Assessment Of Atmospheric Exposure and Deposition To High Elevation Forests In The Eastern United States


xvEPA
           United States
           Environmental Protection
           Agency
           Office of Research and
           Development
           Washington, DC 20460
EPA/600/3-90/058
September 1990
An Assessment of
Atmospheric Exposure
and Deposition to High
Elevation Forests in the
Eastern United States

-------
                                                               EPA/600/3-90/058
                                                               September 1990
AN ASSESSMENT OF ATMOSPHERIC EXPOSURE AND DEPOSITION

TO HIGH ELEVATION FORESTS IN THE EASTERN UNITED STATES
                            Volker A Mohnen
                            Principal Investigator
                        Atmospheric Sciences Group
                        State University of New York
                            Albany, N.Y. 12222

                  Viney Aneja, North Carolina State University
                   Bruce Bailey, Associated Weather Service
                  Ellis Cowling, North Carolina State University
                     S. Michael Goltz, University of Maine
                      James Healey, The Fleming Group
                John A. Kadlecek, State University of New York
                  James Meagher, Tennessee Valley Authority
                Stephen F. Mueller, Tennessee Valley Authority
                    John T. Sigmon, University of Virginia
                 COOPERATIVE AGREEMENT CR 813934-03-0
                              Project Officer

                            Ralph Baumgardner
                    U.S. Environmental Protection Agency
                                AREAL
                     Research Triangle Park, N.C. 27711

-------
                                    ACKNOWLEDGEMENT
"This research has been funded through a cooperative agreement with the United States Environmental
Protection Agency (813-934030) as part of the Mountain Cloud Chemistry Program."

A substantial part of this Report is also presented in Vong et. al., "Network Measurements of Droplet
Deposition of Atmospheric Pollutants", in "Deposition Monitoring: Methods and Results - NAPAP State
of Science/State of Technology Report No. 6", (ed. D. Sisterson), NAPAP, Washington DC, 1989, and in
Vong et. al., "Cloud Water Deposition", in "Atmospheric Processes Research and Process Model
Development - NAPAP State of Science/State of Technology Report No. 2",  (ed. B. Hicks), NAPAP,
Washington, DC, 1989.  The authors of this report wish to thank R. J. Vong for his superb coordinating,
editing and writing efforts.
                                         DISCLAIMER
The contents of this document do not necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial products constitute endorsement or
recommendation for use.

-------
                          TABLE OF CONTENTS



ACKNOWLEDGEMENT	  ii

TABLE OF CONTENTS	   iii

LIST OF TABLES	  v

LIST OF FIGURES  	   ix

FOREWORD	  xvi

1. INTRODUCTION	   1-1

2. CONCLUSIONS  	   2-1

      CLOUD FREQUENCY	   2-1
      CLOUD AND PRECIPITATION CHEMISTRY  	   2-1
      LIQUID WATER CONTENT 	   2-1
      WET, DRY AND CLOUD DEPOSITION	   2-2
      OZONE	   2-3
      HYDROGEN PEROXIDE	   2-4
      OXIDES OF SULFUR AND NITROGEN	   2-4
      METEOROLOGY	   2-4

3. PROJECT DESCRIPTION	   3-1

      MONITORING RATIONALE	   3-1
      GEOGRAPHICAL REPRESENTATION .	   3-2

4. ASSESSMENT	   4-1

      CLOUD WATER CHEMISTRY  	   4-1
      FREQUENCY OF CLOUD	   4-16
      LIQUID WATER CONTENT 	   4-21
      METEOROLOGY	  4-24
      GASES	  4-33
      THROUGHFALL 	  4-42
      DEPOSITION	  4-45

5. METEOROLOGICAL ASSESSMENT OF AIR QUALITY	   5-1

      CASE STUDY ON THE IMPACT OF AIR
      MASS ORIGIN  ON AIR QUALITY  	   5-1
      METEOROLOGICAL INFLUENCES ON
      CLOUDWATER CHEMISTRY	   5-1
      GAS CONCENTRATION VS. AIR MASS TRAJECTORIES	   5-6
                                    111

-------
                         TABLE OF CONTENTS
                               (CONTINUED)


6. METHODS AND PROTOCOLS	   6-1

      CLOUD WATER COLLECTION 	   6-1
      FREQUENCY OF CLOUD	   6-2
      LIQUID WATER CONTENT  	   6-5
      GASES	   6-9
      METEOROLOGICAL MEASUREMENTS 	  6-10
      THROUGHFALL 	  6-12
      CLOUD INTERCEPTION	  6-14
      DRY DEPOSITION	  6-17

7. DESCRIPTION OF DATA BASE AND QUALITY ASSURANCE PLAN  	  7-1

      DESCRIPTION OF DATA BASE 	  7-1
      INTERNAL QUALITY ASSURANCE	  7-1

APPENDIX A	  A-l

APPENDIX B	B-l

APPENDIX C	  C-l

APPENDIX D	  D-l

APPENDIX E	  E-l

REFERENCES
                                    IV

-------
                            LIST OF TABLES
TABLE 1-1


TABLE 4-1



TABLE 4-2


TABLE 4-3


TABLE 4-4



TABLE 4-5


TABLE 4-6


TABLE 4-7


TABLE 4-8


TABLE 4-9

TABLE 4-10


TABLE 4-11


TABLE 4-12


TABLE 4-13


TABLE 4-14


TABLE 4-15
A HISTORICAL OVERVIEW OF CLOUD
ACIDITY MEASUREMENTS	  1-3

COMPARISON OF WINTER AND SUMMER
CLOUD ION CONCENTRATIONS AT
WHITEFACE MTN. NY	  4-2

COMPARISON OF MEDIAN SULFATE TO
NITRATE RATIOS AT WHITEFACE MTN., NY  	  4-3

ION CONCENTRATIONS FOR SIMULTANEOUSLY
COLLECTED CLOUD WATER SAMPLES	  4-6

AVERAGE ION CONCENTRATIONS FOR BOTH
PRECIPITATING AND NON-PRECIPITATING
CLOUDS	  4-15

CLOUD FREQUENCY AT MCCP SITES, JUNE TO
SEPTEMBER, 1986-88  	 4-20

STATE TOTALS FOR LAND AREA (HECTARES X 1000)
ABOVE 800 M  	 4-21

SUMMARY OF CLOUD DROPLET SIZE DISTRIBUTIONS
AT WHITETOP MTN., VA	4-23

MCCP HOURLY AVERAGED CLOUD LIQUID WATER
CONTENT SUMMARY FOR 1987-88 	 4-24

SELECTED LONG-TERM CLIMATE STATIONS	 4-25

1986 MONTHLY TEMPERATURE DEVIATIONS FROM
NORMAL	4-28

1986 MONTLY PRECIPITATION DEVIATIONS FROM
NORMAL	 4-28

1987 MONTHLY TEMPERATURE DEVIATIONS FROM
NORMAL  	4-29

1987 MONTHLY PRECIPITATION DEVIATIONS FROM
NORMAL	 4-29

1988 MONTHLY TEMPERATURE DEVIATIONS FROM
NORMAL	'	4-30

1988 MONTHLY PRECIPITATION DEVIATIONS FROM
NORMAL	4-30

-------
                            LIST OF TABLES
                               (CONTINUED)
TABLE 4-16

TABLE 4-17

TABLE 4-18


TABLE 4-19


TABLE 4-20


TABLE 4-21


TABLE 4-22


TABLE 4-23



TABLE 4-24



TABLE 4-25


TABLE 4-26

TABLE 4-27


TABLE 4-28


TABLE 4-29


TABLE 4-30
MCCP AND AIR SITES 	4-34

OZONE EXPOSURE SUM OF SEASON DOSE	4-40

SULFUR DIOXIDE MEASUREMENTS SUMMARY FOR
SELECTED MCCP SITES	4-41

CLOUD WATER H2O2 MEASUREMENTS AT MCCP SUMMIT
SITES FOR 1986-88 		4-42

DEPOSITION ENRICHMENT RATIOS (THROUGHFALL/BULK
PRECIPITATION) AT WHITEFACE MTN.,NY	4-44
MEDIAN VALUE OF SITE ENRICHMENT RATIOS
(THROUGHFALL/BULK PRECIPITATION)	
4-44
ACTUAL (KEQ/HA/YR) AND PERCENT ION CONTRIBUTION
TO TOTAL DEPOSITION AT WHITETOP, VA  	4-45

A REVIEW OF CHEMICAL ION AND CLOUD WATER
DEPOSITION VIA DROPLET INTERCEPTION, VARIOUS
LOCATIONS, INVESTIGATORS, AND YEARS	4-47

FREQUENCY OF OCCURENCE OF VARIOUS SYNOPTIC
METEOROLOGICAL CLASSES BY SITE AND YEAR FOR
GROWING SEASON	4-51

DIFFERENCES BETWEEN SYNOPTIC/TRAJECTORY
SUBCLASSES FOR SELECTED SITES	4-52

MEAN POTENTIAL DEPOSITION BY SUBCLASS AND SITE	4-53

CLASS-WEIGHTED DEPOSITION POTENTIAL BY SITE
FOR 1986-88	4-55

GROWING SEASON CLOUD DEPOSITION POTENTIAL
EXPOSURE BY SITE AND YEAR 	4-56

1986-88 COMPUTED MEAN HOURLY CLOUD DEPOSITION
BY SITE AND SUBCLASS	4-57

COMPUTED GROWING SEASON TOTAL CLOUD DEPOSITION
FLUX BY SITE	4-59
                                    VI

-------
TABLE 4-31

TABLE 4-32

TABLE 4-33

TABLE 4-34


TABLE 4-35
TABLE 4-36a


TABLE 4-36b


TABLE 4-37

TABLE 5-1



TABLE 5-2


TABLE 5-3
TABLE 6-1

TABLE 6-2


TABLE 6-3
TABLE 6-4
           LIST OF TABLES
               (CONTINUED)

MEAN MONTHLY COMPUTED CLOUD DEPOSITION FLUX 	4-61

CHEMICAL SPECIES FOR WHICH DEPOSITION WAS INFERRED . . . 4-62

MCCP WARM SEASON DRY DEPOSITION 	4-64

SITES USED TO ESTIMATE WET DEPOSITION VIA
PRECIPITATION FOR MCCP SITES	4-65

CONCURRENT WARM SEASON WET DEPOSITION VIA
PRECIPITATION IN THE EASTERN USA FROM
NADP/NTN OR MAP3S SITES NEAR MCCP MONITORING
LOCATIONS	4-66

ESTIMATED 1987-88 SULFUR DEPOSITION BUDGETS
AT MCCP SITES	4-68

ESTIMATED 1987-88 ACIDIC NITROGEN DEPOSITION
BUDGETS AT MCCP SITES  	4-69

ESTIMATED CLOUD-TO-WET DEPOSITION FLUX RATIOS	4-70
CLOUD WATER ION STATISTICS CATEGORIZED
BY SYNOPTIC CONDITIONS FOR WHITEFACE MT.
SUMMIT: 1986-88	
5-;
PERCENTAGE OF CLOUD HOURS BY SYNOPTIC
TYPE FOR FIVE MCCP SITES	  5-4

ANALYSIS OF VARIANCE RESULTS FOR FOUR ION
CONCENTRATIONS: MEAN VALUES FOR ALL MCCP
SITES (1986-88) IN jxEQ/L FOR SELECTED
SYNOPTIC-TRAJECTORY CLASSES, ALL CLOUDS  	  5-5

GAS MEASUREMENTS AT MCCP SITES  	  6-9

MCCP SITE SENSOR MOUNTING LOCATIONS AND
HEIGHT ABOVE CANOPY	 6-11

MEAN NUMBER OF THROUGHFALL GAUGES AND
COLLECTION AREA NEEDED FOR 5% ERROR
UNDER A UNIFORM CANOPY (FROM HELVEY
AND PATRIC, 1965)	 6-12

THROUGHFALL BUCKET COLLECTION AREA
REQUIRED FOR 10% RELATIVE STANDARD DEVIA-
TION IN A COMPLEX CANOPY ON WHITEFACE
MOUNTAIN, NY (FROM KADLECEK, 1989)	 6-13
                                    vn

-------
                           LIST OF TABLES
                               (CONTINUED)
TABLE A-l'
TABLE A-2

TABLE A-3


TABLE A-4


TABLE A-5



TABLE A-6
ARITHMETIC MEAN, MAXIMUM, AND MINIMUM
CONCENTRATIONS IN 410 CLOUD/FOGWATER SAMPLES
FOR THE FALLS OF 1983-1986 IN KLEINER
FELDBERG/TAUNUS, 15 KM NORTH OF FRANKFURT	  A-l

GERMAN WEATHER SERVICE MTN SITES  	  A-2

WEEKLY FOG/CLOUD WATER SAMPLES OBTAINED WITH
PASSIVE STRING COLLECTOR AT WUELFERSREUTH  	  A-2

HOURLY FOG SAMPLES OBTAINED WITH ROTATING
STRING COLLECTORS IN PO VALLEY, ITALY	  A-3

MEAN ION CONCENTRATIONS IN HOURLY CLOUD WATER
SAMPLES FROM FOUR TRAJECTORIES FOR ARESKUTAN,
SWEDEN	  A-4

MEAN ION CONCENTRATIONS AT TWO CANADIAN CHEF
SITES IN 1986	  A-5
                                   Vlll

-------
                           LIST OF FIGURES

          PAGE NUMBERS BEGINNING WITH "E" MAY BE FOUND IN APPENDIX E

FIGURE 3-1        THE LOCATION OF THE MCCP MONITORING SITES	  3-4

FIGURE 4-1        COMPARISON OF CLOUD WATER AND PRECIPITATION
                 SULFATE DISTRIBUTIONS: WINTER AND SUMMER	  4-4

FIGURE 4-2        COMPARISON OF CLOUD WATER AND PRECIPITATION
                 NITRATE DISTRIBUTIONS: SUMMER AND WINTER	  4-4

FIGURE 4-3        FREQUENCY DISTRIBUTION FOR 1986-88 SULFATE
                 DIFFERENCES BETWEEN WHITEFACE SUBSITES 1 & 2
                 INCLUDES PRECIPITATING AND NON-PRECIPITATING
                 CLOUDS	  4-7

FIGURE 4-4        MCCP CLOUD CHEMISTRY CONCENTRATIONS: MEAN
                 VALUES IN MICROEQUIVALENT / LITER FOR
                 PRECIPITATING AND NON-PRECIPITATING CLOUDS 	  4-9

FIGURE 4-5        MCCP CLOUD CHEMISTRY CONCENTRATIONS: MEAN
                 VALUES IN MICROEQUIVALENT / LITER FOR
                 PRECIPITATNG CLOUDS	  4-10

FIGURE 4-6        SULFATE CONCENTRATION FOR NON-PRECIPITATING CLOUDS
                 WHITEFACE MTN SUMMIT 1986-88		  4-12

FIGURE 4-7        SULFATE CONCENTRATION FOR NON-PRECIPITATING CLOUDS
                 WHITETOP MT. SUMMIT & MT. MOOSILAUKE SUMMIT	  4-13

FIGURE 4-8        SULFATE CONCENTRATION FOR NON-PRECIPITATING CLOUDS
                 SHENANDOAH SUMMIT & MT. MITCHELL SUMMIT	  4-14

FIGURE 4-9        AVERAGE ANNUAL LOW (<2500 M) CLOUD AMOUNTS (%) IN
                 EASTERN NORTH AMERICA, FOR LAND AREAS ONLY, 1971-81  . . .  E-l

FIGURE 4-10       AVERAGE CLOUD BASE HEIGHT (M) IN EASTERN NORTH
                 AMERICA FOR STRATUS/STRATOCUMMULUS CLOUDS
                 THE DOMINANT LOW CLOUD TYPE, FOR LAND AREAS
                 ONLY, 1971-81 	E-2

FIGURE 4-11       CLOUD AMOUNTS BY TYPE OVER THE APPLACHIANS,
                 BY SEASON, 1985-87	  4-18

FIGURE 4-12       RELATIVE PROBABILITY OF LOW CLOUD (<2100M) FOR
                 THE N. APPALACHIAN REGION (VA TO CANADA)  - SUMMERS
                 1985-87 - BASED ON CLOUD BASE & CLOUD THICKNESS
                 DERIVED FROM US AIR FORCE RTNEPH CLOUD DATA BASE	E-3

FIGURE 4-13       PROBABILITY OF CLOUD BY HEIGHT AND SEASON
                 AT MOOSILAUKE, NH	  E-4
                                    IX

-------
FIGURE 4-14
FIGURE 4-15
FIGURE 4-16
FIGURE 4-17
FIGURE 4-18
FIGURE 4-19
FIGURE 4-20
FIGURE 4-21
FIGURE 4-22
FIGURE 4-23
FIGURE 4-24
FIGURE 4-25
FIGURE 4-26
          LIST OF FIGURES
               (CONTINUED)

PROBABILITY OF CLOUD BY HEIGHT AND SEASON
AT SHENANDOAH, VA 	 E-5

PROBABILITY OF CLOUD BY HEIGHT AND SEASON
AT WHITEFACE MTN., NY  	 E-6

SUMMER (1985-87) PROBABILITY OF CLOUD OCCURENCE
(RTNEPH DATA) FOR THE WEST-EAST TRANSECT
ACROSS THE NORTHERN APPALACHIANS AT APPROX.
44.5° N. LATITUDE	 4-19

LOW-LEVEL CLOUD FREQUENCY DEPARTURE (%) FROM
1965-85 NORMALS FOR 8 NWS AIRPORTS
1986 FIELD SEASON	E-7

LOW-LEVEL CLOUD FREQUENCY DEPARTURE (%) FROM
1965-85 NORMALS FOR 8 NWS AIRPORTS
1987 FIELD SEASON	 E-8

LOW-LEVEL CLOUD FREQUENCY DEPARTURE (%) FROM
1965-85 NORMALS FOR 8 NWS AIRPORTS
1988 FIELD SEASON	 E-9

PROFILE OF CUMULATIVE CLOUD IMP ACTION FREQUENCY
AT WHITEFACE MTN. (1986-88) FOR WARM SEASON
CLOUD HOURS	 E-10

PROFILE OF CUMULATIVE CLOUD IMP ACTION FREQUENCY
AT WHITETOP MTN. (1986-88) FOR WARM SEASON
CLOUD HOURS	 E-ll

LAND AREAS ABOVE MEAN CLOUD BASE (>800M) IN
THE EASTERN US SUSCEPTIBLE TO ACID CLOUD
DROPLET DEPOSITION	 4-22

CLOUD DROPLET SIZE DISTRIBUTIONS FROM
WHITEFACE MTN., NY	 E-12

AVERAGE 850 MB TEMPERATURE (°C) AND PRESSURE HEIGHT
(M) DEPARTURES FROM NORMAL (1957-85 PERIOD) FOR
THE 1986-88 MCCP FIELD SEASONS (JUNE-SEPT) FOR
SIX EASTERN US CITIES	 4-27

DROUGHT SEVERITY (LONG TERM PALMER) OVER THE
EASTERN US MID SUMMER 1986	 E-13

DROUGHT SEVERITY (LONG TERM PALMER) OVER THE
EASTERN US MID SUMMER 1987	 E-14

-------
FIGURE 4-27
FIGURE 4-28
FIGURE 4-29
FIGURE 4-30
FIGURE 4-31
FIGURE 4-32
FIGURE 4-33
FIGURE 4-34
FIGURE 4-35
           LIST OF FIGURES
               (CONTINUED)

DROUGHT SEVERITY (LONG TERM PALMER) OVER THE
EASTERN US MID SUMMER 1988	 E-15

CUMULATIVE DEPARTURE FROM NORMAL PRECIPITATION
FOR THREE HIGH ELEVATION, LONG TERM STATIONS
FOR 1986-88	 4-32

PERCENT FOREST COVER BY COUNTY (REGIONS: MIDWEST,
NORTHEAST, SOUTH, AND MID-ATLANTIC)	 E-16

PERCENT FOREST COVER BY COUNTY (REGIONS: PACIFIC
NORTHWEST, ROCKY MTN. AND WEST)	 E-17

NATIONAL TREND IN THE COMPOSITE AVERAGE OF THE
SECOND HIGHEST MAXIMUM 1-HOUR OZONE CONCENTRATION
AT BOTH NAMS AND ALL SITES WITH 95% CONFIDENCE
INTERVALS, 1978-1987 	 E-18

NATIONAL TREND IN THE COMPOSITE AVERAGE OF THE
ANNUAL SULFUR DIOXIDE CONCENTRATION AT BOTH
NAMS AND ALL SITES WITH 95% CONFIDENCE
INTERVALS, 1978-1987 	 E-19

NATIONAL TREND IN THE COMPOSITE AVERAGE OF THE
NITROGEN DIOXIDE CONCENTRATION AT BOTH
NAMS AND ALL SITES WITH 95% CONFIDENCE
INTERVALS, 1978-1987 	 E-20

NATIONAL TREND IN VOLATILE ORGANIC COMPOUND
EMISSIONS, 1979-1986	 E-21

NATIONAL TREND IN SULFUR OXIDE EMISSIONS
1978-1987	 E-22
FIGURE 4-36
FIGURE 4-37
FIGURE 4-38
FIGURE 4-39
OZONE CONCENTRATION (PPM) EXAMPLES OF TWO SIGMOIDAL
WEIGHTING FUNCTIONS 	 E-23

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
HOWLAND FOREST	 E-24

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
MT MOOSILAUKE SUMMIT 	 E-24

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7.AM & 6 PM
WHITEFACE MT. SUMMIT	 E-25
                                    XI

-------
                        LIST OF FIGURES
                            (CONTINUED)
FIGURE 4-40
FIGURE 4-41
FIGURE 4-42
FIGURE 4-43
FIGURE 4-44
FIGURE 4-45
FIGURE 4-46
FIGURE 4-47
FIGURE 4-48
FIGURE 4-49
FIGURE 4-50
FIGURE 4-51
FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
WHITEFACE MT. SUBSITE 3	  E-25

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
WHITEFACE MT. SUBSITE 4	  E-26

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
HUNTINGTON, NY	  E-26

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
HAMPSHIRE COUNTY, MA	  E-27

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
BEAVER COUNTY, PA	  E-27

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
SHENANDOAH SUMMIT	  E-28

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
SHENANDOAH SUBSITE 2	  E-28

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
SHENANDOAH SUBSITE 3	  E-29

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
BIG MEADOW, VA	  E-29

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
SAWMILL RUN, VA 	  E-30

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
DICKEY'RIDGE WARREN COUNTY, VA	  E-SO

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
WHITETOP MT. SUMMIT	  E-31
                                XII

-------
FIGURE 4-52



FIGURE 4-53



FIGURE 4-54



FIGURE 4-55

FIGURE 4-56

FIGURE 4-57

FIGURE 4-58

FIGURE 4-59

FIGURE 4-60

FIGURE 4-61

FIGURE 4-62

FIGURE 4-63

FIGURE 4-64

FIGURE 4-65

FIGURE 4-66

FIGURE 4-67



FIGURE 5-1


FIGURE 5-2


FIGURE 5-3
           LIST OF FIGURES
               (CONTINUED)

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
GILES COUNTY, TN	 E-31

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
MARION SMYTH COUNTY, VA  	 E-32

FREQUENCY OF CONSECUTIVE HOURS OF OZONE GREATER
THAN 70 PPB BETWEEN THE HOURS OF 7 AM & 6 PM
MT. MITCHELL SUMMIT	 E-32

OZONE EXPOSURE - TOTAL SEASON APR 15-OCT 15 	 E-33

OZONE EXPOSURE SUM OF SEASONAL DOSES	 E-34

DIURNAL OZONE CONCENTRATION	 4-36

DIURNAL OZONE CONCENTRATION	 4-37

DIURNAL OZONE CONCENTRATION	 4-38

DIURNAL OZONE CONCENTRATION	 4-39

MEAN OZONE CONCENTRATION BY SITE 	 E-35

WFM 03 DISTRIBUTION FROM 1973 TO 1988  	 E-36

SO2 CONCENTRATION WHITEFACE MTN SUMMIT  	 E-37

SO2 CONCENTRATION WHITEFACE MTN SUBSITE 3	 E-38

SO2 CONCENTRATION WHITETOP MTN SUMMIT	 E-39

SO2 CONCENTRATION MT. MITCHELL	 E-40

FREQUENCY OF OCCURRENCE FOR H2O2 CONCENTRATIONS
IN CLOUD WATER COLLECTED AT THE WHITEFACE, NY
AND WHITETOP, VA MCCP SITES	 E-41

HOWLAND FOREST: WARM SEASONS 1986 - 1988:
MEAN OZONE VS. 36-HOUR BACK TRAJECTORY  	 E-42

MT. MITCHELL: WARM SEASONS 1986 - 1988:
MEAN OZONE VS. 36-HOUR BACK TRAJECTORY  	 E-42

MT. MOOSILAUKE: WARM SEASONS 1986 - 1988:
MEAN OZONE VS. 36-HOUR BACK TRAJECTORY  	 E-43
                                    Xlll

-------
                            LIST OF FIGURES
                                (CONTINUED)
FIGURE 5-4
FIGURE 5-5
FIGURE 5-6
FIGURE 5-7
FIGURE 5-8
FIGURE 5-9
FIGURE 5-10
FIGURE 5-11
FIGURE 5-12
FIGURE 5-13
FIGURE 5-14
FIGURE 5-15
SHENANDOAH:  WARM SEASONS 1986 - 1988:
MEAN OZONE VS. 36-HOUR BACK TRAJECTORY  	 E-43

WHITEFACE MT: WARM SEASONS 1986 - 1988:
MEAN OZONE VS. 36-HOUR BACK TRAJECTORY  	 E-44

WHITETOP MT:  WARM SEASONS 1986 - 1988:
MEAN OZONE VS. 36-HOUR BACK TRAJECTORY  	 E-44

HOWLAND FOREST: WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	 E-45

MT. MITCHELL:  WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	 E-45

MT. MITCHELL:  WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	 E-46

MT. MOOSILAUKE: WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	 E-46

MT. MOOSILAUKE: WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	 E-47

SHENANDOAH:  WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	 E-47

SHENANDOAH:  WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	 E-48

WHITEFACE MT: WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	 E-48

WHITEFACE MT: WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	 E-49
                                    xiv

-------
                            LIST OF FIGURES
                                 (CONTINUED)
FIGURE 5-16
FIGURE 5-17
FIGURE 5-18
FIGURE 5-19
FIGURE 5-20
FIGURE 5-21
FIGURE 5-22
FIGURE 5-23
WHITETOP MT.: WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	
WHITETOP MT.: WARM SEASONS 1986 - 1988:
FREQUENCY OF OZONE CONCENTRATION VS.
36-HOUR BACK TRAJECTORY	
                                                                         E-49
                                                                         E-50
WHITEFACE MT.: WARM SEASONS 1986 - 1988:
MEAN H2O2 VS. 36-HOUR BACK TRAJECTORY	 E-51

WHITETOP MT.: WARM SEASONS 1986 - 1988:
MEAN H202 VS. 36-HOUR BACK TRAJECTORY	 E-51

HOWLAND FOREST: WARM SEASONS 1986 - 1988:
MEAN SO2 VS. 36-HOUR BACK TRAJECTORY	 E-52

MT. MITCHELL: WARM SEASONS 1986 - 1988:
MEAN SO2 VS. 36-HOUR BACK TRAJECTORY	 E-52

WHITEFACE MT.: WARM SEASONS 1986 - 1988:
MEAN S02 VS. 36-HOUR BACK TRAJECTORY	 E-53

WHITETOP MT.: WARM SEASONS 1986 - 1988:
MEAN SO2 VS. 36-HOUR BACK TRAJECTORY	 E-53
                                     xv

-------
FOREWORD
This report represents an assessment of exposure and deposition of air pollutants to forests, and
interception of clouds by forest canopies. It summarizes data obtained over the first three years of MCCP,
1986 - 1988, with limited exposure data from 1989.

The report is part of a series of reports to the Environmental Protection Agency produced by the
Mountain Cloud Chemistry Project in fullfillment of contractual agreements. These include:

Mohnen et. al., "Exposure of Forests to Gaseous Air Pollutants and Clouds", June, 1987.
Mohnen et. al., "Elevational Gradients in Wet and Dry Deposition of Pollutants", Jan 1988.
Mohnen et. al., "Chemical Deposition to High Elevation Spruce-Fir Forests in the Eastern
United States", Feb 1988.
Mohnen et. al., "Exposure of Forests to Air Pollutants, Clouds, Precipitation and Climatic
Variables, Sep 1988.
Mohnen et. al., "Mountain Cloud Chemistry Project Wet, Dry and Cloud Water Deposition",
Sep 1988.

The main emphasis of MCCP scientists was to publish their results in the peer reviewed literature. Over
fifty publications stemming from MCCP research have appeared, or will appear, in peer reviewed journals.
Over twenty manuscripts are in various stages of preparation for submission to peer reviewed journals. In
addition, numerous papers were presented at scientific meetings which have been published in conference
proceedings. Finally, all principal scientists of MCCP have been involved in the preparation of the State
of Science (SOS) documents dealing with exposure/deposition. A complete list of all these publications
produced by MCCP scientists to date is presented in Appendix D.

Similar efforts to those of the MCCP are carried out in southeastern Canada under the CHEF (Chemistry
of High Elevation Fog) Program. It will therefore be possible in the future to jointly assess the impact of
air pollutants, clouds and precipitation on forests throughout eastern North America.

For a complete discussion of results from the MCCP, the reader is referred to the scientific journals. This
report highlights some major results and summaries of MCCP achievements.
xvi

-------
                                            SECTION 1
                                         INTRODUCTION

          This report is the third in a series of annual summaries of research on the exposure and
deposition of airborne chemicals to forest canopies and the forest floor in eastern North America. It
presents results from the 1986-88 field seasons, with limited exposure data from 1989.  The report is
produced  by the scientists in the Mountain Cloud Chemistry Program, a multi-year study of atmospheric
chemistry and physics sponsored by the United States Environmental Protection Agency.

          The Mountain Cloud Chemistry Program (MCCP) is a principal source of atmospheric
chemical and physical information available to evaluate the hypothesis that acidic and other airborne
chemicals contribute to the observed decline in spruce-fir forests at high elevations in the eastern USA.
The objective  of the MCCP is to characterize the exposure of montane forested ecosystems (largely
spruce fir) to atmospheric inputs (chemical, physical and climatic). To meet this objective, MCCP
conducted atmospheric research to determine:

                 elevational gradients in wet  and dry deposition;

                 the relative significance of various deposition mechanisms to the fluxes of chemical
                 materials into and through forest canopies; and

                 frequency distributions of chemical, physical, and climatic exposure.

          Resource and logistical considerations dictate where measurements can be performed.  Five
high-elevation sites and one low elevation site were selected from 45 «N to 35° N to be as representative
as possible of the geographic and meteorological variability in this large region.

          During the past two decades, high elevation red spruce and fir forests in large areas from
Maine to  North Carolina have shown visible symptoms of injury , decreased radial growth, and increased
mortality  (McLaughlin, 1985; Hornbeck and Smith, 1985; Johnson and Siccama, 1983). Field studies by
Falconer and Falconer (1980), Waldman  et al. (1982), and Dollard et al. (1983) demonstrated that cloud
water can be much more acidic than precipitation, and that substantial fluxes of sulfur and  nitrogen can
result from droplet interception to forest canopies.  Unsworth (1984) calculated that evaporation of
intercepted cloud water from forest canopies can lead to chemical concentrations on leaf surfaces that
are substantially larger than concentrations measured in the drops themselves.

          Because chemicals in cloud water may contribute to the stress experienced by high-elevation
forests in  eastern and western North America (Fowler et al., 1988; Waldman and Hoffman, 1988; Turner
et al., 1988), networks of monitoring sites have been started to characterize the concentration and
deposition of atmospheric pollutants in these forests.  The principal monitoring network in the USA is
the MCCP, which was implemented by the Forest Response Program of the National Acid  Precipitation
Assessment Program (NAPAP).  The data collected by the MCCP is being used in the NAPAP
Integrated Assessment to evaluate the role of airborne chemicals in the changing condition of forests.

          This report documents  typical values, extremes, and sources of variability in chemical
concentrations and deposition from clouds that contact the forest canopy.  Because the data are  limited
in coverage, it is difficult to generalize about spatial patterns. Processes controlling the variability in 5
cloud water concentration and deposition are addressed to define limits on extrapolation.  Both cloud
water chemical concentration and deposition are considered because foliar-mediated processes may reflect


                                                1-1

-------
the concentrations in cloud water while soil-mediated processes reflect cloud water deposition (McCune
and Lauver, 1986; Kohlmaier, et al., 1986; Schaefer, et al. 1989; Reuss et al., 1987).

Numerous investigators have reported data on cloud acidity and, to varying degrees depending
on the focus of their study, on cloud water chemistry. Table 1-1 shows examples of cloud acidity
measurements over several decades and locations. The cloud acidities reported by these authors vary
greatly, indicating that no "average" cloud pH exists. Each pH measurement reflects the chemical
composition not only of the condensed phase, but also of the gaseous phase. The condensed phase
contains a mixture of dissolved gases, aqueous solutions, and solids, which may vary with air mass origin,
cloud type, and height above cloud base.

Characterizing the chemical composition of cloud droplets depends on the objectives of the
investigation. A complete characterization requires, at a minimum, determination of the partitioning of
each species of interest among the various phases (Morgan, 1982), and as a function of size in the
condensed phase (Noone et al., 1988). Considering the difficulties associated with sampling clouds, it is
not surprising that such a complete characterization has not yet been performed. The present under-
standing of cloud chemistry is therefore based on bulk cloud water samples that represent a weighted
average droplet composition.
1-2

-------
                                   TABLE 1-1
                A Historical Overview of Cloud Acidity Measurements

                      Cloud water reported from low elevations
                                 Year of
Reference                      Study

Houghton (1955)               1954
Mrose (1966)                   1957
Lazrus et al. (1970)             1967
Bressan & Larson (1979)        1975
Waldman et al. (1982)           1981
Fuzzi et al. (1983)              1980-82
Munger et al. (1983)           1981-82
Fuzzi et al. (1984)               1982
Jacob et al. (1985)             1982-83
Jacob et al. (1986)             1983-84
Muir et al. (1986)              1985-86
Weathers et al. (1986)           1984
                     Location

                     Northeast USA
                     Germany
                     Puerto Rico
                     Nova Scotia, Canada
                     California, USA
                     Italy
                     California, USA
                     New York, USA
                     California, USA
                     California, USA
                     Midwestern USA
                     Eastern USA
                          Range of
                          pH

                        4.5 - 7.2
                        3.8 - 5.1
                        4.9 - 5.4
                        3.0 - 6.9
                        2.2 - 4.0
                        2.5 - 7.0
                        2.2 - 5.8
                        4.3 - 6.4
                        2.2 - 6.2
                        2.7 - 7.2
                        2.9 - 4.1
                        2.1 - 3.0
                          Cloud collected water by aircraft
Reference

Oddie (1962)
Petrenchuk &
  Drozdova (1966)
Scott (1978)
Scott & Laulainen (1979)
Lazrus et al. (1983)
Daum et al. (1984)
Year of
Study

 1960

1961-64
 1976
 1977
 1980
1981-83
Location

United Kingdom

USSR
Australia
Michigan, USA
Eastern USA
Eastern USA
Range of
  EH

4.4 - 7.2

3.4 - 5.9
4.6 - 7.5
3.7 - 4.0
3.6 - 4.4
3.1 - 6.1
                       Cloud water collected at montane sites
Reference

Okita (1968)
Castillo (1979)
Year of
Study

 1963
 1976
Falconer & Falconer (1980)     1977-79

Weathers et al. (1986)          1984
Hering et al. (1987)            1983
Mohnen & Kadlecek (1989)     1982-87
Location

Japan
Whiteface Mtn,
  NY USA
Whiteface Mtn
  NY USA
Eastern USA
California, USA
Whiteface Mtn,
  NY USA
Range of
  EH

3.5 - 6.5

3.4 - 4.2

2.7 - 4.7
2.8 - 3.1
3.1

2.5 - 4.8
                                       1-3

-------
SECTION 2

CONCLUSIONS
This section presents a summary of the key findings and conclusions of the MCCP during its
first three field seasons. A more detailed presentation of these findings is presented in Section 4 of this
report. A description of the methods and protocols used to derive these findings can be found in
Section 6.
CLOUD FREQUENCY

Regional cloud climatology studies indicate that cloud base height was typically between 800 m
and 1200 m in the Appalachian Mountains. The northern Appalachians were cloudier than the
southern. MCCP data for five high-elevation sites indicated that the Appalachian peaks (> 1400 m)
were in cloud 30% to 40% of the time during the growing season (April-October), and, for one site with
year-round data (Whitetop, VA), more frequently during the non-growing than the growing season.
Elevations of 1000 m and higher experienced cloud at least 10% of all hours. Clouds were present at
some time on at least one out of three days for lower elevations (1000 m to 1400 m), while higher sites
(> 1400 m) had clouds up to 75% of days during the growing season. The higher elevation mountains
had a greater frequency of cloud at night, in contrast to a predominance in the afternoon over
surrounding lower terrain. Approximately 25% of the mountain cloud events were precipitating.

CLOUD AND PRECIPITATION CHEMISTRY

In high elevation spruce-fir forests, observed cloud water pH was approximately 0.6 pH units
lower than precipitation pH. Concentrations of the major ions (NH/+, H+, SO/2", and NOj+) were
substantially higher in cloud water than in precipitation by factors ranging from 5 to 20, depending on
location within the Appalachians. H+, SO./2', and NOj' concentrations greater than 100 jteq/L occurred
infrequently in precipitation but routinely in cloud water (> 40% of cloud hours) at the five MCCP
monitoring sites. Thus, locations below cloud experienced both lower mean concentrations and fewer
extreme concentrations than high-elevation sites in the eastern USA There was a significant altitudinal
gradient within clouds at Whiteface Mountain, NY, for NH/+, H+, SO/2', and NOj"1" ions, with higher
concentrations at or near cloud base. Thus, in the Appalachian Mountains, a step increase in the
concentrations of wet deposited ions was observed as elevation increased from below cloud to cloud base
(approximately 1000 m); further increases in elevation (within cloud) should result in a gradual decrease
in the concentrations to which a forest canopy is exposed.

LIQUID WATER CONTENT

LWC values averaged approximately 0.20 g/m3 at all sites, excluding Whiteface Mountain. The
exception was Whiteface summit, where the average value was 0.46 g/m3. A small amount of data from
Whiteface Mountain, NY, suggest that the liquid water content of a cloud was lower at cloud base than
at higher elevations. Therefore, observed LWC values at any site were most likely a function of site
elevation relative to cloud base height as well as synoptic meteorology.
2-1

-------
WET, DRY AND CLOUD DEPOSITION

One of the principal findings is that deposition of SOj2', H+, NH^+, and NO/ in cloud water
(cloud interception) represented a significant input to montane forest canopies. In estimating cloud
deposition to the forest canopy, MCCP used an improved version of the 1984 Lovett model (CDM).
Cloud impaction events at each MCCP site were classified according to meteorological conditions that
prevail during each hour of the event. A detailed analysis of cloud chemistry and meteorological
variables (wind speed and liquid water content) has demonstrated the usefulness of this technique for
uniquely characterizing conditions during events. In addition to synoptic classification, specific air
trajectory directions computed within a given event type could further characterize event conditions.
This technique was able to explain a major portion of the variance in the chemistry and meteorological
data base, thereby allowing more complete growing season estimates of cloud deposition. Deposition
estimates for MCCP sites could then be made for periods when data were incomplete as long as cloud
frequency and event type would be determined. Event types were (1) pre-warm front, (2) NW sector of
cyclone, (3) post-cold front, (4) warm sector of cyclone, (5) stationary front, (6) marine flow off Atlantic,
(7) cutoff low in upper atmosphere, and (9) cap cloud.

Cloud deposition estimates for each site were made by computing, for each subclass, the mean
water deposition flux using the improved CDM and subclass wind speed and liquid water content. Best
estimates of canopy structure were then used to calculate the gross (pre-evaporation) cloud water flux to
specific forest canopies at each site. Cloud water deposition can exceed the flux from wet (precipitation)
deposition during the growing season at the mountain sites which are frequently exposed to cloud. Most
of the site-differences in cloud deposition could be explained on the basis of differences in canopy
structure. Cloud deposition was found to be highly site specific. Despite an almost 2:1 advantage in
cloud frequency, Whiteface Mountain mean deposition estimates for the growing season were generally
lower than those for Moosilauke because of lower canopy surface area at the specific Whiteface site.
Differences in cloud water deposition between the northern and southern sites were significant and they
were likely caused by differences in such parameters as canopy structure, elevation above cloud base and
synoptic meteorology. It was also interesting to observe the annual changes in cloud water deposition as
a result of changing metrological conditions from year to year. Whiteface Mountain, NY and Mt.
Moosilauke, NH had estimated 1987-88 warm-season sulfate cloud droplet deposition that was of the
same order of magnitude as measured precipitation chemical flux. Mt. Mitchell, NC had an estimated
cloud water sulfate deposition that was about a factor of two higher than precipitation deposition.
Estimated cloud sulfate deposition was up tp 4 times greater than that in precipitation for the Whitetop
Mt. MCCP site.

The Atmospheric Turbulence Diffusion Laboratory deposition model (DRY DEP) was used by
the MCCP to calculate the deposition velocities of sulfur dioxide, ozone, nitric acid vapor, and sulfate
particles from meteorological and site specific biological information. Site input included: major and
minor plant species type, leaf area index for plant species, and site location. All available
meteorological data including canopy wetness and rainfall were also used. The estimates that have been
made suggest that dry deposition of sulfate and nitrate is of greater relative importance at the Howland,
ME and Shenandoah, VA sites than at the other high elevation sites.

It is possible to estimate the importance of cloud deposition to the overall flux of pollutants
received by the forest canopy from the model results. Although cloud interception is highly dependent
upon canopy structure and location above cloud base, it can be nevertheless concluded, that cloud
deposition can deliver to the canopy from one to four times the amount of pollutants as precipitation
does.
2-2

-------
The estimated cloud-to-wet deposition flux ratios for 1987-88 growing season at four MCCP sites are:

Site Sulfate Nitrate

Moosilauke 0.8 1.2
Whiteface-2 1.1 0.8
Whitetop 3.5 4.4
Mt. Mitchell 2.0 2.2
At the cloud free Howland site, dry deposition appears to account for less than one third of the
total acidic substances deposited. The Shenandoah site has a very low cloud impaction frequency due to
its relatively low elevation and dry deposition appears to be of greater relative importance (deciduous
trees). The estimates for Whiteface suggest that dry sulfur deposition accounts for less than a quarter of
the total sulfur deposition flux.

