.1.  GEOGRAPHIC

          The point usually designated as river mile zero on Ohio River main-
stream navigation charts is at the confluence of the Monongahela and
Allegheny Rivers at Pittsburgh, Pennsylvania.   The ORBES study area as
defined in Task I (ORBES 10-18-76) contains areas that are actually in
other river basins, such as the southwestern areas of Kentucky known as
the Jackson Purchase.  Other areas such as West Virginia are deleted from
consideration in the present study.  The Ohio River as defined by the
United States Water Resources Council, drains 203,910 square miles along
its 981 mile length.  The major tributaries in Kentucky are the Big Sandy,
Tennessee, Green and Kentucky Rivers.  In Ohio, the Muskingum, Scioto and
Miami Rivers end at their confluence with the Ohio River.  The White and
Wabash Rivers flow together and form the Indiana and Illinois state line
before joining the Ohio.  The Illinois River although a tributary of the
Mississippi River is included in the ORBES study area.  The Ohio River
joins the Mississippi River at Cairo, Illinois, 981 miles from its origin
at Pittsburgh, Pennsylvania.

          The Ohio River drainage system is almost completely devoice of
natural lakes and swamps.  Only the headwaters of the upper Wabash include
some natural lakes of local importance (13).  Water conservation reservoirs
vary in size from farm ponds to huge multi-state impoundments such as Lake
Bark-ley antbLxHce Kentucky.  'The use of-the Ohio'River for large traffic has
necessitated the construction of navigation dams which dominate the flow
regime of the main stem.  Impoundments, whether built for navigation,
conservation or flood control, have modified the natural assimilative
capacities of the impounded streams.

          The extraction of fossil fuels has given rise to both surface and
groundwater problems throughout the basin.  Coal  mining, especially orphaned
strip and deep mines, have caused acid mine drainage and sedimentation in
local streams.  Oil wells cause problems through brine disposal which most
often affect groundwater but may eventually contaminate surface streams also.

1.6.2.  TOPOGRAPHIC

          The Ohio River Basin is divided by a geologic structure known as
the Cincinnati Arch into two major physiographic regions - Appalachia and
the Eastern Interior or Illinois Basin.  The Appalachian Region is consider-
ably more mountainous than the Eastern Interior Region and soils are more
shallow except in intermontane valleys.

          The Appalachian Region is made up of an elongated structural  basin,
or trough, which is filled with a thick assortment of sedimentary rock units.
The eastern edge of the basin is bounded by the Piedmont and Blue Ridge
physiographic provinces and extends westward to central Tennessee, Kentucky
                             II-B-1-49

-------
and western Ohio.  The basin terminates to the south in Alabama, where it is
overlapped by sediments of the Coastal Plain Province, and extends northward
into New York and Canada (14).  The Coastal Plain Province includes the
extreme southwestern tip of Kentucky at the confluence of the Ohio and
Mississippi Rivers.

          The Eastern Interior Region is a large structural basin that
comprises central and southern Illinois, southwestern Indiana and western
Kentucky.  This regipn is bounded by the Central Lowlands Province on the
north and the Gulf Coastal  Plain on the south.

          Both regions are underlain by bedrock principally of sedimentary
origin varying from limestone .to sandstone, siltstone and shale.  The area
north of the Ohio River contains buried river valleys that were caused
largely by drainage from glaciation and often buried by subsequent
glaciations.

1.6.3.  GEOLOGIC

          Most of the strata in the Appalachian Region were -deposited on the
floors of the Paleozic seas and were repeadtedly squeezed and wrinkled into
long, narrow folds parolled to the Atlantic coast.   The largest remaining
deposits are 35,000 to 40,000 feet thick near the Atlantic coast and the
strata become progressively thinner and less deformed to the west.

          Generally, the Appalachian stratigraphy may be described as
follows: 'IVecambrian'metamorphic and -igneous rocks wi.th minor sedimentary
rocks.  Resting on the Precambrian is the Cambrian limestones, dolomites,
shale, with some sandstones at the base thickening southward.  Ordovician
strata are mined today for their shale, limestone and marble.  The remainder
of the Paleozoic era is dominated by elastics such as shale, sandstone,
siltstone and by limestones.   The Triassic was predominately shale followed
by Cretaceous and Quaternary unconsolidated deposits of sand, gravel, clay
and silt irregularly sedded (15).

          Five groups of formations comprise the Permian and Pennsylvanian
bituminous coal.  They are, in descending order, the Dunkard Group, the
Monongahela Formation, the Conemaugh Formation, the Allegheny Formation and
the Pottsville Group.  The Monongahela Formation consists mostly of red and
gray shale and several important coal seams, including the Pittsburgh, Red-
stone and Sewickly seams.  The lower Freeport coals, the Clarion coal and
the Kittaning coals contain a large part of the Appalachian reserves and
are included in the Allegheny Formation (16).

          Seams average four feet in thickness with average interval be-
tween seams of 80 to 100 feet. .  At least 90 minable bituminous seams have
been identified in the Appalachian region (17).  Estimates of reserves in
the northern part of the Appalachian Region, and primarily the major coal-
bearing states of West Virginia, Ohio, Pennsylvania and eastern Kentucky,
                            II-B-1-50

-------
indicate there are in excess of 14 billion tons for surface reserves and
almost 95 billion tons of underground reserves (U. S. Department of the
Interior, Bureau of Mines, 1975).

          The Illinois Basin, centered in southeastern Illinois, includes
central and southern Illinois, southwestern Indiana, and western Kentucky.
The strata in the basin are similar to the Appalachian basin on the eastern
side of the Ordovician Cincinnati Arch.  Both areas were flooded by the
Paleozoic Interior Sea forming a double bay and two distinct coal  fields
during the Pennsylvanian.

          Ordovician through Mississippian strata contain abundant lime-
stone, sandstone and dolomite.  The coal-bearing Pennsylvanian system is
predominantly shale with thin limestone, sandstone, and coal beds.   The
four groups of coal-bearing strata are, in descending order, the McLeans-
boro, Carbondale, Tradewater, and Caseyville (18).  The Illinois Basin
lost a great deal of the Mesozoic and Cenozoic strata after a Cenozoic
uplift of the entire sequence.  Pleistocene glaciation left drifts of up
to 500 feet and considerable loess deposits.   These glacial deposits and
buried river valleys contain groundwater resources of considerable size
and importance.

          Very near the southwestern border of the ORBES study area is
New Madrid, Missouri.   In 1811-1812 a series of earthquakes occurred there
that were felt over 2 million square miles from the Gulf of Mexico to
Canada. ,Littl.e histor-ical .-information .exists since the area was sparsely
populated, but the New Madrid earthquake probably released more energy than
any earthquake man ever experienced on the North American continent.  As a
result of this recent geological event, the Jackson Purchase Area  has been
declared a seismic 3 or high-risk area.

1.6.4.  CLIMATOLOGIC

          A saying common to every city and township in the ORBES  Region
goes as follows, "If you don't like the weather here in Louisville (or
wherever you happen to be), just waii; till tomorrow.  It'll surely change."
Weather patterns in the ORBES Region are, to some extent, unique.   Pro-
nounced seasons are a part of the pattern with variations in temperature,
sunlight, wind, snow and rainfall within a given season often being extreme.
A chilly day in the low to middle 50's is not unheard of in July in
Kentucky.  Likewise, record highs  in the 70's come along regularly during
winter months.   These happenings, however, last for only a few days at
most and usually a return to "normal" will come about in less than
twenty-four hours.

          Frontal systems with sufficient strength and temperature
differentials to cause significant weather move across the ORBES Region,
generally from west to east, on the average of about once a week in spring
and early summer.  Occasionally, a ridge of high pressure will develop
extending from the Gulf of Mexico up through the Mississippi Basin and
                             II-B-1-51

-------
into Canada which blocks these frontal systems causing local, but quite
severe, stagnation problems in the ORBES Region.  The weather otherwise
changes in accordance with a regular pattern.  Heat waves are not common
but they do occur.

          Extended periods of bitter cold are not unlikely to occur, the
winter of 1976-1977 being a good example.  Occasional droughts are also a
feature of the region's weather and, aside from frequent and devastating
tornados, are probably the most calamitous of regional weather phenomena.
The spring and summer of 1930 holds the record for being the worst drought
in the region's recorded history.

          Generally, the ORBES Region is considered to have a temperate
climate with moderate temperatures and precipitation.  Monthly averages
for air temperature and precipitation are listed in Table II-B-1-27 for
selected stations in the region.

1.6.5.  FLOW CHARACTERISTICS OF THE OHIO RIVER

          The Ohio, when first viewed by explorers in the seventeenth
century, was totally different from the body of water bearing the same
name today.  At that time the river was shallow and coursed by free flowing
clear water.  The bottom was mostly coarse sand and pea gravel of scoured
stone and aquatic plant life along the banks was dense.  Early in the
nineteenth century coal pits began to appear in the upper watershed
around Pittsburgh and the U. S.  Army began its modification programs
which continue until today.  "A system of 46 locks and dams completed in
the 1930's is now being replaced by a new system of 19 high lift dams
and the "river" is now, over 90 percent of its length, a series of long,
narrow lakes.  Its banks are generally high, barren and eroded, and the
river bottom is coarse sand and pea gravel overlain by soft mud and fine
silt.  Fish life, however, still abounds in this waterway which is considered
to be one of the best sport fishing "spots" in the U. S.

          Recurrence intervals for steady discharge at the McAlpine Dam at
Louisv'lle are listed in Table II-B-1-28 for flows in excess of the 43
year average of 110,000 cfs.  Periods of exceptional low flow ( js 15,000 cfs)
are tabulated in Table II-B-1-29.  The almost continuous change in watershed
during the period over which this data were collected makes absolute
prediction of flow recurrence intervals for both high and low flows difficult.
However, a general description is possible.

          The major problem(s) with impact of industrial and utility develop-
ments along the river come about at seasonal low flow.  The most dramatic
period of low flow or record came in 1930 with a 143 day period at the
McAlpine Dam with a flow less than 15,000 cfs.  Other periods  of extended
Tow flow have occurred since, the latest of significance being the
50 day stretch in 1963, and recurrence of such intervals is certain.
                             II-B-1-52

-------
                               TABLE II-B-1-27
             AVERAGE MONTHLY AIR TEMPERATURES AND PRECIPITATION
              AMOUNTS FOR SELECTED STATIONS ON THE OHIO RIVER*
*

Month
January
February
l-iarch
April
May
June
July
August
September
October
November
December
Pittsburgh, Pa.
Air
Temp.
28.9
29.2
36.8
49.0
59.8
68.4
72.1
70.8
64.2
53.1
40.8
30.7
Precip.
(in.)
2.97
2.19
3.32
3.08
3.91
3.78
3.88
3.31
2.54
2.52
2.24
2.40
Cincinnati, Ohio
Air
Temp.
33.7
35.1
42.7
54.2
64.2
73.4
76.9
75.7
69.0
'57.9
44.6
35.3
Precip.
(in.)
3.67
2.80
3.89
3.63
3.80 •
4.18
3.59
3.28
2.71
2.24
2.95
2.77
Louisville, Ky.
Air
Temp.
(°F)
35.0
35.8
43.3
54.8
64.4
73.4
77.6
76.2
69.5
57.9
44.7
36.3
Precip.
(in.)
4.10
3.29
4.59
3.82
3.90
3.99
3.36
2.97
2.63
2.25
3.20
3.22
*A11 data based on standard 30-year-period, 1931  to 1960.
SOURCE:  Butz, B.P., Schregardus,  D.R.,  Lewis,  B.,  Policastro,  A.J., and
           Reisa,  J.J.,  Jr., 1974.   Ohio River  Cooling Water Study.
           EPA 905/9-74-004.  386  pp.
                             II-B-1-53

-------
                               TABLE II-B-1-28
Recurrence
Interval
Unknown
once every
once every
once every
once every
once every
once every
once every
once every
once every
once every
100 years
50 years
25 years
10 years
5 years
2 years
1.25 years
1.11 years
1 .05 years
1.01 years
                        STEADY  FLOW DISCHARGE  AT THE
                                McALPINE  DAM
                                       Flow
                                       (cfs)

                                       1100 x  103
                                        970 x  103
                                        900 x  103
                                        830 x  103
                                        730 x  103
                                        640 x  103
                                        510 x  103
                                        400 x  10;?
                                        350 x  103
                                        320 x  103
                                        260 x  103

                                        110 x  103
                                     (1937 flood)
43 year average
SOURCE:   USGS, Louisville,  Kentucky,  as  reported  in  the  Environmental
           Impact Statement for the  Newburgh  Lock and  Dam: U. S. Army
           Corps of Engineers,  Louisville  District,  Louisville,
           Kentucky (1974).
                             II-B-1-54

-------
        TABLE II-B-1-29

PERIODS OF EXCEPTIONAL LOW FLOW
      AT THE McALPINE DAM
Year
1Q?Q
1930*
icm*
iyoi
1932*
1933*
IQ^A*
IcnR*
13OU
1936*
icn7*
Ly of
1938*
1939*
1940*
1Q/11*
1QA?*
1943*
1944*
iQ
-------
 TABLE II-B-1-29 (continued)

PERIODS OF EXCEPTIONAL LOW FLOW
      AT THE McALPINE DAM
Year
1946
1947
1948
1949
1950
1951*
1952*
1953*
1954*
1955
1956
1957*
1958
1959*
1960*
1961
1962*
1963*
Months With Average
Flow <_ 15,000 cfs
Month
September
NONE
NONE
NONE
NONE
October
September
September
October
November
NONE
September
NONE
August
NONE
September
NONE
NONE
NONE
September
October
Discharge, cfs
11,046



12,355
15,033
10,295
7,316
9,161
11,735
12,885
14,500


12,077
8,206
Longest Set of Days
Running Consecutively
That Year With A Flow
< 15,000 cfs
13
12
29
59
16
29
I
50



        II-B-1-56

-------
                       TABLE II-B-1-29 (continued)

                      PERIODS OF EXCEPTIONAL LOW FLOW
                            AT THE McALPINE DAM
Year
Months With Average
 Flow <_ 15,000 cfs

       Month
Discharge, cfs
Longest Set of Days
Running Consecutively
That Year With A Flow
1 15,100 cfs
1964*
1965*
1 Qfifi*
1.7OD


io7n
September 11,098
August 12,411
WONF 	
MftNF 	
NDNF 	 ;
MONF 	 	 	
-MfiNF 	
28
8




1971
* - Years having a period of low flow (<_ 15,000 cfs) extending for 7 or
    more days.

Source:  These data were compiled by H.T.  Spencer from official  U.S.G.S.
         records and reported in the Environmental Impact Statement for
         the Newburgh Lock and Dam:  U.S.  Army Corps of Engineers,
         Louisville District, Louisville,  Kentucky.
                             II-B-1-57

-------
          A common misconception about the high lift dam system is that it
serves as both a water storage reservoir and barge facility.  In actuality
the dams pass seasonal high flows without storage and subsequently low
flows cannot be augmented significantly from water held within the river
channel itself without endangering barge traffic.

1.6.6.  MINERAL RESOURCES IN ORBES

          The minerals industry is an important economic sector in each of
the ORBES states.  While coal is dominant, all of the states have other
significant mineral resources.  The total  dollar value of minerals production
and the percentage contribution by resource type are shown in Table II-B-1-30.

          The proven fuel reserves and the average coal composition for
the four states are presented in Tables II-B-1-31 and II-B-1-32, respectively.
Much of the coal reserve is high in sulfur content and its use will be
constrained in the future unless coal cleaning or stack gas desulfuriza-
tion come into widespread use.

          Minerals that may be of increasing importance during the study
period are limestone, lime and Devonian shale.  The large scale application
of flue gas desulfurization systems within the region, many of which use
lime or limestone, will result in a significant increase in their production.
Limestone is abundant in the region and major new mines have recently
opened solely to supply scrubbers.

         •Devonian "gas"shales are geologic formations which underlie
extensive protions of the ORBES Region.  Natural gas is trapped within
the shale,  but the quantity is unknown and the technology must be
developed before commercial quantities can be produced.  According to the
FEA, "If a major, rapid development program were undertaken, it might be
possible for sufficient quantities of natural gas to be produced from
local resources to enable these states (including all of the ORBES states),
which, in many cases, are those most affected by the current shortage, to
have an adequate gas resource base" (19).   The quantity of gas from this
source is potentially very large, with one estimate of 4-6 Qcf in
Appalachia  and another estimate of 285 Tcf under five states including
Kentucky and Ohio.  The proven gas reserve in 1974 for the entire United*
States was  estimated to be 237 Tcf, so Devonian shale gas could be a major
additional  fuel  resource.  The ERDA "Eastern Gas Shale Project" has state-
based studies to characterize the Devonian shale resource and there are
field experiments underway in Eastern Kentucky.  FEA estimated that gas
production  from shale formations could be  as much as .09 Tcf/yr by 1985
with conventional technology at a cost of  over $3.50/Mcf (20).

1.6.7.  BIOLOGICAL RESOURCES IN ORBES

          The"biological  wealth of the ORBES Region in terms of timber,
farming, livestock, aquatic resources (commercial fisheries and "put and
take" lakes), and wild habitats and preserves is considerable.  Major
                             II-B-1-58

-------
                       TABLE II-B-1-29 (continued)

                      PERIODS OF EXCEPTIONAL LOW FLOW
                            AT THE McALPINE DAM
Year
Months With Average
 Flow <. 15,000 cfs

       Month
Discharge, cfs
Longest Set of Days
Running Consecutively
That Year With A Flow
< 15,100 cfs
1964*
1965*
IQfifi*
1.7OO
1Qfi7
J.3O/
1QCO
jtyoo
1QCQ
iyoy
1.Q7A
iy I u
1Q71
iy 1 l
1Q79
iy/c.
September 11,098
August 12,411
NOMC 	
wnwp 	
MfiNF 	
MOMF _____
wowc 	
Mf||\ir 	
Mnwr 	 	

28
8







* - Years having a period of low flow (<_ 15,000 cfs) extending for 7 or
    more days.

Source:  These data were compiled by H.T.  Spencer from official U.S.G.S.
         records and reported in the Environmental Impact Statement for
         the Newburgh Lock and Dam:  U.S.  Army Corps of Engineers,
         Louisville District, Louisville,  Kentucky.
                             II-B-1-57

-------
                               TABLE II-B-1-30
                  VALUE AND COMPOSITION OF THE MINERALS
                       INDUSTRY IN THE ORBES STATES

                     Percent of Total  Value by Sector
State
Illinois
Indiana
Kentucky
Ohio
Coal
50
44
85
42
Stone
14
17
6
12
Sand and
Gravel
8
10
1
9
Crude
Petroleum
16
6
3
6
Natural
Gas
1
1
2
5
Other
12
22
3
26
Total Value
Million $
0.826
0.351
1.165
0.807
SOURCE:  Minerals Yerabook, Vol.  II  Area Reports:   Domestic,  U.'S.
           Department of Interior,  Bureau of Mines,  1976.
                             II-B-1-59

-------
                               TABLE II-B-1-31
                            PROVEN FUEL RESERVES
State
Illinois
Indiana
Kentucky
Ohio
Coal
(million
short tons)
139,756
34,779
65,952
41,864
Petroleum
(thousand
barrels)
208,763
30,855
52,548
129,144
Natural Gas
(million
cubic feet)
498,953
86,678
956,296
1,068,372
Natural Gas
Liquids
(thousands
barrels)
942
27
47,118
—
SOURCE:  R. P. Carten, et al_.  "Surface Mined Land in the Midwest:  A
           Regional Perspective for Reclamation Planning," Argonne
           National Laboratory (ANL/ES-43), June, 1974.
                             II-B-1-60

-------
                              TABLE II-B-1-32
                     AVERAGE COMPOSITION OF COALS IN
                             THE ORBES REGION
State
Illinois
Indiana
Western
Kentucky
Ohio
Moisture Volatile Fixed
Matter Carbon
(%) (*) (*)
6.5
10.0
11.3
8.1
12.0
15.2
4.3
7.. 6
11.2
3.3
5.0
12.0
32.3
34.5
37.2
31.7
36.0
40.3
32.3
• 36.0
37.4
30.1
36.0
41.6
40.3
45.0
51.2
39.0
44.0
47.6
41.5
46. .-5
53.7
43.4
48.0
53.2
Ash
(«)
6.8
10.0
12.9
6.2
9.3
12.1
7.6
10., 0
12.2
4.3
8.5
14.6
Sulfur
(%)
81
100
54
100
85
100
69
100
3.0
0.7
3.0
0.0
3.0
1.0
3.0
0.7
Heating
Value
(BTU/lb)
10,450
10,800
13,030
11,010
11,400
12,010
10,830
11,940
12.780
10.280
12,500
13,160
SOURCE:  Same as Table II-B-1-31.
                             II-B-1-61

-------
cash crops include corn, wheat, tobacco and soybeans.   The raising of
cattle for beef is commonplace as is dairy farming and the production
of pork.

          Surface mining interest for a time competed  for farm land in the
region, but present reclamation laws are rapidly bringing about the return
of mined land to agricultural use.  In some instances  land surface-mined,
but properly reclaimed, is more productive than before.   However,  this
cannot be taken as a general rule.  Estimates of the acreage of orphaned
or pre-law strip-mined land range up to 1  million acres  for the region.
Abandoned mining operations are the worst  source of water pollutants in
ORBES in terms of both quantity of pollutants and habitat destruction.

          Unique and diverse habitats are  still numerous in the region.
Conservation efforts had led to the preservation of remaining tracts of
virgin timber.  Wildlife refuges along the Mississippi flyway have long
been established to protect migratory waterfowl and the  region's remain-
ing unique and natural aquatic habitats are protected.

          Extensive forest cover most of eastern Kentucky providing both
timber and wildlife cover.   In addition to this numerous man-made  im-
poundments can be found in ORBES which are biologically  productive.   All
of these areas are protected by law from destruction and pollution but
enforcement has at times been hampered by  the simple logistic problem of
having too few people, no matter how dedicated, to cover too much  land.

          The worst threat to the health and well-being  of the region's
people and biological  resources is clearly chemical pollution.
                             II-B-1-62

-------
                    2.  PRELIMINARY REVIEW OF THE RTC'S
           Each university team involved in the ORBES project had the
responsibility of providing a preliminary review of the RTC's and
associated assumptions.  The UK/U of L team chose, as a means of
evaluation, to undertake development of plausible RTC's on their own for
the sake of comparison to the Task 1 report.  The methodology and out-
come of these deliberations are contained in the following sections.
In summary the UK/U of L team found conflict between their deliberations
and those of the Task 1 group in four major areas.  These are tabulated
below.

           I.  Population projections.  Population projections are the
subject of section 2.2.   Members of the Urban Studies Center at U of L,
particularly the demographers, were quick to point out inconsistencies
in the methods used by the Task 1 group in projecting populations for
the year 2000.  Subsequent to this a team at the Center for Advanced
Computation at the University of Illinois at Urbana-Champaign have made
a similar study of ORBES population projections.  The results of these
studies are tabulated below:

                       Group                          Year 2000

                       U of L Urban Studies Center    20,977,046
                       UI Center                      21,134,433 .
                       ORBES                          23,025,164

The U of L and UI predictions are essentially the same, both being 9% below
the official  ORBES forecast.

           II.  Energy transport out of ORBES.  A major tenet of ORBES
planning has been the concept that virtually 100% of the electrical
energy produced in the ORBES Region is also used there.  The U of L/UK
group could not agree to this and the issue has never been resolved.

           III.  Basic technological and sociological assumptions.
A number of other ORBES assumptions were questioned.  Details of these
considerations are given in section 2.3.

           IV.  Siting.  The question of doing an impact assessment on
facilities sited down to the county level, but no further, came up
repeatedly.  This issue is addressed in the following impact chapters and
is discussed in the summary as well.
                              II-B-2-1

-------
2.1,  The Sequential  Modified Delphi  Process

      The Kentucky Team,  made up of representatives from the univer-
      sities of Louisville and Kentucky,  had a unique opportunity
      and means for carrying out 'a preliminary review of the Regional
      Technology Configurations (RTCs)0   In its original  submission
      proposal, the Kentucky Team had incorporated  into  its  research
      design a sequential  Modified Delphi  Process  as a means for
      carrying out Task 1  of the EPA  work  plan—that is,  the devel-
      opment of plausible future Regional  Technology Configurations.
      Although the development of such RTCs was not required of  the
      university teams  involved in the ORBES project, EPA did approve
      the Kentucky Team's Delphi undertaking,,   The  architects of the
      Kentucky Team proposal  had three principal purposes for carry-
      ing out the Task  1  assignment via  the Delphi  Process.

      First, and most important, the  process would  provide a learn-
      ing process by  which the 27 members  of the Kentucky Team could
      be made aware of  the diverse and complex issues involved in
      the ORBES study.   Secondly, the process  would provide  an ex-
      cellent means for bringing together, organizing and unifying
      the multi-disciplined team which,  like all such groups, could
      be expected to  encounter difficulties that naturally arise
      from the lack of  common backgrounds.  And thirdly,  the carry-
      ing out of the  Task 1 exercise  independently  of ORBES1  Task 1
      Team would provide  a measure of or preliminary review  of the
      RTCs developed  by the ORBES Task 1 Team.

      The Modified Delphi  Process did achieve  these purposes—not
      perfectly certainly, but still  to a  remarkable degree,.   Al-
      though designed to  parallel  the development steps  followed by
      the ORBES'  Task 1 Team, the Kentucky Team's Delphi  exercise
      did not produce precise, quantified  RTCs for  the future.   It
      did, however, produce consensus on all  the key issues  involved
      in the development  of RTCs, i.e., the basic assumptions and
      projections which are the necessary  foundations for the con-
      struction of plausible  futures.

      In all, 27 individuals—nearly all of them drawn from the pro-
      fessorial  ranks of  the  two Kentucky  institutions—participated
      in the Delphi Process which took place generally during Sep-
      tember, October and  November of 1976,,   Of the 27,  15 were
      drawn from the  University of Louisville  and 12 from the Uni-
      versity of Kentuckyo

      There were five formal  meetings of the group.   The  format  for
      each was generally  the  same.  Appropriate papers and background
      information were  provided each  participant for his  study well
      in advance of each  of the meetings.   Guideline forms were  in-
      cluded to focus the  discussions on the various steps that  had
      to be taken in  the  development  of the plausible RTCs.

      The first meeting in the Delphi Process  was held September 16
      in Louisville at  the University of Louisville's Urban  Studies
                        II-B-2-2

-------
 Center.  The  same  site was chosen for all of the other sessions
 save one which was held  in Lexington on the University of Ken-
 tucky's campus.  The first meeting had as its specific purpose
 the organization of the  team and its familiarization with the
 ORBES  study project.  The EPA work plan for the project was
 discussed in  considerable detail and participants were assigned
 to develop various information and working papers appropriate
 to the undertaking.

 Difficulties  in obtaining statistics and other detailed descrip-
 tive data on  the other states within the ORBES region made it
 necessary to  delay the start of the Delphi discussions until
 November 15.  By then the managers of the Kentucky Team had ob-
 tained the descriptive data assembled by the ORBES' Task 1 Team
 which  met the data needs for the Delphi Process.  The descrip-
 tive data—and only that—from the Task 1 Team's efforts were
 circulated well in advance along with an extensive array of pa-
 pers and articles related to energy issues to the Kentucky Team
 participants.  At this meeting the Kentucky team participants
 were asked to concentrate on the development of technological
 and sociological assumptions that necessarily would have to be
 made as a foundation for the RTCs.  To facilitate the develop-
 ment of consensus, a questionnaire form was' used in which each
 of the participants was allowed free choice in stating what he
 felt the assumption should be, given his own disciplinary back-
 ground.

The results of the Round I discussions and the tabulations of
 the questionnaires were circulated and became the basis of
 more concentrated discussions at the Round II session held in
 Lexington on  November 22.  At this meeting general consensus
 was achieved  on all basic assumptions, technological and socio-
 logical, and  these were used as a basis for developing projec-
 tions  through the year 2000 in all  of the significant issue
 areas  involved in the ORBES project,  A paper setting down with
 considerable  specificity the projections of the assumptions
 arrived at in Rounds I and II was circulated well in advance of
 the Round III meeting which was held November 29—along with a
 form allowing each participant to prepare to note any excep-
 tions  he might have to either the finalized assumptions and
 their  projections.  The Round III session was perhaps the live-
 liest of all those held and clearly demonstrated the utility of
 the Delphi Process in focusing persons with diverse backgrounds
 into consensus.

 Consensus achieved by the Kentucky Team in many instances dif-
 fered  significantly with the assumptions and projections ar-
 rived at by the ORBES Task 1 Team.   These differences were the
 principal points of discussion at the final  or Round IV session
 of the Delphi Process.   The primary purpose of this session
was to compare the results of the Delphi Process with those of
 the Task 1 Team and to determine whether the Kentucky partici-
 pants felt that adjustments should be made in their conclusions.
 Following are the projections of assumptions derived from the
                    II-B-2-3

-------
 Modified  Delphi  Process  carried  out  by the  Kentucky  Team.
 These projections  are  drawn  from the summary  document  that  pro-
 vided the basis  for the  Round  IV Session  at which  the  balance
 of the ORBES1  Task 1 Team's  report was distributed and compared,
 Only  at this  point did the majority  of the  Kentucky  Team partic-
 ipants know what conclusions had been  reached by the ORBES  Task
 1  Team.

 2.1.1.        FINAL DELPHI  PROCESS  PROJECTIONS
               Technological and Sociological  Assumption's
               Ohio River Basin  Energy Study

 DEMOGRAPHIC ASSUMPTIONS  AND  PROJECTIONS

 a.  Birth Rate:  Birth rates in  the  Ohio  River Basin will remain
    constant  at  the 1970-1975  state  levels  through the year 2000.
    These crude  birth  rates  are  16.2 births per 1,000  population
    in Ohio,  15»4  in Illinois, 16.8  in Indiana, and  17.0 in Ken-
    tucky.

 b.  Death Rate:  Death rates in  the  Ohio  River Basin will remain
    constant  at  the 1970-1975  state  levels  through the year
    2000.  These crude death rates are 9.2  deaths  per  1,000
    population in  Ohio,  10.8 in  Illinois, 904 in Indiana, and  .
    10.2  in Kentucky.

.c,.  Migration:   Net out-migration will  continue at the 1970-1975
    level  for Ohio,  Illinois and Indiana.   Net in-migration will
    continue  for Kentucky.   Net  migration rates for  Ohio are
    -2.5,  -2.9 for Illiruis, -2.6 for  Indiana, and +3.2 for Ken-
    tucky.  These  migration  trends will produce a  redistribution
    of population  within the Ohio. River Basin, population moving
    out of the northernmost  counties to the southern counties  in
    Kentucky,,  It  is also anticipated  that  the trend towards ur-
    ban decentralization will  continue.

 d.  The previously listed rates  have been combined to  produce  a
   , population projection for  each state  in the Ohio River  Basin,
    as well as the total region:
              Crude  Birth        Crude Death      Net Migration
State            Rate                Rate                Rate
 Ohio             16.2                 9o2               -2,5

 Illinois         15,4                10.8               -2,9

 Indiana          1608                 9.4               -2.6

 Kentucky         17.0                10»2                3.2
                     II-B-2-4

-------
                        2000                    1975
                  Population  Percent     Population   Percent

                   8,426,837    40.2       7,530,200     40.9

                   3,541,875    16.9       3,394,500     18.4

                   4,623,301    22.1       4,100,500     22,3

                   4,360,550    20.8       3,396,000     18,4

                  20,952,564   100.0      18,421,200    100.0

 SOCIOECONOMIC ASSUMPTIONS

 a.   The rate of industrial  growth will  definitely  increase  in
     the Ohio River Basin, both for existing  and new industry.
     Much of the growth is expected to  occur  in  chemical  pro-
     duction, mining,  construction, energy-related  industry,
     agriculture and food  production, and  recreation,,

 b.   Reflecting the above  assumption, the  Gross  Regional  Product
     (GRP) will increase significantly  and at a  somewhat  higher
     rate than will  the Gross  National  Product (GNP).   GNP is ex-
     pected to grow about  2.3  percent per  year through  the year
     2000, while that  of the ORBES region  will  increase at the
     rate of 205 percent per year.   The results  are shown in  the
     fbl'low'ing projection:

 Comparison of GRP and GNP in  Billions  at  1974 Constanl Dollars

                                 1974          1985         2000

 ORBES                           210.34        276.00      400.00
 US                            1,397=40      1,795.00    2,524.00
 ORBES  as percent of US          15.05         15«38      15.80

.c.   Relatively,energy consumption by sector  will change  very
     little through the year 2000,  with the greatest increase be-
     ing shown in industrial usage in keeping with  the  expanding
     industrial growth projected above.  The  projections, ex-
     pressed in percentages, are as follows:

                                1975       1985         2000

'Residential                      20%         18%         19%
 Commercial                       13          13           11
 Industrial                       37          43           48
 Agricultural                      1            1            1
 Transportation                  29_          25_          21_

                    TOTALS      100%       100%         100%
                     II-B-2-5

-------
.do   Relatively, energy consumption  by  sources will change sig-
     nificantly.   Energy  production  by coal and by nuclear facil-
     ities will  expand importantly.  Solar energy will become a
     factor, albeit a minor one, primarily for residential pur-
     poses.

     The relative  importance of petroleum, natural gas and hydro/
     geo-thermal sources  will decrease.

     The projections of energy consumption by source, expressed
     in percentages, is as follows:

                               1975        1985         2000

     Coal                        37%         38.0%       44.0%
     Petroleum                   32          31.0        20.0
     Natural Gas                 28          23.0        16,0
     Hydro/Geo-Thermal            1            .5          »3
     Nuclear                      2           7.0        13.7
     Solar.                      JO          	.J5         6.0

                      TOTALS   100%        100oO%      100.0%

e.   Transportation.  Automobiles will continue to serve as the
     primary movers of workers and recreation-seeking adults
     throughout  the Basin.  However, the relative importance of
     the private automobile for getting to and from work will de-
     cline.  By  1985 the  percentage of work-oriented trips by
     private car will have declined to 84 percent, and by the
     year 2000 to 75 percent,,

     This will be because economics will force substitutions,
     such as increased carpooling, reliance on improved mass
     transit, etc.

     1.  Gasoline consumption will decline, primarily because of
        increased costs, improved efficiency of cars and legis-
        lation against "guzzlers," improved mass transit and
        greater acceptance and/or legislation for conservation.

     2.  By the year 2000 it is projected that the average miles
        per gallon consumed by American automobiles will be be-
        tween 20 and 30 MPGs.

POLITICAL ASSUMPTIONS

a.   It can be projected that conservation will  become a social
    movement.  This movement, both through popular commitment
     and as a result of legislation, will promote conservation
     of natural resources and the protection of the air, water
     and land from pollution.

b.   It also can be projected that there will  be an increasingly
     high degree of regulation within the Basin of:  nuclear
                    II-B-2-6

-------
     power  plants,  both  strip and deep mining, air pollution and
     water  pollution,,

     1»  Utility companies will experience a high to moderate de-
        gree of regulation.  But land use, petroleum products
        and prices will experience only moderate to light regu-
        lation,,

     2.  The effectiveness of such regulations—local, state and
        federal—will vary.  The most effective regulation will
        be focused on nuclear power plants and both strip and
        deep mining.  The regulation of air pollution, water
        pollution and utility companies will be somewhat less
        effective.  Least effective, and perhaps even ineffective,
        will be the regulation of land use, petroleum products
        and prices.

c.   Research.  Thn funding of research and development will gen-
     erally increase through the year 2000.  More specifically:

  •   1.  Funding for research and development in coal and solar
        energy will increase tremendously.

     2.  The funding for nuclear power research will increase
        somewhat.

     3.  But the .funding of research into matters related to pe-
        tro'leum products, natural gas and hydro/geo-thermal en-
        ergy will remain about the same.

d.   Oil Imports.  It is not unlikely that there will be another
     oil embargo or embargoes between now and the year 2000 and
     that the oil exporting nations will  continue to increase
     prices.

TECHNOLOGICAL ASSUMPTIONS

a.   Tributary, mainstream and dry siting of facilities are all
     feasible in the ORBES Region.  Siting at mine mouth will be
     preferred to siting at load centers  due to the cost advan-
     tage of shipping electricity as compared to coal.

b.  The principal energy sources to be used in the Basin are
    coal and uranium.   The fossil fuel  to uranium mix  assumed
    for the year 2000 for 1000 MW units  is 80 percent  fossil
    fuel and 14 percent uranium.  It is  further assumed that 6
     percent of the ORBES energy supply will come from  solar
     systems (predominantly residential).  Of the fossil  fuel
    component, 44 percent will  come from coal, 20 percent from
    oil and 16 percent from natural  gas.  These projections are
    significantly different from those made by the Task 1 Study
    Team.

Co  Significant conversion of coal  to methane (synthetic natural
                     II-B-2-7

-------
     gas)  is considered very  likely  by the year 2000.  The con-
     version plants will  likely be mine-mouth facilities.

 d.   Conservation of electricity will be a significant factor in
     determining demand for the year 2000,

.:6.   Adequate barge and rail  facilities exist for development of
     both  coal and nuclear power parks (sites producing 4,000 or
     more  MW).  Feasible  sites for power parks are along the Ohio
     and on the Tennessee and Cumberland Rivers below Kentucky
     Lake  and Lake Barkley,

 f.   Adequate water is available throughout the ORBES Region for
     industrial and/or power  plant development with the restric-
     tion  of Dust Bowl years  such as 1930 recurring.

 g.   It has been assumed  that new plants built between now and
     the year 2000 will have  S02 controls if coal fired and will
     be under very strict control if nuclear.

 h.  -Two assumptions not addressed by the Modified Delphi Team
     but addressed by the Task 1 Group which must be stated be-
     fore  scenario construction are:

     1.  Sixteen percent of all electrical energy generated in
        one year in the United States is generated in the ORBES
        Region.  It is assumed that this percentage will remain
      •  constant through the year 2000.

     2.  Demand for energy in the ORBES Region roughly equals
        consumption.  That is, the  net exports (from the four
        states, not the ORBES Region) are zero, and it is
        assumed that they will remain zero through the year
        2000o

 i»   Breakthroughs.  As a result of research and development, it
     can be expected that there will be important breakthroughs
     in all matters related to energy production.

     1.  The most likely breakthroughs will  occur in improved
        methods of coal conversion,  both in liquefidation and
        gasification.

     2.  Equally certain will  be breakthroughs in improving the
        efficiency of the transmission of electrical power which
        will reduce losses in transmission.

     3.  Important breakthroughs are expected also in the conver-
        sion of solid wastes into energy resources, and in great-
        ly improved conservation measures.

     4.  It is also likely that research in matters related to
        nuclear power, particularly in development of fast-
        breeder reactors and in safety controls, will produce
                     II-B-2-8

-------
    breakthroughs.

5.  There will be improved technology in solar electrical pro
    duction.
                 II-B-2-9

-------
2.2.  POPULATION PROJECTIONS

        Planning for the future has become a necessary part of socio-
political life in our society.  Thus, it has become increasingly important
for policy and decision makers to have access to the most accurate and
up-to-date population information available.

        In no instance is this more true than in the Ohio River Basin
Energy Study (ORBES).  Given the critical and urgent purposes of the pro-
ject, ORBES requires the best possible population estimates that can be
developed, particularly concerning the present and future population size
of the ORBES Region for the years 1975, 1985, and 2000.

        Shryock and Siege! (1973) point out in the United States Department
of Commerce publication, Methods and Materials of Demography, that popu-
lation estimates viewed broadly can be divided into several types on the
basis of their time rer'erence and basis of derivation:

        1)  intercensal estimates, which relate to a date intermediate
            between two censuses and take the results of these censuses
            into account.

        2)  postcensal estimates, which relate to a past or current
            date following a census and takes that census and possibly
            earlier censuses into account.

        3)  projections^ which relate to dates following the last census
            for which no current reports are available.

        In the case of the ORBES project, a postcensal population estimate
for 1975 and population projections for 1985 and 2000 are required.  Post-
censal estimates are generally made by utilizing actual postcensal data
from the recent past in the form of vital statistics, tabulations from
population registers, or statistics that are merely correlated with popu-
lation change (e.g., school enrollment).  Population projections, on the
other hand, are usually based on a model developed from certain assumptions
regarding the principle components of population change (fertility, mor-
tality and migration) and base year data that illustrate certain analytical
relationships.

        The 1975 population size for counties within the ORBES Region can
be obtained from the annual population estimates produced under the auspices
of the Federal-State Cooperative Program for Local Population Estimates.
These annual  population estimates are reported in the U.S. Census Bureau's
publication,  Current Population Reports (Series P-25 or Series P-26), for
all counties  in the United States.

        However, the size of the population for 1985 and 2000 (i.e., popu-
lation projections) for counties within the ORBES Region is an entirely
different matter.  First, the U.S. Census Bureau does not currently produce
county population projections.  Second, there is no uniform procedure for
data input and methodology for producing county population projections, as
                             II-B-2-10

-------
there is in deriving the annual U.S. Census County population estimates.
Third, population projections provided by the individual states partici-
pating in the ORBES project use disparate methodologies in deriving their
county population projections.  Therefore, population projections for. the
years 1985 and 2000 for counties within the ORBES Region are far less  ;
standardized than the 1975 county population estimates for the ORBES Region.

        The use of different methodologies and data input for deriving
county population projections will, of course, produce diverse enumerations
of the future population size.  Since population size has direct impli-
cations for the future energy needs and consumption of a county, it is of
more than an academic interest to have the most up-to-date, complete and
standardized set of county population projections for the ORBES Region.
Furthermore, it would be advantageous for a standardized set of population
projections for the counties in the ORBES Region to incorporate into the
assumptions of the model the most current county specific data available.

        It is not feasible, given the budget constraints in Phase I of the
ORBES project, to develop and produce these types of county population
projections for the ORBES Region.  Nevertheless, given the importance of
population information in this study, it seems appropriate that funds be
allocated to produce a set of standardized population projections for
counties in the ORBES Region during Phase II of the project.

        The Kentucky ORBES team generated, as a part of the development of
plausible future Regional Technology Configurations (RTC's), a new set of
county population projections for the ORBES Region using a standardized
procedure.  Several social, physical and natural scientists from the Uni-
versities of Louisville and Kentucky participated in this process.

        After reviewing the actual  birth, death and migration rates for
the states in the ORBES Region between 1970 and 1975, it was the consensus
of the Kentucky Delphi Project Team that the following fertility, mortality
and migration assumptions would likely reflect the future population trends:

        1)  Fertility:  Birth rates in the Ohio River Basin will remain
            constant at the 1970-1975 state levels through the year 2000..
            These crude birth rates are 15.4 births per 1000 population
            in Illinois, 16.8 in Indiana, 17.0 in Kentucky, and 16.2 in
            Ohio.

        2)  Mortality:  Death rates in the Ohio River Basin will remain
            constant at the 1970-1975 state^evels through the year 2000.
            These crude death rates are'39r8 in Illinois, 9.4 in Indiana,
            10.2 in Kentucky, and 9.2 in Ohio.

        3)  Migration:  Net out-migration will  continue at the 1970-1975
            level for Illinois, Indiana, and Ohio.  Net in-migration will
            continue for Kentucky.   Net migration rates are -2.9 per 1000
            in Illinois, -2.6 for Indiana, -2.5 for Ohio, and +3.2 for
            Kentucky.   These migration trends will produce a redistribution
            of population within the Ohio River ;Basin (i.e., population
                             II-B-2-11

-------
             moving  out  of  the  northernmost  counties  to  the  southern  counties
             in  Kentucky).   It  is  also  anticipated  that  the  trend  towards
             urban decentralization will  continue.

         The previously  listed  rates  have been  combined  to produce a  popu-
 lation  projection for each county in the Ohio  River  Basin by  means of  the
 following  continuous compounding  formula:
                      P  -  P e
                    •  Ht    po

        Where:        Pt =  the  population  size at the  end  of  the  period
                           (e.g.,  1985 or  2000);

                      PQ =  the  population  size at the  beginning of  the
                           period  (e.g. , 1975);

                      e =  the  base of the natural system  of  logarithms
                           (i.e.,  2.71828);

                      r =  the  growth rate;

                      t =  the  number of years.

        Although  this extrapolation procedure provides  a  complete  and
..standardized .set  of population projections,  it does not necessarily
 represent  the only  viable  alternative for generating  county  population
 projections.  In  fact, this model could be improved by  incorporating
 intj  it additional  data which  are specific to the ORBES counties,  as well
 as trending techniques which would alleviate the necessity of holding
 specific rates constant throughout the projection period.  Concomitantly,
 therer are other  methodologies  for deriving  population  projections that
 can utilize county, age, sex and  race specific data which also could be an
 improvement over  the  alternative  selected by the Kentucky Delphi Team.
 However, current  funding prohibits the undertaking of any of these tasks.

        The remaining section  of  this chapter reports total  population for
 the counties in the ORBES  Region , for the  years 1975,  1985, and 2000.  The
 1975  data  represent county population estimates produced  by  the  Federal -
 State Cooperative Program  for  Local Population Estimates.  The county
 population projections for 1985 and 2000  will have two  figures reported:
 1) "KDP",  the new set of county population projections  produced  vis-a-vis
 the Kentucky Delphi Project; and  2) "ORBES"  the initial county population
 projections collected from the  participating states.

        The following table is  offered to summarize the differences between
 the "KDP"  and "ORBES" population  projections for the  ORBES Region  and
 individual states.
                             II-.B-12

-------
                                                     TABLE II-B-2-1

                                          Population Estimates and Projections for the ORBES

                                                        Region 1975,  1985 and 2000
   STATE
                     1975J
POPULATION
                           KDP
                                       1985 Population
ORBES''    %
KDP
                                                                                    2000 Population
ORBES*
   Illinois    3,394,500     18.4    .  3,453,523     17.8     3,522,706  17.3
                                                                      3,572,712    17,0     4,040,331   17.6
_  Indiana     .4,100,500     22.3      4,298,031     22.2     4,514,100  22.2        4,616,192    22,0     5,122,200   22.2
I—(


7s  Kentucky    3,396,000     18.4      3,753,988     19.4     3,775,814  18,6        4,361,192    20.8     4,406,794   19.1
   Ohio
7,530,200     40.9      7,876,903     40.6     8,530,686  41,9
                                                                      8,426,950    40,2     9,455,839-*  41.1
   TOTAL ORBES 18,421,200   100.0     19,3-82,445    100,0    20,343,306 100.0
   REGION
                                                                     20,977,046   100.0    23,025,164  100.0
    U.S. Census Bureau. Current Population Reports. Series P-26, Numbers 75-13, 75-14, 75-17 and 75-35.
    Washington, D.C. 1976
   9
    Ohio River Basin Energy Study  (ORBES). Task !_: Development of Plausible Future Regional Technology
    Configurations. Table lb-1. October, 1976.
   3
    This figure was calculated by  the ORBES Management's Task I Team on the assumption that the ratio
    between the population in Ohio within the region and the total population for Ohio remains constant
    between 1985 and 2000.

-------
                 TABLE II-B-2-2
                             )




POPULATION  ESTIMATES AND PROJECTIONS FOR





    ILLINOIS  COUNTIES  (ORBES  REGION)






                         POPULATION
FIPS
Code
001
003
005
009
Oil
013
017
019
021
023
025
027
029
033
035
039
041
045
047
049
051
053
055
057
059
County
Name
Adams
Alexander
Bond
Brown
Bureau
Calhoun
Cass
Champaign
Christian
Clark
Clay
Clinton
Coles
Crawford
Cumberland
De Witt
Douglas
Edgar
Edwards
Ef f ingham ..
Fayette
Ford
Franklin
Fulton
Gallatin
1975
70,000
11,800
14,500
5,400
36,400
5,550
13,800
163,400
36,400
16,200
15,100
29,400
47,800
19,400
10,200
16,900
18,500
21,400
7,300
27,400
20,500
14,900
40,900
42,500
7,200
KDP
71,200
12,002
14,749
5,493
37,024
5,594
14,037
166,202
37,024
16,378
15,359
29,904
48,620
19,733
10,375
17,190
18,817
21,767
7,425
27,870
20,852
15,156
41,601
43/229
7,323
1985
ORBES
66,889
10,342
15,660
4,975
38,197
5,380
13,872
176,783
36,960
15,751
15,704
30,645
50,515
20,182
9,791
16,759
18,800
20,347
7,495
28,434
21,498
14,756
41,075
42,928
7,259
2000
KDP
73,039
12,312
15,130
' 5,634
37,980
•v,739
14,399
170,494
37., 980
16,903
15,756
30,676
49,875
20,242
10,643
17,634
19,303 •
22,329
7,617
28,590
21,390
15,547
42,676
44,345
7,513
ORBES
81,518
9,482
16,845
5,179
43,034
5,260
15,625
217,102
39,881
16,770
16,360
31,880
59,173
21,791
11,211
18,674
20,632
22,682
7,7.94
32,969
23,393
16,030
43,890
51,690
7,600
                 II-B-2-14

-------
TABLE II-B-2-2 (continued)



                POPULATION
FIPS
Code
061
063
065
067
069
071
073
075
077
079
081
083
087
091
095
099
101
105
107
109
113
115
117
119
121
123
125
127
129
131
County
Name
Green
Grudy
Hamilton"
Hancock
Hardin
Henderson
Henry
Iroquois
Jackson
Jackson
Jefferson
Jersey
Johnson
Kankakee
Knox
La Salle
Lawrence
Livingston
Logan
McDonough
Me Lean
Macon
Macoupin
Madison
Marion
Marshall
Mason
Massac
Menard
Mercer
1975
16,600
27,500
8,300
21,900
5,000
8,200
55,100
32,600
51,900
11,100
33,800
19,400
8,700
95,800
61,100
108,400
17,200
40,900
30,500
39,400
116,600
127,300
46,200
247,600
40,000
13,100
18,000
13,800
10,700
17,300
198
KDP
16,885
27,972
8,442
22,276
5,086
8,341
56,045
33,159
52,790
11,290
34,380
19,733
8,849
97,443
62,148
110,259
17,495
41,601
31,023
40,076
118,599
129,483
46,992
251,845
40, '€86
13,325
18,309
14,037
10,883
17,597
5
ORBES
16,050
30,660
9,159
21,303
5,309
8,900
59,325
33,181
53,021
1I,'J37
34,462
20,155
9,415
101,520
65,244
110,773
19,060
42,784
29,389
45,255
128,249
128,835
47,660
250,335
42,115
13,342
16,773
12,773
11,002
19,115
2000
KDP
17,321
28,694
8,660
22,851
5,217
8,556
57,492
34,015
54,153
11,582
35,268
20,242
9,078
99,959
63,753
113,106
17,947
42,676
31,824
41,111
121,662
132,827
48,206
258,350
41,737
13,669
18,782
14,399
1.1,165
38,051
ORBES
15,980
35,510
10,366
24,364
5,385
10,429
68,187
34,898
65,599
12,091
39,394
22,970
10,421
114,930
76,448
126,375
19,932
46,193
32,490
60,422
153,306
152,473
55,239
293,100
45,635
13,748
22,854
14,087
12,803
23,392
     II-B-2-15

-------
TABLE II-B-2-2 (continued)



                POPULATION
FIPS
Code
133
135
137
139
143
145
.147
149
. 151
153
155
157
159
1'6 3
165
167
169
171
173
175
179
181
183
185
187
County
Name
Monroe
Montgomery
Morgan
Moultrie
Peoria
Perry
Piatt
Pike
Pope
Pulaski
Putnam
Randolph
Richland

-------
                        TABLE  II-B-2-2 (continued)

                                      POPULATION

FIPS
Code
189
191
193
199
203

County
Name
Washington
Wayne
White
Williamson
Wood ford
1975

14,700
17/000
16,300
52,100
29,400
1985
KDP
14,952
17,292
16,580
52,993
29,904
ORBES
15,310
17,820
16,249
53,577
30,472
2000
KDP
15,338
17,738
17,008
54,362
30,676
ORBES
15,960
18,506
16,647
61,153
34,702
SOURCE:  U.S. Census  Bureau.  Current Population Reports.
         Series P-26,  No.  75-13  Washington, D.C. 1976.

         UniversityT''of Louisville.  Kentucky Delphi Project.
         Prepared by  the Urban Studies Center^University
         of Louisville, Louisville,  1976.

         Bureau of  the Budget. Illinois Population Projections
          (Revised,  1976).  Springfield, Illinois. 1976.
                             II-B-2-17

-------
                TABLE II-B-2-3




POPULATION  ESTIMATES AND PROJECTIONS FOR




     INDIANA  COUNTIES  (ORBES  REGION)






            1975,  1985 and 2000





                         POPULATION
FIPS
Code
001
003
005
007
009
Oil
013
OT5
017
019
021
023
025
027
029
031
035
037
041
043
045
047
049
051
053
County
Name
Adams
Allen
Bartholomew
Beriton
Blackford
Boone
Brown
'Carroll
Cass
Clark
Clay
Clinton
Crawford
Daviess
Dearborn
Decatur
Delaware
Dubois
Fayette
Floyd
Fountain
Franklin
Fulton
Gibson
Grant
1975
27,300
290,600
59,400
10,700
15,900
32,300
9,700
17,600
39,600
82,900
24,300
30,200
8,600
25,700
31,200
23,300
129,200
31,800
27,600
56,200
18,300
17,400
17,200
31,300
84,700
198
KDP
28,642
304,889
62,321
11,226
16,683
33,888
10,177
18, "4 6 5
41,547
86,976
25,495
31,685
9,023
26,964
32,734
24,446
135,553
33,364
28,957
58,963
19,200
18,465
18,046
32,840
88,865
5
ORBES
29,500
326,100
66,100
10,700
16,900
35,500
11,700
20/000
41,600
95,900
25,200
31,300
9,900
27,400
33,800
25,400
139,000
34,900
30,600
63,500
19,300
19,100
19,700
33,800
88,700
2000
KDP
30,781
327,651
66,973
12,064
17,927
36,418
10,937
19,844
44,649
93,469
27,398
34,050
9,696
28,977
35,178
26,271
145,673
35,854
31,119
63,365
20,633
19,618
19,393
35,291
95,499
ORBES
32,800
384,900
76,200
10,200
18,100
41,900
15,800
23,200
42,500
119,100
28,000
31,300
11,100
27,500'
37,800
30,000
144,100
39,500
34,600
70,100
19,600
21,400
22,000
36,400
92,000
                II-B-2-18

-------
TABLE  II-B-2-3 (continued)
                POPULATION
FIPS
Code
055
057
059
061
063
065
067
069
071
073
075
077
079
081
083
085
093
09,5
097
099
101
103
105
107
109
113
115
117
119
121
County
Name
Greene
Hamilton
Hancock
Harrison
Hendricks
Henry
Howard
Huntington
Jackson
Jasper
Jay
Jefferson
Jennings
Johnson
Knox
Kosciusko
Lawrence
Madison
Marion
Marshall
Martin
Miami
Monroe
Montgomery
Morgan
Noble
Ohio
Orange
Owen
Parke
1975
28,100
68,300
40,000
23,200
61,000
53,700
87,300
34,900
34,200
22,900
24,200
27,400
20,700
70,100
39,800
52,300
40,000
138,100
789,000
37,900
11,100
39,900
90,800 -
34,400
47,900
32,400
4,600
17,200
13,200
15,500
198
KDP
29,482
71,658
43,226
24,760
63,999
56,340
91,593
36,616
35,882
24,026
25,390
28,747
21,718
•73,547
41,757
54,872
41,967
144,890
827,796
40,393
11,646
41,862
95,265
36,091
50,255
33,993
4,826
18,046
13,849
16,262
5
ORBES
31,100
98,200
56,500
30,400
77,800
55,400
103,500
38,700
37,300
28,000
25,700
.30,400
23,800
90,400
39,400
63,700
47,000
145,100
824,200
45,600
11,100
42,900
99,600
35,900
57,300
37,700
5,600
. 18,300
15,600
17,800
2000
KDP
31,683
77,008
45,100
26,158
68,777
60,547
98,430
39,350
38,560
25,820
27,285
30,893
23,339
79,038
44,874
58,968
45,100
155,707
889,595
42,732
12,515
44,987
102,377
38,786
54,007
36,531
5,185
19,393
14,882
17,476
ORBES
35,600
170,000
86,100
41,300
110,700
58,400
124,200
42,100
41,200
34,500
27,900
.34,200
29,200
132,100
36,600
82,100
53,800
152,100
865,800
55,700
10,700
43,600
113,100
37/600
74,800
46,100
7,200
18,900
19,900
21,000
      II-B-2-19

-------
TABLE II-B-2-3 (continued)




               POPULATION
FIPS
Code
123
125
129
131
133
135
137
139
143
145
147
149
153
157
159
161
163
165
167
169
171
173
175
177
County
Name
Perry
Pike
Posey
Pulaski
Putnam
Randolph
Ripley
Rush
Scott
Shelby
Spencer
Starke
Sullivan
Tippecanoe
Tipton
Union
Vanderburgh
Vermillion
Vigo
Wabash
Warren
Warrick
Washington
Wayne
1975
18,600
12,000
22,700
12,800
27,500
29,100
22,600
20,200
18,800
38,800
. 17,300
20,500
19,600
112,800
16,400
6,600
163,000
16,700
110,500
35,400
7,900
33,600
20,000
76,800
198
KDP
19,595
12,590
23,816
13,429
28,852
30,531
23,711
21,193
19,724
40,708
18,151
21,508
20,564
118,346
17,206
6,925
171,015
17,521
115,933
37,141
8,288
35,252
20,983
80,576
5
ORBES
19,400
12,000
26,200
13,600
29,400
31,400
25,200
22,600
22,500
43,400
19,600
24,100
18,500
7,100
15,000
7,000
156,900
18,200
111,400
36,000
8,600
48,600
21,400
81,200
200.0
KDP .
20,971
13,530
25,594
14,432
31,006
32f810
25,481
22,775
21,197
43,747
19,506
23,114
22,099
127,182
18,491
7,441
183,782
18,829
124,588
39,913
8,907
37,884
22,550
86,592
ORBES
20,200
11,900
32,400
14,600
30,400
34,400
29,700
24,900
27,100
49,700
22,600
28,100
17,300
7,400
14,500
7,400
144,400
21,200
104,000-
36,500
8,800
77,9-00
24,000
82,400
     II-B-2-20

-------
                        TABLE II-B-2-3 (continued)
                                      POPULATION
                        1975
1985
2000
FIPS
Code
179
181
183
County
Name
Wells
White
Whitley

24,800
21,400
24,700
KDP
26,019
22,452
25,915
ORBES
28,500
24,300
29,600
KDP
27,962
24,128
27,849
ORBES
' 34,400
27,400
38,200
SOURCE;    U. S. Census Bureau.   Current Population Reports.
          Series P-26, No.  75-14.  Washington,  D.C. 1976

          University of Louisville.  Kentucky Delphi
          Project.  Prepared  by  the  Urban Studies Center,
          University of Louisville,  Louisville,  1976.

          State Board of Health.   Indiana County Population.
          .Projections 197.5-.2OOP.   Division of  Research,
          School of Business.  Indiana  University,
          Indiapolis, 1976.
                             II-B-2-21

-------
                TABLE II-B-2-4




POPULATION  ESTIMATES AND PROJECTIONS  FOR




  .  KENTUCKY.  COUNTIES (ORBES REGION)






                        POPULATION
FIPS
Code
001
003
005
007
009
Oil
013
015
017
019
021
023
025
027
029
031
033
035
037
039
041
043
045
047
049

County
Name
Adair
Allen
Anderson
Ballard
Barren
Bath
Bell
Boone
Bourbon
Boyd
Boyle
Bracken
Breathitt
Breckenridge
Bullitt
Butler
Caldwell
Galloway
Campbell
Carlisle
Carroll
Carter
Casey
Christian
Clark

1975
14,400
13,600
10,800
8,400
30,700
9,300
32,800
37,100
18,900
52,300
22,800
7,400
15,700
15,100
33,500
10,100
13,500
29,100
85,000
5,600
8,600
21,700
14,100
69,800
26,400

KDP
15,914
15,030
11,936
9,283
33,929
10,278
36,250
41,002
. 20,888
57,800
25,198
8,178
17,351
16,688
37,023
11,162
14,920
32,160
93,940
6,189
9,504
23,982
15,583
77,141
29,177
II-B-2-22
1985
ORBES
12,679
12,628
10,714
9,363
33,016
10,710
35,565
44,190
21,062
54,264
23,456
8,255
15,722
17,375
47,480
11,612
16,334
27,028
99,900
5,803
9,548
23,385
13,714
71,202
29,094

2000
KDP
18,490
17,463
13,867
10,786
39,420
11,941
42,116
47,637
24,268
67,155
29,276
9,502
20,159
19,389
43,015
12,969
17,334
37,365
109,142
7,191
11,043
27,863
18,105
89,625
33,898

ORBES
12,238
13,041
12,095
10,043
36,722
11,856
44,228
54,220
21,987
57,220
25,461
9,889
17,448
19,099
77,188
13,790
21,322
25,258
108,430
5,902
10,458
28,354
14,525
84,534
33,179


-------
TABLE II-B-2-4 (continued).