Based on these research results, it can be concluded that any assessment of forest damage at
high elevations must take all delivery mechanisms into account, and in particular, cloud deposition.
However, substantial uncertainties may be associated with the deposition estimation procedures available
to the MCCP for estimation of both dry deposition and cloud water interception to mountain forests and
many of these uncertainties are as yet undefined.

OZONE

The long-term ozone exposure indices indicate considerable variability among the high elevation
sites and at a given site from one year to the next. For example, the sum of season dose (total of hourly
concentrations exceeding 70 ppb, expressed in ppm-h) for the 1988 growing season at the highest
elevation subsite at each of the MCCP locations ranged from 12.51 ppm-h at Mt. Moosilauke to 45.17
ppm-h at Mt. Mitchell. Temporal variability is well illustrated at Mt. Mitchell, where the same dose
statistic was 8.34, 5.14, and 45.17 ppm-h for the 1986, 1987, and 1988 monitoring seasons. The severe
drought of 1988 along with the accompanying clear skies and stagnation episodes undoubtedly
contributed to the high ozone exposures observed at many sites that year.

Quantifying the spatial and temporal variability in ozone episodes also reveals substantial
variability. For example, during 1987 the MtMitchell summit had eight episodes of ozone exceeding 70
ppb for at least eight consecutive hours, whereas during 1988 the figure increased to 33 episodes of this
magnitude and duration.

Important differences between the high elevation and low elevation sites are revealed in the data.
High elevation sites are often above the ground based temperature inversion at night, therefore the
ozone concentrations do not drop off to near zero as they do at valley bottom or low elevation flat
terrain sites. Therefore, the diurnal cycles and the exposure patterns are substantially different at
mountain versus valley stations, even those in close proximity to one another. The two primary effects
of this difference are to produce higher mean ozone concentrations and longer episodes at the high
elevation sites compared with nearby valley bottom locations
2-3

-------
HYDROGEN PEROXIDE

       The mean concentrations of hydrogen peroxide in the gas phase at Whitetop Mountain and
Whiteface Mountain during the warm season were between 0.6 ppb and 1 ppb. The maximum
concentration was 6.1 ppb. The mean concentrations of hydrogen peroxide in cloud water at Whitetop
and Whiteface Mountains during the warm season were between 32 and 36 pmol/1 with a maximum
concentration of 182 pinol/1.
OXIDES OF SULFUR AND NITROGEN

       The concentrations of SO2 and NOX measured at the MCCP sites were found to be quite low
(1-2 ppbv) most of the time.  Occasional excursions above 10 ppbv, presumed to be associated with
pollutant plumes, rarely lasted more than a few hours.


METEOROLOGY

       Regional surface and upper air climatological measurements indicate that there was a significant
contrast in weather conditions relative to long term normals between the Northern and Southern
portions of the Appalachians from 1986 to 1988.  Drought and warmer than normal conditions
dominated the Southern Appalachians throughout the period.  In some areas this drought was
unprecedented over the last 100-300 years.  The Northern Appalachians had relatively unexceptional
weather overall in terms of persistent trends.
                                               2-4

-------
                                           SECTION 3

                                    PROJECT DESCRIPTION
MONITORING RATIONALE

       The Mountain Cloud Chemistry Program (MCCP) has been designated by the EPA as a
principal source of atmospheric chemical and physical information available to evaluate the hypothesis
that acidic and other airborne chemicals contribute to the observed decline in spruce-fir forests in the
eastern USA.  The objective of the MCCP is to characterize the exposure of montane forested
ecosystems (largely spruce fir) to atmospheric inputs (chemical, physical and climatic).  To meet this
objective, MCCP conducted atmospheric deposition research to determine:

               elevational gradients in wet and dry deposition;
               the relative significance of various deposition mechanisms to the fluxes of chemical
               materials into and through forest canopies; and
               frequency distributions of chemical, physical, and climatic exposure.

       Resource and logistical considerations dictate where measurements can be performed.  Five
high-elevation sites and one low elevation site were selected from 45°N to 35°N to be as representative
as possible of the geographic  and  meteorological variability in this large region.

       Well-defined methods were not available to measure some inputs to high-elevation forests, and
therefore many of these inputs must  be estimated.  Pollutant concentrations were measured at the
boundary of forest canopy, and deposition was calculated using models. Measurements included gaseous
and aqueous chemical measurements, cloud frequency, cloud liquid water content, and climatic
measurements (deposition model inputs as well as a direct stress).

       Few of the physical and chemical measurements required were easily performed.  For example,
no  field measurement technique existed in the commercial sector for some measurements, such as
hydrogen peroxide and cloud  liquid water content.  Even relatively simple procedures (e.g., ozone
measurement or collection of particulate samples) required  modification when applied to the harsh, wet
mountaintop  environment.  So, the MCCP elected to perform certain experiments at a single site as part
of a method evaluation experiment prior to use at all six sites. Therefore, some parameters may be
available at some sites and not others.

       The meteorological monitoring, conducted concurrently with the chemical exposure
measurements, was designed to provide frequency of cloud exposure, deposition model inputs, and
climatic information.  Frequency of cloud immersion is  critical in defining the extent, both in time and
space, of the  cloud water deposition  process  and in identifying opportunities for chemical contributions
to,  and subsequent impacts on, montane ecosystems.

       The measurement approach taken by the MCCP and its contribution to the three MCCP
objectives are:

        •      Primary Exposure Parameters: Site-specific measurements of cloud and rain water,
               gaseous sulfur and nitrogen compounds, and oxidants are made every hour or are
               directly converted into hourly concentration values for use in exposure estimates.
                                               3-1

-------
Derived Deposition Parameters: Wet deposition of pollutants by precipitation can be
directly calculated from hourly and weekly totals and, in general, provides a yardstick for
comparison of other deposition processes. Cloud water deposition is estimated via
indirect physical and chemical measurements. Process-oriented models are needed to
estimate dry deposition and the material fluxes into and through the forest canopy.
• Inter-site Characterization: To enhance understanding of exposure patterns at individual
sites, some inter-site comparison was performed to determine if any distinguishing
regional characteristics could be observed in the data (e.g., meteorological, elevational,
or latitudinal differences in cloud water chemical composition).

GEOGRAPHICAL REPRESENTATION

The Mountain Cloud Chemistry Program (MCCP) consists of six sampling sites in the
northeastern and southeastern USA. Five sites are located at high elevations that experience frequent
cloudiness: Whiteface Mountain, NY; Mt. Moosilauke, NH; Shenandoah, VA; Whitetop Mountain, VA;
and Mt. Mitchell, NC. One low-elevation MCCP site in Howland Forest, ME, does not experience
substantial cloudiness, but was included for comparison of total deposition and forest condition. The
location of the MCCP sites are depicted in Figure 3-1.

At Whiteface Mt., Whitetop Mt. and Mt. Mitchell, the main sites are located on the summit of
mountains, while the main sites at Shenandoah and Moosilauke are along a ridgeline. Sub-sites are
located on the slopes of Whiteface, Shenandoah, and Mt. Mitchell. At the Shenandoah and Mitchell
sub-sites, instruments are mounted on scaffolding towers; at Whiteface, a mobile trailer is equipped with
a collector attached to a pole that is hoisted up and down on an event basis.

The Howland Forest site (45°13'N, 68°43'W) is located at 65 m elevation between Howland and
Edinburg, Maine, 35 miles north of Bangor. The forest is spruce with lesser numbers of balsam fir,
hemlock, and white pine. A second site is located 2 km west of the first at an elevation of 60 m in an
extended spruce-fir stand. This site has uniform terrain and extensive fetch ideally suited to classical
micrometeorological measurements.

Mt. Moosilauke, New Hampshire (43°59'N, 71°48'W), is one of the most southern peaks of the
White Mountains. It is located about 50 km southwest of Mt. Washington and about 10 km northeast
of the USFS Hubbard Brook Experimental Forest and Watershed. The site at Moosilauke is at 1000 m
and is partially shielded from the prevailing westerly winds. Consistent with its mid-elevation location,
the site does not experience the high frequency of orographic clouds that can be seen on the summit
(1465 m). The forest composition ranges from mixed hardwoods at lower elevations to spruce-fir (about
10% spruce) at mid-elevations, and pure balsam fir at high elevations.

Whiteface Mountain (44°23'N, 73°59'W) is located in the northeastern Adirondack Mountains in
New York, at an elevation of 1483 m. The summit is above tree line, providing access to regional air
flow (sub-site 1). A four-story building houses instrumentation and is the primary site for cloud water
collection. The Whiteface Mountain-Lake Placid Turn sampling site (sub-site 2) is located at 1245 m
adjacent to a balsam fir canopy. A meteorological tower has been installed for routine monitoring, but
cloud water collections are made from a mobile tower along the road. The Integrated Forest Study
(IPS) sampling and monitoring activities take place at 1025 m (sub-site 3), an elevation frequently near
or below cloud base. The forest canopy on the lower slopes consists of balsam fir, white birch, and red
spruce. Sampling for three precipitation networks (MAP3S, NTN, and MODES) and U.S. EPA and
NOAA dry deposition monitoring are conducted at the base support area (sub-site 4; 620 m).
3-2

-------
The Shenandoah, VA site (38°7'N, 78°20'W) is in the Shaver Hollow Watershed, located in the
north-central sector of the Shenandoah National Park. Slopes in the watershed average 30% to 50%,
facing generally northwest. Meteorological towers are erected at three sites in the watershed: sub-site 1
at 1015 m, sub-site 2 at 716 m, and sub-site 3 at 524 m. The tower locations are representative of the
surrounding deciduous forest canopy (primarily, northern red oak, chestnut oak, and red maple, with
several northern hardwoods and eastern hemlock in wetter coves).

Whitetop Mountain (36°38'N, 81°36'W) is located in the Mt. Rogers National Recreation Area
of the Jefferson National Forest in southwestern Virginia, 6 km southwest of Mt. Rogers, the highest
peak in Virginia. The Whitetop Mountain summit research station (at 1689 m) straddles the main
ridgeline of the Appalachian range, strategically located to intercept air from several directions. Most
measurements are made 5 to 8 meters above ground on a platform over a trailer located in a clearing.

The southernmost MCCP site is located in Mt. Mitchell State Park, North Carolina (35°44'N,
82°16'W), one mile southwest of Mt. Mitchell, which is the highest peak in the eastern USA (2038 m).
The summit is covered with Fraser fir, and the region from 1500 m to 1800 m is mixed fir and spruce.
A 16.5 m meteorological tower is installed on Mt. Gibbs at 1950 m (sub-site 1); additional gas and
meteorological measurements are performed at 1775 m (sub-site 2).
3-3

-------
Whitet
                                             n
                                           Forest  S

                                                 •^
                                             **t. Moosilauke
      Figure 3-1: The location of the MCCP monitoring sites
                          3-4

-------
SECTION 4

ASSESSMENT
The process of estimating exposure and deposition of atmospheric pollutants involves the
assessment of a number of important factors. Measurements of cloud water chemistry, gas concentration,
frequency of cloud, cloud liquid water content, other meteorology, and throughfall are factors that must
be considered to adequately characterize exposure and deposition. In this section, each of these factors
is considered in some detail.
CLOUD WATER CHEMISTRY

This section presents results for cloud water chemical composition for all sites. The 1986-88
growing season mean values for NO/, NH^+, H+, and SO^2" are presented along with their frequency
distributions for non-precipitating clouds. Summer and winter concentrations are compared, as well as
differences with elevation at one well-studied site, Whiteface Mountain, NY. Because concentrations of
pollutants in cloud water are highly variable, possible meteorological controlling factors are also
investigated. ^2^2 concentrations in cloud water are presented.

Unless otherwise stated, the cloud water chemistry data represent unweighted concentrations.
Although the concentrations are expected to vary with liquid water content, few analyses have shown a
significant correlation between cloud water concentrations and liquid water content (LWC). One of the
reasons for the lack of demonstrated correlation is the relative absence of accurate time-resolved LWC
data; although MCCP measured LWC during 1987-88 there are enough missing LWC data (about 30%)
for that period to make any LWC-weighing scheme problematic. Mohnen (1988c), however, compared
LWC weighted and unweighted methods for the 1987 MCCP data and did not reach different
conclusions for different weighing schemes.

We have characterized central tendencies in the MCCP cloud water chemistry data using mean
values. Schemenauer (1986) compared mean, median, and daily median statistics for summarizing cloud
water data and reached similar conclusions, but sometimes achieved a shift in emphasis, for different
summary statistics. To avoid problems associated with skewed data distributions, we either display the
data distributions (for qualitative comparisons), transform the data (for statistical analyses), calculate
median values, or utilize both mean and maximum values (to characterize exposures).

Seasonal Variation

Although the MCCP only collects cloud water samples during April-October, previous cloud
water chemistry data from the Whiteface Mountain, NY summit site provide insight into seasonal
variability in ion concentrations. The median and maximum concentrations, the ratio of sulfate to
nitrate ion, and the distribution of concentrations in precipitation and cloud water were compared for
summer and winter sampling periods at Whiteface Mountain by Kadlecek et al. (1985; 1988). The
nearby MAP3S event-based precipitation chemistry data distributions were compared to cloud water
chemistry distributions to demonstrate the elevational dependence (in-cloud versus below-cloud) of
exposure.

During summer at Whiteface, under non-plume conditions, SO2(gas) lifetimes are less than one
hour if clouds are present (McLaren et al., 1984). Once an air parcel has been entrained into a cloud,
ambient SO2 is rapidly oxidized by ^2^2 m tne aqueous phase to sulfuric acid; thus, future sulfur

4-1

-------
transport is governed by cloud droplet, rain, and/or aerosol dynamics. Also, direct deposition of sulfur to
the high-elevation canopy by cloud water impaction is substantially greater than it would be if the SC>2
had remained in the gas phase.

       Summertime chemical processes do not apply to wintertime conditions because of differences
between summer and winter temperatures, solar flux, absolute humidity, and meteorology.   To examine
the consequences of these differences, cloud water was collected in winter to create a data base similar
to that obtained during summer. Table 4-1 summarizes the concentrations of the four major cloud ions
observed during January and February of the four winters sampled at Whiteface Mountain, NY; for
comparison, the same ions are shown for June-August.  Samples taken during cloud-only conditions  have
been included.
                                           TABLE 4-1

       Comparison of Winter and Summer Cloud Ion Concentrations at Whiteface Mountain, NY
                                (from Mohnen and Kadlecek, 1989).
Year

1983
1984
1985
1986
                                     January-February (jieq/L)
                   H+
                   SCy2'
median max    median max    median max
129
204
324
191
 589    100
1148   -160
 776    366
1585    168
                       554     54
                      1632     95
                      1100    218
                       972     73
205
680
458
522
                                         NO/
                                     median max
 71
140
250
 86
                                                            #of
                                                            samples
 258    107
 838    156
 631    150
1131    247
days
with
data

  8
 11
 11
 22
                                       June-August
Year

1981
1982
1983
1984
1986
1987
    H+
median max    median max    median max
                                                 NOf       # of
                                             median max    samples
-
141
166
251
145
148
-
1660
3162
977
1413
1778
154
242
224
316
142
194
3562
1224
3050
1202
1318
2224
68
134
130
147
64
113
732
788
855
612
770
920
60
74
69
83
46
68
816
590
1338
285
752
1344
289
171
304
124
162
140
                                                               days
                                                               with
                                                               data

                                                                32
                                                                33
                                                                29
                                                                21
                                                                23
                                                                25
        Ion concentrations varied as much among winters as they did between winter and summer.
Variability within individual seasons was large, with standard deviations typically 50% to more than
100% of the seasonal medians (the concentrations are not normally distributed).  Seasonal medians of
different years varied greatly,  often by a factor of two to three, and particularly in winter.  One reason
for large differences in winter concentrations is that large-scale meteorological conditions, once
established, tend to persist, but they do not necessarily repeat  each winter.  Also, samples were taken in
only part of each winter.  For example, if an event is relatively clean, most samples will have
concentrations of a few tens of jteq/L; if the event is relatively polluted, the concentrations will generally
                                               4-2

-------
be a few hundreds of jteq/L.  If the block of samples obtained tends to emphasize a different set of
transport conditions than the unmonitored cloudy periods, the seasonal statistics will be influenced.

        Sulfate concentrations were only slightly greater in summer than in winter.  While this may
seem inconsistent with the much lower rate of SC>2 transformation  in winter, the few available
measurements indicate that the lower liquid water content of winter clouds (Kadlecek et al., 1988) tends
to concentrate the available species in about half the water, suggesting an average wintertime
atmospheric clear air sulfate loading equal to  about half that of summer.

        Since the rapid oxidation of ambient SC>2 characteristic of summer clouds is not available during
winter, and since atmospheric loadings of nitrate are similar (summer and winter), a shift in the sulfate
to nitrate ratio was expected.  This shift  is apparent  in Table 4-2.


                                            TABLE 4-2

             Comparison of Median Sulfate to  Nitrate Ratios at Whiteface Mountain,  NY.


                       Year                   Summer                Winter

                       1981                     1.3
                       1982                     1.6
                       1983                     1.6                   0.7
                       1984                     1.9                   0.6
                       1985                      -                    0.7
                       1986                     1.5                   1.0
                       1987                     1.4

        Shown in Figures 4-1 and 4-2 are the summer and winter concentration distributions of the
dominant anions in cloud water and precipitation from Whiteface Mt.  Across most seasons, the cloud
water concentrations of the dominant ions ranged over three orders of magnitude. However, individual
cloud events  or substantial periods of events lasting many hours had relatively constant  ion
concentrations.  The distributions  in Figures 4-1 and 4-2 resulted from the mixing of many different
events, each with its own relatively constrained concentration range.

        These summer and winter cloud  data can also be  compared to MAP3S (event-sampled)
precipitation  data from Whiteface Mountain in Figures 4-1 and 4-2. Nearly all of the precipitation
samples in each season had anion concentrations less than 200 neq/L.  During summer for most years,
both precipitation distributions had similar shapes, peaking in the 40 to 100 neq/L range for SOj2'.
During winter, NO/ resembled the summer distribution, although it shifted slightly to higher
concentrations.  The precipitation SO,/2"  distribution was shifted substantially to lower concentrations;
more than 80% of the winter samples had less than 40 /ieq/L, consistent with the reduced oxidation rate
of SO2 in the winter atmosphere.

        In addition to cloud liquid water content, differences between seasonal trajectory patterns suggest
that the winter inter-annual differences probably result in large part from differences in meteorological
flow, emphasizing that average meteorological conditions do not repeat each year (Kadlecek et al., 1988).
For example, trajectories ending at Whiteface  Mountain, NY, during January, 1985 (the month in which
most winter data were taken), showed the flow to be unusually restricted to southwestern sectors,
resulting in relatively more samples with  higher  ion concentrations. Conversely, during  1983, more
trajectories traversed areas with low emission sources, such as the Atlantic Ocean or regions to the


                                                4-3

-------
             120
                     Rain
                                                              Winter
                                                              Summer
                   .51   23456789  10
                   Aqueous Concentration in 10~"4mole»/liter

Figure 4-1  Comparison of both summer (June-Aug) and winter (Jan-Feb)
          SO^" distributions for both cloud and rain water.  Values
          shown are percent of cloud samples at the Whiteface Mountain,
          NY, summit MGCP site and percent of precipitation samples at
          the Whiteface Mountain, NY,  MAP3S site.
             100
           3
           O
           U
           O
              80
              60
              40
              20
           o
           DC.
Clouds
                              Winter

                              Sumrm
                     1   2         5      7.9
                   Aqueous Concentration In 10"* moles/liter
Rgure 4-2  Comparison of both summer  (June-Aug) and winter  (Jan-Feb)  NC^"
          distributions for both cloud and rain water.  Values  shown are
          percent of cloud samples at the Whiteface Mountain, NY,  summit
          MGCP site and percent of precipitation  samples at  the
          Whiteface Mountain, NY, MAP3S site.
                                   4-4

-------
northwest. The mean concentrations of sulfate and nitrate during 1985 were more than twice their 1983
values. This provides some insight into the inherent seasonal variability due to meteorological
conditions such as cloud and precipitation extent, air mass path and speed, and mixed layer depth (Pratt
et al., 1986). The winter seasonal averages were based on individual cloud events, each of which had
concentrations identified with particular meteorological conditions, but which did not respond to
short-term changes in SC>2 or NOy levels since neither has a rapid oxidation rate.

Figures 4-1 and 4-2 depict differences between winter and summer cloud water SO,/2' and NOj"
but these seasonal differences are small relative to differences between precipitation and cloud water
SO/2' and NO/. This suggests that at Whiteface Mountain, NY, the MCCP strategy of sampling only
growing-season cloud water does not introduce uncertainties for the chemical concentrations as large as
uncertainties for the LWC in winter clouds, where and how frequently cloud immersion occurs for both
winter and summer, and prediction of cloud droplet deposition.

Elcvational Variation

The height of cloud water sample collection above cloud base is expected to be an important
source of variability in both cloud water chemistry and liquid water content. Nucleation of pre-existing
aerosol, condensation of diffusing water vapor, and droplet growth occur with vertical motion of clouds
overrunning mountaintops. The Junge (1963) formulation for nucleation scavenging implies that the
concentration of cloud water solute varies with the inverse of liquid water content (LWC). Therefore, if
LWC increases with height above cloud base, the concentrations of any solute with a substantial aerosol
phase precursor should decrease with height above cloud base.

From aircraft cloud water sampling, it has been observed that liquid water content of clouds has
been found to typically increase to a maximum at some level above cloud base and eventually decline
toward the top of clouds, while the chemical concentrations of pollutants dissolved in cloud water usually
are highest near cloud base, and fall off with increasing altitude (Petrenchuk and Drozdova, 1966; Daum
et al., 1984; Romer et al., 1985; Lazrus et al., 1983; ten Brink et al., 1987; Scott and Laulainen, 1979).

The Mountain Cloud Chemistry Project (MCCP) extended the time-restricted aircraft
measurements through observations of entire events at two Whiteface Mountain, NY sites: Whiteface-1
(summit, 1483 m) and Whiteface-2 (Lake Placid turn site, 1245 m). The concentrations of SO/2', NO/,
H+, and NH^+ in 108 coincident-paired hourly cloud water samples have been compared to determine
whether height above cloud base is an important determinant of cloud water acidity. Table 4-3
summarizes the data for SO^, NOj% H+, and NH^+ concentrations for all coincident-paired samples
collected during 1987-88. As can be seen in the table, the mean concentrations for all four ions at the
lower site (WF2) are nearly twice those recorded at the summit (WF1). To statistically compare the ion
concentrations at WF1 and WF2, the ion concentration data were log transformed. It is observed that
the log of each of the ion concentrations approximates a normal distribution, so a t-test can be applied
to the paired differences. Table 4-3 also presents the calculated results for the paired differences of the
transformed ion concentrations. For each ion, the t value is well above the critical value for rejecting
the null hypothesisof equal means with a confidence level of 0.05. Reisinger (1987) analyzed two cloud
events in 1986 and 1987 at Whitetop Mt. for vertical varibility of ion concentrations in cloud water with
similar results.
4-5

-------
                                            TABLE 4-3

            Ion Concentrations Statistics for Coincident-Paired Cloud Water Measurements.
                  Results are for 1987-88 precipitating and non-precipitating clouds.


                WF1           WF2          Number       Aw. of Paired                    t
Ion             Mean          Mean         of Obs          Differences                   Statistic
               lanoW         fsnol/l                         tog(WF2)-k)g(WFl)

H+              122            218            107              +0.300                     -7.0
SCv2'             79            151            108              +0.470                     -9.8
NO/             48            85            108              +0.478                    -14.7
Cl-               3              7            108              +0.745                     -9.6
NH/             74            142            108              +0.593                    -11.4


        Figure 4-3 presents a histogram of the differences in SO^2' concentration for the simultaneous
samples from two Whiteface sites (calculated as [SO4] at WF2 - [SO4] at WF1).  By converting the
original concentrations to differences between the two sites, substantial covariation is removed from  the
results, thereby allowing the elevational effect to emerge. During 92% of the sampled hours, the lower
elevation site receives cloud water with higher concentrations of SOj2' than the summit  (similar results
were obtained for NO/ and H+ differences).  The analysis presented in Figure 4-3 confirmed that both
sites tend to experience relatively "clean" events or tend  to experience  relatively "dirty" events at the
same time, and that the lower site had consistently higher concentrations.

        Histograms for both simultaneous and all other  samples collected at the  two Whiteface sites (the
summit had 8 times more data) suggest that hours with  [SO^2"] greater than 500 
-------
  FREQUENCY  DISTRIBUTION  FOR  1886-1988  SULFATE  DIFFERENCES
                  BETWEEN WHITEFACE  SUBSITES 1  & 2
      INCLUDES PRECIPITATING AND  NON-PRECIPITATING CLOUDS
  cr
  w
  6-

  «
              100 - 0 0-100 100-200 200-300 JOO-400 400-500 500-600 JOO-700 700-W  >SOO
  WF1
  WF2
                     ION  CONCENTRATION  (Mequiv/1)

             ION CONCENTRATIONS  FOR SIMULTANEOUSLY  COLLECTED
           CLOUD WATER SAMPLES  FROM WHITEFACE  MTN.  1987 - 1988
                                (108  SAMPLES)
              FOR  PRECIPITATING  AND NON-PRECIPITATING  CLOUDS
                      I
                          I
                                      \
                                          I
     I
I
I
I
I
                   MEAN VALUES IN MICROEQUIVALENT / LITER
        LWC = 0.46
        LWC = 0.19
          OTHERS!!]
                               SO,
          iNH.4   DOTHERS
                 toot

                  791

                  021

                  SOS

                  381



                  18S- -
       '  I  '  i  '  I  '  I  '' 'I  '  I  '  I  '  I
   -800  -700 -800 -500 -400 -300 -200  -100  0
                               1400

                               1300

                               1200

                               1100

                               1000

                                too

                                aoo
 |  i  |  .  |  .  |  i  |  .  |  ,  |  I
100  200  300  400  SOO  800  700  800
      RIGHT SCALE: CLOUD  BASE ELEVATION(m) AND  CUMULATIVE FREQUENCY
              LEFT  SIDE: LIQUID WATER  CONTENT (LWC) IN (g/m1)
                           SOURCE: MOHNEN (1988)
Figure 4-3  Frequency  of occurrence  for the differences in SO^
          concentration for the  two Whiteface  Mountain, NY,  sites
          (calculated as [SO^] at  WF 2 - [SO,/,]  at WF1) for
          simultaneous, hourly,  samples.
                                         4-7
-------
Whiteface Mountain must exist for other mountain sites in the eastern USA.  However, it seems likely
that it occurs in many locations based on the available vertically resolved aircraft and surface
observations of cloud water chemistry, cloud nucleation and scavenging processes.

Data Summary

        In this section, the 1986-88 results for the five high elevation MCCP sites are summarized.  The
overall chemical composition of the cloud water samples is presented, along with differences between
precipitating and non-precipitating clouds.  Extreme cloud water solute concentrations are described.
Statistical modeling indicated a significant (P < .06) site-to-site variation in SO/2' and  NOy
concentrations; thus the means and their frequency distributions are examined to identify any spatial
patterns in the data.

        Figure 4-4 presents the unweighted mean concentrations of H+, SO/2', NOj", CT ("other"
anion), and NH/+, and "other" cations (sum of Ca2+, Mg2+, K+, and Na+) for all MCCP cloud
samples taken during the 1986-88 growing seasons (generally May to October).  The samples were
primarily sulfuric acid with substantial amounts of the two nitrogen species but relatively low
concentrations of other anions and cations.   The samples were in approximate charge balance, suggesting
that the major ionic species have been measured on the average.

        Figure 4-5 classifies the data as precipitating and non- precipitating clouds.  The concentrations
in non-precipitating clouds were much higher than in  precipitating clouds, as was noted by other
investigators (Falconer and Falconer, 1980;  Schemenauer, 1986; 1988).

        It is suggested  that MCCP measurements may overestimate liquid water content (LWC) for
precipitating clouds.  Falconer and Falconer (1980), Schemenauer (1986; 1988), and Mohnen and
Kadlecek (1989) suggest that most cloud water samplers do not completely exclude rain drops (especially
the ASRC sampler). Therefore, at this time, neither LWC nor cloud water chemical concentrations have
been accurately characterized for precipitating clouds.

        Non-precipitating cloud data are probably more important than  precipitating cloud data for
determining forest canopy exposure and cloud water deposition because  they represent  chemical inputs
during a time  period when the precipitation is not occurring. However, substantially increased
concentrations in precipitating cloud water over precipitation (as collected under the NADP/NTN
protocol), coupled with increased water deposition due to cloud interception by the forest canopy,
suggest  that precipitating clouds contribute  to total deposition in as yet  undefined quantities.
Non-precipitating clouds must be handled separately in cloud water deposition calculations.

        Figures  4-6 through 4-8 present the distributions of SO/2" ion for non-precipitating cloud water
at six sites (both Whiteface summit and Whiteface-2 are included).  A substantial number of hours at
each site (15% to  40%) had cloud water H+ and SO/2" concentrations greater than 500 jteq/L, which we
define as an "extreme"  value.  Distributions for H+, NO/, and NH/+ are similar in shape to those for
SO/2", but are dispersed around different mean values as given in Table 4-4.  The "extreme" SO/2"
values were evenly distributed between concentrations of 500 to lOOOeq/L. "Fjctreme" NO/ and NH/+
are defined as those greater than 250 neq/L; these values were evenly distributed between concentrations
of 250 to 500  neq/L. Mohnen (1988c) presented the distributions of all  ions.

        Sulfate is a very important ion in the determination of cloud water acidity.  For example, the
"extreme" concentrations of SO/2"  alone in non-precipitating clouds (500 jteq/L) would  support a pH of
3.3; this type of calculation defines the "potential acidity" that would occur if no other  cations or anions
were present in  the cloud water solution. The near-maximum value of 1000 iieq/L [SO/] observed at
most  MCCP sites translates to a potential acidity of pH 3.0.  Cloud water SO/^" concentrations over


                                                4-8
-------
             MCCP CLOUD  CHEMISTRY CONCENTRATIONS
                MEAN  VALUES IN  MICROEQUIVALENT / LITER
       FOR PRECIPITATING AND NON-PRECIPITATING CLOUDS (1068-88)
WF1
WF2
MSI
SHI
WT1
MM1
                         I
        OTHERSD  NO,'!
                          so.M
                                     I
          I
    T
I1 T
IB'
INH.'  DOTHERS
-tOO -800 -700 -000 -SOO -400 -SOO -200 -100  0

                           (ALL SAMPLES)
                                       i T i  •  i  ' i '  i^  i^ r^i^
                                      100 200  300  400 SOO  600 700  100 100
             SITE         SITE LOCATION

             WF1     Whiteface  Mountain. NY

             WF2     Whiteface  Mountain, NY

             MSI     Mount Moosilauke, NH

             SHI     Shenandoah, VA

             WT1     Wbitetop Mountain, VA

             MM1     Mount Mitchell, NC
                                           ELEVATION (m)

                                               1463

                                               1245

                                               1000

                                               1015

                                               1689

                                               1950
                                          2-
Figure 4-4  Mean concentrations of H  ,  SO^  ,  N03 , other anions  (Cl )
          NH4  ,  and other cations  (sum of Ca2+, Mg2+, K+, and Na+) for
          all  MCCP cloud samples for  1986-88.

                                  4-9
-------
             MCCP CLOUD  CHEMISTRY CONCENTRATIONS
                MEAN VALUES IN MICROEQUIVALENT  / LITER
                  FOR PRECIPITATING CLOUDS  (1986-88)
WF1
WF3
MSI
SHI
WT1
MM1
        OTHERSQ  NO/I
SO/
INH/  DOTHERS
     ' i ^i ^i1  i^  i '  i  ' r'  i  ' i  '  i  ' i  '  i  ' i  '  i  ' r ^r^ i  '
  -900 -800 -700 -600 -500 -400 -300 -ZOO -100 0  100 200  300 400  500 100  700 100 »00
                            (ALL SAMPLES)
                                        r\
Figure 4-5 Mean  concentrations of H+, 804  , N03 , other anions  (Cl~),
         NH4+,  and  other cations stratified by precipitating and
         non-precipitating cloud type, for all MCCP samples, 1986-88.
                                4-10
-------
Figure 4-5  (Continued)
                  MCCP  CLOUD CHEMISTRY  CONCENTRATIONS

                     MEAN VALUES  IN MICROEQUIVALENT / LITER
                     FOR NON-PRECIPITATING CLOUDS (1088-66)
    WF1
    wra
    MSI
     SRI
    WT1
    MM1
             OTHERSD   NO,'!
SO/I
      INH/   DOTHERS
      -tOO -800 -TOO -tOO -900 -400 -300 -ZOO -100  0   100 tOO  JOO 400  600 (00 700  SOO 100
                                  (ALL SAMPLES)
                    MCCP CLOUD  CHEMISTRY CONCENTRATIONS

                      MEAN  VALUES IN MICROEQUIVALENT / LITER
                      FOR NON-PRECIPITATING CLOUDS (1987-88)
      WF1
      MSI
      WT1
      MM1
              OTHERS
                         NO.'
  SO.'
H4
NH."
                                                               OTHERS
           i  |  .  |  T  |  i  |  i  j i  |  i -|  r—j-
         800 -700 -tOO -SOO -400 -300 -200 -100   0
             I  '  1  '  I  '  I  '  I  '  I  '  I
            100  200  300  400  SOO  600  700
                      Upper  b»r: LWC weighted mem
                      Lower  bar: Unweighted mean for  same samples
                                           4-11
-------
SE
w
S3
cr
w
a:
Cb
               WHITEFACE MT.  SUMMIT  --  1986-1988


       SULFATK CONCENTRATION FOR  NON-PRECIPITATING CLOUDS
            0-50  50-110 100-150 150-200 200-250 250-JOO JOO-J50 J50-<00 «00-UO 450-500 >500



                   ION  CONCENTRATION (^equiv/1)
 o
 z
 w
 O"
 W
              WHITEFACE  MT. SUBSITE  2 --  1986-1988


        SULFATE CONCENTRATION FOR  NON-PRECIPITATING CLOUDS
              0-5t  50-100 100-150 150-200 200-259 250-:00 JOO-350 J50-UO 400-450 450-500  )500



                    ION  CONCENTRATION (^equiv/1)
 Figure 4-6  Frequency of occurrence for hourly  80^2   ion ±n non-

           precipitating cloud water at Whiteface-1  (summit) and

           Whiteface-2, NY.


                                       4-12
-------
or
u
05
b.
               WHITETOP  MT.  SUMMIT  --  1986-1988


       SULFATE CONCENTRATION  FOR NON-PRECIPITATING CLOUDS
            fl-50  51-100 100-150 150-200 200-250 250-300 300-350 nO-UO<00-no 450-500  >50)



                   ION  CONCENTRATION  (/zequiv/1)
 u
 tL-



 M
               MT.  MOOSILAUKE  SUMMIT  --  1986-1988


        SULFATE  CONCENTRATION FOR NON-PRECIPITATING  CLOUDS
             0-50 50-100 100-150 150-200 200-250 250-300 300-350 350-UO(«0-<5« 450-500  >5»l



                    ION  CONCENTRATION  (/zequiv/1)
 Figure 4-7  Frequency of occurrence  for hourly  SO^    ion in non-

           precipitating cloud water at Whitetop,  VA, and Moosilauke,

           NH.


                                   4-13
-------
u
z:
w
»
o-
w
                SHENANDOAH SUMMIT  --  1986-1988

       SULFATE  CONCENTRATION  FOR  NON-PRECIPITATING CLOUDS
       50
45 -

40 -

35 -

30 -

25 -
       20 -      _





            [•••••••••I
            0-50  51-100 100-150 150-200 200-250 250-300 JOO-J50 HO-W 4«0-<50 450-500 >50I
                   ION CONCENTRATION (/xequiv/1)
  •z.
  w
  E>
  o-
  td
  o:
                 MT.  MITCHELL  SUMMIT  --  1986-1988


        SULFATE CONCENTRATION  FOR NON-PRECIPITATING  CLOUDS
              0-50  50-100 100-150 ISO-200 200-250 250-900 300-»0 150-400 <«0-450 451-510  >500



                    ION CONCENTRATION  (/xequiv/1)
                                                   _
Figure 4-8  Frequency of occurrence for  hourly SO^   ion  in non-

          precipitating cloud water  at Shenandoah, VA and Mt.

          Mitchell, NC.



                                 4-14
-------
2000 jieq/L occur, but rarely.  A few hourly cloud water samples of about pH 2.8 were observed at some
MCCP sites during most warm season sampling campaigns.  Cloud water with pH of 3.0 was routinely
sampled at the MCCP sites.

       The "extreme" concentrations for H+, SO,/2', NO/, and NH^+ were observed to occur most
frequently at the beginning of a cloud event when LWC was low or at the end of some events where
LWC was rapidly decreasing; an increase in the sulfate-to-nitrate ratio sometimes  co-occurred with this
decrease  in LWC.  This is consistent with the evaporation of liquid water, concentration of cloud water
solutes, and vdlitization of HNOj during the dissipating cloud stage. This qualitatively supports
calculations by Unsworth (1984) and Milne et al. (1988) which suggest that evaporation on both the
sampler and the forest canopy can lead to  concentrations that are enhanced over what is advected to  the
canopy.


                                           TABLE 4-4

       Average Ion Concentrations (jimol/1) for Both Precipitating and Non-Precipitating  Clouds
                                          for 1986-1988
                        at Six High-Elevation Locations Monitored by MCCP.

                                                                                            #of
   Site               H+            SOj2-         NOy          Or            NH^+    hours
Mitchell, NC           398            489           174             31            184        624
Moosilauke, NH*       263            257           132             16            107        328
Shenandoah, VA       171            176            94             13             93        230
Whiteface-1, NY        171            205            73              5             97        987
Whiteface-2, NY*       255            352            92              7            157        120
Whitetop, VA          174            321           144             16            152        656

*  Did not collect cloud water during 1986
** See Figures 4-6 through 4-8 for the sulfate ion concentration distributions in non-precipitating clouds

        Chloride (presented in Table 4-4) and sodium concentrations (not presented) suggest that the
seasalt contributions to the sulfate means are minimal.  The MCCP cloud water data are well
approximated by considering only the four major ions.  The sum of cation concentrations is useful
primarily for identifying neutralization of H+ in individual samples by crustal-like material.

        It is difficult to  extrapolate from these overall concentration means and distributions to other
high elevation forests in the eastern USA  The  two southern-most sites,  Mt. Mitchell, NC, and
Whitetop, VA, have the largest mean H+, SO^2', NOj", and NH^+ concentrations and the greatest
frequency of extreTme values (defined here as > 500 /icq/L) (greater than  30% of hours), followed  by
Moosilauke, NH, with intermediate values for mean concentrations and 25% of samples representing
extreme values.  The Whiteface summit (site 1), NY, and Shenandoah, VA had the lowest means and
less than 15% of samples constituting extreme H+ and SOj2' concentrations.

        The Whiteface-1, NY and Shenandoah, VA sites display unusual  characteristics when LWC and
cloud immersion frequency, respectively,  are compared to those of the other MCCP sites.  Therefore
site-to-site differences among the remaining MCCP sites are considered.   When the data for the
Whiteface-2, NY (1250 m) site are compared to those for  the site at Moosilauke, NH, a picture of cloud
water chemistry emerges that is consistent with the results of the Canadian CHEF program
(Schemenauer, 1986; 1988).  When compared to the northern Appalachian sites,  the MCCP data for


                                               4-15
-------
Whitetop, VA, and Mt. Mitchell, NC, and IPS data for a site in the Great Smoky Mountain Park, NC
(Lindberg et al., 1988) suggest higher mean H+, SO/2', NO/, and NH/+ cloud water concentrations for
southern Appalachian sites.

        The dependence of cloud water concentrations on synoptic weather type and observed differences
in the relative frequencies of cap cloud, warm sector, marine flow, and post-cold frontal synoptic types
between northern and southern MCCP sites suggest  that the north/south differences in cloud water
concentrations may be related to cloud climatology.  The southern sites frequently are cloudy under the
stable, warm-sector synoptic type (high concentrations), while the northern sites experience cloudiness
associated more frequently with frontal passage (lower concentrations). Additionally, when air mass
trajectories shift from southwest to northeast at the northern sites, the concentrations of H+, SO/2',
NO/, and NH/+ normally decrease while the southern  sites receive high concentrations under northwest
flow.
FREQUENCY OF CLOUD

        The frequency of cloud immersion, the mass concentration of liquid water (LWC), and its
distribution across droplet sizes are important for understanding chemical fluxes to a forest canopy.
Collectively, these data describe the presence of liquid water in the atmosphere.

        High-elevation forests of the Appalachian Mountains are often immersed  in cloud.  Previous
estimates of cloud impaction frequency on mountain summits in the southern Appalachians range from
30% of all hours  in the Great  Smokey Mountains to 40% at Mt. Mitchell (USFS, Northeastern Forest
Experimental Station, personal communication). Estimates for the northern Appalachians are 50% at
Whiteface (Nicholson and Scott, 1969), 55% at Mt. Washington (NOAA, 1975) and 40% at the summit
of Mt. Moosilauke (Lovett et al., 1982).  Siccama (1974) estimated that the upper slopes of the Green
Mountains are in clouds 30% to 50%  of all hours.  Many of these estimates were derived from visual
observations as a  biproduct of  forest ecosystem research and do not represent actual cloud
measurements.  The following material addresses recent regional and mountaintop studies dedicated
specifically to cloud.

Regional Cloud Studies

        Warren et al.(1986) developed global cloud climatology statistics from surface airport observa-
tions  for the 1971-1981 period on a 5  deg. latitude by 5 deg. longitude grid. Over the eastern U.S., the
low cloud types (<2500 m) in  order of decreasing average cloud amount are: stratus/stratocumulus,
cumulus and cumulonimbus. The combined mean sky cover (Figure 4-9 - Appendix E) for the three low
cloud types ranges geographically from 24% to 46%, with the highest  values found over the northern
portions of the eastern U.S.  The stratus/stratocumulus type is strongly dominant, accounting  for at least
75% of the overall low cloud amount.  The average cloud base height for this class  (Figure 4-10 -
Appendix E) is generally in the range  of 700-900 m. Cumulus clouds have higher  average bases, typically
between 1000 m and 1200 m; cumulonimbus bases average in the 800-1000 m range.

        A 21-year study (1965-1985) of low cloud observations  from 12  eastern U.S. airports located near
five of the MCCP sites found that the incidence of low cloud (<2134 m) exceeded 40% at all locations
with the highest values in the north (Mohnen,  1988). In the vicinities  of the southern MCCP sites  of
Whitetop, Shenandoah, and Mitchell, low clouds were reported 42% to  49% of the  time.  In  the
northern Appalachians near Whiteface and Moosilauke, the occurrence  ranged from 44% to 63%.  Cloud
base height at all locations was most frequent in the 900 m to  1200 m range.
                                               4-16
-------
       The U.S. Air Force Real-Time Nephanalysis (RTNEPH) global cloud database for low clouds
(<2.1 km) was examined for the 1985-87 period for a large portion of the Appalachian Mountain region
(southwest Virginia north to southeastern Canada). The trend over this period was for decreasing cloud
frequencies,  consistent with the expanding drought in the  southeastern U.S.  Figure 4-11  (Appendix
E)illustrates this trend and shows the dominance of stratiform cloud types (in agreement with Warren et
al., 1986). Cloud base height was found to average between 700 m and 900 m. After taking into account
cloud thickness as well as cloud base, the mean probability of low cloud as a function of elevation was
found to be  greatest between 900 m and 1300 m  (Figure 4-12 - Appendix E).  Approximately 30% of
low clouds were precipitating:  19% in summer, 42% in  winter.

       An examination of the individual RTNEPH grid cells nearest three MCCP sites indicates local
variability.  Near Moosilauke (Figure 4-13 - Appendix E), cloud is most likely to occur between 1000 m
and 1300 m.  Near Whiteface (Figure 4-15 - Appendix E), highest cloud probabilities are within the
broad range of 800-1600 m. Near Shenandoah (Figure  4-14 - Appendix E), cloud probability is lower
than at either Moosilauke or Whiteface but is relatively strong between 800 m and 1300  m and strongest
between  1000 m and 1100 m.  It should be kept in mind that these cloud height values apply to
relatively large-scale cloud elements in  the general vicinity of the referenced mountains and are not
necessarily representative of site-specific cloud elevations.

       The continuity of cloud heights across the Appalachians is shown in Figure 4-16.  This figure
presents  a cross-section  of a west-east transect of cloud probability (according to the RTNEPH database)
across the northern Appalachians in summer (June-August, 1985-87 averages) at a latitude of approxima-
tely 44.5  degrees N.  The cross-section  represents a height of 0 to 2 km and a width of 70 deg. W to 78
deg. W longitude. The diagram indicates a zone  of maximum cloudiness at a height of 800-1000 m from
southern Ontario province across the northern Adirondacks into the Green Mountains of Vermont.  The
zone lifts to near 1200 m across New Hamphsire and to near 1600 m over southwestern Maine. A
second transect (not shown) approximately 85 km south of the first one spanning the central Adiron-
dacks and the Mt. Moosilauke area, exhibits the same trend across New York, Vermont and New
Hampshire.  This suggests that the preferred zone of cloudiness  over the high mountain regions of the
northern Appalachians  is between 800 m and 1200  m with a tendency for clouds in the lower portion of
the range over the Adirondacks and Green Mountains.  In the central  and southern Appalachians, the
results of a single transect analysis across eastern West  Virginia  through the Shenandoahs of Virginia
suggest a slightly higher zone of preferred cloudiness-1000 m to 1400 in-over the high mountains of
West  Virginia, but a lowering of the  zone into the 700-1000 m range east of the Appalachian's major
axis.
Site Specific Measurements

        Table 4-5 presents the percent frequency of cloud impaction for the combined 1986-88 field
seasons for the five MCCP summit sites.  The top half of the table indicates that cloud frequencies range
from 11% to 37% of all hours during the summer season among the sites. Due to the differences in
station elevation,  latitude, and local topography, it is not possible to discern spatial trends in these
results.  Temporally, the general decline in cloudiness  from  1986 to 1988 is reflective  of the drought
conditions which prevailed over the southeastern U.S.  and the shift from cool, wet weather in 1986 in
the north to drought conditions in 1988.  Approximately 25% of the cloud events were precipitating.
The bottom half of the chart indicates that, on a daily basis, clouds occur frequently: 32%-42% of the
days experienced cloud at the two lowest elevation sites, and 68% to 77% of the days on the three
higher  mountains.

        In an attempt to put the cloudiness of the MCCP field seasons into the perspective of long-term
climate, the frequency of low-level (<2100 m) clouds was determined for eight local National Weather


                                               4-17
-------
 oo
 CJ
  I
 T3
n
                                                                        oo
                                                                        oo
                                                                         3
                                                                        00
                                                                        r-
                                                                        CO
                                                                         ex,
                                                                        00
                                                                        r-
                                                                        oo
                                                                        oo
VO
00  pr
 3  £
oo  O
    oo

^  &j
00  W
 ex, oo
00
                                                                        oo
                                                                        £
                                                                        00
                                                                        UH
                                                                        00
                                                                         p
                                                                        CO
                                                                        00
                                                                         ex
                                                                        00
                                                                        00
                                               H
                                       4-18
-------
         Summer (1985-87)  Probability of Cloud Occurrence (RTNEPH data)
         for the West-East Transect Across the Northern Appalachians at
         Approx. 44.5  deg. N Latitude.  Cross-section Represents a Height
         of 0 to 2  km  and  a Width of 70 deg. W to 78 deg. W.
78 W
                       f          t        t
U.S.-Canada     Whiteface  Mtn.         Mt.  Washington
   Border                   Mt  Mansfield
70 W
         West-East Transect and Land Areas
              Above 800 m in Elevation
                 78 W
                               Figure 4-16

                                       4-19
-------
Service airports for the same time period and compared with a 21-year climatology (1965-85) previously
developed for these airports.  The results are presented in Figures 4-17 through 4-19 (Appendix E)
which show the percentage departures from normal for each of the three field seasons.  In 1986, positive
anomalies occurred in the Northeast, near normal conditions existed in the central Appalachians and
below normal cloudiness occurred in the southern Appalachians. During the 1987 field season, above
normal cloudiness was again observed in the Northeast but values generally were not as high as those in
1986. Below normal cloudiness was observed throughout the central and southern Appalachians.  In
1988 the dryness spread into the Northeast where lower than normal cloud amounts were observed,
especially during the  early portions of the field year. In southern regions, cloud frequencies were once
again below normal with the exception of one airport (Asheville, NC)  which had a normal frequency.
These airport trends  are consistent with the 3-year trends observed among the MCCP sites. This finding
suggests that the northern MCCP sites experienced  above-normal cloudiness in both 1986 and 1987 and
below-normal cloudiness in  1988.  It also suggests that the southern sites experienced below-normal
cloudiness during all  three field seasons. What is not known, however, is the impact that large scale
trends in cloudiness have on localized, orographically-induced clouds.


                                           TABLE 4-5

                    Cloud Frequency at MCCP Sites, June to  September, 1986-88.

                  Site             Elevation (m)        1986   1987   1988   Mean

                               Cloud frequency (% of hours in cloud)

               Whiteface, NY          1483              45     40     25      37
               Moosilauke, NH         1000              25     21      6      19
               Shenandoah, VA        1015              18      7      6      11
               Whitetop, VA           1689              38     28     26      30
               Mitchell, NC            1950              35     28     23      29

                          Percentage of total days experiencing some cloud

               Whiteface, NY          1483              79     80     73      77
               Moosilauke, NH         1000              52     51     22      42
               Shenandoah, VA        1015              46     25     25      32
               Whitetop, VA           1689              76     67     62      68
               Mitchell, NC            1950              84     75     68      76


        Cloud base observations taken at Whiteface and Whitetop  (Figures 4-20 and 4-21 - Appendix E)
indicate that the frequency of cloud at approximately the 1000 m elevation level is about  10% at
Whitetop and  15% at Whiteface.  Taken together with the direct station measurements at Shenandoah
and Moosilauke which  are both at or near the  1000 m level, a north-south trend in  cloud frequency
becomes apparent: 10-11% frequency at approx. 1000 m in the south and 15-19% frequency in the north.

        Diurnal cloud activity indicates a preference for cloud impaction at night (7 pm - 7 am) on
mountaintops, as previous MCCP reports have illustrated (Mohnen, 1988c). Some sites experience more
than twice as much cloud during the early  morning than the afternoon.  Moosilauke and Shenandoah did
not exhibit such a diurnal trend, probably due to their lower elevations and their relative  position
downwind from other mountains.  In contrast to the mountaintop trends, regional airports and the
                                               4-20
-------
RTNEPH database exhibited more cloudiness during daylight hours. This suggests that mountain
cloudiness at night is localized and probably linked to orographic mechanisms.


Land Areas Susceptible to Cloudwater Deposition

        There is general agreement from regional and site-specific cloud studies in the eastern U.S. that
elevations above 800 m are frequently exposed to cloud.  Based on this agreement, the location and area
of land above 800 m was determined. Figure 4-22 depicts these land areas which  total 25 million
hectares.  Table 4-6 lists total land area for 14 states above five elevations: 800 m, 1000 m,  1200 m, 1400
m and 1600 m.
                                            TABLE 4-6

                      State Totals for Land Area (hectares x 1000) Above 800 m,
                                        In 200 m Increments
                       State
800-    1000-   1200-
1000m  1200m  1400m
GA
KY
MA
MD
ME
NC
NH
NY
PA
SC
TN
VA
VT
WV
883
216
7
477
548
570
718
1091
255
75
1215
5670
420
5794
149
50
0
0
74
346
210
175
0
*
698
356
30
2586
9
1
0
0
9
151
56
33
0
0
238
154
0
541
1400-
1600m

  0
  0
  0
  0
  *
 51
 20
  *
  0
  0
 79
 33
  0
  1
> 1600m    Total
0
0
0
0
0
15
3
0
0
0
56
2
0
0
1041
267
7
477
631
1133
1007
1299
255
75
2285
7214
451
8922
        * less than 500 hectares
        Land areas based on Defense Mapping Agency digital elevation data.  Elevation values for every
        30 seconds of latitude and longitude.

        Land area calculations by K. Hermann of NSI Technology Services Corporation.


LIQUID WATER CONTENT

Droplet Size

        An important parameter in cloud droplet deposition modeling is droplet size because it affects
the collection efficiency of the canopy. At any given time, a range of droplet sizes will  exist in a cloud.
The diameter of a particular droplet depends on its time history.  Factors affecting droplet size include
the size, composition, and air concentration of cloud condensation nuclei (CCN), the length of time
droplets are exposed to supersaturated air, the degree of supersaturation and, in "supercooled" clouds,
                                               4-21
-------
Land Areas Above Mean Cloud Base (>800m)
in the Eastern U.S. Susceptible to Acid
      Cloud Droplet Deposition
               Figure 4-22
                     4-22
-------
the concentration of ice crystals.  Entrainment of dry air can affect droplet sizes because it lowers the
saturation ratio and introduces additional unactivated CCN into the cloud.  The coalescence of droplets
of different sizes is generally not important for non-precipitating clouds overriding mountaintops,
although precipitation can reduce cloud droplet concentrations appreciably.  These and other processes
influencing droplet size are described in detail by Byers (1965), Mason (1971), and Rogers (1976), among
others.  Droplet size distributions are highly variable in time and space.

        In continental clouds, droplet concentrations tend  to be higher, the mean droplet diameter tends
to be smaller, and the size spectrum tends to be narrower  than in marine clouds  (Mason, 1971).  A
survey of cloud droplet size data by Mason (1971) showed  that the modal diameter of the droplet size
distribution was usually 5 to 15 /an  for many non-precipitating continental clouds and  15 to 30 /on in
non-precipitating marine clouds, with the difference due mainly to the availability of CCN.

        Droplet size spectrum in clouds impacting elevated terrain have been measured using a Forward
Scatter Spectrometer Probe (FSSP)  at three MCCP  sites:   Mt. Mitchell, NC (1950 m), Whiteface
Mountain, NY (1483 m), and Whitetop Mountain, VA (1689  m). Clouds impacting the sites form in
airmasses that are mostly continental in nature and  are characterized  by relatively high CCN
concentrations. Data have been collected in precipitating and non-precipitating frontal and cap clouds.
DeFelice (1988) presents cloud droplet spectrum for the Mt. Mitchell, NC, site.  Table 4-7 summarizes
results of the FSSP measurements at Whitetop Mountain,  VA.  A description of the FSSP instrument
and its measurement uncertainties is given in Section 6 of  this report.


                                            TABLE 4-7

               Summary of Cloud Droplet Size Distributions at Whitetop Mountain, VA.
 Data were collected during five mainly post-frontal cloud  events in October, 1987 (Valente et.al., 1989).

                                                    Mean              Std  Deviation

               Number of observations                903
               Averaging time (min)                    5
               Liquid water  content (g/m5)fl           0.21                 0.1
               Volume mean diameter (/tmol/1)        11.9                 4.
               Number concentration  (#/cm3)         388                221.

a Measured with FSSP
        A joint intensive effort was made by Rudi Pueschell's group at NOAA-ERL and SUNY-Albany
ASRC during the summer of 1982 at Whiteface Mountain, NY to document in detail the aerosol/cloud
spectrum prior, during, and after cloud occurrence.  LWC was calculated by integrating over individual
size distributions (Mohnen, 1987) and size distributions were lumped together into four categories
(Figure 4-23  - Appendix E).  These were used by MCCP as input for the cloud water deposition model
(described later in this report).

        For each case, a plot of the number distribution and a statistical summary for each mode of the
cloud droplet distribution (gamma distribution, one to three modes) were provided.  One goal was to
establish a relationship between liquid water content of clouds and cloud droplet size distribution.  If
cloud droplets below 2 uta radius  are disregarded, the remaining cloud spectrum shows a distinct trend
with liquid water content:  total cloud droplet number concentration increases slightly and the slope of
the size distribution towards large particles becomes less steep; that is, it "opens" towards larger cloud


                                               4-23
-------
droplets.  It therefore appears possible that a set of "standardized cloud spectrum" could be defined that
correspond to liquid water content classes, such as < 0.05, 0.05-0.1, 0.1-0.25, 0.25-0.40, 0.4-0.6, and > 0.6
g/m3.

        MCCP's approach was then to tentatively assign a cloud droplet spectrum to a particular range
of liquid water content and test and refine this relationship at Whitetop during the  field intensives of
October 1987 and June 1988.  The outcome of these combined efforts was an MCCP-wide
recommendation for droplet spectrum to be incorporated in the modified Lovett model (Mueller, 1990)
for estimation of cloud interception/deposition. The initial cloud droplet size spectrum with assigned
LWC values is presented in Figure 4-23 (Appendix E).  The 1982 Whiteface Mountain LWC data were
in general agreement with LWC values obtained directly during the 1987-88 field  seasons using the
MCCP Valente device.  The results of this investigation suggest that the MCCP approach should provide
satisfactory cloud deposition estimates from a model that inputs measured LWC values and
"standardized" cloud droplet spectrum, with cloud droplet spectrum from Whiteface  Mountain and
Whitetop Mountain field experiments.  The sensitivity of these conclusions  to the known variability of
drop size distribution (Paluch, 1988; Fuzzi et al, 1983; Pruppacher and Klett, 1980) was not calculated.

        Information on the liquid water content in clouds impacting the five mountain summits is
summarized in Table 4-8 for the growing season (April-October).  The MCCP filter technique described
by Valente et al.  (1989) was used.  All data  represent hourly sampling intervals with continuous cloud
(visibility less than  1 km).  There are differences across sites; the difference between Whiteface summit
and all other sites is one of the most striking results, and can be explained on the basis of height above
cloud base of the sampling site.


                                            TABLE 4-8

              MCCP Hourly Averaged Cloud Liquid Water Content Summary for 1987-88.
                                       All values are in g/m5.

                   Whiteface      Mitchell        Whitetop       Moosilauke     Shenandoah
                    19871988      198719JB      19871988      1987 19j»      1987 1988

Mean              0.39  0.44      0.28 0.30       0.20  0.21      0.18  0.22       0.16  0.22
Std.  Deviation     0.22  0.44      0.21 0.21       0.12  0.10      0.13  0.12       0.10  0.16
Median            0.36  0.36      0.22 0.27       0.19  0.22      0.16  0.19       0.13  0.17
# Observations    96    103      77   292       372  245       250   124       17   57
METEOROLOGY

Field Year Summaries

        Meteorological data are collected to aid in the interpretation of the cloud water chemical data,
as inputs for cloud water and dry deposition models, and to document the climatic conditions at the
MCCP sites. The following parameters were recorded on an hourly basis during the 1986-88 field
seasons:  temperature, relative humidity, precipitation, wind speed and direction, total solar radiation
(horizontal surface) and atmospheric station pressure.  Summaries of these parameters for  each subsite
are provided in Appendix C.
                                                4-24
-------
Climate Perspective

        This section discusses the temperature and precipitation climate of the Appalachian region
during the course of the MCCP field program (1986-88) by examining weather records for several  low
and high elevation weather stations in the proximity of the MCCP sites. Most of the MCCP sites were
established in 1986 and therefore cannot be used to address long-term trends.  One assumption made is
that if a number of regional stations surrounding an MCCP site show similar climatic trends, then that
MCCP site likely experienced the same trend but not  necessarily of the same magnitude. High
correlation coefficients (0.80 - 0.98) derived from daily temperature comparisons between MCCP and
regional weather stations for the 1986-88 period support this assumption.

        Three separate studies were used to evaluate the representativeness of the 1986-8 field seasons.
In one study 18 existing first-order and cooperative observer weather stations (Table 4-9) were used in
the analysis. They were divided into nine northern and nine southern stations, together with station
elevation above sea level and period of record.  Six of these stations were at elevations comparable to
the MCCP sites.  Approximately 40 years of records are available for most sites.
                                           TABLE 4-9
                               Selected Long-Term Climate Stations
        Northern Appalachians

     SITE

Bangor.ME
Jackman.ME
Orono,ME
Concord,NH
Mt. Washington,NH
Pinkham Notch.NH
Burlington, VT
Mt. Mansfield,VT
Lake Placid,NY

Southern Appalachians

Big Meadows.VA
Burkes Garden,VA
Luray.VA
Elkins.WV
Bristol.TN
Asheville,NC
Banner Elk,NC
Celo,NC
Grandfather Mt.,NC
ELEVATION(M)   LENGTH OF RECORD
      58
     360
      18
     104
    1908
     613
     101
    1207
     573
    1079
    1006
     366
     594
     459
     652
    1143
     817
    1615
1953-88
1948-88
1948-88
1948-88
1948-88
1948-88
1948-88
1954-88
1948-88
1948-88
1948-88
1948-88
1948-88
1948-88
1948-88
1948-88
1948-88
1955-88
    The climate of the 1986-88 period was compared to the long-term climatologies by deriving monthly
deviations from the long-term normals for both temperature and precipitation. Tables 4-10 through 4-15
present results in terms of above- and below-normal deviations, defined as follows. A plus(minus)
                                              4-25
-------
indicates the month was above(below) the long term mean (LTM). A +A(-B) designates that the month
deviated by at least one standard deviation above(below) the LTM and can be described as an
above(below) normal month.  A +MA(-MB) indicates the month  registered at least two standard
deviations above(below) the LTM and can be classified as much above(below) normal. An N indicates a
normal month (deviation = 0.0). A 1, 2, or 3 following the +/- designation is used to show that the
month ranked first, second, or third, respectively, over the length of record (usually the past 40 years).
For example, a -MBS would be used to represent a month whose  mean temperature or precipitation was
at least two standard deviations below the LTM and ranked as the third most below normal month for
that site.  The discussion below of each field season's climate incorporates the results from these tables.

       In a second study, the climatology of the upper atmosphere as measured by daily National
Weather Service rawinsondes  (OOZ) was determined for six eastern U.S. cities for the  1986-88 field
seasons and compared to an available 29  year (1957 -  1985) climatology.  Figure 4-24 depicts field
season departures from normal for average temperature and pressure height for the 850 mb level which
is, at an average height of approximately  1500 m above sea level, comparable to the height of several
MCCP summit sites. Temperatures in the northern Appalachians were near or below normal in  1986,
near normal in 1987 and above normal in 1988.  In the south, seasonal temperatures  were above normal
for all three years. The 850 mb pressure heights were near or below normal at the northern-most sites
throughout the three seasons.  In the  south, the predominance of a high pressure  ridge is reflected by
near and above normal height.

       A third study, illustrated by Figures 4-25 through 4-27 (Appendix E) shows the mid-summer
drought severity throughout the eastern U.S. for the three field seasons according  to the Palmer Drought
Severity Index (Palmer,  1965).  This analysis, generated bimonthly by the NOAA/USDA Joint
Agricultural Weather Facility, depicts  areas experiencing prolonged and abnormal  dryness or wetness.
This slowly changing index reflects long-term moisture runoff, recharge, deep percolation and
evapotranspiration conditions for relatively low elevation land areas.  It is not necessarily representative
of high elevations which can experience added moisture inputs from direct cloudwater deposition.
Nonetheless,  this index is indicative of large-scale weather conditions that influence temperature and
moisture conditions at all elevations.  The figures are  generally reflective of the three field seasons as a
whole and are discussed in greater detail  below.
                                                4-26
-------
            Average 850 mb Temperature (°C) and  Pressure Height (m)
        Departures  from  Normal  (1957-1985  period) for  the  1986-88 MGCP
             Field Seasons (June-September) for Six Eastern U.S. Cities.
                                     PORTLAND, ME
                                     + 1 ••
                  PITTSBURGH, PA
                    + 1 -.
                 TEMP
                                                      GREENSBORO, NC
                                                        + 1 •
                                                     TEMP
 HUNTINGTON, WV
   + 1 .,
TEMP
 HT
   -10 •
         86  87 88
                                Figure 4-24
                                       4-27
-------
                                       TABLE 4-10
                      1986 Monthly Temperature Deviations from Normal
NORTHERN SITE

Bangor,ME
Jackman, ME
Orono,ME
Concord,NH
Mt. Washington,NH
Pinkham Notch,NH
Burlington, VT
Mt. Mansfield,VT
Lake Placid,NY

SOUTHERN SITE
Big Meadows,VA
Burkes Garden,VA
Luray.VA
Elkins.WV
Bristol.TN
Asheville.NC
Banner Elk,NC
Celo,NC
Grandfather Mt.,NC
NORTHERN SITE

Bangor,ME
Jackman,ME
Orono.ME
Concord.NH
Mt. Washington,NH
Pinkham Notch,NH
Burlington.VT
Mt. Mansfield.VT
Lake Placid,NY

SOUTHERN SITE
Big Meadows,VA
Burkes Garden.VA
Luray,VA
Elkins,WV
Bristol,TN
Asheville, NC
Banner Elk, NC
Celo, NC
Grandfather Mt.,NC
APR    MAY    JUN
                JUL
        AUG    SEP
                OCT
+MA1
+A3
+
+A1
+MA1
+MA1
+A2
+MA1
+A3
+
+
+A
+
+
+
+A
+
+
+ +
+ -MB2
-B3
+ -B3
+ -B
+ -B
+ -B
+A2
+ -B
+
. + +A
+ +
+
+ +A
+ +
+ +A3
+A
-FA2
-
-B4
-
-B

-
-
-
-
+
+A1
+
+A3
+A
+A2
+A2
+A2
+A1
+
-B -B2
-
.
-
-B3
-
.
-B
-B3
+A
-B
+
+
-B +
+A
+A
-B +
-
-
-
-
-
-
-
-
-
+
+
+
+
+
+
+A
+A
+
TABLE 4-11
1986
APR
+
-B2
-
-B


-B2
.
Monthly Precipitation Deviations from Normal
MAY JUN
+
+A +
+
+A
+ +
+
+ +
+A +
JUL
+A2
-
+A
+A
+A
+
+
+
AUG SEP
+A
+A +
+A3
+A
+A +
+A
+A +A
+A3 +
OCT
-MB1
-
-B
-
-
-B2
-
+
-B2
        +MA2   +A3
-B2

-MB2
-B2
-B2
-B2
-B2
-B

-B3
        +A2
        -Bl
        -B2
        -B3

        -Bl
+A

-Bl
+MA1


 N

+A

+A
                                           4-28
-------
                                      TABLE 4-12
                      1987 Monthly Temperature Deviations from Normal

NORTHERN SITE     APR   MAY   JUN     JUL     AUG    SEP     OCT
Bangor,ME           +A2     +       +
Jackman,ME          +A      -        -                -B
Orono,ME           +A      -1-       +      -        -        +
Concord,NH          +A      +               +      -B
Mt. Washington,NH   +A2     +       +      +
Pinkham Notch,NH    +A3     +       +      -       -B        N
Burlington.VT        +A1     +       +      +       -        +
Mt. Mansfield,VT     +MA2    +       +      +A
Lake Placid,NY       +A2     +       +      +       -        +

SOUTHERN SITE
Big Meadows,VA     -B3      +       +      +A       -
Burkes Garden.VA    -B3     +A      +      +
Luray.VA             -        +      +A      +
Elkins.WV            -       +A      +      +A
Bristol,TN           -B       +A2     +      +A
Asheville,NC         -B       +A3     +      +
Banner Elk,NC       -B       +A2    +A      +A
Celo,NC             -B       +A      +      +A
Grandfather Mt.,NC    -       +A      +      +A
                                +A

                                +A
                                +A2
                                -I-A3
                                +MA2
                                +A1
                                +A
                                 -MB2
                                 -MB2
                                 -B3
                                 -B2
                                 -B2
                                 -Bl
                                 -B
                                 -Bl
                                 -B
NORTHERN SITE

Bangor,ME
Jackman.ME
Orono,ME
Concord,NH
Mt. Washington.NH
Pinkham Notch,NH
Burlington.VT
Mt. Mansfield,VT
Lake Placid,NY

SOUTHERN SITE
Big Meadows.VA
Burkes Garden.VA
Luray.VA
Elkins,WV
Bristol.TN
Asheville,NC
Banner Elk,NC
Grandfather Mt.,NC
                                      TABLE 4-13
                      1987 Monthly Precipitation Deviations from Normal
APR   MAY    JUN
        JUL
       -B
+A

+A3

+MA1
-B

 N
+MA1   +
+MA1
+A3
                        -B2
                                 AUG    SEP
-Bl

-B
                                -B
                                -B
                                -B
                -B
                 OCT
                                         +A
                                 +A
                                        +MA2
+MA1
+
+A
+
+MA1
+MA1
+
-
-B3
-B
-B
.
-
.
.
+MA1
+
+
-
-Bl
-B
.
-B
N
-
+
-Bl
-B
-Bl
-B3
+A
+A
+A
+A
+A
+A
-
.
-B
-B2
.
-B3
                                          4-29
-------
NORTHERN SITE

Bangor,ME
Jackman.ME
Orono,ME
Concord,NH
Mt. Washington,NH
Pinkham Notch,NH
Burlington,VT
Mt. Mansfield,VT
Lake Placid,NY

SOUTHERN SITE
Big Meadows,VA
Burkes Garden.VA
Luray.VA
Elkins,WV
Bristol,TN
Asheville,NC
Banner Elk,NC
Celo,NC
Grandfather Mt.,NC
                                      TABLE 4-14
                      1988 Monthly Temperature Deviations from Normal
APR
MAY

+A

+A
                JUN
 -B
        -B3
                -B3
                -B
                -B
                -B
 +
 N
        -B
JUL
+A3
+A
+A
+A
+A
+A
+A2
+A
AUG
+
+A3
+A3
+
+A
+A
+A
+A3
                                                            SEP
                                        -B
                                                            -MB1
                 +      +A
                 +MA2   +MA2
                 +A      +MA1
                 +      +A
                 +      +A
                 +      +MA1
                         +A3
                 +      +A3
                                                             N
                        OCT
                                                                    -B2

                                                                    -B3
                                                                    -MB1
                                                                    -B3
                                                                    -B
                                                                    -MB2
                                                                    -B3
                        -MB1
                        -MB1
                        -B
                        -Bl
                        -MB1
                        -B2
                        -B2
                        -B2
                        -Bl
                                      TABLE 4-15
                      1988 Monthly Precipitation Deviations from Normal
NORTHERN SITE    APR    MAY    JUN
Bangor,ME
Jackman.ME
Orono,ME
Concord,NH
Mt. Washington,NH
Pinkham Notch.NH
Burlington, VT
Mt. Mansfield,VT
Lake Placid,NY

SOUTHERN SITE
Big Meadows,VA
Burkes Garden,VA
Luray.VA
Elkins,WV
Bristol.TN
Asheville,NC
Banner Elk,NC
Celo,NC
Grandfather Mt.,NC
+MA1
+A

+A2
-B
        -B3

         +
        -B3
        +MA2
        +A2
        -B2
        -B
        -B
                -B2
                                    -B
                                    -B

                                    -Bl
                                    -B2
                                    -Bl
                                    -B2
                                            JUL
                                 AUG    SEP
         +      -MB1
 -I-      +MA1   -B
 +       +      -Bl
+MA1   +A3
        +A3     +
 +      +MA2
         +      -B
        +A
 +       -I-      -B
                                  N

                                 -B
                                         +MA1
                                         +
                                                 OCT
                                                                     -B3
                                                 -B
                                                  +
                                          4-30
-------
        The following material discusses the regional climate of each year.

        1986:  The April through October period of 1986 produced contrasting weather conditions
between the northern and southern portions of the Appalachians.  In the north, it was cool and wet
after a warm spring.  April ranked as the warmest on record at most stations and had relatively light
precipitation.  This led to mild to moderate Palmer drought conditions in this region by mid-June. In
June and July a series of low pressure systems moved across the Great  Lakes and New England,
reversing this trend while convective type precipitation dominated the region in August. By late August,
and for the rest of the fall, this region was characterized by cool and moist  or very moist conditions.

        In contrast, the southern zone was warm and dry overall. A warm spring was followed by a hot
summer. This was due to a strong upper level ridge which controlled the weather until August. Some
synoptic scale systems broke the pattern in August resulting in a cool and wet month. Despite this
reprieve, drought conditions in this zone were not alleviated. In April, southern Virginia and western
North  Carolina were classified to be enduring a severe to extreme drought with  northern  Virginia in a
moderate drought. Some stations in June and July experienced near-record warm temperatures and low
precipitation amounts. By late August and onward, the drought was extreme to the south and upgraded
to severe in the northern sections of the southern zone.

        1987: In the  north, the warm spring was followed by a warmer  than normal June  and July and a
cooler  than normal August through October. April was one of the warmest on  record. Monthly
precipitation varied due to its convective nature.  Frontal waves began  to infiltrate the region beginning
in September. Mild  to moderate Palmer drought conditions were prevalent in this zone by mid-June due
to a dry May, but by early October the region had reversed that  trend to moist conditions.

        In the south, the drought conditions seen in 1986 continued and in some cases unprecedented
weather conditions occurred.  The upper level ridge held control over the region until September in the
northern part of the  zone and until October to the south.  This led to warmer and drier than normal
conditions for most of the season. The wettest April on record led to moist conditions in Virginia and
lessened the drought in North Carolina to moderate levels. By early August, the drought had increased
to severe levels in the entire  zone.  The remnants of a tropical storm in September brought plentiful
rains to some portions of the zone.  By the end of October, the  southern part of the zone was still in a
severe  drought while the northern portion had a temporary respite. October was also one of the  coldest
on record.

        1988:  There was much less of a difference in the climate between the north and  south.   Both
regions experienced a hot July and August and cold October, and generally  dry conditions prevailed.  An
upper level ridge dominated the eastern U.S. until mid-August in the north and late August in the
south.  July and August were  among the hottest on record.  After that, cool conditions generally occurred
as outbreaks of cool  air were able to sweep down from the north.  Record and near record cold  again
occurred in October.  This year's weather regime brought drought to the northern zone and prolonged
the drought in the southern  zone.  June was one of the driest on record in  the south. By early July,
moderate to severe drought levels were attained in the north and extreme levels in the south. With the
change in the pattern by September, the northern zone had transformed to a  moist region and the
southern zone had eased from an extreme  drought area to a region of moderate to severe drought.

        1986-88 Summary:  Two outstanding climate features mark the  1986-88 period:  1)  the large
contrast in weather conditions between the northern and southern portions  of the Appalachians,  and 2)
the predominance of drought and above normal temperatures in the southern Appalachians throughout
the entire period. The southern drought has been cited as an unprecedented event for the last 100-300
years (e.g. Bergman et al., 1986; Cook et al., 1988).  Figure 4-28 illustrates a virtually continuous deficit
in precipitation over  this three year period at Celo, NC.  By comparison, the  northern Appalachians had


                                               4-31
-------
5
O


z
O
h-
5!

o
1U
oc
Q.
        Figure 4-28:     Cumulative departure from normal precipitation for three

                        high elevation, long term stations for 1986 through 1988
                                            4-32
-------
relatively unexceptional weather overall in terms of persistent trends. Cool and wet weather dominated in
1986 and mixed conditions followed in 1987.  It was not until the 1988 field season that, according to
the Palmer drought index, moderate to severe drought levels infiltrated the northern Appalachians.


GASES

          Most of the nation's long-term ozone, sulfur dioxide and oxides of nitrogen (NO,.) monitoring
stations are located in or near population centers.  For the most part, the main  focus for air quality
monitoring has been the determination of concentrations to which large numbers of people are exposed.
Only in recent years has there been an interest in documenting ozone levels in rural areas where
agriculture and forestry concerns dominate.  Of the approximately  1400 ozone sites in the EPA's
Aerometric Information and Retrieval System (AIRS), only 33% have been designated as rural or
remote.  Of the 1831 sulfur dioxide sites in AIRS, approximately 33% are designated as rural or remote.
Of the 674 nitrogen dioxide sites in AIRS, only 24% are designated as rural or remote.  In the United
States, ozone, sulfur dioxide, and oxide of nitrogen  concentrations  are monitored by  local and state
governments.  In  addition to AIRS, other databases containing  hourly sulfur dioxide, nitrogen dioxide,
and ozone monitoring information exist.  In early 1983, the National Park Service established a
nation-wide network to collect hourly mean ozone,  sulfur dioxide, and NO^ data.  Since 1985, the MCCP
has monitored ozone, sulfur dioxide, oxides of nitrogen, hydrogen peroxide, and  other air quality and
cloud chemistry parameters in forested, montane ecosystems across  the Appalachian  Mountains.

        The challenge to identify monitoring  data that are characteristic of exposures experienced in
forested areas is great. Data derived from the MCCP, as well as EPA's national forest research
monitoring program in the early 1980s, provide some assistance in describing the air quality exposures
which United States forests receive.  However, in truth, there are a small  number of air quality
monitoring sites located in forests. This is because the majority of commercial and public forest lands
appear to be located in areas that  are more than  50 kilometers from metropolitan centers and the
monitors are mostly located in population centers.  Figures 4-29 and 4-30 (Appendix E)  represent some
forest resources (public, private, and state owned) across the United States, identifying those counties in
the U.S. where forest resources dominate.

        For perspective, a box plot of trends  in annual concentrations is presented in Figures 4-31
through 4-33 (Appendix E). While the absolute numbers  may not necessarily reflect  the level of forest
exposure to these air pollutants, they nevertheless show that no upward trend has occurred over the past
decade. The ambient levels of sulfur dioxide even show some statistically  significant decline. The
temporal trend in Oj, SO2 and NOX are consistent with the emission trends for  oxides of nitrogen,
volatile organic compounds and  sulfur dioxide shown in Figures 4-34 and  4-35 (Appendix E).

Ozone

        Lefohn and Pinkerton (1988) characterized  ozone levels across the United States using
hourly-averaged ambient ozone monitoring data for an eight-year period, 1978-1985,  for eight forested
areas of the United States.  The analysis focused on the annual number of occurrences of hourly
averaged ozone concentrations greater than or equal to 0.07, 0.08,  and 0.10 ppm during the growing
season (April-October) as well as during the  early (April-June) and late (July-October) portions of the
growing season.   On the average, the authors reported that elevated ozone concentrations occurred more
often in the Piedmont/Mountain/Ridge-Valley and Ohio River Valley areas than  in the Pacific
Northwest, Upper Great Lakes, and Northern New England/New York areas.  In the eastern United
States, 1978, 1980, and 1983 were generally the years with the most occurrences  of elevated  ozone
concentrations.  In these years, the later part (July-October) of the  growing season experienced more
elevated concentrations than the earlier part.  The area-wide ozone statistics were derived mainly from


                                               4-33
-------
urban-oriented ozone monitoring stations, which may overestimate the number of elevated ozone periods
which would be observed in the majority of commercial forests in an area.

       The selection of rural/remote monitoring sites for this assessment was based in part on the
distribution patterns  of the hourly ozone concentrations ("ozone diurnal") and also on the location of
monitoring sites relative to forest regions.  Attempts to gain a coherent picture of ozone exposures
across the United States is difficult. Ozone exposures experienced at a specific site in one year may not
resemble those that are experienced at the same site the next year.  Large year-to-year variation of ozone
concentrations measured at sites influenced by urban sources occur  (EPA, 1989; Lefohn et al., 1989a).
For example, ozone exposures at some urban-influenced sites in the United States were  higher in 1983
than in 1984 and  1985 (Lefohn  and Pinkerton, 1988; EPA, 1989).  In addition,  in 1988,  ozone exposures
experienced at many locations across the United States were higher than in previous years (EPA, 1989).
In addition to year-to-year variability, ozone exposures vary from site to site;  local meteorological factors
influence ozone exposures. In addition, sites that are located at higher elevation experience different
exposure profiles  than those located near ground level (Lefohn et al., 1989c).  Furthermore, because of
the importance of the higher  concentrations, attempts  to average ozone values across a geographic region
may lose the identity within a region of those sites  that experience the highest exposures.  Thus, careful
attention is required to adequately describe ozone exposures that occur within geographic regions, as well
as across years.

       The ozone monitoring sites selected for characterizing forest exposure in the geographic area
where forest damage has been observed are listed in Table 4-16. The EPA Mountain Cloud Chemistry
Program  (MCCP) is  a major  source of atmospheric chemical and physical information in the
northeastern United  States for addressing the forest decline hypotheses.


                                           TABLE 4-16
                                       MCCP and ATR Sites

                                  AIRS
Site Name                        ID #   Site Code        Years     Elevation (m)

Rowland Forest                              HF           87-88            65
Moosilauke                                   MS           87-88         1000
Whiteface 1                                  WF           86-88         1483
Whiteface 3                                  WF            87          1026
Huntington Co., NY             36031-0005  HU           86-88          500
Hampshire Co., MA              25015-6001  HA           86-88          312
Beaver Co., PA                   42007-0003  BE           87-88         1300
Shenandoah 1                                SH           87-88         1015
Shenandoah 2                                SH           87-88          716
Shenandoah 3                                SH           87-88          524
Big Meadows, VA                51113-0003  BM           86-88         1067
Dickey Ridge                     51187-0002  DI           86-88          631
Sawmill Run, VA                51015-0004  SM           86-88          453
Whitetop Mtn, VA                           WT           86-88         1689
Marion County, VA              51173-0005  MA           86-88          710
Giles Co., TN                    47055-0001  GI           86-88          244
Mt. Mitchell 1                                MM          86-88         1950
Mt. Mitchell 2                                MM          87-88         1775
Mt. Mitchell 3                                MM           88           750
                                               4-34
-------
        There has been considerable discussion in the literature concerning the selection of optimum
exposure indices for vegetation effects purposes (Lefohn and Benedict, 1982;  1985; Tingey, 1984; Lefohn
and Jones, 1986; EPA, 1986; 1988; Lefohn and Runeckles, 1987; Krupa and Kickert, 1987; Lefohn et al.,
1988; Runeckles,  1988; Ashmore, 1988; Parry and Day, 1988; Lee et al., 1988; Hogsett et al.,  1988;
Lefohn  et al.,  1989b).  The general focus has been on mathematically deriving a "biologically-meaningful"
dose surrogate, such  as a cumulative index that describes the highest exposure, while not ignoring the
lower, biologically less important exposures. Possible approaches for weighting different air pollutant
concentrations have been discussed extensively.  For example, Lefohn and Runeckles (1987) proposed an
exposure index that used  the sigmoidal weighting of individual hourly average concentrations  of ozone
and summing over time.  The sigmoidal weighting function was multiplied by each of the  hourly average
concentrations. The  lower, less biologically effective concentrations were included in the integrated
exposure summation.  They were not eliminated because no threshold was used.  An example of two
experimental sigmoidal weighting functions (W95 and W126) used by Lefohn et  al. (1988) is shown in
Figure 4-36 (Appendix E).

        In the United States, some exposure indices of ozone that have been used in the past to
characterize hourly mean concentrations over monthly and seasonal periods are:

(1)     The sum  of all hourly mean ozone concentrations using no threshold concentration.  This  is
        commonly referred to as "total dose" (SUMO);

(2)     The sum  of all hourly mean ozone concentrations (W126) where each hourly concentration is
        weighted  by a sigmoidal  weighting function;

(3)     The sum  of all hourly mean ozone concentrations equal to or greater than 0.070 ppm (SUM07);

(4)     The number  of hourly mean ozone concentrations equal to or greater than 0.070 ppm;

(5)     The seasonal means of the average of the daily 7-hour (9:00-15:59 h) concentrations;

(6)     The seasonal means of the average of the daily 12-hour (7:00-18:59 h) concentration.

        For characterizing ozone episodes, the consecutive numbers of hours when ozone equalled or
exceeded 0.070 ppm during any period of the day was determined.  These data are presented  in Figures
4-37 through 4-54 (Appendix E). For characterizing seasonal ozone exposure, the sum of all hourly
mean ozone concentration equal to or greater than 0.070 ppm between  the hours of 07:00 and  18:59 was
calculated from April 15 through October 15.  This  analysis was confined to the 12-hour daytime period.
In future analysis, it  is advisable  to include the full 24-h period. Trees  may be sensitive to ozone during
the nighttime periods.  These totals termed "sum of season dose (ppm-hr)" are also shown in Figures 4-
37 through 4-54 (Appendix E) as well as presented separately in Table 4-17 and in Figures 4-55 and 4-56
(Appendix E)  for 1986-88.