              POPULATION
FIPS
Code
051
053
055
057
059
061
063
065
067
069
071
073
075
077
079
081
083
085
087
089
091
093
095
097
099
101
103
105
107
109

County
Name
Clay
Clinton
Crittenden
Cumberland
Daviess
Edmonson
Elliott
Estill
Fayette
Fleming
Floyd
'Fran'klin
Fulton
Gallatin
Garrard
Grant
Graves
Grayson
Green
Greenup
Hancock
Hard in
Harlan
Harrison
Hart
Henderson
Henry
Hickman
Hopkins
Jackson

1975
20,900
8,600
9,000
6,800
81,200
9,500
5,700
13,300
189,700
12,000
40,100
37,300
9,500
4,400
10,000
11,700
32,300
18,200
10,800
33,800
7,400
72,000
39,800
14,600
14,700
36,900
11,500
6,500
42,900
10,400

198
KDP
23,098
9,504
9,947
7,515
89,740
10,499
6,299
14,699
209,651
13,262
44,317
41,223
10,499
4,863
10,747
12,931
35,697
20,114
11,936
37,355
8,178
79,572
43,986
16,135
16,106
40,781
12,709
7,184
47,412
11,604 '
II-B-2-23
5
ORBES
23,187
8,575
9,839
. 6,739
96,722
10,083
6,609
14,314
210,539
12,160
44,879
39,999
12,317
4,829
10,388
11,322
35,443
20,146
11,407
36,233
8,488
102,838
43,534
16,973
16,204
42,268
12,110
7,132
44,882
12,713

2000
KDP
26,836
11,043
11,556
8,731
104,263
12-, 198
7:,319
17,078 .
243,580
15,408
51,489
47,894
12,198
5,650
12,840
15,023
4 41,474
V23,369
13,867
43,400
9,502
92,450
51,104
18,747
18,875
47,381
14,766
8,346
55,085
13,482

ORBES
31,103
9,302
11,682
6,409
121,830
11,027
7,097
16,442
263,400
12,610
49,853
45,460
15,130
5,595
10,740
12,596
,38,969
23,545
11,934
38,071
•; 9,585
128,378
54,642
19,884
17,333
51,171
13,491
7,834-
54,442
16,710


-------
TABLE II-B-2-4  (continued)




               POPULATION
FIPS
Code
111
113
115
117
119
121
123
125
127
129
131
133
135
137
139
141
143
145
147
149
151
153
155
157
159
161
163
165
167
169
County
Name
Jefferson
Jessamine
Johnson
Kenton
Knott
Knox
Larue
Laurel
Lawrence
Lee
Leslie
Letcher
Lewis
Lincoln
Livingston
Logan
Lyon
McCracken
McCreary
McLean
Madison
Magof f in
Marion
Marshall
Martin
Mason
Meade
Menifee
Mercer
Metcalf e
1975
700,700
22,100
20,500
130,500
16,800
26,300
11,600
31,300
12,100
7,000
12,500
26,600
12,700
17,700
8,700
22,100
5,900
60,300
14,300
10,200
47,400
11,400
16,600
22,300
10,800
16,800
17,800
4,400
17,600
8,400
KDP
774,393
24,424
22,656
144,943
18,567
29,066
12,820
34,592
13,373
7,736
13,815
29,398
14,036
19,562
9,615
24,424
6,521
66,642
15,804
11,273
52,385
12,599
18,356
24,645
11,936
18,567
19,672
4,863
19,451
9,283
1985
ORBES
809,394
26,059
22,523
142,170
16,792
26,394
12,208
29,943
12,319
8,458
14,139
31,384
13,809
18,207
8,569
24,716
6,261
64,126
14,489
10,333
45,758
12,292
20,034
23,178
13,519
17,616
19,817
4,790
19,719
8,053
2000
KDP
399,716
28,377
26,323
167,565
21,572
33,770
14,895
40,190.
15,537
8,988
.16,050
34,155
16,307
22,727
11,171
28,377
7,576
77,427
18,362
13,097
60,863
14,638
21 ,315
23, 634
13,867
21,572
22,856
5,650
22,599
10,786
ORBES
949,491
34,679
24,076
146,840
20,958
31,692
.12,569
33,267
14,864
11,717
17,992
43,392
15,284
20,279
9,163
27,342
6,455
67,068'
17,333
12,043
46,019
15,762
23,360
26,785
20,797
17,948
19,794
6,009
23,055
7,815
     II-B-2-24

-------
TABLE II-B-2-4  (continued)




               POPULATION
FIPS
Code
171
173
175
177
179
181
183
185
187
189
191
1'93
195
197
199
201
203
205
207
209
211
213
215
217
219
221
223
225
227
229
County
Name
Monroe
Montgomery
Morgan
Muhlenberg
Nelson
Nicholas
Ohio
Oldham
Owen
Owsley
Pendleton
Perry
Pike
Powell
Pulaski
Robertson
Rockcastle
Rowan
Russell
Scott
Shelby
Simpson
Spencer
Taylor
Todd
Trigg
Trimble
Union
Warren
Washington
1975
12,100
17,200
10,500
30,300
24,400
6,800
20,000
18,400
7,900
5,200
10,400
28,000
68,800
8,600
40,300
2,300
12,800
17,100
11,500
18,900
19,700
14,100
5,700
18,200
11,000
9,000
5,600
16,400
62,400
10,400
I
1985
KDP
13,373
19,009
11,604
33,487
26,966
7,515
22,103
20,335
8,731
5,747
11,494
30,945
76,036
9,504
44,538
2,542
14,146
18,898
12,709
20,888
21,772
15,583
6,299
20,114
12,157
9,947
6,189
18,125
68,963
11,494'
I-B-2-25
ORBES
11,195
19,362
11,812
33,953
28,502
6,784
21,970
22,506
7,839
5,634
11,080
32/683
74,063
9,663
42,471
2,268
14,653
19,657
11,441
23,260
20,285
15,104
6,267
19,263
12,392
9,253
5,944
17,391
74,140
12,090
2000
KDP
15,537
22,085
13,482
38,906
31,330
8,731
25,681
23,626
10,144
6,677
13,354
35,953
88,341
11,043
51,746
2,953
16,436
21,957
14,766
24,268
25,295
18,105
7,319
23,369
14,124
11,556
7,191
21,058
80,123
13,354
ORBES
10,360
25,003
14,658
43,103
34,760
7,116
26,101
28,372
7,561
6,085
12,585
45,242
88,620
11,945
45,803
2,374
17,673
17,808-
12,602
25,913
20,570
17,594
6,879
20,187
14,332
9,320
6,657
20,443
95,304
13,051

-------
                         TABLE II-B-2-4  (continued)

                                      POPULATION
                        1975             1985                  2000
FIPS
Code
231
233
235
237
239
County
Name
Wayne
Webster
Whitley
Wolfe
Woodford

15,600
14,100
28,400
6,100
16,600
KDP
17,241
15,583
31,387
6,742
18,356
ORBES
15,627
15,482
26,178
6,486
16,675
KDP
20,031
18,105
36,466
7,833
21,315
ORBES
17,362
18,989
29,007
7,258
17,608
SOURCE:  U. S. Census  Bureau.   Current Population Reports.
         Series P-26,  No.  75-17,  Washington, D.C. 1976

         University  of Louisville.  Kentucky Delphi Project
         -Prepared by the 'Urban :Studies Gen.ter., University
         of Louisville, Louisville,  1976.

         Kentucky Department of Transportation.  Revised
         Battelle Projections.  Frankfort, 1976.
                             II-B-2-26

-------
                               TABLE II-B-2-5

               POPULATION ESTIMATES  AND PROJECTIONS FOR

                . .    OHIO COUNTIES  (ORBES REGION)

                            1975,  1985 and 2000


                                       POPULATION
FIPS
Code
001
003
005
007
009
Oil
013
015
017
019
021
023
025
027
029
031
033
037
041
045
County
Name
Adams
Allen
Ashland
Ashtabula
Athens
Auglaize
Belmont
Brown
Butler
Carroll
Champaign
Clark
Clermont
Clinton
Columbiana
Coshocton
Crawford
Darke
Delaware
Fairf ield
1975
22,500
109,300
43,600
102,000
51,500
41,500
82,300
29,900
244,100
24,100
32,300
154,900
108,000
32,600
111,800
35,000
50,000
55,000
50,800
84,900
198
KDP
23,536
114,331
45,607
106,695
53,870
. 43,410
86,088
31,276
255,335
25,209
33,787
152,030
112,971
34,101
116,946
36,611
52,301
57,532
53,138
88,808
5
ORBES
21,004
124,269
48,555
124,013
50,177
43,861
86,447
30,860
271,410
23,927
36,297
186,659
121,677
35,417
124,006
36,924
58,206
56,877
49,778
29,324
2000
»
• KDP ORBES*
25,179
122,315
48,792
114,145
57,632
46,442
92,100
33,460
273,166
26,970
36,146
173,344
120,860
36,482
125,112
39,166
55,954
"61,549
56,849
95,009
* population projections  for  2000  were not provided for Ohio
  counties during the initial collection of data.
                              II-B-2-27

-------
TABLE II-B-2-5 (continued)




                 POPULATION
FIPS
Code
047
049
053
055
057
059
061
065
067
071
073
075
079
081
083
087
089
091
097
099
101
103
105
107
109
County
Name
Fayette
Franklin
Gallia
Geauga
Greene
Guernsey
Hamilton
Hardin
Harrison
Highland
Hocking
Holmes
Jackson
Jefferson
Knox
Lawrence
Licking
Logan
Madison
Mahoning
Marion
Median
Meigs
Mercer
Miami
1975
26,400
866,100
28,100
67,300
125,700
39,700
905,000
31,900
17,900
31,500
22,200
25,300
28,900
94,400
43,700
60,400
114,000
37,400
31,400
307,100
67,600
98,600
21,300
37,600
87,300
KDP
27,615
905,965
29,393
70,398
131,486
41,527
946,655
33,368
18,724
32,950
23,222
26,465
30,230
98,745
45,711
63,108
119,247
39,121
32,845
321,235
70,712
103,138
22,280
39,331
91,318
1985
ORBES
28,409
983,526
26,234
76,328
166,634
43,978
1,048,534
32,020
18,310
33,366
23,924
29,053
30,453
108,564
43,581
62,870
130,919
41,385
31,972
325,301
73,562
97,266
22,533
39,571
103,662
2000
KDP ORBES*
29,544
969,228
31,446
75,314
. 140,667
44,427
1,012,760
35,698
20,031
35,251
24,843
28,313
32,341
105,640
48,904
67,592
127,574
41,853
35,139
343,667
75,649
110,341
23,836
42,077
97,695
     II-B-2-28

-------
               TABLE II-B-2-5 (continued)


                             POPULATION
FIPS
Code
111
113
115
117
119
121
127
129
131
133
135
139
141
145
149
151
153
155
157
159
163
165
167
169
175
SOURCE


County
Name
Monroe
Montgomery
Morgan
Morrow
Muskingem
Noble
Perry
Pickaway
Pike
Portage
Ereble
Richland
Ross
Scioto
Shelby
Stark
Summit
Trumbull
Tuscarawas
Union
Vinton
Warren
Washington
Wayne
Wyandot
1975
15,600
558,000
13,500
24,500
80,200
11,100
28,500
43,800
20,500
132,900
35,900
130,400
60,800
80,800
40,200
384,200
535,300
241,300
80,300
28,800
10,300
87,700
59,500
91,000
22,300
: U. S. Census Bureau.
Series P-26, No. 75-3
University
r»w^«~~,«,3 i~ - ,
1985
KDP OP.BES
16,318 18,403
615,064 719,767
14,121 13,462
25,628 24,796
83,891 84,223
11,611 11,368
29,812 31,274
45,816 47,183
21,444 22,508
139,017 152,770
37,552 39,981
136,402 151,895
63,599 70,903
84,519 83,272
42,050 45,957
401,884 437,610
559,939 629,577
252,406 270,482
83,996 86,160
30,126 28,497
10,774 10,435
91,737 111,040
62,239 60,675
95,189 105,441
23,326 24,234
2000
KDP ORBES*
17,458
658,014
15,107
27,417
89,750
12,422
31,894
49,015
22,941
148,725
40,175
145,927
68,040
90,421
44,987
429,948
599,039
270,032 •
89,862
32,229
11,526
98,143
66,585
101,836
24,955
Current Population Reports.
5. Washington, D.C.,
of Louisville. Kentucky Delphi
1976
Project.
Louisville, 1976.

State of Ohio.  Population Projections. Department of Economic
and Community Development.  Columbus, 1975.
                   II-B-2-29

-------
2.3.  ANALYSIS OF SCIENTIFIC METHODOLOGIES USED IN TECHNOLOGY ASSESSMENT -
      PRELIMINARY REVIEW OF RTC's

           The validity of any scientific or technological assessment is
usually directly proportional to the scrutiny with which the assessment is
subjected to a methodological critique.  In fact, one might argue that
good science is characterized more by critical, methodological insights
than by a particular content.  If this is true, then the validity of any
assessment of the Regional Technology Configurations of the Ohio River
Basin Energy Study is directly proportional to the scrutiny with which
the assessment itself is subjected to a methodological critique.  The
purpose of this chapter is to provide just such an analysis of the
ORBES assessment.

           To.evaluate the methodology of this assessment, however, requires
that one ask some critical questions about issues not solely "scientific."
Such questions do have a precedent, however, in the writings of some of
the greatest scientists the world has known.  Reading Albert Einstein's
numerous articles, lectures, and books soon convinces one of the breadth
of his scientific comprehension of areas as diverse as physics, religion,
philosophy, and political science, and the depth of his commitment,
not only to the study of quantum theory and relativity, but also to the
search for worldwide nuclear disarmament, civil libertarian government,
and world peace.  Perhaps the magnitude of his greatness is due to his
often-expressed belief that a great scientist could never be merely that,
but must also be a great critic, humanist, or philosopher.  Otherwise,
Einstein '.thought,-a "scientist could never understand -why his theories were
correct methodologically speaking, or whether his assumptions were
desirable, ethically speaking.   He was painfully aware of the ethical
responsibilities of scientists, not only because he knew of tne potential
for harm of nuclear energy, but also because it was his colleagues in the
Prussian Academy of Science who, in the early thirties, condemned him
publicly for his ethical stance in opposition to Hitler's flagrant
violations of citizens' civil liberties.

           Before evaluating the methodology used in the ORBES assessment,
including the critical logical, scientific, and ethical assumptions of the
study, it would be well to review the key components of the four basic
ORBES Regional  Technology Configurations.  In this way, one can obtain a
clear perspective on what is being analyzed.

           In part because of the Technology Assessment Act of 1972, the
present ORBES study has been directed by EPA to assess the impact of
energy conversion facilities, both existing and probable, on the American
Midwest.   After reviewing various projections of U.S. energy proudction
made in recent years, most ORBES group members agreed that their technology
study should focus on analyses  of impacts of the extreme (low growth and
high growth) energy scenarios.   Although the occurrence of these extremes
is unlikely, it was felt that the impact of actual energy growth would lie
somewhere between that calculated for the two extremes, and that (by basing
the studies on these two) one would have described the "boundary conditions"
for such an assessment.
                            II-B-2-30

-------
           The high-growth projection used by the ORBES group was taken
from a Bureau of Mines report (see 1), which predicted a national demand
for 163 quadrillion BTUs by the year 2000.  This amounts to a 5.8%
annual electrical growth rate.  Its low-growth counterpart was taken
from the Ford Foundation Energy Policy Report (see 2), which projected a
national demand for 117 quadrillion BTUs by the year 2000.  This repre-
sents a 2.5% annual electrical growth rate.  Alvin Weinberg's projections
are essentially the same as those of the latter group (the "Ford Tech Fix"),
although arrived at by different means (see 3).  For both the high and
low projections, the percentage of the total U. S. energy supplied by the
ORBES Region was assumed to be 11.2%.

           Based on these high-growth and low-growth projections, the
scenarios used by the ORBES group to describe the distributions of energy
conversion facilities and their characteristics are four in number:

           1)  the Bureau of Mines high-demand projection for an 80%
               coal-generated, and 20% nuclear-generated, energy mix
               for the region;

           2)  the Bureau of Mines high-demand projection for a 50%
               coal-generated, and 50% nuclear-generated, energy mix
               for the region;

           3)  the Ford Tech Fix low-demand projection for a 100%
               coal-generated energy base for the region; and
                                                                       *
           4)  the Ford Tech Fix low-demand projection for a 100%
               nuclear-generated energy base for the region.

Built into the four above scenarios are a number of assumptions made by a
majority of members of the ORBES group, viz.,  (A) that uranium and coal
would be the only fuels available in the year 2000 in sufficient quantity
to be of importance;  (B) that the 4% of electricity currently being
generated in the ORBES Region by sources other than coal or uranium is
not significant, and should not be considered;  (C) that solar, wind, and
hydroelectric power, although alv/ays available, will be limited in quantity
in the region in the year 2000;  (D) that oil would be depleted and/or too
expensive for the utilities to buy by the year 2000;  (E) that uranium and
coal will not be too expensive for the utilities to buy by the year 2000;
(F) that electrical power produced in the ORBES Region will  not be trans-
ported in significant quantity outside the region (assumption (F) will be
valid only in the event of a low-energy-demand scenario, such as (3) or
(4), since the high-energy demand scenarios will  be regional  only if an
enormous increase in population occurs); and  (G) that all currently
planned energy conversion facilities will  be completed by 1985.

           In addition to these seven- assumptions, a number of demographic
presuppositions, specific to the BOM study, to the FTF study, and to the
ORBES study, were made.   Since these specific presuppositions will  be cited,
                            II-B-2-31

-------
analyzed, and later critiques, chiefly by urban and economic demographers
on the ORBES study team, these assumptions will not be noted here, but
will be briefly analyzed, insofar as they are relevant, in the sections
immediately following this one.

           After developing the four scenarios according to high and low
energy demand projections and on the basis of the 80/20; 50/50; 100; 100
"mix" of coal and nuclear, members of the ORBES team next considered the
location of all electric energy conversion facilities.  On the basis of
the respective energy "demands" implicit in each of the scenarios, the
ORBES group outlined, on a county-by-county basis, the location of
electric energy conversion facilities in the region.  Illinois Conversion
Facilities are given in Table II-B-1-6, Kentucky Conversion Facilities
in Table II-B-1-7, Ohio Conversion Facilities in Table II-B-1-8, and
Indiana Conversion Facilities in Table II-B-1-9, Chapter 1., Section
1.2. of ORBES, Volume II-B, Preliminary Technology Assessment Report,
University of Kentucky and University of Louisville, May 15, 1977.

           Regional Technology Configuration Sitings and Plant Characteris-
tics for Kentucky, Illinois, Ohio, and Indiana, are given respectively,
in Tables II-B-1-10, II-B-1-11, II-B-1-12, and II-B-1-13 of Section 1.2.
of Chapter 1. of ORBES, Volume II-B, Preliminary Technology Assessment
Report, University of Kentucky and University of Louisville, May 15, 1977.
The locations of existing and planned electrical energy conversion
facilities have been tabulated according to generation capability and are
reported in the .ORBES Task 1 Report .(.see 4).

2.3.1.  ASSESSMENT APPROACHES AND METHODOLOGIES

           As was noted in a previous report by the ORBES ciroup (see 4,
p. 1 g/h-51), the current study has a number of weaknesses because "many
approximations and value judgments are unavoidable."  In this section of
the paper various methodological assumptions and value judgments, made
in the course of the study, will be evaluated, not only because it is
generally true that all science is value/theory-laden and hence never
wholly "objective," but also because it is true., in the case of any
particular conclusion, that its validity is only as certain as that of
the assumptions or premisses on which it is based.  Moreover, such a
methodological evaluation is made not because it is ever possible to
exclude value judgments and assumptions from one's analyses, but because
good scientific procedure demands that one realize the logical and
ethical implications of the assumptions one chooses or is forced to'make.

2.3.2  METHODOLOGICAL ANALYSIS OF THE ASSESSMENT APPROACH IN GENERAL

           There are, at least, two sorts of logical, scientific, and
ethical assumptions built into the methodology employed by the ORBES
group.  The first sort characterize the study as a whole, and acceptance
of these assumptions is implicit in consideration of all aspects of the
four Regional Technology Configurations.   The second sort of presuppositions
characterize, not the entire ORBES group procedure, but rather are hypo-
                            II-B-2-32

-------
theses accepted within the framework of at least one (but not all)
Regional Technology Configurations.  These two classes of assumptions
might be termed "general" and "particular", respectively.

2.3.3.  EVALUATION OF THE DELPHI METHOD IN GENERAL

           The most critical  assumptions of the ORBES group, with
respect to methodology in general, focus on use of the Delphi Method
(see 5, 6) throughout the study and on a whole group of premisses
upon which the four Regional  Technology Configurations are predicated.
Employed as a means to gain consensus on issues about which there is
considerable disagreement, the Delphi Method allows progress to be made,
and conclusions to be agreed to, in otherwise problematic areas.
Basically it is a means whereby most members of a group are polled,
often by means of an initial  questionnaire followed by successive ones,
in which repeated iterations of the questionnaire are amended further
and further so as to produce the specific statements with which the
greatest number of group members can agree.  As such, the Delphi  approach
is a method of aggregation; it does not include intervals of debate or
dialogue used to persuade group members on a particular issue.   Rather
it consists merely of a tabulation of responses to particular theses, and
successive modificaton of those theses so as to promote the greatest
possible degree of agreement with them in repeated tabulations.  The
effect of the method, of course, is not only to establish a "middle
ground" with which most group participants can agree, but also  to
eliminate all minority positions from consideration.

           The obvious logical fallacy involved in use of the Delphi
Method is commission of the argumenturn ad popuTurn, the appeal to the
people, also called the "fallacy of consensus."  Clearly a statement is
logically and factually valid either because of the empirical evidence
or the logical inferences on which it is based, and not because of the
number of people who support it.  The majority can always be wrong in
their opinion, and their consensus is logically irrelevant to the truth
of the statement in question.

           In the case of the approach of the ORBES study team, this
logical fallacy was committed by virtue of the use of the Delphi  Method
to determine such things as  (a) correctness of population projections,
(b) extent of future use of nuclear energy, and  (c) availability of solar
power to meet future energy needs.  Clearly questions such as (a), (b),
and (c) are ones to which empirical investigation and (in part) logical
inferences are relevant.   Valid conclusions regarding (a), (b), and (c)
could never be obtained merely by aggregating responses of study group
members, not only because using such an aggregation as a basis  from
which to draw conclusions is  logically fallacious, but also because such
a gaining of consensus presupposes that any and all opinions are  of equal
worth and scientific plausibility, and hence are to be only quantitatively
(never qualitatively) assessed.
                            M-B-2-33

-------
           A second logical fallacy, whose commission is implicit in use
of the Delphi Method, is that of begging the question, called petitio
principii.  One falls prey to this fallacy by assuming, rather than
arguing for, what one is trying to prove.  In the case of employment of
the Delphi Method by members of the ORBES group, particular questions
(such as  (a), (b), (c) above) were begged, and particular conclusions
were assumed, rather than empirically and logically substantiated.  These
conclusions were assumed, rather than empirically substantiated, because
they were based on what data ORBES group researchers had previously ob-
tained from various government agencies.  Clearly the only logical way to
"check" the empirical plausibility of such things as population projections,
or safe levels of exposure to radioactivity, is to engage in a direct,   •
experimental, data-based study in which one seeks to confirm (replicate)
or disconfirm the findings of previous researchers.   If one merely uses
old data to check the validity of old conclusions, then clearly one is
begging the critical  questions of whether the old conclusions and the
old data are valid.  For example, in the Task 1 report of the ORBES
group, instead of using good social-science methodology to obtain precise
and accurate population projection studies, some project members used
population projections from various agencies, e_.£.,  Oak Ridge Laboratories,
who clearly have a bested interest in the projections.  (See 4, lb-2).

           Even if one does not use data from sources who have vested
interests in the issue for which the data is used (and much of the ORBES
data is not from such "vested interest groups"), there is still the
problem of using previously collected data, rather than experimentally
establishing one's own.   This could involve one in a third logical fallacy,
viz., the argumentum ad verecundiam, the appeal to authority.  Clearly a
particular theory or statement is correct, not because of the authority
to whom it is ascribed or from whom it is taken, but because there are
sound empirical  and/or logical reasons for accepting it.   For example,
one does not seek scientifically to substantiate the level of magnesium
content in the Ohio River at the Kennedy Bridge by asking a group of
people to read a specially selected body of published data on magnesium
content there, and then by polling them, find the most common, therefore
most . ertain, answer.   Instead one would either perform,  or pay an
expert in the field to perform, a significant number of methodologically-
valid experiments to  determine magnesium content.  Likewise, one ought hot
to seek to engage in  ORBES group technology assessments by simplyy accepting
the data of other "authorities."  One ought to seek  to replicate that data,
as a check on it, and thus avoid the logical  fallacy of argumentum ad
verecund i am.

           When  all of these logical  fallacies were  discovered at one of
the ORBES group  meetings, it was decided that the only way to have avoided
committing them would be to have had adequate funds  to do a solidly-
based empirical  study in all  facets of the ORBES group researchers.
Such funds would be many, many times those actually  allotted and the
length of the ORBES project in time would have to be considerably longer
than it is now scheduled to be.  ORBES group members felt that it was
preferable to do even an inadequate, consensus-based, question begging
                            II-B-2-34

-------
study rather than to do none at all.  All project team members agreed as
to the inadequate funding in terms of which the group is forced to do its
work.

2.3.4.  EVALUATION OF OTHER ASSUMPTIONS OF THE APPROACH IN GENERAL:
        PROBLEMS WITH ALL FOUR SCENARIOS

           Besides the logical and scientific problems implicit in the
ORBES group assessment because of the Delphi Method, there are a number
of other methodological difficulties which characterize the study as a
whole.  These might be discussed best by dividing them into two categories:
doubtful scientific assumptions implicit in all four energy scenarios and
doubtful/undesirable ethical assumptions implicit in all scenarios.

2.3.4.1.  SCIENTIFIC ASSUMPTIONS

           There are at least six critical scientific assumptions which
seem to be suspect.   Each of these will be stated and discussed with respect
to plausibility.  .As was indicated in the introduction, one of the first
and most basic of these questionable scientific assumptions is that only
coal  and uranium will be-able to provide major sources of energy for the_
U. S. by the year 2000 (see 4, le-1).  This statement seems.questionable
cause of the lack of agreement concerning available uranium reserves.
There seems to be no doubt as to the fact that coal will be a major
source of energy, yet even pro-nuclear advocates, such as Weinberg, warn
.that--their .biggest problem is the availability of adequate supplies of high-
grade uranium:  "the single most vulnerable part of the nuclear enterprise
is the uncertainty as to the availability of uranium ore" (7, p. 183).
This problem is critical, because contemporary light-water reactors run
on enriched uranium 235, of which the supply is dwindling; low-cost,
high grade uranium will be all gone before the year 2000 (8,  pp. 146-153),
and unless something unforeseen is done, will  cause "a rapid  decline of
the nuclear industry" (9; see also 10, 11, 12, 13, 14, 15, 16).  In fact
this year American utilities began loading their reactors with 10% imported
uranium, a ratio that will escalate upward until it reaches 100% by 1984
(8, p. 149; 9, 10, 11, 12, 13, 14, 15, 16).  If in 1984, 100% of uranium
for U. S.  reactors must be imported, how can uranium be a major source of
energy in the year 2000?  Moreover, how can nuclear energy render the
U. S. energy-independent?  Instead of being dependent upon the Arab Nations,
it appears that nuclear-dominated U. S. will be 100% dependent on Britain,
France, Australia, Canada and South Africa.  Moreover, by the end of this
year, 1977, some light-water reactors may be forced to shut down because
of the unavailability of fuel and because there are no reprocessing
plants in the U. S.  (8, pp.  155-56; 17, 18, 19).  If such plants existed
and were workable, the U. S. might be able to "buy some time" in its
energy crisis; at present, however, the plants have been called "critical"
and "unworkable" (8, p. 158), and hence no solution to the problem of the
uranium shortage.

           Even European utilities have been cancelling planned reactors
because of uncertainty over future uranium supplies (20, p.  54).  Hence
                            II-B-2-35

-------
 it  is puzzling why the U. S. would gamble on future uranium reserves or the
 ability to reprocess uranium when otherwise it could be energy-independent
 through coal power.  In the face of an acknowledged shortage of uranium
 which has been admitted by ERDA, the U. S. Geological Survey, and the
 utilities, one wonders about the plausibility of the ORBES group assumption
 that uranium and coal will be able to supply major U. S. energy needs in
 the year 2000.

           A corollary to the preceding assumption is a second presupposi-
 tion central to all four energy scenarios of the ORBES report.   This is
 the assumption that non-nuclear-non-coal forms of energy will not be able
 to  help meet the energy demand by the year 2000.  This claim is doubtful
 for several reasons, the most obvious of which is ERDA's insistence that
 by the year 2020, solar energy can meet 25% of the total U. S. needs (8,
 pp. 146-47, 94; 21, p.  13); if this is possible by the year 2020, clearly
 a significant part of U. S. needs can be met by solar energy  by the
year 2000.  Moreover, according to a recent report by the Federal Energy
 Administration (22, p.  1-9), the prospects for solar energy are even more
 hopeful; solar energy can produce 25% of our present energy needs by 1990.

           Another reason why this second assumption, viz., the unavail-
 ability of nonnuclear/noncoal energy sources by the-year 2000, is
 questionable is that the relative allotments of funds to develop nuclear,
 coal, and solar energy is decided by government, and not by popular vote.
 Currently, the 1977 ERDA budget has allocated nuclear funds, as compared
 to allocations for solar funds, at a ratio of 1_8 t.p_ 1_, 1845 million
 dollars-to 116 million--dollars (23, p. 2).  This-ratio-is all the more
 disconcerting when one realizes that a National Science Foundation Report
 in 1976 said ERDA should be spending $189 million for solar d?velopment,
 and ERDA is spending only about half that amount (23, p. 1).  It seems
 that the government's earlier investments in nuclear power have paved the
way for it to continue to move in the direction of more nuclear power, and
 the ORBES group has not questioned the factual basis of this "business
as usual" assumption.  Clearly a developed solar technology is available
 (see 24; 25; 26;  27; 28, pp. 16-20; 29; 30), and estimates of the cost of
solar space heating systems range f^om $5,000 (28, p. 17) to $3,000 for
a single house (31, pp.  14-15).  These estimates sound high until one
realizes that total solar costs, averaged over a fifteen-year period,
amount to current utility rates in an average East-cost city (131, p. 15).
Moreover, the advantage of solar power is that it is available every-
where, it is clean and safe, its technology is understood and its fuel
 is "free."  Some proponents even maintain that there are three reasons
why solar power will soon be cheaper than fission power:  low maintenance
costs, infinite capacity, and location of heaviest energy demands (31,
pp. 14-15).   Even one of the staunchest nuclear supporters, Hans Bethe,
admits solar energy will be widely used in 25 years (32, p. 24).

           A third questionable assumption made throughout all four energy
 scenarios is that uranium and coal will both be inexpensive enough to use
as major sources of U.  S. energy by the year 2000 (4, p. le-2).   Whereas
                            II-B-2-36

-------
 the price of coal  has not shown an extremely dramatic rise,  this is not
 the case for uranium, and hence this scientific assumption seems doubtful,
 ERDA studies in 1975 said that uranium must cost less than $16 per pound
 to compete with power from coal plants'  the current price of uranium is
 $40 or more per pound which suggests that coal  is a cheaper  source of
 power than is uranium (8, pp.  146-47; 28, pp.  12-14, 16-18;  33, p. 44).

            Moreover, when it is assumed that both uranium and coal will
 be inexpensive enough to use by the year 2000,  one wonders,  "inexpensive
 enough for whom?"   The Alabama Public Service  Commission, for example,
 blamed its nuclear facility for the highest utility rates ever paid by
 Alabama consumers  (Birmingham Post Herald,  April  14, 1974).   Consolidated
 Edison of N. Y. faced bankruptcy because of its nuclear plant (Washington
 Star News, April  24, 1974), and the shutdown of Boston Edison's Pilgrim
 Plant cost consumers $9 million per month (Nuclear Industry, June 1974).
 Moreover power from the Vermont Yankee Plant is costing five times as
 much as the 4-mill-per-kilowatt hour originally projected.   After the
 plant is shut down to correct radiation emissions, officials estimate
 the cost will be  10 times as much to the consumer as was oricn'nally
 projected by the  pro-nuclear industry (Vermont  Press Bureau, 1973).

            Although some famous proponents  of  nuclear power  have claimed
 that when coal  and fission plants both operate  at 40% capacity, costs to
 the consumer for  power are equal  for both coal  and fission (32, p. 30),
 arguments such as  Bethe's seem to suceed only  because they ignore the
 large capital costs of nuclear plants, and  fail  to take account.of all
"•"aspects of nuclear costs.   According to G.  Cody of'Princeton,  when all
 relevant parameters are computed, coal costs the consumer only half as
 much as nuclear power (21, p.  13; 30, p.  43).   Bethe's argument fails to
 consider and compute the ne_t energy of fission  power, and instead only
 compares the cost  of coal  with an equivalent amount of uranium 235.

            Even though the above financial  details are enough  to cause
 one to question the ORBES assumption that coal  and nuclear power will
 both be inexpensive enough to  supply energy to  the U.  S.  by  the year
 2000, two remaining financial  points illustrate the situation  even more
 dramatically:  (1) Nuclear plants average 50% efficiency, and  fossil  fuel
 plants average 65% efficiency,  but nuclear  power plants  were built on
 the cost-benefit  assumption that their efficiency would  be 80% or more.
 Since fossil fuel  plants are cheaper to  build,  since nuclear plants  have
 not attained the  level  of efficiency predicted  of them,  and  since nuclear
 plants operate at  38% efficiency after 6 years,  nuclear  power  will  soon
 be too expensive  to use, if it is not already  (34, 35,  36, 37,  38, 39,
 40, 41, 42, 43, 44, 45, 46, 47, 48,  49,  50,  51,  52, 53,  54,  55,  56).
 Another problem indicating this same point  is the fact that   (2)  all
 nuclear plants  and utilities are beset with  capital  and  repair costs
 that are at least  twice as great as  predicted.   Hence it is  not sur-
 prising that 2/3  of planned nuclear plants  have been deferred  in  the
 last two years, and that costs to consumers  have risen between 28 and 67%
 in utilities owning a nuclear  power installation  (57;  58; 8, pp.  236-27;
 34, 35, 36, 37, 38, 39, 40, 41, 42,  43,  44,  45,  46, 47,  48,  49,  50,  51,
 52, 53, 54, 55, 56).
                             II-B-2-37

-------
           If the preceding remarks are accurate, then one wonders, not
only why some persons have assumed that uranium will provide a source of
inexpensive power.  Also one wonders, if clean and safe coal can supply
our energy needs for hundreds of years (59, 60, 61), then why should
one assume that a more expensive source of power (uranium) will be a major
supplier of energy needs by the year 2000?

           A fourth scientifically questionable assumption exlicitly made
in the ORBES report is that all currently planned energy conversion
facilities slated to begin operation by 1985 will, in fact, be completed.
While the rationale for making this assumption is plausible, it seems
questionable for a number of reasons.  First, "more than 2/3 of all
previously announced nuclear plant construction projects in the U. S. .  . .
in the last two years alone," have been cancelled or postponed (57; 58;  8,
pp. 236-37).   Not only, therefore is this assumption factually doubtful,
but it seems to beg one of the very important questions addressed by the
ORBES group.   This question is whether the low-growth scenario of the
Ford Tech Fix is, in fact, accurate.  If it is, then no new power plants
would have to be built (between now and the end of the century) after
all those slated for operation in 1985 have been completed.  If it is
not, then more plants will have to be built during this time.   The assump-
tion implicitly suggests that the Ford Tech Fix is inaccurate; otherwise
it would make no sense to assume that all plants needed by the year 2000
will be built by the year 1985.  For this reason, the assumption seems to
be both factually doubtful as well as methodologically in error.  Moreover,
if the ORBES group were to conclude, after careful investigation, that the
current plan and projections followed by the U. S. government for energy
development were logically, scientifically, economically, or ethically
erroneous, then one would expect this conclusion to have some pimpact on
government energy policy.  The assumption of the ORBES group seems to be
that no policy-level impact will be felt until after 1985, in the event
that the ORBES group's conclusions do not support those of ERDA.  This
raises the issue, not only of another respect in which the fourth assump-
tion "begs the question," but also of the validity of a number of
specific energy decisions needing to be made before 1985 in the ORBES
Region.  For example, a decision will probably have to be made, before
1985, regarding the controversial Marble Hill Nuclear Plant, planned for
a site on the Ohio River 30 miles above Louisville.

           A fifth assumption, specifically made throughout the ORBES
study, is also doubtful  from a scientific point of view.   This is that
all plants will  meet all  existing pollution controls (4,  p. le-2).  This
thesis is doubtful for many reasons.  First, it has never been the case
in the past that all plants have met all  pollution controls.  Secondly,
to assume that all plants will meet such standards is to ignore the
obvious facts of human fallibility, and corruptibility, as well as to
forget the clear failures of various aspects of technology in the past.
Such a refusal to admit these failures seems not only historically
erroneous but also scientifically dangerous.  Such refusals seem to
signal a sort of Promethean madness, despite the obvious  pragmatic
                            II-B-2-38

-------
 benefits  that  such  an  assumption would mean  in terms of a simplified energy-
 scenario  model.

            For example,  how  can one assume that all plants will meet all
 pollution standars  when  33%  of all nuclear power plants inspected have
 violated  safety  rules, when  861 "abnormal events" were admitted by the
 AEC  to  have occurred in  1973, and when 371 of those events in one year
 were said by them to have  had a potential for being extremely hazardous
 to the  public  (30,  p.  5; 62; 8, pp. 36-37)?  Or, to cite another example,
 how  can one be sure that laws governing pollution control will, in fact,
 be followed?  According  to the National Environmental Policy Act which
 became  law in  1970, the  government must prepare an environmental impact
 statement for  reactor  plants before they are built (see 8, p. 72).  For
 all  practical  purposes this  law has been ignored by certain branches of
 the  government and  by  nuclear-industry proponents, and even the courts
 have not  prohibited construction of such plants before the environmental
 impact  statement had been  prepared, as in the case of the Palisades plant
 and  the Calvert  Cliffs plant (63; 64; 8, pp. 76-82).

            Another  reason why one might doubt the assumption that all
 pollution standards will be met by all plants is that, according to
 Ralph Nader, the regulatory-industrial complex has engaged in an "extended
 cover-up  of failures,  costs, risks, damage," with the effect that "moni-
 toring  breaks  down" (30, p. x).  Moreover, in assuming that all power
 plants will  meet all existing pollution standards, one forgets the v/ell-
.documented fact ,that .government .regulatory agencies do not always serve
 the  interests  of consumer  protection, but sometimes, of industry
 lobbyists.   For  example, a recent EPA document proposed to govern regula-
 tion of nuclear  power  plants has set standards for nuclear power operations
 which will  permit the  nuclear industry to deliver, to the public virtually
 any  dose  of radiation  it finds convenient for its operations.  The
 following is a direct  statement from page 69 of the EPA document:

      "A variance is proposed to permit temporary operation in the presence
      of unusual operating  conditions so as to assure the orderly delivery
      of power. . .  .  The  proposed standards are designed to govern
      regulation of  the industry under normal operation, and therefore a
      variance  is provided to be exercised by the regulatory agency, to
      accommodate unusual and temporary conditions of facility operations
      which  deviate  from such normal operation. ...  It is anticipated
      that such unusual operation will occur, at some facilities more
      often  than at  others, and that every effort will be made to minimize
      such  operation by the regulatory agency."  (65, p. S6650)

 If government  agencies are, as charged, really giving such elaborate
 "permission" for deviances, then one wonders about the validity of
 assuming, with the  ORBES group, that all  government pollution standards
 will  be met  by all  power plants.

            There is now abundant evidence that governmental regulatory
 agencies  for nuclear power, for example, have suppressed critical information
                             II-B-2-39

-------
 for the last  ten.years,  and  have  ignored  recommendations from its own key
 scientists  regarding  safety  measures  needing to be taken.  Moreover, even
 now some of this critical  information has only come to  light as a result
 of investigations  by  the New York Times,  because of documents leaked to
 the Union of  Concerned Scientists, and  documents demanded under the Freedom
 of Information  Act (66,  67).   Clearly the AEC, NRC and  ERDA have their
 own Watergate.  Carl  Hocevar,  senior  engineer at the Thermal Reactor
 Safety  Division of Aerojet Nuclear Company, which is a  government subcon-
 tractor,  resigned  in  September, 1974  and  charged that the AEC was ignoring
 safety  problems of nuclear power  plants and was maintaining an atmosphere
 of secrecy.   He charged  that,  "There  is no'way anybody  can work within the
 system  and  speak out  freely"  (68).  It  should be noted, however, that
 federal  scientists and engineers  have leaked data to the media about how
 governmental  agencies for nuclear regulation were suppressing information;
 as a result of  this evidence,  a Congressional Economic  Committee on energy
 wrote that  "the AEC had  developed a serious credibility gap. . . by
 suppressing unwelcome evidence of danger  and by demoting or firing
 researchers who have  pushed  their findings too vigorously.  In view of the
 huge federal  investment  in nuclear plants, Congress might want to in-
 vestigate the extent  of  danger in nuclear plants" (8, pp. 81, 71-72,
 78-84).   As a matter  of  fact,  the AEC became the NRC and ERDA in January
 1975 because  of "long standing criticism  and threatening court suits,"
 and so  Ford was forced to abolish the AEC (8, p. 84). which in the
 testimony of  Shaw,  the official in charge of all reactor programs, "was
 indeed  censoring research reports" (69, p. 9).

            Even if one assumes that,  human fragility notwithstanding,
 government  regulatory agencies such as  the EPA will not knowingly or
 intentionally subvert their  own pollution standards, a  problom still
 remains  with  the major assumption under consideration.  This is that
 successful  regulation of polluton depends, in part, on  industry's com-
 pliance.  Can all  power  industries be assumed to meet all pollution
 standards?  In  the case  of nuclear plants, for example, actual radiation
 levels  measured near current U. S. power  plant sites have exceeded, by a
 factor  of 50,000,  the levels "officially" reported to the EPA, and on a
 numbe-  of occasions, the presently permissible levels were exceeded for  .
 both external radiation  and  internal  doses from fission products in the
 local milk  (70, 71, 72,  73).

            Another difficulty with this same assumption, viz., that all
 plants will meet all pollution controls,  is that it implicitly pre-
 supposes  that all  existing pollution  standards will, in fact, be adequate
 to assess the health and safety effects of various forms of energy between
•now and  the year 2000.   Such a presupposition is clearly dangerous, be-
 cause almost  daily the government arrives at new conclusions regarding
 pollutants  or carcinogens such as saccharin, benzene, or various
 pesticides.   Pollution standards are  getting stricter and stricter and
 stricter, if  anything, and this trend could have a major effect on the
 scientific  accuracy of the ORBES conclusions.  For example, the ORBES
 findings  regarding nuclear power and  coal power could be vastly affected
                            II-B-2-40

-------
by  increasing knowledge of, and government regulation related to, the
dangers of radioactivity both from use of uranium and from use of coal
containing radionuclides.

           Setting accurate pollution standards is a well-known problem,
in  part, because it is difficult to obtain accurate measurements either
to  substantiate or to refute one's theories about desirable levels of
pollution.  With respect to light-water-nuclear technology, for example,
this is an especially difficult problem to solve, since there are
numerous difficulties associated with measurement of levels of radio-
activity.  Dr. John Gofman, professor emeritus of medical physics at the
University of California, pointed this out recently when he noted that
EPA neither knows, within a factor of 100,000 what containment of
Plutonium and other transuranics can be planned, nor can show, on the
basis of military experience, that containment to one part per billion
is  achievable.  Moreover the scientific methodology for estimating
levels of radioactivity i", highly problematic.  For example, "An air
monitor is placed outside a specific industrial exhaust.  If a low
activity is found in that morn''Lor, the entire industry is given a clean
bill of health with respect to its releases.   At the same time, leaky
barrels or leaky tanks elsewhere on that industrial site can be
releasing plutonium into the uncontrolled environment.  . . hence a
false optimism occurs.  Precisely this situation occurred at the Rocky
Flats Plutonium Plant.  Stack monitors will also overlook plutonium
tracked out of plants on shoes -- which has also happened" (65, p.
.S.6.650).  F.or all these reasons, claims either about the accuracy of
levels of radioactivity in particular, or about the accuracy of measured
levels of pollutants in general, seem doubtful.

           It has been found that, for example, owing to radioactive
emissions into nearby air and water, cancer and infant mortality rates
for areas within ten miles of a nuclear plant are two to three times
greater than cancer and infant mortality rates for the population out-
side this ten-mile range (70, 71, 72, 73).  Moreover, studies have also
indicated that, even in areas beyond this ten-mile range, there is a one-
to-one correspondence between higher cancer rates and higher radioactive
emissions, and between lower cancer rates and lower radioactive emissions,
for any given time peridd, for any city within 60 miles upstream of a
nuclear power plant, and for any city within  130 miles downstream of a
nuclear power plant (70; 71; 72, 73; 8, pp. 95-102).   In fact, some
scientists have warned that the annual deathrate in the U. S. could
increase by 25% owing to cancers generated by admissible radioactive
emissions, if nuclear plants were relied upon by the year 2025 to
produce electricity (8, pp. 89, 22-24).  Dr.  Linus Pauling, world-
famous chemist and Nobel laureate, calculated that levels of radiation
which the AEC permitted as acceptable (170 millirods from power plants
per year) would result in annual increases of 12,000 more children born
with gross defects, 60,000 more embryonic deaths, 2220 more cases of
leukemia, and 96,000 more cancer deaths (8, pp. 95, 106-130, 89, 95-102).
Other Nobel laureates maintain that current federal radiation standards
may cause 10% more genetic defects because of increased use of nuclear
                            II-B-2-41

-------
 power  (8,  pp.  96  ff).   Moreover,  government  standards  do  not  adequately
 distinguish  between  sol ubLe  and  i n soluble, radi ati on , and  between  natural
 radiation  and  insoluble particles;  the_se. failures_tg_make_precis.e distinc-
 tip.ns_ca.use_ ,o.ne tq_ ignore .cancer.~rd-S.ks jrom_  i nsojuFj^  p£r^fc1esand  in-
                                          -jrf  two ^o.th£cl .types^oTTa di a t To n .
           Another  example, which suggests that  there must  be  something
wrong with current  standards governing pollution in  the  power  plant  indus-
try, comes from studied  effects  of  radioactivity.  PI utonj,uni-wor_kers_in-
the DowCheimiai 1  P 1 an iijvho^ Jll&ged 1yJ]ad_SP 1
                           ^             _

               ^                       "                       °
oTTHe" same age^living  in  the same area" (74; 8,  p.  130).
 f ~  ~        ' ' ~L        ~ —
           Besides measurement difficulties associated with  levels  of
pollutants, there is another reason for  doubting the accuracy of  current
pollution standards.  This  is that scientsits  simply do not  know  all
there  is to know about  the  effects of certain  pollutants.  For  example,
one  could cite  the unknown  long-term effects of  certain pollutants  as  a
reason for arguing that, by the year 2000, pollution standards  will
certainly be  tightened.  CajTcej^^fnjmJjM^
ap^arjjT}.ywhere_f.r-om_f_i.f_ti£n_Ito thTrtZ.y-e.ars.. after  i nitia.1 exposure .
Hence, since  the government has little  idea of how  to estimate  effects
of "legal" radioactive  emissions  into air and  water in normal nuclear
plant  operation (8, p.  21;  28, pp. 5-7;  and 65), one wonders, both  how
ERDA and the  NRC can set acceptable radiation  levels, and_ how the ORBES
group  can assume that these standards will be  the ones met in the year
•2000.  Aptori,  the "success" of such standards can  only -be -determined
after  citizens  have died,  for example,  of cancer.   Not only  are there  many
scientific parameters about radioactivity, for example, which are simply
unknown  (69,  p. 6; 75,  p.  22), even on  the admission of nuclear propo-
nents, but also there is disagreement among experts as to the degree of
nuclear risk  (33, p. 26).   Either of these problems, alone,  is  enough  .
to cause one  to question the assumption  as to  the adequacy of current
pollution standards.  Moreover, for one  to admit that such standards
are  scientifically problematic because  of one's  lack of knowledge,  and then
for  one to use  these standards as a basis for  drawing conclusions about
the  desirability of various forms of energy, is  to  commit the logical
fallacy of argument urn aci ignorant Jam, the argument  from ignorance.
Given  the current uncertainty about pollution  standards, one argues
from ignorance  if one claims that these  standards provide, either for
no harm to the  public,  or  for an  accurate basis  upon which to assess the
desirability  of various energy scenarios.  Hence the assumption (that  all
plants will meet all existing pollution  controls) seems doubtful  for at
least  three reasons.  First, all  plants  have never  met all existing
pollution controls in the  past.   Secondly, there is no adequate monitoring
process able  to insure  that all controls are always met.  Thirdly,  even
if all existing pollution  controls were  met,. this does not insure an
adequate basis  on which to  evaluate future plants,  not only  because
future pollution controls  are likely to  become stricter than those  now
existing, but also because  existing pollution  controls have  often been
based  on a logically fallacious arqumentum ad  ignorantiam.
                             II-B-2-42

-------
           A sixth doubtful assumption, implicit throughout the entire
ORBES study, is that all new power plants will meet the "best available
technology environmental standards and will also use the best available
construction technology" (4, p. le-3).  This assumption is questionable
for many of the same reasons as is the previous assumption, viz., it
assumes the truth of the counterfactual condition that mistakes will
not be made in the future, despite the fact that the past history of
power technology is replete with accounts of accident, mismanagement,
corruption, and coverup.  In addition, this sixth assumption seems
doubtful for a number of other reasons.  First, there is the problem
of how "the best available" environmental safeguards and construction
technology will be chosen.  Hill the government decide what is the best
available?

           In the area of nuclear plant technology, for example, govern-
ment and industry consultants have alleged they are using the "best
available" technology.  Yet their allegedly superior technologies have
been repudiated by thousands of Nobel Prize winners and nuclear physicists
who have independently studied nuclear power.   On August 6, 1975, the
thirtieth anniversay of the atomic bombing of Hiroshima, 2300 scientists
and engineers demanded a U. S. policy of use only of non-nuclear energy
(76, p. 1).  Numerous such petitions have included the signatures of
more than ten different Nobel Prize winners (8, pp. 233-34).  Moreover,
even if all future power plants did meet the standards of the "best
available technology," this condition would not speak to the stance taken
by the 2,300 scientists and ten Nobel Prize winners, who argue that no
•available technology,' not even "the best," is safe enough to caus'e one
to consider building nuclear plants.  In other words, built into the
assumption that "the best available technology" is used, is t^e pre-
supposition that "the best technology" is a good enough standard for power
plants of a certain kind.  Perhaps "the best" plants of a certain type
are still too dangerous to be considered for full-scale implementation.

           Another difficulty with the assumption that "the best avail-
able" environmental and construction standards will be followed in all
future power plants is that this statement ignores the fact that various
utility companies, whether coal or nuclear, are able (by virtue of their
large capital)  to exert implicit and explicit force on government and
regulatory agencies.  Hence cost-beneficial arguments might be said to
outweigh "best available technology" arguments for purely pragmatic and
political reasons.   To realize the significance of this point, one need
only remember how much more money was available last year in California
to the utilities to fight Proposition 15, and how much less money
interested citizens (who opposed the utilities) were able to raise.   As
one of the Princeton scientists (who led the APS light-water reactor
study) pointed out:  "it is necessary to create a political counter-
balance to the bias of the nuclear industry and the utilities in favor
of 'business as usual'. . . . neither regulatory agencies nor congres-
sional committees can control a powerful industry in the absence of an
informed public and institutional  arrangements which given citizens'
                            II-B-2-43

-------
groups a serious hearing in the decision-making process" (69, p.  11).
The blame for employing political pressures regarding nuclear power
does not rest with utilities and the nuclear industry alone, however.
Part of the problem is also that government agencies have been
responsible both for promoting nuclear power and for regulating it.
Logically the "promoters" cannot be expected not to attempt to force
the "regulators" to agree with their emloyers1  position.  The net
result has been, in the case of nuclear plants, for example, that the
government (AEC5 ERDA, NRC) has suspended its own regulations governing
safety when it has been put under pressure by the nuclear industry (8,
pp. 76, 82; 64).  Moreover, such use of pressure politics by the  power
industry is one of the main reasons why the AEC, for example, lost its
credibility in the area of environmental  and construction standards, and
th-n was split into ERDA and the NSLC two years ago (69, p. 11).

           There are numerous examples of the ways in which industry has
pressured government into accepting less  than "the best available"
standards for power plant construction and for saving the environment.
first, the power industry has lobbied (successfully) for suppression of
studies showing that "the best available" construction and environmental
standards were not being followed.   It was not until 1973, for example,
when the AEC was threatened with a Freedom of Information suit, that it
was forced to release the government study indicating that a nuclear
accident (like the ones that began in Idaho and Detroit) could com-
pletely destroy an area the size of Pennsylvania and would cost between
$17 and $70 billion in property damages (28, pp. 5-7; 8, pp. 83-84).
Besides 'suppressing"results of studies showing environmental or con-
struction-plan hazards, the government agencies have often been persuaded
by industry to employ something less than "the best, available con-
struction standards."  For example, although the American Nuclear
Society says "nuclear power plants  are not sited in fault zones," there
are plants on faults in both Virginia and California, and the Madison,
Indiana fission plant is planned for a spot within the #2 most dangerous
fault zone in the U.S.A., a fault zone so dangerous that an earthquake
in 1811 created the famous Reelfoot Lake  (77, p. 4; 8, pp. 39, 5-7; 33,
p. 27).

           The ultimate reason for industry's pressure on government not
to require the "best available technology" for use in pollution control
and power plant construction is probably  financial.  For example, for
purely financia] reasons, the government  has not forced the nuclear
industry to make improvements in nuclear  plant design and operation
(69, p. 10), and the consumer stands in need of protection against
industry's bias (69,  p. 11).  A little known evidence of this bias is
reflected in the cancer and death .rates for workers in uranium mines and
mills, as well as in the fact that many tailings piles still remain un-
supervised, even after the hazards  of using them for building in  the
Grand Junction, Colorado, area were known (8, pp. 41-51; 30, p. 3).  For
all these reasons, it does not seem plausible to assume, as does  the ORBES
                            II-B-2-44

-------
group, that all new power plants will meet the best available techno-
logical standards with respect to the environment and with respect to
construction.

2.3.4.2.  ETHICAL ASSUMPTIONS BUILT INTO THE ENTIRE ORBES STUDY

           Besides the six technological or scientific assumptions which
seem questionable on empirical and logical grounds, there seem to be at
least four assumptions which are also central  to the entire ORBES report
but which are highly implausible on purely ethical  or philosophical
grounds.  One of the most important of these assumptions is that energy
demand reflects actual energy consumption (4,  p. le-1).  This thesis
seems questionable because it confuses what ought to be the case with
what js_ the case.  Clearly it ought to be the  case  that energy demand
represents actual energy consumption, but this is obviously not what is
the case, since energy demand in the U. S. includes a significant propor-
tion of waste.  That we waste energy, and call it "energy demand" is
evidenced, not only by the fact that with a minority of the world's
population, the U. S. has used a majority of the world's energy, but also
by the fact that the U. S. uses two to three times  the per capita energy
that the industrial European nations use (78,  p. 35).  Moreover, several
other countries, viz., Sweden, Denmark, and Switzerland, all have had and
continue to have a higher per capita GNP than  does  the U. S., although
all these countries have 1 ess than 1/2 of the  per capita energy consump-
tion of the U. S. (78, pp. 35-36; 28, pp. 16-18)7

           A second assumption central to the  ORBES approach, which also
seems questionable from an ethical point of view, is that there will be
sufficient capital available to "inance new coal and/or uranium-fueled
facilities.  There are several difficulties with this assumption.  First
of all, it does not take into account the fact that either coal power or
nuclear power may be so expensive, especially  with  adequate pollution
controls, that one of them or both of them may not  be desirable as part
of the future energy scenario of the U. S.  Rising  financial costs of
nuclear power, for example, have caused the U. S. nuclear industry to
cancel "more than 2/3 of their previously announced nuclear plant con-
struction projects in the U.  S.  .  .  . in the last two years alone" (57, .58),
Besides this, the U. S. government has indicated that uranium must cost
1 ess thaji^ $16 per pound if nuclear power is to be as inexpensive as coal
powern~8> P- 147); contrary to this 1975 ERDA  claim, however, uranium now
costs $40 per pound or more.   Hence it is questionable to assume that
uranium, for example, will in fact be cheap enough  to use in the year 2000.

           Besides all these reasons for doubting the ORBES assumption
that uranium and coal will both be cheap enough to  use by the year 2000,
there is still another problem with this assumption.  It does not address
itself to the question of "inexpensive enough  for whom?"  In other words,
the assumption not only ignores the economic fact that coal and uranium
have increased in price, but also forgets that what the power industry
calls "inexpensive" may not be what the consumer calls "inexpensive."
The ethical problem posed with this ambiguity  with  "expensive for whom"
                             II-B-2-45

-------
is whether it is moral for a giant U. S. governmental-industrial complex
to opt for more expensive nuclear power, and then to pass this cost on to
the consumer so that government and industry can get a return on their
enormous investments in, for example, nuclear power.  From government's
and industry's ethical point of view, nuclear power might be "inexpensive"
because it might afford some sort of return on their capital investments.
From the consumers' ethical point of view, nuclear power might be "expen-
sive" because it might result in higher utilities costs to the average
person.

           In other words, the ethical problem with assuming that nuclear
power, for example, will be inexpensive enough to use by the year 2000 is
that government and industry may assume this on the basis of their own
previous financial commitments and not on the basis of the financial
interests of the consumer.  As one critic put it:  "nuclear power, which
represents a centralized, energy-intensive economy requiring an enormous
capital investment and the monopolistic partnership of big business and
big government," is looking out for its own financial interests (8, p. 252).

           One obvious way in which nuclear power is already not inexpen-
sive is in cost to the taxpayers by virtue of government subsidies.  The
Price-Anderson Act, uranium enrichment, fuel reprocessing, nuclear research,
and waste disposal all represent areas in which the government has been
subsidizing the nuclear industry (78, p. 35; 30, p. 2; 69, p. 11).
Because of these subsidies, which are quite expensive for the taxpayer, the
moral problem which suggests itself is quite clear.  This is that the
government, having invested enormous sums in nuclear fission at the
request of the utilities, will find itself financially forced (because
of .:~ome "return" on its investment) to promote nuclear power, "only
because of the great investments in it and our failure to have developed
appropriate alternatives" (79, p. 4), even if these other alternatives
are less expensive to the consumer.

           A third assumption (employed throughout the ORBES group pro-
ceedings) which also seems to contain implicit ethical problems is that the
location of all projected power plants will be away from population centers'
and major metropolitan areas (4, p. le-2).  This can only be assumed  be-
cause of the negative environmental-medical effects of coal  or nuclear
plants upon human beings.  And if there is an admitted (although implicit)
assumption about the ill effects of power plants, then one wonders why it
should be assumed that such plants will be built at all, since clearly to
assume that plants will be built away from population centers also assumes
they will be built.