        For studying the elevational gradient phenomenon, hourly mean concentrations of ozone data for
the three MCCP Whiteface Mountain, Shenandoah National Park  sites, and the  two Mt. Mitchell sites
were used. To supplement the gradient analysis, data from three additional Shenandoah National Park
sites (Big  Meadows, 1071 m; Dickey Ridge, 631 m; Sawmill  Run, 453 m), from one site located 40 miles
from Whiteface Mountain (Huntington Forest, 500 m) and from two sites near Whitetop Mountain
(Marion, 710 m; Giles, 244 m) were also used. In addition, one site in  Massachusetts  (Hampshire
County, Quabbin  Reservoir, 312  m) and one in Pennsylvania (Beaver County,  1300 m) have been
selected from the AIRS database in order to bridge  the geographic gap  between North and South.  The
results are presented  as diurnal ozone concentrations in Figures 4-57 through 4-60. The diurnal pattern
is useful for characterizing scavenging efficiency (Lefohn et al., 1989c).


                                               4-35
-------






^.

o
t— 1
F-

Cti °°
I-M CO
£-« OS
•^ -*
W 1
Z eo
o ^*
O ""
00
W SK
•^ <=>
•*=-• 00
O -J3
g S
,-j «
— H



?
Ib 4«
! I
cjt1 ^i
Ji i
: : 5

*? t
6-ir •
: : I
LL J,
i T §
': • 1
9T ^
*•* 4t
A* •»
•
9 ••)
t ^ ^)
^ -01
T if
* *
•^ *
j
* *
' I
* «
1 1 • i • i • i • i • i*1 ih| i ' i ' i • i •
co
CM
^^
n •"
:si*
_ ^ " 0 * o
— 2 z 1 H
HO-

t*** ^ •"•
Z*22
•J >• W f-
	 M W J X
— J u >
	 lO JJ, '~' M
*^ hi
ts^ w
 g25S5
0> ° ° - o M
y M tt. «
X 0 *
H 2 H Z
-D-D
m o m o
* a * a
•i p» y» y
— eo
- - N « *
















^^^N
*^r
u
a
E








       MV3W
4-36
-------



z
0
E^
^*4
CsL e°
I-M gQ
E-« o»
W 1
O
Z oo
O ®»
CO
W 2=
~r °
•^ CO
O -*
g s
—
.-J OS
OS
2
Q




	 yrr — »r — « 	
$ * «* *
CJ 'JP Q3 T^
• /'• /'
C^lp HI* 09
A -f W *
' * '
^tcji -
<^TO-»el *
^ Ui
:• \;, \
<»ft i^fj
\ ''•• '•.'•-
tow *tcj-*5
'« >* i^5
'(rtid ''i
• OM up
^.^ ^
cpj^ co fJ *p
CCHC') N v)
ff f f
?? ? f
// i i
1 1 ' 1 ' 1 ' 1 ' 1 ' 1 ' 1 ' 1 **l *\* 1 '
CJ
	 <^ a a
_ S a 2
t^ z a M
— 1~ o - z
— ' o o o _
^aSz?a
> - o •*
!/j W o " > n
wS2^"5
— ^ "°-w"z
•^c ^ z ** . ^
/% ^ ^
CO g^ 2 > "^ ^ **
"-* " < 3 "^ . •<
— ^ OT ^ W 2 ^" M
O « ^ "" *" J
_ »— j H >. - Z 0 J
O M Z H 0 B .
— DC S«S°Sg
h. 0 o H a M
— H w « OT
Q O •* ,
.^ z z QS a z j
— f- •< - H OT 0 -
^^Sjf
0 3 U •< -4 4
BO EO OB 9 a M
r- M P ^ f> to
MUM
•» N n •* 
-------









,_-
 * *
;; / ;
* * 6
*» » »
*
7? f
5 : ^
of-* rt
•* I 1*
•' < .
f\ I
li
•' ". '
^ \
: : ; -.
•j<« f) t
•^tfl W *C
4 > >
WWW
J J J ~
	 U Dd U a
uo o
	 — 4 — O •* o
>-! WWW1"
C»3 f-s W CO to O
— __, ca m oa r
D D D 4
QLH to co co >
o 2Z2^
-^ ^^x»
OH H H H
_ .—. Z Z Z >.
'-' a s a z
O ooo
}T] *• ~ * 2
WWW?
y o o
•^ •< •< °
_r^ h fc. fc Z
w w w ~
1 1: t z
m B a D
* * * m
— ITS
r* N P *

_
— CO
— U N CS '•*
	


















o\
10
^^3"
fl j
a
E















ooooooooooooo
       NOI1VH1N33N03  MV3M
                4-38
-------





o
E-
Qi °°
* * CO
E— » as
W i
"^ *°
*=-< co
O o*
O ~"
CO
[>•[ 2:
•^ °
•^ co
O -*
g a
i-j a:
<3 ^
5
i— i
°



, 	
-H pi
X
crW
73' p*'
•f'f
•4DO)
IW
i
<*
i\
^
-< N 'n.
•n 'N. 'cs.
N ^
Cv«
*
4
or)
44 dp
/Ik
I 1 I j I 1 . 1 . 1 1 1 . 1 I 1 . 1 . l«p 1 .
CO
CM
~Z ^
	 a
09 ,_,
2^"
	 r- ^ z
"^ > o ^
j **
— ^ "S N
-~* >* M£O
__co Q J § ISji E
sfs
"^ M 2 w
~ B •< i
> ai «
" f f
— CO


ooooooooooooo
       NOUVH1N30MOO MV3W
                4-39
-------
       The analysis of the data indicates that there is a distinct ozone elevational gradient at Whiteface
Mt., and that the MCCP sites in the South experience higher cumulative ozone exposures than sites in
the North.  Figure 4-61 (Appendix E) also  indicates increased mean ozone concentration with elevation
for 19 ozone monitoring stations in the eastern U.S.( see Table 4-17 for station codes).  The length of
time that the MCCP has monitored ozone is too short for investigating the existence of any temporal
trend.  However, the Whiteface Mountain summit has an uninterrupted monitoring record since 1973.
As shown in Figure 4-62 (Appendix E), no trend in annual mean ozone concentration is discernable for
this 15 year time period.
Site Name

Rowland Forest (HF)
Moosilauke (MS)
Whiteface 1 (WF1)
Whiteface 3 (WF2)
Whiteface 4 (WF3)
Huntington Co., NY (HU)
Hampshire Co., MA (HA)
Beaver Co., PA (BE)
Shenandoah 1  (SHI)
Shenandoah 2  (SH2)
Shenandoah 3  (SH3)
Big Meadows, VA (BM)
Dickey Ridge, VA (DI)
Sawmill Run, VA (SM)
Whitetop Mtn, VA (WT)
Marion County, VA (MA)
Giles Co., TN  (GI)
Mt. Mitchell 1 (MM1)
Mt. Mitchell 2 (MM2)
         TABLE 4-17

       Ozone Exposure
     Sum of Season Dose
(for daylight hours 7AM - 6PM)

Sum of Season Dose (ppm * hr)
                  .> 0.07 ppm
        1986      1987
N.D.
N.D.
2.29
N.D.
N.D.
5.09
9.14
N.D.
N.D.
N.D.
N.D.
5.56
3.21
11.39
N.D.
4.11
16.38
8.34
N.D.
0.82
7.81
9.68
9.41
3.47
5.74
9.26
13.06
9.49
9.01
6.07
28.50
31.07
26.80
38.54
9.27
16.73
5.14
6.68
1988

 4.16
12.51
20.51
16.50
 N.D.
11.36
34.93
31.70
23.27
39.44
20.88
31.89
40.25
30.16
37.68
26.92
28.91
45.17
19.49
                                              4-40
-------
Sulfur Dioxide

        Sulfur dioxide concentrations are very low throughout the network as indicated in Table 4-18.


                                            Table 4-18

                   Sulfur Dioxide Measurements Summary for Selected MCCP Sites

                                           Max              Mean   # hourly averages

Rowland  HF                              12 ppb           1.07 ppb         1371
Whiteface  WF1                            20 ppb           1.42 ppb         1990
Whiteface  WF3                            16 ppb           0.86 ppb         2962
Whitetop   WT                            47 ppb           2.08 ppb         4310
Mt. Mitchell MM1                         12 ppb           1.41 ppb         996


        Figures 4-63 through 4-66 (Appendix E) show the frequency distributions derived from
continuous SC>2 measurements. Stations operating on a weekly schedule (filterpack) reported similarly
low values  for the warm season May 1 through October 31, 1987.  The northern sites exhibited  a strong
directional  variability on the basis of 36-hour back trajectories.  Air masses from the south and  west had
significantly higher SO2 concentrations than air masses from any other sector.  The southern stations
exhibited a more  even pattern of SC>2 concentration as function of air mass trajectories.  Because SC>2 is
the precursor gas to cloud water sulfate, it is not surprising to find similar directional variabilities for
both sulfur compounds.

        Because of the very low concentration levels of sulfur dioxide  during the warm  season, it is very
difficult to  reliably measure an altitudinal gradient.  It is equally difficult to discern a geographic gradient
from the data set.

Hydrogen Peroxide

        Hydrogen peroxide (P^C^)  plays a key role in the formation of sulfuric acid in the atmosphere
(Calvert et al.,  1985; Seigneur and Saxena, 1988).  The aqueous phase oxidation of dissolved SC>2 by
H22 was applied to Norway spruce.  Joslin et al.  (1988)
showed  that H2C>2 was effectively removed from cloud water upon  contact with red spruce foliage.
Cloud water H2C>2 measurements from the MCCP network can help to quantify the role  of this
compound  as an oxidant and  as a possible agent of forest damage.

        Olszyna et al. (1988)  reported cloud water HjC^ measurements for the MCCP  site at Whitetop
Mountain,  VA. The highest concentration measured (246.6 /trnol/1) was more than twice any previously
reported cloud  concentration. Daum et al. (1984) used an aircraft to sample stratiform  clouds in the
eastern  USA and  reported H2O2  concentrations varying from 0 to  75/tmol/l.  Romer et al. (1985) used
an aircraft  to collect cloud samples at various European locations and found that H2O2 concentrations
varied from 0.05 to 88 jtmol/1. Lazrus et al. (1985) reported that  in 284 cloud  samples at Whiteface
Mountain, NY, ^O^ concentrations varied from 0 to H2nmol/\. Kelly et al.  (1985) reported total
peroxide (assumed to be H2O2) levels for an additional  190 cloud samples collected at  Whiteface
Mountain and found similar H2C>2 levels (0  to 100 jtmol/1).


                                               4-41
-------
        Cloud water H2O2 levels for four MCCP sites from 1986-1988 are presented in Table 4-19.  Not
all cloud water samples were analyzed for H£>2-  Mean concentrations were similar for all sites except
Shenandoah, VA  The low mean concentration reported for this site may be an artifact of the small
sample size. The highest H2O2 levels were recorded at each site during  summer.  Spring and fall levels
were typically less than half of the  summer values.  Kadlecek et al. (1985) reported H2O2 concentrations
in winter clouds  at Whiteface Mountain to be always much lower than 1 /tmol/1. This seasonal trend is
no doubt due to the photochemical production rate of H2O2 in the gas phase, which would be expected
to exhibit a strong annual trend with maximum production in summer and minimum during winter.

        The distribution of H2O2 concentrations in cloud water collected at the Whiteface and Whitetop
sites is shown  in Figure 4-67 (Appendix E).  A higher frequency of very  high H2O2 concentrations  was
observed at Whitetop Mountain, VA

                                           TABLE 4-19

                Cloud Water H2O2 Measurements at MCCP Summit Sites for 1986-88
                                  All concentrations are in jonol/L


Site                    Minimum      Maximum     Mean          # Samples

Whiteface,                 1.6            136.4         37.7                 54
Shenandoah, VA          0.3            49.0         12.3                 22
Whitetop, VA             0.8           246.6         44.1                141
Mitchell, NC              0.3            196.0         41.9                236
THROUGHFALL

        The chemistry of throughfall water is a composite of rain, cloud, and dry deposition, tree
emissions and absorptions, evaporation from the canopy, and biological activity on the canopy and in the
collected sample. Also, physical features (terrain slope and aspect, canopy structure and individual
canopy element geometry, and wind velocity) affect where  canopy collection of rain and cloud occurs and
how the directed flow develops within the canopy.  It is difficult to properly attribute the sources and
sinks to the correct mechanisms since they are difficult to  measure separately and have considerable
variations across the canopy and within and between events.

        For a rain event, there is an initial period during which the canopy is wetted until the storage
capacity is satisfied.  The excess accumulation over evaporation drips through the canopy as throughfall
or flows down the tree trunk as stemflow.  The common method for determining canopy storage capacity
involves linearly relating the unperturbed rain amounts (which can be subject to  collection problems in
high wind environments) to the corresponding average  throughfall from many events.  The offset from
the origin provides the storage capacity, assuming that  evaporation is much less than the water held by
the canopy.

        For a cloud  event, the storage capacity may be satisfied only after prolonged exposure to clouds.
While the cloud water may be relatively inconsequential to forest hydrology in most areas, it has the
potential to contribute to foliar/needle damage  (because of the higher concentrations of dissolved species
in the cloud water as compared with rain, and the extended  period of evaporation that further con-
centrates the less volatile chemicals).  However, measurements of evaporation of intercepted water are
rare, further complicating efforts to quantify this process.  Most events at high elevation are a mixture of


                                               4-42
-------
rain and cloud, but the canopy storage cannot be obtained directly because the cloud water enhances
throughfall, without being collected as precipitation. In many cloud events, the throughfall amount
slightly exceeds what was measured as rain, providing an estimate of cloud water deposition, if estimates
of canopy storage for the appropriate canopy composition can be made.

        Seasonal data from the summit of Whitetop Mountain (May 1987 - May 1988) have shown,
however, that differences between throughfall volumes and precipitation volumes are generally poor
indicators of the volume of cloud deposition due to substantial and unknown evaporation losses,
especially during the summer. At this site during the wanner months, monthly deposition volumes under
canopies are  less than or about equal to those in the open.  This pattern occurred despite the fact that
other information (computed cloud water flux and total sulfate deposition in throughfall)  indicated that
cloud deposition occurred during the warmer months as well.

        One  reason for this summer pattern is the higher frequency of orographic clouds  with low liquid
water contents and  high ion concentrations.  These events rarely result in much water reaching the forest
floor, while still accounting for a considerable deposition of ions to the canopy. In contrast, during
periods  when evaporation is less important due to lower temperatures and solar radiation levels, the
accumulation of additional water from clouds can be observed. MCCP research is now in progress in
which the cloud, precipitation, and throughfall chemistries are monitored.  If the dry deposition  of SO^
and Cl are relatively small, their enrichments in  the throughfall can be used to further refine the cloud
water deposition. The biologically and/or chemically more active species H+, NO/, and NH^+  cannot
be used for this purpose since considerable loss of these species is possible.

        Within the last  two  decades, interest in throughfall has broadened  from a hydrological focus to
include  nutrient cycling, particularly fertilization  and leaching (Parker, 1983). Data are now being taken
with sufficient time resolution to monitor the changing chemical environments on the canopy  surfaces.
These results will aid simulation studies designed to provide a better understanding of the beneficial and
harmful effects from extended periods of canopy wetting.

        Initial analysis of Whiteface data indicated that  ion depositions  have a larger variability  than
does the deposition of water.  The data collected apply only to the site  from which they were  taken and
to sites  where the wind regimes and canopy composition and structure are similar. Event-based com-
parisons of throughfall and rain chemistry were made at the same  two sites on  Whiteface Mountain, NY.
The lower site was  affected by rain only, and the upper  site  had additional cloud water deposition.
Throughfall from both sites  included whatever dry deposition had occurred since the last wetting, residue
after evaporation of the last rain/cloud, and emissions from the tree itself and  biology living on  the tree
surfaces. Table 4-20 summarizes enrichment ratios integrated over  the summer season in species
deposition, where the uncertainty was obtained using standard error propagation calculations.  No
estimates  of bias were possible.
                                               4-43
-------
                                           TABLE 4-20

      Deposition Enrichment Ratios (Throughfall/Bulk Precipitation) at Whiteface Mountain, NY
               Water (cm)

               H+
               NK/+
               N0i
               so/-
               K+
               cr
Under Canopy
Upper Site

8.51 ± 0.48

0.95 ± 20%
< 0.35
0.86 ± 23%
1.45 ± 19%
9.4 ±44%
2.51  + 27%
       Open Rain
       Lower Site

       8.31 ± 0.08

       0.71 ± 13%
       1.11 ± 22%
       1.65 ± 19%
       1.16 ± 14%
       > 13
       > 2.0  + 25%
        Of particular interest is the loss of the first three species at the upper (balsam fir) site, either on
the canopy or in the collected sample. Given that this site also had a substantial input of these species
from cloud water (current estimates suggest deposition on the same order as from rain), this suggests
that about half the H+ and the NO/ and at least 75% of the NH^+ that should have been deposited
were not present at the time of analysis.  Samples were refrigerated immediately after the event and the
analysis was completed within  a few days.  Sulfate and chloride have reasonable enrichments given
contributions from dry and from cloud at the upper site.  Foliar leaching is responsible for the ten-fold
increase in K+ deposition.

        This section does not  provide a comprehensive review of throughfall with an investigation of the
role of cloud water deposition; rather, it is a characterization of the gross features of the throughfall
compared with the precipitation obtained during the same time.  For each weekly pair of samples, the
ratio of deposition in throughfall  to open  precipitation was calculated. The median values are presented
in Table 4-21.
                                           TABLE 4-21

               Median Value of Site Enrichment Ratios (Throughfall/Bulk Precipitation).

        Site                    SO^2-          NO/          NHf+         H+
Shenandoah, VA - upper        1.3
Shenandoah, VA - middle       1.3
Shenandoah, VA - lower        1.1
Moosilauke, NH               1.3
 1.8
 1.7
 1.2
 1.0
0.9
1.0
1.0
0.8
0.5
0.2
0.1
0.9
        Not all sites have the same enrichment behavior because the relative input strengths differ, the
fraction of total deposition due to cloud water affects the ratios, and the canopy type helps create its
own chemical and biological environment.  While dry deposition of various ionic species can be es-
timated and is in some cases relatively small, foliar exchange can be either a major source or sink.  The
degree of interaction between the acidic deposition and the canopy is reflected primarily in the loss of
H+ and NH^+, and an increase in base cations in the throughfall samples.

        Similar weekly integrated throughfall measurements made during  1987 and 1988 at two sites in
spruce-fir forests of the Black Mountains showed that while the yearly H+ average was 24% to 34% of
                                               4-44
-------
the total ions in rain, it was only 16% to 23% in throughfall (Robarge, 1989).  Similarly, NH4+ was 8%
to 10% in rain, but only 3% to 5% in throughfall.  Similar trends for these ions have been reported in
other canopies (Bredemeier, 1988; Joslin et al., 1988; Waldman and Hoffmann, 1988; Olson et al., 1981;
Rodenkirchen, 1986) and in controlled experiments (Kaupenjohann et al., 1988; Kelly and Strickland,
1986; Evens 1982; Scherbatskoy and Klein, 1983). Recent data (May 1987 - May 1988) from Whitetop
Mountain give further evidence of the importance of the foliar exchange (see Table 4-22).


                                           TABI£4-22

         Actual (keq/ha/yr) and Percent Ion Contribution to Total Deposition at Whitetop, VA
             Percentages for cations and anions are treated separately  (Joslin  et al., 1988).

                       H+     NH,+  Ca2+  Mg2^  Na+   K+     NO/   SO^  Or
Throughfall &          1.27    0.21    1.26    0.27    0.16    0.35    0.72    2.50    0.37
  Stemflow             36%   6%      36%  8%      5%    10%    20%    70%    10%

Precipitation           0.50    0.14    0.07    0.02    0.03    0.01    0.20    0.52    0.04
                       66%   18%     9%     3%     4%     1%    26%    69%    5%

Cloud                  63%   26%     5%     2%     3%     1%    32%    64%    4%
        Precipitation and cloud water had similar proportional contributions to total deposition;
throughfall was different from both.  Because of the large, unquantifiable amount of foliar exchange,
cation deposition in throughfall (with the possible exception of Na"1") was of little use in estimating
cloud water or dry deposition.  Of the major  anions, nitrate appeared slightly depleted (see also
Lindberg et al., 1987) and chloride was somewhat enriched  (both foliar absorption and leaching of
chloride has been observed; Leonardi and Flueckiger, 1987). Sulfate appeared the least affected by foliar
exchange.   In addition, the concentrations and total quantities of sulfur deposited and passing through
these forest canopies was large relative to the size of total foliar pool.  The time period for which
throughfall sulfate is used to  represent total sulfate deposition should be sufficiently long so that
differences in canopy residue sulfate are small relative to the amount passing through.  The reproducibil-
ity and completeness of wash-off,  the  importance of dry deposition of aerosol sulfate and sulfur dioxide,
and the sequence of meteorological systems affecting the site affect the utility of  this approach.


DEPOSITION

Cloud Water Droplet Deposition

        Estimates of cloud water  and chemical deposition to high-elevation forests are now reviewed.
Because cloud water is usually collected over  hour-to-event time intervals during  intensive  sampling
campaigns, few full measurements exist for seasonal or annual  cloud water deposition.  Annual cloud
deposition estimates that have been published are usually extrapolated from short-duration, growing
season  (generally April-October) cloud water  chemical composition and model or throughfall-based cloud
water interception rates.  Extrapolation of cloud water or chemical fluxes to  annual values is highly
uncertain  because winter chemical concentrations are generally not known (see section on winter data
from Whiteface Mountain, NY) and because cloud water interception rates in winter depend on
unknown rime ice collection efficiencies.
                                               4-45
-------
Historical Perspective —

       Table 4-23 summarizes past estimates of chemical and water'deposition due to the interception
of wind-driven cloud droplets.  The MCCP has been evaluating the cloud deposition model (CDM) and
revising the technique used to  estimate cloud deposition at its high elevation  sites.  These estimates are
presented later  in this section. This historical perspective is  presented to give background for the later
discussion of deposition at MCCP sites. Cloud water chemical  deposition values presented in Table 4-
23 have been converted to a  consistent set of units (kg/ha/mo) for ease of intercomparison among
estimates.  Cloud water deposition is reported as an annual estimate (cmtyr, extrapolated from shorter
time periods) to facilitate  comparison  with annual precipitation totals.  All deposition  estimates
presented here  are based either on canopy throughfall data or model-derived cloud water fluxes.

       Lovett  et al. (1982) reported estimates of sulfate and  nitrate deposition from clouds on Mt.
Moosilauke, NH, using cloud water fluxes  computed from a model (Lovett, 1984).  He estimated that
cloud water deposition contributed about 70% of the total (cloud and precipitation) chemical fluxes for
SO^2' and NO/.

       Mueller and Weatherford (1988) computed cloud deposition on Whitetop Mountain, VA, for a
26-day period in the spring of 1986.  Using Lovett's model (Lovett, 1984), they computed cloud deposi-
tion for every hour during the study period, although some meteorological parameters and ion con-
centrations had to be estimated to fill in missing data.  The cloud water SO^ flux was between 5.3 and
9.1 kg/ha/mo, while NOj flux was between 2.8 and 5.4 kg/ha/mo.  The ranges reflect projected variation
in unmeasured  model input data.
                                                4-46
-------
                                            TABLE 4-23
          A Review  of  Chemical Ion and  Cloud Water Deposition  Via Droplet
             Interception,  Various  Locations,  Investigators,  and Years.

Sitefl
MS.NH
MS.NH
MS.NH
MS.NH
UF1.NY
UF.NY
UF3.NY
WF3.NY
UF.NY
UF.NY
UT.VA
UT.VA
WT.VA
UT.VA
MM.NC
MM.NC
CL.NC
MM.NC
MM.NC
MM.NC
MM.NC
MM.NC
SH.VA
SH.VA
SH.VA
SM.NC
SM.NC
SM.NC
RT.QUE
HN.CA
GDF.GB
I,
Ref.°
Lovett7
Lovett
Lobett7
Mohnen2
Mohnen2
Mohnen2
Lindberg
Lindberg5
Mohnen2
Mohnen2
Mueller7
Mohnen2
2
Mohnen
Mohnen2
Saxena
Stogner
Dasch
Lin5
Lin5
Lin5
Mohnen
Mohnen2
Krovetz7
Sigmon
Mohnen2
Lindberg7
Lindberg
Lindberg5
Schemenauer
Ualdmen7
DoUarcr
Cloud Droplet Deposition Frequency H2° Flux site Coll.
H+ H»4+ MO/ SO/" H20 Cloud Method" Elev Per.'
[kg/ha/K>]
.20
.
.
.04
.18
.05
.04
-
.
.
.
.22
.20
.
.17
.12
(.03)
.09
.18
.11
.09
.13
.01
.02
.04
.14
.05
.
.11
.01
.01
1.
.
.
0.
2.
0.
0.
-
.
.
.
2.
11.
.
1.
0.
(0.
0.
1.
0.
0.
1.
0.
0.
0.
1.
0.
.
.
0.
-
4


24
3
8
5C




6
8

2
8
3)
6
3
7
1
1
13
21
26
0
5**


03

8.5
.
_
0.8
4.8
3.0
1C
1.2
.
_
4.3
7.7
10.4
_
3.2
3.2
(0.9)
1.9
4.0
2.8
2.6
2.2
0.5
0.8
1.3
2.2
1**
1.1
.
0.5
0.5
11.5
.
.
1.7
11.2
7.0
2
1.3
_
_
8.1
13.8
13.1
_
9.8
7.3
(2.2)
5.2
8.0
5.6
5.5
8.9
0.6
1.2
1.3
7.2
4
4
.
0.3
0.7
lcm/yr} Dflir] (•)
68
34-98
153
(18)
(127)

13
5-31
112
1
.
_
.
25-56
35-77
-
18
32
75
57
.
.
9
5
.
.
37
9-70
77
2
-
40
40
40
23
42
27
-
5-13
30
10
.
31
31
27-41
25-52
-
62
28
41
30
27
27
7
7
7
25
.
10-35
44
.
7
model
th.fall
model
model
model
th.fall
model
model
model
model
model
model
th.fall
model
model
model
th.fall
model
model
model
model
th.fall
model
th.fall
model
model
model
model
assumed
assumed
msmt.
1220
1220
1220
1000
1483
1200
1225
1225
1200
900
1686
1686
1686
1686
2020
2020
1987
2020
2020
2020
2020
2000
1014
1014
1014
1740
1740
1740
970
780
850
80-81


87
87
87
86-88
86-87
86
86
86
87
87
86
86
86
86
86
87
88
87
87
86-87
86-87
87
86
86-88
86-87
85
82-83
82
a MS = Moosilauke, NH;   UF  = Whit'eface Mtn,  NY;  WT = Whitetop, VA;  MM = Mt. Mitchell, NC;
  CL = Clingmans Dome,  NC;  SH = Shenandoah, VA;  SM = Great Smoky Mtns.  NC;  RT = Roundtop,  Quebec
,  HN = Henmnger Flats,  CA;  GDF = Great  Dunn Fell, Great  Britain
  1 = Peer-reviewed journal article or book  chapter;   2 = Technical report  subject to EPRI  or US EPA
  review;  3 = Unpublished  data summary or submitted manuscript
C NO/  and NH^ were estimated from total  N cloud deposition and the overall  NH^/NOj ratios given by Lovett
 , (in Lindberg & Johnson, 1989)
  Annual  cloud water flux was determined  by:  (1) a version of the Lovett model (1984); (2)  collecting
  throughfall under the canopy and correcting  for precipitation and evaporation; or (3) direct
  micrometeorological   measurements.
e Collection period refers  to the years when data were collected.  Most data are from short-term, intensive
  sampling performed during the growing season (April-October).  None represent rime ice conditions.
note:  Data in parentheses  were calculated from information provided in the  paper; Mohnen's  Whiteface Mt.
  estimates are based on data from sites  at  1483 and 1245m.
                                                 4-47
-------
        Lindberg et al. (1988) used Integrated Forest Study (IPS) data to estimate cloud, precipitation,
and dry deposition at a high-elevation site in the Great Smoky Mountains.  Cloud water SO4 and NOj
deposition were estimated to be 7.2 and 2.2 kg/ha/mo, respectively, both two to five times greater than
estimates from a nearby low-elevation site for the period January-April 1986.

        More recent deposition estimates from the IPS appear in discussions by Lovett, Knoerr and
Conklin, and Ragsdale in a draft summary of the IPS (edited by Lindberg and  Johnson 1989).  Cloud
water H+ deposition at these two sites was estimated to be 30% to 40% of total  H+ deposition. The
IPS scientists concluded that the primary sources of uncertainty in cloud water deposition were
immersion time, cloud water amount, and ion concentrations.  These IPS results were only presented as
a graphical summary with no presentation of methodology, chemical concentrations, or quantitative
uncertainties.

        Following Lovett's work, Saxena and co-workers estimated cloud water and chemical deposition
on Mt. Gibbs (Mt. Mitchell  State Park, NC), while assuming canopy structure  and collection efficiency
for cloud droplets (Saxena et al., 1989a; Lin and Saxena, 1989; Stogner and Saxena, 1988; Saxena et al.,
1989b). Table 4-23 illustrates some of the cloud deposition  estimates for Mt. Mitchell.  There is a wide
range of cloud water deposition values from two research institutions  (2 to 20  kg  of SO4/ha/mo).
Although variation in cloudiness and pollutant concentrations are primary causes  of differences between
years, the variety of results for one year suggests that the early Mt. Mitchell cloud deposition modeling
results are uncertain.

        One of the Mt. Mitchell area estimates (Dasch,  1988) from nearby Clingmans Dome has been
compared with coincident data collected by Saxena et al. (1989b).  The representativeness of the two
results remains unknown because of the lack of direct measurements of cloud water deposition.  A
comparison  of cloud frequency versus cloud water flux from  all investigators indicates that Dasch's cloud
water deposition was anomalously  low compared to cloud frequency in other published results (see Table
4-23).


Status of the Cloud Deposition Model -

        The cloud deposition model (COM) used by the MCCP is an improved version of the model
originally developed by Lovett (1984).   Modifications to Lovett's model are briefly described by Mohnen
(1988a). Work related to the CDM has concentrated on examining the following:

        * the difference in model performance caused by using a site-specific vertical profile of forest
        canopy surface area versus the model  default profile based on data from Lovett's Mt. Moosilauke
        forest (Lovett, 1984)

        * the difference in model performance caused by using site-specific vertical profiles of wind
        speed versus the default profile based on Lovett's site

        * the difference in model performance caused by changing from the original Lovett droplet
        collection efficiency  scheme, based on individual tree components  (twigs  and branches), to an
        experimental scheme designed to consider (crudely)  the effect of tree morphology ("bulk
        collector" versus individual branches and twigs); several experimental schemes were tried, each
        used a slightly different technique for merging bulk collection efficiency (applicable to the most
        dense portion of a tree crown) with Lovett's tree component collection efficiencies (Thorne et.
        al.,  1982)


                                               4-48
-------
        * the importance of using site-specific meteorological data (for model input) versus data from a
        nearby site

        * overall model performance relative to canopy throughfall measurements

        Data have been collected from two spruce stands near the summit  of Whitetop. These include
canopy-top wind speed and related  profiles of speed, spatial variation in cloud liquid water, vertical
airspeed at canopy top, and canopy throughfall (TF) rates.  One TF plot (A) was located about 75 m
northeast of the Whitetop summit,  while the second (B) was located about 125 m east-northeast of A.
A larger TF database exists for plot B because of the construction and operation of an automated TF
volume measurement system.  Plot A is considered a near-summit site (local slope less than 5%)  and its
architecture is fairly  open (leaf area index=4.8; maximum canopy  height=15 m), whereas plot B  is  a
north-facing slope site (local  slope almost 20%) that is closed with very little tree mortality (leaf area
index=11.0; maximum canopy height=17.5 m).  The leaf area index values provided here represent the
ratio of the total (full-sided)  leaf area to the associated ground area.

        Preliminary findings of the  CDM evaluation are now summarized:

        * model performance was little affected by the surface area (SAI) profile used;  this is consistent
        with expectations from a previous analysis (Mueller 1990) of model sensitivity to this feature
        (computed cloud water flux is not sensitive to modest uncertainty in the height of the SAI
        profile maximum)

        * model performance was greatly affected by wind speed profile changes-the default profile of
        Lovett produced lower cloud deposition estimates than actual profiles-  the difference was on
        the order of a factor of 4 at plot B and a  factor of 1.5 at plot A

        * several modified  (to test different methods for incorporating bulk collection efficiency into the
        Lovett model framework) droplet collection efficiency schemes were tested-much data are still
        needed to better understand the relationship between collection efficiency and tree morphology;
        the scheme used to make subsequent cloud deposition estimates for this report (Mueller, 1990)
        produced 60-70% lower deposition estimates than Lovett's (1984) original scheme

        * site-specific meteorological data (measured canopy-top wind speed and spatially-adjusted liquid
        water  content) were found to result in much better model performance at plot B than data
        collected at the mountain summit

        * overall, the CDM performed better in simulating cloud flux to the closed stand (plot B) as
        opposed to the open stand  (plot A), but the difference was likely due to a greater uncertainty in
        model input wind data for plot A.  Using  the most updated version of the model and "best"
        inputs, flux was overestimated at plot A by 20-50%, depending on which version of the modified
        collection efficiency scheme was used, with computed  flux bias ranging from -20-30% for plot B.


Current Modelling Results  -

        The version of the  CDM used for the spruce-fir sites in this  analysis of cloud deposition was
configured with the modified version of the droplet collection efficiency scheme that produced the least
bias for the Whitetop evaluation. This modified scheme used measured bulk droplet collection  efficiency
(Joslin et. al.,  1990) in place  of Lovett's individual tree component efficiencies for those portions  of the
tree crowns having the highest leaf-to-total surface area ratios (ie. ratios >  0.85). This represents a


                                                4-49
-------
major change  from the 1988 analysis (Mohnen, 1988a).  The uncertainty in deposition calculations
caused by using this particular parameterization is  unknown.  Default SAI profiles (similar to those
used in the  1988 analysis) were used for all sites except WT where site-specific  profiles were known.
Another major change from the 1988 analysis was  the use of a = 0.20 in place of a = 0.27 for the wind
speed extinction coefficient.  The default (Lovett) value of a = 0.27 and the  value that appeared to be
most representative of conditions at Whitetop a = 0.10 yield considerably different results.  The best
value  for each of the other MCCP sites was not known.  A compromise was to  use a = 0.20, due to a
lack of evidence favoring any specific value over another. Site-specific wind speed profiles were used in
all WT CDM runs.

        The methodology followed for this analysis was very similar to that used in the 1988 MCCP
report (Mohnen, 1988a).  Cloud impaction  events at each MCCP site were classified according to the
synoptic meteorological conditions. A detailed analysis of cloud chemistry and  meteorological variables
(wind speed and liquid water content) was conducted by Vong et al. (1989).  In addition to synoptic
typing, Vong et al. found  that specific air  trajectory directions computed within a given event type can
further characterize event conditions. For example, ion concentrations  measured during events classified
as Varm sector" (with respect to a cyclonic weather system) were sometimes different  when  segregated
according to whether associated backward air trajectories crossed the heavily industrialized Ohio River
Valley.  This technique was able to explain half of the variance in the chemistry and meteorological
variables, and can  be used to estimate non-monitored, multiple-hour means  for the different synoptic
types.  Deposition estimates for MCCP sites can now be made  for periods when data  are incomplete  as
long as cloud frequency and event type  can be determined.

        In this analysis, cloud event synoptic type was determined for each hour of each growing season
at each site for the 1986-88 period. The results of the synoptic/trajectory (subclass) analysis performed
by Vong et  al. (1989) were used to develop subclass means of aqueous ion concentrations for non-
precipitating clouds, wind speed and liquid water content.  These parameters form the basis for
estimating cloud deposition to high elevation forests. This approach, by reducing the chemistry and
meteorological variance within each subclass, enables subclass means  to be  used in place of hourly data
for computing deposition   flux in a manner that removes the bias associated with non-random, part-time
cloud sampling.  The frequencies of the various synoptic  meteorological classes for each site used in this
analysis are summarized in Table 4-24.  Detailed definitions of these classes are found in the 1988
MCCP deposition  report (Mohnen, 1988a).  In summary, these classes are:

                       class 1:  pre-warm front
                       class 2: NW sector of cyclone
                       class 3:  post-cold front
                       class 4:  warm sector  of cyclone
                       class 5: stationary front
                       class 6: marine flow  (off Atlantic)
                       classes 7&8:  cutoff low in upper atmosphere
                       class 9: cap cloud
                                                4-50
-------
                                             Table 4-24

                 Frequency of Occurrence of Various Synoptic Meteorological Classes
                                 by Site and Year for Growing Season

Site

MS


WF


MM


WT



Growing
Season
May 15 -
Oct 15

Jun 1 -
Sep30

May 1 -
Oct 15

May 1 -
Oct 15


Year

1986
1987
1988
1986
1987
1988
1986
1987
1988
1986
1987
1988



a




Percent Frequency by Class
1
26
27
31
16
27
9
0
0
0
0
0
0
2
3
18
7
3
10
1
3
3
1
1
5
2
3
15
8
9
33
26
36
15
13
18
12
20
16
4
36
26
51
32
20
39
43
43
39
57
41
40
5
0
0
0
0
0
1
3
3
1
1
2
2
6
11
19
2
6
5
0
35
21
27
28
21
21
7&8
9
2
1
8
0
7
0
5
3
0
4
13
9
0
0
0
2
12
6
1
11
11
2
6
6
b
Overall
Cloud
27
22
7
44
40
23
32
29
22
38
30
26
a.  Frequencies include only those cloud hours that could be
   classified.  Frequencies may not sum to 100 due to roundoff.
b.  Growing season cloud frequency (%).
        Some of the synoptic meteorological classes contained only a few hours per site.  Hence, a
complete event characterization was not possible  for all classes/subclasses at all sites.  The subclasses
that proved  to be of most importance were those involving trajectories crossing or avoiding  the Ohio
Valley within synoptic class 4, north versus other trajectories in synoptic class 3, and east versus other
trajectories in synoptic class 6.  Examples of some of the more extreme differences found between
subclasses are provided in Table 4-25. All data in the table are for precipitation-free hours only (this
avoids the likely contamination  of cloud water samples by precipitation).
                                                4-51
-------
                                            Table 4-25

           Examples of Differences Between Synoptic/Trajectory Subclasses for Selected Sites

Site
MS
WF
MM
WT

Synoptic Subclass
Class
6 East traj.
Other traj.
3 N traj.
Other traj.
4 Ohio traj.
Other traj.
6 East traj.
Other traj.

Speed
(m/s)
3.9
3.6
11.3
10.5
11.0
9.0
5.9
5.7

Mean
LWC
(gfr*5)
0.18
0.16
0.42
0.46
0.16
0.24
0.24
0.23
a
Conditions
[H]
Oeq/L)
135
275
30
80
756
383
113
400

[S04]
Oeq/L)
101
66
27
116
947
317
130
586
a.  Average of given parameter over all hourly measurements in
   subclass.  LWC=liquid water content;  [H]=cloud hydrogen ion;
           cloud sulfate ion.
        Cloud deposition estimates for each site were made by computing, for each subclass, the mean
water deposition flux using the COM and  subclass-mean wind speed and liquid water content.  Best
estimates of canopy structure (Mohnen, 1988a) and growing season averages of  other, less important
CDM inputs (i.e., temperature and pressure)  were used to estimate the gross (pre-evaporation) cloud
water flux to specific forest canopies at each site.  However, these deposition estimates are known to be
sensitive to canopy structure and the local  influences of topography and forest  inhomogeneity on
meteorology.  Thus, the "potential" cloud deposition was also examined to determine site-to-site
deposition-related variability not dependent on model performance.

        Potential deposition (K,) for ion X was defined as

                              Fp =uW[X]

where u is canopy  top wind speed, W is liquid water content and [X]  is the cloud water concentration
of X. The deposition of cloud water and, consequently, X, to a forest canopy is known to be highly
correlated with the product u W (Mueller and Imhoff, 1989).  The mean potential deposition was
computed, when possible, for each subclass as
                       Fp  = [{u W [X]ct  }/(u W)] u W-* l^j     ,         (1)


where [X]ct was the cube-root transformed cloud water concentration of X and the triple product of
uW[X]c, was averaged over all hours in a  subclass.  The subclass mean solute concentrations [X] used
to estimate Fp as in (1) are calculated as water content and wind  speed weighted mean values, where the
solute concentration is in cube root-transformed units.  The concentration is back-transformed to


                                                4-52
-------
after the weighted mean is determined.  Use of the cube-root transform, avoids errors associated with
skewed data distributions.  Weighting [X] by W and u minimizes the influence of any inter-variable
correlations on the calculated triple products (uW[XJ) for each synoptic class.  The factor  f^y in (1) is
a unique adjustment applied  to the computed  Fp for  each site to correct for placement of the wind
speed sensor (canopy top wind speed is most relevant to deposition flux), and the spatial differences, at
some sites, between the location where W and [X] were measured and the canopy of interest. The basis
for most of these adjustments were previously described in  Mohnen (1988a).   For Mt. Moosilauke, f^y
= 1 because  no adjustments were needed.  For Whiteface Mt., f^y = 0.33, the product of the
adjustment factor (0.33) for converting the summit u to canopy-top u, the adjustment factor (0.357) for
converting the summit W to  canopy-top W, and the adjustment  factor (2.8) for converting the summit
[X] to canopy-top [X] assuming that the elevational difference in [X] was due only to dilution (note that
the water content and [X] adjustments cancel  out).  For Mt. Mitchell, the f^* of 0.62 converted u from
that measured at 16.5 m above ground to that at canopy-top (about 9 m).  For Whitetop Mt. f^/y was
computed to vary with wind direction (a direction frequency-weighted adjustment was computed for each
subclass - values ranged from 1.14 to 3.41),  and the u measured at 7.4 m at the summit was converted
to a u value for the nearby canopy top (about 15m).

        The results of the F., analysis  are summarized in Table 4-26.  Synoptic classes 2, 7 and 8 are
rare and, therefore appear together under "other classes" in the table.


                                            Table 4-26

                          Mean Potential Deposition by  Subclass0 and Site

                                          (a)  Sulfate ion

                       Class/Subclass*         Mean Potential Deposition
                                                 MS     MM     WF    WT

                            I/none               0.40       -       0.48
                            3/Ntraj.               -      0.39      -     0.20
                            3/non-N traj.          -      0.97     0.17   0.10
                            4/OH traj.           0.23      0.44     0.67    1.69
                            4/non-OH traj.        -      0.54     0.28   0.39
                            6/Etraj.             0.05      0.33      -     0.29
                            6/non-E traj.          -       -        -      1.63
                            9/none                 -       -        -     0.55
                            other classes          -      1.20     0.002   0.78
                                               4-53
-------
                                             Table 4-26 (continued)


                           Mean Potential Deposition by Subclass0 and Site

                                           (b)  Nitrate ion

                       Class/Subclass*         Mean Potential Deposition
                                                  MS     MM    WF    WT
I/none
3/N traj.
3/non-N traj.
4/OH traj.
4/non-OH traj.
6/E traj.
6/non-E traj.
9/none
other classes
0.23
-
-
0.14
-
0.02
-
-
-
_
0.12
0.31
0.20
0.18
0.16
-
-
0.51
0.14
-
0.05
0.24
0.19
-
-
-
0.002
_
0.03
0.10
0.50
0.11
0.20
0.69
0.23
0.33
                       a.   Results shown only for those subclasses having a minimum of 10 hr of
                            observations.  At Whitetop (WT), Fp is for plot A (summit).
                       b.  "None" indicates no trajectory subclass considered.


        In simple terms, R, is the canopy-top horizontal cloud flux of ion X. Thus, high values of R, are
associated with windy conditions, dense  clouds, and/or high cloud water ion content. Mean values of Fp
vary greatly from one subclass to another and from one site to another.   If all other conditions
affecting cloud deposition were identical at  all the sites  (this means, among other things, that site-to-site
variations in cloud droplet size distribution and canopy structure would be insignificant), then the mean
hourly cloud deposition flux would vary according to R, alone.   Values of Fp>l /*eq m~2 s'1 were found
for only two sites. Class-weighted mean  values  of R, for each site (Table 4-27) show that the greatest
overall cloud deposition potential (using the subclasses shown in  Table 4-26 and weighting subclass-
mean Fp values by the relative frequencies of observed wind speed for each subclass) for sulfate and
nitrate is at Whitetop.  This conclusion  is valid  only for those specific forest canopies considered here.
For any given mountain, different forest canopies exposed to different meteorological conditions  could
have considerably lower or  higher deposition potentials.
                                                 4-54
-------
                                            Table 4-27

                       Class-weighted Deposition Potential by Site for 1986-88
                       Site0           Class-weighted Deposition Potential (jceq/m2s)
                                                    Sulfate         Nitrate
MS
WF
MM
WT
0.27
0.35
0.54
0.64
0.16
0.12
0.19
0.24
                       a.  The stand on MS is on a ridge, but not at the highest point on the
                          mountain.  The stand on MM is near the mountain summit.  The stand on
                          WF is several hundred meters below the summit (no trees are on the WF
                          summit).  The WT stand is near the summit.


        Another measure of deposition potential-the growing season potential exposure (Pe)--takes into
account the frequency of cloud  impaction.  As defined here, Pe represents the total cloud ion mass
passing horizontally at canopy top through a unit area during a  growing season.  Thus, Pe=t
where t is the growing season cloud  exposure time and  is the weighted mean value of Fp listed
in Table 4-27.  Combining the data in Tables 4-24 and 4-27, one finds that the potential growing season
exposure to sulfate and nitrate  cloud deposition was also a maximum at the Whitetop site for every
year of the analysis.  If the average cloud collection  efficiency was known for a given stand (this is what
the COM is designed to compute)  and the total collecting surface area of the stand per unit of ground
area (SAI)  was known, then the values in Table 4-28 could be used to estimate the growing season
cloud deposition total.
                                               4-55
-------
                                            Table 4-28

             Growing Season Cloud Deposition Potential Exposure' (Pe) by Site and Year

       Site            Year           Growing Season Potential Exposure to Cloud Deposition0
                                             Sulfate (eq/m2)        Nitrate (eq/m2)
MS



WF



MM



WT



1986
1987
1988
1986-88
1986
1987
1988
1986-88
1986
1987
1988
1986-88
1986
1987
1988
1986-88
1.0
0.8
0.3
0.7
1.6
1.5
0.9
1.3
2.5
2.3
1.7
2.2
3.5
2.8
2.4
2.9
0.6
0.5
0.2
0.4
0.6
0.5
0.3
0.5
0.9
0.8
0.6
0.8
1.3
1.0
0.9
1.1
       a  The Pe units used here are not to be interpreted as depositable [H]™ equivalents per unit of
       ground area, but equal instead to the growing season horizontal flux ofjH]^ equivalents
       through a vertical plane  at canopy top.


       Hourly cloud water, sulfate and nitrate deposition estimates  computed for each site using the
CDM and class-mean meteorological and chemical variables are listed in Table 4-30 by class/subclass.
Deposition was computed for both CDM evaluation stands on Whitetop (plots A and B mentioned
earlier in this  section) to illustrate the  expected differences in deposition at two sites in relatively close
proximity on the same mountain.  WT Plot A was the open stand of low  density and plot B was closed
and dense.
                                                4-56
-------
                              Table 4-29
1986-88 Computed Mean Hourly Cloud Deposition by Site0 and Subclass

                            (a)  Cloud water

Class/subclass                      Mean Computed Flux (mm/hr)
                                 MS    MM    WF   WT(A) WT(B)
   I/none
   3/N traj.
   3/other traj.
   4/OH traj.
   4/non-OH traj.
   6/E traj.
   6/non-E traj.
   9/none
   other classes
   0.11


   0.15

   0.11
0.36
0.31.
0.21
0.25
0.26

0.09
0.15
0.11

0.10
0.09
0.09
                   0.06
                   0.04
0.29
0.19
0.49
0.22
0.46
0.60
0.27
0.29
0.68
0.21
1.29
0.18
0.26
1.25
0.55
0.63
Class/subclass
ClassAsubclass
(b) Sulfate ion

       Mean Computed Flux (eq/ha/hr)
    MS    MM    WF   WT(A) WT(B)
I/none
3/N traj.
3/other traj.
4/OH traj.
4/non-OH traj.
6/E traj.
6/non-E traj.
9/none
other classes
0.43
1.15
1.81
0.54 1.96
0.79
0.11 0.50
-
0.90
1.07
0.13
-
0.12
0.30
0.26
-
-
0.32
0.01
„
0.59
0.30
2.40
0.56
0.60
3.53
2.83
0.87
.
1.30
0.33
4.94
0.48
0.39
6.79
5.41
1.65
(c) Nitrate ion

       Mean Computed Flux (eq/ha/hr)
    MS     MM    WF   WT(A) WT(B)
   I/none
   3/N traj.
   3/other traj.
   4/OH traj.
   4/non-OH traj.
   6/E traj.
   6/non-E traj.
   9/none
   other classes
   0.24     -      0.06
           0.38     -      0.22     0.49
           0.59    0.03    0.16     0.18
   0.32    0.94    0.04    0.71     1.46
           0.28    0.10    0.22     0.19
   0.05    0.21     -      0.29     0.19
                           1.67     3.22
           0.39    0.17  .  0.96     1.83
           0.41    0.003   0.43     0.81
a.   Values shown only if hours of data exist for a subclass
                                 4-57
-------
        The large variations between subclasses in computed water and  ion flux was expected because of
the large variations in meteorology and chemistry.  Uncertainty in these flux estimates is unknown, but is
expected to be larger for MS, MM and WF because of the greater  uncertainty in the inputs to the
model for these sites,  It is also expected to be larger for those subclasses for which relatively few hours
of data were collected (a mininum of 10 hours was used as a  selection criterion).  The WT flux
estimates may be 20% - 30% higher  for both plots, based on  results of the  CDM testing at these sites.
However, some  assumptions were necessary (e.g., previously analyzed relationships  between summit
wind speed at 7 m and canopy top wind speeds at both plots was assumed  to hold for the entire 3-year
period)  to run the CDM for both plots A and B because site-specific wind speed and liquid water
content  data, representative of canopy top, were not always part of the subclass analysis done by Vong
et al.  (1989).

        One interesting feature of note is the difference in computed cloud deposition between WT plots
A and B.  Deposition is sometimes  greater at one and sometimes greater at the other, depending  on
subclass. This occurs because of the variation in prevailing wind direction among subclass type.
Subclasses with a prevailing direction from southeast through west produce lower water deposition
because of the sheltering effect of the mountain-plot B is in  the  lee of the mountain under  these
conditions, near the location of the lee eddy that has been observed frequently in videotaped  images of
the mountain. However, wind speeds are very similar for other directions, and the greater surface area
at plot B causes deposition estimates to be  much higher during these conditions.

        Yearly growing season estimates  of cloud deposition were  computed using the synoptic  class
frequency data and the mean hourly  flux estimates shown in Table 4-29.  For each site, the
subclass-mean flux estimates include  subclasses that account for more than 75% of  the total  cloud
impaction hours.  The mean hourly flux during the  unrepresented periods was assumed to equal the
computed mean averaged over the represented periods. This  was not expected to introduce much  bias
into the estimated deposition fluxes.  However, int the future, other estimation techniques should also be
tested.  Table 4-30 summarizes the computed growing season  deposition totals for each site.
                                                4-58
-------
                                   Table 4-30

           Computed Growing Season Total Cloud Deposition Flux by Site
Site    Year
       Cloud Water
Flux (on)         [H]
  Cation Deposition (eq/ha)/(kg/ha)
[NH4]        [K + Mg + Ca + Na]
MS 1986
1987
1988
86-88
WF 1986
1987
1988
86-88
MM 1986
1987
1988
86-88
WT(A) 1986
1987
1988
86-88
WT(B) 1986
1987
1988
86-88
12.9
10.1
3.5
8.8
11.6
10.5
6.0
9.4
33.5
27.5
21.8
27.6
53.1
39.4
34.3
42.3
77.3
61.2
54.3
64.3
486/0.5
349/0.3
141/0.1
325/0.3
144/0.1
156/0.2
81/0.1
127/0.1
1205/1.2
1114/1.1
855/0.9
1058/1.2
1514/1.5
1189/1.2
1047/1.0
1250/1.2
2454/2.5
1983/2.0
1764/1.8
2067/2.1
189/ 3.4
135/ 2.4
55/ 1.0
126/ 2.3
107/ 1.9
96/1.7
64/1.2
89/ 1.6
377/ 6.9
377/ 6.9
278/ 5.0
344/ 6.3
780/14.0
602/10.8
525/ 9.5
636/11.4
1229/22.1
985/17.7
866/15.6
1027/18.5
23/-
17/-
11-
161-
161-
111-
10/-
14/-
100/-
113/-
83/-
99/-
149/-
110/-
93/-
117/-
221/-
170/-
144/-
178/-
                                      4-59
-------
                                            Table 4-30 (continued)
       Site           Year                        Afflon Deposition (eq/ha)/(kg/ha)
                                             [SO4]              [NO3]              [a]
MS



1986
1987
1988
86-88
435/21
313/15
127/6
292/14
250/16
178/11
73/5
167/11
18/0.6
13/0.5
5/0.2
12/0.4
       WF            1986                   220/11            54/3               6/0.2
                       1987                   202/10            67/4               5/0.2
                       1988                   128/6            31/2               3/0.1
                       86-88                  183/9            51/3               6/0.5

       MM            1986                  1091/52            406/25              68/2.4
                       1987                  1052/51            396/25              64/2.3
                       1988                   791/76            296/18              48/1.7
                       86-88                  978/60            366/23              60/2.1

       WT(A)         1986                  1737/83            667/41              87/3.1
                       1987                  1337/64            522/32              62/2.2
                       1988                  1173/56            463/29              54/1.9
                       86-88                 1416/68            551/34              68/2.4

       WT(B)         1986                  2817/135          1053/65              105/3.7
                       1987                  2244/108           858/53              82/2.9
                       1988                  1981/95           769/48              74/2.6
                       86-88                 2347/113           893/55              87/3.1
        First note that uncertainties in canopy structure, and hence wind speed profile and droplet
collection efficiency could easily make these values biased by a factor of two for all sites but WT. These
values indicate changes relative to previous MCCP estimates of cloud deposition.  For example,
compared to the 1988 estimates for the 1987 growing season, the current 1987 sulfate deposition totals
are higher for MS (+81%), lower for WF (-78%), higher for MM (+70%) and lower for WT(A)
(-16%).  To some extent these  differences reflect  changes in the COM, but are mostly due to the more
detailed  characterization of the forest canopy which became available to MCCP through the efforts of
the USDA Forest Service.  In addition, improvements in characterizing the meteorology and chemistry
associated with the various cloud  event synoptic classes also necessitated a recalculation of the 1987
interception values.  At WF the differences are also due to different  assumptions concerning the wind
speed profile, canopy structure  and spatial variability of cloud water ion concentrations.

        Another interesting finding is the large deposition flux estimated for WT(B)  compared to
WT(A).  This is in spite of the fact  that plot B at WT is occasionally sheltered in the lee of the
mountain and intercepts less  cloud water during such periods. The primary reason that WT(B) estimates
are larger than those for WT(A)  is the much greater canopy density  of the former.  COM performance
for WT(B)  has been tested (Mueller et al., 1990) and found to  be good (within 20% of throughfall
rates).   At WT, unlike the other  sites, different wind directions were considered when computing
deposition for the synoptic classes.  This was done to maximimze the accuracy of the wind speed

                                                4-60
-------
adjustments needed to convert from 7 m  speed at the summit to canopy top speed.  There is a
possibility that the partitioning of wind directions for the synoptic classes (which  was done using only
1988 data) was not representative of all years of the analysis.  The data indicate that west-northwest is
the most  prevalent wind direction for most synoptic classes, with south- southeast often being the
second most prevalent.  Overall, W and WNW winds-which do not place the WT(B) site in the
mountain lee-appear to be twice as frequent as SE and SSE winds. Thus, on a long-term basis,  the
sheltering effect may not be as important  as previously believed in its influence on deposition at  this
site.

        The greatest computed growing season cloud deposition fluxes  were found for WT and MM
which also had the longest growing seasons. The differences between MS and WF may be greater than
they appear here because  of the uncertainty in model input data for these two  sites.  The mean
monthly computed cloud deposition flux is provided  in Table 4-31 to facilitate comparison of deposition
rates  independent of growing season length. Despite an almost 2:1 advantage in cloud frequency, WF
mean monthly deposition estimates were  generally lower than those for MS. The primary reason for
the difference is the lower  canopy surface area  at the  modeled WF site. Differences between the
northern (MS and WF) and  southern (MM and WT) sites are so great that they are likely to be real
despite  uncertainties in the model, and they exist independent of the longer growing season in the south.


                                             Table 4-31

                          Mean Monthly Computed Cloud Deposition Flux

Site    Year           Computed Water        Computed Ion Flux (kg/ha/month)
                       Flux (mm/month)      [H]    [NH4] [SO4]   [NO3]  [Cl]
MS



1986
1987
1988
86-88
25
20
7
17
0.10
0.07
0.03
0.06
0.67
0.47
0.19
0.44
4.1
2.9
1.2
2.7
3.0
2.2
0.9
2.0
0.12
0.09
0.04
0.08
WF    1986                   29            0.04    0.48     2.6     0.8   0.05
        1987                   26            0.04    0.42     2.4     1.0   0.05
        1988                   15            0.02    0.28     1.5     0.5   0.03
        86-88                  23            0.03    0.39     2.2     0.8   0.04

MM    1986                   60            0.22    1.21     9.4     4.5   0.43
        1987                   49            0.20    1.21     9.0     4.4   0.40
        1988                   39            0.15    0.89     6.9     3.3   0.30
        86-88                  49            0.19    1.11     8.4     4.0   0.38

WT(A) 1986                   95            0.27    2.51    14.9     7.4   0.55
        1987                   70            0.21    1.94    11.5     5.8   0.39
        1988                   61            0.19    1.69    10.0     5.1   0.35
        86-88                  75            0.22    2.04    12.1     6.1   0.43

WT(B)  1986                   138            0.44    3.95    24.1    11.7   0.67
        1987                   109            0.35    3.17    19.2     9.5   0.52
        1988                   97            0.32    2.78    17.0     8.5   0.47
        86-88                  115            0.37    3.30    20.1     9.9   0.55
                                                4-61
-------
       The cloud deposition information contained in this report has been produced using
state-of-the-science computational methods.  The state of the science is such that considerable
uncertainty remains in the deposition estimates despite recent advances in our knowledge concerning
cloud deposition to  forest canopies. However, the  deposition potential-as defined previously--and the
computed  deposition fluxes both indicate trends toward higher cloud deposition at the southern sites.
Differences between northern and  southern sites are due to site specific parameters and not latitude.
The southern sites have a significantly different canopy structure than the Whiteface Mtn.  site.  The
southern sites are also exposed to a different mix of cloud event types (as defined by synoptic
meteorology and computed air trajectories) when compared to  the northern sites, and different event
types  have been shown to be characterized by differences in conditions affecting cloud deposition. More
work  needs to be done to better define conditions at each site before more accurate deposition estimates
can be obtained.
Dry Deposition

Model Data Requirements --

        The Atmospheric Turbulence Diffusion Laboratory (ATDL) model used by the MCCP,
DRYDEP, calculates the dry deposition velocities of sulfur dioxide, ozone, nitric acid vapor, sulfate, and
nitrate from meteorological and site-specific biological information.  Site information includes: major and
minor plant species type, leaf area index for plant species, and site location.  Stomatal resistance
parameters include plant species, light response coefficients, minimum stomatal resistance, and optimum,
maximum, and minimum temperatures for stomatal function.  All  available meteorological data such as
wind speed, standard deviation of direction, global radiation, air temperature, relative humidity, canopy
wetness, and rainfall are also used.

Model Application --

        The current big-leaf model does not directly account for nonuniform terrain.  Only Rowland met
the model requirements.  Therefore, to estimate dry deposition at  the other sites, the model was run
with the aerodynamic resistance set to zero in  an  attempt to estimate the maximum effect of an uneven
surface. Table 4-32 shows the warm period and the species for which deposition flux density was
inferred at each  site.
                                           TABLE 4-32

                         Chemical Species for Which Deposition was Inferred

Site                              Period                 Species

Rowland, ME                   4/7 - 10/12        O3  SO2 HNO3  SO4 NO/
Whiteface, NY                  6/1 - 10/6         O3  SO2                   SO^
Moosilauke, NH                 5/15 -  10/21       O3
Shenandoah, VA                4/1 - 11/16        O3  SO2 HNO3  SO4 NOj
Mitchell, NC                    5/1 - 10/21        O3  SO2                      NO,.
Whitetop, VA                   4/26 -  10/18       O3  SO2
                                               4-62
-------
Model Sensitivity Testing --

        The effect of estimating deposition velocities when setting aerodynamic resistance to zero varies
by chemical species.  For example, the effect for ozone, which has considerable canopy resistance, would
be less than the effect for HNOj, for which canopy resistance is assumed to be zero.  Ozone deposition
velocities for 1987 Rowland data showed a 14% increase when calculated with zero resistance.  HNOj
deposition velocities calculated with and without aerodynamic resistance  for Shenandoah showed an
offset of 0.36 cm/sec, and an increase of 59% when calculated with zero resistance.

     .   Some uncertainty was included  in leaf area index (LAI) values used for most  sites; only the
values for Mitchell were obtained by direct measurement. Therefore, the sensitivity of deposition velocity
to changes in LAI was evaluated using ozone at Rowland.  The primary  and secondary tree species have
LAIs of 5.5 and 0.2, respectively. Deposition velocity (Vd) was calculated with primary LAI values from
2.5 to 11.0, with the secondary LAI held constant.  For each of seven values of LAI,  the resulting Vd
values were averaged for the period of record.  Paired values of normalized LAI and  Vd were deter-
mined using the accepted values as the  base. The results showed that if estimated LAI values were in
error by as much as  25%, Vd would either be overestimated by 12% or underestimated by 15%.  This is
probably within the uncertainty of the model estimates. The current model also tended to overestimate
deposition velocities for chemical species for which canopy resistance is important, such as ozone and
SO2, (Matt and Womack, 1988; Matt et al., 1987).


Model Results ~

        The warm season total deposition flux density (kg/ha/mo) is a  function of deposition velocity,
concentration, and season length. Table 4-33 displays the available dry deposition values for 1987 and
1988.  The values were  derived by multiplying weekly deposition velocities derived from the model  by the
appropriate weekly concentrations for the warm period at each site.  Estimates based  on the mean  were
substituted for missing data. Estimates for 3 to 6 missing weeks yielded reasonable results.  However, as
in the case of SC>2 and  SO^ for Shenandoah-1, substituting estimates  for more than 50% of the warm
period led to anomalous results.

        During 1987 ozone values for Whiteface seem low due to  moderate deposition velocities and the
short season.  For that year,  Moosilauke and Whiteface, if scaled to the Rowland warm season, would
have ozone deposition values of approximately 6.81 and 9.75 kg/ha/mo, respectively.  On the other hand,
if Shenandoah was similarly scaled down, ozone deposition would still  be the largest with 9.29 kg/ha/mo.

        Although ozone exposure was considerably greater in 1988 than in 1987, deposition flux density
was  approximately the same for both years for three northern sites due to a marked decrease in
deposition velocities during 1988. Deposition velocities also decreased somewhat at the southern sites,
but this was offset by very much higher average ozone concentrations.

        Sulfur dioxide concentrations were also higher during 1988 at all reporting sites.  However, SO2
deposition velocities were uniformly lower, producing only modest increases in deposition flux density at
Rowland and Whiteface.  Only Mt.  Mitchell recorded a substantially higher SO2 flux  density during
1988.

        In general, for chemical species with sufficient available data, northern sites received less  dry
deposition than southern sites. However, within each region there was marked variability in seasonal
ozone and SO2 deposition. Also, seasonal deposition totals did not adequately reflect  weekly site-specific
variability.


                                                4-63
-------
                                          TABLE 4-33

                               MCCP Wann Season Diy Deposition

Site                          Elev.(m)                  Diy deposition flux density
                                                           (kg/ha/mo)

                                                       O3    SO2    SQ4  HN05   NOj   NO2

Rowland, ME (1987)                65                 5.48    0.22   0.11    0.29    0.002
          (1988)                                       5.39    0.28   0.15    0.37    0.03

Moosilauke, NH (1987)            1000                 5.%
          (1988)                                       6.57           0.31    1.61    0.01

Whiteface, NY (1987)             1483                 6.46    0.30   0.36
          (1988)                                       6.09    0.51   0.28

Shenandoahl, VA (1987)          1015                14.06    3.06   0.80    0.007
          (1988)                                      14.83

Shenandoah3, VA (1987)           524                13.36    1.32   0.19    1.03    0.005
          (1988)                                      15.11

Mt. Mitchelll, NC (1987)          1950                 7.70    0.39                          0.20
          (1988)                                      12.7    1.48           1.64            0.21

Mt. Mitchell2, NC (1987)          1775                 7.81
          (1988)                                      10.56

Whitetop Mt, VA (1988)          1689                11.95    0.75


Precipitation Deposition

          In this section, wet deposition for the 1987 and 1988 warm seasons at the MCCP sites is
estimated using standard NADP/NTN or MAP3S measurements of precipitation amount  and chemistry.
These estimates were obtained from NADP and MAP3S reports for the network stations closest to the
MCCP sites or, in one case, for an  NADP-type collector operated at the Shenandoah MCCP site.

          To represent wet deposition via precipitation at the Rowland, ME, site, Mohnen (1988a) used
data from the NADP/NTN site at Greenville, ME, located approximately 100 km west of Rowland.
Other MCCP sites had NADP/NTN type samplers no more than 25 km distant.  Details  for these
stations are  shown in Table 4-34.

         .For comparison with MCCP cloud water deposition, wet deposition only for the warm season
are reported as  monthly deposition.  The growing season is  longest at the Shenandoah and Rowland
sites, extending  from early April at  both locations to mid-November and early October, respectively.  Mt.
Mitchell and Whitetop Mountain have the next longest growing seasons, followed by Mt. Moosilauke and
Whiteface with  the shortest (June to early October).  For consistency the concurrent growing season


                                              4-64
-------
(June to October) was used for 1987 data and all available data (until August 23) were used for the
same period during 1988.


                                          TABLE 4-34
               Sites Used to Estimate Wet Deposition via Precipitation for MCCP Sites.

MCCP Site            Wet Deposition Site           LaL        Long.      Elevation

Howland Forest, MA   NTN Greenville ME09         45°9'       69°9'         322 m
Whiteface Mtn, NY    NTN Whiteface NY98          44°3'       73°!'         622 m
                      MAP3S Whiteface             	(co-located)	
Moosilauke, NH       NTN Hubbard Brook NO2     43°6'       71°2'         250 m
Shenandoah, VA       NTN Big Meadows VA28       38°!'       78°6'        1074 m
Whitetop, VA         NTN Whitetop VA29          36°8'       81°6'        1689 m
Mt Mitchell, NC       NTN Clingmans Peak NC45     35°44'      82° 17'      1987 m


       MCCP pollutant deposition and climatological data tend to form two distinctly different groups.
Because of physical proximity, and similar influences from synoptic systems, the Howland, MA, Whiteface
Mountain, NY,  and Mt. Moosilauke, NH,  sites are considered together as northern sites and the
Shenandoah, VA, Whitetop, VA, and Mt.  Mitchell, NC,  sites are grouped together as southern sites.

       The wet deposition data for the two groups of sites are shown in Table 4-35 [kg/ha/mo] for
SCXf2", NOy, NH/+, and H+ ions in precipitation (not cloud) collected at  the NADP/NTN sites.  At
Whiteface Mountain, NY, both NADP/NTN and MAP3S collected precipitation, but during 1987 the
NADP/NTN site was hit by lightning, so only the MAP3S data were used.  Rain gauge data were used to
calculate precipitation amount. About 98% of the precipitation was analyzed for chemical composition
during 1987.  Missing samples were replaced by actual precipitation amount for that week and the
volume-weighted mean concentrations of growing season data for that site and year. A further discus-
sion of the calculation procedures is  presented by Mohnen (1988a).

       The data indicate that the most westerly sites in the  north, Whiteface, NY, and Moosilauke, NH,
received greater wet deposition via precipitation  than did the more northeasterly location in Maine.  Big
Meadows, VA, and nearby Shenandoah had the most NH/+ deposition from rain.

       These data suggest that the highest elevation site, Mt. Mitchell, NC, had the largest wet flux for
SO/2" and NO/.  Hubbard Brook, NH, (a low elevation site in the north), Shenandoah, VA (a medium-
-elevation site in the  south), and Whitetop Mountain, VA, (the second highest site) had similar wet
deposition for 1987.  The lowest site, Greenville, ME has the least wet deposition via precipitation.

       For comparison, available data for the 1988 growing  season are presented in Table IV-18 for
several MCCP sites.  Data for all of the same sites as for 1987 are not currently available.

       The 1988 precipitation chemistry data indicate that the southern sites exceeded the northern
sites in SO/2' and NOf deposition, with the differences being greater for SO/=.  Shenandoah had the
most NH/"1"  deposition.  These values for  wet deposition via  precipitation are about 20% to 50% of the
deposition via cloud water interception during 1987.
                                              4-65
-------
                                           TABLE 4-35

            Concurrent Warm Season Wet Deposition via Precipitation in the Eastern USA
                from NADP/NTN or MAP3S Sites Near MCCP Monitoring Locations
             (for the period 26 May to 6 October, 1987, or 31 May to 23 August, 1988.)
                                          [kg-ion/ha/mo].
                            Data are (invalidated and  subject to change.

       Location               H+        NH,+      SO^~      NO/      Year

       Greenville, ME           *        0.11        1.28        0.65        1987

       Whiteface, NY           *        0.33        2.27       ,0.95        1987
                              .032        1.11        1.18        0.93        1988

       Hubbard Brook, NH      *        0.28        2.84        1.46        1987
                              .041        0.11        2.16        1.10        1988

       Whitetop, VA           *        0.31        3.20        1.32        1987
                              .053        0.14        2.84        1.17        1988

       Mt.  Mitchell, NC         *        0.46        5.15        2.26        1987
                              .051        0.12        2.69        1.19        1988

       Big  Meadows, VA        *        0.48        2.32        1.43        1987
                              .016        0.92        2.90        1.34        1988

       Shenandoah, VA         *        0.35        3.11        1.25        1987
                              .045        1.76        2.80        1.19        1988
  Not reported
Wet. Cloud and Dry Deposition
        Cloudwater deposition is the factor which differentiates mountains from lowlands and the
relative importance of cloud and rain deposition is discussed here.  It is hypothesised that the principal
deleterious effects on forests which might be associated with such deposition are direct effects of acids
on foliage, mobilization of aluminum because of ions deposited to soils, or excess nitrogen fertilization
influence on winter damage.  Since most of these effects are more significant for warm season deposi-
tion, the analysis intervals were selected for each site to correspond roughly with the periods between
local forest bud break and fall dormancy.  These periods differed to some extent for the  MCCP  sites.
        However, cloudwater deposition calculated for Whitetop Mountain seems to stand out above the
rest. Since this peak is substantially lower than Mount Mitchell which is only about 90 miles southwest
of Whitetop,  the difference is difficult to explain.
                                               4-66
-------
        It is apparent from Tables 4-31, 4-33 and 4-35 that cloud deposition based on current models
estimate, is comparable to deposition of rain at all summit locations. Dry deposition of sulfur-bearing
compounds is reported as sulfur rather than as SC«2 or sulfate anion (Table 4-33).

        If the geographic gradients suggested by the data in Table 4-33 continue in future estimates, then
some interesting conclusions could be drawn.  First, the  Shenandoah site seems to experience the
greatest dry deposition.  In the case of both sulfur and nitrogen budgets, dry deposition is significant in
the three Northern sites.  This gradient reflects the wet deposition gradient reported in both the
National Academy of Sciences findings on acid rain and  in the 1989 NAPAP interim assessment.  The
more easterly sites, Moosilauke and  Howland are very much lower in pollutant burden than are the
Whiteface and southern sites. With the exception of Moosilauke and Howland, the MCCP site warm
season sulfur deposition  is 20-40 kg/ha-year.  This level is about that reported for annual deposition at
the more contaminated low elevation sites in rain.

        It is not yet clear whether cold season  sulfur deposition should really be factored in to the
mountain estimates. There is some  evidence of SC>2  loss from snowpack and the excess  ions in snow-
pack melt are of much greater consequence in  the acidification of lakes than they might be in determin-
ing ion concentrations in frozen soils. In any case, it is clear that mountaintop sulfur deposition is quite
significant, approximating or exceeding that of  near-source low elevation sites.

        For all of the data reported so far, dry deposition sulfate ion in fine particles is  very small
relative to wet deposition and in many cases relative to that of SC>2 dry  deposition as well. Dry
deposition of nitrogen is more uncertain because of the lack of dependable data  for nitric acid.
However, ammonium ion dry deposition accounts for a substantial amount of the total nitrogen
deposited; this material appears to be mainly in the form of ammonium sulfate.  The season-long values
for Shenandoah for nitric acid try deposition may be  biased somewhat low of the actual values, because
most of the determinations were performed relatively late in the study period.  In any case, the nitric
acid dry deposition appears to be quite significant for that  site, and probably for the other southern sites
as well.

        Overall growing  season fertilization rates are 10-20 kg N/ha-season for the southern sites and
lower for the northern ones. This application of nitrogen is driven mainly by wet deposition and the
estimates could go up or down as model predictions are  improved.

        Table 4-36 summarizes the average monthly sulfur  and acidic nitrogen (excludes  ammonium ion)
deposition fluxes determined to-date for the MCCP sites for the 1987-88 growing seasons. At the cloud-
free Howland  site, dry deposition appears to account  for less than one-third of the total  acidic sulfate
and nitrate ion deposited.  The Shenandoah site has a very low cloud impaction frequency due to its low
elevation, and dry deposition is probably of greater relative importance there (it could exceed 50% of the
total deposition flux) than at any of the other high elevation sites.  Estimates for Whiteface suggest that
dry sulfur deposition accounts for less than 30% of the total sulfur  deposition flux.

        The greatest amount of information is  available for comparing estimated cloud and wet
deposition at the four high-elevation spruce-fir  sites (Moosilauke, Whiteface, Whitetop and Mitchell).
Cloud-to-wet deposition  ratios (Table 4-37) vary from near unity at Moosilauke and Whiteface to near 4
at Whitetop.  In terms of acid deposition, clouds  seem to be of greater relative importance at the two
southern sites (WT and MM) than at the two northern sites.
                                                4-67
-------
Site
Year
                     TABLE 4-36

                          (a)

Estimated 1987-88 Sulfur Deposition Budgets at MCCP Sites'1

                    Growing Season Mean S Deposition (kg S/ha-mo)b

          Wet (Measured)        Qoud (Computed)     Dry (Computed)0
Howland, MEd    87
                 88
                87-88
                0.43
                                                         0.15
                                                         0.19
                                                         0.17
Moosilauke, NH* 87
88
87-88
Whiteface, NY 87
88
87-88
Shenandoah, VA 87
88
87-88
Whitetop, VAf 87
88
87-88
Mitchell, NC 87
88
87-88
a. Not all data were
b. Average fluxes for
0.95
0.72
0.84
0.76
0.39
0.58
1.04
0.93
0.99
1.07
0.95
1.01
1.72
0.90
1.31
available in time for inclusion in
combined 1987-88 period were c
0.97
0.40
0.69
0.80 0.27
0.50 0.35
0.65 0.31
1.80
3.83
3.33
3.58
3.00
2.30
2.65
this report.
omputed by averaging values given separately
       for the two years.
c.      Sulfuf (S) flux computed as the sum of individual computed SO2 gas and SOj aerosol fluxes
       (approximate maximum rates due to assumption that aerodynamic resistance within forest canopy
       was zero).
d.      Greenville, ME  NADP data used for wet deposition fluxes.
e.      Hubbard Brook, NH NADP data used for wet deposition fluxes.
f.      Near-summit (plot A) site.
g.      Nitrogen in ammonium ion not considered.
h.      Nitrogen (N) flux computed as the sum of individual computed HNOj gas and NOj aerosol
       fluxes (approximate maximum rates due to assumption that aerodynamic resistance within forest
       canopy was zero).
                                             4-68
-------
                                         TABLE 4-36 (continued)
Site
Estimated 1987-88 Acidic Nitrogen Deposition Budgets at MCCP Sites'^

                         Growing Season Mean N Deposition (kg S/na-mo)6

              Wet (Measured)       Cloud (Computed)     Dry (Computed)*
Year
Rowland, MEd    87
                 88
                 87-88
                  0.15
                                                             0.07
                                                             0.09
                                                             0.08
Moosilauke, NHe 87
88
87-88
Whiteface, NY 87
88
87-88
Shenandoah, VA 87
88
87-88
Whitetop, VAf 87
88
87-88
Mitchell, NC 87
88
87-88
a. Not all data were
b. Average fluxes for
0.33
0.25
0.29
0.21
0.21
0.21
0.28
0.27
0.28
0.30
0.26
0.28
0.51
0.27
0.39
available in time for inclusion in
combined 1987-88 period were c
0.50
0.20
0.35
0.23
0.11
0.17
0.23
1.31
1.15
1.23
0.99
0.75
0.87
this report.
omputed by averaging values given separately
       for the two years.
c.      Sulfuf (S) flux computed as the sum of individual computed SC«2 gas and SO^ aerosol fluxes
       (approximate maximum rates due to assumption that aerodynamic resistance within forest canopy
       was zero).
d.      Greenville, ME NADP data used for wet deposition fluxes.
e.      Hubbard Brook, NH NADP data used for wet deposition fluxes.
f.      Near-summit (plot A) site.
g.      Nitrogen in ammonium ion not considered.
h.      Nitrogen (N) flux computed as the sum of individual computed HNOj gas and NOj aerosol
       fluxes (approximate  maximum rates due to assumption that aerodynamic resistance within forest
       canopy was zero).
                                             4-69
-------
                                            Table 4-37

                           Estimated Cloud-to-Wet Deposition Flux Ratios

        Site                   Year                  Sulfate                 Nitrate
MS


WF


WT°


MM


1987
1988
87-88
1987
1988
87-88
1987
1988
87-88
1987
1988
87-88
1.0
0.6
0.8
1.1
1.3
1.1
3.6
3.5
3.5
1.7
2.6
2.0
1.5
0.8
1.2
1.1
0.5
0.8
4.4
4.4
4.4
1.9
2.8
2.2
a.       Near-summit plot.


        In summarizing the deposition results, MCCP and IPS researchers have estimated that
cloudwater deposition provides a substantial fraction of the total chemical deposition to high-elevation
eastern  USA forests.  Lindberg and Johnson (1989) estimated that cloudwater contributes approximately
25% to  50% of total SO^, N, and H+ deposition at two high-elevation  sites on Whiteface Mountain,
NY and in the Great Smoky Mountains National park, NC.  MCCP results indicate that cloudwater
SO^2', H+, NO/, and NH^+ deposition exceeded wet deposition via precipitation for three sites located
above 1400 m, while two lower elevation sites  received cloudwater chemical inputs that were at least
one-half of precipitation chemical deposition.  Preliminary estimates for dry deposition are considerably
smaller  than either cloud or precipitation deposition at all MCCP sites except one site located in a
deciduous forest at Shenandoah,  VA (Mohnen, 1988a; Krovetz et al., 1989).  While  the current large and
undefined uncertainties that are associated with cloud water and dry deposition  models make firm
estimates for complex terrain impossible, there is a general consensus in the literature that cloud water
deposition represents a substantial increment over precipitation and dry inputs to high-elevation forests
in the eastern USA
                                               4-70
-------
                                           SECTION 5

                     METEOROLOGICAL ASSESSMENT OF AIR QUALITY


CASE STUDY ON THE IMPACT OF AIR MASS ORIGIN ON AIR QUALITY

        A simple case study was performed at a northern MCCP site to investigate the relationship
between synoptic weather patterns and cloudwater ion and gas phase pollutant concentrations.  If
meaningful results could be obtained, then there would be a basis to extend this type of analysis  to other
sites and to other synoptic  categories.

        In this case, ion concentrations during warm sector synoptic conditions were investigated for the
1986-88 field seasons at Whiteface Mountain.  Warm sector conditions are defined as the southeast
sector of the classical cyclone model or the west side of a high pressure system. Stagnant conditions
sometimes occur with semi-permanent "Bermuda High" type situations which often lead to air pollution
episodes.  Therefore, warm sector conditions were classified as "non-stagnant" and "stagnant", the latter
requiring that warm sector, high pressure conditions at or to the south of the site were persisting for at
least 72 hours.

        Table 5-1 shows the concentrations of selected ions for cloudwater and pollutant concentrations
for gases and sulfate aerosol for three categories: non-warm sector (category 0); non-stagnant warm
sector (category  1) and stagnant warm sector  (category 2) conditions based on hourly samples.  For all
ions, gases  and sulfate aerosol, both warm sector concentrations are greater than other sector concentra-
tions by a factor of 2 to 3,  indicative of the relatively warm and humid conditions, together with
favorable south to southwesterly trajectories, which favor high pollutant levels. Stagnant conditions
exhibited the maximum  sample  concentrations for all ions, gases and sulfate aerosol.


METEOROLOGICAL INFLUENCES ON CLOUDWATER CHEMISTRY

        Air mass back-trajectories during all hours at all MCCP sites originated most frequently  from
the western quadrant, and least frequently from the southeast.  For cloudy periods, back trajectories
favored the southwestern quadrant,  with a secondary frequency peak in the eastern quadrant, especially
for southern sites.

        Synoptic scale mechanisms responsible for cloud formation differed for northern and southern
portions of the Appalachians.  Nearly two-thirds of cloud events at Whiteface Mountain, NY, were
associated with pre-warm frontal synoptic conditions.  However, at Whitetop Mountain, VA, over
two-thirds of all  cloud events were associated with warm-sector and marine flow conditions.

        The amount of acidic deposition to a mountain site depends on factors such as cloud frequency,
windspeed, liquid water  content, and the history of the air mass in which the clouds are  formed.  To
relate cloudwater composition with  large-scale circulation features, cloud events collected during the
1986-88 field seasons were classified according to nine synoptic features responsible for cloud production.
This scheme classifies cloud events according  to a mountain's location relative to surface weather map
features for  hours when cloud was observed at the sites.
                                                5-1
-------
                                      TABLE 5-1

                            CLOUD WATER ION STATISTICS
                       CATEGORIZED BY SYNOPTIC CONDITIONS
                         FOR WHITEFACE MT. SUMMIT: 1986-88
       HYDROGEN ION (hourly measurements: /rniol/1)

Synoptic Cat.  Frequency      Percent        Mean
Maximum
               Synoptic Categories:
                            0 - not in warm sector
                            1 - warm sector, non-stagnant conditions
                            2 - warm sector, stagnant conditions
Minimum
0
1
2
1&2
All
382
139
68
207
589
SULFATE ION (hourly
Synoptic Cat.
0
1
2
1&2
All
Frequency
384
139
68
207
591
NITRATE ION (hourly
Synoptic Cat.
0
1
2
1&2
All
Frequency
384
139
68
207
591
64.9
23.6
11.5
35.1
100
measurements:
Percent
65.0
24.5
11.5
35.0
100
measurements:
Percent
65.0
23.5
11.5
35.0
100
134
335
320
330
203
MttlOl/l)
Mean
78
203
214
206
123
jimol/1)
Mean
61
130
191
150
92
1288
1412
1778
1778
1778

Maximum
547
668
1112
1112
1112

Maximum
752
632
1344
1344
1344
0.3
0.3
50.1
0.3
0.3

Minimum
0.4
0.3
40.8
0.3
0.3

Minimum
0.7
0.3
18.4
0.3
0.3
AMMONIUM ION (hourly measurements: ianol/l)
Synoptic Cat.
0
1
2
1&2
All
Frequency
384
139
68
207
591
Percent
65.0
23.5
11.5
35.0
100
Mean
80
181
227
196
121
Maximum
770
652
920
920
920
Minimum
0.7
0.5
43.0
0.5
0.3
                                          5-2
-------
                                 TABLE 5-1 (Continued)

                           GAS CONCENTRATION STATISTICS
                       CATEGORIZED BY SYNOPTIC CONDITIONS
                         FOR WHTTEFACE MT. SUMMIT: 1986-88
       OZONE (hourly measurements: ppbv)

Synoptic Cat.  Frequency      Percent       Mean
Maximum
             Synoptic Categories:
                    0 - not in warm, sector
                    1 - warm sector, non-stagnant conditions
                    2 - warm sector, stagnant conditions
Minimum
0
1
2
1&2
All
6590
1292
565
1857
8447
SULFUR DIOXIDE
Synoptic Cat.
0
1
2
1&2
All
Frequency
4581
1025
456
1481
6062
78.0
15.3
6.7
22.0
100
41.9
57.3
69.7
61.1
46.1
127.1
132.6
135.3
135.3
135.3
11.0
23.0
32.0
23.0
11.0
(hourly measurements: ppbv)
Percent
75.6
16.9
7.5
24.4
100
HYDROGEN PEROXIDE (hourly
Synoptic Cat.
0
1
2
1&2
All
Frequency
594
90
188
278
872
Percent
68.1
10.3
21.6
31.9
100
Mean
0.8
2.4
2.3
2.3
1.2
measurements:
Mean
0.6
0.8
1.5
1.3
0.8
Maximum
13.7
17.7
20.0
20.0
20.0
ppbv)
Maximum
6.0
2.8
3.8
3.8
6.1
Minimum
0
0
0
0
0

Minimum
0.0
0.1
0.0
0.0
0.0
                                          5-3
-------
        Four of the nine synoptic types are associated primarily with the main sectors of a cyclone.  The
nine synoptic types are: a) pre-warm-frontal  clouds, b) northwest-sector of cyclone clouds, c) post-cold-
frontal clouds, d) warm-sector synoptic types  including the southeast sector of the cyclone and stagnant
conditions, e) stationary-front cloudiness, f) easterly flow that advects moisture inland from the Atlantic,
g) and h) cutoff-low-aloft  types (these two  categories were created based on the location of the cutoff
relative to the site), and i) cap cloud which involves no important synoptic scale features within several
hundred km of the site and high  cloud bases  at nearby airports.