           Another problem with the assumption that future power plants
will be built away from metropolitan areas is that, ethically speaking,
it seems to vitiate the principles of equal rights, equal opportunities,
and equal justice under law.  It seems to assume that it is.morally
justifiable for a rural citizen to bear the ill  effects of a power plant,
but that it is not morally justifiable for an urban dweller to bear
these hazards.  Is a city dweller "more equal" than a native of the
                             II-B-2-46

-------
country?  Or, since this assumption implicitly admits that pollution and
health related dangers are much greater near a power plant, do those liv-
ing near power plants really receive equal justice under law?  Or, are
equality and justice and protection from environmental pollution only
rights to be bought by the rich, but not the poor, since the rich are
free to move away from, for example, areas of higher radioactivity?

           Besides the ethical difficulties imbedded in this assumption
about locating power plants away from areas of heavy population, there
are a number of difficulties, ethically speaking, with the whole frame-
work of ORBES assumptions regarding power-plant siting (4, pp. 1 g/h-5
and 1 g/h-6).  Although numerous excellent criteria for power-plant
siting, including community attitudes regarding the possible plant and
public acceptance of the plant, are used, the list of ORBES criteria
for power plant siting does not mention such ethical constraints as
whether a particular site will insure equal justice of nearby residents.
This omission takes one back to the ethical difficulties mentioned in
connection with previous assumptions.  Clearly a particular power plant
site may be disadvantageous, not for economic or environmental reasons
only, but also for ethical reasons.  Perhaps it would be impossible, for
example, to evacuate a certain area quickly, because of the site of a
particular nuclear power plant.  Or, perhaps existing pollutants in air
and water from other sources would mean that the addition of a power
plant, on a particular site, would increase the health danger to such
an extent that local inhabitants could not be said to be receiving equal
justice, as compared to health dangers faced by other persons.  For all
the~se reasons it seems questionable, from an ethical point of view,
whether the assumed criteria for power plant siting are exhaustive and
hence truly just, in terms of all  people affected by a particular power
plant.

2.3.5.  METHODOLOGICAL ANALYSIS OF THE ASSESSMENT APPROACH IN PARTICULAR

           Besides all the ethical, scientific, and logical difficulties
characterizing certain assumptions made throughout the ORBES group
researches, there are a number of problematic assumptions made within
each of the particular energy scenarios which are part of the ORBES
framework.  In the next section of this paper, each of the scenarios
will be methodologically evaluated with respect to those ethical,
scientific, and logical difficulties which seem unique to it.

2.3.6.  METHODOLOGICAL PROBLEMS WITH THE BOM SCENARIOS (1 and 2)

2.3.6.1.  LOGICAL PROBLEMS WITH THE BOM SCENARIOS

           Logically speaking, perhaps the greatest difficulty with the
Bureau of Mines'  Scenarios (one and two) is that these projections of
energy demand (5.8% electrical growth rate per year) are inconsistent
with several  other well-known and widely accepted studies of energy-
demand projections.   The BOM projections are inconsistent with the Ford
Tech Fix projections (2), and the projections of Alvin Weinberg (3), as
well as  with the results of the ORBES group.
                             II-B-2-47

-------
2.3.6.2.  SCIENTIFIC PROBLEMS WITH THE BOM SCENARIOS

           In order to determine empirically whether the BOM scenarios for
energy demand between now and the year 2000 are accurate, one must deter-
mine whether the BOM projection of 5.8% average annual electrical growth
rate (1) is scientifically sound.  There are several reasons why this
projection seems too high.  First of all this projection assumes an in-
crease in per capita energy demand.  This increase seems unlikely, both
because since 1974, per capita energy demand in the U. S. has not in-
creased, and because most governmental experts are convinced that this
per capita zero-growth scenario will continue into the future (20, pp. 18-
21, 48; 80, pp. 49-56).  Moreover the same sort of low-energy growth,
because of zero per capita growth, is being experienced in other countries,
especially in Europe, where energy consumption has dropped (80, p. 54).
Current energy uses in California are running at approximately 50% of
industry's projections even though California is our fastest growing
state (33, p. 6).

           Scientifically or empirically speaking, one of the reasons why
the BOM projections are too high is that the methodology used in deriving
them does not take into account the fact that higher prices for energy
might cause citizens to use less (4, p. 1 g/h-4).  Clearly, increases and
even an absence of decreases in the cost of power will mean that less
power is used.  The Federal Power Commission has repeatedly critized
industry for refusing to take "price effects" into account when making
energy demand projections (81).  .Despite the FPC's caution, the BOM
scenarios (one and two) fall into error on precisely this point, viz.,
not taking price into effect, and hence giving estimates for future
energy demand which are too high.

           The BOM 5.8% annual growth rate statistics also probably fall
into error because they were calculated on the assumption that residential -
sector uses of energy would increase between now and the year 2000 (4,
p. lc-57).   This seems to be a false assumption, both because pej^ capita
energy demand has not increased, and because constraints such as "peak-
hour pricing" have significantly reduced the amount of energy used
residentailly (82, 83).  Moreover, the BOM scenarios also seem wrongly to
assume, as the price of energy goes up, and as we move into a shortage of
fossil  fuels, that the single largest consuming sector of energy will be
waste heat from electrical generation (4, p. lc-59).  Again, if previously
cited figures about energy usage in highly industrialized European nations
are correct,  then there is reason to believe that U. S. energy production
can be accomplished much more efficiently.

           Another highly implausible "scientific" assumption made in the
BOM methodology for energy demand projection is that the breeder reactor
will be introduced (4, p. 1 g/h-4).  This is factually implausible, both
because the current U. S. presidential administration has stopped plans to
introduce the breeder, and because a number of difficult scientific problems,
regarding the breeder, have not yet been worked out, such as successful
and safe reprocessing of spent fuels for use in the breeder.
                            II-B-2-48

-------
           Yet .another highly doubtful assumption in the BOM methodology
is that solar power is not taken into account as a possible source of
energy by the year 2000 (4, p. 1 g/h-4).  For all the reasons cited in
section 3.212, it seems scientifically inaccurate to make such an assump-
tion about the unavailability of solar power, especially since ERDA has
indicated that solar energy could supply 25% of all  U.  S. electrical
energy needs by the turn of the century.

2.3.6.3.  ETHICAL PROBLEMS WITH THE BOM SCENARIOS

           One of the biggest ethical problems with scenarios (1) and (2)
of the ORBES group researches is that these BOM scenarios are predicated
on the assumption that, between now and the year 2000,  there will be no
efforts at conservation which will  reduce energy demand (4, p. 1 g/h-4).
Not only does this assumption seem factually incorrect, since pricing
will have the effect of forcing more conservation, but  also this view of
no conservation seems ethically erroneous.   It presumes that, waste and
shortages of fossil fuels notwithstanding,  Americans will not take steps
toward a more efficient use of energy, but  will continue on a course of
"business as usual,"  Ethically speaking, this course of action can only
mean that we have confused energy needs with energy demands.  Moreover,
if Americans confuse needs with demands, and make no effort to conserve,
then we will be saying, ethically,  that retaining our energy-based
luxuries is more important than preserving  the health of the planet.
Both of the BOM scenarios assume nuclear energy will be used; since all
nuclear plants increase, although to a low  degree, the  leve of radio-
activity in the-human environment,  in accepting increased levels of
radioactivity in exchange for high  energy demands (in which no effort
to conserve is made), the BOM scenarios seem to sanction a sort of
"survival of the fittest" ethic whose radioactive consequences will be
felt for hundreds of generations to come.  This ethical consequence (of
the BOM assumption regarding no increased conservation  of energy)
follows because,  both internationally (The  International Commission on
Radiation Protection) and nationally ("Code of Federal  Regulations,"
Title 10), it is accepted that "genetic and somatic effects of radiation
in humans occur in linear proportion to dose even at low exposure*;" (75,
p. 16).  If any dose of radiation is harmful, then it is ethically
questionable to cause such dosages  to increase simply because we are too
lazy not to waste energy.   And, as  was pointed out earlier, in section
2.3.4.  of this paper, to confuse energy need with energy demand is
ethically fallacious because it attempts to justify what ought to be,
in the future, with what is, in the present.

           A seond ethically-questionable assumption, unique to the BOM
scenarios, is that electrical  power may be  transported  out of the region
in which it is produced (4, p. Ic 57-lc 59).  This assumption clearly
violates principles of equal justice under  law and equal protection,
especially equal  health protection, if one  realizes  that any increase in
the level of pollutants in the environment  adversely affects the health
of those who are subjected to them.  It seems ethically unreasonable to
accept the fact that one group of people will  bear the  risks and
                             II-B-2-49

-------
 responsibilities  (.§_•£•, pollution) of energy production while another
 group  of  people enjoy  the benefits of such energy production.  Accepting
 the  BOM assumption  that energy will be transported out of the ORBES
 region thus  seems to amount to creating a class of second-class citizens,
 a minority who bear the ill effects of what is enjoyed by the majority.

 2.3.7.  METHODOLOGICAL PROBLEMS WITH THE FTP SCENARIOS (THREE AND FOUR)

 2.3.7.1.  LOGICAL PROBLEMS WITH THE FTF SCENARIOS

           The only possible logical problem with the Ford Tech Fix
 projections  of 2.5% average annual electrical demand increase is that
 they are  inconsistent  with the projections of the Bureau of Mines', just
 discussed.   This  inconsistency is only a problem, however, if it can be
 sceintifically shown that the BOM projections are correct, which seems
 unlikely, given the. preceding considerations.

 2.3.7.2.  SCIENTIFIC PROBLEMS WITH THE FTF SCENARIOS

           Although the possible logical problems with the r'TF scenarios
 (three and four)  seem  slight, are there any scientific difficulties
 with these projections?  Perhaps the only signficant scientific question
 is whether it is, in fact, correct to assume (as is done in the FTF
 methodology) that with the low energy growth scenario, there will be no
 significant  damage  to  the economy.  According to FTF calculations, the
 low energy growth scenarios (three and four) will do no great damage to
-thre 'economy-  and will result'in'less than 4% decrease in the GNP over the
 next twenty-five years (4, p. 1 c 69).  This assumption demands such
 scrutiny  because  it is an economic truism that, in the past, high energy
 demand parallels  high GNP, and vice-versa.  Moreover because citizens can
 use less  energy without substantially hurting the GNP, does not scienti-
 fically establish that they will do so.  Although justifiably, the
 methodology  of the  FTF assumes that what can be done will be done.

 2.3.7.3.  ETHICAL PROBLEMS WITH THE FTF SCENARIOS

           Perhaps  the greatest ethical assumption built into the FTF
 methodology  has to  do with the rights of government.   According to
 scenarios three and four, prices and government policies will in the
 future encourage greater energy efficiency (4, p. 1 g/h-4).  While this
 statement seems true, from an economic point of view, some might question
 it from an ethical  point of view.  Does government have the right,
 ethically speaking  to develop policies that deliver control of energy
 usage  away from industry, and into the hands of regulatory agencies?

           From a free-enterprise, laissez-faire ethical perspective, one
 might  argue  that power companies have a right to create energy demand as
 well as to meet it.  Moreover, one might argue that government control of
 pricing,  so  as to insure more energy conservation, would have a harmful
 effect not only on  free enterprise but also on the rights of the poor to
 meet their energy needs.   Presumably, with higher energy costs, it will
                             II-B-2-50

-------
be the poor, and not the rich, who are forced to conserve.  However, this
need not be a direct consequence of the FTP assumption regarding govern-
ment pricing and policy-making for the power industry.  It could also be
argued that government could control pricing (and consequent energy demand)
by charging higher energy users more or by providing a sliding-scale
system of rates for energy usage.  Nevertheless, the key ethical problem
posed by the FTF's assumption, of government's right to control energy
consumption, is that of balancing individual rights with group rights.

2.3.8.  METHOLOGICAL PROBLEMS WITH THE NUCLEAR SCENARIOS (ONE, TWO, AND
        FOUR)

2.3.8.1.  LOGICAL PROBLEMS WITH THE NUCLEAR SCENARIOS

           As was indicated previously, the argumentum ad ignorantiam is
committed whenever one argues on the basis of his ignorance about certain
facts to a conclusion regarding those facts.  For example, it was pointed
out earlier that one cannot argue that a certain level of radioactivity
is safe when one's only grounds for so arguing are that one has no clear
knowledge of the harmful effects of radiation.

           In the case of scenarios (1), (2), and (4) of the ORBES group
research, these nuclear scenarios all  assume, methodologically speaking,
that nuclear power will continue to be used between now and the year 2000,
and that it is safe.  One of the chief reasons why these plants are said
to be safe is that the probability of a core meltdown has been calculated,
by government physicists, to be very small.  Hence, with no occurrence of
core meltdown, it is argued, nuclear plants will not endanger the public.
Such an argument, central to all the nuclear scenarios under consideration,
seems to commit the fallacy of argumentum ad •ignorantiam, however, because
the emergency core cooling system (which would protect against a core melt-
down) has never been tested on a standard-size nuclear fission plant.
Moreover, all scale-model tests of the emergency-core cooling system for
light-water reactors have failed (8, pp. 79-81; 69, pp. 7-9).  In the
face of this absence of accurate scientific data regarding the ability
to protect against core meltdown, o-^.e who argues that core meltdown will
not occur seems to be guilty of an argumentum ad ignorantiam.

2.3.8.2.  SCIENTIFIC PROBLEMS WITH THE NUCLEAR SCENARIOS

           With respect to scientific methodology, perhaps the greatest
problem with all the nuclear scenarios (one, two, and four) of the ORBES
group is the assumption that nuclear power is safe and that it will
supply at least 20% of the total U.  S. energy needs by the year 2000.
The main difficulty with this assumption is that the only allegedly
complete study of nuclear power safety (84), the Rasmussen Report
(known as WASH 1400) has (in both its original and amended form) been
condemned for methodological and scientific errors by the scientific
community.  The American Physical Society, the most prestigious group of
scientists in the world, criticized the Rasmussen Report on two major
                            II-B-2-51

-------
grounds:   (1) the mathematical calculations of the accident probabilities
are invalid owing to the fact that a number of important parameters were
ignored  (chance of accidents in transportation of radioactive materials;
long term  genetic effects of radiation on workers and residents even if no
"accident" occurred; disposal of radioactive waste; risks of sabotage,
theft and  terrorism; accidents in fuel reprocessing plants, uranium mines
and mine waste disposal sites); and  (2) false scientific assumptions
invalidated claims about the long-term health effects of nuclear fission
plants.  For example, the APS criticized the Rasmussen Report for assuming
that persons downwind from a reactor would receive radiation only for
one day after an accident, whereas in reality the radioactivity would
irradiate  people over a large area for an extended period.  "Correction
of this one error alone" would increase lethal cancers and genetic
defects by a factor of 25.  Because of these mathematical and conceptual
problems,  as well as "ommissions in the calculations," the computations
of accident risks and related health problems are wrong by several orders
of magnitude (78, p. 3.3; 8, p. 23; 33, pp. 30-31).

           The EPA also condemnad the Rasmussen Report for methodological,
scientific and mathematical errors and concluded that the Report's
calculations of casualties from nuclear power plants were ten times as
great as admitted (8, p. 31).  Numerous other scientific critiques, in-
cluding one done by a group of scientists from MIT, condemned the report
on scientific grounds and argued that a correct use of mathematics would
place the  number of major power plant accidents at a level 100 times
greater than that calculated in the report (8, p. 23; 85; 86).   In short,
the scientific community, the EPA,_APS and numerous other groups of experts
have found the mathematics and scTentif ic JBg^hgjg\°?JL-°f--^-ne QjiJx reactor
                 " "88).  Hence it seems questionable whether, from a
scientific point of view, nuclear fission plants can truly be said to be
safe, as is assumed within all three of the ORBES nuclear scenarios.

           The nuclear scenarios also seem to be on (questionable grounds
by virtue of the assumption that although nuclear accidents have occurred
in the past, they will not occur in the future.  Within the three nuclear
scenarios, reactor accidents in the past have been either ignored or down-
played, as have problems with radioactive wastes (such as leaks into the
environment).  Good, empirical, scientific procedure has simply not been
followed with respect to learning from past empirical situations.  For
example, the estimating techniques employed in the Reactor Safety Study,
WASH 1400 (the Rasmussen Report) enabled the AEC and ERDA to deem
"impossible" an accident (such as the one that has already occurred in
an Alabama reactor, failure of the emergency core cooling system) that "has
already occurred."  How scientifically valid are such research methodologies
when this accident was calculated by AEC/ERDA to have a probability of one
in a billion-billion, yet when this disaster has already occurred (78, p.
32; 8, pp. 156, 164-70, 110-11, 36-40)?  Hence there is reason to believe,
contrary to the assumptions made in the nuclear-energy scenarios (one,
two, ..and four), that nuclear energy is not yet scientifically reliable
and safe.

           Another scientific problem with the nuclear scenario is that it
                              II-B-2-52

-------
"is  assumed  that  there  is an accuracte scientific base from which to assess
 the validity  of  radiation  standards.  This assumption has (in section
 2.3.4.)  already  been shown to be  false, and scientific investigations
 continue to build  a case against  this thesis  (see 89).  Moreover, even
 according to  pro-nuclear scientists, such as  Alvin Weinberg, nuclear
 energy  is somewhat questionable,  since it is  certain that, contrary to
 all  three nuclear  scenarios, if the U. S. were to supply all its future
 energy  needs  with  coal and not nuclear power, then the environment would
 be  less  polluted than  it is now,  even though  there are a substantial
 number  of nuclear  reactors in operation (3, pp. 156-57) now.  From the
 point of view of pollution, therefore, it is  questionable whether the
 three nuclear scenarios are safer than the 100% coal scenario (number
 three).

 2.3.8.3.  ETHICAL  PROBLEMS WITH THE NUCLEAR SCENARIOS

            Besides the above-mentioned scientific problems with the
 three nuclear scenarios, there are a number of ethical problems.  Perhaps
 the gravest difficulty associated with the assumption of a fell-blown
 nuclear technology by  the year 2000 is that regardless of whether one
 believes light-water reactors are safe or not, consumers ought not to be
 required to take this  risk without appropriate insurance coverage.  The
 government  has repeatedly refused to guarantee adequate insurance
 coverage in the  event  of a power  plant accident and no insurance company
 will contract with consumers to cover more than 1% of the expected damage
 in  the  event  of  a  core meltdown such as occurred in part in Idaho, Ala-
•bama, and Michigan recently.  (Parenthetically, one might ask, if
 nuclear  power is so "safe," why do industry and government refuse to
 cover more  than  a  small percentane of possible damages from nuclear
 power?   Clearly  either nuclear power is not safe, or the insurance indus-
 try and  government are missing a  chance to make a lot of money from
 nuclear  plant disaster insurance.)  The federal government, in its own
 studies  (which it  suppressed until 1973 when  it was forced to release
 these results because of the Freedom of Information Act), has admitted
 that property damage alone (not counting loss of life and injury to
 humans)  in  the event of a nuclear accident would run between $17 billion
 and $70  billion  and would contaminate an area the size of Pennsylvania
 (28, p.  6;  78, pp. 33-34; 8, pp.  11, 22, 54,  55; 30, pp.  57-58).  If
 these are the government estimates of property damage, why is the govern-
 ment limitation  (reaffirmed roughly a year ago) on the total insurance
 coverage only $560 million?  (A little mathematics reveals that this
 means that  citizens have access,  at^ mos_t, to  insurance coverage for
 roughly  3%  of their property damages.)  Why is the AEC/ERDA estimate of
 damages  thus  inconsistent with the provision  of the Price-Anderson Act
 which limit insurance coverage?   Clearly it is immoral to require a person
 to  take  such  a risk and then to refuse to give him adequate insurance
 coverage for  97% of his property  losses and for 100% of his health losses.

            Another ethical problem implicit in all of the ORBES nuclear-
 energy  scenarios,  is the right of present generations to leave radiactive
 wastes  to future generations.  One wonders, again whether such a system
                             II-B-2-53

-------
 of waste storage creates  a  problem  of  equal justice  under  law.   Is  it
 just to place this  burden on  future generations?  What  is  the morality
 of using a technology whose wastes  will  endanger generations of  people
 for thousands and thousands of years to  come?  Although one might grant
 that it is ethical  to take  a  risk upon oneself, certainly  the morality
 of imposing these risks,  on unborn  generations to come, is quite another
 question,  somewhat  like the problem of taxation without representation.
 Currently  ERDA has  no workable plan for  storing the  radioactive  wastes
 from the nuclear plant industry  (90, p.  56), and what is worse,  there
 may be r\o_ way to store them.  As George  Wald, Nobel  Prize  winning biol-
 ogist from Harvard,points out:   there  is no place on earth where v/e can
 guarantee  geographic, geological, and  political stability  far millions
 of years,  so that the radioactive wastes causing death, cancer,  and
 genetic damage will  be guaranteed not  to enter the environment for  the
 hundreds of thousands of  years when they are lethal.  The  EPA has said
 government storage  plans  for  nuclear waste are inadequate, and even
 nuclear proponents  (such  as Alvin Weinberg) question our right to make
 tons and tons of plutonium, the  most poisonous material ever known  to
 man (a millionth of a gram  will  kill a human being)  while  having no idea
 of how to  protect ourselves and  future generations from this poison (8,
 pp. 161-62, 70). Moreover  the radioactive waste which  the government
 claimed was "successfully stored" in Hanford, Washington has been leak-
 ing since  1944;  115,000 gallons  of  potent radioactive waste leaked  from
 April  to June, 1973,  went undetected for 55 days, and reached the
 Columbia River,  10  miles  away from  the site of the spill.  Water plants,
 fish,  and-wild fowl  in the  Columbia River were .contaminated with radio-
 activity from the spill (8, pp.  163-66).  The obvious ethical question
 is:  how can a society which  cannot secure waste for 10 years, do so
 for 100,000 years in  order  to protect  future generations?  As Gofman
 points out:  "What  is really  at  issue  is a moral question  -- the right
 of one generation of  humans to take upon itself the  arrogance of
 possibly compromising the earth  as  a habitable place for this and
 essentially all  future generations.  .  .  visiting cancer on this  and a
 thousand generations  to come  and the prospect of genetic deterioration
 of humans  that will  ensure  as increase in most of the common causes of
 death  in future  generations"  (8, p.  214; 33, p. 31).

            The waste  issue  and our  responsibility to future generations
 is all  more pressing  because  some radioactive wastes, such as plutoni.n
 239 must be totally isolated  for 1/2 a million years, it is so lethal
 that one tablespoon,  left in  the open  atmoaGbere^is enough—to give"
 200,000,000 people  lung cancer (30,  pp.  33V~3T-36).  Moreover 500,000
"ga~llons of hip-level  radioactive waste, in 18 known leaks, have already
 leaked into, and contaminated, the  environment .at storage  sites  (28,
 pp. 1,  3-5, 9-11).   Just  the  radioactive waste from  the weapons  testing
 program of the fifties has  been  estimated to have caused 300,000 infant
 deaths because of genetic damage from  minute amounts of radiation in
 the air and water (28, p. 4).

            Besides  all the  difficulties of storage of radioactive wastes,
 another problem  implicit  in all  three  nuclear-energy scenarios is whether
                             II-B-2-54

-------
 any pragmatic value can justify the loss  of human  life.   Even  pro  nuclear
 scientists admi^t unequivocally that any  plans  for use  of nuclear  power
 will  result in an increase in  cancer deaths.  The  only  difference  between
 pro and anti-nuclear people on this score is  the number of such  deaths
 which they claim will  result from nuclear power plants.   If pro  nuclear
 people, like Bethe, are correct in saying that  low-level  radioactivity
 released in the air and water  will  only^cause a slight  increase  in cancer
 rates,  are we justified in promoting the  nuclear industry because  of  heavy
 capital investments already made in it?  ^Ij-a return,on our capital in-
 vestments in nuclear energy worth five, or fifty,  or  five hundred  lives
 lost  because of cancer each year?'  Clearly--s,ome cancer  deaths  might be
 morally justifiable if nuclear power were the only-way  to provide  energy
 to feed, clothe, and house even greater numbers of people.   But  nuclear
 power is not the only way to do this, and not the  best,  if the ethical
 price is cancer and if, contrary to the assumptions of  the nuclear
 scenarios, uranium will not be available  after  1985 and if solar technol-
 ogy will be available by the year 2000.   The  pressing ethical  problem,  in
 the face of danger from aj_l_ levels of radioactivity,  is:   by what  ethical
 principle does one sacrifice human life for economic  gain, when  there are
 other ways to insure economic  gain without loss of human  life?

            Besides the dangers to human life  from  adoption of  the  assump-
 tions implicit in the thre^R^-e^F—s-Geaajiiji&jthere  are  other ethical
 problems that pose a thre^to civil  libertiesT^If the three  possible
 nuclear scenarios operateN^:cFffr^efi*e-44^e--ttTecurren^ framework  within
 which nuclear energy decisions are.ma.de currently, citizens will probably
 have  very little voice in what affects them.

            If nuclear risks have potential  for  hurting  citizens, then
 citizens ought to have a major voice in deciding whether  or not  to take
 those risks.   And if they do not have such a  voice, then  their rights to
 due process are being violated.   Dr.  John Gofman,  a medical  biologist
 and head of the Committee for  Nuclear Responsibility  clearly believes
 citizens do not have any power,  for he maintains:   "Nothing has  suited
 the promotional  nuclear-power  interests better  than keeping alive  the
 misconception that a decision  pro or1 con  nuclear fission  power rests  upon
 esoteric technical  arguments at so-called 'public  hearings'  where  concerned
 citizens have been led, like lambs  to the slaughter,  into the  promoter's
 arena to contest a variety of  valves, filters,  cooling  towers, and mis-
 cellaneous other items of hardware in specific  nuclear  plants'" (8, p. 214).
 As Albert Einstein once said,  "The future of  nuclear  power must  be decided
 in the  town square of America" (91, p. H1780).  Clearly citizens do not
 now have the power to make such  decisions.  Agencies  do no t_tru1y_repre-
jeatu:Jii^enj_^nterests_ since all  objections^to nuclear  power plant
 construction must be raised within the quasi-legal  format established
 by the  NRC.   Government and industry standards  cannot be  challenged at
 such  hearings, but only whether the utility will really meet such  standards.
 Hence the real problems of licensing and  approval  of  nuclear power plants
 are counted as irrelevant to the proceedings.   When,  for  example,  the
 Peach Bottom Plant was challenged because the AEC  had refused:  (1) to
 consider the cumulative effects  from low-level  radiation  from  all  three
                             II-B-2-55

-------
reactor units;. (2) to compel the utility to provide an evacuation plan,
and  (3) to consider transportation risks involved in taking radioactive
materials to and from the plant, instead of hearing these ojbections,  the
licensing board dismissed all of them as irrelevant, since government
standards did not require these questions to be answered (8, pp.  211-16).
Even the NRC, in a recent study, admitted that public utility commissions
on the state level "have little or no influence" on the safety and
reliability of nuclear reactors (8, p. 192).

           With all these problems of failure to follow regulations pro-
tecting the consumer, forced subsidy of the nuclear industry, and power-
plant hearings which are not truly open, some of the best nuclear physicists
in the country maintain that there is neither an effective means  for open
discussion of all issues nor a way for the public to make decisions in-
stead of merely' expressing opinions as to what kind of power it wants.  Un-
fortunately, concludes one study, this lack of citizen voice "is  the pattern
of nuclear power plant, siting controversies generally" (92, pp. 28-35).

           The ethical problem- associated with possible violations of civil
liberties do not extend merely to the instances when citizens might have
been denied a voice in nuclear plant siting decisions.  Rather the conse-
quences to civil  liberties, within a nucler power persepctive, extend
much farther, and include possible theft, sabotage, and terrorism.

           Recently, a 20-year-old chemistry student in college,  using
only published information, was able to correctly design an atomic bomb
in only five weeks.  This is a crucial point, since only 7 to 8 pounds
of plutonium are needed to make a sizeable bomb, and since nuclear materials
plants have repeatedly caused sensational headlines when it was discovered
that they were more than 200 pounds short of plutonium, or that an em-
ployee, concerned about lax safety regulations, had mysteriously  died.
In one instance,  13 pounds of plutonium could have been obtained  by
clearning the air filters at the plant (8, pp.  178-79).  Statistics
indicate that 2% of everything shipped in the U. S. today is pilfered,
that internal crime in business costs industry up to $30 billion  a year
in the U. S., and that organized crime has made deep inroads into the
transportation industry (8, pp. 180-81, 82-85).  Moreover, terrorist
threats on nuclear plants have already occurred; a 14-year-old science
honors student in Orlando blackmailed the city under threat of blowing
it up; guerrillas have already seized atomic plants in Argentina, and
people carrying "lethal weapons" have gotten well inside the San  Onofre
nuclear plant in California (8, pp. 170-77; 78, p. 35; 28, p. 4ff).
Even if one were unable to make a bomb from plutonium, simple dispersion
of one tablespoon of plutonium in the air could kill  hundreds of  thousands
of people (8, pp. 106-119; 30, p.  39; 93; 94).

           The whole question (of the morality of providing an environment
in which there is increased risk of terrorism,  theft of plutonium,  and
sabotage) raises  yet another ethical  problem.
these dangers
It is easy to see how the U.  S.,  in  order to  guard  against
rs in the nuclear industry,  could turn  itself into  a  garrisoned
                            II-B-2-56

-------
state.  This would, in turn, have enormous implications on freedom,
society, and privacy; in short, the social cost would be extreme be-
cause it would mean an increase of police powers and a decrease of
civil liberties (30, p. 20; 95).  For example, the Virginia Electric
Power Company has requested the state to allow VEPCO to establish its
own police force and to arrest persons anywhere in the state; VEPCO
claimed the AEC regulations required they have this authority (78,
pp. 34-35).  The consequences of nuclear power to civil liberties can
be illustrated dramatically by a question posed by a young Harvard law
student in 1975:  Suppose a nuclear bomb threat has been made and the
bomb is almost ready to go off at an undisclosed location on the highly
populous East coast.  The police have the terrorist in their company.
Should the police use whatever horrible means of torture they can de-
vise in order to attempt to save the lives of hundreds of thousands of
people (8, pp. 183, 181-185; 28, p. 1; 94)?

           Because of the serious problems created by the presence of
permanently-lethal readioactive materials, the ethical problems arising
in the three nuclear scenarior. "seem to have very far reaching conse-
quences.  For this reason, one might be tempted to ask about the
methodological-ethical desirability of the coal scenarios (one, two,
and three), and especially, of the 100% coal  scenario, number three.

2.3.9.  METHODOLOGICAL PROBLEMS WITH THE COAL SCENARIOS

2..3.9..1,  LOGICAL PROBLEMS WITH THE COAL SCENARIOS

           Insofar as all three coal scenarios of the ORBES group contain
thi assumption that coal-produced energy is clean and safe, to that
extent, these scenarios are guilty of the fallacy of argumenturn ad
ignorantiam.   As with the effects of-radioactivity from nuclear plants,
so with coal  plants, the long-term effects of their pollutants are not
known.  Hence one cannot be certain that coal power is environmentally
clean and safe.  Not only are scientists unsure of adequate standards
for sulfur dioxide emissions from coal plants, but also, they are not
able yet to determine how harmful are the effects of burning coa'is with
radionuclides.  For both these reasons, there seems to be a significant
amount of ignorance about the effects of coal-produced energy; hence it
would be logically fallacious to argue that one has certainty about the
environmental effects of coal-burning plants.

2.3.9.2.  SCIENTIFIC PROBLEMS WITH THE COAL SCENARIOS

           With reject to scientific and technological difficulties
built into the three coal scenarios, there are also a number of problems.
Besides the problems with assessing the environmental effects of sulfur
dioxide and of burning coal containing radionuclides, there are numerous
remaining scientific questions regarding methodologies for sampling and
analysis of trace elements in American coals.  Great variations exist in
trace element content in the same coal seam,  not to mention variations
from basin to basin.  Moreover, the environmental problem of what happens
to trace elements following combustion is one which is still partially
                            II-B-2-57

-------
unclear, because of inadequate information regarding pollution from these
metallic elements.  Hence, as with nuclear power, the scientific lacunae
in coal technology pose a problem for all  scenarios in which decisions
(about the desirability of obtaining energy from coal) must be made.

2.3.9.3.  ETHICAL PROBLEMS WITH THE COAL SCENARIOS

           Many of the same ethical problems arising within the nuclear
scenarios are found also within the coal scenarios, viz., insuring civil
liberties of citizens in siting decisions; ascertaining that all residents,
whether they live near a pollution-source such as a power plant or not,
have equal justice and equal  protection.  Since the ethical problems
raised by the coal scenarios  seem to be a subset of those raised by the
nuclear scenarios, I shall not repeat them here, but simply refer the
reader to sections

2.3.10.  SUMMARY AND CONCLUSIONS

           If the preceding remarks are even partially correct, then  some
of the major problems facing  one within each of the four ORBES energy
scenarios are not scientific, but philosophical.  Inasmuch as these
problems are largely ethical, they are much more difficult to resolve
than purely technical questions.  Just to have outlined a few of these
difficulties, hov/ever, is to  take a step toward their solution and
toward government recognition of them.
                            II-B-2-58

-------
                                REFERENCES

                     (Cited  In  order  of  appearance)


 1.   U.  S.  Bureau  of Mines,  United  States Energy  through  the  Year  2000--
     Revised.   Bureau of  Mines,  Department  of  the Interior, December,  1975.

 2.   Ford  Foundation Energy  Policy  Project  Team,  A Time to Choose;
     America's  Energy Future.   Ballinger Publishing  Company,  Cambridge,
     Massachusetts,  1974.

 3.   Allen L. Hammond,  "Nuclear Moratorium:  Study Claims That  Effects
     Would Be Modest, Foresees  Low  Growth Rate for Total  Energy Demand,"
     Science 195  (1977);  156-157.   (This article  is  a  review  of Alvin
     Weinberg's three-volume tome,  Economic and Environmental  Implications
     of_ ^ U_. SK Nuclear Moratorium.   Oake Ridge,  Tennessee: Institute  for
     Energy Analysis, Oak Ridge Associated  Universities,  1976.

 4.   ORBES, Task  1,  "Development of Plausible  Future Regional Technology
     Configurations," October 18, 1976.

 5.   Norman Dalky,  "Delphi Method:  An Experimental  Study of  Group
     Opinion,"  Rand  Corporation, 1969.

 .6.   .Erich Jantsch,  Techno!ogy  Forecasting  .in  Rer.sp.ecti.ye.

 7.   Alvin Weinberg, "The Short Term  Nuclear Option,"  Report  of the  Cornell
     Workshops  on  the Major  Issues  of a  National  Energy Research and
     Development  Program,  1973,  Chapter  III.

 8.   M.  C.  Olson,  Unacceptable  Risk.  New York, Bantam, 1976.

 9.   Acting Federal  Energy Administration (FEA) Deputy Administrator
     John  Hill, testifying before Joint  Committee on Atomic Energy,  from
     Bureau of  National Affairs, May  8,  1975.

10.   George Getschow, "Utilities To Face a  Scramble  for Uranium as More
     Nuclear Plants  Come  on  Stream,"  Hall Street  Journal, September, 1974.

II.   Ron Chernow,  "U.  S.  Confronts  Uranium  Crisis,"  Philadelphia Inquirer,
     May 5, 1975.

12.   "Uranium'Backfires on Westinghouse," Business Week,  August 18,  1975.

13.   M.  C.  Day, "Nuclear  Energy: A Second  Round  of  Questions," Bulletin of
     the Atomic Scientists,  December, 1975.

14.   "Why  Atomic  Power Dims  Today," Business Week, November 17, 1975.
                             II-B-2-59

-------
15.  "U. S. Needs Nine New Colorado Plateaus To Fill Uranium Need,"
     Not Man Apart, mid-January, 1976.

16.  "GAO Warns of Uranium Shortages," Critical Mass, February, 1976.

17.  David Burnham, "Atomic Reactors May Have To Shut," New York Times,
     January 5, 1975.

18.  "Nuclear Fuel Services May Never Reopen," Not Maji Apart, mid-June,
     1975.

19.  M. C. Olson, "Nuclear Fuel:  The Hot Shuffle," Progressive, April,
     1975.

20.  F. Von Hippel and R. H.  Williams, "Energy Waste and Nuclear Power
     Growth," Bulletin of the Atomic Scientists, December, 1976.

21.  G. Cody, "Letters," Scientific American 234 (April, 1976).

22.  Project Independence Final Task Force Report:  Solar Energy, U. S.
     Federal Energy Administration, Washington, D. C., November, 1974.

23.  Mike Gravel, Energy Newsletter, U.  S. Senate, February, 1975.

24.  Catalog on Solar^ Energy^ Heating and Cooling Products, ERDA-75,
    ..Oc.to.ber, 197.5.

25.  Mike Gravel, Energy Newsletter, U.  S. Senate, September, 1974.

26.  "Harnessing the Sun," Forbes Magazine, October 15, 1974.

27.  "Solar Heat Looks Better as Fuel  Costs Skyrocket," Washington  Star-
     News, Friday, September  13, 1974.

28.  Mike Gravel, "Fact Sheet on Nuclear Power," U. S. Senate, 1974.

29.  K. Parker, "Sun Power Comes of Age  across State," The Providence
     Sunday Journal, December 28, 1975,  p. Gl.

30.  Center for the Study of  Responsive  Law, A Citizen's Manual on  Nuclear
     Energy 1974, Washington, Center for Study, 1974.

31.  B. Chalmers, "Letters,"  Scientific  American 234 (April, 1976): 14-15.

32.  Hans Bethe, "The Necessity of Fission Power," Scientific American
     234 (January, 1976).

33. R.  Doctor,  J. Harding, and J. Holdren, "The California Nuclear  Safe-
     guards Initiative," Sierra^ Club Bulletin, May, 1976.

34.  T. Ehrich, "Atomic Lemons:  Breakdowns and Errors in  Operation Plague
     Nuclear Power Plants," Wall Street  Journal, May 3, 1973.
                            II-B-2-60

-------
35.  "Uneconomical A-Plants," (Honolulu) Star-Bulletin and Advertiser,
     June 23, 1973, a Christian Science Monitor report.

36.  D. D. Comey, "Nuclear Power Plant Reliability:  A Report to the
     Federal Energy Administration,: September 10, 1974.

37.  G. Bukro, "Economics of Nuclear Power Hit," Chicago Tribune,
     September 29, 1974.

38.  "Critic Says Atom-Power Reliability Gap Could Cost $100 Billion
     in Next 15 Years," Wall Street Journal, October 8, 1974.

39.  M. C. Olson, "Nuclear Energy:   It Costs Too Much," Nation,
     October 12, 1974.

40. J1. P. Donsimoni, R. Treitel and I. C. Bupp, "The Economics of
     Nuclear Power," Technology Review, February, 1975.

41.  P. Margen and S. Lindhe, "The  Capacity of Nuclear Power Plants,"
     Bulletin of the Atomic Scientists, October, 1975.

42.  J. Margolis and H. Kelly,  "Electric Utility Business:  Even When
     It's Good It's Bad," Chicago Tribune, May 26, 1975.

43.  J. Margolis and H. Kelly,  "Why the Power Cost Controversy Is
     Heating Up," Chicago Tribune,  May 25, 1975.

44.  D. Burnham, "Federal Study Charges Little Concern for Utilities
     in Reliability of Reactors," New York Times. March 9, 1975.

45.  P. Milius, "Major Battle Is Brewing on  A-Plants," Washington Post,
     June 23, 1975.

46.  "Five Utilities Delay Six  Units, Cancel  Another,"  Wall  Street
     Journal, September 10, 1974.

47.  S. Nordlinger, "Nuclear Plants Lose Favor," Baltimore Sun. March 31,
     1974.

48.  "Utilities Doubt Nuclear Plants," Critical  Mass, April, 1975.

49.  J. Mateja, "Money Is Mightier  Than the  Atom," Chicago Tribune,
     December 7, 1975.

50.  "The Bills Are Electrifying,"  Newsweek.  April 8, 1974.

51.  "PE's Rates Up 63.6% in One Year," Philadelphia Inquirer. August 25,
     1974.

52.  J. Kifner, "Shutdown of a  Nuclear Plant, 17th in 19  Months,  Spurs
     U. S. Debate," New York Times, March 31, 1974.
                            II-B-2-61

-------
53.  "Palisades To Pay Pipers," Not Man Apart, December, 1974.

54.  "Electric, Gas, Phone Bills Jump—Some Double or More,"  U> S^. News
     and World Report, May 13, 1974.

55.  D. Clendinen, "Jacksonville's Giant Business Gamble," St. Petersburg
     Times, June 8, 1975.

56.  "A City That Reached for Riches and Got Headaches Instead," U_. S. News
     and World Report, September 1, 1975.

57.  A. Spake, "Critical Mass '75," Progressive, January, 1975.

58.  "Critical Mass Conference Marks Increase in Citizen Activity,"
     Critical Mass, December, 1975.

59.  "Coal Research Unit To Be Reorganized," New York Times, February 24,
     1973.

60.  "Breakthrough Allows Use of 'Dirty' Coal," Milwaukee Journal,
     September 25, 1974.

61.  "Sulfur Can Be "Scrubbed" from Coal Power Wastes," Wisconsin  State
     Journal, September 26, 1974.

62.  David B.urnham, "AEC Penalizes .Few Nuclear .Facilities Despite  Thousands
     of Safety Violations," New York Times, August 25, 1974.

63.  "Nuclear Power Plants:  Do They Work?"  BPI Annual Report. 1973.

64.  Calvert Cliff's Coordinating  Committee v. AEC, 449 F.  2d 1109, 1128,
     Circuit Court of Appeals, Washington,  D.  C. 1971.

65.  John W. Gofman, "Radiation Doses and Effects in a Nuclear Power Economy:
     Myths vs. Realities," Congressional Record 122 (May 6,  1976).

66.  D. Burnham, "Data Show AEC Suppressed  Studies on Dangers of N-
     Reactors," New York Times News Service, The Courier-Journal and Times,
     Louisville, Kentucky, November 10, 1974,  p. A3.

67.  "Nuclear Safety:   Who Can Be  Believed?"  The Courier-Journal,  Louis-
     ville, Kentucky,  Editorial, November     1974.

68.  L. Arnold, "Debate Livens as  Nuclear Objections Grow,"  The Hoosier
     Waltonian 13 (Winter, 1974):   8.

69.  J. Primack and F.  von Hippel, "Nuclear Reactor Safety," Bulletin  of
     the Atomic Scientists 20 (October, 1974):  7-9.

70.  E. J. Sternglass,  "Significance of Radiation Monitoring Results for
     the Shippingport  Nuclear Reactor," Department of Radiology, University
     of Pittsburgh.
                             II-B-2-62

-------
71.  Lawrence K. Cohen, "Pre-Operational Environmental Radioactivity
     Monitoring Program at the Beaver Valley Power Station," Quarterly
     Reports. NUS-884, 915, 916, 950.  NUS Corporation, Rockville, MD,
     1972.

72.  L. B. Lave, S. Leinhardt, and M. B. Kaye, "Low-level Radiation and
     U. S. Mortality, W. P. 19-70-1," Carnegie-Mellon University,
     Pittsburgh, PA, July, 1971.

73.  E. J. Sternglass, "Infant Mortality Changes Associated with Nuclear
     Waste Discharges from Research Reactors into the Upper Ohio Watershed,"
     Department of Radiology, University of Pittsburgh, February -15, 1972.

74.  J. Brimberg, "Dow Workers'  Radiation Exposure Called High," Denver
     Post, August 14, 1970.

75.  J. P. Holdren, "Hazards of the Nuclear Fuel Cycle," Bulletin of the
     Atomic Scientists 30 (October, 1974).

76.  Mike Gravel, "Energy Newsletter," U. S. Senate, August, 1975.

77.  Knob and Valley Audubon Society Newsletter 3 (September-October, 1974).

78.  Ralph Nader, "Nuclear Power:  More than a Technological Issue,"
     Mechanical  Engineering, February, 1976.

79.  G. Speth, A. Tamplin, and T. Cochran, "Bypassing the Breeder,"
     Environmenta1  Action, April 12, 1975.

80.  Von Hippel  and Williams. "Energy Waste in the COAT Scenario,"
     Bulletin of the Atomic Scientists, December, 1976.

81.  "Power Generation, Conservation, Health, and Fuel Supply," Federal
     Power Commission, March, 1975.

82.  L. McCormick,  "Vermonters Test Cheaper, Off-Peak Electricity Use,"
     The Christian  Science Monitor, Friday, March 12, 1976.

83.  "Demonstration of Peak Load Pricing Urged," The Dallas  Morning News,
     Sunday, February 22, 1976,  p.  21A.

84.  U. S. Atomic Energy Commission, "An Assessment of Accident Risks in
     U. S. Commerical Nuclear Power Plants," Summary Report, August, 1974
     (Washington: U. S. Government Printing Office) 1974-585-745.

85.  Budnitz, Lewis, ejt cil_., "Report to the American Physical  Society by
     the Study Group on Light-Water Reactor Safety," Reviews of Modern
     Physics 47  (Summer, 1975):  S1-S123.

86.  The Study Group on Light-Water Reactor Safety, "Nuclear Reactor
     Safety—the APS Submits Its Report," Physics Today 28 (July,  1975):
     38-43.
                            II-B-2-63

-------
87.  B. Augustine, "AEC Finds Safety in Numbers," Environmental  Action,
     October 12, 1974:  3-8.

88.  C. Lerza, "AEC:  Sinking in a Sea of Criticism," Environmental
     Action. October 12, 1974:  4-5.

89.  T. F. Mancuso, Alice Stewart, George Kneale, "Radiation Exposures .
     of Hanford Workers Dying.from Cancer and Other Causes," presentation
     at the Tenth Mid Year Symposium of the Health Physics Society,
     Saratoga Springs, New York, 11-13, October, 1976.

90.  "Can Nuclear Wastes Be Stored?" Newsweek, November 18, 1974.

91.  "Nuclear Powerplant Construction Ban," Congressional  Record  121,
     Washington, March 14, 1975:  H1780.

92.  R. A. Richardson, "The Selling of the Atom," Bulletin of the Atomic
     Scientists 30, October, 1974.

93.  D. Krieger, "When Terrorists Go Nuclear," Center Report, June, 1974.

94.  J. D. Schleimer, "The Day They Blew Up San Onofre," Bulletin of the
     Atomic Scientists 30, October, 1974.

95.  A. Tamp!in, T. Cochran, J.  Speth, "We Do Not Have  To  Rely on Nuclear
     Fis.s4.on," The .Washington Post, Washington., D.  £.,  June 9, 1974.
                            II-B-2-64

-------
                              Acknowledgements

                                 Section
     The author would like to thank the many people who supplied data,
reports and information for this study.  Special thanks goes to persons in
the U.S. EPA Region IV and V Offices and state agencies who supplied air
data.  The assistance of the Oak Ridge and Argonne National Laboratories as
well as various U.S. EPA and ERDA offices is also most appreciated.
     Criticisms of the current state of emissions data and regulations  should
not be construed to reflect upon either the agencies or personnel involved,
for the significant improvement in air quality in recent years is indicative
that they are doing a commendable job under difficult circumstances.
                                  1TB-3-I

-------
3.1 AIR IMPACTS           /
3.1.1 ASSESSMENT PERSPECTIVE
3.1.1.1 INTRODUCTION
      Air pollution is widely viewed as the most important environmental
problem in the United States.  Among the various source categories power
generating facilities are the greatest impactors on the air resource on a
quality basis as shown in Table 3.1-1. The relatively high density of
power plants in the ORBES region and the use of indigenous high sulfur
coals make the percentage impact from power plants even higher for
the region.

                             TABLE HB-3-1-1
             DISTRIBUTION OF AIR POLLUTION SOURCES IN THE
                            UNITED STATES (1)

Source

Transportation
Miscellaneous
Industry
Stationary fuel combustion
(power plants)
Solid waste disposal

Mass basis
percent
51.4
14.6
14.0

15.8
4.2

Air quality basis
percent
16.4
18.9
27.0

34.9
2.8
      Among the factors  that cause air impacts to be of such importance
      that transport to receptors occurs  through random processes which
are difficult to define and that the exposure and uptake by the receptor
cannot be controlled.  Of the  physical environment impacts (air, water,
and land) that will result from the ORBES scenarios, the most serious
constraints to future growth will be due to air quality considerations.

      The overall  known  air  impacts of fossil fuel-electric cycles from
resource extraction to  end use are an order of magnitude greater than
the known   impacts from a nuclear-electric cycle of the same production
under normal (i.e. non-accidental) operating  conditions.   Therefore the
preliminary  analysis focused on the fossil-fuel cycle.  The current
electrical production by fuel type in the four-state study  area is given
in Table 3.1-2.

              Coal     Oil       Gas    Nuclear

Percent of
Production    89.7     4.8        .8      3.2          1.5

                             TABLE EB-3.1.-2
        ELECTRICAL PRODUCTION IN ILLINOIS, INDIANA,  KENTUCKY
                      AND OHIO BY FUEL TYPE (2)
      Since  coal currently accounts for 94% of the total fossil fuel production
sector  and since that percentage will be increasing during  the study period,
the analysis addresses only the impacts from  coal.  The term "fossil fuel"
as used in the impact  assessment sections refers to coal.
                               IT 8~3~

-------
3.1.1.2 BASELINE DATA

   A.  Emissions Data
      The most recent emissions data for coal fired power plants were obtained
from state air pollution control agencies and U.S. EPA RegionsTV and V.  These
data for participates and sulfur oxides are summarized in Appendix 3.1-A. While
some of the data bases were considered to be of good quality, overall there are
major discrepancies which must be resolved before the data can be considered
adequate, except for comparative and qualitative analysis.  An example of the
discrepancies found is shown in Table 3,/-3  which lists the existing
emissions for Kentucky power plants.  The EPA National Emissions Data
System (NEDS) data in that table are the latest published available (May, 1976)
and are for the year 1973 (3).  The Kentucky data are for the years 1974 and
1975. The EPA-NEDS data show total loading for particulates to be 3.5 times
the total for Kentucky data and twice the sulfur oxide levels.
      While emissions were greatly reduced during that period due to
regulations, it is highly unlikely that the levels  of decrease indicated by
the table actually occurred over a one-to two-year period.  Even if both
sets of data were correct for their respective time periods,  there are  still
problems, because major national studies which  are currently underway
and whose results are intended to be used in policy decisions are using
the older NEDS data (in one case,  1972). The NEDS data is normally
supplied by state agencies, so it would be expected that the files should
be the same with some reasonable time lag.  This does not seem to be the
case, however; and of the NEDS data requested for the four ORBES states,
it was recommended that the data be obtained directly from  two of the state
agencies. Reasons for these discrepancies include excessive lag  times and
errors in data entry. In one case  it was reported that errors noted by state
agencies could not be changed in the NEDS system.

      Inconsistencies were also noted within the data files.  At least one
state agency file of power plant emissions in tons per year consisted of an
unlabeled mixture of maximum possible emissions and actual average emissions
based on plant capacity factors. Since the average capacity factor is  on the
order of fifty percent, this could result in a one-hundred percent error or
more.  Both maximum possible emissions and average emissions rates data
are needed for short and long time period analyses respectively, but the
data need to be segregated and the basis clearly specified.

      The emission factors that are often used to compute total emissions
are another source of error.   As pointed out in the "1973 National Emissions
Report, " emission factors are statistical averages and "are  not necessarily
precise indicators of emissions from an individual source"  (3).  The report
also points out that percentage error increases as the geographic  area
decreases. There is apparent uncertainty regarding how much of the sulfur
in coal leaves an uncontrolled stack as SQt.  The U.S. EPA and most state
agencies use (for bituminous  coal) a 0.95 factor, while TVA uses  a 0.90 factor,
and a National Academy of Engineers report (4)  cited a use of 0.85.  That factor
is  known to vary with coal type, and a recent Teknekron study, an"lntegrated
Technology Assessment of Electric Utility Energy Systems "(5) , bituminous
(0.95), sub-bituminous (0.85), and lignite (0.75).
                            ors-3-3.

-------
AQCR CODE
     72
     77
     78
     79
    101
    102
    103
    105
  TOTAL
                                   TABLE II-B-3.1-3
                  COMPARISON OF POWER PLANT EMISSIONS DATA
                   From EPA-NADB (a) and Kentucky NEDS System (b)
       Particulate
STATE (b)      EPA-73 (a).
20,892         218,062
 8,263           3,784
 2,680
 1,088
 1,860
25.593
18,439
17,974
96,789
  5,760
  2,572
 22,525
 11,840
 73,399
337,942
                                                              SO.
STATE (
-------
      The Teknekron study gave the following quality rating of emission
factors for bituminous coal fired boileri hydrocarbons and aldhydes-poor
(error in the range of 100 percent), carbon monoxide - good (error probably
25 percent) radioactive emissions - good (error probably 50 percent) particulate
polycyclic organic matter (PPOM) - very poor (error around an order of magni-
tude) (5).  The quality of trace element emission factors can be inferred from
that report to be poor to very poor, with errors typically in excess of 70 percent.

      In determining existing emission rates, it is commonly assumed that
control equipment operates within its design  specification.  This is also
questionable and effects the reliability of emissions calculation.  In at least
one case a major new power plant in the ORBES with high efficiency precip-
itators (in excess of 98 percent removal) has had major problems with its
equipment and was several years in reaching a seventy percent removal
efficiency.  During this period it has been reported that permits for the
construction of another nearby power plant and Prevention of Significant
Deterioration (PSD) determininations were made assuming that the plant
with equipment problems would be operating within design specification.
      In addition to observations already discussed, there were others that
bring emission rate data base into question.  The base included entries current
from 1974 until 1977, and many changes could have occurred since the earlier
date, including changes in fuel type and composition, changes in capacity
factor, and changes in control equipment.  Discrepancies were observed in
plant capacity and fuel type.

      It should be noted that the observations made regarding the reliability
of the data have been based on preliminary evaluation of the most recent data
made available and from informal verbal discussions with control agency
personnel. These data files, which are continually updated and improved
are most important since they are used in models which evaluate ambient air
quality, in determining the adequacy of State Implementation Plans (SIP)
and their emissions limitation,  in designating Air Quality Maintenance Areas
and in making Prevention of Significant Determinations (Section 3.1.1.3).
Therefore, it is strongly recommended that Phase n of ORBES include a
thorough review of the emissions data  base, establish a measure of its
reliability, determine the possible effects based oh that reliability, and
make improvements as required.
      The comments regarding the often poor quality of the existing data
base should not be construed as an reason for not making an assessment or
as a reason to defer policy decisions indefinately.  As the Report of the
Inter-Agency Working  Group on Health and Environmental Effects of
Energy Use stated:
            We cannot wait for every detail of the  hierarchy from "data
      sources" to "policy analysis" to  be fully completed before applying
      the information and analysis tools now available to,the pressing
      issues... .The only alternative to this is mas'slve experimentation
      with policy based on subjective  decisions and some form of prayer ,(35*
                                 HB-3-5

-------
      In subsequent years the ORBES study will likely address transport
of pollutants into as well as out of the Basin.  TVA has recently compiled
an emissions data base for a large area which includes the *outh westerNl
part of the ORBES region. (7) . The data should be useful for that analysis
considering that the predominent winds from that area are from the southwest
to the northeast (7) and would carry emissions into the ORBES region.

3.1.1.2
   B. Ambient Air Quality                                   ,|j
      The air  quality in the ORBES region ranges from some of the most
highly polluted areas in the country to large  rural areas with "clean"_fir.
Steubenville and Louisville are among areas in the ORBES region that have
perennially appeared in published lists of areas with poor air quality.
      A detailed analysis of the existing air quality data base was not   \
within scope of the first year of study, and it is recommended that this   I
could be done  in Phase n.  However, based on preliminary evaluation of
selected data and extrapolation of the findings regarding the current state
of emissions data, it is anticipated that there are major problems with the"
available data  base. The Council or Environmental quality reported in its '.
sixth annual report that in 1973-74 EPA Region V evaluated selected state
and local agency monitoring sites (a number  of which were in the ORBES
region) and reported that "serious deficiencies were found" and that .
data from these sites must be interpreted cautiously" (6) . Deficiencies
were noted in  analytical methods, location, operation, and data reduction.
Improvements  have probably been made  since that time, but it is unlikely
that  data can be used without extensive analysis and consultation with the
monitoring agency.

      In 1973, as a result of the case of "NRDC, Inc. et al vs EPA", all
state implementation plans were disapproved by EPA as failing to provide
for the maintenance of standards. The states were then required to analyze
areas in which expected growth by 1985 could result in the national ambient
air quality standards being biolated.  Counties or county  groups which were
projected to be in violation of any National Air Quality Standard (NAAQS)
by 1985 were classified as AQMA's (8).  The use of county-based areas has
largely replaced the use of original Air Quality Control Region  (AQCR's) in
regulation and compliance management.  Since the AQMA designations in 1975,
additional areas have been identified by EPA for "attainment and maintenance"
of the national ambient air quality standards. These determinations are
based partially monitored standard violation  (or "non-attainment") reported
by the states to EPA.  The additional areas designated for Kentucky have
been as follows: Jefferson County - oxidants, CO; Boone, Kenton and
Campbell Counties - Oxidants; and Boyd County - SO- (9).
                                                 £
      Table 3.1-4 shows the ORBES counties designated as AQMA's or "non-
attainment" areas for particulate or sulfur oxides and in which ORBES Task I
sited plants. As can be seen in Table 3.1-4, essentially all the site assignments
in AQMA's occurred in Ohio.  A total of 17,000 MWe capacity or about a third
of the BOM 80-20 scenario total was assigned to AQMA's in Ohio, which brings
into  serious question the capability of the state's air resource to support that

-------
           TABLE  3.1-4
Counties Designated AQMA's in which
  ORBES Scenario Coal-fired Power
    Facilities Have Been Located
STATE
COUNTY (AQMA Pollutant)
BOM
80/20
Scenario
BOM
50/50
TECH FIX
100% Coal
ILLINOIS

St. Clair (TSP  SO2,  Oxidants)


INDIANA

Clark (TSP, SOg)

KENTUCKY

Henderson (TSP)


OHIO

Belmont (TSP, SO2)


Butler (TSP, Oxidants)


Clark (TSP, SO2)

Clermont (TSP, Oxidants)


Franklin (TSP)

Montgomery (TSP, SO2)


Hamilton (TSP,Oxidants)


Warner  (TSP, Oxidants)
            Low BTU Gas
            Plant (1985)
Low BTU Gas
Plant (1985)
            1000 MW (1990)     1000 MW (1990)
            High BTU Gas
            Plant (1995)
            High BTU Gas
            Plant (1995)

            3000 MW (1995
            2000)
High BTU Gas
Plant (1995)
High BTU Gas
Plant (1995)

2000 MW
(1995 - 2000)
            2000 MW (1995)     2000 MW (1995)
            3000 MW
            (1990 - 1995)
2000 MW
(1990 - 1995)
            2000 MW (2000)     2000 MW (2000)
            2000 MW
            (1985 - 1990)

            2000 MW
            (1990 - 1995)

            3000 MW
            (1995 - 2000)
2000 MW
(1985 - 1990)

1000 MW
(1990 - 1995)

2000 MW
(1995 - 2000)
600 MW (1997)


600 MW (1998)

600 MW (1998)


1200 MW (1996)

600 MW (1999)


600 MW (1996)


600 MW (1999)
              JJb-3-7

-------
level of growth.  A total of 11,000 MW was assigned to the Cincinnati
Interstate AQMA, which includes Butler, Clermont,  Hamilton and Warren
Counties in Ohio. At the level of the current analysis (county level) these
sitings represent an inconsistency in the RTC.
      It is common practice in site screening studies to totally exclude
AQMA's as potential sites (10,11) .  However, in reality the AQMA1 s were
designated by counties for administrative purposes and the AQMA may
actual apply to a sub-county area.  Depending on the specific location of
that area, more of a constraint could be presented against locating a power
facility in an adjoining county, than to a further removed location in the
same county.  Such determinations are made by detailed site specific
analysis.
      An indication of the distribution of current air quality within ORBES
can be obtained from Figure 3.1-1 which shows the designated Air Quality
Maintenance Areas for total suspended particulates (TSP) and sulfur dioxide
Designated AQMA's for oxidants are the following: Illinois - Madison, St.
Clair, Monroe; Indiana - Marion; Ohio - Butler; Kentucky - Boone, Campbell,
Kenton.
      High levels of sulfate are experienced in most of the ORBES region as
shown in Figures 3.1-2 and 3.1-3.  The area in those Figures have the highest
sulfate levels also have a high percentage of the nation's coal fired electric
generating capacity and the highest SO- emission density as shown in Figure
3.1-4. Ambient sulfate data and analysis for the ORBES region are available
from the ongoing SURE (Sulfate Regional Experiment) and MISTT (Midwest
Interstate Sulfur Transformation and Transport) research programs (12,13).
The location of the SURE monitoring sites and sample data are presented in
Appendix 3.1-B

3.1.1.3 REGULATIONS
      A. Emission Regulations

      The utility industry has been vocal in its frustration over the
continual changes in air pollution emissions regulations that apply to
their facilities. The findings of the  first year of the study indicate
that there is indeed a basis for the "moving target" complaint, that overall
the current state of these regulations can best be described  as a muddle,
and that there seems to be no resolution in sight.  Because of State Imple-
mentation Plan (SIP) remands, AQMA reviews, legal actions, and other
factors, some of the ORBES states have as many as two or three set of
regulations; and it is obvious from discussions that in many cases staff
level control agency personnel themselves are not sure which regulations
apply.  The effect of this situation on the ORBES study is quite significant,
and the following discussion is divided by new source and existing source.
The reader is also referred to other  studies which give extensive analysis
to the environmental and economic ramifications of various emission limitation
levels.  (11,14,15,16,17,18.19).

-------
          Air Quality Maintenance
              .Area Pollutants
              Sulfur Dioxide
               Particulates
      Sulfur Dioxide and Particulates
Figure 3.1-1  Air Quality Maintenance Areas

    for Sulfur Dioxide and Particulates
               ITB-3-?

-------
   FIGURE  3.1-2Geographical Distribution of Typical Urban Sulfate Levels  in the
   United  States (Annual Means)  (reproduced from ref.  8)
       5.0-6.9 ngtm3
       1.0-2.9 (ig/m3
     5-1 >9.0 jig/in3
    ~~l 3.0-4.9 fij/mj
    	    	„_._             .           ^J                       ..."
. FIGURE 3.1-3 Geographical Distribution of Typical  Nonurban Sulfate Levels in
 the United states. (Annual Means)  (reproduced  from ref* 8)
      I  1 < I tonAm'
      77ft  1-20 lom/Vm>
      IBi >20 tons/km* •
    FIGUP.E3.1-4  Nationwide Geographic Variation in SO2 Emission Density
    .(reproduced from ref.  8)

-------
      New Sources: The current maximum allowable emission rates for new
coal fired power plants are given in Table 3.1-5.These rates are uniform nationally
and have remained fixed since their promulgation.  U.S. EPA Prevention of
Significant Deterioration (PSD) Determinations (see the following section) can
require lower emission rates.

      Allowable Emission Rate:  Lb Pollutant/10  Btu Input	


      Particulate              Sulfur Oxides      Nitrogen Oxides
                             As SO2             As NO2

         .1                     1.2                 0.7

Table 3.1-5 NEW SOURCE PERFORMANCE STANDARDS FOR COAL FIRED POWER
            PLANTS WITH INPUTS OF 250 x 106 BTU OR GREATER

      While the standards given in Table 3.1-5 were used in the Phase I
analysis, it is likely that changes will be made before the third year of the
study is complete.  The Clean Air Act ammendments which are before Congress
are more stringent in requiring the use of Best Available Control Technology
(BACT) on all coal fired power plants, regardless of the fuel's sulfur content.
Resivion in the NSPS for power plants are being considered and may be issued
by EPA in 1978.  Reportedly the current plan also required BACT.  The require-
ment of BACT would have a major impact on the ORBES  region, one of which
could be to make the use of low-sulfur western coals uncompetitive in most of
the region. Future sulfate regulations, which EPA plans on developing in 1981
(20),  could also affect the allowable rates of SO-, a precursor to sulfates.
      The national air quality criteria documents are being revised, with
drafts completed according to the following EPA timetable: photochemical
oxidants (and related hydrocarbons) - August 1977; nitrogen oxides -
February 1978;  car-bon monoxide - August 1978; sulfur  oxides (and associated
particulate) - Au^ueO979; and particulate matter - August 1979 (21) .  If
revised air quality^sfandards are required, then changes in emissions  limita-
tions would likely follow.

      Existing Source Regulations:  At the beginning it  was believed that it
would be a relatively simple matter to determine the current emissions  from
existing power plants and how they would vary during  the study period. This
was found not to be the case at all; and, as the Teknekron utility system technology
assessment pointed out, "What one would expect to be a straightforward problem—
specifying sulfur emissions limitation to be met by existing coal and oil fired
power plants—turns out to be somewhat complicated" (11) .  The determination of
the current emission rates for ORBES power plants turned out to be very time
consuming and a major element of the analysis.
      Another factor that make specific allowable emission rates difficult
to determine is that different approaches that are taken  by the various states
in their SIP's.  The most common approach is to specify a set of allowable
emission rates (usually in pounds per million Btu input) which apply,  based
on the air quality classification of the area in which the plant is sited.
                              EI3-3-IJ

-------
These areas are managed as either Air Quality Control Regions, for which
there are three classifications, or as counties.  In Kentucky for example,
there are five classifications by county, each with specific emissions
limitations.

      In some  cases there is disagreement between the state and U.S. EPA
as to what the classification is and therefore differing interpretations of
what the allowable emission rates are. Other states, Illinois for example,
have regulations based on ambient air quality levels and may not require
control unless  a given concentration level is exceeded in the vicinity of
a plant. Because of its very nature, that type of limitation is  under
continual review as air quality monitoring dictates.

      The time frame in which plants will be in compliance with their
allowable emissions standards is also very difficult to determine.  The
statutory deadlines have passed, and many plants are still not in compliance
and are either  operating under compliance schedules, negotiating compliance
schedules, or operating in violation of regulations.   For the initial analysis
it was considered to be adequate, and realistic, to assume that all existing
plants would be in compliance with participate regulations by 1980 and half
would be in compliance with SO-, regulations by 1980, with the remainder
by 1985.                     ^

      The reliability of the projections of the emissions from existing plants
is very significant in the study since they are projected to make up   SO  %
of the total coal-fired plant SO- emissions in the year 2000 for Scenario I
and 76% for Scenario HI.  The degree of uncertainty caused by varying
interpretations of Indiana's regulations as shown  in Figures -5.1-6 and .3.1-3.

      Projections based on U.S. EPA's interpretation of emissions  limitations
gave a loading of 208,195 TONS SO2 /yr., while the total computed from the
NEDS listing of allowable emissions reported by the state air pollution agency
was 772,932 TONS SO./yr for the year 2000.  (The allowable U.S. EPA emission
rates were obtained from a computer output entitled "The SASD Interpretation
of the State Implementation Plan SO, Regulations for Coal-Firing as of 12-31-75.)
While this situation hopefully does not exist for all of the ORBES states, the
projected air loadings for Indiana can vary by a factor of 3.7 depending on
which interpretation is followed.
      It is recommended that the Phase II of the ORBES study determine in
greater detail the applicable emissions regulations, compliance schedules,
discrepancies  between state and EPA interpretations  and when and how they
can be resolved.  From that analysis a consistent set of allowable emission
rates or a range of likely allowable emission rates should be developed.
      B. Prevention of Significant Deterioration (PSD)

      Significant deterioration considerations due to current and proposed
regulations pose the most severe of the physical environment constraints to
the numbers, siting and size of power generating facilities  in the ORBES
region.

-------
      The Clean Air Act of 1970, as amended, stated that one of its basic
purposes is "to protect and enhance the quality of the Nations's  air resources
so as to promote the public health and welfare..."  The EPA's interpretation
of that language and its approval of state implementation plans which did not
provide for preventing significant deterioration were challenged in court.
The courts ordered the EPA administrator to review all implementation
plants  and disapproved any plan that failed to effectively prevent significant
deterioration. Subsequently regulations were promulgated in December,  1974,
calling for the establishment of "classes" of varying incremental increases
(22) in TSP and SO_. Those classes are qualitatively described as follows:
                  c.
      Class I      Areas in which almost no changes in air quality would be
                  allowed

      Class II     Areas in which moderate changes would be allowed, but
                  where stringent air quality constraints would be imposed.

      Class III    Areas in which air quality would be allowed to deteriorate
                  down to national standards.


      The entire county was originally designated as Class II with provision
for the states to recommend class changes for specific areas.  It was determined
from EPA Region V and the Kentucky Division of Air Pollution Control that all
of the ORBES region is currently Class II.  The allowable PSD increments are
given in Table 3.1-6. The possible impact of PSD has been described in three
recent studies (10,11,18),  from which material has been condensed and is
included in Appendix B. 1 and to which the reader is referred for a more
comprehensive treatment.
      Table 3.1-7, prepared from the Argonne  National Laboratory study,
contains the allowable PSD  class increments and the  concentrations that
could result from the types and sizes of power facilities used in the ORBES
study.  That study shows that large power parks or clusters are likely to
be severly constrained by PSD Class II increments.  Table 3.1-7 also shows
that those increments will also constrain siting or require emissions reduction
for power plants on the order of 1000MW or larger.

      Designations of Class I areas under current EPA  regulations or mandatory
designation under the proposed amendments to the Clean Air Act of certain types
of national lands (e.g., parks) of a given size could cause extensive siting
constraints.  Plants of 1000 MWe capacity at the SO- NSPS would be excluded
from a 90-mile buffer zone  around Class I areas, and plants of 3000 MWe
capacity at 10% of the SO- NSPS would be limited from a 30-mile zone, based
on the Argonne analysis*X10) • Depending on the specific location, it is likely
that PSD will not present serious constraints to coal gasification facilities.

-------
TableB-3.1-7
Comparison of NAAQ and Estimated Maximum Concentrations from ORBES^Coal Utilization Facilities
                            SOURCE:  Reference 10
Maximum Concentration pg/m*



	 	 ...


	

1 COAL FIRED POWER PLANTS j






hi
M
03
i

-------
         In making PSD determinations the EPA models ambient concentrations
   that will result in the plant vicinity including the effect of other power plants
   within a twenty kilometer radius (23) and the background level.  Thus the
   cumulative local effects of plants are considered.  Based on this type of
   analysis the allowable emission rate for the proposed plant is specified. It is
   not clear if the consideration of nearby power plants is limited to a 20 kilometer
   (12 mile) radius in ail determinations. If that is the case it does not seem to be
   adequate, according to the Argonne analysis which indicates maximum 24 hr.
   SO- concentrations of over 20|J.g/M3 at a 20 mile distance from a 1000 MW power
   plant meeting NSPS  and oven 60|ag/M3 at a 20 mile distance from a 3000 MW power
   power plant meeting  NSPS (10) .  The concentration levels from existing
   uncontrolled plants would be several times those amounts.  Since the
   allowable 24 hour SO, increment is lOOjig/M3 it appears that power plants
   within 50 to 100 miles or more should be included in the determination.
         The March 25, 1977^determination for the Wise's Landing Plant  in
   Trimble County in setting the allowable SO- emission rate states:

         Sulfur dioxide emitted to the atmosphere from the boiler shall not
         exceed....0.85per million Btu.. .by using a control  device which
         has a control efficiency of 90 percent. EPA has determined that
         control devices of 90 percent efficiency are available  (authors
         emphasis)  (24).

         The March 1, 1977 SD determination for the Ghent Power Station
   states:
         The following general statements can be made concerning BACT
         (Best Available Control Technology) for power plants. .BACT for
         sulfur dioxide would  consist of either low sulfur coal (less than
         0.72%) or a flue gas desulfurization FGD system.. .The maximum
         emissions of .. .SO_ which will be  allowable are.. .1.21b/million
         Btu heat input.... u3)

   In the case of the Ghent Plant it is proposed to use low sulfur coal while the Wises
landing plant proposes to scrub. The apparent difference in the interpretation of
BACT in  these two cases causes further confusion about the BACT.

         The allocation of PSD increments for a given location  are based on a
   "first come" basis.  Thus a  proposed facility which meets NSPS could receive
   all of  the PSD increments for a given site.  The effect of this policy could be
   to encourage premature installation of generating capacity and also could
   result in piecemeal unplanned allocation of the air resource.

         Since PSD determinations are of such importance, the procedure used
   should be carefully reviewed to.determine uncertainties.  In one case the
   contribution of a nearby power plant was reported evaluated with the assumption
   that the control equipment was operating within design specifications, when in
   fact the equipment  had performed poorly since installation and never approached
   design limits.
                                     JT3-3-I5

-------
      The reliability of the modeling procedure should also be examined, for
the models are the least reliable for the short-term average period which is
most critical.  In addition to the preceding considerations, ORBES phase II
should examine the energy facility configurations with respect to PSD and
changes that may occur in PSD policy in the next few months.

3.1.1.4 LIMITATIONS OF ANALYSIS

      The methodology that was chosen, and the one commonly used for
environmental impact statement and "first cut" analysis of the air quality
impact of a given action proceeds  as follows:

      1.  determination of the current emission sources

      2.  determination of future emission sources

      3.  application of current  emission rates to existing sources to
         determine baseline emission rates

      4.  application of compliance emission rates to existing sources at
         some future date

      5.  application of NSPS to  future sources

      6.  calculation of the total pollutant loadings from existing and
         new sources for the base year and for selected years in the
         future

      7.  impact analysis based on the trends and levels of the projected
         loadings

      The results  of this analysis are given in section 3.1.4.1. Items 1 and
2 above were determined by ORBES Basic - I.  The Reliability of 1 is on the
order of 10%, and the  difference in the scenarios themselves give an indication
of the range of uncertainty regarding "plausible" future levels. The accuracy
of items 3 and 4 was discussed in Section 3.1.1.2 and 3.1.1.3, respectively,
and are of poor quality.  The NSPS's (item  5) are known and were assumed
to apply during the study period but are actually likely  to change as discussed
in Section 3.1.1.3. The  reliability of the total loadings calculations is no better
than the reliability of its  parts. The reliability of the impact analysis (item 7)
depends upon the level on which it is being done and the use of the results.'
The projections made for Phase I are for qualitative and comparative
assessments and therefore the overall accuracy was felt to be adequate,
However, in Phase n a more detailed level of analysis will be made which
will involve modeling and dose-response evaluations and will require a
much more reliable data base.

-------
3.1.2 OVERVIEW OF IMPACTS BY ENERGY CYCLE
3.1.2.1 NUCLEAR-ELECTRIC CYCLES

      The surface mining of uranium-bearing ore results in fugitive
particulate emissions from various operations, and gases (sulfuric oxides
(SO  ), nitrogen oxides (NO ), hydrocarbons (HC ), carbon monoxide
(CO3^ from machinery.  The Icale of these impacts is much less than those
from a surface coal mining operation based on equivalent energy production
(25) . The enrichment of the uranium-bearing ore potentially produces the
most severe air impacts of the nuclear cycle due to large amounts of electrical
energy (assumed to be produced by coal).  The enrichment process requires
the equivalent of about 4% of the net product of a boiling water reactor  and
5% of a pressure water reactor (11) .  Of the three enrichment facilities in the
United States, two are located in the ORBES region (Paducah, KY. and
Portsmouth,  OH) and one just outside of the region (Oak Ridge, TN) .  The
Portsmouth facility requires 2000 MWE (26), which is produced by coal fired
power plants  within the ORBES region and  the Paducah plant consumes about
the equivalent of the power produced by the coal fired TVA Shawnee Plant
(1750 MWE) .  The combined installed capacity required for these two plants
is over six percent of the total installed capacity within the region and is
therefore a significant air impact.  The nuclear-electric power plant itself
produces negligible particulates and gaseous emissions. All radioactive
emissions must meet standards which reportedly limit exposure increases
to slightly more than that naturally occurring.  Studies have reported  the
same order of magnitude of airborn radio-nucide emissions from coal fired
plants as nuclear plants.

      Large amounts of heat must be rejected from a nuclear-electric plant;
over a third more than a coal fired plant of equivalent production as shown
in Table 3.1-8.
1,000 MWE Efficiency Thermal input
Power Plant percent MW
nuclear
fossil
32_ 	 a
_- - * — •« C
JS"**^ ' \
f TM I
3,125
2,570
Total heat
rejectedMW
2,125
1,570
Heat rejected
to water MW
2,125
1,250
TABLE IIB-3.1-8: HEAT REJECTION OF FOSSIL AND NUCLEAR,
                 POWER PLANTS (27).

      All of the waste heat from nuclear plants is rejected into water
initially, while about twenty percent of the waste heat from coal plants
is emitted with the hot stack gas.  The impacts from heat rejection into
water from nuclear and coal plants are the same for a given type of
cooling system except for the relative difference in total amount rejected.
The effective limitations on  once-through cooling have resulted in the
widespread use of cooling towers which introduce large amounts of heat
and moisture into the atmosphere. These additions are minute compared
to the natural heat and water balances and are of no global concern as
discussed in Section 3.1.3.5. However, they can be of local significance
causing convective turbulence, fogging, icing, reduced visibility, and
aesthetic problems from both cooling structures and plumes.
                       US-3-17

-------
3.1.2.2. FOSSIL FUEL ELECTRIC CYCLES

      The generation of electricity from fossil fuels inherently involves
certain unavoidable thermal and physical impacts due to thermodynamic
limitations and the chemical composition of the input fuel. The thermal
efficiency of conventional systems appears to be limited to under forty
percent of the systems that offer higher efficiencies (combined cycle
and MHD) are not projected to have a significant impact during the study
period.  If coal were all carbon and combustion were complete, then
only carbon dioxide, water vapor,  heat and nitrogen oxides would be
released. However, in reality coal is chemically very complex and all
of the components plus products of incomplete combustion (carbon
monoxide and hydrocarbons)  must  eventually be deposited in  air, land,
or water receptors.
      As discussed in section 3.1.1.1 coal is projected to dominate the
other energy cycles in the total amount of energy produced. The known,
quantifying air impacts of coal electric cycles on a unit energy output
basis are on an order of magnitude greater than nuclear-electric and
synthetic fuel cycles as discussed in  Section 3.1.2.1 and 3.1.2.3.  The
specific impacts of coal-electric generation are described in sections
3.1.3 and 3.1.4 and a comparison of the scenario impacts is given in
section 3.1.4.1.

3.1.2.3 SYNTHETIC FUELS FROM COAL
      The total number of synthetic fuel plants projected by ORBES Task I
(one high Btu and one low Btu plant per state) for the BOM Scenarios,
represents a small portion of  the total energy production from  coal in the
region.  The actual magnitude of the syn-fuels industry could be much greater
according to national policy and government support provided. For example,
Kentucky is currently supporting projects that could lead to a commercial
scale high Btu gasification plant and  a liquefaction plant, three low Btu
gasification facilities, a solvent refined coal plant, a combined product
plant, and other types of conversion  and cleaning facilities. Projects are
also on-going in other ORBES states and there are aggressive  state programs
designed to encourage the development of the syn-fuels industry.

      The pollutant loadings  shown in Table 3.1-9 for unit gasification
facilities are estimated to be from one half to an order of magnitude
smaller  than  those of a 1000 mw coal-electric generating plant. The unit
gasification plants were sized so that their output would support a 1000 mw
electric plant operating at a 33 percent conversion factor.  It should be
noted that these types of plants have  not been built on the commercial
scale arid emission factors are based  on standards for related operations
and on control technology that could be applied. The Agonne  National
Laboratory assessment of the  health and environmental effects of coal
utilization in the Midwest assumed  a higher number of gasification plants
in its scenarios and reported  the following:

            Emission of airborne pollutants from the gasification train (with
      the possible exception of some  trace elements whose ultimate fates are
      not completely known)  are not expected to be a significant problem.
      Particulate emissions from the  gasification train will be  virtually
                                    UB-3-/2)

-------
                                  Table IIB - 3.1-9

                      Annual Pollutant Loadings from Coal-Electric
                           Plants and Coal Gasification Plants

                                Loadings in Tons/Year
                                                      Unit Low BTU    Gasification Plant (c)
Pollutant

so2
Part.
NO*
CO
HC
1000 MW Coal-
Electric Plant (a)

28.62
2.39
16.70
.91
.34
Unit High BTU Eastern
Gasification Pit... Appalacial Coal Interior Coal
(b)
Ohio
7.87 8.4
.88 .39
8.50 3.76
.11 .28
. 04 . 084
111.
6.6
.36
4.45
.33
.099
Kjr.
9.2
.6
4.41
.31
.092
Capacity Factors:  .60 for 1000 MW power plant, .90 for coal gasification plants
A.
B.
Emission Factors:  SO2> Particulates,  NOX - NSPS
CO - .038 lb/106 BTU input (11)
HC - . 014 lb/106 BTU input based on . 3# HC/TON coal input (29)

Unit plant: 250 X 106 SCFD output (1000 BTU/SCF gas) Emmision factors
from reference (10) Eastern Interior Coal
C.     Unit plant: 2500 X 10  SCFD output of 100 BTU/SCF gas annual loadings from
       ERDA Synthetic Fuels Commercialization Program Draft Environmental Statement (30).

-------
      nonexistent  ...  In many cases  . .  . auxiliary power production
      will contribute more environmental pollutants than the main gasifica-
      tion process stream (10).

      However, there are many areas of uncertainty and the Council on
Environmental Quality stated that "The  necessary environmental informa-
tion for standard setting and other decisions on commercialization of
coal-based synthetic fuels probably will not be available by the mid-
1980's" (28).

      Thus for both nuclear-electric (Section 3.1.2.3) and coal gasifica-
tion cycles the majority of the total air borne emissions come from
coal-fired boilers for which  the impacts are described in Sections 3.1.3
and 3.1.4.
3.1.3 ALTERATION OF BASIC ATMOSPHERIC PROPERTIES
3.1.3.1 INSOLATION

      Emissions from power plants can contribute to insolation reduction
on local, regional, and global scales in several ways. The first, and
probably most significant, is due to fine particles which are very
effective in reflecting, diffusing, and absorbing light.  The accumula-
tion of such fine matter in the atmosphere is believed to cause a slight
reduction in insolation on the global scale  (4).  Particulate emissions
from modern power plants with high efficiency precipitators are in the
smaller size ranges, with 30 to 40% below 20 microns (3|).

      Sulfates represent 40-60% of the respirable suspended particulate
(RSP) in most areas and therefore represent a major component of atmos-
pheric fine marticles (32).

      Nitrogen oxides in urban power plant plumes may enter into the
complex reaction chains that produce photochemical smog, which can cause
insolation reduction on a regional scale.

      All of these mechanisms through which plant emission may contribute
to insolation reduction are the object of a number of current research
projects.  No quantitative statements can be made at this point, but it
can be safety assumed that scenarios that involve the largest amount of
coal combustion will result in the greatest insolation reduction.  The
problem in insolation reduction is considered to be a relatively low
priority impact area as it relates to the  ORBES study.

3.1.3.2 VISIBILITY
      All of the mechanisms described  in section 3.1.3.1 by which power
plant effluents can cause insolation reduction can obviously also contribute
to visibility reduction. The smog and haze that can result to some degree
from generating facilities emissions can decrease contrast between distant
obj ects and lower visibility.  The exact contribution of power  plants is
not known,  but a National Academy of Engineering  committee concluded
that "it is clear that the generation of electricity does make a signifi-
cant contribution to the man-made production of visibility-reducing aerosols
and particles. "  Sulfates  have been identified specifically as major con-
tributors to reductions in visual range  caused by atmospheric aerosols (34) .
Induced fogging from cooling towers can also cause local short-range visi-
bility problems depending upon the weather conditions and local  topography.

                              XTB-3-2.0

-------
      The EPA has compiled national visibility isopleths for July, 1974,
and is planning to assemble 20*0? such data (33) .  These isopleths were
calculated from human observer visibility measurements as recorded at
146 airports.  The EPA plans to compare these visibility isopeths with
those for sulfate, ozone, and stagnation and perform case history studies
of the role of long-range transport for  selected time periods.

      An example of the visibility isopleths is shown in Figure 3.1-5.
For that day the area of greatest reduction covered most of the ORBES
region and much of the industrialized Northeast.  Preliminary analysis
of visibility data indicates that the ORBES region does have significant
existing visibility reductions which may be due in large part  to naturally
occuring conditions. However, contribution from industrial activity and
coal-electric generation activity can only worsen the condition; therefore
the higher the levels of these activities by scenario, the greater the impact
on visibility reduction.  Subsequent years of the ORBES study should
monitor  the research described above to develope a more quantitative
evaluation.

3.1.3.3  AIRFLOW PATTERNS AND TURBULENCE FACTORS

      Modification to airflow patterns and turbulence from energy facilities
will be local and not major impacts.  Because of the termendous amount of
heat rejected by large generating stations, there will be local convective
turbulence and "heat island" effects similar to those observed over urban
areas.

3.1.3.4  NOISE

      Noise from coal-electric plants can come from a number of operating
areas and from a general increase in vocal activity.  The coal unloading  and
handling systems may be major sources with additional sources being high
speed pumps,  transformers, cooling towers, and draft fans.  The discharges
from safety valves and blow-off valves is a source of considerable annoyance
but can  be controlled relatively successfully.  A National Academy of Engineer-
ing committee stated that successful solutions were available for noise problems
"as evidenced by the fact that generating stations have been accepted as  close
neighbors in urban  areas" (4). It is questionable to extrapolate the "acceptance"
of noise annoyance is highly subjective and could vary greatly depending on the
socio-economic status of the plant neighbors among  other factors.  Also a given
level of  ambient noise increment may be far more noticeable and adverse in a
"quiet" rural setting than the same  increment would be in an existing  "noisey"
urban area.
      The scenarios that involve the greatest numbers of plants will produce
the highest level of noise impact because of the greater area and population
affected. Overall, the noise impacts are not viewed as major.

3.1.3.5  HEAT  AND MOISTURE

      The amount of heat rejected by energy plants is described in Section
3.1.2.  While the total amount is very large and represents a waste of energy,
it is not of global concern because the  total contribution of man's activity to
the total heat load is estimated  to be much less than  0.1%. Similar analysis
on the local scale concluded that a 1000 MW power plant represents a large
but not enoumous impact.  Likewise the amount of moisture added from cooling
towers is insignificant compared to the global balance and will cause only

-------
                                                     BEXT  FOR  27 JUL  74  1200C
Figxore 3.1-5 National Visibility
Isopleths for 27 July 1974
(from reference 33)

-------
minor changes in the average temperature and humidity in the plant vicinity
when the terrain is flat to gently rolling.  (4) . However, if the terrain is
rugged, serious problems can arise from  a cooling tower plume reaching
the ground. Given the appropriate weather condition fogging, light drizzle,
or freezing could occur.  An Academy of Engineering committee further stated
that "no cooling tower is likely to create thunderstorms in areas where nature
does not, and generally clear regions are not suddenly going to become cloudy.
At most, the tower plume could influence flightly the frequency and location of
natural phenomena" (4) .  Any significant adverse heat and moisture impacts
appear to be site specific in nature and beyond the scope of this analysis.

3.1.4 COMPARATIVE IMPACTS OF SCEANRIOS

      The dominance of coal-fired electric power plants in terms of scenario
air quality impacts, both because of numbers of facilities and emission rate
per facility, was discussed in Section 3.1.1. and 3.1.3.  Because of that
dominance the regional aggregate air impact analysis of energy facilities
can be reduced to consideration of only the coal-electric plants.  More
localized impacts in terms of emission rates  and expected local concentrations
have been discussed in general terms in other sections and more site specific
analysis is beyond the scope of Phase I.

      The total annual loadings for sulfur dioxide from coal-electric plants
and shown for the ORBES portion of each  state and for the total region in
Figures 3.1-6 through 11 and for particulates in Figures 3.1-12 through 17.
The procedure used to develop those data and its reliability was described
in Section 3.1.1.4.  The reliability of the input data was such that the figures
should be considered qualitative and for comparative purposes.  However, they
are useful in several ways.  First they illustrate the very large portion of the
future  air pollution loadings that are "built in" because of existing facilities.
For example the emissions from "existing" plants in the year 2000 for sulfur
dioxide make up from forty to ninty percent of the total, depending upon the
scenario.  This means that future energy  development could be constrained
unless emissions from existing plants are reduced even further than now
required.
      The figures also show that from a state and regional air quality
perspective the Ford Tech Fix scenarios are not significantly different from
each other.  Conclusions can also be drawn regarding background air quality
levels and the amounts of pollution carried out of the region by long range
transport relative to those observed today.  The regional  and state level
ambient .background levels should generally correspond to pollution loadings,
that is a significant decrease in loadings should result in lower background
levels, while future emissions rates of the same magnitude will cause the
same order of magnitude ambient background levels.  This would not be true
for sub-areas in the region which experienced a high density of emissions,
such as stretches along the Ohio River.  The same general reasoning can be
applied to transport of pollutants, especially sulfates, out of the region.  The
power  plants portion of the sulfates transported out of the region will vary
directly with the magnitude of the loadings.

      Data on the local impacts of facilities, such as maximum ambient concentra-
tions and PSD consideration  are presented in Section 3.1.1.3. The next level of
comparative analysis would be to determine loadings on a county and multi-county
bases and evaluate exposure factors.  However, the ORBES  study is an iterative

-------
process and there is sufficient basis to cause revision the RTC's from air quality
and other considerations before going to more detailed analysis.

      The U-shape of the loading curves for some scenarios also raises
policy questions. Any scenario in which total loadings decline rapidly
reach as minimum, and rise again, bring up the question of the desirability
of "overshoot" or going past a target level.  There may be overiding localized
factors which may necessitate going through a minimum such as the ORBES
total in the BOM 80-20  (Figure 11) scenario. For example the minimum loading
level observed in 1985 is comprised of higher local densities than the higher
levels observed for  the year 2000, represent a greater number of dispersed
plants emitting at lower rates. However, based on consideration of the total
loadings alone, and background levels along with other regional impacts
related to those loadings, it would seem most desirable to reach a future
target level by a continual decline which does not go through a significant
minimum.  Such an overshoot represents excessive control and therefore
excessive cost to both the consumer and producer. On the other hand, if
the minimum level is actually  the desirable target level, then future policy
should be formulated to prevent increas^ beyond that level.
                                   tf B-3-2.4

-------
    2.0 r
o:
a:
UJ
a.
o


o
    1.5
    1.0
                        FIGURE 31.-6

          S02 EMISSIONS FROM KENTUCKY POWER

              PLANTS  BY SCENARIO
                                                  BOM 80-20
                                              BOM 50-50


                                              TECH  COAL

                                              TECH  NUCLEAR

                                           —-•EXISTING
                                              PLANTS
     1975
         1980
1985
  1990

YEAR
1995
2000
   2.0
tr
<
UJ
LU
QL
O
1.5
    1.0
                        FIGURE 3.1-7


          S02 EMISSIONS FROM  ILLINOIS  POWER

              PLANTS BY  SCENARIO
                                                  BOM 80-20


                                                  BOM 50-50
                                              TECH COAL

                                              TECH NUCLEAR
     1975
         I960
1985
 1990

YEAR
1995
2000
                           U&-3-Z5

-------
   2.0 r
o:
<
bJ
cc
UJ
CL

    1.5
    1.0
                                                    BOM 80-20
                                                    BOM 50-50


                                                    TECH COAL

                                                    TECH NUCLEAR

                                                    EXISTING
                                                    PLANTS
    -5
                        FIGURE 3.1-8
          S02 EMISSIONS  FROM INDIANA^   POWER

              PLANTS  (STATE  ESTIMATE*)
                                JL
                                                  J
     1975
             1980
                   1985
          1995
        2000
   2,0
1.5
CC
<
LU
tr
LJ
Q_
P

z
o
13   .5
                       1990

                      YEAR

*  Interpretation of allowable emissions        !
   from existing power plants, see section 3.1.112.
                           FIGURE 3.1-9

             so,  EMISSIONS'. FROM INDIANA  POWER
                 Lw I ₯ I I SM^N^ I ^S • ~9\^ "ll» -*^ • • •  ...— --  --

                 PLANTS (EPA  ESTIMATE*)
                                                    BOM 80-20


                                                    BOM 50-50
                                                   TECH COAL
                                                   TECH NUCLEAR
     1975
              1980
                   1985
 1990

YEAR
1995
                                        2000

-------
o:
<
UJ
cr
UJ
a.
a:

UJ


a:
UJ
o.

c/)
z
o


z
o
_J
   2.0 r
    1.5
    1.0
— i   .5
     1975
   6.8
   -5,1
    1975
                     FIGURE- 3.1-10


          S09  EMISSIONS FROM OHIO POWER
              PLANTS  BY SCENARIO
             1980
1985
  1990

YEAR
                  FIGURE 3-. 1-11


          SO, EMISSIONS ORBES  TOTAL
              1980
1985
 1990

YEAR
1995
1995
                                                   BOM 80-20
                                                   BOM 50-50


                                                   TECH  COAL

                                                   TECH  NUCLEAR


                                                   EXISTING

                                                   PLANTS
2000
                                                  BOM 80 - 20


                                                  BOM 50-50
                                                  TECH COAL

                                                  TECH NUCLEAR
2000
                          ££-3-2/7

-------
   200
o:
S  150
>-
tr
u
a.
O
h-
en
D
O
X
I-
   100
 50
    0
                         FIGURE 3.1-12

         PARTICULATE EMISSIONS FROM KENTUCKY POWER

                     PLANTS BY SCENARIO        •
     1975


  200
cc
<
LJ
>
(T
LJ
Q.
150
   100  -
Z)
O
X
H-
50
    0
     1975
                           BOM 80-20

                           BOM 50-50


                           TECH-COAL
                           TECH-NUCLEAF
                           EXISTING
          1980
1985     1990
    YEAR
1995
2000
         PARTICULATE EMISSIONS FROM ILLINOIS POWER
                       PLANTS BY SCENARIO

                      FIGURE 3.1-^13
                       I
                            I
                                             BOM 80-20
                                             BOM 50-50
                          TECH-COAL
                          TECH-NUCLEAR
                        -•EXISTING
                          I
          1980     1985     1990

                       YEAR
                1995
       2000
                        JTB-3-2.B

-------
    200
 cr
 <
 LJ
 o:
 LJ
 a.
CO
D
o
X
 150
    100
 50-
               FIGURE 3.1-14                  ~
      .PARTICULATE   EMISSIONS FROM
      INDIANA  POWER   PLANTS  BY SCENARIO
                                               BOM 80-20
                                               BOM 50-50
                            TECH COAL
                            TECH NUCLEI
                           . EXISTING
                            PLANTS
o:
<
Id
UJ
Q_

CO

O
Q
CO
Z5
O
'X
     1975
   200
150
100
 50
          1980
1985
  1990
YEAR
1995
2000
                                               BOM 80-20

                                               BOM 50-50
             FIGURE 3.1-15  .
      PARTICULATE   EMISSIONS  FROM
      OHIO POWER   PLANTS  BY  SCENARIO
                            TECH COAL
                            TECH NUCLEA
                         --EXISTING
                            PLANTS
     1975
          1980
1985
  1990
YEAR
1995
2000

-------
   800
                FIGURE 3.1-16

PARTICULATE EMISSIONS FROM TOTAL ORBES POWER

             PLANTS BY SCENARIO
cr
cc
LJ
Q_
   600
   400
O)
=>
   200
                                               BOM 80 -20
                                                BOM 50-50
                                      TECH-COAL
                                      TECH-NUCLEAF

                                    — EXISTING

                                      PLANTS
     0
      1975
                              1
    1980
1985    1990

    YEAR
1995
2000

-------
                Appendix

                  3.1-A
        Power Plant Emissions  Date
for Participates and Sulfur Dioxide by State
                3TB-3-3I

-------
                                                EMISSION INVENTORY FOR KENTUCKY POWER PLANTS (1)
                                                                                                                  Prepared by IMMR for ORBES








N
53

Uj
I
u>
1°





Utility
KEUC
KEUC
KEPC
OWMU
BREC
HEPL
LOGE
LOGE
LOGE
KEPC
TVA
KEUC
KEUC
TVA
EKPC
BREC
KEUC

Plant Name
Pineville
Ghent
Dale
Owensboro
Hawesville
Henderson
Paddys Run
Cane Run
Mill Creek
Big Sandy
Shawnee
Brown
Green River
Paradise
Burnside
Reid
Tyrone

(County)
(Bell)
(Carroll)
(Madison)
(Daviess)
(Hancock)
(Henderson)
(Jefferson)
(Jefferson)
(Jefferson)
(Lawrence)
(McCracken)
(Mercer)
(Mulhenberg)
(Mulhenberg)
(Pulaski)
(Webster)
(Anderson)
County
Code
200
580
720
920
1580
1760
1920
1920
1920
2140
2460
2760
2960
2960
3460
4020
4140
AQCR
Code
101
79
102
77
77
77
78
78
78
103
72
102
72
72
105
77
102

P articulate
1,860
1.088
5.528
1.416
1,505
153
234
2.047
399
18,439
9.790
18.023
6.116
4.986
17.974
5.189
2,042
	 jjjj
(T\
Part A11CZ)
(420)
(1,804)
(1.296)
(2.050)
(3,154)
(207)
(1.567)
(5.710)
(1.629)
(8.390)
(6.630)
(11.615)
(1.460)
(10.693)
(3.955)
(14.503)
(251)
ussion usuro
so2
2.130
42.682
7.638
69.577
97.700
__
4.764
137.657
63,389
49.577
249,750
76,132 •
31,512
537.529
33,619
78.706
1.599
ates ions/ tea
(2)
S02 AIT '
(4.176)
(85.911)
(20,680)
(22.158)
(31.600)
(4,383)
(8.425)
(43.539)
(15,899)
(132,440)
(118,320)
(103,430)
(15,782)
(181.927)
(44,480)
(63,637)
(3,648)
r 	
NOx
612
8,877
4,578
16.265
13.284
	
1.792
15,196
7.917
23.953
71.440
13,483
4,616
103.127
7.407
—
1,588

HC
10

49
154
221
__
23
290
132
399
710
225
77
1,031
123
—
17

CO
34

163
513
737
	
87
752
440
1,331
2,380
749
257
3.439
412
—
53
Sulfur
Content
1.6
2.5
1.2
3.5
3.5
	
3.4
3.8
3.8
.9
2.8
1.6
3.2
4.1
2.2
3.5
.8
                       TOTAL
96.789
(75.334)   1.483.941    (900.435)    294,135
3.461   11.347
(1)  Obtained from the Kentucky Division of Air Pollution Control
(2)  Allowable emissions under current regulations

-------
  Tables of power plant
emissions for 111., Ind.,
 and Ohio will be included
    in the final draft.

-------
           Appendix

            3.1-B
Location of  SURE Monitoring Sites
        and Sample Data
            OTB-3-36

-------
  SURE Class  I  Stations

     100A Montague, MA
     102   Scranton, PA
     103A Indian River,  DL
     lO^B Philo,  OH
     106   Rockport, IN

  SURE Class  II Stations

       1   E. Beverly, MA
       2   Fall River, MA
       3   Albany, NY
       4   Osv/ego, NY
•4       5   Dunkirk, NY
*       6   Roseton, NY
       7   Allegheny, PA
       8   Lewisville, PA
       9   Brush Valley,  PA
      10   Gettysburg, PA
      11   Delmarva, DL
      12   Gavin;  OH     >•—.
      13   Clifty Creek,(OH
      l4   Big  Sandy, OH^""1
      15   Breed,  IN .
      16   Munroe, MI
      17   Port Huron, MI
      18   Kincaid, IL
      19   Collins, IL
      20   Picv/ay, OH
      21   Jay, ME
      22   Toronto, Ont
      23   Huntington, NY
107   Land-between-Lakes, KY
109   Chapel Hill, NC
111A  Roanoke, IN
112   Lewisburg, WV
      Loves Mill, VA
      Hytop, AL
      Giles City, TN
      Paradise, KY
      Memphis, TN
      Hanover, NH
      Benton Harbor, MI
      St. Louis, MO
      Niles, OH
      Minneapolis, MN
      Galesburg, IL
      Mount Storm, WV
      Chesterfield, VA
      Yorktown, VA
      Riverbend, NC
      Weatherspoon, NC
      Atlanta, GA
      (Upstate New York)
      Columbia, SC
      Cayuga, NY
      Dan River, NC (?)
      Lafayette, IN
                    25
                    26

                    28
                    29

                    31


                    36
39
4o
                    ^3

                    47

                    49
(Future SURE air monitoring sites
  reproduced from reference 12)
                                                                                    STATE OUTLINE
                                                                                UNITED STATED
                             I  SURE Class I Station

                             o  SURE Class II Station
                                             Vl/77

-------
«
U>
 I
CP
           I   2   3   4  5  6   7  8   9  10  II  12   13  14  15  16  17  18  19  20 21   22 23 24  25  26
            (Past sites for SURE  air  monitoring stations, reproduced from reference  12)
                                i    Sites for Annual Data  Set  as Indicated by Square  (01)

-------
                                                               Table 6-2


                                        STRATIFICATION OF SO °°. TSP,  AND SO, BY THREE WEATHER TYPES
 i
Ui
 i

UJ
Synoptic
Weather

No Precipitation


Precipitation


Fog


TSP
No Precipitation


Precipitation


Fog


so2
No Precipitation


Precipitation


Fog




c*
a
N
c
a
N
c
0
N

C
a
N
c
a
N
c •
0
N

c
0 '
N
C
a
N
c
o
N
.ockport , :
(K
11.9
6.7
100
11.8
6.2
114
16.4
9.5
54

60.4
42.4
100
49.1
27.5
114
59.9
26. 5
54

12.6
4.8
. 60
13.1
5.0
71
21.2
22.7
33
ullivan, :
10
7.8
6.9
50
8.8
3.2
64
11.6
3.4
30

36.8
19.6
SO
31.9
14.6
64
50.2
22.2
30

21.1
14.0
47
22.3
11.9
64
30.1
19.4
31
ollins, 11
o
9.2
5.1
151
9.9
5.9
100
15.2
10.5
36

59.3
34.6
151
55.1
44.6
100
63.3
34.3
36. "

27.7
26.0
145
23.2
19.4
86
40.4
45.8
37
.adison, I(
X
12.0
7.1
118
12.2
6.3
118
18.8
10.0
57

49.0
31.0
118
38.0
21.3
118
62.9
25.8
57

41.9
25.1
94
31.2
17.7
94
51.0
33.7
50
aukegan , ]
• Z
8.5
8.3
114
10.6
6.9
67
14.5
9.5
34

55.4
36.7
114
49.5
30.1
67
86.6
47.6 '
34

16.8
17.3
113
23.6
19.6
67
23.8
15.9
33
awrencebui

10.5
7.1
147
11.1
7.4
126
19.6
11.9
66

61.2
39.8
147
45.8
28.8
126
78.4
38.2
66

44.9
34.3
113
36.6
37.2
107
46.4
33.6
56
awrencebui
^
11.2
7.9
138
11.2
8.3
121
20.3
11.7
62

59.3
40.7
138
41.1
29.1
121
84.1
43.4
62

46.7
41.6
118
32.6
26.6
108
60.3
35.4
59
untington,
X
8.9
4.3
77
10.4
.8.1
129
14.5
9.4
117

38.0
15.8
77
33.6
20.7
129
50.1
23.0
117

23.5
18.5
76
16.7
19.3
131
18.8
22.4
116
heeling, V
s
15.9
11.1
117
14.5
8.6
112
20.2
14.3
59

67.9
35.7
117
55.8
28.5
112
77.7
43.5
59

53.5
44.2
102
46.4
32.4
88
80.8
65.0
50
cranton, E
tO
9.3
5.2
158
10.3
5.6
111
13.2
6.2
40

61.4
40.1
158
37.8
22.2
HI
63.0
36.6
40

47.8
44.6
149
40.7
29.2
101
65.8
42.7
33
z
6
4)
10
0
7.8
5.7
75
6.6
3.4
62
12.8
7.7
17

28.2
21.5
75
18.6
105
62
38.9
20.3
17

9.9
6.7
50
7.7
5.2
46
17.8
13.0
14
X
z
c
10
•H
**
7.3
5.0
140
9.6
65
118
12.4
8.0
52

37.6
20.1
140
36.7
20.6
118
46.7
18,5
52

19.3
17.9
126
22.0
21.3
105
16.4
11.5
50
                                                                                                                                 c/i


                                                                                                                                1
                                                                                                                                 h-1
                                                                                                                                 CD

                                                                                                                                 C/i


                                                                                                                                 M
h"
rt
O
h
H-
2
OP
                                                                                                                                 0)
                                                                                                                                 o
                                                                                                                                 O.
                                                                                                                                 C
                                                                                                                                 O
                                                                                                                                 CD
                                                                                                                                 O
                                                                                                                                 3
                                                                                                                                 CD
               is where  concentration,  o is standard deviation,  and N is  sample  size.

-------
            Appendix

              3.1-C
Siting Constraints to Coal Fired
Power Plants Posed by Prevention
 of Significant Deterioration
Policy.  (Taken from "Issues and
Impacts Associated with Proposed
   Prevention of Significant
Deterioration and Non Attainment
Amendments to the Clean. Air Act"
          Reference 18)

-------
           (Reproduced from reference 18)
                                  CHAPTER III

                                SITING ANALYSIS


      The purpose of the siting analysis  was  to determine whether planned
 powerplant capacity could be sited given the proposed PSD  increments  and
 other major sources of pollution.

      Below,  the  findings of  this  analysis are  presented, in four parts.  First,
 the  effects  of the  proposed  PSD increments on  the size and location of indi-
 vidual powerplants  in  isolation are summarized.  Then the  findings of a co-
 location analysis — which assessed the  effects of the PSD increments on
 powerplant siting when numerous powerplants  and other major sources of pollu-
 tion are located in the same geographic  area — are presented.  Next, the
 conclusions  drawn from these analyses  are discussed.  Finally, the relevance
 of these conclusions to the  PIES analysis is discussed.

 CONSTRAINTS  ON INDIVIDUAL  POWERPLANTS                   1          '

      Atmospheric  dispersion  modelling  is  the analytic technique that must be
 employed to  assess  the  effects  of the proposed  PSD increments on the-  size and
 location of  individual  powerplants  in  isolation.  Such analyses have been
 performed, by EPA  and by Environmental Research and Technology, Inc. (ERT)  for
 FEA.—   The  findings summarized below are based on a review of these studies.

      Findings

      For  a 1,OOP MW coal-fired ppwerplan.L_that-.CQmpJLies with new_source per-
 formance _s_tandards_(NSPS) and has a 5,QO-fg^.t_stac)^,_^e^^ects'^of^the_PSD'
 increments can be summarized as follows:

          1.  The increments for sulfur dioxide (S02>  limit power-
           .  plant sitings more than the increments  for total sus-
              pended particulates  (TSP).   By  themselves, the TSP
              increments can be met readily by  controlling  parti-
              culate emissions to  NSPS.

          2.   The Class I increments are  so small  that  a new plant
              would have to be located at least a  certain  radius
              from the  boundary of a Class I  area.  The  size of
I/  EPA work reported in Summary of EPA Analysis  of the  Impact of  the  Senate
    Significant Deterioration Proposal distributed in May,  1976.   There  is
    not yet a final report on the ERT work.
                                  ZTB-3-<4

-------
      (reproduced..from reference  18)
                this radius is uncertain.  EPA has estimated this
                "buffer zone" radius to be 60 miles.  ERT has esti-
                mated 75 to 150 miles. 2/

            3.   The Senate Class II increments would not prevent the
                construction of a powerplant as specified above, so
                long as terrain features do not rise more than 500
                feet above the top of the stack within about 15 miles
                of  the plant.  The House Class II increments might
                prevent construction since the House three-hour aver-
                age S02 increment is less than one-half the Senate.
                increment.

            4.   According to EPA, _in flat terrains two_. 1 . OOP ._MW
                powerplants^ could be located very _nejar._ each other .
                However,  in moderate terrains, twoJl.jD.OJl_MW_lpQw_er-
                plants_would_have to be located about 28 miles apart
                in  order to avoid" viplation~bf the Class II incre-
               ments.., _ In_h JLl 1_y_ terra i n ,  hhpy— u^yi A have to be
               located  about 40 miles  eipartj-j/

           5.  Class III increments  that  were  about twice the
               Senate Class II increments would enable about twice
               the powerplant capacity to be located at any one
               site in  a Class III area as  in  a Class II area.
               The House Class III increments  would probably not
               do this, however, because  the House  Class III three-
               hour increment is actually less than the Senate
               Class II three-hour increment.  (650  vs.  700 ug/m^) .

           6.  The size of the Class I buffer  zone  radii and Class
               II separation distances reported above can be reduced
               by controlling emissions below  new source performance
               standards levels .        •

      Uncertainties

      However ,  there are substantial — undez±aintie5_^ssp.ci.afced_with_ atmospheric
 dispersion jnod^ls^..__The.se_unce.rtajLn^ies__are indicated  by the  range  of  the
 estimates^ of_the__(^a_ss_J_JkuJf^r_zpj}e^^                    150  miles.

      The se ujxcejda.inties are also indicated by  d_jff_er^nt_(astimatej_gf_the
 Class_II separation_.di.st.ances.   The EPA estimates are  rej?orjbed_above —  0;
 to 40. .miles depending _on_tarxain .   I" a study for FEA,  ERT  has _repjor ted. -that
. the uncertainties are so ^r^at^haJL-^e_Jlm.iju.n!um plant separation_distance
 could range from O-lOO^jniles, or greater . "Q/
 2/  These estimates apply to the PSD provisions proposed by the House.  The
     Senate provides that for Federal.Class I areas, the Federal Land Manager
     must make a finding as to the impact of any new source on the "air quality
     related values" in the Class I area under his jurisdiction.  Hence, the
     actual radius might be larger or smaller than indicated above.

 3/  It is possible that these separation distances would be greater than  indi-
     cated above if the House three-hour increment (which is about one-half
     the Senate three-hour increment) wer^ enacted.

 4/   Memo  from Arthur Bass,  Senior Staff Scientist of ERT to Roger Morris of
     FEA dated April 27,  1976.

-------
       (Reproduced  from  reference  18)


      Further,  these  uncertainties are evident by the way EPA's assessments
 of  the effects of  the Class II increments have changed over time.   Initially,
 j,n  Technical Support Studies,  dated January,  1975,  EPA reported that "a
 powerplant of  600-jOO MW shouTd"Ti?t~e'ndahlfeT~tH^
J24-hour^averagestjl_and. thalf "a lj000 MW 2lant_would~ail[ow~most of  the plants
 to  meet the allowable increments, while a 500 MW plant would ensure that all
 plants would meet  the allowable increment."$/  Hence,  the thrust of these
 atmospheric dispersion model_fi£^ings^a^jbha1^^^nts_perhaps_as_J.arge as
j., OOP MWjrould be  sited, in  Class JCI_ar_ea.sj"_relative to the EPA 24-hour "aver-
 age incremenF~TwH:Cch~"is~the same in the proposed Senate amendments but about
 10  percent higher  than in the  proposed  House  amendments).   No mention was
 made that  the  size of these plants were limited by  hilly terrains.   However,
 recently in "Summary of EPA Analysis of the Impact  of  the'Senate Significant
 Deterioration  Proposal,"  which was distributed in May,  1976,  EPA reports that
 "between an 1,100  to greater than 4,000 MW coal-fired  powerplant ...  could
 be  built in areas  of flat or moderate terrain."§/  Hence,  the thrust of EPA's
 more recent findings is that 1,000 MW is now  the minimum size (in  non-hilly
 terrains)  and  much larger plants are probably allowable.

      Finally,  there  are uncertainties associated with  what the Class II  incre-
 ments, might be.  The findings  presented above relate to the Senate  increments.
 As.mentioned,  the  House 24-hour average increment is about 10 percent lower
 and the  House  three-hour  average  is over 50 percent lower.   The House three-
hour average could be much  more restrictive than the constraints listed  above.

      Accordingly,  the  findings  of any analysis  (including  those presented
below)  that relies on atmospheric dispersion  models should be interpreted
within the  perspective  that such .modeling J.3..JS t i 11, a ..very _ ijnprecisejsc ience.

CO-LOCATION ANALYSIS

      In order  to assess the  effects  of  the proposed PSD  increments  on the   .
 siting of powerplants when  all  planned  powerplants  and other  major  emissions
sources are considered  together,  a  series of maps were developed.   These maps
provide a visual representation of  the  effects of the proposed  PSD  provisions
on  the siting of powerplants and  other major,  facilities.   See Pages 111-11-17.

      Class  I Increments                                 .                     .

      Map 1  shows the mandatory Class I  areas according to  the House version
of  the PSD  provisions.  A 60-mile buffer zone has been drawn  around the peri-
meter of each area to indicate  the  area within which a 1,000 MW powerplant
meeting NSPS and having a 500-foot  stack would violate the Class I  incre-
ments.

     Map 2  illustrates the effect of the Senate mandatory Class  I areas.  A
comparison  of the  first two maps  indicates that  the Senate has more Class
I areas, particularly in the East.
5/  Pages 101-102 and 103, NTIS. PB-240-215.

6_/  Pages 6-7.

-------
                                    REFERENCES

                                    Section 3.1
  1.   L.  R.  Babcock,  "Ratings  of  Pollutants by Effect",  J.  Air Pollution
      Control Assoc.,  1,  p.  727-729,  1972.

  2.   "Development  of Plausible Future Regional Technology  Configurations"
      Ohio River Basin Energy  Study Tasic 1 Report,  October 18, 1976.

  3.   1973 National Emissions  Report,  U.S. Environmental Protection Agency
      National Air  Data Branch, EPA-450/2-76-007,  May,  1976.

  4.   Engineering for the Resolution  of the Energy-Environment Dilemma,
      National Academy of Engineering, Washington, D.C., 1972

  5.   An Integrated Technology Assessment of  Electric Utility  Energy Systems
      Draft First Year Report  Vol.  II.. U.S.  Environmental  Protection Agency,
      Office of Research and Development, 1977.

  6.   The Sixth Annual Report  of  the  Council  on Environmental  Quality, CEQ,
      December, 1975.

  7.   V.  Sharma, et al., "Regional Atmospheric Transport of Coal-Fired Power
      Plant Emissions", published in  Health,  Environmental  Effects, and
      Control Technology of  Energy Use, U.S.  EPA Office of  Research and
      Development,  EPA 600/7-76-002,  no publication  date given (meeting
      Feb., 1976 proceedings).

  8.   Guidlines for Air Quality Maintenance  and Analysis Vol.  14.  Designation
      of Air Quality Maintenance  Areas, U.S.  EPA,  EPA-450/4-75-002, December,
      1975.

  9.   Federal Register. Vol. 41,  No.  134, p.  28600,  July 12, 1976.
 10.   A Preliminary Assessment of  the Health  and  Enviro'ment'al Effects  of
      Coal  Utilization  in  the  Midwest.  Draft Volume  I, Argonne National
	    Laboratory,  January,  1977.
 11.   An Integrated Technology Assessment  of  Electric Utility Energy Systems
      Draft First  Year  Report  Vol.  II., U.S.  Environmental Protection  Agency,
      Office of  Research and Development,  1977.

 12.   G.  M.  Hidy,  et al.,  Design of the Sulfate Regional Experiment  (SURE),
      Vol.  I;  Supporting  Data and Analysis,  Electric Power Research Institute,
      EPRI  EC-125,  February, 1976.

 13.   William E. Wilson, Sulfates  in the Atmosphere:  A Progress Report on
      Project  MISTT (Midwest Interstate Sulfur Transformation and Transport),
      U.S.  EPA.  Environmental Sciences Research  Laboratory, EPA-60017-77-021,
      March, 1977.

 14.   Air Quality  and Stationary Source Emission  Control, Prepared for the
      Committee  on Public  Works, United;-States Senate, Serial No. 94-4,
      March 1975.

-------
15.  D.W. Locklin, et al.,  Power Plant Utilization of Coal, Battelle Columbus
     Laboratories, September, 1974.

16.  Frank T. Princiotta, Sulfur Oxide Throwaway Sludge Evaluation Panel
     (SOTSEP), Vol. II. Final Report - Technical Discussion.  U.S. EPA,
     EPA-650/2-75-010-6, April, 1975.

17.  R. Bright, et. al., Air Quality Policy Analysis of Electric Utilities;
     A Regional Perspective; Argonne National Laboratory, March, 1975.

18.  Issue and Impacts Associated with Proposal Prevention of Significant
     Deterioration and Non-Attainment Amendment to the Clean Air Act;
     Federal Energy Administration, FEA/G-76/313, June 25, 1976

19.  Coal-Fired Power Plant Capital Cost Estimates, Electric Power Research
     Institute, EPRI AF-342, January 1977

20.  "Environmental Reporter Current Developments" Vol. 7, No. 25, Oct. 22,
     1976

21.  "Environmental Reporter Current Developments" Vol. 7, No. 21, September 24,
     1976

22.  V.P. Biniek and Marialt. Grimes, "Air Quality: Prevention of Significant,"
     Library of Congress Congressional Research Service, Issue Brief Number
     IB74039, August 19, 1976.

23.  "Pre-Construction Review and Preliminary Determination for Units 3 and 4
     Proposed for Construction at Kentucky Utilities Company's Ghent Power
     Station"  EPA Region IV Air Programs Branch, March 1, 1977.

24.  "Pre-Construction Review and Final Determination for the Trimble County
     Generation Station Units 1, 2, 3 and 4 Proposed for Construction Near
     Wises Landing, Kentucky" EPA Region IV, March 25, 1977.

25.  "Energy Alternatives:   A Comparative Analysis," The Science and Public
     Policy Program, University of Oklahoma; Norman, Oklahoma, May 1975.

26.  "Moody's Public Utility Manual," p. 56, 1976.

27.  Way, S., "MHD Prospects in the USA," Energy International, £, No.  4,
     p. 14-18, 1971.

28.  Enviroment and Conservation in Energy Research and Development, Council
     on Enviromental Quality, September, 1976.

29.  Energy From Coal: Guidelines for the Preparation of Enviromental Impact
     Statements, Battelle Columbus Laboratories, PB-242960, April, 1975.

30.  Synthetic Fuels Commercialization Program Draft Environmental Statement,
     ERDA-1547, December, 1975.

31.  V. Smil, Energy and the Environment; A Long Range Forecasting Study,
     The University of Manitoba, 1974.

-------
32.  Statement of Sulfates Research Approach,   U.S.  Environmental Protection
     Agency, EPA-60018-77-004, February,  1977.

33.  William E.  Wilson, Memorandum:  "National  Visibility Isopleths," U.S.  EPA
     Environmental Sciences Research Laboratory,  Research Triangle Park,  N.C.,
     April 4, 1977.

34.  William E.  Wilson, "Sulfates in the Atmosphere: A Progress Report on Project
     MISTT,"  U.S. EPA Environmental Sciences Research Laboratory, EPA-60017-021,
     March, 1977.

35.  Report of the Interagency Workings Group on  Health and Environmental
     Effects of Energy Use, Council on Environmental Quality,  PB-237 937,
     November, 1974.

-------
3.2 WATER IMP ACTS

3.2.0 METHODOLOGY

      The Adverse Impact Matrix or AIM was designed to assure full subject
coverage for water impact in the ORBES study. Five major column headings
represent the category^ production. The sixth column allows for cross-
referencing so interdependencies of topics are considered. The row titles
are 11 impact categories under the major headings of changes in ground
and surface water quantity and changes in water quality.  The matrix
topic squares were first evaluated as to severity of potential adverse
impact - light, moderate or heavy.  Although the AIM is designed to
evaluate adverse impacts, the beneficial impacts are also considered in
the body of this section.  There are several modifications of the original
matrix that might prove useful.  Matrices could be compared as to long
term verses short term, regional verses localized, or the matrix could
evaluate the need for data.