        A site-cloud climatology was developed for Whiteface Mountain, NY, Mt. Moosilauke, NH,
Whitetop Mountain, VA,  Shenandoah, VA, and Mt. Mitchell, NC.  As Table 5-2 shows, at the northern
sites (NY and NH), the post-cold-frontal type (c) and warm-sector synoptic types (d) comprised most of
total cloud hours. At  the southern sites (VA and NC), the warm-sector type  (d) was most frequent
while the second most frequent cloud synoptic type (f) was a result of easterly flow.


                                             Table 5-2

             Percentage of Cloud Hours by Synoptic Type for Five MCCP Mountain Sites
                                      (1986 - 88 Field Seasons)

        Synoptic
          Type         Whiteface      Moosilauke     Shenandoah    Whitetop      Mitchell

a. pre-warm front       16.5            25.5            6.5           0.1
b. northwest of cyclone   5.8            11.5            4.5           4.2             2.5
c. post cold front       32.4            11.0           23.9           16.0           15.2
d. warm sector          30.7            31.8           11.3           45.4           41.9
e. stationary front        0.3             0.1            3.2           1.5             2.5
f. easterly marine flow    4.0            15.9           43.7           23.1           26.8
g. low aloft (W-flow)     3.4             2.0            1.8           5.0            2.9
h. low aloft (SE-flow)     1.4             2.2            5.1           0.1
i. cap cloud              5.5             -              -            4.6            8.2


        Two  types of statistical models were  applied to the  data for 1986-88. The first model, partial
least-squares  regression on latent variables (PLS), was used to identify combinations of meteorological
parameters that predicted linear combinations of chemical concentrations (Vong et al., 1988a; Geladi and
Kowalski, 1985). The linear combinations  were formed from correlated variables.

        The original data consisted of hourly cloud water samples and corresponding hourly meteoro-
logical parameters of trajectory, synoptic type, cloud type, and local winds. Trajectories were obtained
from a mixed-layer model that uses rawinsonde measurements to define the transport field (Kahl and
Samson, 1988).  The chemical data were square-root transformed before analysis to keep extreme
samples from dominating the results. Hourly samples from continuous cloud events were aggregated
into sub-events with consistent trajectories.

        A six-component  PLS  model identified significant covariance structure in  the meteorological
variables that was related to  the  chemical data.  The correlations that were revealed among synoptic
types and mixed layer  trajectories were physically realistic.  For example,  the "marine" synoptic type
occurred with easterly trajectories and high cloudwater chloride concentrations,  the warm-sector synoptic
type (d) occurred frequently  with SW trajectories and high SO^2" concentrations,  and the post-cold-


                                                 5-4
-------
frontal synoptic type (c) occurred with NW trajectories and low SOj2' concentrations.  This agreement
added confidence to the use of the two types of meteorological variables (synoptic type and mixed-layer
trajectory).  PLS identified significant covariance in the chemical data also.  H+, NH4+, SOj2-, and H+
were correlated.

        Based on the PLS modeling, the following synoptic-trajectory classes were posed as the most
important influences on cloudwater chemistry at the four sites for further analysis:

               Synoptic type (d)  = warm-sector, SW trajectory
               Synoptic type (d)  = warm-sector, all other trajectories
               Synoptic type (i) = cap-cloud, SW trajectory
               Synoptic type (i) = cap-cloud, all other trajectories
               Synoptic type (c) = post-cold frontal, NW trajectory
               Synoptic type (c) = post-cold-frontal, all other trajectories
               Synoptic type (f) = marine-flow, SE to NE trajectory
               Synoptic type (f) = marine-flow, all other trajectories
               All other synoptic type (a, b, e, g, h) and trajectory combinations

        The second model, analysis of variance (ANOVA), was used to test the hypothesis that this
smaller number of meteorological  variables would describe variability in cloudwater chemistry. Classes 3
and 4 were combined with 1  and 2 because both represented stable atmospheric conditions, and there
were few observations of cap cloud compared to synoptic types c, d, and f.  The ANOVA  model (Box et
al., 1978; Vong et al., 1988b) tested for differences in cloudwater chemistry for the seven remaining
synoptic/trajectory classes.

        This model was highly significant (p < 0.001).  The three main effects were significant; there
were differences between sites, cloud-only (non-precipitating) and mixed rain/cloud (precipitating) types,
and synoptic-trajectory classes. Table 5-3 gives mean ion concentrations for all MCCP sites  for selected
synoptic-trajectory classes and cloud types.
                                            TABLE 5-3
                      Analysis of Variance Results for Four Ion Concentrations:
 Mean Values for all MCCP Sites (1986-88) in *eq/L for Selected Synoptic-Trajectory Classes, All Clouds

                                                        Ion

Synoptic-Trajectory Class                   SO^2'   H+       NH4+      NO/

Warm sector - Ohio Valley Trajectory       361      304       161         133
Warm sector - Other trajectories            246      190       117          88
Post cold front - NW trajectories            131       90        59          43
Marine flow - E trajectory                  134      138        47          61

Mean of all 9 classes  (all cloud)             199      170        86          78
Mean: non-precipitating cloud              262      220       116         104
Mean: precipitating cloud                  113      101        47          43
      The site-class interaction terms confirmed different warm-sector synoptic type dependencies of
SO/2' on trajectory. The highest SO^2', H+, NH^+, and NO/ concentrations occurred with the
                                                5-5
-------
warm-sector synoptic type (d) when the trajectories pointed toward the Ohio Valley.  This is consistent
with the notion of transport from that high emission area.

      Remaining variation within each synoptic-trajectory class is expected to be related to variation in
gas- and aqueous-phase chemical processes, scavenging, and meteorology that could not be described in
the synoptic-trajectory  classification scheme.  For example,  the height of the monitoring site above cloud
base should affect the amount of available  liquid water and, therefore, aqueous phase concentrations, but
testing that effect for all sites was not possible because cloud base data were not always available. For
one site (Whiteface Mountain, NY) there were consistent differences between two collection sites located
at different elevations when 106 hours of simultaneous data were compared; the lower site had higher
concentrations. The modeling approach described here was able to characterize several factors that
influence the composition of mountain cloudwater composition in the eastern USA


GAS CONCENTRATION VS. AIR MASS TRAJECTORIES

Ozone

      Mean ozone concentration is  not strongly a function  of 36-hour air mass trajectory sector, as
shown in Figures 5-1 through 5-6 (Appendix E) for the 1986-88 period.  Northern sites have somewhat
higher mean values for southeasterly to southwesterly trajectories while  southern sites exhibit a fairly
uniform distribution.  A clearer pattern emerges from high ozone concentrations.  Figures 5-7 through
5-17 (Appendix E) focus on trajectory dependency for ozone values greater than 70 ppb and 100 ppb.
Rowland, Moosilauke and Whiteface  favor west-southwesterly trajectories during ozone episodes whereas
the southern sites exhibit a broader and more northerly distribution of trajectories. This pattern is
consistent with the weaker, more stagnant  airflow conditions found over the southeastern states in
summer and their proximity to the center of semi-permanent high pressure systems.   In the  Northeast,
ozone episodes are usually associated with  backside flow around a stagnant high pressure  system  (e.g.
Bermuda High).

Hydrogen Peroxide

      Gas  phase measurements for  H^ at Whiteface and  Whitetop are shown in Figures 5-18 and 5-19,
(Appendix E) respectively. Despite the overall lack of variability in concentrations, mean values  at
Whiteface  are highest  for southwesterly trajectories, which  is consistent  with the pattern found for ozone.
Hydrogen peroxide is a secondary product  resulting from complex photochemical reactions initiated by
ozone.  The directional dependence at Whitetop is less clear, but peak H^ values there are associated
with westerly trajectories.

Sulfur Dioxide

      In summer, mean S02 concentrations are low, averaging under 3 ppb at all sites.  As with the
other gases, S02 values (Figures 5-20 to 5-23 - Appendix E) are higher with west to southwesterly
trajectories for sites  in the north.  Southern sites exhibit higher concentrations for north to northwesterly
flow.  The differing dependence of gas concentrations on trajectory between northern and southern sites
suggests a  common air mass source region for relatively  high concentrations-the Ohio Valley.
                                                 5-6
-------
                                            SECTION 6

                                  METHODS AND PROTOCOLS
      This section discusses field collection and laboratory analysis methods for cloud water chemistry,
liquid water content, and supporting measurements such as meteorology, throughfall, and precipitation.
Uncertainties are discussed with each method.

      The MCCP collects cloud water only during the growing season, the onset of which varies
depending on site location.  At most  sites, sampling is conducted for intensive campaigns of ap-
proximately three to six weeks, during which every cloudy period is sampled. At least three intensive
sampling periods have been conducted during the growing seasons (approximately April-October) of 1986
through 1988 except for two sites that started during 1987.

      Cloud water collection is begun as soon as possible after the start of a cloud  event, which is
defined by both visibility and volume.  When a stationary object approximately 1 km from the collection
site is consistently obscured from view for more than 15 minutes, a cloud event is defined to have
started. Collection is discontinued when either the stationary object is no longer obscured from view, or
less than 10 ml of sample has been collected over a 20  minute  period. Liquid water content and
aqueous phase chemistry are monitored during these cloud events. Gases, meteorology, precipitation
chemistry, and throughfall chemistry are measured during cloudy and non-cloudy periods throughout each
intensive sampling campaign.


CLOUD WATER COLLECTION

      Mohnen and Kadlecek (1989) discussed the processes governing cloud collector efficiency and
design.  Passive cloud water collectors, such as the MCCP designated ASRC design (Falconer and
Falconer, 1980), depend on ambient wind speed to impact droplets onto  0.4 mm wires strung between
two circular disks. Active cloud water collectors have a blower or fan  to provide the velocity difference
between the droplet and the collecting string that is necessary for impaction. Factors that govern the
overall efficiency for collector accumulation of liquid water include:  impaction on the inertial collector
surface (based on Langmuir stopping distance); accumulation/coalescence of individual droplets into  a
film or larger droplet; transport of the water from the collector surface to the storage container via
gravitational, centrifugal, or pressure  forces; evaporation of deposited cloud water from the collector
surface; and resuspension (blow off) of water due to high wind speed at the collector surface. Various
collector designs attempt to maximize efficiency for droplets above 2 to 5 nm diameter (a smaller cut-off
size would allow collection of unactivated aerosol).  However, the reduced  collection efficiency below 5
itm diameter allows small droplets to escape collection.

      During winter, supercooled cloud water freezes on impact with the collector strings.  Collection
efficiency is reduced due to the increased surface  area of the strings covered by rime, so  the accumulated
rime is collected manually as often as possible to minimize this  effect (Kadlecek et  al., 1988). MCCP
does not perform winter sampling.

      Two types of cloud water collectors are used at MCCP sites, the ASRC passive collector and the
Cal-Tech active string collector. Both have been  found to collect water with equivalent chemistries over
wind speeds ranging from about 3 to  30 m/s (Mohnen and Kadlecek, 1989).  At most sites, the collectors
are placed on the top platform of towers above the surrounding canopy.  The only  exceptions to this are
Whiteface Mountain, where the collector is on the roof of the summit research lab, and  Whitetop
Mountain, where it is on a platform built over the research trailer.


                                                6-1
-------
      Once a cloud event has begun, clean tubing and a collection bottle are attached to a collector and
it is deployed.  Collection bottles are changed hourly, and a portion of each hourly sample is analyzed in
the field for pH.  Five to 10 ml is preserved for hydrogen peroxide analysis. Each bottle is refrigerated
until it is sent to a laboratory for chemical analysis.  At the end of each event, the cloud collectors are
detached and rinsed with deionized water until the conductivity of the rinse water is within % 10
^Siemens. Collectors are covered until the next cloud event. When more than 125 ml is collected at the
end of one hour, the sample is  split into two aliquots; one split is sent to the Central Analytical Lab
(CAL) and the other to the site-associated analytical lab.  Further details are provided in the MCCP
Standard Operating Procedures Manual  (1989); Quality Assurance procedures and results are described.


FREQUENCY OF CLOUD

      For forest exposure and cloud water deposition estimates, it is necessary to understand when and
where liquid water is present, at what mass concentration, and how the water is distributed among
different droplet sizes. These data are referred to collectively as describing the presence of liquid water
in the atmosphere.

      The emphasis of the cloud frequency analyses conducted by the Mountain Cloud Chemistry Project
(MCCP) is to estimate cloud impaction  at mountain summits and the overall  cloud distribution in the
Appalachian Mountains.  Because routine observations are scarce on mountains, other data sources were
necessary to estimate long-term horizontal and vertical cloud distribution.  This section describes the
methodologies and associated uncertainties in determining regional and site-specific cloud frequency
estimates. The data sources considered  are National Weather Service (NWS) surface airport observa-
tions, US Air Force Real-Time Nephanalysis archives, and site-specific  mountain cloud measurements
conducted by the MCCP.

Regional Estimates

      Airport observations have been analyzed to obtain cloud climatologies for the eastern  USA
(Warren et al., 1986).  Surface airport observations  consist of hourly measurements of cloud base height,
coverage, and type. Cloud base heights have been measured since the mid 1920s  using either a fixed- or
rotating-beam ceilometer. This technique has an estimated accuracy of,+ 30  up to 1000 m in elevation,
and of +. 10% above  1000 m (WMO, 1976).  Data are archived by the National  Climatic Data Center
for all reporting airport stations.

      A limitation to airport observations is that few stations are located within  mountainous areas.
Airport-mountain comparisons  have been investigated by Bailey and Markus (1987) and Imhoff and Malo
(1989).  A common finding is that airports are good indicators of mountain cloudiness for large-scale
cloud systems but are poor indicators of small-scale cloud systems typically associated  with orographic
effects (e.g., cap cloud).

      Bailey et al. (1989) have examined the temporal and spatial variations of low-level clouds (below
2000  m) in the eastern USA using the Real-Time Nephanalysis (RTNEPH), the  global cloud archives
produced by the US Air Force.  An earlier version, called 3-Dimensional Nephanalysis (3DNEPH) was
begun in 1971  (Fye, 1978) and  replaced in 1984 by  the improved RTNEPH version.  Both versions use
all conventional surface and rawinsonde data, pilot reports and  satellite data to produce three-
dimensional cloud information.  RTNEPH data are organized according to a horizontal grid system
superimposed upon a polar stereographic projection.  Grid point spacing is 47.7 km at 60° latitude
where projection is true. Each grid point contains  the following data for every 3-hour period:  type of
low, middle and high clouds, present weather, maximum cloud top, minimum cloud base, total percent


                                                6-2
-------
sky coverage, and percent coverage for 15 fixed layers. Gordon et al. (1984) have used 3DNEPH analysis
of observed monthly mean cloud amounts to calculate radiative fluxes with a cloud model.  Hughes and
Henderson-Sellers (1985; 1986) have compiled a global cloud climatology for 1979. Schulz and Samson
(1988) used 3DNEPH data to determine non-precipitating low cloud frequencies for central North
America for 1982.

      The uncertainties in the RTNEPH data archives consist of the limitations in each data source that
was included.  Limitations to the surface airport readings have been discussed previously.  The satellite
visual data processor cannot distinguish between clouds and bright areas of snow or ice.  Therefore,
unless accurate snow or ice data were available, the analyses may be in error. In addition, small cloud
elements over high variability backgrounds (coastlines and mountainous areas) may not have been
detected.  The 47.7 km grid spacing precludes resolution at smaller space scales, and thus the RTNEPH
analyses are best used for regional analysis.  Low-level clouds, particularly for multi-layer cloud condi-
tions, are best detected by surface and rawinsonde observations and least by satellites.  In areas where
airport station density  is low, low-cloud information may not be well represented.

Site-specific Methods

      Three different techniques for estimating the frequency of cloud  impaction have  been used by the
MCCP since the 1986 field season. From the project's onset, relative humidity has been used at most
sites to estimate periods of cloud presence in lieu of an affordable direct measurement technique. The
Whitetop site has always used a reflectometer to  determine the presence of cloud, but its cost  (in excess
of $10,000) precluded its use throughout the MCCP network.  Beginning with the 1988 field season, the
new Mallant optical cloud detector was installed at most sites.   This section describes the three
techniques while the next section inter-compares  the techniques.  A fourth technique (Krovetz et al.,
1988) has also been used on an experimental basis, particularly at the Shenandoah site. This technique
will not be discussed here because it was not used to derive official presence of cloud  estimates.

      Relative humidity (RH) is sensed by a Rotronics MP-100 combination temperature/RH probe
housed in a naturally aspirated radiation shield. Cloud presence  is determined subjectively by analyzing
the time series trace during saturation and near-saturation conditions.  The tendency for the sensor's
response in these conditions to drift upward with time has precluded the use of a wholly objective RH
cloud threshold  (such as a minimum RH value) to define cloud  events.  Average RH values are recorded
hourly based on 5-second samples.  As verified by field observations, the onset of cloud is typically
marked by a sharp rise in RH to near 100% followed by a leveling off, and cloud dissipation is marked
by an abrupt drop of several percentage points.

      The uncertainties inherent in the RH technique are several: the element of subjective interpreta-
tion, the sensor's 5% accuracy specification, the use of hourly averaged values, and its  slow  response
time relative to  the two optical techniques.  Overall, it is suspected that this technique  may somewhat
overestimate the frequency of cloud due to  the fact that near-saturation conditions can occur in the
absence of clouds such as during heavy or prolonged rain events.

      A backscatter reflectometer (visibility  sensor Weathertronics Model 8340)  has been used  by TVA
since 1986 as  a cloud detector.  According to Valente et al. (1989), "observations at Whitetop have shown
that by using a signal of 0.15 (5% of the output range) as the definition of presence of cloud, haze and
liquid precipitation are not mistaken for cloud impaction." The instrument  has a visibility range of
1000+m to 10 m, corresponding to a  signal output of 0 to 100% of scale, respectively. The detection
threshold of 5% of scale corresponds  to a visibility of about 260 m according to the manufacturer's
literature.  This  instrument has not been used at  MCCP  sites other than Whitetop.
                                                6-3
-------
     The uncertainties with the reflectometer are not well defined. However, the method used to obtain
hourly signal values will tend to overemphasize the presence of cloud within an hour in which cloud was
present less than half the time.  This occurs because the signal's scale is logarithmic whereas hourly
averages are derived arithmetically (private communication, TVA).  Hourly average reflectometer values'
are based on 5-minute intermediate average values derived from 20-second samples. Given that  the first
5% of scale (cloud absent) corresponds to a visibility range of about 740 m while the remaining  95% of
scale (cloud present) corresponds to about a 250 m visibility range, arithmetic averaging of percent of
scale values within an hour will give greater weight to "cloud present" signals. This can be a source of
discrepancy when comparing different measurement techniques during variably cloudy hours.

     The Whitetop  site has also used a time-lapse video recording system to estimate cloud base
elevation.   The system is located about 4.5 km northwest of the mountain at an elevation about  760m
less than the site elevation of  1689 m.  The location provides a clear view of the mountaintop from one
direction and is limited to daytime observations. Cloud base heights are derived hourly through  manual
interpretation of the video record.

     The Mallant Optical  Cloud Detector is a forward-scattering optical device originally developed by
the Energy Research Foundation in The Netherlands (Mallant and Kos, 1989). Following successful
laboratory and field tests (Valente et al.,  1989), a modified design was built by Associated Weather
Services and deployed at all summit sites in 1988.  It has an adjustable detection threshold corresponding
to liquid water content at  an assumed mass median droplet diameter.  Hourly recordings indicate the
percentage of time with cloud  present based on 5-second samples.  Each sample produces a binary signal
corresponding to cloud presence (=1) and absence (=0).

     The Mallant instrument's detection threshold is set at an equivalent LWC value of approximately
0.04 g/m3 for a droplet mass median diameter of about 11 microns. This is roughly equivalent to a
visibility of 350 m (Atlas and Bartnoff, 1953). The uncertainty in this threshold  is on the order of ±0.02
g/m3 due to the uncertainties in the FSSP-based laboratory calibrations and on the imprecision of the
adjustment potentiometers. Under ambient field conditions, the actual detection threshold will also be  a
function of the cloud's drop size distribution. Therefore, this instrument's response to cloud, especially
"thin" clouds, is variable.  It has also been observed that this instrument can be susceptible to  droplet
accumulation on the optical lenses which results in a reduction in  the sensitivity to the presence of
cloud.  The detector does  not  contain heaters and therefore is suitable only for above-freezing cloud
conditions.

Comparison of Methods

      This section presents results of several intercomparison tests between the three primary  techniques
used to estimate presence of cloud.

Mallant vs. Reflectometer-

      Several intercomparisons of the Mallant cloud detector and the  TVA reflectometer have taken
place, beginning with the 1987 liquid water content instrument "shoot-out" at Whitetop  Mountain
(Valente et al., 1989).  This first intercomparison used an original Mallant prototype whereas  all later
intercomparisons used the modified design.  In all intercomparisons the TVA reflectometer was  desig-
nated an arbitrary "standard" to define periods when cloud was present or absent. The first intercom-
parison, covering a two-week period in 1987, found that the Mallant prototype had 98% agreement with
the reflectometer during cloud events and 93% agreement during non-cloud events.

      Four other intercomparisons were conducted during 1988 and 1989 using four different Mallant
instruments.  These  tests comprised over 5700 field hours.  The Mallant  design agreed with 96% to 99%


                                                6-4
-------
of the cloud/no cloud observations taken by the reflectometer. When clouds were indicated by the
reflectometer, the Mallants agreed 92% to 95% of the time for three of the instruments, and 80% for
the other.  The lower agreement of this one instrument was the result of it not responding to cloud for
a 24-hour period within one extended cloud event.  It is suspected that water droplets accumulated on
its lenses, reducing the instrument's sensitivity.  When  the reflectometer indicated  that clouds were absent,
agreement with all four Mallants ranged from 98% to 100%.
Relative Humidity vs. Mallant-

      Concurrent estimates of hourly cloud presence using the RH and Mallant techniques were made
throughout the 1988 field season at four MCCP summit sites: Mitchell, Moosilauke, Shenandoah and
Whiteface. A minimum of 2300 hours of simultaneous values were taken at each site.  Based on its
favorable intercomparison  results with the TV A reflectometer, the Mallant detector was designated the
standard in the intercomparisons with RH. An hour with cloud as defined by the Mallant required the
instrument to be detecting cloud for at least 50% of the hour. Overall agreement between techniques on
an hourly basis ranged from 87% to 96%.  During cloud conditions as defined  by the Mallant, agreement
ranged from 81% to 94%.  During no-cloud conditions, they were in agreement 88% to 96% of the time.

RH vs. Mallant and Reflectometer-

      The early 1989 field  season was the first opportunity for the standard MCCP RH sensor to operate
together with the Mallant and reflectometer instruments at Whitetop Mountain. Based on over 1100
hours of operation,  RH agreed with the reflectometer 91% of the time, and 90% with the Mallant.
During cloudy conditions the agreement was 86% and 85%, respectively.  During non-cloudy conditions,
the agreement was 94% and 91%, respectively.


LIQUID  WATER CONTENT

      The MCCP initiated network-wide measurements of cloud liquid water content (LWC)  in 1987
using the TVA-Valente filter collection method. This section presents the results of a review of LWC
and drop size measurement methods and a field comparison of several LWC instruments.  Most LWC
measurement instruments have been designed for aircraft operation where the speed of the aircraft is an
aid to operation.  Ground-based instruments are typically modified versions of aircraft systems, and can
be classified according to their operating principle as thermal, optical, or inertial impaction methods.

      The Johnson-Williams liquid water instrument consists of a heated wire that is cooled by the
evaporation of liquid water droplets.  An instrument time constant of approximately 1-2 sec may cause
underestimation of water contents (Spyer-Duran, 1968).  A further limitation is a limited response to
droplets larger than  30 ion. The instrument must be calibrated in a cloud tunnel to achieve the expected
accuracy.

      The NHRL Nimbiometer and the CSIRO (King) Probe are heated-wire instruments. Improve-
ments were made by Merceret and Shricker (1975) by operating the instrument at a constant tempera-
ture below 100°C where the power dissipated is proportional to the square of the liquid water content.
King et al. (1978) improved the instrument by increasing the cooling by liquid water relative to cooling
by dry air.  Measurements  agree with other instruments to about 5% at LWC of 2 g/m3.   The advantage
of this design is that wet calibration is not required.  However, at  low air speeds typical of ground-based
systems, the errors are proportional to \-^2 (King et al., 1978), making it necessary to vary the aspira-
tion speed with wind speed. Particle Measuring Systems, Inc., has developed a King Probe prototype to


                                               6-5
-------
measure LWC of up to 1 g/nv3.  The probe is oriented using a wind vane.  A field test at Whitetop
Mountain, VA, indicated that further development and testing are needed because of short circuits
during wet, cloudy periods, sensitivity to radio frequency interference, and rain droplet contamination of
the inlet (at wind speeds >  3 m/s).

      Chylek (1978) investigated an optical technique that uses the extinction of radiation of a wave-
length of 11 ^m and found that infrared extinction should be directly proportional to liquid water
content. Several attempts to use this technique in cloud and fog have been encouraging (Gertler and
Steele, 1980; Jiusto and Lala,  1982), but practical problems have limited its full use.  Comparison of a
laser transmissometer (Jiusto and Lala,  1982) with a droplet spectrometer (FSSP-100) revealed good
agreement up to water contents of about 1 g/m3. The presence of many large droplets (> 28 ^m)
degrades Chylek's (1978) approximation and contributes to forward scatter errors.  At LWC less than 0.5
g/m5, the transmissometer system was within  10% of filter measurements and 35% of FSSP-100 values.
This promising technique is still in the development stage.

      Another instrument estimates LWC by measuring scattered intensity in the near forward direction
(0.25  to 5°).  BIyth et al. (1984) designed an instrument with a 26 cm path with discrete detectors at five
different angles  from the scattering axis. Based on  theoretical calculations, the signals are weighted to
provide a signal proportional to LWC.  Comparison with drop spectrometers and infrared transmis-
someters show excellent agreement.  This instrument still requires development to construct a field unit
suitable for routine observations.

      Gerber (1984) proposed a similar instrument  based on the principle that light scattered by the
droplets in the near-forward direction is strongly correlated with infrared extinction, and is thus also
proportional to  LWC according to the Chylek (1978) relationship.  Gerber Scientific offers the  Particle
Volume Monitor (PVM), capable  of measuring in situ, and in real time, the  integrated volume of
particulates suspended in the atmosphere with a stated precision of 0.002  g/nv'.  A narrow beam from a
780 nm laser diode irradiates the droplets in the open air along a 40 cm path.  Calibration results
indicate % 10% agreement with an infrared transmissometer in controlled fogs (Gerber, 1988).

      A prototype instrument developed by Scott McLaren of ASRC SUNY-Albany, NY, uses a chilled
mirror dewpoint sensor to measure the  dewpoint of two differently treated streams of sampled air, and
switches streams at five minute intervals.  The first  stream contains air that has been heated  to  evaporate
the cloud droplets into the air stream, and the second stream is filtered to remove the droplets. By
measuring the small difference in dewpoint between these air streams, the LWC can be calculated from
psychrometric principles, provided that significant changes in LWC do not occur between the two
adjacent averaging periods.

      Filters and impactors  are the oldest method of measuring liquid water  content. Examples are
reported by Houghton and Radford (1938) and Wattle et al.  (1984). The operating  principle is simply
to draw a large  volume of cloudy air through a filter and then weigh the accumulated water.   Using large
air flows and large  collection areas minimizes the size selection collection problems.  With light wind
and droplets below precipitation size, this technique comes close to providing an absolute measurement
of LWC.  One limitation is that during periods with less than 100% relative  humidity, evaporation may
result in an underestimate of LWC.  The method requires constant attention and averages over the
one-hour sample period.

      Impaction techniques  are similar to filtering, but they emphasize droplet sizing instead  of mass
concentration.  A glass slide with  a gelatin coating  (Jiusto, 1965) or another  material sensitive to water
drops is used for collection.  With gelatin, droplets leave impressions that are twice  the diameter of the
impinging droplets.  The slides are analyzed with an optical microscope to determine drop size  distribu-
tion and water content. This technique  may discriminate against either the large or small end of the


                                                6-6
-------
spectrum, depending on the collection method.  Baumgardner (1983) estimated a 32% error in measure-
ment.

      Optical particle spectrometers measure the size distribution of particles by light scattering. Size is
determined by a calibration of the light pulse amplitude from particles of known size. LWC is calcu-
lated by integrating the droplet volumes from the measured droplet spectrum and dividing the total mass
of liquid water by the sampled air volume. Different optical configurations compensate for the size
dependence of the scattering efficiency. Virtually all designs exhibit some uncertainty at sizes of the
order of the wavelength of the light source.  Perhaps the greatest obstacle to using these instruments in
cloud is that the aerosol configuration discriminates against large particles such  as cloud droplets
(Davies,  1968).

      Particle  Measuring Systems, Inc., developed the Forward Scatter Spectrometer Probe (FSSP-100)
with a unique sample inlet configuration. The laser light source  is focused to a 100 nm diameter;
scattered light is split to separate the light scattered by particles in the center of the  beam from the
signal, making it possible to avoid the aerodynamic focusing used in other systems. The signals from
particles that pass through edges of the sample volume are eliminated by measuring the  time it  takes for
a particle to cross the beam and comparing this with a running average.  This design alleviates inlet
problems, but  the use of forward scattering with small collection angles has resulted in a scattering
response with  several multi-valued regions  (Pinnick et al., 1981), especially between 0.5 and 2 nm and
near 8 urn diameter. FSSP measurement uncertainty under  the best conditions is about 34%, and an
uncorrected system  could have errors from 54% to  105% (Dye and Baumgardner, 1984; Baumgardner,
1983). FSSP-100 sampling efficiency remains open; in low wind conditions the sample should be
representative.  The probe should be aligned into the wind and, in gusty conditions, correction to the
sampling rate may be necessary.   To compensate for variable winds, modifications that continuously
adjust  the fan  speed are  available.

      The TVA-Valente  instrument (Valente, 1988) used by the MCCP collects  droplets by filtration
from a metered  air volume in an Ertalyte plastic cartridge filled with eight layers of a high collection
efficiency polypropylene mesh. Inlet velocity is matched to  the mean wind velocity to minimize non-
isokinetic sampling  errors.  A rain shield minimizes collection of precipitation-size droplets; however,
significant precipitation-induced  errors are likely at wind speeds  above 10 m/s.

      Liquid water content (LWC) measurement in the MCCP follows the same collection schedule as
that described for cloud water. The LWC  collector is deployed near  the cloud water collector at the
start of an event. The collector cartridge is weighed at the beginning and end of an hour and the hourly
LWC is calculated by dividing the weight gain by the sampled air volume.  The inlet  velocity is adjusted
to match the mean  wind velocity to minimize anisokinetic sampling errors.  A rain shield is used to
minimize collection of precipitation-sized droplets.  The sampler must also be pointed onto wind and an
inlet size is selected depending on wind speed. Detailed operational procedures can be found in the
MCCP Standard Operation Procedures (1989).

      During 1988,  MCCP used  optical cloud detectors to record the hours with patchy clouds.  The
collection mesh  and range of inlet velocities were designed to collect cloud droplets from 3 to 200 ^m in
diameter with  greater than 95%  efficiency.  A rain shield excludes rain (droplet  radius > 500 urn) and
drizzle (200-500 /»m radius) falling at an angle of greater than 15° from horizontal. However, during
high winds combined with precipitation and cloud, some precipitation will be sampled.  LWC measure-
ments  of clouds by the MCCP therefore differentiate between precipitating and non-precipitating events
because the technique may overestimate LWC during precipitating events.
                                                6-7
-------
Droplet Size Distribution from LWC

      Cloud droplet size distributions (or size spectra) are measured using impactors or a forward
scattering spectrometer probe  (FSSP) that measures the degree of scattering of a narrow laser beam as
particles pass through the beam (Knollenberg, 1981).  The amount of scattering depends on particle size.
The instrument can detect droplet diameters from  < 1 itm to 100 iaa.

      Because median droplet size is important for cloud droplet interception processes but is not
routinely monitored by MCCP, it is sometimes estimated from measured liquid water content (LWC).
MCCP scientists have, investigated the relationship between the median droplet volume diameter (dn)
and LWC for stratus and stratocumulus clouds.  The parameter dn is  defined as the  droplet diameter
that divides the total cloud liquid water into equal parts by volume  (Best,  1951).  Stratus and stratocum-
ulus clouds were examined because of their similarity to clouds observed from mountaintop forests.
Seventeen data pairs, representing 96 measured droplet size distributions, were  used  to determine the
correlation between dn and LWC. One-half of the variance in dn was explained by variance in LWC,
suggesting that LWC and droplet size distribution  are related for mountaintop clouds but that con-
siderable size spectra variability is associated with other factors.

LWC from Cloud Sampler Collection rate

      An additional size spectra feature having  implications for cloud  deposition modeling is the large
variation in the size spectrum for a given value of LWC. Large fluctuations in droplet number con-
centration and diameter imply that, despite a general tendency for the size spectrum to vary with LWC,
a large amount of noise exists in the data.  If droplet-leaf capture efficiency  is non-linearly dependent on
droplet  size, then the large variability in the size spectrum will result  in a biasing of cloud deposition
estimates.

      A data analysis technique for estimating LWC has been examined at several MCCP sites.  The
collection of cloud water (at a volume collection rate Rc) by a passive ASRC string  collector such as
used by Falconer and Falconer (1980) is known to depend on the dimensions of the strings, cloud
droplet  diameter, wind speed (M), and LWC.  A given droplet size  distribution has a characteristic
diameter that is presumed to be representative  of the droplet size spectra. This diameter is not known.
Assuming a constant linear relationship, several investigators have used-  regression to determine an
empirical relationship between LWC and RC/M (e.g., Saxena et al., 1989).
      At Shenandoah, Krovetz et al. (1989) compared LWC (gravimetric measurement) from the ratio
        , where V is the measured collector cloud water volume and S is the Stokes number (Fuchs,
1964).  S  was computed using the known string diameter, measured wind speed, and a constant droplet
diameter. They reported a correlation (r2  = 0.77) between 11 pairs of measured and estimated  LWC
values.

      These regressions assumed  that the effect of droplet size on droplet collection efficiency is
constant.  However,  measured droplet spectra have been found to be highly variable over short time
intervals,  even when liquid water content was  relatively constant (Mueller and Imhof, 1989). Therefore,
estimated median  droplet size from LWC data or LWC from a passive sampler collection rate are
considered to be a crude approximation of the distribution or concentration of liquid water in the
atmosphere.
                                                6-8
-------
GASES

      Gaseous measurements in MCCP are made using a combination of standardized and research
techniques.  The ultraviolet (UV) photometric method for ozone and the UV fluorescence method for
sulfur dioxide are the most well-defined gas measurements, with traceable quality assurance standards.
Measurement methods for nitric acid, sulfur dioxide, and hydrogen peroxide lack traceable quality
assurance standards, but have been involved in inter-comparison, laboratory, and field studies.  Table 6-
indicates the measurements that are made at each MCCP site.

Ozone

      Ozone measurements at MCCP sites are made with the UV photometric technique described by
Bowman and Horak (1972), which is equivalent to the US EPA's reference method (Federal Register,
1980). The TECO Model 49 used in the MCCP has a minimum detectable limit and precision of 2 ppb.
Ambient ozone values are 40 to 50 ppb at MCCP sites.


                                           TABLE 6-1
                                 Gas Measurements at MCCP Sites

   Site                                    Gas Measurement

Rowland, ME                     Ozone, sulfur dioxide, nitrogen oxides
Mitchell, NC                     Ozone, sulfur dioxide, nitrogen oxides
Moosilauke, NH                  Ozone
Shenandoah, VA                  Ozone, sulfur dioxide, nitrogen oxides
Whiteface,  NY                    Ozone, sulfur dioxide, nitrogen oxides, hydrogen peroxide
Whitetop, VA                    Ozone, sulfur dioxide, nitrogen oxides, hydrogen peroxide


Sulfur Dioxide

      Sulfur dioxide (802) is measured using continuous and integrated methods.  Continuous detection
of the characteristic fluorescence by  SO2 when irradiated by  ultraviolet light was described by Okabe
(1973).  These monitors have been designated as equivalent to the US EPA's reference method (Federal
Register, 1979).  The TECO model 43 used  for SO2 has a minimum detectable limit of 2 ppb with a
precision of 5 ppb.  The maximum SO2 concentrations at MCCP sites are in the 20 ppb range with a
mean of approximately 2 ppb.  This  instrumentation thus provides a good estimate of the upper limits of
SO2 at MCCP sites, while lower SO2 limits are not well defined.

      Integrated measurements of SO2 are made using the filter pack or the annular denuder system.
The filterpack system has been characterized by Anlauf (1986) and Sickles (1987).  In the filterpack unit,
SO2 is collected on a Whatman 41 filter impregnated  with sodium carbonate in  glycerol (Huygen, 1963).
The minimum detectable limit for SO2 using the filterpack system is estimated to be 0.5 ppb.  Factors
influencing the sensitivity of the method include sampling time, analytical sensitivity, and sampling
environment. A comparison of the filterpack method  and a  TECO SO2 monitor indicated agreement
over a range of ambient concentrations (Anlauf, 1986).

      The annular denuder system described by Possanzini (1983) is a multi-constituent sampler than can
be used to  measure a number of atmospheric gases: SO^ HNOj, NHj, and HONO.  Diffusion denude-
rs  have been used by several investigators to separate  gases and aerosols, to collect gases for analysis, or
                                               6-9
-------
as preseparators (Forrest, 1984; Shaw, 1982; Perm, 1972).  The overall sensitivity of the technique is
comparable to the filterpack method for 803.

Nitric Acid

     Nitric acid measurements in MCCP are made using either the fllterpack or the annular denuder
system (ADS).  Because of possible biases in HNOj' and NOj" sampling with the filterpack, a total
inorganic nitrate value is obtained from ion chromatographic analysis of the Teflon and nylon filters.
Using total inorganic nitrate as a measure of nitric acid could result in overestimation of nitric acid.
Nitric acid and nitrous acid are collected on a sodium carbonate coated annular denuder in the ADS
system.  For long sampling times (greater than 48 hrs), the nitrous acid can be oxidized to nitrate on
denuder surface; this gives an overestimation of the nitric acid due to the potential biases listed above.

Hydrogen Peroxide

     Gaseous hydrogen peroxide is measured with an automated monitor described by Lazrus (1986).
This technique reacts peroxide with p-hydroxyphenylacetic acid in the presence of peroxidase.  The
minimum detectable limit is estimated to be 0.05 ppb with a precision of 0.01 ppb. Hydrogen peroxide
levels range from 0.1 to 3.0 ppb at the MCCP sites.  When the monitor was compared with laboratory-
generated standards, the two agreed within  10% across a concentration range of 0.06 to 128 ppb
(Kleindienst,  1988).


METEOROLOGICAL MEASUREMENTS

     The MCCP meteorological measurement program was designed to complement the cloud water
collection and analysis as well as to define  the climate of high-elevation Appalachian Mountain forests.
Each MCCP site measures wind speed, wind direction, temperature, relative humidity, solar radiation,
barometric pressure, and precipitation. Beginning with the 1988 field season,  cloud presence was
measured with an optical cloud detector. All MCCP sites except Whitetop use the same package of
sensors and follow the protocols of measurement established by the MCCP Quality Assurance (QA) Plan
(1988) and detailed in  the MCCP's  Standard Operating Procedures (1989) and the Meteorological Site
Technician's Handbook.  The Whitetop site follows a measurement protocol that is similar but not
identical to the MCCP QA Plan.

      Wind speed and  wind direction are measured with the R.M. Young Wind Monitor (Model
# 05103).  Wind speed is sensed by a helicoid propeller that rotates a six-pole magnet that in turn
generates an AC sine wave with a frequency proportional to wind speed.  Wind direction is sensed by a
thermoformed plastic vane  that transmits its angular  position via a coupling to a 10 K ohm poten-
tiometer.  The sensor is calibrated at the start and end of each field season, and is subjected to at least
one Quality Control (QC) check during the season.  Most sites perform the QC check monthly.  Wind
speed is checked via an 1800 rpm motor and a torque disk for starting threshold, and wind direction  is
checked versus reference directions.

      Temperature and relative humidity are measured with the Rotronic Instrument Model MP-100
combination sensor. A capacitive element  senses  relative humidity and a resistive device (RTD) senses
temperature.   Both transducers are  housed inside  a Gortex cap and wired to a PC board that outputs
voltages corresponding to the ambient conditions.  The sensor probe is housed in a naturally-aspirated
radiation shield. The sensor is calibrated at the start and end of each field season versus a collocated
reference.  During the field season, weekly QC checks are performed using a psychron.
                                               6-10
-------
      Solar radiation is measured with a Li-Cor pyranometer (Model # L1200S).  The sensor uses a
silicon photodiode to measure solar radiation on a horizontal surface. The sensor is calibrated at the
start and end of each field season versus a collocated reference.

      Barometric pressure is measured with a Met One, Inc. barometer (Model # 090-B).  A piezoresis-
tive diaphragm senses the pressure, and its signal  is converted by PC circuitry to a voltage.  The sensor
is calibrated at the start and end of each field season against a collocated reference.

      A Weathertronics tipping  bucket gauge (Model # 6021-B) is used to measure precipitation.  Each
0.1 mm of precipitation results in a tip of the bucket.  At the start and end of each field season the
sensor is calibrated by applying  a known volume of water at a standard rate to the sensor.  During the
field season, monthly QC checks are performed in the same manner. This parameter has not been
measured at the Whiteface Mountain peak since the 1988 field season due to high winds.

      Meteorological sensors were installed to best represent weather conditions in the vicinity of the
cloud water collector without interfering with the  collection process.  Most sites locate their cloud water
collector on towers or roofs above the local forest canopy top. Thus, most meteorological sensors  are
similarly exposed adjacent to the cloud water collectors.  Precipitation and barometric pressure, however,
are measured near the surface.  Table 6-2 lists each site's sensor mounting location and height above the
local canopy.

      The purpose of the meteorological measurement program is to define the weather conditions at
each site's  cloud water collector station. Therefore, these measurements are site-specific and are not
necessarily representative of weather conditions  elsewhere.   This is especially true  for wind measure-
ments, which are strongly influenced by local topography and surface roughness.  In addition, the
meteorological monitoring program is designed  for growing season (April-October) only.
                                            TABLE 6-2
                  MCCP Site Sensor Mounting Locations and Height Above Canopy

                                     Height Above Canopy (m)

Site
Howland*
Moosilauke
Whiteface*
Shenandoah-1
Shenandoah-2
Shenandoah-3
Whitetop
Mitchell-1
Mitchell-2
Mitchell-3
Mount
Type**
R
T
R
T
T
T
R
T
T
T
Wind
Speed/Dir.
9.1
9.0
17.0
5.5
8.2
8.2
2.0
11.9
9.0
9.4

Temp/RH
9.1
6.0
16.0
3.6
3.7
3.7
0.0
7.3
6.0
4.9

Pressure
2.0
-10.0
9.0
-10.0
-16.5
-21.9
-4.5
- 7.3
-18.0
-13.2

Precip.
2.0
- 8.0
-11.0
-16.5
-21.9
-3.0
-5.9
-18.0
-15.2
Cloud
Freq.
N/A
4.1
15.0
1.7
N/A
N/A
0.0
5.4
N/A
N/A
a T - from tower
  R - above roof platform
b heights are relative to ground level
c not measured since 1988
                                                6-11
-------
THROUGHFALL

        Weekly integrated collections of throughfall and bulk or wet-only precipitation were made
during the summer of 1988 at some MCCP sites to compare throughfall inputs to the soil with what
would have been deposited had only rain occurred.  Ten 16 cm diameter funnels were deployed randomly
along the cardinal directions from the center of a 0.1 hectare circular plot. Each collector was mounted
approximately 1.5 m above ground level. A nylon wool plug inserted in the neck of the funnels and
replaced weekly prevented  large material from falling into the collection bottle attached to the funnel.
Weekly throughfall data were obtained from a proportional composite of at least 10 collectors, and the
precipitation data were from a single collector, either wet-only or bulk.  To minimize chemical changes
between collection and analysis, about 1 ml of chloroform was added to each bottle at  the beginning of
the week (occasionally preservative evaporation occurs).

        A bulk precipitation (BP) collector identical  to a throughfall collector is deployed in  an open
area.  The collection schedule is identical to that of throughfall. The BP is weighed in the field and a
250 ml  aliquot is reserved for lab analysis.  The wet-only precipitation collector is located adjacent to the
BP collector.  An Aerochem Metrics collector is used for wet-only (WO) collection and standard NADP
procedures for buckets are followed. Collection frequency  and sample handling are identical to that for
throughfall and bulk precipitation.  Laboratory analysis  methods and procedures for TF, BP, and WO
are identical  to those for cloud water analysis.

        The measurement of throughfall has involved a variety of collectors,  usually insufficient in
number given the spatial variability in deposition. Buckets, funnels, rain gauges, and troughs of various
sizes have been used (Reigner,  1964; Hamilton and Rowe,  1949; Goodell, 1952).  The uniformity of
terrain and canopy dictate  the effective collection area needed to adequately estimate throughfall  to a
canopy  location. The intention is typically to sample randomly, collecting from, canopy features accord-
ing to their relative densities.  For example, Helvey and Patric (1965) determined the number  of gauges
necessary to sample throughfall to a 5% error depending on rain amount (see Table 6-3).
                                            TABLE 6-3
            Mean Number of Throughfall Gauges and Collection Area Needed for 5% Error
                       Under a Uniform Canopy (from Helvey and Patric, 1965).
                Throughfall                                   Total Collection
                  (cm)                # of Gauges             Area in m2

                  < 0.5                   46                          0.25

                 0.5-1.0                   18                          0.06

                 1/0-1.5                   14                          0.04

                  > 1.5                   13                          0.04
                                                6-12
-------
        In a more complex and higher elevation canopy typical of the conditions on Whiteface Moun-
tain, the throughfall variability is larger, requiring more collection area to obtain a 90% confidence level
for a 10% coefficient of variation (Kadlecek,  1989), as shown in Table 6-4.  At the upper site, 17
buckets (28 cm diameter) were  used, and at the lower site 15 buckets were  used.  The ratio of highest to
lowest  collected volume during  a single event ranged from about 2 to 8, with the higher factor applying
primarily to low-volume events.  Collectors with larger surface areas, such as troughs, give data with a
smaller range, and those with smaller collection areas produce a larger range.  The principles are the
same for any canopy; the actual values depend on the canopy properties.


                                            TABLE 6-4
          Throughfall Bucket Collection Area Required for 10% Relative Standard Deviation
               in a Complex Canopy on Whiteface Mountain, NY (from Kadlecek, 1989).


                                                      Collection Area in m2
        Precipitation                  1200 m elevation                       600 m elevation
            (cm)                         balsam fir                          beech, birch, maple

           < 0.25                         2.9                                    8.8

          0.25-0.76                        2.4                                    3.3

          0.76-1.52                        3.0                                    0.8

           > 1.52                         1.3                                    0.6


Note:  These averages were obtained from 8 events at the upper site and 13 events at the lower site.


        Similar data from  the 1000 m  elevation (spruce, fir, beech) based on a regular 25-bucket grid
(bucket diameter of 28 cm, bucket spacing of 5 m) showed that had only three collectors been used,
there was a  90% probability that the three-bucket average would have differed from the 25-bucket
average by more than 30%. This is not to say that 25 collectors were enough, only that too few
collectors can give biased results because the canopy directs the flow unevenly and, in the case of cloud
water, exposed features are the preferred deposition sites. Longer sampling times  tend to average out
some of these differences because, as the wind vector changes during events, several segments of the
canopy are sampled. Sampling at the sub-event level is expected to require more collectors for the same
uncertainty discussed here.

        Reynolds and Leyton (1963) recommend 20 rectangular troughs with a total collection  area of
about 10 m2 to obtain a standard error of 5% to 10%.  Eidmann (1959)  used trough collectors with an
opening of about 1 m2. One method of calculating how  many collectors are necessary is to first deter-
mine by experiment the variation in collection over the study area; then,  the number of collectors would
be equal to  the squared ratio of the standard deviations  to the desired standard error of the mean
(Kittredge, 1948; Helvey and Patric, 1965).
                                               6-13
-------
CLOUD INTERCEPTION

        Particles greater than a few urn in diameter (including cloud droplets) are deposited onto
surface by impaction and sedimentation. Impaction is primarily due to the inability of particles  in the
airstream to follow rapid changes in air flow; their inertia carries them into surfaces protruding into the
airflow. Sedimentation is the result of the acceleration of particles downward under the influence of
gravity.  After a short time, aerodynamic drag balances gravity for small particles, and the particles fall at
a constant velocity. A detailed description of the deposition process is beyond the scope of this section:
the interested reader is directed to  Fuchs (1964), Chamberlain (1975), Bache (1979a), and Waldman and
Hoffmann (1987) for further information.

        The deposition process for cloudwater is a function of the free stream velocity of the droplets,
the concentration of droplets (liquid water content of the cloud), the droplet size distribution, and the
size, shape and distribution of elements protruding into the flow.  While there are several empirical
descriptions of particle deposition to vegetative canopies (e.g., Bache, 1979b; Thorne et al., 1982; Grant,
1983), the actual deposition process is not well understood.  One attempt at relating impaction efficiency
(deposition) to these basic physical parameters is (Bache, 1979a):

        E.I.  = St2/(St +  0.6)2         (1)

where:

        E.I.  = efficiency  of impaction
        St   = Stokes number, defined as:

        St = (Dp - Da)-u-c-dp2/9vdc   (2)

where:

        Dp   = density of the particles
        Da   = density of air
        u    = velocity
        c    = Cunningham slip correction factor
        dp   = diameter of the particles
        v    = dynamic viscosity
        dc   = characteristic length scale of the collecting surface

Stokes calculations have previously been applied to deposition processes by Davidson and Friedlander
(1978).  Unfortunately, there are few field data with which to evaluate these relationships.  Turbulent
transport over and among non-homogeneous surfaces, the driving mechanism for cloudwater deposition,
requires more research before the process will be fully understood and viable relationships identified.

         Shuttleworth (1977) described a  simple, steady-state computational model of fog/cloudwater
deposition to, and evaporation from, a uniform  vegetation canopy.  The model is based on the analogy
of electrical resistance in  a  direct-current circuit similar to the "big leaf" concept first defined by
Monteith (1965) and subsequently  applied to  deposition by Wesely and Hicks (1977), Unsworth (1981),
and Hicks et al. (1987), among others.

         Lovett (1984) adapted the Shuttleworth model to a balsam fir forest canopy on Mt.  Moosilauke,
NH.  Lovett included the effects of vertical variation in canopy structure through the construction of a
multiple-layer model in which the vertical turbulent transport of droplets is controlled by the


                                                 6-14
-------
aerodynamic resistances between model layers and between the top layer and the air above the canopy.
The droplet size distribution in the model is based on a distribution function described by Best (1951),
with all droplet diameters occurring in the range of 0 to 30  m.  While research carried out in forests
has raised questions about the applicability of aerodynamic theory for fluxes over forests (e.g., Stewart
and Thorn, 1973; Thorn et al., 1975; Knoerr and Mowry, 1977), the model has been applied to estimate
the flux of both water and ions in cloudwater to forests (Lovett and Reiners, 1986).

        The Lovett model forms the  basis of two cloud deposition models used to estimate the con-
tribution of cloudwater deposition to  the total deposition budget  at the MCCP sites  (Mueller  1989;
Krovetz et al., 1989). Modifications to the model have been made for the Whitetop Mountain site and
the Shenandoah site, and manuscripts describing this work  are currently in review. The modeling work
at Whitetop has been directed towards spruce forests, while the work at Shenandoah has been directed
towards deciduous forests.

        The canopy structure portion of the model had to be generalized to enable computations at
locations other than the  original Mt. Moosilauke study site.  The spruce version of the model (CDM-S)
allows the input of projected (silhouette) leaf area index which is used to calculate the full-sided leaf
area index.  Unless other input data are provided, the spruce model uses the vertical profile of total
surface area index (sum of leaf area index and total non-leaf surface area index) and the distribution of
surface area by canopy component type determined by Lovett for his balsam fir canopy.  Canopies
shorter than the Moosilauke canopy studied by Lovett (10.6 m) are assumed to have the full crown of
the Moosilauke canopy with shortened boles.  Conversely, the boles are lengthened for canopies of more
than 10.6 m in height, producing a smaller crown-to-total height ratio (a modification  consistent with
Whittaker et al., 1974).

        The spruce model also allows modifications  to both the wind speed profiles and the eddy
diffusivity profiles within the canopy.  These modifications  are not based on experimental data, but they
can be used for computational comparisons of model output.  The spruce model also allows the user to
select the cloud  droplet size distribution relevant to the cloud type being modeled if the original
distribution (Best, 1951), is considered inappropriate, and to input a measured  cloud liquid water content
if available.

        The deciduous version of this model (CDM-D),  developed at the Shenandoah site, is  a hybrid of
the original Lovett  (1984) model, the  model described by Bache (1979b), and recent  research on
deciduous forest canopies.  The distribution of both leaf and twig/branch surface area are determined
using the grid  technique  of Aber  (1979). The deciduous model uses the Best (1951) droplet size
distribution with additional droplet size categories and a  mean droplet diameter of 20  m. Liquid water
content can either be entered directly or calculated from an empirical formula relating capture efficiency
of the cloud water collector to liquid  water  content. Droplet capture efficiency  was changed to cor-
respond to the morphology of deciduous leaves;  it is computed for planar objects having characteristic
sizes determined from field sampling of fallen leaves (see Bache, 1979a). Wind  speed profiles above and
through the canopy are based on work in a deciduous forests by Sigmon et al.  (1984),  in complex terrain
by Mowry (1980), and basic micrometeorological theory.  Exponential decay of wind speed is used above
the canopy, while linear decay is used within the canopy.

        The differences in structure between the spruce and deciduous versions of the model  result  in
different outputs for identical meteorological conditions.  While differences in deposition between the
two types of canopies would be expected from theories concerning the capture efficiency of leaves versus
needles (as well  as the morphology of coniferous vs.  deciduous canopies), separating actual differences in
deposition from  model errors is difficult. No comparison of the models has yet been performed, and
very few validation data are available.  A priority for future work is to validate and compare the models
to identify differences in  deposition between the canopy types.


                                                6-15
-------
Cloud Deposition Model Application

        Lovett tested his original model against a limited amount of data collected on Mt. Moosilauke,
NH. The CDM-S has undergone limited testing at Whitetop Mountain, VA; for six cloud events during
1987, mean water flux (0.33 mm/hr) appeared to agree with the mean measured throughfall rate but only
36 percent of the variance was described. Another model test in a new spruce stand resulted in
model-predicted water flux of 0.4 mm/hr compared to a measured throughfall rate of 0.3 mm/hr.
Cloudwater collection rate explained 84 percent of the variance in throughfall rate. These results are
not conclusive because of uncertainties in the inputs  (such as forest stand characteristics, liquid water
content, and windspeed)  (Mueller and Imhoff,  1989). Due to limited testing, the uncertainties in cloud
droplet interception modeling are unknown at  this time.

        The response of the original Lovett cloudwater deposition model to varying meteorological and
canopy structure inputs was summarized by Lovett (1984) and Lovett and Reiners (1986).  Some key
results and sensitivities are:

               Computed cloudwater flux showed a near-linear dependence on input wind speed.
               Cloudwater deposition was proportional  to liquid water content (LWC); this
               proportionality defined a deposition  velocity generally in the range of 10 to 70 cm/sec
               for windspeeds of 2 to 20
               Cloudwater flux, as simulated  by the model, was a complex function of the distribution
               of surface area among various canopy component types  (e.g., the relative amounts of
               needle and twig sizes).
               The modeled contribution of droplet sedimentation compared  to impaction decreased
               rapidly as wind  speed increased above 2  m/sec.
               Computed cloudwater deposition  was sensitive to droplet size  distribution.
               Cloudwater flux increased locally  downwind of a forest edge or gap; edge effects could
               raise droplet deposition velocities from about 40 to 200 cm/sec.
               Cloudwater flux was insensitive to the vertical droplet eddy diffusivity profile.

        The sensitivity of the CDM-S model (used by MCCP) to changes in model parameters has been
examined by Mueller (1989) (see Table II-5).  He calculated that the cloudwater flux parameter was most
sensitive to lack of canopy uniformity as represented by a simulated forest edge.  Cloudwater deposition
at the edge was computed to be four to five times greater than in a closed forest.

        The second most important model parameters, in terms of output sensitivity, were cloud  liquid
water content  (LWC) and drop size distribution.  As with Lovett's original model, the CDM-S and
CDM-D  produced computed fluxes  that vary linearly with LWC. When the droplet size spectrum is
assumed to co-vary with  LWC, model response became non-linear; different forms of the drop size and
LWC relationship result  in computed flux differences by  a factor of two.  The  model is sensitive to the
concentration  of larger droplets  because of collection efficiency and fall  speed  dependencies on droplet
size.

        Droplet capture efficiency (computed  from the Stokes equation 2, above) was the third most
sensitive parameter in  the CDM-S version of the cloudwater deposition  model (it was second in the
CDM-D).  Mueller  (1989) tested the CDM-S model sensitivity to capture efficiency by varying it across
the range of possible values derived from the uncertainty in the experimental  data (Thorne et al., 1986)
and found that deposition varied by 100 percent.  Differences between CDM-S and CDM-D  result from
different capture  efficiency related to morphological differences between conifer needles and  deciduous
leaves.

                                               6-16
-------
        Other parameters had less influence on cloudwater flux calculations. Evaporation was important
for computed  net cloudwater flux when testing model simulations against canopy throughfall data.  The
effect of net radiation data on model performance is not known because net radiation was not measured
by MCCP (Mueller, 1989).

        The relative insensitivity of the CDM-S to vertical in-canopy profiles of surface area, wind
speed, and eddy diffusivity is fortunate because these parameters are poorly characterized for the MCCP
sites and nearby forests.  Of these three parameters,  the surface area profile appears most important
because it determines the computed wind speed  profile in the CDM-S and it affects estimates of the
vertical turbulent transport rate.  Unlike deciduous canopies that tend to have surface area more evenly
distributed in  the vertical, spruce-fir canopies tend to have the surface area concentrated in a shallow
crown space.  Differences of 10 to 20 percent in the  vertical location of the  canopy surface area
maximum are  relatively unimportant for predicting cloud water deposition with CDM-S.  Lovett (1988)
calculated that simultaneously increasing three key model input parameters (cloud liquid water content,
cloud frequency, and wind speed) by 25 percent  resulted in an increase from  154 cmfyr to 300 cmfyr
cloud water deposition using the Lovett (1984)  cloud droplet interception model.  Decreasing these three
parameters by 25 percent reduced the predicted  cloud water flux from 154 to 65 cm/yr.  Therefore, the
cloudwater deposition model errors for Lovett's  sensitivity analysis were linearly related to the product of
the windspeed, liquid water  content, and cloud frequency errors.  (Uncertainty in  these three measure-
ments and in cloudwater chemical composition are discussed in SosyT chapter 6).   Lovett concluded that
uncertainty in the input data in addition to extreme variability in time and space make cloudwater
deposition modeling impractical for the estimation of cloud water deposition over  hectare spatial scales
or seasonal temporal scales.

DRY DEPOSITION

        Dry deposition is more difficult to measure directly than wet deposition,  and the corresponding
data set is therefore meager. Methods to  measure dry deposition include the use of surrogate surfaces,
mass balances, and micrometeorological techniques.

        The use of surrogate surfaces has been  reviewed by Stevens (1985).  Uncertainties in this method
result from unknown collection efficiencies, unknown relationships of surrogate to natural surfaces, and
exposure of the surrogate surfaces to  wet forms of deposition.  Mass balance approaches have been used
by Eaton et al. (1978) and Galloway and Whelpdale  (1980), among others.  In this approach, dry
deposition results as the residual  in a mass balance calculation.  Therefore, at best, dry deposition
estimates can be as good as the combined variability of the measured parameters. Micrometeorological
techniques include at least five methods (Hicks et  al., 1980): (1) the eddy correlation method (the  only
direct method for determining mass flux density); (2) the variance method; (3) the eddy accumulation
method; (4) the gradient method; and (5)  the modified Bowen ratio method.  Micrometeorological
techniques produce more reliable results than surrogate surfaces or mass balance methods, but routine
implementation is difficult and expensive.  In the MCCP, models requiring readily available data sets
have been used to estimate dry deposition.

The Inferential Model - Development and Description

        Models used to estimate  dry  deposition  take two  general forms, both using the concept of
electrical resistance.  The more complex form, which considers the vertical distribution of canopy
structure and air transport structure, has been used in models by Murphy et al. (1977), Shreffler (1978),
Bache (1979a & b, 1984), Slinn (1982), and Davidson et al. (1982).  The simpler form has often been
referred to as  the "big-leaf concept, first introduced by Monteith (1965).
                                               6-17
-------
        The MCCP uses the best available simple modeling tool to convert dry species concentrations  to
dry deposition flux densities in adopting the inferential big-leaf model under development by the
ATDD/NOAA.  This model conceptualizes the system as an Ohm's law analogy of mass flowing to and
from a surface through a sequence of series and parallel resistances. These include an aerodynamic
resistance that is a function of the turbulent exchange properties of the atmosphere, a quasi-laminar
boundary layer resistance that accounts for the role molecular diffusivity plays adjacent to leaf surfaces,
and a surface canopy resistance that accounts for leaf transfer processes involving  stomatal,  cuticular, and
mesophyll resistances as well as soil resistances. Simplified models of this type, which simulate canopy
transfer processes as if the canopy was a single or "big leaf, show exchanges driven by the concentration
gradient of the species in question (ozone, sulfur dioxide,  particles, etc.) limited by the  appropriate set  of
resistances in the diffusion pathway. Measured data or submodels must provide estimates of the
resistances and the source or sink strengths.  Errors may occur for environments having complex terrain
or patchy surfaces.

        When using big-leaf models for water vapor and sensible heat exchange (e.g., Stewart and Thorn,
1973; Jams  et al., 1976), internal leaf resistance is negligible.  For other gases, such as carbon dioxide
and air pollutants, leaf sinks or sources are taken into account by defining a virtual resistance to a zero
sink or by using a relationship between the environment and the source or sink strength. Such methods
have been used to predict carbon dioxide  exchange (Waggoner, 1969; Sinclair et al., 1976). For most
atmospheric pollutants, the zero-sink case is appropriate, and a deposition velocity is defined as the
inverse of the sum of the resistances (Hicks et al., 1987). Such models have been  used to predict
pollutant flux (Matt et al.,  1987; Hosker and Lindberg, 1982; Unsworth, 1980; Wesely and Hicks,  1977).

        Currently, estimations of resistance come from  climatological and physiological data.  Hicks et
al. (1985) propose estimating aerodynamic resistance from the standard deviation  of the horizontal wind
direction and mean horizontal  wind speed. Based on work by Brutsaert (1975)  and Garratt and Hicks
(1973), they also propose the quasi-laminar boundary layer resistance as a function of the Schmidt and
Prandl numbers.  Canopy surface resistance is the most difficult to estimate where the primary transfer
resistance is  through the stomata but may also include cuticular and mesophyll resistances.  An addition-
al complication arises when resistances vary through the canopy depth and a weighted average value for
the big leaf must be derived.  In some experimental work,  the canopy surface resistance is evaluated as a
residual by measuring the flux, the  total resistance, the aerodynamic resistance, and the quasi-laminar
boundary layer resistance (Baldacchi, 1987).  In the model, gross stomatal resistance is combined in
series with gross mesophyll resistance,  while cuticular resistance is combined in  parallel with these, and
the gross canopy resistance is  determined by weighing with the leaf area index (Hicks, 1985).  A review
of stomatal  resistance concepts is presented by Jarvis (1971).  Gross stomatal resistance can be modeled
as a function of PAR (Baldacchi  et al., 1987; Burroughs and Milthorpe, 1976) as mediated  by the effects
of water stress, humidity, and  temperature (Rodriguez and Davies, 1982; Fisher  et al., 1981; Jarvis, 1976).
Mesophyll and cuticular resistances, although measured experimentally, have yet to be adequately
modeled.
                                                6-18
-------
                                           SECTION 7

              DESCRIPTION OF DATA BASE AND QUALITY ASSURANCE PLAN
DESCRIPTION OF DATA BASE

        The Data Management Center (DMC) of the Fleming Group in Albany, NY, is the central
collection, processing, analysis, and distribution point for data collected at MCCP sites and laboratories.
DMC activities include data acquisition and entry, storage and documentation of data files, data valida-
tion, reduction, and certification, report generation and distribution, data archiving and analysis, and
model testing and evaluation.  Types of data handled are:

               physical measurements (e.g., presence of cloud, precipitation, cloud LWC)
               meteorological measurements (e.g., temperature, wind speed and direction)
               aqueous phase measurements (e.g., cloud, precipitation, and throughfall chemistry)
               gas phase measurements (e.g., ozone, sulfur dioxide, hydrogen peroxide)
               ancillary data  (e.g., site latitude, longitude, and elevation).

        These data are stored in an ORACLE relational database. The computer hardware is a DEC
VAX 11/750 with the VMS version 4.1 operating system. The database is implemented using ORACLE
version 5.0.20.  All programming applications are written in C using the Digital version 2.1 compiler.

        MCCP data are available from the Data Management Center by written request to The Fleming
Group, 55 Colvin Ave., Albany, NY 12206. Timeseries data sets have been developed for the  analysis of
multiple parameters in time sequences. The MCCP data are organized into hourly records containing all
measurements available for that hour, including meteorological, gas, physical, and chemical data.  The
data are further organized by site, month, and year, and are available on floppy disk or magnetic tape.
These data sets are available for  1986 to 1988 data.
INTERNAL QUALITY ASSURANCE

        Internal quality assurance/quality control (QA/QC)  for the MCCP is coordinated and managed
by the Fleming Group, in Albany, NY. QA consists of:

               development of standard operating procedures (SOPs) and quality assurance plans;
               preparation of training sessions and site technician certification;
               writing annual QA/QC summaries;
               conducting system audits; and
        •      data validation and  certification.

        Standard quality control practices for the field (i.e., methods of tracking performance) include
training (by on-site personnel), station checks, station logs, zero/span checks, precision measurements,
and calibrations.  All station checks, routine measurements, and record keeping follow the MCCP SOP
and QA plan.  Table III-l summarizes the data quality objectives for the MCCP.

        Additional quality control is provided by a site QA coordinator who verifies that monitoring
procedures are followed, checks all instrument calibration records, and confirms any effect of problems
on the final data. The network QA officer informs  the project manager of problems at each site.

                                                7-1
-------
        The QA office also monitors laboratory quality control procedures for both the on-site laborato-
ries and a central analytical laboratory (CAL), the Illinois State Water Survey (ISWS), that analyzes
samples from all the sites.  The latter is the same laboratory that analyzes the NADP/NTN rain samples.
The CAL established the QA/QC protocols for the laboratories and prepares QC check solutions.

        Performance of site labs with respect to CAL are evaluated by split sample comparisons. A
split is a sample that has been divided into two aliquots in the field.  The results from the analyses of
these split samples are used to determine the precision between the two labs. Table III-2 summarizes
the splits analysis results  for 1986, 1987, and 1988.

        An internal systems audit for each site is performed annually by the QA Office.  Inspection
includes evaluation of all aspects relevant to  the goals of the program and adherence to the Project QA
Plan.  Upon completion,  a report is written to describe the results of the audit and recommending
actions for improvement.

        Quality Assurance  reports to management have been prepared on an annual basis.  Copies of
these reports may be obtained from The Fleming Group or the EPA
                                                7-2
-------
                                         APPENDIX A

                                  EUROPEAN MONITORING

        We present this review European cloud water chemistry to provide a context for viewing USA
and Canadian data.  Few true "networks" exist; site specific data for Sweden and Italy are included for
comparison with FRG and North American data.

GERMANY

        Since 1916, several researchers in the Federal Republic of Germany have collected fog/cloud
water using string collectors (Linke, 1916; Rubner, 1931; 1935; Grunow, 1953; 1957;  1958; Baumgartner,
1958a; 1958b). The interest in fog/cloud chemistry has grown in recent years primarily in response to an
increase in forest damage.  It has been hypothesized that chemical species in fog/cloud water intercepted
by forest canopies contribute to that damage. Hence, several institutions developed a variety of fog/clo-
ud collectors.  To ensure data compatibility and  quality, in 1986 the Federal Ministry for Research and
Technology (BMWFT) funded a comparison of collectors at the Center for Environmental Research,
University of Frankfurt (Enderle and Jaeschke, 1988).

        Although there is currently no cloud monitoring network in Germany, several groups have
initiated limited field experiments.  Of special interest are cloud chemistry data obtained over an
extended period of time, such as reported by the University of Frankfurt group (Schmitt, 1989), the
German Weather Service in Hamburg (Kroll and Winkler, 1989), and the University of Bayreuth
(Trautner, 1988).

        The Frankfurt group obtained cloud samples from Little Feldberg  Mountain, a  region close to
urban centers and heavily industrialized complexes, including chemical  factories. These results are shown
in Table Al-1.  The Weather Service group operated five mountain sampling sites during an 18-month
period from  October 1986 through  May 1988: Hohenpeissenberg (HP), Kahler Asten (KA), Wasserkuppe
(WK), Grosser Arber (GA), and Schauinsland (SL).  Each fog event contributed one sample for
chemical analysis. These values, presented in Table Al-2, were standardized to a uniform liquid water
content (LWC) of 0.2 g/m3 to avoid additional variability due to different LWCs.


                                          TABLE Al-1

        Arithmetic mean, maximum, and minimum concentrations in 410 cloud/fogwater samples
          for the Calls of 1983 - 1986 in Kleiner Feldberg/Taunus, 15 km north of Frankfurt
                                       Values are in
                      Mean          Maximum      Minimum

                      831              5620          87.4

        NO/          778              7067          28.6

        Cl-            243              2257          17

        NH^+         1197             5210          55.4

        H+           158              5011           0.0126


                                              A-l
-------
                                           TABLE Al-2

               German Weather Service Mountain Sites (Active impactor-type collector)
                           All values are normalized to LWC of 0.1

                                              Station
Ion Species
Oteq/L)

H+
NO/
NO/

# of events
LWC (g/m5)
               HP
151
110
102
172

381
  0.025
               KA
200
219
229
321

126
  0.057
               WK    SL
 43
108
 82
139
 85
 73
 47
133
 20     22
 0.055   0.042
                      GA
148
260
223
310

195
  0.127
        The Bayreuth group (Trautner, 1988) collected a total of 25 weekly samples at a mountain site
in the Bavarian Fichtelgebirge  (Wuelfersreuth). Each fog event yielded one sample, and no attempt was
made to differentiate precipitating and  non-precipitating events.  The mean ion concentrations are listed
in Table Al-3.

                                           TABLE Al-3

               Weekly Fog/Cloud Water Samples Obtained with Passive String Collector
                                 at Wuelfersreuth (Fichtelgebirge)
                          September 1984 - March 1986 (Fall/Spring only).
LWC (g/m5
Ion Species (/teq/L)
  so/2-
  NO/
                              Mean
                                3.9
               336
               334
               253
                              Standard
                              Deviation

                                    4.0
                   191
                   192
                   149
                       Maximum

                          3.4
                          0.137

                        892
                        936
                        569
                              Minimum

                                6.5
                                0.003

                               65
                               68
                               44
        Appreciable regional difference sin the chemical composition of cloud water have been observed
for these mountain measurements. The mean concentrations varied greatly not only among different
stations but also within stations.  Since each investigator adhered to his/her own standard operating
procedures for cloud/fog collection and analysis, it is difficult to prepare a  quantitative summary from the
results to date in Germany.  It  is clear, however, that cloud/fog water was enriched compared with
precipitation water for all major ions (including those not summarized in Tables Al-1 through Al-3).
The results show that high-elevation forests were exposed to substantial concentrations of chemicals in
clouds/fogs.
                                               A-2
-------
ITALY

       A field test to investigate microphysical and chemical characteristics of fog was organized by
FISBAT (Istituto Perlo Studio dei Fenomeni Fisici e Chimici della Bassa e Alta Atmosphera) in the
eastern Po Valley during February and November of 1984 (Fuzzi et al, 1988).  Twelve American and
European institutes participated (Atmospheric Sciences Research Center, Albany, NY  (USA); Atlanta
University, Atlanta, GA (USA); Italian National Electricity Board (Italy); Lawrence Berkeley Laboratory,
Berkeley, CA (USA); Technical University Vienna (Austria); University of Bologna (Italy); University of
Frankfurt (FRG); University of Padova (Italy) and University of Vienna (Austria).  Rotating string
collectors collected fog on an hourly basis.  Twenty samples were obtained during February 1984, and 87
samples from six events were obtained during November  1984.  The mean, maximum, and minimum
concentrations for the major chemical components in the Po Valley fog are presented  in Table Al-4.
The average anion to cation ratio was 0.96,  an indication that no important constituents  were missed.
Four ions accounted for most of the ionic strength of the fog water solutions:  H+, Nrfy+, NO/, and
SO^2'. The average liquid water content  (LWC) was 0.08 g/nv3 (maximum 0.29 g/nr*)  measured by a
high-volume filter technique.


                                          TABLE Al-4

           Hourly Fog Samples Obtained with Rotating String Collectors in Po Valley, Italy
                                   February and November, 1984.

                                Mean              Maximum            Minimum
                         Feb      Nov       Fcb      Nov        Feb      Nov

H+                       200       160       3600     1200         10        1
SO^-                    1000      900       6300     2100        350       150
NO/                    1100      900       8200     3300        290       80
NH^+                   2100     1400       8100     4200        700       300
H2O2                     <  0.1

LWC (g/m5)                  0.08                 0.29


SWEDEN

       A ground-based passive string collector sampled water" from stratiform clouds for approximately
five weeks during the summers of 1983 and  1984 in the mountains of central Sweden (Areskutan, 1250
m asl; Ogren and Rodhe, 1986). These measurements were considered representative to air arriving over
central Scandinavia. Cloud liquid water content was  measured only in 1984,  using a heated-rod impactor
King probe (King et al., 1978).  A total of 179 cloud water samples (resolution less than one hour) were
obtained and a wide range of concentrations was encountered; for example, sulfate concentrations ranged
from 1 to 1600 neq/L.   Table Al-5 shows  the mean ion concentrations classified into four air mass
arrival sectors. Transport of air from  the North Atlantic Ocean (NW) was associated with low concentra-
tions, while high concentrations occurred  with transport from industrial regions in Europe  (S). Long-ra-
nge transport appears  to be the primary factor controlling the chemical composition of cloud water at
the sampling site in central Scandinavia.
                                               A-3
-------
                                          TABLE Al-5

           Mean Ion Concentrations in Hourly Cloud Water Samples from Four Trajectories
                                for Areskutan, Sweden, 1250 m asl
            A Passive String Collector Sampled Precipitating and Non-precipitating Clouds
                                in the Summers of 1983 and 1984.
                                           Trajectory

                         NE           NW           W              S
Ion Species (jieq/L)

H+                      33            13             38             370

                         34             6        '     30             700
                       max 150       max 19        max 220        max 930

NO/                     9             2             10              68

NH4+                    17             0.5             1              NA


LWC           Averaged over all directions = 0.16 g/nv*

No. of Samples           47            41             33               4


CANADIAN MONITORING

       There  are areas of forest  at elevations above 600 m scattered throughout southern Quebec, along
the Gaspe Peninsula and in northern New Brunswick. Because higher elevation forests show
unexplained  damage symptoms in central Europe and the eastern United States, the Chemistry of High
Elevation Fog  (CHEF) program was started in late 1985 to measure wet and dry deposition as well as
the meteorological conditions at several higher altitude sites.  The objectives of the CHEF program are
similar to those of the Mountain  Cloud Chemistry Project  (MCCP) in the USA (Schemenauer, 1986;
Schemenauer et al., 1988; Schemenauer and Winston, 1988).  The CHEF program operates 12 months a
year on two  mountains (Mt. Tremblant and Roundtop Mountain) and for four months on a third
mountain (Mt. Epaule).  Paired fog/cloud water samples are collected with a passive string collector that
is very similar  to the MCCP cloud water collector. The program is designed  to sample every fog event
on each day of the year. The sampling periods can vary from 1  to 24 hours depending on operator
availability.  Because an operator is at the field  station only eight hours per  day, when fog is expected
overnight a pair of collectors are  exposed at the end of the afternoon and the samples collected the next
morning.

       The CHEF program is still in the initial stages of data  analysis and interpretation. Cloud/fog
samples at Mt. Morency were collected between 25 June and 24 September 1986.  The data are
presented as sample means in Table Al-6.  The concentration of all ions was 10 to 20 times  higher in
non-precipitating cloud water than in  precipitation.  The precipitating cloud/fog samples had
intermediate values but more closely resembled  the precipitation data.  The dominant cations were H+
and NH^+ and the dominant anions SO/' and NO/. Fog/cloud samples from Roundtop collected
between 1 May and 25 September 1986 are also shown in Table Al-6.  Note that the small number of


                                              A-4
-------
samples precludes any comparison of precipitating and non-precipitating clouds without more recent
CHEF work.

        The principal CHEF sampling locations are at altitudes near the typical cloud base heights for
southern Quebec (845 to 970 m).  Thus, the main depth of the cloud is usually above the sampling
location and the likelihood of precipitating being mixed with cloud events is high. Any indication of
precipitation at the field sites during the exposure of the passive fog collector results  in the sample being
classified as a "precipitation fog/cloud" event. The CHEF data are in agreement with those reported for
northeastern MCCP sites.  Schemenauer (1986) reported that the CHEF monitoring sites in Quebec are
in cloud approximately 44% of the year, with lower elevations (about 500 m) experiencing clouds for
about 23% of observations.

        Schemenauer et al. (1988) collected paired samples to screen the data for possible
contamination.  These results describe the combined experimental precision related to uncertainties in
cloud water collection (ASRC-type passive sampler), chemical analysis, and handling procedures. About
3% of the CHEF samples were eliminated due  to a "significant" difference in concentration between the
sample pairs. The remaining 17 pairs of samples had a median difference of 4 neq/L for sulfate (mean
total concentration was 199 jteq/L) and 1.9 jteq/L  for nitrate (mean concentration was 65 jieq/L); the
median difference was about 2% of observed concentrations.

        Except  for the length of the sampling period the CHEF and Mountain Cloud Chemistry Project
(MCCP) protocols  are similar. Samples from MCCP and CHEF have been exchanged for comparisons
of analyte concentrations.  It is anticipated that for non-precipitating cloud water that the CHEF and
MCCP cloud water concentration data are directly comparable.


                                          TABLE Al-6 .

                   Mean Ion Concentrations at  Two Canadian CHEF Sites in 1986
                              (from Schemenauer and Winston, 1988).
                      Morency           Roundtop         Roundtop
                      (970 m)            (970 m)            (845 m)
                      NP*    P         NP     P         NP     P

# samples               7    20          3     57          17    81
pH                     3.42   3.66        4.52    3.65         3.89  3.71

Ion species (iieq/L)
H+
SOj2'
NOy
NH,+
383
522
87.5
215
218
301
51.1
141
29.9
44
21.9
52.4
224
244
104
116
129
227
72.9
149
197
248
104
125
a NP = non-precipitating clouds
  N  = precipitating clouds
                                               A-5
-------
                                         APPENDIX B

                                        DATA REPORTS

       This section lists the principal data reports and publications from MCCP, ISF, and CHEF cloud
water routine monitoring activities. A brief summary of each is provided.

       Mohnen and Kadlecek (1989) presented typical winter and summer events to contrast
time-dependent histories of chemical concentrations for the two seasons. During  summer, sufficient
hydrogen peroxide (H2C>2) existed to oxidize available SC>2 to SO^, temporarily depleting the H2C>2.  In
winter, H2O2 and oxidation rates were low; thus, sulfate concentrations in cloudwater were generally
independent of SC>2.  A chemical cloud  climatology for Whiteface Mountain was presented for data from
1982-1987.  Summer mean hydrogen ion concentrations ranged from a low of 174 /ieq/L in 1987 to a
high of 331 2 were higher in summer than in fall (0.8 versus 0.15 ppb) and were strongly correlated with ozone,
temperature, and dew point.  Daytime H2C>2 exceeded nighttime values by 26%. Cloudwater acidity data
are not presented here.   During spring, summer, and fall of 1986 at Whitetop Mountain, the mean
cloudwater [H2O2] was  26 jimol/1 (for 100 samples), while the mean rainwater H2C>2 was  10 ionol/1 (28
samples).  A maximum  aqueous H2C>2 of 247 /imol/l was observed, the highest value reported to date in
the literature; 13% of summer cloud samples showed greater than 120 junol/l ^C^.  Comparison of
rain and cloud H2C>2 during periods of poor vertical mixing in the atmosphere suggested that
levels aloft exceed those at the ground.


                                               B-l
-------
       Mueller and Weatherford (1988) computed cloud deposition on Whitetop Mountain, VA for 26
cays in the spring of 1986. Using Lovett's model (Lovett, 1984), the computed cloud deposition for the
study period.  The cloudwater SO^ flux was between 5.3 and 9.1 kg/ha/mo, while NOj flux was between
2.8 and 5.4 kg/ha/mo.  The ranges reflect projected variation in unmeasured model input data.

       Lindberg et al. (1988) used Integrated Forest Study (IPS) data to estimate cloud, precipitation,
and dry deposition at a high-elevation site in  the Great Smoky Mountains.  Cloudwater SO^2' and NO/
deposition were estimated to be 7.2 and 2.2 kg/ha/mo, respectively, both two to five times greater than
estimates from a nearby low-elevation site for January-April 1986.  This is the only published paper from
the IPS cloud water studies.

       More recent deposition estimates from IPS appear in discussions by Lovett, Knoerr and Conklin,
and Ragsdale in a draft summary of the IPS that was edited by Lindberg and  Johnson (1989). This
report does not present cloud water concentrations and presents cloud deposition only in the form of
graphs.  Cloudwater H+ deposition at these two sites was estimated to be 30% to 40% of total H+
deposition. The IPS scientists concluded that the primary sources of uncertainty in cloudwater
deposition were immersion time, cloudwater amount, and ion concentrations.

       Other than Lovett's original work at Mt. Moosilauke,  the best documented estimates of cloud
water chemical exposure and deposition are from the Mountain Cloud Chemistry Program (MCCP) by
Mohnen and co-workers (1988a; 1988b;  1988c).  The CDM-S and CDM-D models (Mueller, 1989;
Krovetz et al., 1989) were used to predict  deposition to the spruce-fir forests  on  five mountains in the
eastern USA.  Sigmon et al.  (1989) and Joslin et al. (1988) estimated  deposition at Shenandoah and
Whitetop Mountain, VA, from canopy throughfall measurements.
                                               B-2
-------
                  APPENDIX C








       SUMMARY OF METEOROLOGICAL DATA




FOR THE MOUNTAIN CLOUD CHEMISTRY PROJECT SITES




                   1986 - 1988
-------
               SUMMARY OF METEOROLOGICAL DATA
                   SITE:   HOWLAND FOREST
      Periods of Record
Average Temperature (C)
1987:  Apr 11 - Nov 11
1988:  Mar 25 - Nov 15

1987
1988
Max
Min
Mean
Average
1987
1988
Mean
APR MAY
8.4 12.1
5.4 13.2
25.6 31.1
-5.4 -2.8
6.6 12.7
Relative Humidity
59.4 57.3
68.9 62.0
65.1 59.6
JUN
16
17
35
1
17
(%)
67
66
67
.9
.0
.9
.6
.0
:
.8
.2
.0
JUL
18
21
33
4
19

73
77
75
.8
. 1*
.0
.9
.7

.6
.5*
. 1
AUG
17.
20.
• 34.
3.
18.