      The presentation of the coal and nuclear energy production AIM's
(table 3.2.-1,2) is not for the purpose of data presentation. The matrices
are merely  a thought-organizing process to assure coverage of environmental
impact topics as comprehensively as possible.
      The evaluation of the four scenarios is admittedly qualitative and
subjective. The intent of this approach, necessitated by breadth of subject
matter and time constraints, is twofold. This analysis hopes to optimize
effort by directing attention to the most critical areas while also defining
needs for data or tools to quantify impacts.

3.2.1. CHANGES IN SURFACE WATER QUANTITY

3.2.1.1 DRAINAGE NETWORK

      Drainage networks must be modified during and after surface mining
to stabilize spoil and disturbed slopes.  Diversion ditches above high walls
and lateral  drains along benches are the most common forms of these localized
modifications.  (See figs. 3.2.1.1. -1.2,3)  Head of hollow fills are reclamation
features that use spoil to fill relatively narrow, steep-sided ravines.  The
ravines are usually terraced down the slopes and provided with underdrains
(see fig. 3.2.1.1. -4) .  Other surface water diversions may be necessary to
prevent site flooding. This is done by simple rerouting of water courses or
by pumping.  Overland flows should be diverted from spoil or mineral stock
piles when  there is potential for leaching of toxic materials. The adverse or
beneficial impacts of drainage network modification are generally measured
by impact on hydrologic factors and will be considered under 3.2.1.2.

      Regional drainage network modification may be the result of mining,
irrigation projects, or flood protection. Regional evaluation of such actions
could, at best, involve the summation of separate smaller actions or the
gross averaging of parameters over  large non-homogeneous areas.  Except
for mining impacts, the secondary impacts from population and industrial
growth would be equal for the BOM scenarios.  The BOM 80-20 sceanrio
would rate highest if this scenario requires the greatest production of coal.

-------
tf

#
 I
LM
X. ENERGY
\PRODUCTION
MAJOR\^T1VITY
IMPACT >^
CATEGORY N.
CHANGES IN
SURFACE H20 QUAN.
DRAINAGE NETWORK
HYDROLOGIC FACTORS
WATER USE FACTORS
CHANGES IN
GROUND H20 OUAN.
HYDROLOGIC FACTORS
WATER USE FACTORS
CHANGES IN
WATER QUALITY
THERMAL
AESTHETIC
SETTABLE SOLIDS
INORGANIC CHEM.
ORGANIC CHEM.
MICROBIOLOGICAL

&
Ul
o
L
L
L
H
M
L
H
M
H
L
H

MIN
UJ
1
V)
M
M
L
M
M
L
H
H
H
L
H

TRANSPORTATION g
L
L
L
.L
L
L
M
L
L
L
L

RECLAMATION
L
L
L
L
L
L
L
L
M
L
L

COAL ADVERSE IMPACT LEVEL
BENEFICIATION
L
L
M
L
M
L
M
M
H
L
M

PROCESSING
L
L
M
L
M
H
M
L •
H
M
H

END USE
L
L
M
L
M
M
M
L
M
M
H

SECONDARY
IMPACTS
L
L
M
L
L
H
H
M
H
M
H

INTERACTION
WITH OTHER
IMPACTS
.
















-







                                  TITLE- COAL ENERGY  PRODUCTION

                                        ADVERSE  IMPACT LEVEL



                                           Table II-3.2.0.-1
L-LIGHT


M - MODERATE


H -HEAVY

-------
00
-C
J)
N. ENERGY
\PRODUCTION
MAJOR\^TIVITY
IMPACT N.
CATEGORY >v
CHANGES IN
SURFACE H20 QUAN.
DRAINAGE NETWORK
HYDROLOGIC FACTORS
WATER USE FACTORS
CHANGES IN
GROUND H20 QUAN.
HYDROLOGIC FACTORS
WATER USE FACTORS
CHANGES IN
WATER QUALITY
THERMAL
AESTHETIC
SETTABLE
INORGANIC CHEM.
ORGANIC CHEM.
MICROBIOLOGICAL

S3
UJ
o
L
L
M
M
M
L
M
M
H
L
L

SURFACE S
• 5
M
M
M
M
M
L
H
H
H
L
L

TRANSPORTATION z
o
L
L
L
L
L
L
L
L
L
L
L

RECLAMATION
L
M
M
L
M
L.
L
L
L
L
L

NUCLEAR ADVERSE IMPACT LEVEL
BENEFICIATION
L
M
H
M
R
H
L
H
H
L
L

PROCESSING
M
M
M
M
H
H
M
L
L
L • .
M

END USE
L
L
M
L
M
M
M
L
M
M
H

SECONDARY
IMPACTS .
L
L
M
L
L
H
H
M.
H
M
H

INTERACTION
WITH OTHER
IMPACTS

























                                   TITLE- NUCLEAR ENERGY PRODUCTION
                                         ADVERSE IMPACT LEVEL

                                          Table II-3.2.0.-2
L-LIGHT
M - MODERATE
H - HEAVY

-------
        3d  Bench

       __ J^;- Spc        _^_
     Original  Ground  Surface
     -Diversion  Ditch
                                Spoil  Bank
BackfiUed  Bench /_
                      Ground Sur face
                                              Pipe  or  other
                                              Drainage  Structure
PIPE  BENO
                                                         Riprap   Ditch
                       WATER  DIVERSION
                            Figure II-3.2.J.1.-1
                                                              drawing
                                               in reference  No.  6

-------
                             Compacted Fill To Prevent
                             Ponding At Toe
Original Grogn.d.'Surface^
      mmttmzz^"
^\  Diversion
 \_nii
                                      Ditch
                  CROSS   SECTION OF
   DRAINAGE  DITCH ON UPHILL SIDE OF  A SPOIL  PILE
                                       Top Of Spoil
                                            Slope
            Terrace—v Slope      ,s£y£^^
 Toe Of
 Spoil
                   CROSS SECTION OF
            DIVERSION DITCH APPLICATIONS
                       Figure II-3.2.1.1.-2
                                         Reference No. 6

-------
-Diversion  Ditch




   •Hi^hwal1
                                  Excess Spoil from

                                   ^.  8 2Qd  Pits
       Mineral  Seam—             <.*„,.-
                  Original  Ground Surface
                                                         Toe  of

                                                         Fill
 -Diversion Ditch
    Highwall



r^,  \     Finished  Grade

^|    \    Surfaces     __
                                  /-Reverse  Terrace  Slope




                                            \
      fmeral  Seam
                 Original Ground Surface-


                         4TH   STEP
                                                        Toe  of

             BOX-CUT. MINING

                         Figure II-3.2.1.1.-3
                                               Reference  No. 6
                          JJT3-3-5Z,

-------
            Strip  Mine Bench
      Crowned
      Terraces
                           PLAN
                                           Original   —.
                                        Ground  Surface \
                                          Highwall

                                          Fill

                                  Lateral  Drain
Crowned.
Terraces,
                            Rock Filled
                            Natural  Drainway
                 CROSS   SECTION   OF
          TYPICAL  HEAD-OF- HOLLOW   FILL
                         Figure  11-3.2.1.1.-4
                                                      drawing
                                          in reference No. 6

-------
Uranium mining, although outside the ORBES region  , would require
diversion of surface flows for tailings ponds and reclamation efforts.
The competition for water in uranium mining areas is generally higher
than in the coal mining areas in the ORBES region.  The production of
western low-sulfur coal may also cause regional critical water situations
in the West.
      Regional impacts tend to be long-term whereas localized impacts
tend to be short-term. Flood control, navigation, irrigation,  and reservoir
projects are long-term commitments that are often renewed, rebuilt, or
replaced beyond their original predicted life span.  Mining operations
have their greatest impact locally, and modifications to drainage networks
are expected to stabilize and lessen with time.  The data on recovery of
surface stream recovery from mining effects is scarce but growing.  The
prediction of recovery time is not well-developed at present.

3.2.1.2 HYDROLOGIC FACTORS

      Surface mining has a major impact on local hydrographs, especially
during storm events.  Water diversion structures often intercept overland
flow and speed its deliverance  to surface water courses.  Detention ponds
for sediment control have an opposite effect.  Overland flows and small
streams are collected and retained while  sediment falls from suspension.
Retention ponds must be properly constructed and maintained to avoid
potential hazards.  Underdrains for groundwater interception do  not
modify the hydrologic regime as much as diversion structures and
sediment ponds.

      Underdrains may extend the "die-off" of storm hydrographs.  Where
toxic overburden must be protected from leaching, there will be local increase
in storm peaks and volume of surface flows. Deep mining and associated reclama-
tion efforts to prevent acid mine drainage may effect local surface hydrographs
to a varying extent, depending on the local ground-surface water relationships.
Dewatering operations during mining increase base flows of streams. Large
dewatering volumes discharged into small streams may change their character
and storm carrying capacity.

      The BOM 80-20 scenario will have  the greatest effect on local hydro-
graphs because of the demand for local coal resources.  The other scenarios
would be ranked according to area disturbed during mining.  The major
floods in Kentucky around Pikeville in 1977 are expected to propogate
studies defining strip mining effect on storm hydrographs.

      Hydrographs on the Ohio River are largely the result of dam and
reservoir management.  The lower the natural flow the more management
determines river flows.  The types of water resources projects expected
to change the main stem flows are up-stream reservoirs for supplementing
low-flows and modification of tributaries to support growing transportation
needs.  Urban and industrial growth increases the percentage of impervious
area in basins. This type of development often encourages channel widening
to handle increased storm peaks.  Unless non-structural flood control measures
are taken, increased storm peaks will be passed downstream necessitating
further modification until the impact reaches a body that can safely accept
the flow.

-------
      The high level of growth that would result from the BOM scenarios
might overload present flood control capacities in the Ohio River Basin.
These scenarios would be accompanied by pressures to modify tributaries
for industrial traffic.  The Ford Tech scenarios would require little more
than maintenance and replacement of present structures.

      Figure 3.2.1.3-1 uses 1970 National Power Survey figures for
consumptive cooling uses. For the years 1975 through 1985, the following
values for fossil fueled plants were utilized:  12CFS/100MWE for once through
cooling, 16 CFS/1000 MWE for cooling ponds,  and 28 CFS/1000 MWE for cooling
towers.  The scenarios were evaluated with lower consumptive values: 10 CFS/
1000 MWE for once through, 14  CFS/1000MWE for cooling ponds,  and 16 CFS/
1000MWE for cooling towers. Nuclear power plants with  a higher heat rejection
rate use 1.5 to 3 times more water.  The following values in CFS/100 MWE were
used for 1975 to 1985 nuclear plants: 17 for once through, 22 for cooling ponds,
and 40 for cooling towers. The scenario values used were;  15, 20, and  35
respectively. Name plate ratings and a linear relation between flow and MWE
were assumed.
      Although coal mining processes in the ORBES region consume
a relatively small amount of water, localized impacts on groundwater use
may be severe. Uranium mining uses a considerable amount of local water
resources in tailings ponds  at the mine site. The refining of uranium at
two plants within the basin and the power  production that support the plants
must be considered. For these reasons, the BOM 50-50 will be the most
consumptive scenario.  The ranking of scenarios would follow the order
in figure 3.2.1.3.    One possible estimate for total future consumption
could be calculated from setting cooling water needs as some percentage
and back-calculate the total. Other methods for total growth in consumption
may be tied to population growth, industrial output, or total earnings.
3.2.1.3  WATER USE FACTORS

      The United States Geological Survey estimated the mean annual
runoff of the Ohio River Basin as  125 b.g.d. (billion gallons per day)
and the annual flow exceeded in 90% of  years as 75 b.g.d.  Of the 35
b.g.d. that is withdrawn, 0.9 b.g.d. is consumptive use. According
to these 1970 estimated figures, the consumptive use is 0.72% of the mean
annual runoff or 1.2% of the annual flow exceeded in 90%  of years.  With-
drawals equal 28% of the mean and 47% of annual flow exceeded in 90% of
years. Although the consumptive figures  imply consumptive use is a
minimal fraction of the annual runoff, the percentages may vary consider-
ably within the year and within the basin.  The chemical and physical
impacts  on the 28 to 47% withdrawl will be  considered under sections
 32.3.1. - 3.2.3.6.
      The following quote from the  "Ohio River Cooling Study" gives a
basinwide estimate of withdrawl and consumption for steam-electric cooling
water needs.
                                rr 13-3-55

-------
  2800 h
  2400 h
  2000 h
1600 h
o
£
o
u_
0)
I 1200

o
O
    800 h
    400 h
                                                    Bom 50-50
                                                    Bom 80-20
                                                 Tech 100 N

                                                 Tech 50-50

                                                 Tech 100 C
      1975
           1980     1985     1990     1995
                         Year
                CONSUMPTIVE FLOW BY SCENARIO
                      Figure I1-3.2.1.3.-1
                                                 2000
                          JIB-3-56

-------
            In both the Ohio River basin and on the Ohio River
      the electric power industry asserts the largest demand on
      the available water resources, and nearly all of the power
      industry is for cooling and condensing the steam used to
      produce electric energy.  In 1965 an estimated 31 billion
      gallons per day were drawn from the Ohio River basin for
      man's use, and about one billion gallons of this total were
      used consumptively. The electric power industry withdrew
      19 billion gallons per day (61% of the total water withdrawn)
      for cooling purposes.  About 8.5 billion  gallons per day with-
      drawal were taken from the Ohio River and the lower reaches
      of its principal tributaries.

      For the same year, 1965,  the U.S.G.S. estimated a cooling water
42% share nationwide as compared to EPAfs 61% figure for the Ohio River
Basin.  Table 3.2.1.3-2 demonstrates the expected larger share for cooling
water expected by the year 2000.  In the ORBES region, the larger share
for steam electric power is probably compensated by a smaller share for
rural domestic and agriculture.


Table II-3.2.1.3.2 Estimated Water Withdrawl and Projected Requirements
                  by Purpose, United States in Billion Gallons Daily,  (%)

Year                    1950                 1970               2000
Rural Domestic
and Agriculture
114
( 57 )
114
( 36 )
156
( 19 )
Public supplies             14                 27                  51
(municipal and            (7)              .(7)               (6
industrial)
Steam-electric              40                 170                 470
    Power               ( 20 )               ( 46 )              ( 58 )
Other self-                 37                 47                 127
supplied                 ( 19 )               ( 13 )              ( 16 )
industry

                                                                  5
                                           Adapted from reference

-------
3.2.2. CHANGES IN GROUND WATER QUANTITY

3.2.2.1 HYDROLOGIC FACTORS

      Groundwater varies considerably between and within river basins.
Availability and quality of groundwater are dependent upon recharge
rates, types of substrata that convey and contain the water, and the
degree of development of the aquifer.  Availability and quality are
site specific, but generalizations can be made for large areas.

      In the Ohio River basin, moderate to plentiful groundwater
supplies are available throughout most of the glacial till areas and
the alluvial valleys of other portions of the basin.  The unconsolidated
deposits to the north of the Ohio River contain large groundwater storage
reservoirs in buried flow channels framed by preglacial systems.  For
the most part, groundwater reserves of the  glacial till and the small
area of the Gulf Coastal Plain in the lower portion of the Ohio River Basin
are plentiful and adequate, except for  large concentrated municipal and
industrial water supply needs.  However, the effect on streamflow of
groundwater withdrawal may prove a restraining factor in groundwater use.

      The mineral content of groundwater is generally higher than that of
surface water. Significant problems of excessive chloride exist in several
areas, and problems associated with excessive mineral concentrations and
hardness are also encountered. High iron content is also common through-
out the basin.
      Water quality from the consolidated aquifers of sandstone, limestone,
and dolomite and unconsolidated aquifers of sand and gravel are generally
hard, due to excessive calcium magnesium concentration in  the aquifer bearing
rock, although the degree of hardness varies significantly.  Other groundwater
quality problems include dissolved solids and iron.

      In some places the groundwater resource is inadequate for other than
domestic and rural use, although this problem is often due  to improper well
locations or outmoded supply and distribution systems. A few areas have
highly saline bedrock aquifers or p"oor'unconsolidated aquifers, prohibiting
major ground water use. Extensive'lowering of bedrock aquifer water levels
often occurs in metropolitan areas.  This  condition results in increased pump-
ing costs and depletion of future water availability.  Contamination of shallow
aquifers by waste disposal and of deep aquifers by leakage from water in multi-
aquifer wells has also occurred.7
      General groundwater resource studies in Ohio, Indiana and Illinois, and
Kentucky show that in the alluvial flood plains of rivers in these states expected
yields from individual wells can be expected to exceed 1.1 cfs. Expected yield
per well drops significantly away from the flood plain to 0.2-1.0 cfs in the
extreme northern areas of the river basins that lie in glacial drift to under
0.01 in the southern portions where glacial deposits diminish and Devonian
or Mississippi shale predominates.  No effort has been made to quantify in
detail the groundwater availability in the four-state area.

-------
      Adverse impacts resulting from the interruption of aquifers during
mining can not be avoided.  Lowering of water levels of wells, drying up
of springs and seeps, and reduction in streamflow would occur in an area
around the mines where aquifers were disrupted.  The location and extent
of depression around mined areas would vary, depending on various aquifer
properties.  Replacement of aquifers with unclassified overburden would alter
groundflow patterns and could reduce aquifer storage capacity.  If large quantities
of groundwater were withdrawn from thick sand and shale aquifers, some
subsidence may result.

      Underground mines may further interrupt aquifer flow patterns
because of fracturing resulting from subsidence. Strip mining in certain
areas necessitates the removal of aquifers. The present surface mining
bill before the 95th Congress, if passed, will require pre-mining evaluation
of aquifers. Under this bill, mining companies will be held responsible for
degradation of groundwater quality and quantity on the mined  area and
surrounding effected areas.  Where mining occurs below surrounding
land levels, direction of flow will also be changed.
      Since detailed inventories of groundwater exist for limited areas, only
qualitative judgments are possible at this  stage. The ranking  of scenarios
would be tied with the amount of coal to be mined, especially from surface
mines in the ORBES region.  The BOM  80-20 scenario will have the greatest
impact on groundwater hydrologic factors. Pre-mining inventories, if
required by law, will aid in both quantification and minimization of these
impacts.

3.2.2.2 WATER USE FACTORS
      In many parts of the Ohio River Basin, high quality groundwater
exists in various sedimentary deposits related to glacial deposits, along
present water courses, or in buried river valleys.  Most high  quality
aquifers are moderately to fully  developed.  Further development of these
resources could compete with present usage. Induced infiltration can
increase productivity of aquifers,  but water quality degradation usually
occurs.
      Increased population growth will also stress the quantity and quality
of groundwater systems. Water for domestic uses, especially in small
communities and rural areas, will be obtained from wells. In conjunction
with withdrawals for mine dewatering and agricultural purposes, these
increased withdrawals for domestic use may result in  aquifer depletion
in some areas.
3.2.2.3.1.1 REGULATIONS  & THERMAL REGIME

      The Ohio River Valley Water Sanitation Commission, or ORSANCO,
is an interstate compact composed of representatives from the federal
government and the following states:  Illinois, Indiana, Kentucky,
New York,  Ohio, Pennsylvania, Virginia and West Virginia.
      The ORSANCO Pollution Control Standards No.  1-70 and 2-70,  adopted
November, 1970 apply to all non-contract cooling water and other discharges
of heated water to the Ohio River.  These Standards are:

        All sewage, industrial wastes  and cooling water from municipalities
      or political subdivision, public or private institutions, or installations,
      or corporations discharged or permitted to flow into the  Ohio River from


                                IT B- 3-5*7

-------
      the point of confluences of the Allegheny and Monongahela rivers at
      Pittsburgh, Pennsylvania, designated as Ohio River mile point 0.0 to
      Cairo Point,  Illinois. Located at the confluences of the Ohio and Mississippi
      Rivers, and  being 981.0 miles downstream from Pittsburgh, Pennsylvania,
      shall be so regulated or controlled as to provide for reduction of heat content
      to such degree that the aggregate heat-discharge rate from the municipality
      subdivision, institution, installation or corporation, as calculated on the
      basis of discharge volume and temperature differential (temperature of
      discharge minus upstream river temperature) does not exceed the amount
      calculated by the following formula, provided, however, that in no case
      shall the aggregate heat-discharge rate be of such magnitude as will
      result in a calculated increase in river temperature of more than 5 deg.F:
Allowable heat-discharge rate (Btu/sec) =
      62.4 x river flow (cfs) x (TA - TR)  x 90%
      Where:
      T. = Allowable maximum temperature (deg.F.)
          in the river as specified in the following table:

                          TA                       TA
                           A                        A
            January       50          July          89

            February      50          August       89

            March         60          September    87
            April          70          October       78

            May           80          November     70

            June          87          December     57
TR  = River temperature (daily average in deg.F)
      upstream from the discharge

River flow   =  measured flow but not less than critical
               flow values specified in the following
               table:

                   River Reach                                  Critical
      From                           To                          fl°w
                                                                 in cfs*

Pittsburgh, Pa. (mi.0.0)              WiUow Is. Dam  (161.7)       6,500

Willow Is. Dam  (161.7)               Gallipolis  Dam (279.2)       7,400

Gallipolis Dam  (279.2)                Meldahl Dam (436.2)         9,700

Meldahl Dam (436.2)                  McAlpine Dam (605.8)        11,900

McAlpine Dam (605.8)                Uniontown Dam (846.0)       14,200

Uniontown Dam  (846.0)               Smithland Dam (918.5)       19,500

Smithland Dam (918.5)               Cairo Point (981.0)           48.100

* Minimum daily flow once in ten years
                                     2X13-3-^0

-------
      At its May 24, 1973 meeting the Commission adopted the following
guideline for administrative and enforcement purposes with regard to the
thermal discharge provisions of ORSANCO Pollution Control Standards
No. 1-70 and 2-70:

            "The foregoing requirement (thermal discharge provisions of
      Standards No. 1-70 and 2-70) shall be deemed to be met if the
      aggregate heat-discharge rate does not exceed the calculated
      allowable heat-discharge rate during more than two percent of
      the days in the 12-month period ending with any month and in no
      case shall in a calculated river temperature more than 3 deg .F
      above the corresponding allowable maximum temperature; the vari-
      ance is applicable only in those instances when the daily-average
      river temperature upstream from the discharge approaches or
      exceeds the specified allowable maximum temperature as the result
      of meteorological conditions .  In no case  shall the aggregate heat-
      discharge rate be of such magnitude as will result in a calculated
      increase in river temperature of more than 5 deg.F" .

      State standards and ORSANCO regulations generally agree on stream
criteria. Although there are varying definitions for mixing zones. The
critical flows specified in the ORSANCO Standards are based on U.S. Corps.
of Engineers estimates of probable 1-day in 10-year minimum flows to be
expected after 1975.  Recently these values, especially in the upper Ohio
River have been questioned because some of the authorized tributary
reservoirs may not be constructed.
      The greatest thermal impact on surface waters occurs at the energy
production stage when energy resources are consumed for steam electric
power.  The water withdrawn by a power company circulates through the
power plant condensors and absorbs most of the heat retained by the steam
after it leaves the turbine and before the condensate is returned to the feed
water heaters and boilers. The amount of heat  rejected depends on fuel
source, size of plant, cooling system type and receiving body.

      A typical nuclear plant of 31 percent thermal efficiency releases about
50 percent more heat to cooling water than a fossil-fueled plant of comparable
power output. Once-through cooling is the most efficient transfer of waste
heat but has the greatest adverse impact on receiving waters. Future installa-
tions in the ORBES area are expected to employ cooling towers,  either natural
or mechanical draft.

      An ORANSCO evaluation of thermal discharges to the Ohio River was
based on the effect on river temperature relative to accepted stream tempera-
ture standards and the degree to which individual discharges compiled with
applicable effuent limitations in the 10-year period, 1964-1974. Daily
average temperatures at ORSANCO's 11 electronic monitors on the Ohio  were
analyzed and the temperature limits  specified in state stream standards
were exceeded on only four days.  The study indicates that thermal dis-
charges to the Ohio River, during the period 1964-1974, had a relatively
localized influence on the river temperature. This conclusion is based  on

-------
data from electronic monitoring devices located on the basis of need for
multi-parameter stream quality evaluation.  It represents a macro evalua-
tion of overall thermal quality of the Ohio River.  Evaluation of the
impact of thermal discharges in the 3 dimensional or mixing zone case will
require more detailed data on a case by case basis.1

3.2.3.1.2 THERMAL DISCHARGE IMPACTS

      The greatest impact on phytoplankton and zooplankton has probably
been directly or indirectly from the damming of the  Ohio River. Fish popu-
lation studies should be performed to determine spawning habitats and give
these areas special protection from thermal or chemical discharges. Cooling
water intakes should also be located to minimize entrainment of biota or
disturbance of spawning habitats. Since future steam electric facilities
are expected to employ cooling towers,  thermal discharges should level off
or possibly decrease as once-through plants are retired. The largest impact
will be from secondary effects of the various scenarios.

3.2.3.1.3 SECONDARY EFFECTS

      The change in thermal regime for the two Bureau of Mines scenarios
should be similar.  The secondary growth accompanying the increased energy
production could cause stretches along  the Ohio where the temperature would
rise and fall with regulation limits.  These areas, already aggravated by
river damming would see decreases in species diversity and increases in
populations hardy enough to survive.  Occasions for algal blooms, especially
blue-greens,  and sewage fungus  (near  sewage treatment plants) will increase
causing problems with municipal water  intakes.  Increasing temperature
accelerates consumption of dissolved oxygen by high BOD wastes.  Near these
outfalls  critical D.O. levels may be reached before the waste can be assimi-
lated in a more  diluted stream. Higher  temperatures will also decrease the
solubility of oxygen and slow re-aeration rates.  Many undesirable compounds
increase their solubility with temperature.
      The two Ford Technical Fix scenarios should have nearly identical
impacts. With gradual elimination of oncethrough plants and limited industrial
growth, the chances for localized  lessening of thermal impacts are good.
Regional improvements will occur  only if BOD loads, chemical wastes and
other pollutants are lessened. In  any case the economics of cooling towers
and increased water consumption should be weighted against the benefits for
off river cooling.
      The U.S. EPA promulgated "Effluent Guidelines and Standards  for Steam
Electric Power Generating" effective November 7, 1974 (40 CFR 423; October
8, 1974 Federal Register).  The inventory classifies plants in two categories:
"A" which includes  EPA designations "small unit" and "old unit" and  "B"
corresponding to the guideline "generating unit". The guidelines do not
establish thermal limits for units in the  "A"  class. Best practicable—=-
treatment, defined as no discharge of heat, is established as the effluent
limitation for plants in the "B" category. Section 316(a)  of P.L. 92-500
provides a mechanism for requesting exemption from thermal limitations  in
NPDES permits  for both class "B" plants where thermal limitations may be
included in the permit on the basis of compliance with stream standards.

-------
       Discharges must also meet applicable NPDES effluent limitations and
 monitoring requirements.  There are region-to-region differences in monitor-
 ing requirements not only for once-through condenser cooling water but for
 other characteristics of power plant waste discharges.  For thermal dis-
 charges the monitoring requirements should include determination and report-
 ing of daily average heat rejection in Btu/hr.

 3.2.3.2 AESTHETIC FACTORS

       Aesthetic impacts are the most difficult to quantify and even attempts
 at qualitative descriptions are largely subjective. Aesthetics may be the
 measure of mood or atmosphere of an awe-inspiring waterfall or the feeling
 of isolation when one stands alongside a pristine river devoid of any sign
 of man's influence. Pristine areas fill some with a feeling of oneness with
 nature, and these unique areas provide an appreciation of the mystery of
 nature and natural forces.  Scenic vistas may invoke moods ranging from
 deeply religous to that of human insignificance.  The Bureau of Reclamation,
 under its present environmental evaluation system, relegates up to 15% of
 the impact from water resources projects to aesthetic impacts.

       Certain areas of energy production activities may meet resistance or
 increased costs due to aesthetic values.  These values must be  recognized
 and listed in environmental impact statements as required by the National
 Environmental Protection Act. The classic case of aesthetic degradation is
 orphaned strip mines. The increased turbidity and rust color impacted to
 surface waters is the most common complaint against the mining industry.
 Construction of reservoirs are highly;  controversial undertaking. These
 projects can provide recreation and accessibility to aesthetic beauty while
 inundating wooded and geologic shorelines.

       The construction of power plants and new transmission right of ways
 has a high potentials for disturbing recreational or sports uses of water-
 ways . Nuclear power plants especially have a negative impact on nearby
 recreation since some segments of the population choose to avoid such
 areas, even where the plant is not in view.

       Today with proper reclamation and time, some mined areas can be restored,
 and certain areas even improved, but some lands will always be judged
, irreplaceable.  Private landowners, or the taxpayer-voters through our

Apolitical system, may decide to forgo aesthetic beauty for economic
 productivity.
       The BOM scenarios with a large number of new fossil fueled and nuclear
 power plants  would  have a large negative impact on the aesthetic enjoyment
 of several stretches of the Ohio River but little impact on already heavily
 developed areas. Off-river siting might require construction of reservoirs
 for  sufficient water supply.  Since the BOM 80-20 scenario requires the
 greatest demand on regional coal resources, the combined impact from mining
 and new plant construction would cause more degradation than the BOM 5050
 scenario.     t.
 3.2.3.3 SETTI&BLE SOLIDS AND DEBRIS
       Suspended material in surface water and sediment loads  transported by
 streams in the ORBES region have a direct effect on the cost and feasibility
                                        -ZT3-3-63

-------
of water resources utilization. Sediment loads decrease life expectancies of
reservoir projects and increase costs due over-sizing for sediment storage.
Often communities immediately downstream from reservoirs complain of
sedimentation increasing after project construction.

      Water from streams or impoundments with high sediment loads requires
treatment before most uses.  Suspended sediments may directly effect fish
populations by interfering with oxygen absorption in the gills or by destruction
of spawning habitats. Turbid water also decreases light penetration, therefore
photosynthesis.  The loss of food production causes sometimes unpredictable
perturbations in food chains. High and sustained sediment loadings can cause
loss of recreational and sports fishery incomes.

      Sedimentation in navigational channels results in higher maintenance
costs through increased dredging.  Sediment deposits in stream channels can
reduce channel capacity, thus increasing flood potential.  Also during floods,
the deposition of mud on the flood plains and the cost of clean up often represent
the major damage by floods.

      Soil loss or sediment transport is commonly predicted by modeling based
on parameters such as land use type, soils, precipitation and slopes. The  two
land types most often associated with increased stream sediment loadings are
agriculture and surface mining while one of the .lowest sediment yields is from
forest cover. Proper soil conservation practice^and reclamation efforts can
significantly reduce solids loadings to surface waters. Sedimentation ponds
are a common integral part of strip mining reclamation. Suspended solids
may also be contributed  by point sources  associated with power plant ash or
flue gas scrubber sludges but these sources are considered negligible or
controllable.
      Within the Appalachian Region, average annual sediment loads range
from about 20 tons per square mile to as high as 3,000 tons per square mile.11
This wide range can be partly explained by differences in climatic factors,
such as quantity and intensity of precipitation and length of growing season,
and topography.

      One method for calculating sediment loads is assuming a regional average
Btu content of coal, average thickness of seam and a rate of 2.54 tons sediment
per square acre mined for each inch of runoff or 1600 tons per square mile.
Electrical production is converted from watts to Btu's and loadings of around
.2 tons per 1012Btu are the final result.10 The  shortcoming of the residual
calculation method is that temporal and spatial variables are ignored. Sediment
that is eroded may be trapped in a sediment detention basin or it may collect
in a stream bed for a year until a major storm removes a year's accumulation.
New mining and reclamation methods  accompanied by new federal legislation should
reduce sediment loading even with increased production. Efforts should be
directed towards collation of data from ongoing reclamation studies so that  more
accurate erosion rates can be applied to sediment loading.  Temporal distribu-
tion and intensity of rainfall events are also critical in consideration of sediment
movement.

      Where sediment control devices are not required construction of urban
or industrial areas is a considerable  sources of solids.  Interstate highways
as well have the potential for severe damage from erosion and sedimentation.

      Fine particles, especially clays and organic particles, are capable of
absorbing and releasing trace metals and pesticides.  Fines in urban runoff

-------
are particularly difficult to remove before discharging to surface waters.
The pollutants carried by these particles may effect benthic biota when
they settle from suspension.

      The BOM 80-20 scenario will produce the highest sediment loads mainly
as a result of strip mining and secondarily from growth in construction  of
urban and industrial areas. The BOM 50-50 scenario will produce only a
slightly lower loading while the Ford scenarios will likely allow consider-
able improvement of present conditions.

3.2.3.4.1 RESOURCE EXTRACTION

      Extensive literature exists on the  subject of acid coal mine waters,
including its causes, mechanism, treatment, biologic agents, and disposal.
When pyritic material is exposed to air and sulficient moisture, the end
product is sulfuric acid and iron salts, usually iron oxy-hydroxide.  Not
all coal mines produce acid drainage; some are even alkaline.  Many streams
contain alkalinity capable of neutralizating at least some acid mine waters.

      Acid mine waters, when discharged to streams, increase total dissolved
solids, sulfate, iron, manganese, and hardness.  The corrosive acid waters
shorten the life of ordinary concrete and metals used  in bridge piers, dams,
pumps, turbines, boats, and barges.  The pH of streams may fall below
tolerance levels for many species of fish. Precipitated iron salts on stream
bottoms smother benthic and planktonic biota, disrupting food chains for
higher animals like fish and shellfish.

      Brines associated with natural gas or petroleum development more
often affect groundwater but may produce erratic and extreme changes in
dissolved solids concentration and chemical type of receiving streams.  Large
amounts of water are used in tailing ponds associated with uranium or mining.
Discharges, from these ponds if allowed, can also raise dissolved solids levels
in receiving streams.

      Reclamation efforts aimed at sediment control may, in areas of acid
forming strata, accelerate formation of acid drainage. Thus, the chemical
effects of physical treatment must be taken into account.
3.2.3.4.2 STEAM GENERATION

      Steam electric power generation facilities must treat intake water before
use as steam feed or cooling water.  Steam feed water must not cause scale,
corrosion, or cracking. Additives for this purpose include di-or trisodium
phosphates, sodium nitrate, ammonia, or cyclohexylamine.  Cooling tower
waters must not cause scale, corrosion, or biologic growth.  Chemicals that
are commonly added to cooling water include chromates, zinc, phosphates
and silicates for corrosion control; chlorine,  hypochlorite,  chlorophenols,
quaternary amines, and organometallic compounds for bacterial growth control;
acids and alkalis for pH control necessary to prevent scale formation; lignin-
tannins, polyacrylamides, polyethylene amines,  and other polyelectrolytes
to reduce silt deposition.4

      EPA has promulated chemical effluent limitations as of July 1, 1977 for
total suspended solids, oil and grease, copper, and iron to  be applied to boiler
blowdown waters.  Cooling tower blowdown for existing plants will have to
meet best available technology economically available by July 1,  1983.  This
includes the following maximum limits:   zinc-hng/1, chromate-.2 mg/1,  and
                                      IFE-3-65

-------
phosphorous-5 mg/1.  Any plant constructed after March 4, 1974, must meet
new source standards for these elements, and that is non detectable. Free
chlorine residuals in all effluents should have no more than 0.5 mg/1 free
residual chlorine.

3.2.3.4.3 COAL RESIDUALS

      The trace elements in coal used in an electric generation station will
be primarily contained in the coal ash, which must be disposed of in an
environmentally sound manner. If scrubbers are required for flue gas
desulfurization, trace elements and high sulfate sludges must also be dealt
with properly. Although there are no standards directly applied to sludge
ponds, the following elements are listed in the EPA proposed regulations for
public water supply intake:

      arsenic           O.lmg/1            lead            O.Olmg/1

      boron             l.Omg/1            selenium        O.Olmg/1

      cadmium          0,01mg/l            zinc            5.0mg/l

      chromium          0.05mg/l            copper          l.Omg/1

                           mercury 0.002mg/l

      In a Radian Corporation  study for the Electric Power Research Institute
18 elements were measured in eight leachate samples.  Of the 11 elements that
are included in the aforementioned standards, selenium, boron, and chromium
exceeded the proposed standards in the majority of cases involved.  12 Leachates
from coal stockpiles should also be monitored for metals.

3.2.3.4.4. NUCLEAR RESIDUALS

      After mining, uranium ore must go through four steps before utilization
in a power plant:  milling-production  of yellow cake, UFg gas production,
enrichment and fuel fabrication. Large amounts of energy are used in the
enrichment stage.  The residuals from coal-fired steam generation to power
the Paducah, Kentucky, and Portsmouth,  Ohio,  enrichment facilities must be
charged to the nuclear residuals account.
      High level wastes from nuclear power plants are stored in sealed
containers and never contact effluent waters. Low level wastes are buried
in certified landfills such as Maxey Flats, Kentucky, where radioactive levels
are monitored in surface and goundwaters.

      Cooling waters are handled the same way as in coal-fired plants  since
they normally do not contain any radioactivity.  Steam  feed in boiling water
reactors does  come into contact with the reactor core.  This steam is then
condensed and revised.  In pressurized water reactors, a dual  coolant system
is used for transferring  energy. The water that comes into contact with the
core remains in a closed loop.  The heat from this system is transferred to
a second loop, which provides steam  for turbine generators.  This steam is
condensed and recycled. The  greatest part of radioactive emissions from
power plants is normally emitted as gases to the atmosphere.

      There will be no attempt to quantify effects of accidental spills or
leaking wastes.  These events  have occurred and will probably continue.

-------
3.2.3.4.5 SECONDARY EFFECTS

      The following discussion aims to delineate water quality problems
expected as a growth by-product of increased energy production and
consumption.

      Iron in surface waters at levels of more than 1 or 2 mg/1 usually
indicates acid wastes from mine drainage or other sources.  Levels as
high as 0.3 mg/1 at municipal intakes will probably require removal
to prevent staining, unpleasant taste and growth of iron bacteria.
Manganese is also a common constituent of acid mine waters.  Federal
drinking water standards provide that iron and manganese together
should not exceed 0.3 mg/1.

      Eutrophication or overenrichment of surface waters is largely a
result of conditions too favorable for aquatic growth and highly dependent
on nitrogen and phosphorus levels. Waste treatment plants are a common
source of nutrients since,  without advanced treatment, their removal is
only incidental. When nitrogen is released in the form of ammonia, there
is a consumption of dissolved oxygen as the ammonia is converted to
nitrates. Nitrogen consumption at levels of 10mg/l-N may cause methemo-
globanemia.  Nursing mothers should avoid high nitrogen content waters.
These waters are also unsuitable for infant feeding formulas. Both
phosphorus and nitrates are additives to cooling waters and steam feed
where copper must be added to control biological growth.  Agricultural
runoff may also contain fertilizers not absorbed by crops or soils.

      Hardness, measured by calcium and magnesium content,  is a
naturally occurring situation which can be aggravated by acid mine
drainage. Hardness causes scale in pipes and boilers if not removed
and hard water requires greater soap consumption for cleaning. This
may in turn raise phosphate levels in receiving waters.  Increased
calcium and magnesium in intake waters for most industrial processes,
especially textile and electroplating, will cause higher pretreatment
costs and sludge or salt wastes production.

      Sodium and potassium are discharged in sewage wastes,  agricultural
runoff,  salt brines from wells and industrial brines.  Large amounts of these
ions  in combination with chloride impart a salty taste and are hazardous for
consumption by people with high blood pressure.  Public health standards
for chloride is 250 mg/1.  Some sodium salts cause foaming in steam boilers,
and a high sodium ratio may limit the use of water for irrigation.

      Sulfate from acid mine drainage, sewage, or industrial discharges can
form scale in steam boilers. In large amounts, sulfate in combination  with
other ions gives a bitter taste to water and can have cathartic effects on
consumers.  Federal drinking standards recommend that sulfate content
not exceed 250 mg/1.
3.2.3.4.6 SCENARIO EVALUATION

      Both BOM scenarios will likely reduce the value of the Ohio River as a
high quality water source. More stringent regulations will be required or
more money will be spent on pretreatment of water.  Some municipalities or
industries may find capital expenditures necessary for development of alter-
native sources  deemed more dependable or even less expensive in the long
run.  The secondary effects from growth associated with the BOM scenarios

-------
include possibly a greater source of inorganic chemicals in waters than the
actual energy production phases. The BOM 80-20 scenario will release a
somewhat greater amount of inorganic chemicals than the BOM 50-50 scenario,
because of increased coal mining and utilization.

      The Ford coal scenario will have slightly more negative impact than the
Ford nuclear scenario. Under both scenarios, present and proposed regulations
may be sufficient to improve present water conditions.

3.2.3.5 ORGANIC CHEMICAL FACTORS

      Recent attention has been drawn to the occurrence of halogenated
organic compounds in drinking water and their relationship to disinfection
by chlorination. Chloroamines and organic compounds, such as chlorinated
hydrocarbons, are known by-products of waste  treatment plant chlorination.
Normally, levels of free residual chlorine amount to 0.5 mg/1 after disin-
fection dosages that range from 2 to 20 mg/1. The chlorine is initially
destroyed by reducing compounds such as iron, manganese, hydrogen
sulfi.de and organic matter. As more chlorine is added, chloro-organic
compounds and chloroamines are formed. Further addition will destroy
some of the compounds formed but others will remain in the effluent.

      Municipalities dependent on water supplies from multi-used streams
or impoundments are justifiably concerned about the contents of their drinking
water. Chlorination of waste waters can be replaced by ozonation or other
disinfection processes, but many possible carcinogens  may pass through
municipal or industrial treatment systems without notice.

      Spills of hazardous chemicals on the Ohio  River are recorded by the
United States Coast Guard, and ORSANCO warns downstream intakes of the
nature and timing of spills.  A cursory review of 400 Coast Guard spill records
for 1974-1975 indicates a variety of sources, causes, and materials.  Commonly,
spills involve barge loadings and oil refineries, but these are usually visible
events. Many accidents, no doubt, often go unnoticed  unless there is some
visual or odor indication.  If industry increases dramatically along the Ohio
main stem and tributaries, then spills and releases of potentially toxic and
carcinogenic materials will likely become more frequent. Many municipalities
may feel the need to develop standby water supplies  or convert altogether to
more expensive sources.

      Agricultural runoff is also a source of halogenated hydrocarbons from
pesticides.  Since the realization that many pesticides are concentrated as
they pass to food chains, many long-life or non-biodegradable products have
been limited or discontinued.  Development of biologic  controls,  disease
resistant crops, and biodegradable pesticides, herbicides, and fungicides by
the year 2000 will have lessened the contribution of organic chemicals to surface
waters by agriculture.
      Removal of organic pollutants from surface waters is producing an ever
increasing load on land disposal facilities .  Waste treatment facilities remove
and concentrate oils, greases, trace metals, and organics  in sludge solids.
Whether sludges are burned or not,  the final residue is land disposed. These
wastes along with a considerably larger  amount  of solid wastes from municipal
garbage collection systems are sources for organic contamination of ground-
water.

-------
      EPA is expected to control use and disposal of highly toxic organics
such as polychlorinated and polybromated biphenols.  Present EPA promul-
gated effluent standards call for no discharge of PCB's after July 1, 1977.

      The two BOM scenarios will result in a release of nearly equal amounts
of organic chemicals into surface and ground waters.  Although stricter effluent
limitations will decrease continuous emissions of dangerous organic spills from
on river point sources, transportation will continue to be hazardous.  The
rate of growth accompanying the Ford scenarios should allow municipalities
to bring overloaded facilities up to required treatment levels.  Critical D .O.
situations caused by BOD loading from treatment plants should continue to
decline.

3.2.3.6 MICROBIOLOGICAL FACTORS
      Channel modification and navigation dams have changed the Ohio River
from a free-flowing stream to a series  of lake settings. Understandably,  a.
concurrent shift from river flora and fauna to a lake type biota has occurred.
Thermal discharges may have an impact on biota in the mixing zone but the
effects diminish rapidly downstream.  Increases in suspended solids reduce
light penetration necessary for photosynthesis. The resultant loss in plant
production can disrupt established food chains.  Changes in stream or lake
pH's, such as  caused by acid mine drainage, cause shifts in microbial
populations.

      The United States Geological Survey has several biological sampling
stations on the Ohio main stem and reports monthly bacterial counts for fecal
coliforms and streptococci.  Also reported are periphyton biomass and
chlorophyll A and B.  Other biological water data can be obtained from
ORSANCO and EPA's STORET data  system.

      The Council on Environmental Quality noted significant overall improve-
ments in fecal  coliform violation rates in the Ohio main stem.  (1976 Annual
Summary - CEQ) .  These improvements can be expected to continue with
the financial aid provided to municipalities through federal 201 facilities
grants. The BOM high growth scenarios with its associated population
expansion may overload existing facilities before new plants can be built.
Some increase in bacterial pollution might be experienced from poorly installed
septic  tanks.  Further discussion can be found in section 4.1 under ecological
considerations.

-------
                        REFERENCES  SECTION  3.2

 1.   Ohio River Valley Water Sanitation  Commission(ORSANCO), Thermal
     Discharges to the Ohio  River,  1975.

 2.   U.S. Department of the  Interior,  U.S.  Geological Survey, Water for
     Energy Needs, Circular  703.

 3.   Federal  Power Commission,  National  Power Survey. 1970.

 4.   U.S. Environmental  Agency, Ohio River  Cooling Study, by Argonne
     National  Laboratory,  EPA-905/9-74-004, 1974.

 5.   U.S. Department of the  Interior,  U.S.  Geological Survey, Estimated
     Use of Uater in 1970, Circular 676,  1968.

 6.   Energy Research and Development Administration, Synthetic Fuels
     Commercialization Program, Draft  Environmental Statement, ERDA-1547,
     1975.

 7.   Energy Research and Development Administration, A Preliminary
     Assessment of the health and environmental effects of coal utili-
     zation in the midwest.  Draft by Argonne National Laboratory, 1977.

 9.   U.S. Department of the  Interior,  Bureau of Reclamation, Environ-
     mental Evaluation for Mater Resources  Planning, by Batelle Columbus
     Laboratory, 1972.

10.   Council  for Environmental  Quality,  Energy Research and Development
     Administration, et al., Energy Alternatives: A Comparative Analysis.
     by the Science and Public  Policy  Program, University of Oklahoma,
     Norman,  Oklahoma, 1975.

11.   U.S. Department of the  Interior,  U.S.  Geological Survey, Geology and
     Mineral  Resources of  the Applachian  Region, by G. Meyer, Hydro!ogic
     Atlas  HA-198.

12.   Electric Power Research Institute,  Environmental Effects of Trace
     Elements from Ponded  Ash and Scrubber  Sludge, prepared by Radian
     Corporation, EPRI 202,  1975.
                                      ZTB-3-70

-------
                           4.  ECOLOGICAL IMPACTS


           Ecological effects of plant locations under the four scenarios
being considered cannot be determined with the same degree of specificity
as can, for example, the quantities of potential pollutants released from
the stacks and effluent pipes of these hypothetical plants.  The specifi-
city of data provided is on a microscale with respect to a particular
plant's operating characteristics, but location of the plant is specific
only to level of the county in which it may be located.  One cannot scien-
tifically extrapolate from the information provided for the 167 plants
located only generally in county X to the macroscale data available for
the ORBES Region and then back to the microscale data required for the
assessment of a specific ecological impact.   As a result, evaluations of
effects on specific habitats are not possible except in a very general
sense.  Yet, to the ecologist, these habitats represent an important
resource which possibly should be protected.  Chapter 4 then will contain
first, a discussion of many of the possible hazards of power production
in the ORBES Region, and second, a general inventory of some of the habi-
tats of this region which could be affected as outlined in the first section,
but which should be protected.

4.1.  AIR POLLUTANTS FROM NUCLEAR AND FOSSIL FUELED POWER PLANTS AND THEIR
      EFFECTS ON THE BIOTA

           This section contains an outline of areas of environmental im-
pacts resulting "from air pollutants. 'The effects given are primarily from
data reported in studies generally applicable to the ORBES Region.   Similar
dat.'1. for this geographic area are sparse or lacking entirely.   Thus the
relative degree of impact for any given area cannot be specified.  There-
fore this section addresses only the following broad general  questions:
(1) What will be the effects on the ecosystem if a nuclear plant is con-
structed?  (2) What will be the anticipated effects on the ecosystem if the
plant is coal fueled?

4.1.1.  AIR EMISSIONS FROM NUCLEAR PLANTS   .

           If a decision is made to locate a power plant in county X of
Kentucky, Illinois, Indiana, or Ohio, the plant will be coal-fired or
nuclear, and its specifications will be those cited earlier in section 1.
If the option is a nuclear plant, the main biological-ecological concerns
will be the fate of water, heat, and radionuclides entering the water and
terrestrial  environment.  Water and heat can be expected to enter princi-
pally from cooling ponds or cooling towers.

           The emissions from cooling towers have several potential
ecological effects.  For example, shade over some of the immediate ( 1 ).
In addition, moisture will be added to the atmosphere.   The amoutn of
water vapor released from a 1,000 MW nuclear facility operating at full
capacity is projected to be 14,000 gpm ( 2).  The evaporative losses
reported by Parker ( 1 ) for a boiling water reactor (BWR) are 50 cfs, and
                             II-B-4-1

-------
40 cfs for a pressurized water reactor (PWR) for a 25°F range from inlet to
outlet of tower.   The drift losses for the BWR and PWR equal 0.8 cfs and
0.65 cfs, respectively.  This additional  atmospheric moisture conceivably
could result in changes in the rainfall patterns in the vicinity of the
cooling tower.   Since optimum moisture ranges exist for vegetation, the
added water may affect the plant life in  the area.

           Chromates, phosphates and other ions used to inhibit corrosion
and scaling, and chlorine to lessen fouling organism growth,  will be
carried with the drift, and some chemical? will fall  onto soil  and vegeta-
tion and possibly reduce productivity (1 ).  Generally, in small concen-
trations these ions may have no effect or they may even act as  nutrients.
In greater quantities, however, they can  be toxic.  Some plant  species will
be affected while others will remain unaffected.  The reason  for this is
that living plants differ in their requirements for various nutrients.  They
also will vary in their toxic thresholds  for various chemicals.   Some
plants are more sensitive than others.  Probably not all vegetation will
be affected, but if the critical threshold for one species is reached, then
ri, could be replaced by another species or eliminated altogether.  The
reduction of species diversity within these areas could lessen  stability
of the system by altering natural  balanaces.  Also, ions accumulating on
the soil as well as on leaf surfaces can  alter the soil chemistry to the
extent that a particular plant species may no longer be in an optimum
soils location.  With increasing amounts  of ions deposited, the soil may
become too brackish or chemically altered to support vegetative species
common ,to the.area.

           Also included in the gaseous effluents of a nuclear  power plant
is the potential of various radionuclides.   Eisenbud (3 ) prepared a table
reproduced below which lists the radionuclides released fron  a  BWR.


                               TABLE II-B-4-1
                    PRINCIPAL RADIOF.UCLIDES RELEASED IN
                        GASEOUS EMISSIONS FROM A BWR

     Radionuclides                Half-Life           Release rate    /sec

          Kr-85                    4.4 hr                  3 x 102
          Kr-85                   10.7 yr                  1  x 10'1
          Kr-87                   76   min                 7 x 102
          Kr-88                    2.8 hr                  5 x 102
          Xe-133                   2.3 da                  1  x loj
          Xe-133                   5.3 da                  3 x 102
          Xe-135                   9.1 hr                  8 x 102
          Xe-138                  17   min                 2 x 1030
          H- 3                    12   yr                  5 x 10"2
                             II-B-4-2

-------
A PWR poses less danger to the environment than a BWR, under normal operating
conditions, since wastes can be stored in the former for approximately 60
days as opposed to 30 dross for the BWR (4 ).

           The ecological effects of radionuclides released to the environ-
ment can be detected at both the level of the .individual species and at the
ecosystem level by alteration of the food chain.  A discussion of specific
effects of radionuclides on individual species is too extensive an under-
taking for this discussion, but Reichle, e£ al_. (5 ) described the fate of
radionuclides in the biosphere and the discussion below is based on their
study.

           The effect of radionuclides depends upon the type of elements and
the radiation emitted.  Alpha particles are stopped by skin tissue, beta
particles penetrate skin, and gamma raps penetrate to internal organs.
The primary health hazard of alpha and beta emitters to the biota occurs
after ingestion.

           At the ecosystem level the effects of exposure to the ionizing
radiation may involve changes in the frequency of mutations and changes in  .
the vigor of organisms irradiated (6 ).  Factors of particular concern are
length of exposure, and magnification or increase in concentration in certain
organisms.  Those radionuclides deposited in organs or tissues of low turn-
over (e^£., bone or exoskeleton) will be present for a longer period of time,
and thus a greater possibility of deleterious effects exists.  Reichle, e_t
_aj_.  ,(*5 ^recorded 'the following --observations concerning concentrations of
radionuclides in various terrestrial organisms.   It is noted that results
should be anticipated to be variable because of differences in feeding
habits, seasonal changes in diet, selective feeding, and in herbivores by
the type or form of vegetation available.

           1.   For some elements for which information is available,
concentrations decreased in the food chains from plants to mammalian
herbivores; for other elements, however, concentrations increased.

           2.   Calcium and strontium decreased in whole-body concentrations
of invertebrate without calcified exoskeletons;  but,

           3.   In calcium-sink invertebrates (millipedes, isopods and snails),
calcium and strontium were concentrated by factors greater than 150.  In
vertebrates similar results could be anticipated.

           4.   Cobalt, potassium and cesium decrease through invertebrate
food chains by a factor of only 0.5 after two trophic levels (plant--
herbivore—carnivore).

           5.   Phosphorus levels may be greater from plants to insect by
a factor of 10 with little additional change at higher levels.

           6.   Sodium is concentrated continually at each level.
                             II-B-4-3

-------
                                 TABLE  II-B-4-2

                     Biological Half-Lives of Elements in Major Groups of Organisms
                                    Elements with indicated range of half-lives
 Animal group
0.1 to 1.0 day
1.0 to 10 days
              10 to 100 days
               100 to 1000 days
Terrestrial    3
 invertebrates
Rb. Sr. Y, Cu
Ns, P, K,
 Ca. Ci, Co,
 As, Sr. Y,
 Ru.Cs, 1,
 W.Ir
Na, P. Fe, Co,
 Zn, (Ca and
 Sr),2l
> 1000 days
Aquatic ,
Invertebrates Ca, Sr

Fish


P

Na,I


Co, Zn.Cs ?
(SrandMn)
Na, K, Cs, 1

o
Zn2

Ca, Mn, Co,
Zn, Sr, Cs

Mn

S.Co ,Sr

Vertebrates B.Ge.Tc.Os H, Li. C,
.. O.Na.P. K,
Sc, Mn.Co,
.. Ga^Se.'Br,
Rb, Mo, Tc,
Ru.Rh, Ag,
Te, 1, Cs,
Re.Os. Hg,
TlBi


V

H,B,C, N.O,
Na, Mg, Si.
P,S, K,Sc,
V,Mn,.Fe,
Co, Cu, Zn,
Se, Rb, Ru,
•;Rh,Pd, Ag,
In.Sn, Sb,
Te.I.Cs,
Ba. Eu, W,
Ir, Pt, At,
Fr.U

F, Mg. Al,
P.Ca.Ti,
Cr, Mn, Fe,
.Ni, Z-n.-As,
Sr, Zr, Nb,
Cd.I.Cs,
La, Cc, Pr,
Nd, Pm.Sm,
Eu, Gd, Tb,
Dy, Ho, Er,
Tm, Yb, Lu,
Hf, Ta, Au,
Pb.U
Ca, Sr. Y,
Pb. Ra, Ac,
Th, Pa, Np,
.Pu, Am. Cm,
Bk.Cf








    - Primarily annelids  and mollusks.
    - Invertebrates with  calcified  exoskeletons.
    - Primarily adult  insects.
    - Primarily homoiothermic vertebrates.
  Reproduced  from Reichle,  D.E., P.B.  Dunaway and  D.J.  Nelson,  1970.  Turnover
  and Concentration of Radionuclides  in  Food Chains.   Nuclear  Safety 11(1):
  43-55.  As  it  appeared  in:  Schultz  V.  and Whicker,  F.W.,  1972.   Ecological
  Aspects of  the Nuclear  Age:   Selected  Readings in  Radiation  Ecology.
  U.S.  AEC Technical Information Center.  TID-25978.
                                II-B-4-4

-------
           7.  In contrast to its activity in invertebrates, cesium
concentrations increased in higher trophic levels.
systems:
snails.
           Reichle ( 5) also reported concentration increases in aquatic
           1.  Strontium and manganese accumulate in shells of clams and
           2.  Phosphorus can be concentrated to higher factors in both
invertebrates and vertebrates.

           3.  Cesium and potassium concentrations are higher in aquatic
animals than in the surrounding waters but generally no increase is noted
in higher trophic levels.

           4.  Cobalt, ruthenium and zinc are less concentrated in inverte-
brates than algae or plants, but more concentrated than in water.

           A second factor to be considered is the biological  turnover of
the ingested radionuclides.  The turnover is a measure of the rate radic-
nuclides are excreted.  A measure normally used to quantify turnover rates
is the biological half-life, which, by definition, is "that time required
for an organism to lose 50% of its body burden of a radionuclide when
removed to a noncontaminated food source" ( 5).  Table II-B-4-2 lists
biological .half-lives of .elements for major groups of organisms.  The
table should be treated with some caution, however, since excretion of
radionuclides by organisms is dependent on many biological  factors such
as body size, metabolic rate, age, sex, physical  activity,  physical condi-
tions, and deposition in biological sinks and such environmental factors as
media composition and temperature.  For example,  strontium and calcium have
longer biological half-lives in invertebrates with highly calcified exo-
skeletons but in species without exoskeletons, phosphorus,  calcium and
strontium have rapid turnover times.   Cobalt and  zinc act as cofactors for
enzyme systems and thus have slower turnover times, while sodium and
potassium show short biological  half-lives.   Rapid turnover was observed
for iodine except in highly pigmented forms.

           Assimilation of radionuclides varies with differences in food
base.   In invertebrates, assimilation is low is detritus  is the food base.
Cesium-134 and cesium-137 are 53-65% assimilated; strontium-90, 77%;
and calcium-47, 69% assimilated.   Radionuclides from green  plant material
are assimilated in somewhat greater quantities by invertebrates since more
cellular constituency is available.  The following rates  have  been recorded:
Cesium-134 and cesium-137, 73-94%; phosphorus--32, 54-66%; rubidium-86, 100%;
and wolferium-187, 100% ( 5). Flesh-eating predators show the  next higher
assimilative efficiency.  Typical  recorded values are:  cesium-134 and
cesium-137, 79-94%; and calcium-47, 98%.
                             II-B-4-5

-------
           Mammals assimilate radionuclides through a) the gastrointestinal
tract, which normally is the most significant pathway; b) respiratory
system where retention times of compounds vary up to as much as 1,500 days;
and c) the skin which is of minimal significance except for deuterium and
tritium  ( 5 ).

           Aquatic organisms obtain radionuclides either from food sources
or directly from the water.  Food sources differ according to species that
range in size from microscopic plant life to fishes.  These food chain path-
ways determine the distribution of the following radionuclides:  phosphorus-
32, cesium-137,     scandium-46, arsenic-76, copper-64, chronrium-51, iodine-
131, zinc-65, and cesium-134.  Again variations in assimilation will occur
with respect to organism, food chain, and other intrinsic and extrinsic
factors.

           Prediction of the effects of radioactive emissions into the
atmosphere and hydrosphere for the Ohio River Basin can be stated only in
the most general terms until more information on the parameters listed
above is available for this specific area.  Since hundreds of different
species of fish, mammals, and insects occur in the habitats discussed in
this report, and since it has been indicated earlier that these groups do
assimilate radionuclides though at different rates, in all probability
then radionuclides released from the proposed power producing facilities
will find their way into the food chains of this region.  What the short
or long term effects these assimilations will have will depend upon the
organism .affected, j_ve.» .whether its life .span is long enough to record
genetic or somatic changes.  The effect on the ecosystem as a whole will
be dependent upon the individuals affected, and severity of impact.  In
order to answer these questions more field research on ecological  problems
in the Ohio River Basin is warranted.

4.1.2.  AIR EMISSIONS FROM FOSSIL FUELD POWER PLANTS

           If a coal-fired plant is chosen for county X, the air emissions
of concern are again the radionuclides and water vapor, plus those combustion
products typical of fossil  fuel systems.   Parker (l ) quotes the following
concentrations in tons/year released from a 1,100 MWe plant, a cyclone
burner utilizing electrostatic precipitators with an efficient of 97%
and utilizing a fuel  with 10% ash content:  particulates, 5,200; sulfur
oxides, 274,000; carbon monoxide, 3,600;  hydrocarbons, 110; nitrous oxides,
1,988,000; aldehydes, 18; and carbon dioxide, 19,900,000.  In addition the
following radionuclide concentrations in  millicuries/year would be released
to the atmosphere:  radium-226, 5.85; radium-228, 4.8; thorium-228, 3.3;
thorium-232,  6.83.  The loss of water from cooling towers by evaporation
would be 33 cfs and 0.5 cfs from drift.

           The effects of water vapor from a fossil-fueld facility on the
ecosystem is again a multi-faceted problem.  Specific effects of extra
water vapor are, in general, similar to those listed for nuclear facilities.
However, the following additional effects are noted.  First, absorption of
heat which coupled with carbon dioxide can add to the "greenhouse" effect;
                             II-B-4-6

-------
second, the water vapor provides a reactant and/or a reaction medium for
production of additional pollutants which may or may not be present in the
plume as it exits the stacks, e_.£, the transformation of sulfur oxides to
sulfuric mists.

           Other stack gases can also have effects.  For example, inhaled
carbon monoxide alters the oxygen carrying capacity of hemoglobin.  Carbon
dioxide concentrations in the atmosphere, which are rising on an annual
basis, may have the impact of retaining released heat.  Although carbon
dioxide is removed from the atmosphere by green plants, increasing human
activity is reducing the amount of green vegetative cover on a global  scale.
Aldehydes and other hydrocarbons released during the combustion of fossil
fuels contribute to photochemical smog.  The occurrence of industrial-
urban smog in major cities of the region is well documented.  Additional
power plants may enhance this problem.

           Since coal is the resultant product of the partial anaerobic
degradation of plants existing 300 million years ago, it is not surprising
that many of the elements assimilated by these plants can still be found
in the coals.  Thus radionuclides which were assimilated then are released
from the coal during combustion.  These radionuclides are either released
out the stacks or are retained in the ash.  A discussion of some of the
implications of release of radionuclides into the atmosphere was presented
in section 4.1.1.

           The major effects from a coal-fired plant will be from sulfur
oxides and nitrous oxides, as well as other "stack "gases.   These ions and
compounds can affect the vegetation and soil  microbiota.   Animal life  may
be affected by ingestion'of contaminated vegetation or by inhalation of
gases and particulates.

           Particulate matter acts as an adsorption medium and therefore
can concentrate other pollutant gases and vapors already present in the air.
In addition, the particulates may function as condensation nuclei thus
possibly increasing rainfall.  Finally, in the upper atmosphere, particu-
lates in large quantities may scatter incoming solar radiation, a-;d have
a cooling effect.

           Sulfur and nitrous oxides are converted to acid mists in the
atmosphere when sufficient water vapor is present.  These compounds can be
beneficial  to plants in a few instances, e..c[., when they are at low con-
centrations in the soil.   The acute deleterious effect on vegetation,  how-
ever, is significant (tissue damage present)  on exposure to concentrations
of sulfur oxides of 1 ppm for 1  hour or 0.3 ppm for 8 hours (6 ).  When
adsorbed onto the surface of particulates, sulfur oxides  and/or sulfuric
acid may be incorporated into deep lung tissue.   In the gaseous form,  sul-
fur oxides and sulfuric acid mists act as irritants to the eys and upper
nasopharynx.   With an average particle size of 1 micron,  a sulfuric acid
mist of 1  mg/m3 will produce a respiratory response in vertebrates ( 6).
                             II-B-4-7

-------
Acid rain produced by sulfuric acid or other acids in precipitation can
leach nutrients from soils, alter the microbiota and thus exert possible
detrimental effects on ecosystem stability.

4.2.  EFFECTS OF POWER PLANT OPERATION ON AQUATIC ECOLOGICAL RESOURCES

           Aquatic biological effects of cooling water systems of nuclear
and fossil fueled power plant systems can result from thermal effluents,
passage of organisms through the cooling system, entrainment of organisms
on the screens at the system intake, and chemical dissolved in the effluents.

4.2.1.  THERMAL STRESS ON AQUATIC SYSTEMS

           The changes in stream temperature will affect each species in
a different manner.  Some general effects on major groups of aquatic organ-
isms such as fish, waterfowl, reptiles, benthos, and plants are discussed
in this section.

           Thermal tolerance of the various freshwater organisms ranges
from 0-42°C with absolute tolerance varying with individual species.  The
response of a species to a new temperature regimen will  vary for a single
individual and is dependent on the water temperature to which it had been
acclimated.  If insufficient time if available for acclimation and tempera-
ture rises sharply, organisms can experience lethal  thermal shock.
Permanent or temporary changes in water temperature will  have different
effects on organisms depending on their life stage.

           Specific for various species, however, is the critical  thermal
maximum (CTM).  This is the temperature at which an organism is unable to
control locomotor activity and thus looses the ability to escape lethal
thermal conditions.  The CTM for a particular organism depends on  the
temperature to which the organism is acclimated, i_.e_., a fish living at
20°C will have a higher CTM than those acclimated- at 10°C.

           Temperature also will affect metabolic rates,  thereby causing
intern?! chemical reactions to increase.  Increased reactions expressed as
metabolic rate result in greater food demand by the organism.

           Thermal effects on individual fishes range from those causing
death - coagulatin of proteins, liquification of lipids,  cellular  permeability
changes and internal changes which produce toxins - or those decreasing vigor
or affecting reproduction.  Effects on fish populations are dependent on
the season of the year, life stage, preferred temperature,  acclimation,  and
availability of food.  Movement and migratory patterns may also be
influenced in the heated area.

           Beyers ( 7 ) cites several references of the effects of  thermal
discharges on turtles.  Female turtles attain larger body sizes, lay more
clutches of eggs per season and mature sexually at an earlier age.
Apparently increased temperature provided more available food protein which
in turn increased vigor.
                             II-b-4-8

-------
           Brisbin (8 )> studying the response of waterfowl  to heated
effluents, found that waterfowl, free to choose between heated or un-
heated water chose the unheated waters in greater numti-ers and species
diversity. Reduced aquatic and riparian vegetation in the heated water
area may have been the cause of this preference.  Coots, however, were
observed feeding on algal mats in the heated water portion.   An important
finding was that the ducks more sensitive to heated waters were those
valued for sport and as a food source.

           In a study of thermal effluents entering into the Delaware
River, Strangenberg and Pawlaczyk (9 ) found that the number of benthic
organisms began to decrease when water temperatures reached 30°C and
that a tolerance limit of approximately 32°C exists for certain benthic
organisms.  While many organisms are affected at temperatures of about
30 C, others such as dragonflies grow to larger sizes undar these condi-
tions than do other individuals in unheated water.

           Heated water effluents also affect particular green plants.
Since not all plants are equally tolerant of heated water, o. change in
diversity of plant life near heated water effluents could be anticipated.

           Microscopic algae vary in their response to heated effluents
(7 ).  Bluegreen algae, a less desirable form, are more heat tolerant
than desirable species.  However, the ambient water temperature before
heat stress is a factor in determining which species may be present.
Heated .water -in Lake Michigan would possibly -harbor more dia.toms
(desirable species) than bluegreens while in southern locales, where
ambient temperatures are higher, the bluegreens would be favored 0-0).

           Decomposing organic-ms also respond to increased temperatures.
In general, the decomposition of organic matter is enhanced at warmer
temperatures as long as dissolved oxygen concentrations do not become
limiting (7);   If dissolved oxygen decreases significantly, the faculta-
tive anaerobic bacteria dominate but decompose material at a reduced rate
to the normal aerobic process.

4.2.2.  EFFECTS OF IMPINGEMENT BY CONDENSER SCREENS ON AQUATIC ORGANISMS.

           Organisms too large to pass through 3/8 inch screens can be
impinged upon the bar rack and grill over the intakes and may be killed
by asphyxiation, starvation or exhaustion.   The severity of the impact
depends upon the velocity of intake, the swimming speed of the affected
species, its physiologic condition and its  role (niche) in the ecosystem (10).

4.2.3.  EFFECTS OF CONDENSER PASSAGE ON AQUATIC ORGANISMS

           Bacteria, aquatic fungi, phytoplankton, zooplankton, ichthyo-
plankton, benthic invertebrates and fish fry small enough to pass through
a 3/8 inch screen can be drawn into condenser piping.   During transit,
                            II-B-4-9

-------
organisms experience thermal shock, mechanical shock, pressure changes and
exposure to chemicals.  At river water temperatures below 20°C, phytoplank-
ton photosynthesis was stimulated while at temperatures above 20 C it is in-
hibited (10).  Reduced productivity by reduction of chlorophyll (a) content
was recorded as a result of chlorination, but mechanical stress presumably
was less of a problem.

           Mechanical stress, however, results in zooplankton death during
condenser passage, and apparently, stenthermal plankton species and larger
zooplankton experience negative selection pressure during condenser passage
(10).  Fish eggs and larvae sometimes have as much as 70 to 100% mortality
due to thermal and mechanical stress during condenser passage (H).

4,2.4.  EFFECTS OF CHEMICAL DISCHARGES ON AQUATIC ORGANISMS

           Organisms present at the point of effluence into the waterway
are subject to many stresses during blowdown due to the various algicides,
scale and corrosion inhibiting chemicals, bacteriocides, and silt deposition
preventing chemicals.  In addition, the chemicals together may have addi-
tive properties resulting in possible synergistic effects.   The possibility
of synergistic effects of additional chemicals present in the water from
other sources also exists.   The specific long and short term effects from
these additions again will  depend upon the particular species, its life
stage, physiologic condition, etc.

4.2.5.  SEDIMENT QUALITY AND BENTHIC BIOTA

           The information for this section can be found in the report of
Special Study III-I: Radionuclide and metal ion content of late summer
Ohio River Sediments, McAlpine Pool, 1977.  Because many maoroinverte-
brates are filter feeders,  the adsorption of metal ions and radionuclides
onto sediment particles poses another potential route by which these
elements may enter the food chain.

4.2.6.  EFFECTS OF COAL MINING ON AQUATIC RESOURCES

           Because of environmental conditions which existed at the time
of coal formation, certain  seams and the rock material in'association with
the seam contain chemical  precursors of acid mine drainage (AMD).   The
effects of AMD on the aquatic ecosystems are severe and often totally dis-
rupt an aquatic system.   These effects are synergistic because of lowering
of pH which favors increased concentration of salts and metal  ions.  Few
species can tolerate these conditions.   According to most reclamation laws,
the toxic elements must be covered and not allowed near the surface of
graded areas where AMD would be continually formed for many year following
mining.  Thus if the laws are stringently enforced, AMD should occur only
during the mining process and not afterward.   AMD with undesirable quali-
ties should be collected and treated before it is allowed to enter water-
ways.   Accidents occur,  however, and with anticipated increased intensity
of mining the probability of accidental  release of AMD into nearby waterways
                             II-B-4-10

-------
is increased.  If the headwaters are affected, reinvasion is more question-
able, and if rare or unique species existed, reestablishment of fauna is not
probable.  An additional factor is siltation of streams resulting from
erosion, and chemical precipitation of soil particles by.AMD.   This
occurrence will change drainage conditions of streams, cover eggs deposited
by aquatic species and cause respiratory difficulties when silt accumulates
on gill surfaces.

4.2.7.  COOLING PONDS

           The development of cooling ponds as a mechanism for the release
of waste heat from coal fired and nuclear power plants is an alternative
to the direct intake of water from rivers and lakes, and the return of
heated effluent either directly or via cooling towers.  This discussion
addresses some of the positive and negative environmental effects of
these cooling facilities.

           Although design parameters will vary according to the individual
specifications for each plant, approximately 1-2 acres of pond surface
area will be required for eaclv megawatt of generating capacity.   Evapora-
tion from a nuclear facility would be about 28.6 acre-feet per megawatt per
anhum while evaporation from a coal-fired facility would be approximately
17 acre-feet per year for each megawatt (12).  If these parameters are
adhered to, a 1,000 megawatt facility would have a one-two thousand acre
"pond" associated with it.   Obviously, as generating capacity increases,
the "ponds" are in reality reservoirs.  It should be anticipated therefore
that, "if this design alternative is followed, reservoirs of significant
size would be constructed in the ORBES Region.

           The beneficial aspect of reservoirs is that the colling system
is essentially a closed circuit.  This would circumvent the problem of
releasing heated effluents into the streams of the Ohio River Basin.
Associated with this beneficial  effect is the potential for recreation
and economic development of the reservoir area.   As a result, the reser-
voirs could provide a substantial  facility for boating, fishing, camping,
aquaculture as well  as many other uses.   The possibility of multipurpose
use offers an attractive alternative to the use of cooling towers which
also require substantial acreage for development.

           The development of reservoirs, however, is not without environmental
costs.  First, these facilities could remove large blocks of forested area
if site selection criteria are not developed to minimize this possibility.
Secondly, many of the potential  construction sites could be within stream
valleys where the watershed would supply enough water to maintain a normal
operating power pool for the reservoir.   Pump storage facilities are the
other feasible alternative.  With both these facilities, the possibility
exists for the destruction of free- flowing streams which are being
degraded at an accelerating rate.   If the cooling pond alternative is selected,
site selection criteria should be stringent so that the above effects would
be minimal.
                             II-B-4-11

-------
           The development of reservoirs also again raises the questions
that have arisen so many times in the construction of reservoirs.   These
questions have both positive and negative components.  For example, if a
stream is eliminated, some species of fishes common to streams will dis-
appear; however, other species of fishes will flourish in the reservoir
system.  Wildlife would be disrupted or eliminated from the reservoir basin,
but if suitable habitat is provided through wildlife management, wildlife
can be expected to re-establish in the reservoir area.  In many cases,
real estate aquisition will be necessary, and many personal lives and
livelihoods will be disrupted.  The construction of power plants in
rather isolated areas would minimize this .possibility.  These impacts,
of course, cannot be fully avoided, but they can be minized by following
reasonable selection criteria.  To balance these losses, the reservoirs
would provide flood control, recreation, water supply storage, and most
importantly, a method of cooling which potentially is the least environ-
mentally degrading.

           It should be recognized that, even though cooling ponds may
be an alternative to other types of cooling, the quality of t!:e impounded
waters may be less than desirable.  Because of the evaporation rate,
the concentration of dissolved solids, metals, and nutrients would be
increased.  As a result of these excesses, these reservoirs will be
highly productive.  If concentrations reach critical  levels in summer
months, noxious blooms of algae could result.  These blooms could be
minimized through water quality management procedures, and the provision
for adequate flowthrough or low water releases to remove the higher
concentration from'the hypoliinnion or bottom waters during the summer
months.  Even with removal, excesses of metallic ions and other elements
over natural concentrations can br anticipated during fall overturn and
during any hypolimnetic releases.   These excesses will be seasonal, how-
ever, and should not preclude the realization of this laternative.

4.3.  EFFECTS OF POWER DEVELOPMENT ON TERRESTRIAL HABITATS IN THE ORBES REGION

4.3.1.  EFFECTS OF COAL MINING ON TERRESTRIAL RESOURCES

           In order to supply the amount of coal which will be need to fuel
the projected number of fossil fuel plants in the ORBES Region, strip,
contour, and underground mining methods will all be employed in the
Appalachian and Eastern Interior Provinces.  The biological impact from
these methods which are known and well-documented are those which result
from disturbance of the land surface, formation of acid mine drainage,
and siltation damage to aquatic ecosystems.

           Strip and contour mining destroy all existing habitats  in the
areas mined.  Because of reclamation laws enacted in  the various states
during the past approximate ten years, grading and re-establishment of
vegetative cover must be achieved.  That is, an older, more diverse, less
biologically productive ecosystem is replaced with a  younger, less  diverse,
more biologically productive one.    The newly established system can
either be permitted to return to the "climax" ecosystem in time (600 years
                              II-B-4-12

-------
are needed to re-establish the eastern deciduous forest) or be maintained
at the earlier successional state by agricultural use.

           Even though the vegetative cover can be re-established, the fate
of the animal life may be more seriously altered.  When the original habitat
is disturbed, it is assumed that the resident wildlife will attempt to in-
vade similar areas surrounding the original one.  Competition for
existing niches would result in the exclusion of individuals and possible
elimination of entire species.  If any unique areas are destroyed, it is
highly doubtful that these ecosystems would be re-established.  With re-
vegetation, the reclaimed area will be reinvaded by a succession of wild-
life species from the surrounding areas if 1) the area is allowed to re-
vert to its previous climax state and 2) if reinvasion is not prohibited
by the existence of such impediments as highwalls.

4.3.2.   EFFECTS OF THE TRANSPORTATION ON THE TERRESTRIAL RESOURCES OF THE
        ORBES REGION

           Any new transportation corridors (raillines or roads) which will
be needed for the delivery of fuels to the power plants will affect the eco-
systems in that some habitats will be destroyed entirely while others may
be divided by the roadway and thus mobility of the animal life will be
impeded.   Only the vaguest generalities can be expressed here until it is
known which ecosystems will be affected in terms of where roads will
be built.  A discussion of those areas unique to the region is included
later in section 4.4.

           The impacts on the environment associated with actual transport
processes will be the possibility of aquatic damage which would result from
barge accident or spill of 1) nuclear fuel, 2) coal, 3) spent nuclear fuels,
4) limestone sludge and flyash from SOp scrubbers, and precipitators.  Coal
and its wastes will introduce a definite chemical and solids problem to
the aquatic system.  Nuclear waste and coal wastes which contain radio-
nuclides will endanger ecosystems if the nuclides enter food chains, and
could also pose serious health problems to humans.

           Overland transport of the 4 types outlined above may pose some
problems if an accident were to occur.  Since an accidental spill of coal
could be recovered, the terrestrial stress would be expected to be endurable
by the system.  A stress on the aquatic system of the receiving water body
is anticipated to be minimal.  Loss of radionuclides would be different in
that some radionuclides used in the production of power adsorb to clay and
soil  surfaces and may tend to be more permanent.  Since the locations of
accidents are not predictable, their specific impacts also cannot be
delimited but only probable modes of action suggested.

4.4.   TERRESTRIAL SYSTEMS

4.4.1.   FORESTED AREAS

           The upland forested areas of the ORBES Region are a portion of the
North American temperate  deciduous forest biome.  The forest initially
                              II-B-4-13

-------
developed in North America, Europe and Asia during the Arcto-Tertiary
geological time period about 60 million years ago.  Over the extensive
period of human occupation, the Asian and European forests have been
extensively altered by man, and as a consequence, are now small (13).
The North American forest, however, is still present to some extent
primarily because it remained undisturbed until settlement by Europeans
in the 1600's.

           Following settlement the forest has been either completely re-
moved or extensively modified as a result of agriculture, logging,
cities, strip-mining, transportation systems, power line rights-of-way,
and many other man-related functions.  Because of this influence, it
is estimated that only 0.1% of the entire forest remains in a virgin
condition (14).  Fortunately, in those areas where reseeding has
occurred, a return of deciduous forest has been possible.  As a conse-
quence, many plant communities observed in the Ohio Valley today .repre-
sent second growth deciduous forests, or various stages leading
eventually to a new forest.

           Because climatic conditions are variable within the Ohio River
Valley, several plant associations have developed within this region.
The predominant ones are the beech-maple region covering most of Indiana
and Ohio and the western mesophytic forest of portions of Kentucky,
Tennessee, Illinois, Ohio, West Virginia, and Pennsylvania (14).  These
forests and their serai  stages are characterized by a large number of
herbaceous and woody plants that are unique to this type of habitat.

           The animal communities inhabiting the upland forests have a
large diversity of species.  In general, many of these animals are found
in other biomes, but some are associated with the system.  Two broad
groupings are often delineated, the North American deciduous forest biocia-
tion and the North American deciduous forest edge biociation (15).   Since
the deciduous forest group occurs in the late serai stages of plant
community succession and climax vegetation of mature forests, this group
has been eliminated or reduced in many locations and subsequently has
been replaced by the forest edge biociation when fields and other areas are
allowed to return to the natural  state.

           As an inhabitant of the eastern deciduous forest, man is also
dependent upon the region.  The climatic conditions of the biome are
favorable for his activities, for health and energy, high population
densities, and development of modern socieities.   Even when man settles
in open spaces, he usually attempts to duplicate the forest environment
by planting trees around his home.   The chief economic aspect of the
forest is the wood that  it supplies.   About 25% of the nation's hardwood
supply is derived from the 35 million forested acres found in the Ohio
River Basin.

           The value of  upland forested areas extends much beyond economic
utilization.   These areas, for example,  are used extensively for recreation.
                             II-B-4-14

-------
 In the Ohio River Basin area today, over 50 parks, 10 state forests, and
 2 national parks offer a multitude of recreational opportunities.  The
 forests also provide an important element to the hydrologic cycle by con-
 trolling erosion and by allowing rainfall to slowly percolate into the
 soil.  Finally, they provide laboratories for studies of natural environ-
 ments.  The pressures of human use of the environment are now of
 practical importance as well as theoretical scientific interest.  It is
 therefore important that natural areas be preserved for research studies
 so that natural environments can be interpreted (16).

           Man has had considerable impact on the deciduous forest.
 Equipped with a large variety of tools and machinery, his influence can
 be extreme.  As a consequence, some specieis such as the eastern bison,
 Carolina parakeet, waipiti, gray wolf and others have become extinct;
 other forest inhabitants currently have a rare or endangered status.
 To provide the energy and resources required by our economy, modification
 and lateration of the terrestrial  ecosystem has been accelerating.  Strip
 and contour mining for coal is the most destructive to the forest communi-
 ties because all existing habitats are destroyed in the immediate area of
 the mining.  Even though modern reclamation laws require the re-establish-
 ment of vegetation on strip-mined sites, the return of the vegetation to its
 former conditions may be uncertain.  If any unique areas were destroyed
 in the mining process, it is highly improbable they would ever become
 re-established.  Other energy-related activities are equally as destructive.
 For example, haulroads, power plant transmission lines and pipeline
 construction eliminate many forested areas and increases sediment loads
 and runoff until suitable vegetation is established within the construction
 area.

 4.4.2.  RIPARIAN SYSTEMS

           The shifting channels of rivers and the numerous streams of the
 Ohio River Valley provide conditions favorable for a variety of deciduous
 trees.  Because of regular flooding, the vegetation developing in these
 areas often does not reach the upland forest western mesophytic or beech-
 maple climx condition.  For plant ecologists,  these areas are as important
 as sand areas for the study of plant community development patterns  (14).
 In addition of serai  vegetation found along stream banks and in flood
 plains, similar communities also develop along old river channels where
 oxbow lakes, sloughs, and swamps form.   These areas are best represented
 by the river bottom swamplands of western Kentucky and southern Illinois
 (Horseshoe Lake, Heron Pond, Mermet Swamp, Pine Hills Swamp, and Henderson
 Sloughs).   These are unique because many are probably over 1,000 years old.
 During this time, animals normally extremely rare or absent in other Ohio
 River Basin habitats have colonized these unique habitats.   Examples are
 the cottonmouth water moccasin, nesting blue herons, bird-voiced tree frogs,
mole salamanders and the rare and  endangered swamp rabbit.

           Man's influence on riparian vegetation has also been extensive.
 Most floodplain vegetation in those locations  suitable for agriculture has
                             II-B-4-15

-------
been removed, and the cleared floodplains are now under cultivation.  Only
areas too wet for agriculture such as the river bottom swamplands mentioned
above have escaped modification or elimination.

           Since it is estimated that the period required for succession
of floodplain forests may be over 600 years, protection of any remaining
areas should be considered.

           Along streams with steep banks, however, beautiful riparian
forests often remain.  These forests, many of which lie within parks or
forests, provide a valuable aesthetic resource for the region, and are
utilized substantially for recreational  purposes.

4.4.3.  OPEN AREAS

           Open areas of the Ohio River Basin include the Kentucky Bluegrass
Region and those openlands originally occupied by prairie.  Both areas have
been extensively modified for agriculture, and as a result,  the basin is
one of the more productive farming areas in the nation.  These alterations
have reduced the native vegetation to the point that only small patches
such as those in the gorges of the Kentucky River are remaining (17).

           Riparian open areas in the flood plains and along streams con-
sist of natural sand bars and areas removed for agriculture.  Because of
agricultural use, any unique open areas may have already been eliminated.

4.5,  AQUATIC ECOLOGICAL RESOURCES OF THE OHIO RIVER BASIN

4.5.1.  AQUATIC RESOURCES

           The Ohio River today is the descendent of an older Teays River
and the Pleistocene glaciations of the northern half of the  basin which
covered stream valleys of northward flowing streams and diverted them into
the Ohio River.  The Teays River which was in existence 200  million years
ago was mightier than the Mississippi, which was one of its  own tributaries.
Therefore, even though flow characteristics may have changed, the current
Mississippi River and its major tributary, the Ohio, represent the major
drainage system of the North American continent.  The long association of
these systems has provided an uninterrupted opportunity for  the evolution
of freshwater organisms for over 200 million years.

           The series of glacial invasions ending 10,000 years ago also
have had profound effects on the aquatic resources of the basin.   The
glacial margins reached only to the upper one-third of the basin.   As a
result, the Basin contains northern fauna! elements which were pushed to
the southernmost limits of their distribution as well as southern elements
already present.
                             II-B-4-16

-------
            Within the basin  three  aquatic  habitats  other  than  the  Ohio  River
 itself are unique:   a) the subterranean  streams,  seeps, and  pools  found in
 the many caves of the Karst  regions  of southern  Indiana and  Illinois  and
 Kentucky,   b)  the river bottom swamps  and  marshes of  southern  Illinois  and
 western Kentucky, and  c)  the  numerous small,  clear streams  draining  those
 areas of Indiana, Ohio and Kentucky  which  were originally covered  by  the
 eastern deciduous forests.

            We  have,  in another section of  this report, catalogued  the
 sport and  commercial  fishery resources of  the  basin and pointed out that
 the principal  sport  fisheries  are  based  upon the  largemouth  and small-
 mouth basses,  crappie, channel  catfish,  bluegill, other sunfishes, and
 white bass.  Considerable  effort is  also made  by  the  state fish and game
 departments  to provide trout fisheries in  the  region  through hatchery
 releases.   In  general, fishing is  moderate to  good, but Kentucky small-
 mouth bass streams provide excellent fishing.  Historically, the Mississippi
 River drainage,  including  the  Ohio River,  has  supported important  commercial
 fisheries  for  food fishes.   Although pollution, siltation, and navigational
 practices  have adversely affected  these  fisheries,  significant numbers  of
 carp, catfish, and drum are  still  being  taken  from  the Ohio  River, some of
 its larger tributaries (.e.£.,  the  Wabash)  and  from  the larger  impoundments
 of the region.

            The fish  fauna  of the Ohio  River Basin is  one  of  immense variety
 and antiquity.  In each of the states  in the region are found  177  to  210
 species of freshwater fishes or roughly  one-third of  all  the fish  found
 in the
-------
in vertebrate animals to an unusual habitat.  The value of these rare
animals has been recognized and some of the known cave habitats in which
they are found are protected in one form or another (e_.cj_, Mammouth Cave
National Park).  However, the danger is increasing that man's activities
will degrade the quality of subterranean water in the basin to the detri-
ment of these and other unique cave organisms.

           The unique geologic history of the Ohio River Basin which re-
sulted in the great diversity of fishes has also resulted in a diverse and
unique fauna of freshwater mussels.  Freshwater mussel shells were
commercially important in the 1920's and 1930's as a source of material
for manufacturing buttons.  In the 1950's  and 1960's the mussel industry
was revived when Japanese pearl culturists began using pellets of mussel
shell as the "seed" for pearl culture (24).  Today the industry is much
reduced (as are the mussel populations in some areas).  Several mussels
are included on the state lists of endangered species.
                             II-B-4-18

-------
                              LITERATURE CITED
 1.  Parker, F.L., 1974.  Other Ecological  Impacts in Energy, Environment,
       and Human Health, A. J. Finkel (ed.), Publishing Sciences Group,
       Inc., Mass.

 2.  Clark, J.P., 1969.  Thermal  Pollution  and Aquatic Life in Scientific
       Technology and Social Change, Scientific American.

 3.  Eisenbud, M., 1974.  Health  Hazards from Radioactive Emissions in
       Energy, Environment, and Human Health, A.  0.  Finkel  (ed.),
       Publishing Sciences Group, Inc.,  Mass.

 4.  Martin, J.E., Harward, E.D., and Oakley, D.T.,  1969.   Radiation Doses
       from Fossil Fuel and Nuclear Power Plants  in  Pov.'er Generation and
       Environmental  Change, Berkowitz,  D.A., and A.  M.  Squires (eds.),
       Symposium of the Committee on Environmental Alteration, ;\AAS,
       December 28, 1969.

 5.  Reichle, D.E., Dunaway, D.B., and Nelson, D.S.,  1970.   Turnover and
       Concentration of Radionuclides in Food Chain  in Nuclear Safety 11
       (1): 43-55.

 6.  American Chemical  Society, 1969.  Cleaning:  our Environment, The
       'Chemical Basis for Action.  Washington, D.C.

 7.  Beyers, R.J., 1974.  Ecologice"1 Impacts of Energy Production on
       Rivers and Lakes in Energy, The Environment,  and_ Human Health,
       Finkel, A. J.  (ed.), Publishing Sciences Group, Inc., Mass.

 8.  Brisbin, I.L., Jr., 1973. Abundance and Diversity of  Waterfowl
       Inhabiting Heated and Unheated Portions of a  Reactor Cooling
       Reservoir in Thermal Ecology, Augusta, Ga., May 3-5, 1973,
       Gibbons, J.W.. and Sharitz, R.R.  (eds,), AEC  Symposium Series
       (Conf. 1730505)  as referenced in  the above.

 9.  Strangenberg, M.,  and Pawlaczyk, M.Z., 1961.  The Influence of
       Warm Water Influex from a  Power Station upon  the Formation of
       Biocenotic Communities in  a River.   Nauk Pol  WR,  Wroclaw, No. 40,
       Inzyn Sanit I: 67-106.  Water Pollution Abstracts 35 (3),
       Abst. No.  579 as referenced in #8 above.

10.  Butz, B.P.,  Schregardus, D.R., Lewis,  B., Po Licastro, A.J., and
       Reisa, J.J., Jr., 1974. Ohio River  Cooling Water Study. EPA
       905/9-74-004.   386 pp.

11.  Marcy, B.C.   Vulnerability and Survival  of Young Connecticut River
       Fish entrained at a Nuclear Power Plant Canal.   J.  Fish. Res. Bd.
       Canada 30: 1195-1203.
                             II-B-4-19

-------
12.  Energy Alternatives.  A Comparative Analysis.   Prepared for CEQ, ERDA,
       EPA, FEA, FPC, DOI and NSF by the Science  and Public Policy Program,
       University of Oklahoma, Norman, Oklahoma,  May, 1975.

13.  Walter, H., 1973.  Vegetation of the Earth in  Relation to Climate and
       the Eco-physiological Conditions.  Springer-Verlag, New York, 237 pp.

14.  Shelford, V.E.  The Ecology of North America:  University of Illinois
       Press, Urbana, 610 pp.

15.  Kendeigh, S.C., 1974.   Ecology with Special  Reference to Animals and
       Man, Prentice-Hall,  Englewood Cliffs,  N.J.,  474 pp.

16.  Jacobs, J., Lange, O.L., Olson, J.S., and Wieser, W., 1970.
       Analysis of Temperate Deciduous Forest  Ecosystems,  Springer-Verlag,
       New York, 304 pp.

17.  Braun, E.L., 1967.  Deciduous Forests of  Eastern North America, Hafner
       Publishing Co., New York, 596 pp.

18.  Trautman, M.B., 1957.   The Fishes of Ohio, Ohio State University Press,
       Columbus, Ohio, 683 p.

19.  Gammon, J.R., and Gerking, S.D., 1966.   The  Fishes,  pp. 401-425 in
       A. A. Lindsey (ed.), Natural Features  of Indiana,  Indiana Academy
       of Science, Indianapolis, Ind.

20.  Smith, P.W., 1971.  Illinois Streams: A  Classification Based-.on their
       Fishes and an_ Analysis £f Factors "Responsible for  Disappearance o_f
       Native Species.  Biol. Notes #76, 111.  Nat.  Hist.  Survey, 14 pp.

21.  Clay, W.M., 1975.  The Fishes of Kentucky, Ky.  Dept.  Fish and Wildlife
       Resources, Frankfort, Ky., 416 pp.

22.  Hubbs, C.L., and Laeler, K.F., 1975.   Fishes of the  Great Lakes Region.
       Univ. of Michigan Press, Ann Arbor, 213 pp.

23.  Miller, R.R., 1972.  Threatened Freshwater Fishes of the United States,
       Trans.  Amer. Fish Soc. 101 (2): 239-252.

24.  Williams, J.C., 1969.   Mussel  Pishery Investigations.   Tennessee, Ohio
       and Green Rivers Final Report, Ky.  Dept. of  Fish and Wildlife
       Resources, 107 pp.
                             II-B-4-20

-------
                         .5.0 ECONOMIC IMPACTS


 5.1 INTRODUCTION

       This section of the technology mini-assessment provides initial, broad
 based estimates of the primary economic impacts regarding future energy
 conversion alternatives in the Ohio River Basin (ORBES) .  These estimates
 are based on national level scenarios developed by the U.S. .Bureau of Mines
 (BOM) and the Ford Foundation .(Ford Tech Fix) . *
       Two BOM scenarios call for an installed electrical generation capacity
 of 245,000 MW within the ORBES region by the year  2000; the first scenario
 representing an electrial generating fuel mix which is 80 percent coal fired
 and 20 percent nuclear, while the second scenario calls for an installed
 capacity which is 50 percent coal fired and 50 percent nuclear.

       The two Ford Tech Fix scenarios call for an installed electrical capacity
 of 94,000 MW by the year 2000. The first of these scenarios is achieved by
 the addition of new generating units  which are all coal fired, and the second
 scenario is based on the addition of new units which are fueled solely by
 light water reactors.

       Accordingly, this section of the technology assessment makes reference
 to these scenarios as follows:

       Scenario  I   BOM 80/20 coal/nuclear           .  .         .
       Scenario  II  BOM 50/50 coal/nuclear
       Scenario  III Ford Tech Fix, 100 percent coal
       Scenario  IV Ford Tech Fix 100 percent nuclear

       The evaluation of scenarios developed at the national level and disag-
 gregated the regional level presents  major diffiduties, especially with regard
 to the economic impacts.  Clearly the most important of these is the absence
 of price estimates for fuels,  materials and labor,  which in turn precludes
 estimation for the prices of electricity and the elasticity of these prices.
 Energy, like any other commodity, is subject to production and market forces
 and the culmination of these  forces combine to determine  the final demand.
 Moreover, it is  the determination of the final demand for energy  on a regional
 basis that will impart the real-and substantial economic impact.  Energy will
 always continue to be an important factor of production, and the regional
price for energy as a factor of production,  ceteris paribus,  will  be extremely
in determining the occupational and industrial structure of a given region.

      With this  background  limitation, this section of the  technology assessment
instead focuses  on the broad scale estimation of regional employment and
income associated with the construction and operation of electrial generating
 units.  We are particularly interested in a broad evaluation of how particular
 groups of county economies would be impacted if  an electrical generating
 station were built in a specified county.  In addition, we  have provided some
1. See references 1 & 3
                              II-B-5-1

-------
first-cut estimates of the capital formation required to build the number of
power plants specified by each scenario.

      Finally, we have identified selected areas of analysis in the study that
should receive particular attention in subsequent phases: areas that will indeed
permit a more valid estimate of the economic impact of future energy conversion
alternatives in the ORBES region.

5.2. REGIONAL EMPLOYMENT AND INCOME

5.2.1. BASELINE CONDITIONS

       In 1974, the ORBES region had a total population of 17.7 million, total
employment was 5,297,000, the gross regional product was 93.7 billion, or
6.2% of the U.S . gross national product.  GNP per capita in the ORBES
region was $5,295, which was about 2.5% below the average for the  U.S.
(table II-B-5-1).

                              Table II-B-5-1

              .   POPULATION, EMPLOYMENT AND GNP, 1974
Region*
USA
Illinois
Indiana
Kentucky
Ohio
ORBES
Population
(thousands)
211,909
3,881
4,113
3,357
6,341
17,692
Employment
(thousands)
85,936
893
1,267
826
2,310
5,297
GNP**
($ billions)
1,511.7
18.2
21.2
14.9
39.4
93.7
GNP per capita
$(1974) (1958)
5,434 3,530
4,682 3,041
5,149 3,344
4,438 2,883
6,217 4,038
5.295 3,439
*  For the states,  all data refers only to the parts of those states within
   the ORBES  boundaries.
** For regions small than the USA, "GNP"  refers to Gross Regional Product
   (GRP).


SOURCE: U.S.  Statistical Abstract, 1975 and City and County Data Book, 1972


      Employment by sector is not reported for the ORBES region.  However,
the information is available for the four-state region which includes the ORBES
region (table 2).
                             II-B-5-2

-------
                                 Table IIB-2

             EMPLOYMENT BY SECTOR (TOTALS FOR THE STATES
                  ILLINOIS,  INDIANA, KENTUCKY AND OHIO)

 Category          1970        %       1973        %          2000
Total
Agriculture
Mining
Construction
Manufacturing
Tr anspor tation
Wholesale
Retail
F.I.R.E.*
Services
Unclassified
9,125,990
17,538
70,
426,
3,756,
351,
623,
1,683,
500,
1,453,
56,
868
1.90
523
053
899
440
117
709
613
0
0
4
41
5
6
18
5
15
0
.2
.8
.7
.3
.8
.8
.5
.5
.9
.6
9,551,708
20
77
461
3,779
551
631
1,833
546
1,611
38
,129
,698
,193
,233
,562
,271
,519
,801
,653
,649
0.
0.
4.
39.
5.
6.
19.
5.
16.
0.
2
8
8
6
8
6
2
7
9
4
14,423,
30,
177,
908,
3,903,
705,
701,
3,154,
1,220,
3,600,
3,600,
079
565
850
359
947
496
657
900
788
482
482

0.2
1.2
6.3
27.1
4.9
4.9
21.9-
8.5
25.0

SOURCE:  County Business Pattern (State Summaries),  1970 and 1973
* Finance, Insurance and Real Estate


      In 1970, the manufacturing sector was the major employer, accounting
for 41.3% of all employment. Retail trade and services were important employers,
Agriculture and mining together, provided only 1% of the region's employment.
It should be noted that the sectoral employment patterns in the four state region
may not be fully representative of the ORBES region.  Chicago and the industrial
cities along the Great Lakes in Indiana and Ohio are included in the four state
region,  but not in the ORBES region.  Thus, we could expect the percentages
reported for the four state region to overstate the manufacturing  sector and
understate the agricultural and mining sector for the ORBES region.  There
is another reason to believe that agricultural employment is understated:  the
data report only employees covered by FICA.

5.2.2. INCOME PROJECTIONS: SCENARIOS I AND II

      The scenarios (1), applied nationwide,  are projected to generate a
GNP of $2,105 billion (constant 1958 dollars) in the year 2000.  We project
GNP for the ORBES region to be $155.8 billion (1958 dollars) in the year 2000.
Per capita GNP in the ORBES region is projected to be $7,039 billion (1958
dollars) in the year 2000.  Note that this is a little more than twice the 1974
ORBES region per capita GNP (tableH-B-5-lX

      Scenario I is the BOM scenario, assuming that 80 percent of the new
generating capacity will be coal-fired  and 20 percent will be nuclear. Currently,
89.7% of existing capacity in the region is coal-fired (2). Thus the above GNP
projections appear reasonably reliable for Scenario I.
                              II-B-5-3

-------
       Scenario II assumes a 50/50 coal/nuclear mix. Since the ORBES region
 is a major producer of coal, it is predictable that Scenario II, with its lesser
 emphasis on coal utilization,  would result in somewhat lower regional and
 per capita GNP in the year 2000, compared with scenario I.  The quantitative
 estimation of the  difference in income under scenarios I and II is a matter
 for further detailed research.

 5.2.3. INCOME PROJECTIONS:  SCENARIOS III AND IV
       The Ford Technical Fix scenario (3), applied nationwide, is projected
 to generate a GNP of $1,320 billion in the year 2000 (1958 dollars). We project
 GNP for the ORBES region to  be $97.7 billion  (1958 dollars) for  the year
 2000. This amount to $4,413 per capita. The Tech Fix scenario  would permit
 per capita GNP in the year 2000 to increase 28% above 1974 per capita GNP
 (table 1).

       Scenario III assumes all new generating capacity will be coal-fired.
 Since 89.7% of existing capacity in the region is coalfired,  the above GNP
 projections are probably reasonable for Scenario III. However,  Scenario
 IV assumes all  new capacity will be nuclear.  Considering that the ORBES
 region is a major producer of coal, one would expect that scenario IV would
 work to the relative disadvantage of the ORBES region.  Thus regional and
 per capita GNP would be somewhat lower than the above projections, if scenario
 IV were to be followed. How much lower is a question for further in depth
 research.

 5.2.4 EMPLOYMENT PROJECTIONS

      Employment in the year 2000 in the four state region was projected
 (table 2) . The method of projection involved extrapolation of total sectoral
 employment trends in the 1970-73 period, making some necessary adjustments
 in sectoral employment projections. Total employment is projected-to grow
 by 51 percent between 1973 and 2000, while population is projected to grow
 by 34 percent.  We believe the total employment projection is somewhat overstated,
 since it projects continuation of two trends (the increase in the number of
 employee categories covered by FICA and the increase in the proportion
 of women entering the work force).  While these trends will most likely continue,
 the rate of increase is expected to become lower, as workers covered by
 FICA and working women approach saturation levels. Thus, projected employ-
 ment is overstated.

      Projected total employment in the ORBES region is 7,998,000, for the
 year 2000.  Our employment projections for the four state region (table 2)
 indicate that, relative to total  employment, employment in the mining, construction
retail and service sectors will grow more rapidly, while employment in the manu-
facturing and transportation sectors will grow less rapidly.

      We have not made employment projections for each of the four scenarios, '
 such projections would require research beyond the scope of a mini-assessment.
However, we would expect employment in total and by sector, to be somewhat
 sensitive to the scenario followed.  We expect employment to be influenced
by the scenario followed, but to much lesser extent than income and GNP.


                              II-B-5-4

-------
 5.3 CAPITAL REQUIREMENTS FOR ENERGY DEVELOPMENT

       Assuming that the capital cost of nuclear generating capacity amounts
 to  $1.2 billion per 1,000 megawatts, and coal fired capacity is $0.9 billion
 per 1,000 megawatts plus $100 million to establish an underground coal mine
 and facilities to feed the plant, the capital requirements to establish the
 additional generating capacity envisaged for the 1985 to 2000 period where
 estimated be:

                               Table II-B-5-3

                CAPITAL REQUIREMENTS AND GNP  (1985-2000)

 Scenario     Capital Costs* ORBES      Capital Costs* USA    -Projected GNP
               ($ billions)                ($ billions)        (total 1985 to  2000)
I
II
III
IV
180
192
36
42
2277
2428
455
531
50,952
50,952
35,590
35,590
 * Includes capital costs of new generating plant,  transmission facilities
   and underground coal mines.

SOURCE:  Derived from ORBES Task I Report (Ref.2) and personal communications
          with several electric utilities.


      Historically, total business investment on plant and equipment has been
about 7.5% of GNP, and the electric utility industry has accounted for 16% of
that total (i.e., 1.2% of GNP)(4).Table 3 indicates that, in order for the four
scenarios to be followed during the 1985 to 2000 period, the electric utility
industry would need to use the following percentages of GNP for capital
investment in plant and equipment:

      Scenario I                      4.5%                .           .
      Scenario II                     4.8%

      Scenario III  .                  1.3%

      Scenario IV                    1.5%

      Clearly scenarios I and II  would require a massive reallocation of GNP
toward capital investment in electric utilities.  If the historical trend that
electric utilities command 16% of total investment in plant and equipment is
maintained (and this is a reasonable assumption, since new industrial and
other  plants  and equipment will be needed to utilize the electricity produced),
the proportion of GNP allocated to new plant and equipment would need to
increase from the current 7.5% to 28.1% (for scenario I)  arid 30% (for scenario II)
during 1985-2000. This would imply a scarcity of capital  far beyond the
experience of our economy.  Scenarios I and II may be infeasible on the
grounds of capital requirements alone.

                              II-B-5-5

-------
 5.4 DIFFERENTIAL IMPACTS WITHIN ORBES REGION                      .

 5.4.1 BASELINE CONDITIONS  .

       To consider the local area economic impacts of power plant location,
 construction, and operation, counties were chosen as the basic evaluative
 unit.  Abrupt shifts in local employment and income along with the concurrent
 sharp increases in  demand for services, both public and private, are often
 associated with investment projects of the scale and magnitude of contemporary
 generating stations  (5). Actual local economic impacts, however, will be bounded
 by three critical factors, the size and diversity of the local economy, the proximity
 of the plant  site to  urban or metropolitan centers, and the number and size
 of electrical  generating units situated within a county (6) .

       To differentiate intracounty economics by source of income and employment,
 indices of economic diversity were calculated using available census data.
 Employment and income generated within the agricultural and mining sectors
 of each county were summed, and the percentages of total employment and
 income contributed by the agricultural and mining sectors were calculated
 (figures II-B-5-1 & 2).  County economics traditionally dependent on either
 mining or agricultural sectors are well defined by income percentages greater
 than fifteen percent in figure II-B-5-2.  Historically more diversified counties,
 counties whose economies include a relatively large percentage of income generated
 by  their manufacturing and service sectors, tend to be distinguished by agricultural
 and mining sectors  of less economic importance,  an income percentage of less
 than 1.5% in figure II-B-5-2.  Local economies more recently experiencing diver-
 sification toward manufacturing and growth in services, tend to be included
 in the range from 1.5% to 15% in figure II-B-5-2.

      An effort to delineate counties with regard to their proximity to subregional
 economic centers is shown in figure II-B-5-3 where counties are classified
 as metropolitan (SMSA), (7)  are indicated, as are counties situated beyond
 commuting distances (defined as two county widths) from a SMSA or a city
 with at least 20,000 population. Only two counties with Illinois  and  Indiana
 along with the Southeastern portion of Kentucky, remain isolated from both
 metropolitan  areas and urban centers.  Unmarked counties in figure II-b-5-
 3 are within two counties of an SMSA as  a city of at least 20,000  population.

      With the location of future power generating units, specified by  scenario
 in Figure II-B-5-4 through 8, an initial assessment of local economic impacts
 is possible.  Prior to the construction phase, localized economic impacts are
minimal.  County merchants and manufacturers are urilikely to experience
 a rise in purchase orders for their merchandise or products.  Both the specialized
material requirements arid the highly developed supply linkages which satisfy
these requirements tend to minimize local pre-construction procurement even
where the generating unit is located within  a metropolitan areas (8) . Rising
property values are likely as land is purchased for the site but the actual
 effect on land prices and thus revenues collected through taxation will vary
widely in response to local expectations  and tax schedules.
                               II-B-5-6

-------
                  Figure II-B-5-1

     PERCENT EMPLOYED IN AGRICULTURE AND MINING'
                                 Less  then 1.0%

                                   1.0 to 10.0%

                                   Above 10.0%

                           Information Withheld
" Employed persons covered by FICA
SOURCE: County Business Patterns,  1974.
                      II-b-5-7

-------
                     Figure II-B-5-2

          PERCENT OF COUNTY PERSONAL INCOME FROM
                  AGRICULTURE AND MINING
      .'
     \ V- /^{_ J '	!£_'j ''-1	"	•" S •"•""   	^	
                                   Less than 1.5%

                                     1.5 to 15.0%

                                      Above 15.0%

                             Information Withheld
SOURCE: BEA, Local Area Personal Income, 1969-1974.
                       TI-B-5-8

-------
                      Figure II-B-5-3



           PROXIMITY OF ORBES COUNTIES TO SMSA'S
       m^Km\
                          P-;;,•„".:;.;] ,.-'	.^.v-;;1  .-.""i

                          S f-.:]//'tW^&y^.	-
                          ?,;- ;;• :-  ,! ;;\	•'•/:>*,^JttL>*< I.'.-
           SMSA Counties
           Counties separated from cities with 20,000

           population or more by two adjacent  counties
SOURCE: U.S. Bureau of  the  Census
                       II-B-5-9

-------
                            Figure II-B-5-4



                        1975 GENERATING CAPACITY
                                   1 L  -  c1*..^
                                   •,]no-..o 9»»'Bv, ,„
                                   --\.		L «
                                                  | SMI IB*



















                                                	'"T7 - —	' , '-S !	f»  HOC-ING J I... f»*i.ti«VM "1
(^oS°"CMA   r*t *'!• — f V

Plants present or under construction:




     (Total megawatt capacity.)




        Less than 1000
                                         Number of units scheduled for

                                            operation by 1985:-

                                           (Megawatts per unit = 200 or ir.ore)




                                               •   Fossil fuel
        1000 - 1999
   '•//,  2000 - 3499
        3500 or more
                                                     Nucloo.r fuel
                                          SOURCE:  ORBES  Task  I  Report,

                                                   (See Ref. 2)
                              II-B-5-10

-------
                   Figure II-B-5-5

NUMBER OF ELECTRIC GENERATING UNITS BY TYPE OF FUEL:

               BOM SCENARIO I (80/20)


                                             • Coal

                                             n Nuclear
                 1985 MEGAWATT CAPACITY
  Le'ss  than
    1000  MW
 1000 -
1999 MW
 2000 -
3499 MW
3500 MW
or more
SOURCE: Development of Plausible Future Regional. Technology
     ":  Configurations, Ohio River Basin Energy Study Task 1
        Report. October 18,. 1976.
                        II-B-5-11

-------
                     Figure II-B-5-6       -   .

NUMBER OF ELECTRIC GENERATING UNITS BY TYPE OF FUEL-IN
                 BOM SCENARIO. II (50/50)

                                           1000 MW  of

                            •     .          "Coal
                                           a Nuclear
                                                      2000
                 1985 MEGAWATT CAPACITY
  Less  than
    1000  MW
                   1000  -
                  1999 MW
 2000 -
3499 MW
3500 MW
or more
SOURCE: Development of Plausible Future  Regional  Technology
        Configurations,  Ohio River Basin Energy Study  Task I
        Report.  October 18,  1976.
                       II-B-5-12

-------
                     Figure II-B-5-7

NUMBER OF ELECTRIC GENERATING UNITS BY TYPE OF FUEL IN  2000

          FORD TECH FIX SCENARIO III (100% COAL)
                                          600 MW-units

                                           • Coal
                 .1985 MEGAWATT CAPACITY
  Less than
    1000 MW
 1000 -
1999 MW
 2000 -
3499 MW
3500 MW
or more
SOURCE: Development of Plausible Future Regional  Technology
        Configurations,  Ohio River Basin Energy Study Task  I
        Report. October 18,  1976.
                       II-B-5-13

-------
                     Figure II-B-5-8

NUMBER OF ELECTRIC GENERATING UNITS BY TYPE OF FUEL IN  2000

        FORD TECH FIX SCENARIO IV (100% NUCLEAR)   .


                                          1000 MW units

                                          a Nuclear
                  1985 MEGAWATT CAPACITY
   Less than
     1000 MW
 1000 -
1999 MW
 2000 -
3499. MW
3500 MW
or more
 SOURCE: Development of Plausible Future Regional Technology
         Configurations, Ohio River Basin Energy Study Task I
         Report. October 18, 1976.   ...
                         II-B-5-14

-------
      As actual construction approaches, great potential exists for changes
 in local employment and income patterns. Because local availability of the
 highly skilled labor force necessary for power plant construction and operation
 is not considered a primary constraint to location,the possibility  of large scale
 in-migration of workers and associated economic impacts must be considered
 (9) .  Where plants are located within commuting distance of metropolitan areas,
 the required labor force can generally be recruited from the metropolitan area
 with only a small percentage of the work force changing residence to the work
 site locality (10) .  Where the plant site is located far from metropolitan areas
 but within commuting distance of urban centers, the work force will tend to
 disperse residence toward the urban center, where more diverse services
 and consumption alternatives are available (11).  Only where the plant site
 is well isolated from commercial centers can the abrupt effects of large scale
 immigration on small communities be expected.  Even in isolated areas the
 impacts of  a large impermanent work force will be moderated by the dispersion
 of residences within a wide radius of the work site (12) .

      By determining the possibility of local residence and the ability of the
 local economy to satisfy the demand of resident plant employees for goods and
 services, the local economic impact of power plant location can be projected.
 Total impact of changes in employment and income will include both the direct
 increases in employment and income created by the power station itself, and
 the induced additional employment and income generated within commercial
 and governmental sectors which supply goods to the plant and its employees.
 Generally, the multiplier effect of direct employment and income generated
 by the installation of a new  industry will be greater as the size of the region
 considered increases for the larger region will encompass a greater proportion .
 of induced  employment and income.  In smaller areas, income generated by
 new plants tends to flow out from the local economy in return for goods and
 services unavilable locally. The large import sectors of small economies result
 in lower multipliers.  For regions of similar size, income and employment
 multipliers tend to be smaller iprf* regions which exhibit less economic diversity.
 This effect again,  is related to the relatively large "leakages" from undiversifie.d
 economics  (13).

      Local impacts of energy development Tables II-B-5-4,5 and 6 relate the
 number and type of power generating units specified by each scenario to the
 proximity of the host county to metropolitan or urban centers, and the relative
 importance of the agriculture and mining sectors to the local economy.  Each
 unit represents a facility producing 1000 megawatts  except in Scenario III
 where coal fired units ar.e rated at 600 megawatts. Approximately- 1100 workers
will be employed during the construction phase of any single 1000 megawatt
 unit.  Additionally, at least 80% of the work force will be professional or highly
 skilled labor. Construction phases of the total number of units projected for
any one scenario are assumed to be staggered in deference to efficient utilization
of the skilled labor pool. Efficiencies of scale suggest that generating facilities
will encompass two to four generating units and employ an operational staff of
 120 to 150 professional and skilled workers (14).
                               II-B-5-15

-------
                             table I I-B-5-4

              NEW SITES PLANNED WITHIN SMS A'S (1985-2000)


      Scenario          Coal               Nuclear         •    Total
I 40
II 24
HI 10
IV % . 0
3
14
0
5
43
38
10
5
SOURCE:   Development of Plausible Future Regional Technology Configurations,
           Ohio River Basin Energy Study Task I Report (October 18, 1976)
                            II-B-5-16

-------
                               Table! 1-5-5               .'    .   .

        NEW SITES PLANNED OUTSIDE SMSA'S BUT WITHIN COMMUTING
             DISTANCE OF SMSA OR CITY OF 20,000 POPULATION
                     Undiversified
                     County
County Economics Diverse    Unclassified
of Intermediate    County         by
Scenario
I


II


III


IV


Fuel
Coal
Nuc.
Total
Coal
Nuc.
Total
Coal
Nuc.
.Total
Coal
Nuc.
Total
Economics*
33
7
40
20
25
45
3
0
3
0
1
1
Diversity*
40
12
52
28
27
55
10
0
10
0
6
6
Economics* Diversity*'
9 2
4
13
7 2
9
16
3
0
3
0
3
3
 *  Economic diversity determined by sources  of income
**  Unclassified due to lack of published information

SOURCES:  Income  data from "U.S. Dept. of Commerce, Bureau of
           Economic Analysis,  Local Area Per sonar In come, 1969-1974.
           Vols. I, II,  Washington D.C., 1976
           Plant site data  from  "Development  of Plausible Future  Regional
           Technology Configurations, Ohio River Basin Energy Study Task I
           Report  (October 18,  1976).
                              II-B-5-17

-------
                                   Table II-B-5-6

             NEW SITES PLANNED OUTSIDE SMSA'S BUT WITHIN COMMUTING
                  DISTANCE OF SMSA OR CITY OF 20,000 POPULATION
Scenario Fuel
. I


II


m


IV '

Coal
Nuc.
Total
Coal
Nuc.
Total
Coal
Nuc.
Total
Coal
Total
Undiversified County Economics Diverse Unclassified
County of Intermediate County by
Economics* Diversity* Economies* TOTAL Diversity**
18
2
20
14
11
25
5
0
5
2
2
50
13
63
31
35
66
7
0
7
7
7
14
16
30
7
12
19
4
0
4
1
i
82
31
113 3
52
58
110 3
16 •
0
16 0
10
10 . 0
 * Economic diversity determined by sources of employment
** Unclassified due to lack of published information

SOURCE:   Employment data  from U.S.  Dept. of Commerce,  Bureau of the Census,
           County Business  Patterns, Washington D.C. 1974.
           Plant Site data from "Development of Plausible Future Regional
           Technology Configurations, Ohio River Basin Energy Study Task I Report
           (October  18, 1976).
                                   II-B-5-18

-------
       Within metropolitan areas, most of the skilled labor necessary during the
 construction phase will be supplied by the local labor pool. The proportion of
 locally available skilled labor will largely depend upon the wage rate offered
 by the construction firm and local employment levels. The multiplier effects
 of employment and income derived from the construction power plants within
 metropolitan areas will be quite high (15.)  Impact on the local economy will
 tend to be stimulating rather than traumatic since the magnitude of expenditures
 of those directly employed by the construction firm will be small in relation to
 total expenditures within the local metropolitan economy. As mentioned pre-
 viously, no significant local impact will occur due to procurement of construction
 material.  Expenditures for the required materials will be dispersed throughout
 the region and the nation. Termination of the construction phase may signal
 shortrun rises in unemployment within the affected skill areas as workers
 search for new positions locally or await employment elsewhere.  As the
 facility enters the operational phase,  employment and income generated by
 the facility will approach insignificance when compared to the metropolitan
 economy as  a whole.  Power facilities constructed within commuting distance of
 metropolitan areas will create only minor impact on the local economy (16).

       Lack of locally available skilled labor will induce construction firms to
 hire mostly from within metropolitan areas. Again, insignificant procurement
 of construction materials will occur locally and only a small portion of the labor
 force will choose to change residence to the construction locality.  The incidence
 of employment and income multiplier effects will be primarily toward the metro-
 politan area, and toward the host county to a much lesser extent.

       Where power facilities are located beyond commuting distance of metropolitan
 areas but are within commuting distance of smaller urban center, local economic
 impacts of the construction phase would tend to be greater. However, residence
 of the labor force would conform to the availability of housing and services, only
 a small percentage of which could be provided by the locality itself.  Given the '
 experience of large manufacturing firms in rural  areas similar to  those of the
 ORBES region, the multiplier effects would be dispersed over a wide area,
 resulting in relatively small local impacts (17).  The host county  may experience
 the negative effects of providing public services to the power facility while fore-
 going the benefits of the income generated by the facility.  The most commonly
 experienced negative effect is inadequate sewage treatment facilities. (18p.28)'..
 The need for those facilities is immediate and the  cost of providing them is large;
 almost always beyond the budgetary capabilities of the host county. Other negative
 effects, but less severe in nature, include the provision of public safety services
road maintenance and educational facilities (19 p.26). With the termination of
 the construction phase, the effects of unemployment would be dispersed over a
large area.  Entering the operational stage,  the power facility would little affect
the host county's economy through either employment or income.  The economy
of the nearby urban centers would retain income and employment,  since the opera-
tional labor force would most likely choose to reside in the  center.

      Scenarios I and II would differ little in their local area economic impacts
 during the construction phase. Scenario I shows a greater percentage of coal
fired generating units located within metropolitan areas,  while Scenario II
 substitutes nuclear powered facilities in counties contiguous to metropolitan
 areas.  The high concentration of power generating units also   the major water
                               II-B-5-19

-------
 courses of the region corresponds to traditional patterns of urban and metropolitan
 growth.  Only in Russell County, Kentucky, would the construction phase pose
 severe and disruptive demand upon traditional economic patterns.  Here the
 traumatic effects of a large temporary labor force upon a small local economy
 may parallel those of western boom towns.  It must be noted however, that
 Russell County is experienced in these matters, having beeri the construction
 site for Wolf Creek Dam, and is currently the site of structural improvements
 to that dam.  Russell County has a large  supply of vacation homes, which could
 supply temporary housing for construction workers.  Thus, Russell County will
 experience major impacts, but is equipped to handle them fairly well.
                                                          i
       Once plants entered operation,  the effects of Scenario Land II would diverge.
 As discussed  earlier, the predominately  coal based Scenario I would tend to promote
 further development of ORBES region coal fields while the coal nuclear Scenario II.
 would only partially encourage the  development of regional fuel reserves.  However,
 economic  impacts would differ not by the impact of the county hosting the generating
 facility but through the induced employment and income effects within coal producing
 areas of the ORBES region.  Scenario I would reinforce the current trend in develop-
 ment of regional coal reserves while Scenario II would diminish this trend.

       Scenarios III and IV represent siting allocations which minimize the deleterious
 impacts of large increases or declines in demand for skilled labor during construction
 phases. All plants  are located within commuting distance of metropolitan areas
 or urban centers.  For the host counties, no major economic impacts distinguish
 the two scenarios.  However, again, the  two imply vastly different impacts on
 regional coal field areas, depending on the coal vs. nuclear fuel choice. Neither
 during the construction or operational phases would major income or employment
 effects be manifested within the economy  of the  site locality.

      In general, the pronounced or traumatic local economic impacts sometimes
 associated with large scale investment projects  will be minimized in  the ORBES
 region during the construction and  operation of power generating facilities as
 outlined by the four  scenarios.  In part, this relatively small localimpact is
 due to the capital intensive nature of investment in electrical generating units.
 The technological sophistication, design, and volume requirements can seldom
 be met by suppliers within the local economy.  Thus, for the most part the  large
 direct capital expenditures bypass the local economy. The labor force required
 for both construction and operation  is largely professional or highly skilled, and
 therefore usually found in the metropolitan areas. Finally, the spatial proximity
 of the selected sites to areas where  the labor force resides permanently encourages
 commuting to the work site,  rather than large scale in-migration.  Though the
 high energy Scenarios I and II may  cause some negative impacts as construction
 labor forces shift between jobs, this will  be largely limited to metropolitan areas,
 most likely by size and diversity to  offer re-employment opportunities.

      This largely qualitative analysis suggests that the traumatic effects of large
 scale energy development will be minimized within ORBES. In-region shifts
in employment and income associated with energy development will not usually
be urban-to-rural in direction.  Shifts toward river  based  sites will occur and
                               II-B-5-20

-------
 be most pronounced in Scenario II.  There will be income and employment shifts
 toward the in-region coal field, counties in Scenarios I, II and III (in that order of
 magnitude).  For the total ORBES region, income and employment patterns are
 projected in Section 1 of this report) .

 5.5. QUESTIONS TO BE ADDRESSED IN FUTURE PHASES OF THE STUDY

    (a) Scenarios I and II were based on the BOM national scenario and
 scenarios III and TV were based on the Ford Tech Fix national scenario.
 All scenarios implicitly assume energy developments in ORBES region
 will parallel projected nationwide developments.
       However, the ORBES region, with its relatively abundant coal and
 energy resources has a comparative advantage in electricity production
 relative to many other sections of the US.  It would be useful to consider
 scenarios in which the ORBES region expanded its electricity production
 faster  than the US and thus become a significantly more important exporter
 of energy.

   (b) The process of scenario development for ORBES take no particular
 cognizance of the economic and institutional conditions under which the electric
 utility  industry operates.  In particular, the service area concept was ignored,
 as was the possibility that considering service area boundaries certain sub-
 regions of the ORBES region could feasibly remain net importers of energy in
 the future. An economic analysis based on the reassignment of generating
 capacity among ORBES counties based on the service area concept and projected
 growth in energy use within service areas would result in much more realistic
 results and projections.

   (c) The mini-assessment mode of analysis necessarily relies on the use of
 secondary data, and thus on the interpolation  of national and regional trends
 down to the sub-regional and county levels.  A series of case studies, analyzing
the economic impacts of particular projected developments in particular counties
and utilizing data reflecting specific local socio economic conditions and circum-
 stances would be of great value.  As indicated above, the. experiences associated
with energy developments in the west are unlikely to be repeated, in the same
form and to the same degree, in the ORBES region.
   (d) Our preliminary study of capital requirements for developments along
the lines of scenarios I and II cast doubts on the economic feasibility of these
scenarios, as opposed to their desirability. In-depth study of capital requirements
and capital formation is seen as essential.
                              II-B-5-21

-------
                               REFERENCES
 1.    U.S. Department of the Interior, Bureau of Mines,  United States
      Energy  Through the Year 2000 (Revised)  . Washington, B.C.
      December,  1975.
2.    Development of Plausible Future Regional Technology Configuration
      Ohio River Basin Energy Study Task I Report (October 18,  1976),p. 1
      glh 36. .
3.    A Time  to Choose America's Energy Future, Final Report, Ford
      Foundation, Energy Policy Project, Cambridge, Mass;: Ballinger,
      1974.  Chapter 3 and Appendices A,B, andF.
4.    Federal Reserve Bulletins.
5.    Tennessee Valley Authority,  Final Environmental Statement,.
      Hartsville Nuclear Plant, 1975.  See also John S. Gilmore and
      Mary K. Duff.  The Sweetwater County  Boom.  Prepared for the
      Rocky Mountain Energy Company, University of Denver Research
      Institue, Denver, Colorado, 1974.
6.    D. J. Bjornstad. "How Nuclear Siting Affects Local Communities."
      Survey of Business.  May/June, 1976:  pp.  7-10.
7.    U. S. Department of Commerce, Bureau of the Census.
      Census Update/1975. Washington, D.C., 1975.
8.    Alice W. Shurcliff.  "The Local Economic Impact of Nuclear Power."
      Technology Review.  January, 1977. pp. 40-47.
9.     NUS Corporation.  A Regional Siting Survey for Thermal Power
      Plant Sites in the State of Ohio.  Prepared for the Columbus and
      Southern Ohio Electric Company.  Rockville, Maryland, 1973.
                             II-B-5-22

-------
 10.   A..W. Shurcliff. op.  cit.
 11.   Gene R. Summers. Large  Industry in a Rural Area:  Demographic,
      Economic, and  Social Impacts.  Working Paper RID 73-19.  Center
      of Applied Sociology.  University of Wisconsin. Madison, Wisconsin
      1973.  38pp.
12.   Frank Clementa and G.R. Summers.  Rural Industrial Development
      and Commuting Patterns.  Working Paper RID 73-15.  Center of
      Applied Sociology.  University of Wisconsin.  Madison, Wisconsin,
      1973.  12pp.
13.   George Brinkman. "Effect of Industrializing Small Communities. "
      J.ournal of the Community Development Society  4(1): 1973.
14.   Personal Communication.  Kentucky Utilities Co. February,  1977.


15.   G. Brinkman.  op.  cit.


16.   Shurcliff.  op ^ cit,  Clemente and Summers.  op .  cit.


17.   Summers,  op. cit.
18.    Department of Housing and Urban Development, Office of Community
      Planning and Development.  Rapid Growth from Energy  Projects.
      Washington, D.C., 1976
19.    ibid.
                             II-B-5-23

-------
5.6.  INTRODUCTION TO IMPACTS ON HOUSING

5.6.1.  HOUSING MARKETS

        A housing market is the complete set of transactions between
suppliers of housing and households.1  These transactions influence
how much housing goes on the market (either through new construction
or a previously occupied unit) and its price, location, quality,  size
and tenure.  The housing market consists of many actors:  consumers,
direct suppliers and financiers.  These actors generally operate  as in-
dividuals rather than as groups.  Most consumers are individual households;
there are very few large.suppliers of housing; and financing is similarly
dispersed among small organizations.

        There are several  imperfections in housing markets which  can
result in an inefficient balance of supply and demand:   e.g., supply
lags, racial discrimination, and non-substitutable units.  Additionally,
the relative permanence of housing means that the bulk  of the stock in
any year v/as not recently built.  Given the governmental restrictions
(zoning and building codes) on the supply of new housing and some charac-
teristics of the product,  new units are generally not affordable  to most
of the households potentially on the market.  Most families obtain used
units which have depreciated in quality and value relative to the new
stock.

       ;Change in demand for.housing is a function of changes in  popu-
lation, incomes, and supply.  The important change in population  is the
increase or decrease in the number of families and unrelated individuals.
The separate effects of population change (independent  of changes in in-
come and supply) are hard  to determine.  In most housing analysis, the
focus is on the increase in "households," defined as one or more  persons
occupying a housing unit not in group quarters.   This definition  confuses
the effects of population  change on demand with  income  and supply changes.
Even with this difficulty, it is a frequently used measure because of
problems in projecting changes in the number of families and unrelated
individuals.  Projections  of average household size are thus used as sur-
rogates for projections of average family size.

        Changes in income  influence the amount of quality of housing
consumed.  Housing consumption is also influenced by family size  and age.
Income is perhaps the most important demand side variable from a  policy
perspective since it can be changed over short periods.   Consequently, it
is important to examine the relationship between income and housing con-
sumption.  The important measure is the coefficient of  income elasticity
of demand:   the change in  the amount of housing  purchased (AQ/Q)  resulting
from a given percentage change in the consumer's income.^
        1 The unit that demands housing is  normally called  a  household.
        2 Income in regard to housing market analysis  is  measured  by
normal "real" income rather than income in  any particular year or  month.
                             II-B-5-24

-------
Thus the income elasticity of demand is equal to  -    - .   Important dif-
ferences are expected between groups of housing consumers based on tenure,
age, race and size.  The best evidence is that the "income elasticity of
rental housing in the United States is probably in the range of 0.8 to
1.0. "(20)  For owner-occupants, the evidence supports a slightly higher
elasticity than for renters (in the range of 1.1 to 1.5). (21)  Recent
evidence has indicated that the income elasticity for renters may be
slightly lower than indicated above, approximately 0.5. (22)  This is par-
ticularly the case in nonmetropolitan urban markets. (23)  The best in-
terpretation of the low elasticities in these markets is that "stock con-
straints are primarily responsible for the very weak housing expenditure
responses to income changes in smaller cities. "(24)

        As indicated above, changes in supply also influence demand.  This
is expected to be an important problem in nonmetropolitan areas (cities
not  in standard metropolitan statistical areas and with populations greater
than 2,500).  In these communities there are very few builders, a small
labor pool in the construction trades and a lack of suppliers of building
materials.  Consequently, the supply of housing is not very responsive to
changes in demand.  For this reason, housing shortages are expected in
nonmetropolitan areas even if financing is readily available.

        In short term analysis, the most important housing market variable
is the availability of financing.  The money market for housing is cyclically
sensitive:  When the general economy builds to a peak, housing becomes less
.and  less -competitive in the money market (and vice v.ersa).  Thus, the short
term supply of financing will  have a very important effect on the housing
market.  This is, however, expected to be less important in the long run
where cyclical effects are not as important to the analysis.

5.6.2.  HOUSING AND GROWTH DUE TO ENERGY DEVELOPMENT

        The direct effects of the location of power plants on housing
markets are:  1)  population growth; 2) growth in real income; and
3) changes in land use.  Indirect effects will be the result of investment
and employment changes.

        There are two aspects  of population change related to power plant
location.  The labor force required in constructing and operating the
plant may have to come from outside the immediate area.  The temporary or
permanent housing demanded by these workers will  directly affect the local
market.  The important task is to accurately estimate the in-migration of
workers resulting from the power plant.  The second component of population
change is from general growth in the area that results from the location
of the power plant.  This would be the result of location decisions of
other industries who would be attracted to the area as a result of the
power plant.  These industries will in turn attract workers and permanent
residents to the community.3
        3 A decision not to include the indirect growth effects of power
plant siting was made by the ORBES Management Team.   For evaluating housing
markets, this may be unfortunate.   The indirect impact of power plant siting
on growth may be a more important long run impact on housing than the housing
needs of construction workers.


                             II-B-5-25

-------
        There are also important population changes that cannot be attri-
buted to power plant siting.  The local  community will  have a population
dynamic of its own.  Its residents will  age, give birth, die, and move to
other communities.  These are the changes that are reflected in population
projections.  For the purpose of this analysis, the projected population will
be considered the normal growth.  In-migration relating to construction and
operating employment in a particular power plant will  be estimated independent
of the population projection.  In other words, net change will  be considered
the sum of normal growth and growth and growth that is directly related to
the power plant.

        Immediate indirect effects of power plant siting include changes
in investment and employment patterns.   The capital requirements of power
plants will not be met locally, but this will  have a local impact.  Interest
rates will likely be affected since many local firms will need financing to
expand operations.  Investment flow will likely shift away from residential
construction to plants in nonmetropolitan areas.

        Additionally, residential construction workers are likely to have
skills that are in demand in constructing power plants or associated con-
struction.  The farther away from a metropolitan area that the plant is
built, the higher wages will have to be to attract the required skilled
work force to the plant.  Power plant wages for skilled and unskilled con-
struction workers are expected to be higher than those available in resi-
dential  construction.  Local construction workers able to shift to power
plant construction are expected to do so.

        Wages in residential construction may  not be forthcoming until
after the construction of the power plant.   A  similar effect on entre-
preneurial talent is expected.   Consequently,  local residential construction
may lag  well behind even the normal  growth in  demand due to nonmigration
population changes and increasing incomes.   The increased incomes due to
the power plant are not expected to overcome the supply problems in the
housing  market.

5.7.  ENERGY PLANTS AND GROWTH IN THE ORBES REGION.

        So far the discussion has been on the  types of housing market
impacts  that are expected to occur.   No attempt has been made to localize
these or estimate magnitudes.  These issues are obviously more difficult
and the  methodology and data are insufficient  to the task of precise impact
analysis.   The remainder of this chapter will  deal with what little data
are available.  Methodological  issues of an integrated impact assessment are
currently not resolved.   From the point of view of housing analysis, pro-
jections of income, employment, .and population impacts  are essential.   Without
reliable projections in these areas, housing analysis  must remain limited.

5.7.1.  HOUSING IN THE IMPACTED COUNTIES

        A statistical profile of housing in the impacted counties (i.e.
those counties having power plant sitings in any of the growth  scenarios)
is given in Table II-B-5-7.   The data are from the 1970 Census  and are
                            II-B-5-26

-------
                                                                                                                     TABLE   II-B-5-7
                                                                                                                  ILLINOIS  COUNTIES
  -z-.i'. :::.;IJ2 rui'r; '.>.-.:i.         ;.s;7     «,S32       5.M9     5,733       3.163     7.907       2,3)0    1C.856        5.546       5,533      12.426     4.353      5.572     6,849       2.592      35,3.'     2.340       2.070     4.707       6.157    3.50'..231
  -'.-.(-. ^--'.r-'.-t.. V.-.!:s.              73.3      ;;.<        77.1      71.1        =0.5      74.S        72.=      6J.3         75.5        75.9        5?.3      71.4       73.5     79.8        63.9        £7.2      71.5        53.0      77.1        71.2        53.4
'•--.'.". i' t\\ ::;.:'.•;  '.-.:*.-.

Y!:s-:c/ ~.&-je:  rcr Sale.
'.*:»-;/ =.4:c:  Tcr .:.j?.t.
19.2      S.i        15.3      IS.o        23.1      12.4       12.S       4.4
 5.7      C.I         4.5       i.3         5.4       5.1        5.S   '   4.2
  .0136      .0152       .C22i     .0177       .0131     .0173       .0136     .0117
  .C«24      .0459       .0705     .OS03       .1451     .0757       .0524     .05:1

 2.2      2.4         3.3       2.3         2.3       2.3        2.3       2.4

 3.0      2.e         2.6       2.S         2.4       2.7     :   3.4       3.0
                                                                                                                                                                                17.0        :£.s
                                                                                                                                                                                                                 :s.7
                                                                                                                              1.3         6.7         6.:       4.5        5.0      7.3        11.7        11.7       3.3         £.7       5.1         £.5          7.e
                                                                                                                               .C234       .0204       .0:23     .C1S6      .0155     .0065       .0279       .C122    ..0107       .C4;S     .0173       .02E5        .C.Z
                                                                                                                               .0-:~S       .1067       .0453     .0460      .0251     .0711       .1123       .0332     .0515       .:'."     .0753       .::-'.3        .C-5
                                                                                                                              2.3
                                                                                                                                                                        2.5      2.4         2.3         2.S       2.3
                                                                                                                                                                                                                             2.3       2.4
                                                                                                                                         J.7        2.3       3.0        3.1      2.4         2.2         2.5       2.«         2.9       2.<
                                                                                                                                                                                                                                                  2.3
                                                                                                                                                                                                                                                  2.6
                                                                                                                                                                                                                                                               2.3
   "i::si V«lv» - CsT-tr C-:c.            S9.400    59. 3*0      J9.cC3    S8.4:o     $6.500    S9.050      S9.400   S11.6CO       S11.60C     33,705     $13.733   S12.700    $11,300    $9,500      $5,3CO      $12,700   $10,300      $?.!"   ilD.lOO      53.C05      1:3.530
   :..-;••• 3,;lt i»C5 - sire!-. 1370
          5.25
   13"5  rzs. as ?e.-te:.t o* S:aU.
   iy.^  ":?. ^s ?«rce.-.; cf S:atc.
   1375-2CX Sro-iri ^a^.
                                         $73       S£3        £73       172
                                          3.3       3.1        6.3       2.6
                                                                                567
                                                                                 3.D
                                                  '4.3       63.6      75.3       67.4      77.5        64.9      £7.4
                                      14        74         116       33
                                        .;=      .13         .14      .15
                                        .C4      .U         .12      .11
                                          40       116          £4      157
                                            .CS       .20         .07       .33
                                            .07       .17         .07       .25
                                                                .cs:e    -.;oi5       .0051   .  .0034       .0032     .0020
                                                                                      S£5        $76        5105       $35        $53      $77
                                                                                      4.7        4.3         4.4      4.2        3.5'      6.8
$r,2
 5.2
                                                                                                         $72        $75         S124
                                                                                                         • 4.3        5.4          2.C
 47.3       53.0        65.1      67.4      72.5      63.2        67.3        33.3      75.4       77.0      K.S

1.33        S3         226        83        33       134          67        -.300        39         31        92
   .17        .15        .37       .12        .15      .13         .K       2.5        .07        .05       .13
   .1C        .14        .33       .10        .17      .13         .06       2.4        .05        .04       .11
   .0076      .0043      .00=3     .0019      .0126    .0015      -.0011       .0054     .0033       .0034     .0033
                                SI        77.315
                                  .'.5
                                  .12
 i
CO
en
 i

-------
                                                     TABLE II-B-5-7 (continued)
                                                         KENTUCKY COUNTIES
';-.;• -.--.;'.il :-;.!'.->; Units. . 2,5:5 2.333 4.517 3.M3 1,513 1,310 9. £66 11.5CO 3
-f.i-. :•.-.?.--:-.;. •.".;:-,. S3.: ££.5 7£.l 75.0 7i.2 £4.7 77.7 65.4.
!'.:--";'•:-.-« ;' ill =::;."fi;. . 17.1 36. S 35.7 £0.2 20.7 36.4 16.6 12.1
'•'.--. i.C-1 v- rcrj'^^r.s'jcr fiocs. 5.2 9.7 12.1 11.8 6.8 12.0 11.3 10.4
.::;-•../ =j:-: f;r Sale. .0151 .0127 .0073 .CX3 .0052 .0047 .0134 .0134
.::i-:/ ••.:••. Fir -.£?.:. .1300 .1133 .1013 .0317 .0679 .0474 .0591 .0534
.-; . 2.4 2.3 2.5 2.7 2.3 2 . 3.1 2.7
".'•• . 	 " 	 ' ' 2.7 3.0 3.2 2.2 2.6 3.4 3.2 2.8
•'•:'- Yj:.e - j.-.tr Ccc. S7.2CO SE,£;0 58,500 55.500 57,200 SIC. COO $11.300 S7.0CO 57
'•'.-. •: -c;11- *-»:/:rj-;itrs 9.5 6.3 11.0 7.2 £.8 6.2 5.2 6.7.
!>- •.'.'. 32.2 S1.4 51.1 48.3 £2.0 -67.4 32.6 45.7
=..:l " 92 27 177 51 £1 23 279 304
I?" =::. «-, ?..-;j.-.'. or £:;-.. . .25 .22 .45 .10 .15 .13 1.01 1.11
:::• 7:3. j; ?er:e.-.t of S:a:e .23 .22 .43 .31 .13 .13 .06. 1.16
•.3-.5-:XO Gr»t>. ?au .OOC8 .C110 .0035 .0120 .0035 .0102 .OC44 .0122
I — I
F— 1
1
CO
512 2,542 6.776 5.583 3,055 5.335
67. f, £0.5 54.1 55.1 78.0 48.6'
50.7 31.1 14.6 26.1 24.5 13.3
13.0 9.0 . 5.7 9.0 6.3 12.5
.0109 .0337 .0142 .0104 .0075 .0152
.1055 .0702 .0331 .0407 .1433 .0525
2.S 2.4 2.6 2.4 2.4 3.2
3.0 2.9 ' 2.6 2.9 3.3 3.2
000 55.200 $11.503 512,730 53,300 510.700
SS5 S51 371 S73 S55 5107
4.2 10.9 7.7 2.7 9.2 • 3.2
45.2 41.3 27.5 65.3 50.3 22.6
61 100 234 90 79 142
.35 .23 .63 , .51 .2S .55
.35 .21 .61 .41 .27 .'.-,
.0074 .0063 .0033 .0016 .0101 .0319

2.S54 3.453 5.413 2.710 l.{=5 4.455 4.531 553.
63.6 73.9 61.5 65.7 72.5 71.2 75.2 £5.3
44.5 32.6 22.3 23.5 31.8 21.5 2S.S 23.3
8.5 8.6 10.0 12.0 9.1 11.5 10. < 13.1
.01E4 .0253 .0080 .0141 .C"3 .C1C3 .0153 .111!
.1427 .124, .c;59 .C774 .CC53 .03" .1255 .:=K
2.3 2.5 2.£ 2.6 2.7 2.5 2.3 2.S
2.7 3.1 2.3 2.3 2.7 2.9 2.5 2.5
S3.803 57.700 514. 4DO S5.!C3 £12.1:3 57. CIO 55.6:: Sir.ClO
353 S'"3 S83 SCS 5. 3 S;3 Sst S-j
4.6 6.4 3.4 3.7 13.1 7.2 4.£
62.7 33.7 £5.0 40.5 £2.3 57.9 65.3 42. £
01 32 174 114 46 1C* 90 2353
.22 .32 .59 .25 .15 .43 .43
.17 .23 .53 .21 .15 .45 .4:
-.COM3 .0064 .0105 .0025 .",n .0303 .0133 -CltS

CO

-------


                        CA !
                                   ^  I      I  "2
                                  H  «
                                                          
                                                                                                                                        \A
                                                                                                                                        O
                                                                                                                                               -
                                                                                                                                               o
                                                                                                                      UJ
                                           f \67tlO
                                                                                                                                              250/3 22~"
                                                                            S.i
                     s.
        -r A
          \\         ,

           c.v-M\?:'c6. rAC..
-01 or r»>cro"rt''Sfrrvi>56£.»•
   ^ Vfi-1.ue. - Ctw^v Occ,
                                             ,0/051,
                    1,4
                                                                    73-2.
                                                                   3,6-
                                                                                                 ,0/03
                                                                                                 '
             C,,0 •  7.
                                                          3, -
                                                                 3,3
                                              3.3
£     1^    ^    ^    --     ii    ^
Il4o o  ! // /o o j // 7o o 177/0 0 /Wo o f?6/ o o I /i
      1     *     i      '      !     '
                ,            __p
                                         7.7
                                  n
                                   frg.f
 •/l.1?Tlp. ^I^ranr-e-rS^-c.  !


^.000 Tbc, ^?efOHvho^ ^>^Jc|
                                             ,o.3 .5?.r.

1375 /^/  1375 ;357 cc<;

'      .     !      '   '	

i ,V/j  JL/   ,^!,77- /.C3


-------
                                 i   -
                                    ^>  \
    V" c-f  Att  'A>u«.u- IL-AI \-s
     '     .T.VC c/v- A
 aAVT-(rr rri'i CcapiJ.il lUv;7^
 i-OI or r*.;rc~ri.''
                                   .015+

          TC,'.    o\-   y
..w;Cr
-.000   ip
,V .:&.$
                                                3,0
                                               1>
                                               i/77301
!  f75
•I'lS^lo-  AS. Itrtt^'ro-r SWG  I   .
                                                  13
                                             11
                    4/5-
                                                                                                                                                                 T'

-------
                               V
                                     {2
                                         r
VI

o

CD



A
                                                              rs

                                                             \£
                                                                    o
                                                                    v\
                                                                    CO.
                                                                         Q-L
                                                                                      O
                                          UJ




                                          cv



                                          "^J
                                                                                                      v-
                                                                                                                     O

                                                                                                                     •i
                                                                                                                 Cti
                                                                                                                                                           1   ,1.
                                                                                                                                                              Lu
                                                                                                                              Ul
                                                                                                                                            u;
                                      .3 I
{-



I
                                           ?,; 77.7
 •r<£ ~>.\ Ccapitd UV.-TC

0!  Or r^DTC t'
                               3,0
   ^D,
                             J.
                 -if OCC.  .   /?2CO
                                                /0.5"
                                                     7.S-
                 6-6
                                                                 t
                                                                  . OlfO
                                                                                                          -TS;/
                                    7-5"
13,1 I  7/3- //-o   7-f i  ?.7j  /
-------

                                  j


                                  T

                                  ^i

                                  u

       61



       3
              .
     V OoJne.r-C«X-  Um'U.
?0,0
    V" C-r  AU
  1-01  Or
     cn
       &ACA  I'Ov
.1001
                                 Z.3
                                             7.7
                        7.?
.05/5 //0/3
                                       1
                                       '*' I
                                                        .0/5
                                                            2.
.083?

Mv.-;er
                         1^70
                                                       3yV
              l>rfc/-,'rof
             .fcf
                                                                               !      I
                                                          I      i
                                                          i      i

-------
reported for impacted counties and states.  The following descriptive
summary is based on the data presented in Table II-B-5-7.

5.7.1.1.  KENTUCKY

        For Kentucky, the impacted counties are in thinly populated, non-
metropolitan areas (except for Henderson, which is part of the Evansville,
Indiana SMSA).  Most of the counties have a higher than average percentage
of owner occupied units, which again reflects their more rural character.

        The condition of the standing stock is generally worse than in
other areas of the state.  The houses are older and more of them are
lacking some or all plumbing facilities.   Rising incomes due to power
plant siting could result in rising expectations about the quality of
housing desired.  This could be a potential source of trouble since the
nonmetropolitan market is expected to be sluggish in responding to demand.

        Reflective of the poorer quality housing consumed, housing value
and rent are generally lower than in other areas of the state.  Meade
County's very high rents possibly are due to the location of Fort Knox.
Otherwise, higher rents and values are primarily due to the county's
closeness to an urbanized area.

        Mobile homes are generally a larger part of the stock.  Mobile
homes are becoming an increasingly important sector of the market in non-
me'tropoTitan areas. ^  They constitute'the largest supply of moderately
priced homes for owner occupants with lower than average incomes.  The
growth in this segment of the market is particularly sensitive to financing
and has been slower to come out of the recent recession than other housing
construction.   However, it is reasonable to assume that the mobile home,
anchored to a slab or tied down, will be an increasingly important segment
of the market in these areas.

        In most of these counties, less than thirty new units are built on
a yearly average.   This does however appear to be a sufficient supply under
normal growth rates.   Most of these communities have fewer than thirty new
families to house a year.  Vacancies in owned units reflect a fairly tight
market.   In contrast, rental units have higher than normal vacancies and
may be able to absorb some growth without going below frictional  vacancy
rates (the amount required to allow normal mobility among units and to
avoid excessive price inflation).  However, given the small size of the
housing markets involved (all but Henderson have less than 10,000 occupied
units as of 1970), short term supply lags can be expected even under slow
growth conditions.
        4 Mobile homes were 33% of the new construction in nonmetropolitan
areas between April, 1970 and 1974.   In Metropolitan areas the figure was
17%.  See U.S. Bureau of the Census,  Current Housing Report's  H-150-74,
Annual Housing Survey:  1974, Part A, U.S.G.P.O.,  1976.
                             II-B-5-31

-------
5.7.1.  INDIANA  '

        Indiana counties potentially impacted by power plant siting
generally have larger housing markets (the average occupied housing
units over all counties impacted in Indiana is twice that in Kentucky,
8557 compared with 4251).  More of the Indiana counties are in, or
contiguous, to metropolitan areas.   The range of housing market size
is much greater in Indiana:  2130 to 32369.  The latter is for Tippecanoe
Coutny, which composes the Lafayette-West, Lafayette SMSA.  Most of the
counties are, however, nonmetropolitan, which is reflected by several
characteristics of their housing markets.

        Excepting Tippecanoe and Clark,5 the impacted Indiana counties
have higher owner occupancy rates,  more units lacking some or all plumbing
facilities, higher percentages of older structures,6 lower values and
rents, and more mobile homes than the state as a whole.   All of these
characteristics reflect nonmetropolitan markets and are less prevalent
when a county is contiguous to an SMSA.

        Normal growth, as reflected in the population projections, has
all but four of the counties decreasing their share of state population
or maintaining a relatively stable  share (0 to .05 percentage points
shift).  Only Clark and Warrick (in the Louisville and Evansville SMSA's,
respectively) have important increases in  their shares of state population.
Counties which are projected to grow significantly faster than the state
'a"r'e'CTaYk,' Harrison5 Lawrence, Ohio, Poscy, and Warrick.  Excepting Law-
rence, all are in or continguous to large  metropolitan areas.   These
communities will likely benefit f*-om the flow of mortgage funds from the
inner portion of metropolitan areas to their outer reaches.   Labor, sup-
plies, and residential development  talent  should be readily available in
these metropolitan areas and no serious lags between normal  increases in
demand and supply responses are expected.   In the'slower groing areas,
supply lags may be a problem due to the lack of access to metropolitan
housing markets.

5.7.1.3.  ILLINOIS

        The size of the markets in  Illinois (again, considering impacted
counties only) are generally comparable to those in Kentucky,  with one
notable exception.  St.  Clair County, in the Saint Louis SMSA, had 86,347
occupied housing units as of 1970.   This is a large metropolitan market
and should be analyzed separately in any detailed analysis.   When St.  Clair
is excluded from the calculation of average size, the mean occupied housing
unit falls from 9447 to 5399 for the impacted counties.

        All  of the Illinois counties impacted have higher percentages  of
owner occupied units than for the state as a whole.  This is undoubtedly
due to the influence of the Chicago SMSA on the state total.   In any case,
the markets in these impacted counties are concentrated  in owner occupied
units and the vast majority of these are single family.   The housing in the

        5 Given their size and metropolitan character, these counties  will be
excluded from the summary description of housing markets which follows.
        6 Excepting Warrick which is part  of the Evansville, SMSA.


                             II-B-5-32

-------
counties not contiguous to SMSA's is generally older, in worse condition,
and costs less  (value or rent) than in the state as a whole.  The dif-
ferences would  be even more pronounced if the comparison was with metro-
politan counties.

        In terms of normal growth over the period 1975-2000, none of the
impacted counties are expected to grow at a rapid rate.   Only one county
(Mercer, which  is next to the Davenport-Rock Island-Moline SMSA) is ex-
pected to increase its share of the state's population..  Even then the
growth rate is  only 1.3 percent per annum.  The increase in housing units,
as measured by  the 1965-1970 rate of construction, is much less than 1
percent per annum.  Even though it is contiguous to a metropolitan area,
there could be  short term supply lags in this county.  These are not ex-
pected to last  in the long run however.

5.7.1.4.  OHIO

        The most marked difference between the housing markets in the
impacted counties of Ohio as compared with the other states is their
size and metropolitan character.   The average occupied housing units for
the impacted counties in Ohio is 53,108.  Only six of the twenty-three
counties have less than 10,000 occupied housing units.  Thirteen of the
twenty-three are within SMSA's.   An additional  five are contiguous to
SMSA's.

        The largely iratropolitan character of the-Ohio counties is
further reflected by their housing characteristics.  Unlike Kentucky,
Indiana, and Illinois, the Ohio counties are more heavily metropolitan
than nonmetropolitan.   The housing markets are less dominated by home
owners and more by renters.   Several  of the counties have lower percentages
of units lacking plumbing than in the state as  a whole.   On age of struc-
ture, value, rent, and mobile homes,  the Ohio counties are clearly divided
between the metropolitan and the nonmetropolitan.  The metropolitan counties
generally have characteristics of newer, more expensive  units and fewer
mobile homes.   The nonmetropolitan markets have very similar profiles as
those for impacted counties  in Kentucky, Indiana, and Illinois,   in Ohio,
the nonmetropolitan markets  are clustered in the southwestern section of
the state» along the Ohio River between  the Wheeling and the Huntington-
Ashland SMSA's.

        Unfortunately, county population projections for the year 2000 were
not available for Ohio.   Consequently no comments will be made on the pro-
jected growth in these areas.

5.7.1.5.  GROWTH DUE TO ENERGY DEVELOPMENT

        As mentioned above,  there are two aspects of population growth
that may be associated with  energy development.   The first is the direct
relocation of workers  constructing or operating the plant.   The second is
the influence the power plant has on  additional  industrial  development.
The latter is not examined in this chapter.
                             II-B-5-33

-------
        Several researchers have documented the boom and bust growth
cycle that came with recent power plant construction in the Hestern
States. (25)  Whether or not a similar cycle will be experienced in the
ORBES region depends on the location of the plants.  The closer to
metropolitan areas, in travel time, the less direct impact one would
expect on the housing markets of the site counties due to relocation o'f
workers.  Construction workers, expecting perhaps two or three years
employment at a site, will not relocate en masse if they can commute to
the site from their existing homes.  Nearly all the proposed plants in
the ORBES region are within two commuting hours of an SMSA.  Many are
within one commuting hour.  It is doubtful that given this characteristic
of the region that the severe boom-bust cycle in Western States would
occur in the ORBES region.

5.8.  HOUSING IMPACT OF GROWTH

        There are several other housing impacts that cai; be expected.
Certainly a number of construction workers living in the labor .shed of
the impacted counties will choose to work on power plant construction
given its generally high wages.  Using Kentucky as a test, labor sheds
were calculated based on the average commuting distance into the site
county (using 1970 Census data for commuting).  The commuting distance
was estimated on a centroid to ce.ntroid basis, using the site county
and other counties reporting commuting into the site county.  Location
quotients were then calculated as follows:

                     (CEi/CE)  *  (Ei/E)

Where the numerator is the ratio of construction employment in the county
(CEi) to total  employment in the county (Ei) and the denominator is the
ratio of construction employment in the state (CE) to total employment in
the state (E).   These quotients were calcuated for all  counties within
the labor shed of a site.county.  Surplus construction labor was calcu-
lated as the amount needing to be subtracted from the county's construction
and total  employment to result in a location quotient of one.

        Table II-B-5-8 presents the surplus construction labor in each  •
labor shed of a site county in Kentucky.  For comparison purposes, the
peak construction labor requirement of the BOM 80-20 and the BOM 50-50
scenarios are also presented. (26)

        The 'relatively large surpluse construction labor indicated in the
table is probably a result of laborers in these counties commuting into
metropolitan counties for work.  This surplus would largely be attracted
to work in the site county during the construction phase.^  Even if we
assume that all  the surplus construction labor in a labor shed is diverted
to the power plant, most site counties will  need substantially more
        7 This assumes that their construction skills are in demand there.
A more detailed analysis of construction employment is needed to support
this assumption.
                             II-B-5-34

-------
 construction workers during peak construction.  Much of this will come
 from  labor forces  in metropolitan areas.  The higher wages required to
 attract  these workers may also attract more construction workers out of
 the local labor force.  This would result in a deficit of construction
 workers  available  to work on residential building, thus aggravating already
 sluggish supply response in these nonmetropolitan markets.

         Construction workers attracted from greater distances into the area
 by the power plant may increase the demand for housing.  Quantifying this
 change in demand is difficult and the comments here should be interpreted
 as issues needing  further study.  Clearly an alternative hypothesis to
 a large  boom effect, at least in regard to workers moving into the area,.-
 is that  most of the workers will find it to their benefit to commute from
 existing residences rather than move.  The cost of temporary relocation
 of families will have to be compared to the cost of commuting.

         Given that the housing available in a metropolitan market will be
 generally of better quality and larger supply, the worker may decide to
 suffer the temporary inconvenience of commuting.^->  Also the ;cost of relo-
 cating will be higher than a normal relocation to a nonmetropolitan area.
 The supply lags mentioned earlier will likely push the price of housing up,
 as will  increased demand from new workers and increas'es in incomes locally.

         The direct housing impact will most likely be felt in the demand
 for rental units.  As was shown in Table II-B-5-7, most of these counties
.have  very low percentages of total units in the rental sector.  To meet
 increased demand, which is expected to be short-term rather than long-term,
 the most flexible alternative to the local owner or investor is to convert
 owned units into rental.   It will probably be profitable to minimize one's
 own use  of housing space and put as much on the rental market as possible.
 Renting  rooms is obviously a faster and cheaper response than converting
 single family units to apartments.  Local residents may decide to postpone
 their own increases in housing consumption until after the construction phase.
 This will allow them to maximize their income from housing without permanently
 limiting their own consumption, which can be increased at a later time.
 This  lagged effect on increased housing demand due to increases in local
 incomes  may offset a later "bust" phase of the boom-bust cycle when con
 struction ends.

         Short-term supply is mostly influenced by the money market.  Fi-
 nancing  is not likely to be available for short-term upswings in demand due
 to a  large construction labor force required.   Such housing would have to be
 capitalized over a very short period, influencing lenders and borrowers to
 be cautious.   The only new supply solution would be in mobile homes, which
 could be relocated when the construction when the construction force leaves.
 Financing works against this option except where the worker brings his own
 unit or wants to buy a highly mobile unit.  Highly transportable, smaller
 mobile homes or campers requiring little capital would be attractive.   Im-
 pacted communities should be encouraged to control such development through
 land use planning.
                             II-B-5-35

-------
        Two major housing problems result from the above patterns:
1) problems resulting from the influx of campers and trailers (a land
use problem, rather than a housing problem 2er_se); and 2} problems in
the rental market.  Shortages in the rental market are expected to be
acute.  This could severely affect local residents in the rental market,
particularly young, newly formed households, elderly households, and
low-income, large families.  The latter two groups may be seriously
affected by reducing their housing consumption as rents go up.  This is
particularly the case since their income will probably not stay up with
rent increases.

5.9.  DIFFERENCES BETWEEN SCENARIOS

        Unlike environmental issues, the mix of energy plants (percent
nuclear and percent coal) is of little importance to housing issues.
The important difference between the scenarios is sire:  the more con-
struction there is in nonmetropolitan areas, the more impacts there will
be.  These areas have been characterized as having slow supply responses
to increases in price.   This is expected to be even more pronounced in
situations with large but temporary changes in .demand.

        Problems resulting from acute shortages in supply may be offset
in the long run by increased incomes of local residents.  Property tax
reductions could also be expected.   The relative housing impacts of each
scenario are thus hard to determine.  A more detailed analysis of several
issues is necessary.   Employment, income, and residential location effects
must be estimated.  Housing impacts for each county under each scenario
would then have to be examined.   Characteristics of the population8 that
are particularly important to the housing market (e.g., the number and
percentage of elderly residents who are renters) would then be examined.
Supply side characteristics would also be examined to determine the ca-
pacity of the existing stock to absorb temporary residents.   Comparisons
between scenarios would then be possible.
        8 Population and household size projections by age,  sex and race
would be essential to the analysis.
                             II-B-5-36

-------
                            TABLE II-B-5-8

                    Surplus Construction in Labor Sheds

                    '  KENTUCKY SUMMARY TABLE
TOTAL (a)
surplus construction
Site County labor in labor shed
Ballard
Bracken
Breckinridge
Butler
Carlisle
Gallatin
Greenup
Henderson
Lewis
Livingston
Marshall
Mason
McLean
Meade
Owen
Russell
Scott
Trigg
Trimble
Union
Webster
568
26
657
370
687
232
661
232
281
246
.1 0 5.6
75
633
336
96
55
65
264
74
329
629
Peak construction
labor requirement
BOM 80-20 (b)
2000
2000
2000
1000 :
2000
3000
2000
3250 i
1250
2000
2000 :
0 ;
1000
2000 :
: looo •
2500
3250
2000
3000 :
2000
2000
Peak construction
labor requirement
BOM 50-50 (b)
0
3750
2000
0
2000
0
2500
3250
3750
2000
2000
3750
0
2000 '
1000
3750
3250
2000
3750
2000
2000
The labor shed is comprised off all counties within average commuting
distance of the site county  (measured as  straight  line  distance,  county
centroid to centroid).  Exportable construction  labor  (surplus)  was
calculated using the Location' Quotient Method.

Labor requirements were based upon the following estima.te.s:  2000 per
2000 MWe  coal fired plant, 2500 per 2000 MWe  nuclear  plant and
3250 per MCf capacity synthetic fuel  (gas)  plant.
                           •II-B-5-37

-------
       de  Leeuw,  Frank.   "The  Demand  for  Housing:   A  Review of
       Cross-Section Evidence,"  The Review  of  Economics  and Statistics,
       53,  (1971)  p.  1.


H -^  Ibid,  p.  10

       Carlimer,  Geoffrey.   "Income Elasticity of  Housing  Demand,"
       Review of  Economics  and Statistics,  55  (1973)  pp. 528-32.
                /
       Stegman, Michael  and Sumka, Howard.  Nonmetropolitan Urban
       Housing.   Cambridge,  Mass. Ballinger Pub, Co.,  1976.


       Ibid,  p.  77.


  '"    See  Gilmore,  John S., and Duff, Mary K.  Boom  Town  Growth Manage-
       ment.   Boulder, Colorado:  Westview  Press,  1975;  -Rapid Growth
       f-^QiTi. Energy Projects, Dept. of Hous.ing  and  Urban  Devleopment,
       Office of  Community  Planning and Development, "Washington, D.C.
       1976;  Ralph A. Luken, Economic and Social Impacts of Coal Develop-
       ment in the 1970's for  Mercer County, North Dakota,  Old West
       Regional Commission,  Washington, D.C.,  1974.


       Construction  labor requirements were based  on  Housing and Urban
       Development,  op.  cit. p.  3.

   Cf
       Several studies have  found that employment  location  does ~not
       greatly constrain residential location  in metropolitan areas.
       No  similar  research  has been found examining temporary employment
       outside of  the metropolitan area.  See  D. Lansing, John and  Earth,
       Nancy,  Residential Location and Urban Mobility:   A Multivariate
       Analysis,  Institute  for Social Research, Ann Arbor,  1964.
                              II-B-5-38

-------
           6.  SOCIOLOGICAL IMPACT OF PRESENT ENERGY SYSTEMS
           Analysis of what the real impact of power plant construction
and operation has been in the ORBES region would require a considerable
amount of fieldwork rather than reliance on secondary data as this pre-
liminary assessment requires.  However, a few generalizations may perhaps
be made about what kinds of conditions are necessary for communities to
be receptive to and be able to benefit from the siting of power production
facilities.  These generalizations have emerged from our coverage of the
literature and are cited in Volume III-D of this series (1).

           Generally speaking, communities that are urbanizing, econom-
ically diversified, near to large cities, and which have younger popula-
tions are most receptive to development, avoid the boomtown syndrome,
and generally have the revenues, tax base, and economic infrastructure
to finance the construction phase and benefit from the operating phase.
These kinds of communities, usually growing in population, also are
attractive to secondary development that often accompanies power plant
siting.  Also fortunate, during the construction phase, are rural areas
where a commuting labor force comes in daily with minimal  impact to
community structure.   However, these communities may 'fail  to  expand
sufficiently to capitalize on secondary industrial growth due to an
undiversified infrastructure.  Least able to contend with power plant
siting are poor, declining, rural, isolated communities that  have cities
just .large-enough (and distant enough from major cities) to have to house
the construction labor force for a number of years.  Here the classic
boomtown symdrome occurs with an influx of newcomers with heavy demands
on services, transportation networks, housing and recreational-facilities.
Poor and undiversified local economies usually do not have the  revenues to
deal with such strains, and there is usually a lag between the  ability to
generate revenues to meet the needs of the newcomers and when the fiscal
ability to pay for them is adequate.   These kinds of communities see a
decrease in quality of life for almost all community residents, new and
old alike, during the construction phase.  Moreover, there is,  after
this experience, often a cautious attitude on the part of residents
about further development and sometimes a failure to attract  any further
industrialization.   For some communities, revenues from the plant alone
are sufficient and no further growth is desired.

6.1.  RTC CRITIQUE

           The BOM high energy scenarios strike us, from a social
scientists'  point of view, as unlikely to be acceptable to the  region's
residents.  All  the major river valleys would become virtually  wall-to-
wall plants and if associated industrialization occurred,  it  seems likely
the quality of life for the previously rural residents would  be irrevers-
ibly diminshed due to increased urbanization and associated congestion,
pollution and changes in population density and land use.
                              II-B-6-1

-------
           Moreover, we remain unconvinced that demand for electricity would
escalate as radically as these scenarios dictate, given recent federal
energy policy formations and the circularity of utility company reasoning
concerning future demand for electricity.

           Current population migration trends further suggest continued
movement to Southern and Western states and possible stabilization of
populations in the ORBES region rather than the growth some population
projections offered in this project have forecasted.

           The lower growth Ford Tech-Fix scenarios posit a more reasonable
pace of power plant siting and some conservation measures on the part of
electricity consumers.  Impacts on the quality of life of the regi'on's
residents would be primarily local in character rather than river-basin
wide.

           Even the rate of growth in demand posited by these scenarios
may be too high or the social costs of meeting this demand seen as too
high by ORBES region residents.  Of particular concern is emerging informa-
tion on radioactivity hazards associated with coal-fired plants and unresolved
waste disposal problems associated with nuclear plants.  This latter con-
sideration will have an effect if and when the public becomes informed and
concerned.  On the other hand, future severe winters, changes in household
and industrial technologies, electricity shortages at critical times
could work to increase public demand for electricity and acceptance of
more power plants.

           We would like to work with a scenario that is closely aligned
with President Carter's energy policy and with population projections that
more nearly reflect what is likely to happen in countries.  For instance,
some counties are losing population now, though not many, and basic
changes in land use, the agricultural market, or extensive strip-mining
could accelerate or cause out-migration trends.  In short, we think a
more realistic scenario of power plant siting and accompanying pro-
jections for both population growth and demand for electricity would
be most helpful in trying to evaluate, peering dimly into the future,
what the likely effects of energy development on the ORBES region are to be.

6.2.  POLICY RECOMMENDATIONS

           As social scientists, our policy recommendations are based on
what we think siting a power plant will do to residents of the community
of impact.  While the information upon which these recommendations are
based is secondary, we are relying heavily on what..we have learned from
our literature search about the impacts of other plants built in other
parts of the country.   Generally speaking, a review of the impact
assessments we have drawn will show a variety of possible adaptations
to power plant development.

           We think plants should be sited first where there is minimal
disruption during the construction phase and where the community is
                             II-B-6-2

-------
receptive to the development.  Next, plants could be sited in rural
areas, even though there is some opposition, on land not being currently
used in an economically or socially beneficial way.  However, the work-
force for the construction of such plants should live within commuting
distance so as to disturb the community as little as possible.  Where a
high potential for conflict exists among community members about the
desirability of the plant or where a community is likely to undergo a
boomtown syndrome and is fiscally unprepared for it, plants should not
be sited.  Clearly, too, plants should not be sited near scenic or
historic areas for these are also recreational and tourist areas.
            V
           Our analysis of the demographic profiles of these counties
suggests that many of them are really quite rural, with older populations,
and still fairly isolated from the urban mainstream, contrary to what has
often been said about the ORBES region.  These counties, in particular,
seem unlikely counties for development for they lack the population and
economic structure to truly benefit from power plant siting.   As a
matter of fact, they are likely to bear all  the social  and economic
costs, while newcomers who come to operate the plant will  reap most of
the benefits.

           In general, we find that small  communities that may be subject
to booms or "boomlets" need to be prepared in advance for what is to
happen.   Community leaders need to know how to plan for both temporary
and permanent growth and need access to revenues to finance such
expansions until  the plant becomes a .beneficial f-inancial  force in the
community.  Long-range planning seems of the essence if a community is
to reap the full  benefits that a power plant and associated economic
growth can bring.

6.3.  PRELIMINARY ASSESSMENT OF SOCIAL IMPACTS

6.3.1.  INTRODUCTION AND OVERVIEW OF THE IMPACT ANALYSIS

           In order to assess the likely impact of planned and scenario-
proposed power plants on counties in the ORBES region,  we undertook a
rather extensive special study, Social  Aspects of Power Plant Siting,
which is Volume III-D of this report" (T).For this preliminary assess-
ment, we have elected to present only our conclusions as to what we
think is a plausible picture of possible impacts of power plants on
the various counties.   We have also tried to assess what the likely
reaction of the residents would be to power plant development.  The
basis for these predictions is found in the special study.  In order
to understand the framework in which these assessments  were made and
the language used in the assessment, we present below a brief outline
of the special  study and its contents.

           The special study operationalized a paradign for social-
ecological  analysis suggested by Erik Cohen in which he presented four
basic orientations which may be held by persons, groups, or institutions
                             II-B-6-3

-------
toward the environment  (2).  Within each orientation are sub-orientations
which give clearer meaning ot how the environment is perceived.  These
orientations are not mutually exclusive and may often be in conflict
concerning the environment, for instance, the desires of utility companies
to use land for a power plant may conflict with those of a farmer who
owns the land and finds it satisfying.  On the basis of demographic pro-
files we have prepared, we have tried to classify each county as to pre-
vailing orientation or mix of orientations, the latter being sources for
potential conflict concerning power plant siting.

           The environmental orientations are as follows:  the instrumental
orientation basically views the environment as a commodity to be exploited
and used.  The sub-orientations here are technical feasibility, that is,
knowledge of how the environment might be exploited and economic benefit,
that is, whether or not such an undertaking with the environment will be
profitable.  The territorial orientation has to do with who controls use
of the environment and what mechanisms are involved.  The sub-orientations
here are strategic dominance, that is, how land or other features of the
environment and what mechanisms are involved.   The sub-orientations here
are strategic dominance, that is. how land or other features of the en-
vironment may be acquired, or boundaries maintained, and political  orga-
nization, that is, how decisions come to be made concerning environmental
use.  The sentimental orientation refers to how people feel about their
environment.  Its sub-orientations are primordial  belonging and prestige.
These two are quite different, for the former refers essentially to emo-
tional attachment to place or some object in the environment and the lat-
ter to the class .or stratificational  component that is attached to the
environment and from which the user may gain or maintain prestige.   The
symbolic orientation refers to meanings that people have attached to the
env Fronment and sub-orientations are aesthetic and moral-religious.  The
aesthetic sub-orientation has to do with designating the environment,
either formally or informally, as aesthetically pleasing and valuable
just for that.  The moral-religious orientation confers both sacred and
secular meaning to the environment, usually through formal  designation
as a sanctified place.

           The operationalization of these concepts involved making re-
source, land use, and transportation maps of the ORBES region for the
instrumental orientation; mapping social, economic and demographic cha-
racteristics of the population for the sentimental orientation; and making
maps of public and private forests, parks, recreational areas and the
like for the symbolic orientation.   The territorial orientation was not
presently mappable, however it is the mechanism through which use of the
environment is eventually determined in conjunction with the relative power
of the other three orientations.

           In terms of "pure types",  counties  which have been classified
as having a predominantly instrumental  orientation are urban, prosperous
and have a fairly high degree of economic diversification.   Counties seen
as having a sentimental  orientation were generally broken down into two
types:  the primordial belongingness  sub-orientation was attributed to
rural  counties,  housing  usually poor, or relatively poor people who have
been very stable in residential  preference and how they have used the land.
                             II-B-6-4

-------
Many of these counties are isolated from urban centers, sparsely populated
and economically undiversified.  Suburban bedroom communities, and areas
with high housing values and incomes were generally classified as having
the prestige 'sub-orientation.  Counties classified as having a symbolic
orientation usually have a high substantial proportion of their land de-
voted to forests and other recreational useage, or having a large number
of historical sites or an important religious site.

           Some counties present a picture of mixed orientations, and these
may or may not be in conflict.   For instance, a mostly rural, but urbanizing
county, usually on the fringe of a large city, has an instrumental orien-
tation in part.  But depending on the rest of the county's residents, their
orientation may come in conflict with the instrumental orientation.  For
instance, a change in land useage from symbolic (for instance, recreational
sites) to a power plant site would generate considerable conflict as would
trying to site a plant on land for which the residents feel great attachment
to place (primordial belongingness).  Those residents with a prestige sub-
orientation might object to a power plant site in an urbanizing county if
it were to close to their suburbs, however, if it were sitecr in a more
rural part of the county, they may welcome its appearance.  We have tried
to make a plausible assessment of likely community reaction to power plant
siting in both the pure and mixed types of classified counties.  We have
not addressed possible interactional effects should much of this power
plant construction occur simultaneously in geographically proximate counties.

'6.3.1.1.  -KENTUCKY COUNTIES

           Trigg, Livingston, Marshall, Ballard and Carlisle Counties.

           Trigg, Livingtson, Marshall, Ballard and Carlisle Counties are
all located in the southwestern tip of Kentucky, the latter three west of
the Tennessee River and the Kentucky Lake.   These'five Kentucky Counties,
which are in close geographical proximity,  differ in their basic environ-
mental orientations in spite of relative homogeneity of the population
itself in terms of age structure, educational  attainment and income.   Trigg
and Marshall  may be classified as having an aesthetic (symbolic) sub-
orientation primarily because of a large recreational  aspect.   The possi-
bility for conflict between the economic benefit and the aesthetic sub-ori
entations exist in Marshall  because it is also the most urban, the most
affluent, more industrialized and has the most commuting workers.   In short,
it seems ripe for further industrial development such as a power plant site
and associated industrial  development such  as  a power plant site and asso-
ciated industrial development.   Furthermore, given the nature of the popu-
lation, it seems likely that Marshall  County would be open and receptive
to a power plant.  Because the prevailing orientation in Trigg County is
aesthetic, we would expect considerable opposition to power plant siting
there.   It is questionable whether the county has the resources to be
effective in opposition efforts.

           Livingston County, also, with its recent population growth and
high inter-county mobility might also be receptive .to power plant development,
however, given its rural  nature, it would also be innundated by the con-
struction phase boom and experience negative impacts because of the lack of
                             II-B-6-5

-------
 an effective economic infrastructure to deal with demands of migrant
 workers for services, housing and amenities.

           Ballard County with its aesthetic sub-orientation seems an
 unlikely candidate for power plant development because it also is an
 area of generally older people and contains the most valuable farm land
 and the highest incomes.  In short, it seems to be prospering at its own
 pace and the people there probably do have a fairly high sentimental
 attachment to place as well.  Generally, we would predict opposition to
 a plant in this county.   The only mitigating factors are the mixed origins
 of the population, with quite a few being from other states, however, we
 think many of the immigrants to be retirees, and probably unsympathetic
 to development.  However, should a plant be sited there, many workers could
 commute from nearby cities and mitigate any boomtown"n'mpacts:

           Carlisle, being isolated, rural farm in nature and with an
 older population of limited education and income, while possibly being
 receptive to development, would probably experience an intense boom-bust
 phase should a plant be located there with its attendant disamonities.

           Webster, Henderson, Union, and McLean Counties.

           Webster, Henderson, Union, and McLean Counties form a tight
 cluster along the Ohio River just southeast of the point where it con-
 verges with the Wabash River.  Butler County is separate and to the south-
 east of this cluster.  These five Kentucky 'counties slated  for possible
 power plant siting, are physically close to one another but differ greatly
 in what their basic environmental orientations would seem to b~.  Henderson
 and Union, the most affluent and urbanized counties, seem to have a pri-
 marily instrumental orientation, which would generally make them receptive
 to power plant development and likely, also, to have an economic infra-
 structure to deal  with population growth.   Henderson County would not be
 subject to a boom-bust syndrome because of the availability of a skilled
 labor force in nearby Evansville and in Henderson itself.  Union County,
 on the other hand, is likely to expedience a boom-bust cycle from the con-
 struction phase in one or two of its small cities, depending on where the
 plant is located.

           Webster and McLean Counties, and particularly the former, are
 likely to be highly sentimental  in their environmental  orientation and
 somewhat resistant to power plant development because of relative stability,
elderliness and rurality of the population.   It is doubtful  whether effective
opposition to a plant could be brought about should the residents be op-
posed to power plant development given the low educational  and occupational
 levels that characterize these counties.

           Butler County seems to have a primordial  belongingness environ-
mental sub-orientation.   It also seerns to have an aesthetic sub-orientation
considering the proportion of the land which is forested.  The high pro-
portion of the poor and  the lack of economic diversification suggest poor
adaptation to a boomtown . syndrome that would be likely in  this isolated
county.  We also would expect the population to be opposed  to development.
                             II-B-6-6

-------
            Boone, Carroll, Gallatin, Owen and Trimble Counties.

            Boone, Carroll, Gallatin, Owen and Trimble Counties form a
cluster of  counties in northern Kentucky between Cincinnati and Louisville,
all but Owen County border on the Ohio River.  The environmental orien-
tations of  these five Kentucky counties are very homogeneous with the
exception of Boone County.  The latter county, being part of a SMSA, has
a primarily instrumental orientation and, as such, should house a popu-
lation generally amenable to power plant development.  Moreover, .there
would be no  boomtown  syndrome due to the nearby availability of a skilled
labor force.

           Gallatin, Owen and Trimble Counties would seem to be charac-
terized by the primordial belonging sub-orientations given the rural and
recreational nature of land use and the stability and relative elderliness
of the population who are probably highly attached to their homes, land
and lifestyle.  Gallatin and Trimble would not experience booms because
of nearness to urban centers.  Owen County would be inundated by the con-
struction phase and unable to deal with the strains.

           Carroll County presents a less clear picture with its rural
population and one small city that serves as a trade center.  If a power
plant were to be sited in the county, Carrollton would become a likely
boomtown candidate.   The city might general! welcome such development
and given its relatively diversified but small economic infrastructure
not have too much difficulty coping with boomtown demands 'and successfully
attracting secondary industry after the power plant is built.  One might
surmise that the prestige sub-orientation would characterize C^rrollton
residents and the primordial  belongingness sub-orientation characterize
its rural population who might be opposed to power plant development.

           Bracken,  Lewis, Mason and Greenup Counties.

           Bracken,  Lewis, Mason and Greenup Counties in northwestern
Kentucky, while in geographical  proximity, present some variations in
environmental  orientations.   Bracken County is a poor rural  county with
a high proportion of people who are in the later stages of the family
cycle, with children gone and frequently one spouse dead.   There are a
few job opportunities  for those in their working years who are unable to
maintain a viable farming operation.   Most of the public assistance goes
to aid the elderly who are often widowed.   As a consequence, some migrate
out, some commute to nearby counties for work.

           Lewis County is very similar in demographic characteristics to
Bracken County except  for age structure.   Young families are more prevalent
in this county which also exhibits an unusually stable residential pattern,
even for this generally stable area.   The predominant orientations in these
counties is that of  primordial  belonging with some evidence of an aesthetic
sub-orientation in Lewis County, which may account in part for younger
people's presence there.
                             II-B-6-7

-------
           Mason County is the site of a regional trade and manufacturing
center.  Though sharing the general rural characteristics of the area and
the age structure of neighboring Bracken County, Mason County provides
economic options and supplements to farming for both its own residents
and those of adjacent counties.  Mason County presents a mixed environ-
mentla orientation picture.  It seems to be a thriving county with a
fairly diversified economy.  This would suggest receptivity to a boomtown
situation, given adequate planning.  Attitudes toward power plant develop-
ment probably vary because of the difference in age structure of the popu-
lation with Lewis County being more favorable.  A commuting work force
during the construction phase from nearby cities would generally mitigate
construction boom impacts.

           Greenup County is a transition area between the more rural
counties to the west, and urban areas to the north and east.  The population,
young, working class, and somewhat more mobile, are most likely oriented
to the urban areas for economic activities, living standards, the social
services and amenities, but they are also very likely to have their roots
4n the more rural society to the west.  Greenup County seems to ge in
the transition from a sentimental to an economic benefit sub-orientation
as it continues to urbanize.  The youthfulness of its population and its
bedroom community status would seem favorable indicators toward power
plant development, provided the plant were located in a rural area for the
prestige sub-orientation tends to be dominant in suburban areas where some
opposition may exist but is likely to be ineffective.  A commuting labor
force from nearby Ashland or Protsmouth would mean minimal  boomtown effects.

           Meade and Breckinridge Counties

           Meade and Breckinridge Counties lie on the Ohio  River across
the river from the Hoosier Forest in south-central Indiana.  The predominant
orientation of these counties is clearly sentimental  and symbolic given
the farming and recreational use of the land and its stable, older and
poor residents.  Meade County's previous experience with Fort Knox and
mobile populations suggests a somewhat easier possible adjustment to
power plant siting, however, neither of these counties has  the infrastruc-
ture to deal  with a construction boom.  The nearness to rather large cities
suggests the likelihood of a commuting labor force, by and  large, and in
that sense, minimum impact of that phase.  On the other hand, the nature
of the indigenous population, especially in Breckenridge County, suggests
there may be quite a bit of resistance to a power plant project.

           Russell  County

           Russell  County in south-central  Kentucky,  being  poor, very
isolated and rural  in nature, has the primordial  belongingness sub-orien-
tation.  It is also a seeming candidate for the classic boomtown syndrome
associated with power plant development in these kinds of communities.
However, its  lack of economic diversification and an infrastructure to deal
with boomtown demands means that the boom is likely to take place in nearby
Somerset,  in  another county, which does have more amenities and facilities
to offer the  labor force during the construction phase.   Residents then
                            II-B-6-8

-------
will have to deal only with day-to-day impacts.  We predict some conflict
between the instrumental orientation of the utility companies and those
residents whose land is seen as a possible site, for their attachment to
place is likely to be very strong as it is with many Eastern Kentucky
residents.

           Jefferson County

           The prevailing sub-orientation in Jefferson County in which
Louisville is located, given its urban, diversified nature is economic
benefit.  Since the plants planned are additions to an existing plant,
and there is a skilled labor force nearby, it is predicted that the
impact will be minimal unless opposition groups holding contrary environ-
mental orientations emerge.

           Scott County

           Scott County, located in Central Kentucky and slated for a coal
conversion plant, presents rather mixed environmental  orientations to us.
On the one hand, there is a large rural farm population with a predominantly
sentimental orientation.  On the other, the urban influences, which are
primarily instrumental in character and due both to a  fairly good-sized
town in the county and nearby Lexington, suggest the potential  for real
conflict concerning the location of a coal conversion  plant.   Being nearby
residents ourselves, we know the plant is a very controversial  issue.  We
predict, however, that the instrumental orientation will eventually pre-
dominate and the plant be constructed there.   Opposition will probably
come from environmentalists in nearby Fayette County and from farmers.
If it is well-organized, it might be able to delay the plant or possibly
stop its construction.

6.3.1.2.  ILLINOIS COUNTIES

           Fulton, Schuyler, Brown, Cass Scott, Greene and Jersey Counties.

           In Fulton, Schuyler, Brown, Cass,  Scott, Greene and Jersey Counties
along the Illinois River in northwestern Illinois a mix of environmental
orientations is present in the prime power plant sites.  Fulton and Jersey
Counties, bordering SMSA's have a primariy instrumental orientation while
the remaining counties may be generally classified as  sentimental, with
emphasis on the prestige sub-orientation.   Fulton, being a coal-producing
area, with an economically diversified infrastructure  and nearness to
Peoria, should be able to eas.ily absorb power plant development and the
residents favorably disposed toward development.  This should be generally
ture of Jersey County as well, however, their population might be even
more positive, given its youthful age structure.  Its  ability to benefit
is also slightly greater.

           The remaining rural counties, with the exception of Brown County
(discussed below), being economically fairly well off, quite rural and
isolated (except for Cass County) from urban centers and major transpor-
tation lines, have a primarily sentimental orientation toward the environment,
                             II-B-6-9

-------
and its older population suggests a strong primordial belongingness
sub-orientation.  Farming here, as in the rest of Illinois, is a strongly
instrumental activity as well with its business emphasis creating a mix
of orientations.  However, all these counties (except Cass possibly because
of its nearness to Springfield) would be candidates for a fairly severe
boomtown syndrome.  Given the nature of the population and their orien-
tations, it seems likely that power plant development would not be welcome
here and considerable opposition expected.  Moreover, the economic infra-
structure and the size of possible recipient communities suggest diffi-
culties in coping with boomtown strains.

           Brown County, the poorest of this group and with a higher pro-
portion of older people and a strong attachment to place, would also be
subject to what we just discussed above, however, its ability to cope
would be even less and fewer benefits from plant construction would ac-
crue to its residents for they are not of the proper age to benefit.

           Hamilton, White, Lawrence, Jasper and Clark Counties

           The Illinois counties of Hamilton, White, Lawrence, Jasper and
Clark .in southeastern Illinois, while homogenous with respect to predominant
environmental  orientation (sentimental), exhibit rather interesting variations
in the sub-orientations.  For instance, Jasper County, a stable prosperous
farming community with families in the child-bearing and rearing cycles
contrasts sharply with Hamilton County with its remarkably elderly popu-
lation and its poverty.   In the former, the prestige sub-orientation as
well  as an economic benefit/instrumental orientation is probably dominant;
in the latter, primordial belonging.

           All these counties, with the exception of Lawrence, are not boom-
town candidates.  There are cities in Indiana capable of providing most
of the work force, but those farthest away from cities may have "boomlets."
Lawrence County seems likely to experience a boom and seems not too well
equipped to provide needed services and amenities.   The larger nearby city
of Vincennes in Indiana seems to be the most likely recipient of the boom,
howeve,-, mitigating local effects.

           In terms of local acceptance of development, we would expect
Jasper and Hamilton Counties to be the least receptive, for the differing
reasons described above.  Clark, being the most urban would seem the most
accepting, though farming is a prosperous occupation there.  White County,
being close to Evansville, Indiana should experience little negative impact
from power plant construction and possibly welcome the infusion of money
to the county.

           Iroquois, Livingston', DeWitt, Grundy, LaSalle and Marshall
           Counties.

           The prevailing environmental orientation in these north-central
Illinois Counties (Iroquois, Livingston, DeWitt, Grundy, LaSalle and
Marshall) while primarily sentimental  (prestige seems more important than
primordial belongingness) have a strong instrumental  orientation as well,
                             II-B-6-10

-------
as reflected in the high economic returns afforded by the diversified
economic activity, especially in the more urban Grundy and LaSalle Counties.

           Fanning in this area is a highly instrumental activity which
yields not only economic benefit but prestige as well.  Acceptance of power
plant development would probably be highest in Grundy and LaSalle Counties.
With the exception of Iroquois County, no (or minimal) boomtown effects
would be expected given the nearness of skilled worker populations in the
urban areas and the economic diversification and affluence of the smaller
towns.  It is expected that in all, these counties, if prime expensive
farmland is required for power plant development, there will  be some op-
position from farmers.  Again, except for Iroquois County, these counties
would seem likely to be receptive to further development.

           St. Cla.ir, Washington and Perry Counties.

           The Illinois counties of St. Clair, Washington and Perry south-
west of St. Louis SMSA, while in geographical proximity vary widely in
likely environmental orientations though the latter two are quite similar.

           St. Clair, urban and urbanizing, may be characterized as instru-
mental in orientation with a heavy emphasis on the economic benefit sub-
orientation.   Being a coal-producing county, it is a  logical  site for
coal conversion and coal-fired power plants.  Moreover, unless the plants
are sited in the eastern part of the county, no boomtown syndrome would
be expected'given the proximity.of St. Louis.  -Even if a plant were sited
in the eastern part, Centralia would probably be the  focus of boom effects
and leave the county unaffected.  Given these conditions, we would expect
plant siting and development to be unproblematic and  welcomed by residents
of the county.

           Washington and Perry Counties represent the sentimental orien-
tation, though the primordial  belongingness sub-orientation would be much
stronger in Washington County, given the percentage of land in farms and
the elderliness of the population.   High incomes for  both counties indicate
the prestige sub-orientation to be present as well.  Perry County, a coal-
producing county 60 miles from 60 miles from St. Louis, seems a more re-
ceptive site than Washington County.   However, the aesthetic sub-orientation
is there as well, given the recreational and forested areas.   The movement
from farm to rural non-farm and the decline in land devoted to farming
suggest that land for plant siting would be readily available.  We sugest
that public opinion would be quite split about the desirability of a plant,
because the older population and rural population are generally not favorable
to such development.   The nearness to St.  Louis on the other hand, suggests
urban influences and receptivity to development.

           Washington County would seem to be non-receptive to power-plant
development since it is highly agrarian in nature and likely subject to
boomtown conditions for which it is unprepared to cope, since it lacks
economic diversification.   Its location on a major interstate might allow
for commuting workers mitigating some boom effects.  By and large, however,
the county's  characteristics would indicate negativity about that kind of
development.


                             II-B-6-11

-------
           Mercer, Henderson and Hancock Counties.

           Mercer, Henderson and Hancock Counties, river.counties in
Northwestern  Illinois, present a relatively homogeneous picture with re-
gard to their prevailing environmental orientation:  sentiment.  However,
among the sub-orientations, the ties are not simply primordial belonging
but have a very substantial prestige or stratificational component as well.
Given, too, the economic benefit derived from this prosperous farming area,
the instrumental sub-orientation is there too.

           Given these orientations, we think Mercer County is most re-
ceptive to power plant development, largely because there would be minimal
boomtown effects due to a nearby large city.  In the other counties, we
would expect some opposition from farmers should prime farmland be selected
as a power plant site.  The relatively numerous small  towns in Hancock
County and Mercer County suggest some economic diversification is present
though farming dominates land use.   In Hancock County, the boomtown po-
tential is there as is true for Henderson County.   The latter county, being
completely rural, is likely to be opposed to power plant development, given
its relative prosperity.  The same would be generally true of Hancock County
as well.

           Pulaski County

           Pulaski County, Illinois, in the heart of a rather large re-
creational ^area ~at the-southern tip of the state,  probably has a symbolic
orientation to the environment.  The elderliness of the population and
mixed-origin character suggests it is a retirement community and thus also
chatacterized by the primordial belongingness sub-orientation.  It is poor,
suggesting perhaps, as in Eastern Kentucky, there is natural beauty to
offset the relative poverty of the area.  Also, older people generally have
low incomes.  Given these prevailing orientations, we would expect con-
siderable opposition to a power plant site in this county.   Moreover, if
our assessment of aesthetic qualities is correct,  it seems likely that
outside environmentalists would join in an opposition movement increasing
its effectiveness.

6.3.1.3.  INDIANA COUNTIES
           Sullivan,
           Counties.
Greene, Knox, Daviess, Martin, Gibson, Pike and Dubois
           Sullivan, Greene, Knox, Daviess, Martin, Gibson, Pike and Dubois
Counties form a large block of counties slated for power plant development
in southwestern Indiana.  They vary greatly in terms of boomtown potential
and predominant environmental  orientations.

           The counties which  have already experienced considerable power
plant development (Gibson, Pike and Sullivan)  are all  within commuting
distance fo a construction workforce and are likely to be amenable to fur-
ther development.   Two of these counties also  produce coal  and two are part
of SMSAs.  The orientation of  Gibson and Sullivan Counties  would appear to
                              II-B-6-12

-------
be primarily instrumental.  Pike County too has this predominant
orientation as well as an aesthetic sub-orientation, given the high
proportion of the land devoted to forest.

           Candidates for the boomtown syndrome include DuBois, Daviess,
and Greene Counties, and possibly Martin County as well.  DuBois and
Martin are the most economically diversified of these four and better
able to cope with boomtown demands.  However, DuBois County, being
affluent and self-contained probably has a high sentimental environmental
orientation as well as the aesthetic sub-orientation.  It is likely to
be highly resistant to power plant development.  This is generally true
for Martin County as well; however, Martin County has had previous
experience with large-scale development and might be slightly more
receptive, but still basically a likely locus of opposition.

           Daviess and Greene Counties have different environmental
orientations, with Daviess being characterized as having the primordial
sub-orientation and Greene having a mix of the instrumental and the
aesthetic and prestige sub-orientations.  Greene is a coal county,
near a major university, and has an older population so we would expect
the conflict of orientations to manifest itself in this county.  The
instrumental orientation and attendant power plant development will
probably prevail.   In Daviess County, residents would probably be
opposed to development and ineffective in their protests.  The county
would probably be inept in coping with boomtown strains as well.

           Knox County has a mix of orientations present, primarily
prestige and instrumental, and lies within commuting distance from
Evansville and hence is not a boumtown candidate.   Unless college
students organize opposition groups, we would expect little resistance
to power plant development and that the county would prosper from a
plant.

           Clark,  Jefferson, Switzerland, Ohio and Dearborn Counties

           None of the southeastern Indiana counties along the Ohio river.
Clark,  Jefferson,  Switzerland, Ohio or Dearborn, is likely to be boom- .
town recipients although the relative isolation of Switzerland County
suggests the possibility of a temporary workforce staying during the
week and returning home on weekends during the construction phase.

           The urban counties, Clark, Jefferson, and Dearborn, would
appear to have both the instrumental orientation and the prestige sub-
orientation.  If proposed power plant construction in Dearborn County
is an addition to existing plant, we would expect no opposition from and
minimal disruption to the residents.  Clark County too would appear
receptive to power plant development.  We suggest that the considerable
opposition to the proposed Marble Hill plant has its roots in a relative
shift of orientations from instrumental (economic benefit) to prestige.
Having previously experienced considerable development (the proving
ground and existing power plant), and possibly having had negative
                            II-B-6-13

-------
experiences, we would expect less enthusiasm for further development.
Moreover, as people continue to make this area their home, the senti-
mental orientation tends to take hold, and given the relative affluence
of the area, and likely commitment to current prestige and stratifica-
tional patterns, the plant is seen as not needed.  The fact that it is
a nuclear plant is, of course, a factor as well.

           The rural counties of Switzerland and Ohio may be characterized
by the primordial belonging sub-orientation.  The rurality of the resi-
dents and their relative elder!iness suggest that they would not be in
favor of power plant development, though Ohio County would probably suffer
fewer impacts due to its proximity to the Cincinnati SMSA.

           Tippecanoe, Warren, Fountain and Vermin ion Counties

           Tippecanoe, Warren, Fountain and Vermin ion Counties in Western
Indiana present a mix of environmental orientations and likely reactions
to power plant development.

           Tippecanoe County presents a rather complicated picture.  The
economic diversity and progressive urban character of Tippecanoe County
suggest a strong economic benefit sub-orientation an'd thus, receptivity
to development.  However, the Lafayette/West Lafayette urban complex
does not appear able to contribute many workers to power plant construc-
tion, so a boom or "boomlet" situation seems likely and there may be
negative reaction to that.   The prestige sub-orientation is also present
in this county and opposition to the plant might come from groups that
have a stake in keeping things as they are.  Opposition groups would
also be likely to emerge from a nearby university too.  We woi.ld suggest
that receptivity to power plant development would be mixed and conflict-
laden given the mix of orientations.

           Vermillion County, too, has mixed orientations.  The urban
influence spreading from Terre Haute suggests the beginning of an instru-
mental orientation; however, many other sub-orientations are present and
much stronger.   There is the aesthetic sub-orientation and the age,
stability and occupational  structures present in the population suggest
a primordial belongingness  sub-orientation as well.  The decline in
relative wealth bodes well  for receptivity to power plant development;
however, the aesthetic and  belongingness sub-orientations suggest there
will be considerable opposition to the plant.  Nearness to Terre Haute
means a boomtown situation  is unlikely.   We suspect that the instru-
mental orientation will become dominant and receptivity to power plant
development will increase.

           Given their location, Warren and Foundation Counties are
candidates for a boomtown situation, and lack of economic diversification
suggests difficulty in coping with the syndrome.  The primary orientation
is sentimental  and we would predict considerable opposition to the power
plant, given the rural  nature of the population.
                             II-B-6-14

-------
           Posey', Perry, Warrick, Spencer, Crawford and Harrison Counties

           None of the six southwestern counties along the Ohio River
which are under consideration here, Posey, Perry, Warrick, Spencer, Craw-
ford and Harrison is a candidate for the boomtown syndrome because of
proximity either to Evansville or Louisville.   They do, however, differ
in their environmental orientations and likely receptivity to power
plant development.

           Perry and Crawford Counties both have primarily an aesthetic
sub-orientation, being scenic forested areas.   Moreover, Perry is the
site of a seminary, suggesting the moral-religious sub-orientatioh is
present there as well.  The population here is somewhat older and this
factor and the predominant orientation suggest that there would be
considerable opposition to power plant development in these counties,
both from within the county and from residents of nearby counties as
we! 1.

           On the other hand, both the instrumental  and sentimental
(prestige) orientations are present but not in conflict in Posey and
Warrick Counties, which are both affluent and  part of SMSA's.   We
predict that a power plant would be easily accepted by the residents.

           Spencer County, which appears to be urbanizing but as a sub-
urban community, probably has the prestige sub-orientation as its
predominant mode of relating to the environment.  We wouTd expect
opposition to the plant on these grounds unless it were sited in a
very rural area.

           Lawrence and Jackson Counties

           Lawrence and Jackson Counties in south-central Indiana have a
rural non-farm, stable population comprised mainly of married people
with children.  The people work primarily in the manufacturing and trade
areas-  The predominant orientation is sentimental but both counties
have a substantial aesthetic sub-orientation.   Lawrence is a boomtown
candidate county while Jackson might possibly  escape this if workers
commute from Louisville and if the plant were  sited in the southern
part of the county.  Lawrence County, being the poorer of the two, is
likely to be more receptive to power plant, development.

           The economic infrastructure for coping with the boomtown
syndrome is relatively well-developed compared to other rural  areas
in the region in both these counties and we suspect opposition to
the plant would be minimal so long as the aesthetic sub-orientation
did not come into conflict with the instrumental one.

           Jasper County

           Jasper County, Indiana, being affluent, rural  and populated by
families in the child-rearing phase of the life cycle would probably be
                            II-B-6-15

-------
receptive to development of power plants, largely because it can avoid
the boomtown syndrome.  The environmental sub-orientations predominant
here are the prestige and economic benefit.   There is the possibility
of conflict should large portions of prime farm land be required for
plant siting; however, the nature of the population would indicate
favorable attitudes toward power plant development.

6.3.1.4.  OHIO COUNTIES

           Lawrence, Gallia, Meigs, Athens,  Washington, Monroe,
           Belmont and Jefferson Counties

           Lawrence, Gallia, Meigs, Athens,  Washington, Monroe, Belmont
and Jefferson Counties in Appalachian Ohio differ widely in their environ-
mental orientations.  Jefferson and Belmont  Counties, with their reliance
on coal and related industries, high incomes, older and.urban populations
have mixed orientations.  The instrumental orientation is dominant and
possibly the territorial one is present if coal is strip-mined, and
there is a healthy prestige sub-orientation.   Jefferson County's previous
experience with a large power plant, and its  recreational areas suggest
that previous contact with instrumental and  territorial orientations and
the presence of the aesthetic sub-orientation might work against ready
acceptance of further development.   Belmont,  on the other hand, appears
ready and welcoming.  Neither county would experience strong boomtown
effects for Wheeling and Steubenville should  be able to provide the needed
labor force.

           The profile of Washington County  is similar to that of
Jefferson though we suggest that the aesthetic and prestige sub-orienta-
tion are stronger there, and the people less  amenable to power plant
development.

           Lawrence County probably has the  strongest aesthetic sub-
orientation of all.  It is probably used by  nearby Huntingdon and
Ashland residents as a major recreational area.  Plant siting here
would conflict with the predominant sub-orientation, though no boom
would occur.

           Gallia, Meigs and Monroe are the most rural  counties and the
first two are boomtown candidates,  being rather isolated and poorly
served by highways.  Their lack of economic diversification bodes ill
for coping with power plant development.   As  with most rural counties
with older populations, the sentimental orientation is predominant.
Gallia probably has already experienced a boom due to power plant
construction and may or may not be eager for  more though  its unusual
population structure may work for acceptance.  Monroe County, however,
appears to need further economic diversification, and. would presumably
welcome a plant.   Meigs, the most rural,  would probably be opposed to a
power plant, given the age structure of the population.
                            H-B-6-16

-------
           Athens County presents a very mixed picture, as counties
housing universities often do.  The mining of coal and percentage
urban suggest instrumental and territorial orientations.  The out-
migration suggests that it is possibly those with a primordial belong-
ingness sub-orientation that are leaving.  The aesthetic sub-orienta-
tion is present too.  It is our prediction that reaction to the power
plant would be mixed and conflictful.  The instrumental orientation
will probably win out.  There is also a likelihood of a boom, given
Athens' relative isolation; however, the infrastructure seems diverse
enough to handle it.

           Hamilton, Butler, Warren, Montgomery, Miami and Clark Counties

           Clearly, the predominant environmental sub-orientations of
these southerwestern Ohio counties, Hamilton, Butler, Warren, Montgomery,
Miami and Clark are the economic benefit and prestige components.
Warren and Miami Counties are the only ones likely to have the additional
but secondary sub-orientation of primordial belongingness and Butler,
Warren and Clark to have the aesthetic sub-orientation.

           Hamilton, Montgomery and Clark, being SMSA counties are
probably not only receptive to power plant development but will  also
experience minimal  impact.  There are no boomtown candidates in this
group of counties.   This is generally true for Butler County as well.
The more mixed orientational character of Warren and Miami Counties
suggest possible community conflict concerning the desirability of
power plants; however, the general  prediction is eventual acceptance,
though Warren County is likely to be more tenacious in its resistance.

           Franklin, Pickaway, Ross, Pike and Scioto Counties

           Franklin, Pickaway, Ross, Pike and Scioto Counties, extending
from central  Ohio south to the Ohio River, present a complex picture of
environmental orientations.  Franklin and Pickaway, being urban, are
likely to be instrumental in orientation, though the latter county,
being a suburban county, has a strong prestige sub-orientation.   Gener-  •
ally, there should be little impact from nor reaction to power plant
siting due to a commuting labor force.   Farmers, however, are likely
to be opposed because the plant would likely absorb some profitable
farmland; some suburbanites may also be opposed.

           Ross, Pike and Scioto Counties, although fairly distant from
major urban centers, are on a major four-lane highway which should.
mitigate any major boomtown effects.  However, it seems likely that
during the construction phase, much of the labor force would be
workweek residents  of local communities, except for Ross County.
This means a strain on local  temporary housing.   The communities in
these counties are fairly diversified but Pike County, being the
poorest and least diversified, would be most hard pressed to cope
with development.  If a commuting laboring force from Portsmouth and
neighboring Ashland is sufficient to meet construction phase demands,
then Scioto County, being fairly economically diversified, should  be
                            II-B-6-17

-------
able to cope with its "boomlet".  All these counties have substantial
symbolic orientations due to the presence of forests, recreational
areas and  Indian sites.  The rurality and elderliness of the population
also suggests a strong primordial belongingness sub-orientation as well.
Opposition to plants in these counties, thus, is likely to be intense.

           Clermont, Brown and Adams Counties

           Clermont, Brown and Adams Counties in southern Ohio are
clearly not boomtown candidates though it appears Adams County has recently
experienced a modest one due to power plant construction.  However, it
has not yielded great economic returns as yet.  This may be due to the
relative elderliness of the population since older people tend not to
benefit greatly from power plant construction.  Moreover, the rurality
of the county means the economic infrastructure to create additional
benefits was probably weak.  The recreational facilties and the
rurality and age of the populace suggest the dominant orientations
are aesthetic and primordial belongingness.  These orientations
militate against favorability toward future power plant development.

           Clermont County, the richest, most diversified, and urban of
these three, has already experienced power plant construction and
would probably welcome more due to its instrumental  orientation and
strong economic infrastructure.

           Brown County seems to have a strong primordial belongingness
•sub-orientation given its rurality and older population.  Impact of a
power plant would probably be minimal thoughve would expect considerable
opposition.  On the othnr hand, ^rown may wish to imitate its Clermont
neighbor and become more instrumental in orientation and welcome power
plant development.

           Coshocton, Muskingum and Morgan Counties

           Coshocton, Muskingum and Morgan Counties in Ohio's interior
would react differently to power plant siting.  The very rural  and poor
Morgan County would be ill-equipped to deal with a likely boomtown
situation.  Moreover, the predominant environmental  orientation is
primordial belongingness with the aesthetic sub-orientation, meaning
there would be considerable resistance to development.

           Coshocton and Muskingum Counties, being more urban counties,
being more urban and economically diversidied, could cope with a boom,
but that is unlikely given their proximity to large urban areas.   Being
coal-producting counties as well, the predominant orientation is
probably instrumental.   Farming appears to be declining and not very
profitable with a corresponding decline in the sentimental  orientation.
We would expect these counties to be receptive to power plant develop-
ment and capable of benefiting from it.
                            II-B-6-18

-------
           Mahoning County

           Mahoning County in the northeastern urbanized area of Ohio
seems strongly instrumental  in orientation and the effects of siting
a power plant there minimal  given the highly industrial/urban nature of
the county.  The only opposition to be expected is possibly from
those landowners who have to give up their land and do not wish to do so.
                            II-B-6-19

-------
                               REFERENCES

1.   Sue Johnson and Esther Weil.   "Social Aspects of Power Plant
    Siting:  Ohio River Basin Energy Study.   Volume III-D, 1977.

2.   Erik Cohen.  "Environmental  Orientations:   A Multidimensional
    Approach to Social  Ecology."   Current Anthropology,  Vol.  17,
           No.  1, 1976.
                            II-B-6-20

-------
                     7.  IMPACTS ON PUBLIC HEALTH
          A recent (Spring, 1976) ERDA publication produced for the lay-
man and entitled, Living with Radiation, makes the statement. . .  "There
is no evidence to indicate that there is a level  of radiation exposure
below which we can say there is no life-shortening effect at all"  (1).
Similarly, the International Institute for Applied Systems Analysis in a
1976 study entitled, Evaluation of Health Effects from Sulfur Dioxide
Emissions for a Reference Coal-fired Power Plant, noted that. . .  "Dose-
response relationship (s) are not certain and, furthermore, the pollutant
or pollutants responsible for health effects are also uncertain" (2).
It would seem from such comment that an anlysis of health effects  has
little to go on at present.  Actually, a good deal of information  i_s_
available on the subject of public health versus energy production.
This information is, admittedly, subject to wide error bounds but  that
does not preclude its usefulness or relieve present-day researchers from
their responsibility of addressing the topic.

          Possibly the most important issues are these error bounds them-
selves.  For example, the question of the adequacy of .current regulations
covering exposure to external irradiation has recurred throughout  the
study and is definitely an issue of interest to the public in the  ORBES
Region.  A tenet of the ORBES project has been that all plants sited will
meet government regulations specifying SOX, NOX,  particulates, radio-
nucMde-emissions,*etc,; but, the-point has been made at several  public
meetings during the course of this year that industries have met govern-
ment regulations before only to find that the regulations were actually
permitting workers to receive health-damaging doses of chemical  agents.
The vinyl chloride-induced cancer cases that have occurred recently among
workers at a Louisville, Kentucky chemical plant have often been sited as
examples of the regulations "breaking down."  What then are the stated
error bounds on current government regulations for agents affecting ,
health that are likely to be released to the environment by an energy
conversion facility?  An exhaustive study of this question is beyond the
scope of this preliminary assessment.  However, a few of the key issues
are discussed below as examples of the problem faced by the ORBES
researchers.

7.1.   HEALTH EFFECTS DUE TO RADIATION

7.1.1.   RADIATION PROTECTION GUIDELINES

          The attitude that "all is well" with the nuclear industry has
been consistently reinforced by extensive media and advertising campaigns
on the part of the U. S. government to convince the people that "nuclear
power" is safe.  The following  excerpts taken from ERDA pamphlet GPO:1976
0-219-430 entitled "Nuclear Energy" is typical (3):
                                 II-B-7-1

-------
           "What  about  radiation  releases?
               Nuclear plants  routinely  release  very  small
           arnounts  of radioactivity.   Everyone  receives  radia-
           tion from sources  such as  the  sun, rocks, and soil.
           We  all receive  about 50,000 times more radiation
           from this "natural background" than  from a  nuclear
           plant.   In fact, many  of the things  we do - such  as
           watching color  TV, receiving an  X ray,  or traveling
           by  plane - expose  us to significantly  more  radia-
           tion than do nuclear plants.   Care is  taken through
           plant  design features  and  precise monitoring  equip-
           ment to  assure  that  the radioactive  releases  from
           nuclear  plants  remain  negligible.  The following
           table  shows  totla  average  radiation  exposure  from
           various  sources."

                    Average  Annual Radiation Doses
                           (in mi 11i rems)

              Natural
              Sun,  rocks, soil, etc.  ........ 130

              Man-made
              Medical diagnosis.  .  .	72
              Fallout from past nuclear
               weapoos -testtng	   4
              Television  . .	   0.1
              Commercial  products	   1.9
              Air Transport	   1.3
              Commercial  nuclear  power 	   0.003
                                          Total     209.303"

          The pamphlet states  that.  . .  "radioactive releases from nuclear
plants  remain negligible. .  ." and thus  the reader, presumably the laymen
for whom  such material is intended, would logically conclude that the
health  effects associated with nuclear plants are likewise "negligible."
Releases  of radioactive  material from nuclear facilities are indeed
minuscule when expressed as  a ..percent of that allowed by ERDA regulations.
Mr. Martin B. Biles, Director  ERDA1s Division of Safety, Standards and
Compliance, stated  in the preface of Volume 1 of "Environmental Monitoring
at Major  U. S. Energy Reserach and Development Administration Contractor
Sites"  that.  . . "All offsite  exposures  from routine effluent releases in
1975 were less than one  percent of the established radiation protection
'guidelines for the  public.   Additionally, the estimated cumulative 80 Km
man-rem exposure potential from operations at all ERDA sites is less than
0.02 percent  of the estimated man-rem dose due to natural and bapkground
environmental radiation" (4).  The "established radiation protection
guidelines" referred to  by Mr. Biles are generally cited as "ERDA Manual
Chapter 0524" which gives the guidelines listed in Table II-B-7-1 (5).
                               II-B-7-2

-------
                        TABLE II-B-7-1


                PERMISSIBLE AREA RADIATION DOSE


                                           ERDA Standard
 Location                               in nirem per annum

 Worker in a nuclear plant                     5000

 Individual  at site boundary                    500
-Person -1 iv-i-n'g -in "the ''r
   residential area                              170

 Anyone living within a 50 mile
   radius                                         170
                             II-B-7-3

-------
The figures of. 170 mrem per year are equivalent to approximately natural
background plus 5 percent.

          For the U. S. natural background sources can be broken down as
follows:

          Cosmic radiation from outer space.  The dose from outer space
varies considerably with altitude.  For this reason flight at high
altitudes is, to some extent, dangerous for persons not suitably shielded.
Cosmic "rays" are mostly atomic nuclei (protons) with very high energies
originating outside the earth's atmosphere.  The average dose from cosmic
radiation for a person living in the U. S. is from 35 to 45 mrem per year.

          External gamma radiation.  The uranium and thorium decay series
are given in Table II-B-1-21.  These elements, through their decay products,
make up part of the source of natural background gamma radiation.  The
other principal contributor is potassium-40.  Collectively, these elements
contribute on the average across the U. S. a dose of 60 mrem per year (6).

          In te rn a1 ra d i at ion.  The principal contributors to internal
radiation are carbon-14, potassium-40, polonium-210, radium-226 and
radon-222.  The average dose from these elements, all of which occur
naturally, is 25 mrem per year (6).

          Thus one accumulates a natural  dose of 120 to 130 mrem per year
without having to come into contact with a nuclear facility.  The
question-which then needs to be posed, given that the public has been
assured by ERDA of the safety of nuclear power on the basis of its
negligible contribution to total background, is simply. . . ."Does
natural background alone have a significant health effect?"  If the
answer to this question is no, then the additional load of 5 percent of
natural background is probably inconsequential.  If, on the other hand,
the answer is found to be yes, and that natural background is a direct
cause of natural  death, then one must state that a 5 percent additional
load could conceivably reduce by 5 percent the life span of those living
all of their lives within a 50 mile radius of a nuclear facility.

          The situation of a worker receiving a 5000 mrem dose per year,
or a member of the public receiving a 500 mrem dose at a site boundary,
is somewhat different.   These figures represent increases over natural
background of 3900 percent and 390 percent, respectively, as compared to
the approximate 5 percent increase allowed for persons living in the
nearest residence to a nuclear facility.   The ERDA publication cited
previously, Living with Radiation, gives the data provided in Table II-
B-7-2 as a source to aid the layman in coming to an understanding of the
health impact(s)  of such exposures (7).  Such exposures would, according
to ERDA, have no  health effect whatsoever.  Actually, there is some
question as to the validity of this supposition.
                               II-B-7-4

-------
                             TABLE II-B-7-2
                     EFFECTS OF  EXTERNAL  RADIATION
               (Total  Body Exposure  within  24  Hour  Period)

 Dose,  mrem*                             Effect

 500,000                                 Half of the  people  exposed
                                            would die

 200,000-250,000                         At least  one death  would  occur

 100,000-200,000                         Nausea, fatigue and sickness,
                                            but no  fatalities

 50,000                      .            Slight changes in blood chemistry
                                            but the person would  have  no
                                            symptoms he himself would
                                            notice

 25,0.00                                  We -woul.d  .probably find  no
                                            detectable effect

 5,000                                    Annual dose  permitted for a
                                            worker  in  a nuclear facility

 500                                     Dose allowed to person  on site
                                            boundary

 130                                     Natural background  accumulated
                                            over a  period of  one year
                                            (U. S.  average)


 *ERDA's publication,  Living with  Radiation, expresses these  data in rads
  of gamma  radiation  (7j!Conversion  to  mrem  is made  here in an  effort
•  to keep consistent  terms throughout  the report.
                                 II-B-7-5

-------
 7.1.2.   THE  DOUBLING  DOSE  FOR  CANCER

           The  cause and  effect relationships between cancer  induction and
 ionizing radiation are well  documented  for many types of human tissue.
 Arthur  Tamplin,  in his essay on  "Nuclear Reactors and the  Public Health
 and Safety." calculated  on  the basis of this information that some
 17,000  additional cancer deaths  would result in the United States each
 year if the  total population were exposed to an average background of
 170 mrem per annum  (8);  _K£. ,  if exposed to the level permitted by
 current ERDA standards.  Tamplin also noted. . .  "that the nuclear power
 industry,  acting in what I  (Tamplin) would say is a responsible manner,
 has built extra  safety into the  reactors" so as "to limit  the exposure of
 the public to  something  less than 10 mrem per year — that is, less than
 0.1  of  the allowable  guidelines" (8).   The question related  to this
 discussion which needs to  be answered for persons living in  the ORBES
 Region  is  simply. . . "What are  the health effects due to  radiation
 really  going to  be for the  four  scenarios chosen?"  To answer this
 question,  it is  necessary  to first determine what the dose to the public
 vn'll  be, and second,  to  then determine  what the health effect of that
 dose is.

           Doubling dose  is  defined as that dosage which would double the
 natural  or spontaneous incidence of cancer (9).  Doubling  dose for cancer
 in  general can be stated but a more meaningful analysis can  be developed
 by  studying  the  doubling doses for specific tumors.  Table II-B-7-3
.gives the  best estimates of .doubling,,d,ose of radiation for human cancers
 available  as of  1970.  It can  be seen from these data that a dose of
 25,000  mrem, which is the figure quoted by ERDA as "probably" harmless (11),
 could be expected to  easily double the  incidence of thyroid  cancer among
 young adults and possibly the  incidence of leukemia as well.  In addition
 to  this,  Stewart and  Kneale have determined that a dose of 2-3 rads (j_.e^.,
 2000-3000  mrem of gamma  radiation) received by a pregnant woman is
 sufficient to cause the  child  she is carrying to have a 50 percent
 increase in the  incidence of' a number of cancers, including  leukemia and
 brain tumors, during  the first ten years of its life (12).   This computes
 to  a doubling dose for in utero  induction of cancer of 4000  to 6000 mrem.

           Tamplin has proposed three laws as a statement of  summation for
 evidence to date concerning cancer induction by ionizing radiation.
 These are  as follows:

               Law I.  All  forms of cancer, in all probability, can be
           increased by ionizing  radiation, and the correct way to
           describe the phenomenon is either in terms of the  dose required
           to double the  spontaneous incidence rate of each cancer or,
           alternatively, as  an increase in the incidence rate of such
           cancers per rad of exposure.

               Law II.   All  forms of cancer show closely similar doubling
           doses  and closely  similar increases in the incidence rate per rad.
                                II-B-7-6

-------
                            TABLE II-B-7-3
                  BEST ESTIMATES OF DOUBLING DOSE OR
                     RADIATION FOR HUMAN CANCERS
Human Cancers by
Type or Location

Leukemia
Thyroid cancer
  in adults
  in children
bung-cancer
Breast cancer
Stomach cancer
Pancreas cancer
Bone cancer
Lymphatic and other
  hematopoetic organs
Carcinomatosis of
  miscellaneous origin
Doubling Dose
    mrem	

30,000-60,000

100,000
5,000-10,000.
 •v-.17 5 ,.000
 ~~ 100,000
 — 230,000
 •^125,000
 — 40,000

  ~ 70,000

  "- 60,000
Source:  These figures are based on data published by Tamplin in
         Reference (9).  Tamplin's data is expressed in rads rather
         than mrem, and all  doubling doses beyond those for thyroid
         in the table are shown as approximate.   A conversion factor
         of one rad = one rem was used for gamma radiation (10).
                                II-B-7-7

-------
               Law III.  Youthful subjects require less radiation to in-
          creae the incidence rate by a specified fraction than do adults.
          The developing fetus requires a still  lower amount of radiation.

          Based upon these laws and the data available at the time (ca.  1970)
Tempi in suggested the assigned doubling doses listed in Table II-B-7-4
as reasonable for all forms of cancer (13).

          The permitted dose of 5000 mrem per annum for a person over 18
working in a nuclear plant has long been an accepted standard as a safe
dose (14).  This dose is a factor of ten below that which would, according
to ERDA, give. . . "slight changes in blood chemistry but the person
would have no symptoms he himself would notice"  (15).  And, the regulation
of 5000 mrem per year maximum for a worker has indeed been rigorously
enforced by all ERDA facilities and contractors.   The wearing of film
badges in such facilities is obligatory, and the  badges are read at
frequent intervals to insure that no worker receives more than the
maximum permissible dose.  There is something of  a problem, however,
with accepting this figure of 5000 mrem per annum as safe in that such
a dose appears to be right at the threshold of that which will double the
incidence of cancer in young adults.
          A recent study by T. F. Mancuso, A. Stewart and G. Kneale
documenting the incidence of cancer among workers at the Hanford,
Washington Nuclear Facility has cast additional doubt upon the safety
of the permitted "5000 mrem per annum dose '(T6).  The Hanford plant
was originally designed and operated to produce plutonium for nuclear
weapons.  Construction on the plant began in 1943 and at one time
nine production reactors were in operation.  Today only one production
reactor operates at Hanford (17).

          Radiation monitoring of employees has been in operation
continuously since 1943 and sufficient time h & passed to allow for
health effects with long latency periods (i_.e_., some forms of cancer)
to develop.  As stated by Mancuso, Stewart and Kneale. . . "Time
has now elapsed for most of the non-survivors to be men who died 10 or
more years after leaving the industry" (18).  A summation of their
rather exhaustive analysis of the records on these workers is given
in Table II-B-7-5.

          It can be seen from these data that the permitted dose of
5000 mrem per annum is, in fact, double the doubling dose for "all
.RES neoplasms"- if it is assumed that a 2500 mrad dose as recorded
by a Hanford vyorker's film badge was due solely to gamma radiation
which has a biological effectiveness of 100 percent.  No mention is
made in the Mancuso, Stewart and Kneale report of any analysis being
available of badge records with reference to the type of radiation
received, although an assumption that exposure was due primarily to
                               II-B-7-8

-------
                            TABLE II-B-7-4
               ASSIGNED DOUBLING DOSE AND  INCIDENCE  RATES
Adults
Youthful subjects
  (less than 20 years of age)
Infants in utero
Approximately 100,000 mrem with approxi-
mately a 1% .increase in incidence rate
per year per 1000 mrem of exposure

Between 5000 and 100,000 mrem as the
doubling dose with a 1 to 2% increase
in incidence rate per year per 1000
mrem of exposure

Approximately 6000 mrem as the doubling
dose with a 17% increase in incidence
rate per year per 1000 mrem of exposure
Source:  These figures are based  on  data  published  by Tamplin  in
         Reference (13).   Tamplin's  data  are  expressed  in  rads  rather
         than mrem.   A conversion factor  of one  rad  = one  rem was used
         for gamma radiation  (10).
                                II-B-7-9

-------
                            TABLE II-B-7-5
       ESTIMATED DOUBLING DOSES FOR CRITICAL PRE-DEATH YEARS(a)
Radio-sensitive
  cancers	

Bone marrow

Pancreas

Lung

All RES neoplasms

All cancers
   Critical pre-death periods

Years before   Estimated doubling
  death          dose in mrads
Proportion of all  deaths

  Observed   Expected
9
0
14
11
13
809
7400
6100
2500
12200
0.62
-**
1.39
5.45
1.82
19.02
0.30
0.85
3.26
1.15
15.15
    The .years. .before ..death which .showed the maximum contrast, compared wi.th a
    standard group of all non-cancer deaths.

    Taken from United States Vital Statistics for deaths of white males (1960)
Source :   Radiation Exposures of_ Hanford Workers Dyinfl from Cancer.
         Mancuso, T. F. , Stewart, A., and G. Kneale, University of
         Pittsburgh, under contract E(ll-l )-3428, presented at the
         Tenth Midyear Symposium of the Health Physics Society, Saratoga
         Springs, New York, October 11-13, 1976, Revised Draft March,
         1977, Table 16.
                              II-B-7-10

-------
gamma radiation is probably valid.  It may not, however, be possible at
this date to determine exactly what the proportion of gamma to beta
radiation was for a specific worker.  Thus it must be stated that a
2500 mrad per annum dose as recorded by film badges at Hanford over the
years is approximately equal to about 2500 mrem but the exact figure is
unknown.  In any event, the doubling doses reported for the Hanford
study are much lower than estimates based on atomic bomb survivors
and as the authors have stated,.."They are unlikely to go unchallenged" (19),

          The summary of the Hanford study is very brief and is quoted
below (20).
             (v
               Data from the Hanford study have shown that
          sensitivity to the cancer-rinduction effects of
          radiation is at a low ebb between 25 and 45 years
          of age.  Nevertheless, at younger and older ages
          there is probably a cancer hazard associated with
          low level radiation which probably affects bone
          marrow cancers more than other neoplasms and cancev-s
          of the pancreas and lung more than other solid tumors.u

7.2.  PUBLIC HEALTH ASPECTS OF FOSSIL FUEL COMBUSTION

          The array of pollutants emitted from a fossil fuel power plant
is more diverse than those from nuclear plants.  This diversity has been
reduced-*here -to
-------
          Carcinogenic effects ascribed to SO;? were based on its ease of
 passage through cell membranes and the formation of such ions and free
 radicals as  sulfite, bisulfite and S02.  Sulfite and bisulfite are removed
 by enzyme systems within the body but S02 is longer lived and since it
 can  break disulfide bonds, is a possible mutagen.  Other studies indicate
 that S02 may predispose a system to effects by other chemicals.  Exposure
 to benzopyrene did not induce carcinogenesis in the absence of S02 (21).

          Effects upon the respiratory system per se by S02 include
 inhibition of ciliary action with high dosages and induced thickening
 of the mucus layer over cilia thus impeding their movement (21).

          Certain relationships have been developed by Buehring, et al.
 (22) to estimate quantified human health impacts of S02 emissions from
 coal-fired plants.  Their model was based on a reference 1000 MW plant
 located in Wisconsin and consists of two submodels, a health submodel
 based on a study by Finlea, et^ al_. (23) and a dispersion submodel
 based on results from a typical power plant.  Such parameters as S02
 emissions, population and site charateristics, and backgrounc S02 con-
 centrations  were considered in the human health model.

          The reference plant is a 1000 MW plant operating at 70% annual
 capacity, a  net heat efficiency of 36%, and has equal demand in all
 seasons.  Sulfur dioxide emissions are at a rate of 1.2 pounds per
 million BTU's, the stacks are 152 meters high by 5 meters in diameter,
*gases -exit the -stacks at a temperature 'of 148.7°C -and -at a flow rate
 of 334 nr/sec.  Annual average SO  concentrations within 80 km of the
 power plant  were calculated using^a Graussian plume model and were
 calculated separately for urban and rural  settings since greater
 turbulence exists in the air flow over a city (22).

          Via the health effects computer model five parameters of public
 health were  examined.   The results of their model indicate excess days
 of aggravation of heart and lung disease in the elderly and excess
 astham attacks would be anticipated to occur more frequently with
 premature mortality, excess acute lower respiratory disease in children,
 and  excess risk of chronic respiratory disease symptoms in adults
 anticipated  to occur less frequently.  These effects are projected to
 be the result of high pollution levels either over the long- or short-
 term.  "This study has indicated that the quantified health impacts of
 S02  emissions from a single coal-fired power plant can amount of thou-
 sands of days of human illness, and some premature fatalities." (22)

          These same parameters were examined with respect to health
 impacts of sulfate aerosol, again via computer model, in an EPA
 report (23).  Table II-B-7-6 is from that report as given in the
 Argonne report (21).
                                 II-B-7-12

-------
                                                 TABLE II-B-7-6
                                       HEALTH IMPACTS OF SULFATE AEROSOL
CO
 I
 I
I— '
CO
            Health Effect
Population at Risk
Assumed Baseline
Frequency of
Disorder within
Population at Risk
Pollutant
Concentration
Threshold
For Effect
Effect Increase as %
of Baseline per
Pollutant Unit Above
Threshold
            Mortality
            Aggravation of
            Heart and Lung
            Disease in
            Elderly
            Aggravation of
            Asthma
            Lower Respiratory
            Disease in
            Children
Total Population
The prevalence of
chronic heart and
lung disease among
the 11% of the
population older
than 65 years is 27%

The prevalence of
asthma in the general
population is 3%

All children in the
population or 23.5%
of population
            Chronic Respiratory
            Disease
              Non Smokers       62% of population
                                age 21 or older
              Smokers
38% of population
age 21 or older
Daily  death rate of
2.58 "per 100,000

One out of five of
population at risk
complain of symtom •
aggravation on any
given day
One cut of 50
asthmatics experiences
an attack each day

50% ot children have
one attack per year
2% prevalence


10% prevalence
25  g/m3 for     2.5% per 10  g/m3
one day or more

9  g/m3 for      14.1% for 10  g/rrT
one day or more
6  g/nr for
one day or more
33.5% per 10  g/m"
13  g/m3 for     76.9% per 10  g/m3
several years
                                                  10  g/m3 for     134% per 10  g/m3
15  g/m3 for     73.8% per 10  g/m3
several years
            SOURCE:  A_ Preliminary Assessment of_ the_ Health and Environmental Effects of_ Coal  Utilization TJT_ tne Midwest.
                       Vol. I. Energy Scenarios, Technology Characterizations, Air and Water Resource Impacts, and
                       Health Effects.  Argonne National Laboratory, January, 1977, p. 181.

-------
7.2.2.  HEALTH EFFECTS OF NITROUS OXIDE EMISSIONS

          Nitrous oxides (NO ) are produced from organic nitrogen com-
pounds in coal and oxidation'  of atmospheric nitrogen during the
combustion process.   Exposure of animals to concentrations under 0.5 ppm
showed irreversible emphysema-like conditions while an acute inflammatory
response was manifest upon exposure to this concentration for 4 hours.
The effects were not manifest in human subjects until several hours
after exposure.  Habituation effects^ as discussed in section 7.2.1.
occurred with respect to effects of acute inflammation only (21).

          Nitric oxide in an aqueous solution can be converted to the
nitrite ion (N0|) which can be further converted to a nitrosamine if
the organic amide base is available.  The possibility is present then
for lung cancer due to this pollutant and the hypothesis is being
researched that stomach cancer can also be an effect when particulates
are ingested.  And according to available data, it appears that NOo
will reduce ciliary action similar to the discussion in section 7.2.1. (21),

7,2.3.  HEALTH EFFECTS DUE TO PARTICULATE EMISSIONS

          Respirable particles are generally defined to be those in the
O.Ol-lQ^m diameter size range and, coal combustion releases significant
amounts of particles in this range.

          Due to the environmental conditions present during coal forma-
tion, .most naturally occurring elements'C-an .-be .found in coal deposits.
These elements are released upon combustion in the gaseous or solid phase
depending on volatility and association with other elements and compounds.

          In the human respiratory system particles with diameters of
0.01-1/tm are deposited mainly in alveoli while larger particles are
usually deposited in nasopharyngeal and tracheobronchial regions.
Particulates are then removed from the respiratory system by the action
of cilia, mucous and the cough reflex, to the pharynx where swallowing
or expectoration occurs.  If swallowed, the toxic action of the particu-
late matter, or other elements and compounds adsorbed onto its surface,
may occur in the stomach.

          Alveolar macrophage cells function to cleanse alveolar sur-
faces through their phagocytotic action.  After engulfing a foreign
substance the cell  is eliminated from this system through the ciliary
action described above, or it migrates through the alveolar wall  and
enters the lymphatic system.  Danger of possible toxic action exists
at this point if the compounds associated with the particle can be
dissolved in the tissue fluid.

          Particulates are cleared from the respiratory system at rates
varying from a half-life (assumed to be the amount of time needed to
remove 1/2 the amount of particulates) of 2-6 weeks if removed by
                                II-B-7-14

-------
either of the processes above, to months or years if deposited in the lung.
If participate matter remains for any extended length of time, it is
usually surrounded by fibrous material and removal  by natural methods
is extremely difficult, if not impossible (21).

          Synergi-stic effects are many when particulates are involved.
Toxic effects of irritants absorbed or adsorbed onto particulates are
magnified in that the particles hold the toxin in place, close to sensi-
tive tissues.  Further changes may occur because of this association;
for example, SO^ adsorbed to particulates can be converted to sulfuric
acid which is a more potent irritant (21).

          Particulates containing silica can cause excess fibrous tissue
formation and lead to silicosis, pneumoconiosis, etc.  This particular
effect may be less significant than others implicated as being caused by
power plant emissions.

          Implicated carcinogens, such as nickel, chromium, beryllium
and arsenic can be carried to lung tissue by particulates and thus
affect the respiratory system indirectly (21).

          Direct effects on the nervous system can be" caused by inspira-
tion and assimilation of lead, tellurium, mercury,  arsenic, selenium,
nickel, chromium, and vandium (21).

,7.2.4.  RADIOACTIVE EMISSIONS FROM COAL-FIRED PLANTS

          Eisenbud and Petrow compared the relative biological signifi-
cance of fossil-fuel  radioactivity with emissions from nuclear plants
and concluded. . . "That an electrical generating station that derives
its thermal energy from such fuels discharges relatively greater quan-
tities of radioactive substances into the atmosphere than many power
plants that derive their heat from nuclear energy"  (24).  These workers
calculated that a typical  1000 MW coal-fired plant would release 28
millicuries of radium-226 per year.  Eggermont has  estimated that the
yearly release for a 1000 MW plant w..uld range between 0.0047 and 570
millicuries depending on the source of coal  (26).  Martin, Harward and
Oakley studied the Ra-226, Th-232 and U-238 emissions from a 1960 MW
Tennessee Valley Authority Widows Creek Plant (25).  A summation of
their data is shown in Table II-B-7-7.

          It must be concluded from the work cited  here that coal-fired
plants do indeed emit a sizeable quantity of radioactive material to the
environment, primarily as part of the fly ash.  Typical concentrations
of Ra-226, Ra-228, Th-228 and Th-232 in fly ash are shown in Table II-B-7-8.

          Questions related to the public health impact of radioactivity
in fly ash which need to be answered for the ORBES  Region are:
                                II-B-7-15

-------
                                                  TABLE  II-B-7-7
 I
CO
f
I—•
en
                               AIRBORNE  RADIOACTIVITY  CONCENTRATIONS AND  DOSE RATES
                                      AT THE  WIDOWS  CREEK PLANT  ON  5/13/693
Station
no.
Location
Azimuth

Distance
Air concentration
(*Ci/cm3 x 10-15.)
Ra-226 Th-232 . U-238
Dose rate
(;
-------
                            TABLE II-B-7-8
           ANALYSES OF RADIOACTIVITY IN COAL AND OIL FLY ASH

                                 Concentration
                                 (pCi/g dry fly ash)
xSample
Appalachian coal ash
Utah coal ash
Wyoming coal ash
Japan coal ash
Alabama coal ash
Venezuela petroleum ash
TVA coal plants
Coal ash (Australia)
Oil fly asha
Turkey Point
Coal fly ashb
..Colbert, -TV A
Coal fly ashd
Widows Creek, TVA
Note: pCi = picocuries =
Ra-
226
3.8
1.3
---

2.3
0.21
4.25
7.98

0.18

2.3

1.6
10~1?- curies;
Ra-
228
2.4
0.8
1,3
1.5
2.2
0.49
2.85
---

0.17

.3.1

2.7
1 curie =
Th-
228
2.6
1.0
1.6
1.6
2.3
0.67
2.85
—

0.82

*. — —

2,8
3.7 x 1010
Th-
232
_•.—
—
—
—
—
—
2.85
—

0.17

3.1

2.7
disin-
       tegrations per second.
a Average of 6 samples for Ra-226, Th-228, and .Th-232;  Ra-228 assumed in
.  equilibrium with Th-232.
D Average of 5 samples for Ra-226, Th-228, and Th-232;  Ra-228 assumed in
  equilibrium with Th-232.
c Average of 12 samples; Ra-228 assumed in equilibrium  with Th-232.
d Average of 26 samples for Ra-226; Ra-228 assumed in equilibrium with  .
  Th-232.

SOURCE:  Martin, J.E., Harward, E.D., and D.T. Oakley,  "Radiation doses
           from fossil-fuel and nuclear power plants."   Power Generation
           and Environmental  Change, ed. by D.A.  Berkowitz and A.M.
           Squires, 1971, p.  108.
                                II-B-7-17

-------
               1.  What is the quantity of ash produced at each plant
                   presently operating?

               2.  What is the amount of radioactivity in the ash
                   from these plants?

               3.  What is the fate (j[.e_., transport) of this material?

               4,  What is the estimated dose in mrem per annum for
                   bone and lung tissue for persons living their total
                   life span within 2, 5, 10, and 20 miles of these plants?

               5.  What are the answers to questions 1 to 4 for each
                   of the 1000 MW coal plants sited for the ORBES study?

7.3.  IMPACTS OF THE FOUR SCENARIOS

          An exhaustive analysis of the public health effects of the
four scenarios, although clearly n'eeded and desired by the public, is
not expected as part of this first year's preliminary assessment.
However, such an analysis is forthcoming and will be developed more
fully before the study is completed.  The first year's effort has been
directed mainly at the more pressing or obvious public health problems
such as unemployment, which indeed may be a serious public health effect
of the low growth scenarios, radiation damage and effects due to sulfate
aerosols.  However, plants were noted sited below the level of county for
"the Tirst year making "it'difficult'to assess the impact'of any of these
variables in great detail, particularly those associated with socio-
log-!cal factors such as unemployment.  Also, background levels of sulfate
aerosols are needed before any statements can be made about health
effects at a specific site.  The.atmospheric dispersion of sulfur
dioxide, and thus of sulfate aerosols, will, however, be developed for
specific location(s) in the coming year.  The impact analysis of public
health effects has thus been reduced to talking about "typical" 1000 MW
coal or nuclear fired facilities located in either rural or urban
settings.  This sort of analysis has been done many times before and
there is no need to do it all  over again here.  What is needed is a
point by point analysis of what will happen, in terms of public health
in this instance, at a specific location in the ORBES Region given that
a conversion facility of a specific nature is to be located there.  To
do this for 167 plants for one of the high growth scenarios is an awesome
task, yet public health and siological effects are site specific.  Also,
a detailed analysis of transport of materials to target populations
away from that site is essential to making any sensible analysis of the
problem.

          In brief it appears  that we do have information available as
to the major health effects of energy Telectricity) production.  Given
specific siting and transport  data we should be able to use this
information to assess, with some degree of certainty, the health impacts
                                II-B-7-18

-------
of the four scenarios.  It would be premature to try to compare the
respective health effects, both good and bad, of the four scenarios
at this stage - except possibly for the workers in the plants them-
selves.
                                II-B-7-19

-------
                            LITERATURE CITED
 1.  Living with Radiation.  Francis L. Brannigan, Energy Research and
       Development Administration, Division of Safety, Standards and
       Compliance, ERDA-76/89, UC-41,  1976, p. 11.

 2.  Evaluation of_ Health Effects from Sulfur Dioxide Emissions for a_
       Coal-fired Power Plant.  Bueh Ring, W.A., Dennis, R.L., and
       A. Hfllzl, International Institute for Applied Systems Analysis,
       2361 Laxenburg, Austria, Research Memorandum RM-76-23, 1976,,p. 2.

 3.  Nuclear Energy.  Energy Research  and Development Administration,
       Office of Public Affairs, Washington, D.  C. 20545, GPO: 1976
       0-219-430, EDM-1016 R(9-76).

 4.  Environmental Monitoring at_ Major U_. S_. Energy Research and
       Development Administration Contractor Sites.  Division of Safety,
       Standards, and Compliance, U. S. Energy Research and Development
       Administration, ERDA 76-104, UC-11, Issued August, 1976,
       Preface to Volume 1.

 5.  Standards for Radiation Protection.  U. S.  Energy Research and
       Development Administration Manual, Chapter 0524, US ERDA,
       Washington, D. C., April 8, 1975.

 6.  Ibid. 4, p. 189.

 7.  Ibid. 1, Figure 4, p. 10.

 8.  Pov/er Generation and Environmental Change.   Edited by David A.
       Berkowitz and Arthur M. Squires, Symposium of the Committee on
       Environmental Alteration, American Association for the Advance-
       ment of Science, December 28, 1969, The M.I.T. Press, 1971, p. 49.

 9,  Ibid. 8, p. 50.

10.  Ibid. 1, p. 24.

11.  Ibid. 1, p. 10.

12.  A. Stewart and G. W. Kneale, "Changes in the Cancer Risk Associated
       with Obstetric Radiography," Lancet 1, 104-107 (1968).

13.  Ibid. 8, p. 53.

14.  Basic Radiation Protection Criteria.  Recommendations of the National
       Council on Radiation Protection and Measurements, NCRP Report No.  39,
       January 15, 1971.
                                  II-B-7-20

-------
15.  Ibid. 1, p. -10.

16.  Radiation Exposures of_ Hanford Workers Dying from Cancer and Other
       Causes.  Mancuso, T.F., Stewart,  A., and  G.  Kneale,  University
       of Pittsburgh, Under Contract E(ll-l)-3428,  Presented at the
       Tenth Midyear  Symposium of the Health Physics Society, Saratoga
       Springs, New York, October 11-13, 1976,  Revised Draft March, 1977.

17.  Ibid. 4,  Vol. 2, p. 1105.

18.  Ibid. 16,  p. 2.

19.  Ibid. 16,  p. 8,

20.  Ibid. 16,  p. ii.

21.  Argonne  National Laboratories.  1977.  A preliminary assessment  of
       the health and environmental  effects of coal utilization  in the
       midwest.  Draft.

22,  Buehring,  W.A., Dennis,  R.L. and Hfllzl, A.  1976.  Evaluation of
       health  effects from sulfur dioxide  emissions for a reference coal'
       fired  power plant.  International Institute for Applied Systems
       Analysis.  Research memorandum.  RM-76-23.

.23.  ..Nelson,,  l'I,.C.,,J
-------