68.
74.
71.

9
0
2
4
9

1
1
0
SEP
13.
13.
28.
-2.
13.

74.
74.
74.

5
7
7
5
6

4
3
3
OCT
7.
6.
26.
-5 .
7.

72.
73.
72.

5
4
9
0
0

2
6
9
Total Precipitation (mm)
1987
1988
Mean
Average
1987
1988
Max Hr.
Mean
55.1 62.7
48.5 13.0
51.1 37.8
Wind Speed (m/s) :
2.8 3.1
3.2 3.1
8.9 10.9
3.1 3.1
Resultant Wind Direction
1987
1988
Mean
Average
1987
1988
Max
Mean
Average
1987
1988
Max
Min
Mean
30.5 258.1
6.3 243.8
15.8 250.9
Solar Radiation (
333.1* 237.5
242.2* 299.0*
869.0 906.3
277.4 263.3
Pressure (mB ) :
1013.8 1012.9
1003.8 1010.8
1025.2 1026.2
973.4 994.2
1007.8 1011.9
54
47
51

2
2
8
2
.9
.2
.0

.7
.8
.4
.7
53
68
59

2
1
6
2
.6
. 8*
.6

.4
. 8*
.7
.2
25.
121.
73.

2.
2.
9.
2.
4
9
6

5
0
0
o
146.
41.
93.

2.
2.
7 .
2.
3
1
7

6
6
Q
6
56.
84-.
70.

2.
2.
8.
2.
1
O
•-J
2

6
6
8
6
(degrees ) :
241
290
285
W/m2
220
340
956
270

1007
1004
1019
987
1006
-- No Data, * Data Recovery
.4
.4
.9
):
.5
.7*
-8.
.3

.8
.5
.6
.2
.2
50%
220
243
230

197
182
954
191

1007
1009
1017
994
1008
.6
. 9*
.0

.6
.0*
.3
.4

.2
.5*
.7
.8
. 1
to 85%,
257.
234.
245.

213.
175.
880.
195.

1007,
1009.
1020.
995,
1008
0
4
7

9
,5
5
,2

.7
.4
.3
.8
.6
264.
264.
264.

140.
203.
801.
167.

1008,
1010.
1026,
984.
1009
2
5
4

3
,9*
7
,2

.3
.5
. 0
,3
.4
# Data Recovery
233.
244.
239.

99.
110.
658.
104.

1010.
1011.
1028.
988.
1010.
9
0
0

8
2*
5
2

.2
4
.6
1
.8
f < 50%
-------
               SUMMARY OF  METEOROLOGICAL  DATA
                  SUBSITE:   MT.  MITCHELL  1
Periods of Record:
Average Temperature (C):

1986
1987
1988
Max
Min
Mean
Average
1986
1987
1988
Mean
APR MAY
11.6*
12.8
10.8
20.5
2.0
11.7
Relative Humidity
__
90.2
71.4
79.6
1986;
1987
1988:
JUN
13.
13.
13.
22.
0.
13.
(%):
79.
84.
73.
79.
: May 12
: May 14
: May 10
JUL
3
3
6
1
4
4

7*
4
3
1
14
15
14
23
6
14

82
84
84
83
.0*
. 1
.8
.5
. 1
.8

. 3*
.4
.2
.9
- Nov 20
- Oct 22
- Oct 1
AUG
12
14
15
23
-0
13

90
88
88
88
.2
.6
.2*
.5
.6
.9

.0
.0
.1*
.7
SEP
11
10
12
19
2
11

87
90
88
89
.6
.3
. 0*
.9
.2
.3

. 9*
.8
.5*
.2
OCT
7
4
-
17
-10
6

77
60
-
70
.3
.8
-
.2
.2
.2

.9
.7
-
.5
Total Precipitation (mm):
1986
1987
1988
Mean
Average
1986
1987
1988
Max Hr.
Mean
Resu Itan
1986
1987
1988
Mean
Average
1986
1987
1988
Max
Mean
Average
1986
1987
1988
Max
Min
Mean
57.2
79.2
• -- 26.9
52.4
Wind Speed (m/s):
6.7
4.2
5. .6
21.0
5.6
t Wind Direction
271.9
266.5
284.6
275.0
Solar Radiation (
165.3
162.6
301.1*
1014.8
226.8
Pressure (mB ) :
807.0
810.4
806.1
817.3
798.4
807.7
2.
145.
36.
61.

6.
6.
5.
19.
6.
3
3
8
6

1
5
8
2
1
0
51
96
58

8
5
5
21
6
. 0*
.3
.8
.5

. 3*
.3
.5
.6
.0
4
47
105
49

6
6
5
20
6
.3
.2
. 2*
.4

.4
.5
.6*
.0
.2
22
306
145
161

6
6
6
18
6
.9
.6
.3*
.0

.4
.6
.5*
.9
.5
186
3
-
123

7
7
-
19
7
.9
.3
-
.8

. 7
.4
-
. 9
.6
(degrees) :
293.
289.
292.
291.
W/m2)
220.
231.
355.
1075.
272.

808.
808.
808.
818.
796.
808.
3
7
6
9

6
5#
5
9
1

5
5
5
7
9
5
296
293
270
285

206
-
316
971
266

809
810
811
817
801
810
. 2*
.4
.7
. 1

. 1*
—
. 6*
.3
.7

.4*
.9
.3
. 1
.5
.7
298
349
245
300

140
-
285
946
192

812
810
810
823
802
811
.9
.8
.4*
.8

.9
-
.9*
.6
.0

.8
.2
.3*
.6
.2
.1
284
83
265
206

150
-
269
910
193

815
807
808
828
794
810
.8
.4
.5*
.9

.4
-
.5*
.7
.2

.8
.3
. 9*
. 1
.7
.8
295
297
-
296

135
-
-
860
135

811
804
-
821
793
808
1—1
1
-
• -

c,
-
-
3
• 1

. 3
. 6
-
.4
.0
. 6
-- No Data,   * Data Recovery 50% to 85%,   #   Data  Recover  <  50%
-------
               SUMMARY OF METEOROLOGICAL DATA
                  SUBSITE:   MT.  MITCHELL 2
      Periods of Record
Average Temperature (C)
1986:  Jun 28 - Nov 20
1987:  May 18 - Oct 23
1988:  May 12 - Sep 30

1986
1987
1988
Max
Min
Mean
Average Re
1986
1987
1988
Mean
Total Free
1986
1987
1988
Mean
APR MAY
— —
15.1
12.3
19.7
5.0
13.4
JUN
17.5
15.6
15.6
24.6
2.2
15.7
JUL
17.2*
17.6
17.0
25.0
8.8
17.3
AUG
—
17.0
17.5
26.3
10. 1
17.3
SEP
14.5*
12.7
13.4
22.1
3.9
13.4
OCT
8.7
3.3*
—
21.0
-7.9
8.0
lative Humidity (%):
__
80.9
66.4
72.3
ipitation (mm)
__
40.6
19.3
28.0
73.8
83.2
67.7
75.1
:
11.4
121.9
40.6
78.7
71. lit
77.6
78. 1
77.2

200.4
54.1
95.8
116.4
--
83.2
83.0
83. 1

--
58.9
122.2
90.5
81.3
87.3
84.5
67.2

23.0*
309.6
126.5
170.5
__
56. 6*

56.6

164.6
0.04*
—
142.4
Average Wind Speed (m/s):
1986
1987
1988
Max Hr.
Mean
Resultant
1986
1987
1988
Mean
Average So
1986
1987
1988
Max
Mean
__
2.8
4.3
12.4
3.7
Wind Direction
__
257.5
261.0
259.6
lar Radiation
__
159.5
229.2
1027.5
200.6
3.3
3.9
3.3
13.0
3.6
(degrees
287.9
277.9
305.6
291.6
(W/m2):
187:7
213.9
252.9
1027.5
231.8
3.8
3.6
3.9
16.6
3.8
):
284.9
281.0
260.8
275.5

221.8
201.8
208.2
1052. 1
210.8
3. 1*
3.7
3.2
11.6
3.4

290. 1*
283.4
260.1
274.8

155.0*
177.7
193.2
979.6
180.3
3.9
4. 1
3.6
13.9
3.9

264.8
251.3
255.8
257. 1

170.2
157.5
152.4
954 . 2
159.9
4.2
5.3*

12.1
4.3

266.5
286.6*
—
269.2

153.0
204. 1*
--
861.9
159.9
Average Pressure (mB):
1986
1987
1988
Max
Min
Mean
__
824.9
820.1
839.3
812.1
822.1
828.6
822.6
823.1
832.8
812.2
823. 1
831.8
825.0
826.3
838.6
815.6
827.7
831.5*
824.5
825.6
835.5
817.6
826. 1
832.5
821.7
824.5
837.5
809.6
826. 1
829.3
817.7*
--
809.6
838.8
827.7
— No Data,  * Data Recovery 50% to 85%,  * Data Recovery < 50%
-------
               SUMMARY OF METEOROLOGICAL DATA
                  SUBSITE:  MT .  MITCHELL 3
Periods of Record
Average Temperature (C)
                             1988:  May 14 - Sep 29

1988
Max
Min
Average
1988
APR MAY
17.2
28.2
2.3
Relative Humidity
68.4
JUN
19
31
4
(%)
71
.7
.9
.3

.4
JUL
21.
33.
9.

80.
AUG
6
0
2

2
22
33
12

84
.0
.8
.3

. 1
SEP
18
28
6

85
. 1
.7
.9

.9
Total Precipitation (mm):
1988
Average
1988
Max Hr.
Result an
1988
Average
1988
Max
Average
1988
Max
Min
22.1
Wind Speed (m/s):
1.4
6.1
t Wind Direction
193.8
Solar Radiation (
245.4
1060.7
Pressure (mB ) :
927.3
934.9
920.3
64

1
4
.3

.2
.3
105.

1.
5.
9

1
4
. 71

1
4
.4

.0
.2
117

1
3
.3

. 1
.7
(degrees ) :
222
W/m2
258
1015

929
940
917
.9
):
.9
.8

.5
.3
.8
198.

222.
987.

932.
941.
923.
7

5
2

0
0
7
210

204
957

930
942
921
.5

.6
.9

.6
.9
.7
211

159
892

930
940
917
.8

.6
.4

.9
.3
.3
                                                           OCT
-- No Data,  * Data Recovery 50% to 85%,  # Data Recovery <  50%
-------
               SUMMARY OF METEOROLOGICAL DATA
                   SITE:   MT.  MOOSILAUKE
      Periods of Record
Average Temperature (C)
1986:  Jul 12 - Oct 22
1987:  Apr 9 - Oct 15
1988:  Jun 2 - Oct 18

1986
1987
1988
Max
Min
Mean
Average
1986
1987
1988
Mean
APR MAY
__
6.5 9.9
__
21.7 25.7
-4.5 -3.1
6.5 9.9
Relative Humidity
__
65.4 66.2
— —
65.4 66.2
JUN
—
14.0
14.1*
28.6
2.4
14.0
(%):
--
77.4
61.6*
70.5
JUL
15.9
18.0*
18.9
28.7
4.5
17.8

85.4
80.9*
74.9
79.8
AUG
13.5*
14.7
17.1
28.7
1.2
15.3

82.1*
77.3
82.6
80.6
SEP
9.8
10.9
10.8
20.8
-0. 1
10.5

79.8
84.6
79.3
81.3
OCT
4.7
4.8
4.2
19.4
-5.9
4.6

76.8
76.7
86.0
79.8
Total Precipitation (mm):
1986
1987
1988
Mean
Average
1986
1987
1988
Max Hr.
Mean
Resultan
1986
1987
1988
Mean
Average
1986
1987
1988
Max
Mean
Average
1986
1987
1988
Max
Min
Mean
__
24.6 30.2
__
24.6 30.2
Wind Speed (m/s):
--
4.6 4.6
— —
13.3 13.3
4.6 4.6
t Wind Direction
__
53.5 307.7
__
53.5 307.7
--
163.8
26.4
100.9

--
4.7
4.9*
17.8
4.8
(degrees^
--
306.5
307.5*
306.9
83.3
—
68.8*
74.4

3.2
3.4*
3.3
13.0
3.3
) :
312.0
312.2*
303.4
308.8
101.6
128. 5#
139.7
125.2

4. 1*
4.1
4.0
15.4
4.1

265.0*
305.9
288.6
294.8
82.3*
141.0
51.8
81.4

4.4*
4.7
5.0
15.4
4.7

--
313.4
296. 1
304.8
23.9
46.7
9.1
25.2

4.5
5.3
4.6
21.3
4.8

311.6
327.9
273.7
303.4
Solar Radiation (W/m2):
__
313.1* 227.3
— —
890.3 923.1
313.1 227.3
Pressure (mB ) :
— —
905.5 908.0
— —
915.2 918.3
880.5 892.8
905.5 908.0
--
197.1
259.5*
966.0
224.3

--
905. 1
904.3*
915 .8
887.3
904.8
176.9
209.7*
226.1
968.9
207.3

907.8
908.8*
909.7
919.8
894.4
908.9
183.6*
208.5
225.3*
994.4
206.8

908. 1*
908.4
909.2
918.3
892.0
908.6
144. 1
125.6
151.7
816.8
140.3

908.4
907.3
908.8
921.7
887.1
908.2
102.0
iio.o-
65.9
703.5
92.2

908.0
904.7
907.7
921.0
886.3
906.2
 -- No Data,  * Data Recovery 50% to 85%,  tf Data recovery
                               50%
-------
               SUMMARY OF METEOROLOGICAL DATA
                   SUBSITE:  SHENANDOAH 1
      Periods of Record
Average Temperature (C)
1986:  Apr 24 - Nov 24
1987:  Apr 20 - Nov 15
1988:  Apr 14 - Nov 3

1986
1987
1988
Max
Min
Mean
Average
1986
1987
1988
Mean
APR MAY
15.1 14.0
9.6 15.0
6.7 13.7
23.6 27.0
-1.9 -0.4
9.2 14.2
Relative Humidity
48.9 68.4
64.4 64.6
55.4 72.0
57.0 68.5
JUN
18
19
17
29
4
18
(%)
61
68
66
66
. 1#
.4
.4
.8
.2
.3

. 8#
.2
.2
.2
JUL
20
21
21
33
6
21

84
69
69
71
. 2#
. 8*
.6
.9
.5
.5

. 3#
. 9*
. 1
.4
AUG
17
20
21
32
4
19

85
69
72
76
.0
.3
.0
.5
.0
.4

.5
.4
.7
.0
SEP
15
16
14
26
1
15

80
75
79
78
.2
.7
.9
.4
.5
.6

.7
. 2*
.3
.5
OCT
10
8
6
25
-4
8

69
59
64
64
. 1
.2
.4
.4
.0
2

.3
.3
.3
. 2
Total Precipitation (mm):
1986
1987
1988
Mean
Average
1986
1987
1988
Max Hr
Mean
2.3 123.7
19.3 154.2
10.9 217.9
11.9 165.3
Wind Speed (m/s):
5.7 5.0
5.5 4.0
6.4 4.2
15.2 14.1
6.0 4.4
Resultant Wind Direction
1986
1987
1988
Mean
Average
1986
1987
1988
Max
Mean
Average
1986
1987
1988
Max
Min
Mean
301.1 265.0
12.1 259.0
294.1 214.6
208.1 246.2
Solar Radiation (
288.0 243.7
255.1* 230.4
219.5 220.1
979.3 1038.0
242.7 231.4
Pressure (mB ) :
902.0 901.9
901.0 906.1
895.5 903.2
908.3 911.9
884.9 892.2
898.4 903.7
-
91
41
66

4
4
4
12
4
-
.4
.1
.3

.5*
.0
.3
.7
.3
6
49
66
52

4
3
3
10
3
.9*
.0
.5
.4

. Itf
.7
.6
.3
.7
184
40
74
100

4
4
4
13
4
.4
.6
.9
.6

.2
.2
.0
.6
. 1
21
265
66
114

4
4
3
12
4
. 1
.9
.0
o

.2
.4
.9
. 1
.2
30
30
5
24

4
5
4
13
5
.7
.5
.8
- -J

.8
.3
.8
.5
.0
(degrees ) :
283
294
300
293
W/m2
286
258
264
1113
268

903
904
904
916
896
904
-- No Data, * Data Recovery
. 1*
.5
.6
.3
):
.3*
.6
.6
.0
.8

.6*
.2
.6
.8
.3
.2
50%
291
295
273
284

250
240
233
984
238

903
906
907
916
897
906
. 8#
.4
. 1
.4

. 3#
.3
.9
.0
.4

. 0#
. 1
.7
.0
. 1
.5
to 85%,
195
205
236
212

201
228
219
1037
216

905
906
907
914
898
906
.1
.2
.5
.3

.3
.2
.0
.0
.0

.6
.1
.3
.5
.3
.3
# Data
262
209
257
243

173
155
180
915
170

907
904
907
917
890
906
.0
.5
.5
.8

.6
.2
.7
. 1
.2

.3
.7
.7
.2
. 1
.6
Recovery
288
290
284
288

142
167
145
834
151

905
905
904
919
887
905
.4
.6
.9
.0

. 1
. 1
. 1
.8
.4

.8
. 1
.4
.3
.2
. 1
< 50%
-------
               SUMMARY OF METEOROLOGICAL DATA
                   SUBSITE:   SHENANDOAH 2
      Periods of Record
Average Temperature (C)
1988:  May 17 - Nov 22
1987:  Apr 22 - Nov 17
1988:  Apr 16 - Nov 2

1986
1987
1988
Max
Min
Mean
Average
1986
1987
1988
Mean
APR
—
11.3
9.7
25.5
2.1
10.3
Relative
__
59.4
46.9
51.5
MAY
18.6
17.9
16.8
29.8
3.7
17.6
Humidity
68.0
64.4
61.7
64.0
JUN
21
22
20
32
7
21
(*)
60
64
58
61
.6
.5
.4
.3
.2
.5

.6
.7
.0
. 1
JUL
23.
23.
24.
36.
9.
23.

68.
66.
61.
65.

6
5*
5
1
7
9

6
1*
7
4
AUG
20.
22.
23.
34.
8.
22.

72.
66.
64.
67.
SEP
2
8
9
9
4
3

4
2
3
7
18
19
17
29
6
18

69
73
70
71
.7
.1
.8
.3
. 1
.5

.6
.0
.5
.0
OCT
13.7
10.8
9.6
27.8
-0.6
11.4

60.1
54.0
55.2
56.4
Total Precipitation (mm):
1986
1987
1988
Mean
Average
1986
1987
1988
Max Hr.
Mean
--
16.3
10.2
12.4
98.0
125.2
201.2
149.7
94
89
38
74
.5
.2
.6
.1
1.
19.
86.
47.
5
1*
4
6
--
30.
79.
38.

2
0
4
13
218
76
100
.0
.9
.5
.7
24.6
32.3
31.8
29.6
Wind Speed (m/s):
--
3.6
4.1
12.6
3.9
2.1
2.2
2.9
11.5
2.5
Resultant Wind Direction
1986
1987
1988
Mean
Average
1986
1987
1988
Max
Mean
--
276.3
311.4
298.5
192.6
206.4
165.8
187.9
2
2
2
8
2
.6
.3
.7
.2
.5
2.
2.
2.
10.
2.
6
4*
2
8
4
2.
2.
2.
9.
2.
5
8
6
3
6
2
2
2
9
2
.4
.5
.4
.0
.4
2.3
3.0
3.0
9.9
2.8
(degrees ) :
253
246
285
261
.4
.2
.7
.8
256.
183.
195.
216.
4
4*
1
3
182.
176.
175.
178.
8
0
2
0
195
191
191
192
.1
.0
.7
.6
267.0
274. 1
287.3
276.1
Solar Radiation (W/m2):
--
218.6
184.3
965.6
196.5
217.0
228.2
216.6
1030.5
221.5
272
257
242
1075
257
.5
.2
.7
.4
.5
240.
248.
227.
970.
237.
5
2*
6
8
2
192.
223.
212.
973.
209.
1
7
1
3
0
169
158
169
880
166
.5
.7
.6
. 1
.0
132.5
151. 1
136.0
808.0
139.9
Pressure (mB) :
1986
1987
1988
Max
Min
Mean
--
936.1
934.8
975.6
922. 1
935.3
935.1
940.6
942.2*
969.0
930. 1
939.8
937
937
-
945
929
937
.0
.6
-
.3
.0
.3
937.
939.
938.
947.
927.
938.
9
7*
4*
8 '
4
5
939.
939.
938.
949.
929.
939.
3
8
4
9
2
2
941
938
940
950
922
940
.4
.5
. 1
.5
.3
.0
940.7
940. 1
937.9
955.9
921.4
939.6
-- No Data,  * Data Recovery 50% to 85%,  # Data Recovery < 50%
-------
               SUMMARY OF METEOROLOGICAL  DATA
                     SITE:   WHITEFACE
Periods of Record:
Average Temperature (C):

1986
1987
1988
Max
Min
Mean
Average
1986
1987
1988
Mean
APR MAY
— —
15.8
11.7
20.5
2.9
13.0
Relative Humidity
__
86.1
67.5
83.3
1986:
1987
1988:
JUN
8
11
7
23
-2
9
(%)
76
80
72
76
.0*
.0
.3*
.9
.9
.1
;
.4*
.8
.0*
.8
: Jun 19
: May 29
: May 26
JUL
11
13
14
31
0
13

88
85
.81
85
.9
.3
.6*
.3
.5
.2

.1
.5
.6*
.3
- Oct 28
- Oct 13
- Oct 17
AUG
9
11
12
27
-4
11

92
80
89
87
.7
.5
.8
.4
.2
.3

.2
.0
.1
.5
SEP
6
6
6
17
-4
6

89
89
83
87
.4
.2*
.6
.7
.5
.4

.0
.8*
.1*
. 1
OCT
0
-0
-0
13
-10
0

88
82
91
88
.7
.5
.2
.3
.9
.2

. 7*
.9
.7
.2
Total Precipitation (mm):
1986
1987
1988
Mean
Average
1986
1987
1988
Max Hr.
Mean
__
0.0
—
__
Wind Speed (m/s):
__
12.8
9.9
21.9
10.8
Resultant Wind Direction
1986
1987
1988
Mean
Average
1986
1987
1988
Max
Mean
Average
1986
1987
1988
Max
Min
Mean
__
277.0
276.0
276.3
Solar Radiation (
-_
190.1
247.4
934.7
229.6
Pressure (mB ) :
__
849.8
848.1
852.3
843.5
848.7
3
40
-
32

10
9
9
22
9
.3*
.9
-
.8

.3*
.3
.7*
.7
.6
54
66
-
60

8
8
7
23
8
.4
.8
-
.7

.4
.2
.6
.6
.1
47
23
-
39

8
7
8
25
8
.2
.4#
-
.8

.7
.6
.9
.3
.4
62
3
-
34

9
10
10
24
10
.5
.0*
-
.9

.2
.4*
.7
.2
.1
24
1
-
20

9
8
7
30
8
.9
.Ott
-
.5

.4
. 9*
.8
.0
.9
(degrees):
275
297
286
290
W/m2
225
189
248
1018
217

847
846
843
856
831
845
. 9*
.4
.8*
.4
):
.5*
.6
.0*
.5
.9

.7*
.4
.5
.8
.0
.2
283
77
260
200

190
203
174
955
191

848
848
848
859
833
848
.8
.6
.6*
.6

.3
.3
.6*
.9
.7

.9
.6
.8
.2
.4
.8
274
97
263
216

154
184
159
927
165

849
852
846
863
832
849
.9
.3
.9
.2

.0
.0
.0
.9
.0

.2
.6
.9
.7
.6
.4
266
271
263
267

110
100
129
774
114

848
853
845
865
822
848
.3
.5*
.7
.0

.1
. 9*
.5
.6
.0

.7
. 3*
.1
.7
.6
.9
256
257
231
250

77
92
46
669
71

845
849
842
859
821
845
.8
. 1*
.9
.6

. 1
.0
.7
.2
.7

.2
.1
.0
.4
.4
.1
— No Data,   * Data Recovery 50% to 85%,   #  Data Recovery <  50%
-------
              SUMMARY OF METEOROLOGICAL DATA
                     SITE:  WHITETOP
Periods of Record:
Average Temperature (C):

1986-
1987
1988
Max
Min
Mean
Average
1986
1987
1988
Mean
APR MAY
7.01* 9.8
1.0* 12.5
4.9 9.9
20.4 21.4
-11.9 -4.7
4.3 10.7
Relative Humidity
62. 6# 83.4
77.6* 81.2
70.6 67.8
70.8 77.5
1986
1987
1988
JUN
14.9
14.4
14. 1
24.0
1.3
14.5
(%):
84.6
85.2
69.1
79.8
: Apr 1
: Apr 1
: Apr 1
JUL
16.7
16.5
16.6
25.3
5.0
16.6

86.9
82.3
75.9
81.6
- Oct 31
- Oct 31
- Oct 31
AUG
13.9
15.9
16.8
25.3
1.2
15.6

91.3
85.6
81.1
85.9

SEP
12.8
13.4*
12.5
19.5
3.0
12.8

91.6
87. 3#
91.2
90.6

OCT
7.1
4.7*
2.9
18.0
-8.6
4.9

86.2
65.2*
71.4*
75.4
Total Precipitation (mm):
1986
1987
1988
Mean
Average
1986
1987
1988
Max Hr.
Mean
48.0 193.0
220.2 93.0
121.7 78.2
128.7 121.4
Wind Speed (m/s) :
2.4# 2.5
3.9* 2.3
3.7 2.8
14.6 10.2
3.5 2.5
Resultant Wind Direction
1986
1987
1988
Mean
Average
1986
1987
1988
Max
Mean
Average
1986
1987
1988
Max
Min
Mean
. 328. 4# 207.9
129.2* 243.6
177.2 240.7
195.2 230.9
Solar Radiation (
232. 8* 176.3
172.8 225.9
197.1 247.0
1068.7 1083.7
194.1 216.8
Pressure (mB ) :
828. 8#
826.0 836.3
826.8 833.1
839.1 842.2
810.3 822.5
826.8 711.0
74.2
152.9
63.5
97.6

2.4
2.5
2.2
8.1
2.4
(degrees
200.2
267.4
121.3
198.0
W/m2):
235.5
228.2
292.1
1090.7
251.2

834.1
835.8
836.2
846.3
824.0
835.4
90.4
66.0
103.4
86.6

1.9
1.9
2.5
7.9
2.1
):
229.3
283.6
241.9
252.0

230.8
235.3
230.7
1083.8
232.3

835.6
838.0
839.0
845.1
827.0
837.6
172.7
57.7
95.8
108.2

2.3*
3.0
2.9
9.5
2.8

191.7*
232.7
272.2
241.6

179.6
193.7
216.7
1013.6
197.0

839. 9#
837.7
838.3
844.1
830.1
838. 1
141.7
85.1
143.3
123.4

2.7*
3.6tt
3.3
10.9
3.2

191.7*
158.8*
2.1
85.3

141.2
151.3
147.7
996.0
146.7

838.1
836. 0*
837.4
844.8
822.6
837.4
95.0
23.6*
83.1
71.9

3.7
3.0
—
13.8
3.4

191.7
97. 1*
188.0
163.9

137.4*
174.6
148.7
871.8
155.1

835.4*
824.3
830.7
846.6
820.0
829.8
-- No Data,   * Data Recovery 50% to 85%,   * Data Recovery < 50%
-------
               SUMMA-RY OF METEOROLOGICAL DATA
                   SUBSITE:  SHENANDOAH 3
      Periods of Record
Average Temperature (C)
1986:  May 21 - Nov 23
1987:  Apr 24 - Nov 18
1988:  Apr 15 - Nov 1

1986
1987
1988
Max
Min
Mean
Average
1986
1987
1988
Mean
APR
--
11.4
11.4
25.8
2.6
11.4
Relative
--
52.5
44.9
47.3
MAY
18.5
18.3
17.6
31.2
3.6
18.0
Humidity
68.3
64.7
64.8
65.3
JUN
22.
21.
21.
34.
6.
21.
(%):
61.
62.
59.
61.
JUL
4
8
4
9
6
9

9
3
6
3
24
23
25
38
9
24

74
63
65
66
.0#
.6
.0
.4
.2
.3

.9#
.9
.9
.5
AUG
20.
22.
25.
37.
7.
22.

75.
63.
67.
69.
7
1
0
5
1
6

9
7
9
3
SEP
19
18
19
31
6
18

70
71
71
70
.4'
.3
.0
.5
.0
.9

.5
.1
.1
.9
OCT
14.1
9.6*
10.4
30.5
0.5
11.6

60.4
52.5*
57.0
57.2
Total Precipitation (mm):
1986
1987
1988
Mean
Average
1986
1987
1988
Max Hr.
Mean
Resultan
1986
1987
1988
Mean
Average
1986
1987
1988
Max
Mean
Average
1986
1987
1988
Max
Min
Mean
--
0.0
10.9
7.5
--
123.7
204.0
163.9
--
101.
35.
69.

6
3
0
-
45
85
65
-
.0
.1
.1
180.
31.
75.
93.
8*
5
9
3
29
225
80
110
.0
.0
.3
.7
22.4
33.3*
34.5
29.6
Wind Speed (m/s):
--
2.1
2.4
6.3
2.3
t Wind Di
--
359.6
13. 1
119.9
Solar Rad
--
209.6
205. 1
957.3
206.5
Pressure
--
959.2
954.6
968. 1
942. 1
956.0
1.3
1.6
1.8
5.7
1.6
rection
187.4
179.8
169.8
176.6
iation (
234.7
223.8
227.4
1041. 1
226.9
(mB):
958.5
964.1
961.8
970.7
950.4
962.3
1.
1.
1.
5.
1.
6
5
7
7
6
1
1
1
4
1
.6*
.4
.4
.9
.4
1.
1.
1.
5.
1.
6
7
5
5
6
1
1
1
5
1
.5
.5
.5
.3
.5
1.5
1.9*
1.8
5.5
1.7
(degrees) :
170.
157.
159.
162.
W/m2)
269.
255.
262.
1054.
262.

959.
961.
963.
975.
951.
961.
3
3
2
3

6
3
3
4
4

8
6
2
5
0
5
181
166
188
169

343
243
233
982
250

961
963
965
974
953
964
.2#
.4
.0
.5

.4*
.0
.0
. 1
.4

.2#
.7
.6
.7
.4
.1
182.
166.
166.
172.

224.
220.
221.
973.
222.

962.
963.
965.
973.
954.
963.
4
6
8
1

6
0
7
2
0

6
7
0
2
6
7
179
165
166
170

170
159
177
911
169

964
963
966
977
948
964
.3
.6
.1
.6

.5
.9
.4
.3
. 1

.8
.1
. 1
.9
.1
.6
170.8
91.3*
126.4
135.2

136.2
142.3*
140.4
835.3
139.2

964.4
963.5*
965.4
980.2
947.9
964.6
-- No Data,  * Data Recovery 50% to 85%,  # Data Recovery < 50%
-------
-------
  APPENDIX D




MCCP Publications
-------
Aneja, V.O., Bradow, R.L. and Jayanty, R.K.M., 1988.  "Organic Chemical Characterization of Clouds in
High Elevation Spruce-Fir Forests at Mt. Mitchell, North Carolina".  Proceedings of the 1988
EPA/APCA International Symposium on Measurement of Toxic and Related Air Pollutants, pp 227-236.

Aneja, V.P., Claiborn, Bradow, R.L., Paur, R.J. and Baumgardner, R.E., 1989.  "Dynamic chemical
characterization of montane clouds", Atmospheric Environment, in press.

Aneja, V.P., Claiborn, C.S., Li, Z. and Murthy, A, 1989.  "Exceedences of the National Ambient Air
Quality Standard for Ozone Occurring at a Pristine Area Site", Journal of Air Pollution Control and
Waste Management, in press.

Aneja, V.P., Claiborn, Li, Z. and Murthy, A.,  1989.  "Measurements at high elevations of Oxidants in the
Eastern United States and their role in Forest Decline", in Man and his Ecosystem, L.J. Brasser and
W.C. Mulder, Eds., Elsevier Science Publishers B.V., Amsterdam, Vol. 2, pp 189-194.

Aneja, V.P., Businger, S., Li, Z., Claiborn, C.S. and Murthy, A, 1989.  "Ozone Climatology at high
elevations in the Southern Applachians", Journal of Geophysical Research, submitted.

Aneja, V.P., Claiborn, C.S., Murthy, A. and Kim, S.D.  "Characterization of the Chemical and Physical
Climatology for evaluation of the role of air pollution in forest decline", in preparation for submission to
Atmospheric Environment.

Aneja, V.P., Businger, S., Li, Z., Claiborn, C, and Murthy, A, 1989.   "Ozone Climate at Mt. Mitchell,
North  Carolina, and its Association  with Synoptic Episodes", Proceedings of the 82nd Annual Meeting of
Air and Waste Management Association, Vol. 89-36.5, pp 1-28.

Aneja, V.P., Sookin, D., Murthy, A, Paur, R.J.,  Baumgardner,  R. and  Kronmiller, K., 1990. "Description
and performance of a cloud and rain acidity/conductivity (CRAC) Automated Sampling System",
submitted to the Journal of Air Pollution and Waste Management.

Bailey, B.H. and D.J. Smalley. 1987. Meteorological/climate aspects of mountain cloud chemistry
monitoring.  In: Proceedings of the Sixth Symposium  on Meteorological Observations and
Instrumentation, Jan. 12-16, 1987.  New Orleans, LA American Meteorological Society.  Boston, MA

Beltz, N., W. Jaeschke, G.L. Kok, S.N. Gitlin, AL. Lazrus, S. McLaren,  D. Shakespeare, and V.A
Mohnen. 1987.  A comparison of the enzyme fluorometric and the peroxyoxalate chemiluminescence
methods for measuring ^2^2- J. Atmos. Chem.  5:311-322.

Bradow, R.L. and Viney P. Aneja, 1988. "Aerosol Compositional Effects on Mountain Clouds:
Attributing Sources of Acid Deposition", in Atmospheric Aerosols and Nucleation, Paul E. Wagner and
Gabor Vali, Eds., Chapter 8, pp 40-43, Springer-Verlag, Berlin.

Claiborn, C.S. and Aneja, V.P., 1990.  "Measurements of Atmospheric Hydrogen Peroxide in the gas-
phase and in cloudwater at  Mt. Mitchell State Park, N.C.", in preparation for submission to Atmospheric
Environment.

Davis, J.M.,  Seabaugh, S.S., Bradow, R.L., and Monahan, J.F., 1990. "A Trajectory Climatology and Case
Study of Ozone Occurrence at the Mountain Cloud Chemistry  Program (MCCP) Site 1 at Mt. Gibbs,
North Carolina", submitted Atmospheric Environment.

DeFelice, T.P.  In press.  Occurrence of extreme episodes  of acidic deposition on coniferous forests in
Mt. Mitchell State  Park: aqueous phase.  Water. Air. Soil  Pollut.


                                               D-l
-------
DeFelice, T.P. 1989. Characterization of extreme deposition of air pollutants in Mt. Mitchell State Park:
potential for forest decline and opportunity for cloud deacidification. Doctoral dissertation. Department
of Marine, Earth, and Atmospheric Sciences, North Carolina State University. 200 pp.

DeFelice, T.P. and V.K. Saxena. 1988. Temporal and spatial distribution of ionic composition and acidity
in clouds: comparison between modeling results and observations, pp. 16. In:  Programme International
Congress of Geochemistry & Cosmochemistry, Aug. 29-September 2, Paris, France.

Estes, M.J.  and J.T. Sigmon. 1987. Comparisons of chemical compositions of  mountain stratiform clouds
and valley fog in the Shenandoah National Park, pp. 59-60. In: Proceedings of the Sixth Symposium on
Meteorological Observations and Instrumentation, Jan.  12-16, New Orleans, LA American
Meteorological Society. Boston, MA.

Fehsenfeld, F.C., J.W. Drummond, U.K. Roychowdhury, P.J. Galvin, E.J. Williams, M.P. Buhr, D.D.
Parrish, G.  Hubler, A.O. Langford, J.G. Calvert, B.A. Ridley,  F. Grahek, B. Heikes, G. Kok, J. Shelter,
J. Walega, CM. Elsworth, R.B. Norton, D.W. Fahey, P.C.  Murphey, C.  Hovermale, V.A. Mohnen, K.L.
Demerjian,  G.I. Mackay and H.I. Schiff.  Intercomparison of NO£ measurement techniques. In press
JGR-Atmospheres. 1989.

Galvin, P.J., V.A. Mohnen and U. Roychowdhury 1987.  Advanced mobile measurement laboratory for
mountain cloud chemistry research.  Proc. 6th Symposium  on Meteorological Observations and
Instrumentation, Jan. 12-16, New Orleans, LA.  AMS, Boston, MA.

Galvin, P. and V. Mohnen.  1987. Measurement of ozone and  other oxidants at mountain sites in the
eastern U.S., pp. 395-410.  In: Proceedings of the North American Oxidant Symposium, Feb. 25-27,  1987,
Quebec, Canada.

Gilliam, F.S. and.J.T. Sigmon. 1987. Relationships  between throughfall  chemistry and the chemical fluxes
in dry deposition and mountain clouds, pp. 63-65. In: Proceedings of the Sixth Symposium on
Meteorological Observations and Instrumentation, Jan.  12-16, New Orleans, LA. American
Meteorological Society. Boston, MA.

Gilliam, F.S., J.T. Sigmon, M.A Reiter, and D.O. Krovetz. In press. Elevational and spatial variation in
daytime ozone concentrations in the Virginia Blue Ridge mountains: implications for forest exposure.
Can. J. Forest Res.

Harrison,  E.A., B.M. Mclntyre, and R.D. Dueser.  1989. Community dynamics and topographic controls
on forest pattern in Shenandoah National Park, Virginia.  Bull, of Torrev  Bot. Club. 6:1-14.

Hertel, G.D., S.J. Zarnoch,  T. Arre, C. Eager, V. Mohnen and S.  Medlarz. Status of the Spruce-Fir
Cooperative Research Program. Presented at the 80th  Annual Meeting of APCA, New York, NY, June
21-26. 20 pp.

Hornig, J.F., C.J. High, and P.O. Thorne. 1988. Instrumentation for obtaining meteorological and
precipitation  information at multiple remote forest sites, pp.183-190.  In: G.D. Hertel [Tech. Coordin.]
Effects of Atmospheric Pollutants on the Spruce-Fir Forests of the Eastern United States and the
Federal Republic of Germany. Proceedings of the United States/Federal Republic of Germany Research
Symposium, Oct.  19-23, 1987, Burlington, VT. General Technical  Report NE 120, USDA-Forest Service,
Northeastern Forest Experiment Station, Broomall, PA 543  pp.
                                               D-2
-------
Joslin J. D., Mueller S. F. and Wolfe M. H. (1990)  The use of artificial and living collectors in the
testing of models of cloud water deposition to forest canopies (submitted to Atmospheric Environment).

Kavender, K.A. 1988. Measurement of a vertical ozone concentration profile in a slash pine forest.
Masters Thesis, Environmental Engineering Sciences Department, University of Florida, Gainesville, FL.
77pp.

Keene, W.C. and J.N. Galloway. 1988. Biogeochemical cycling of formic and acetic acids through the
troposphere: an overview of current understanding. Tellus. 40B:322-334.

Krovetz,  D.O., M.A. Reiter, and J.T. Sigmon.  1988. An inexpensive thermocouple probe-amplifier and its
response to rapid temperature fluctuations in  a mountain forest.  J. Atmos. Oceanic Technol.
5(6):870-874.

Krovetz,  D.O., M.A. Reiter, J.T. Sigmon, and  L.S. Gilliam. 1988. Assembly and field testing of a
ground-based presence of cloud detector. J. Atmos. Oceanic Technol. 5:579-581.

Krovetz,  D.O., J.T. Sigmon, M.A. Reiter, and  L.H. Lessard.  In press. An automated system for air
sampling with annular denuders at a remote site. Environ. Pollut.

Lefohn, A.S. and V.A. Mohnen 1986.  The Characterization of Ozone, Sulfur Dioxide and Nitrogen
Dioxide for Selected Monitoring Sites in the Federal Republic of Germany.  JAPCA. Vol. 36, No. 12,
1329-1337.

Lefohn, A.S., D.S. Shadwick and V.A Mohnen.  The characterization of ozone and sulfur dioxide
concentrations at a select set  of high-elevation sites  in the eastern United States. In Press
Environmental Pollution. 1990.

Lefohn, A.S. and V.A Mohnen.  1986. The characterization of ozone, sulfur dioxide and nitrogen dioxide
for selected monitoring sites in  the Federal Republic of Germany. JAPCA. 36:1329-1337.

Lefohn, A.S., V.C. Runeckles, S.V. Krupa, and D.S.  Shadwick. 1989. Important consideration for
establishing a secondary ozone standard to protect vegetation. JAPCA. 39(8): 1039-1045.

Lessard, L.H. and J.T. Sigmon. 1987. Measurements of concentrations of some reactive atmospheric gases
and fine primary particulates with annular  denuder atmospheric samplers, pp.60-62. In: Proceedings of
the Sixth Symposium on Meteorological Observations and Instrumentation, Jan.  12-16, New Orleans, LA.
American Meteorological Society. Boston, MA.

Lin, N.-H.  1988. Investigations on cloud chemistry and acidic deposition at Mt. Mitchell, N.C. Using a
cloud deposition model.  Master's Thesis. Department of Marine, Earth, and Atmospheric Sciences,
North Carolina State University. 149 pp.

Mallant, R., K.  Elsholz, and B. Bailey. 1989. Fog detection with a low-cost forward scattering optical
device. In: Proceedings of the Symposium on the Role of Clouds in Atmospheric Chemistry and Global
Climate, Jan. 29-Feb. 3,  1989, American Meteorological Society, Anaheim, CA.  pp. 221-223.

Markus, M., B.  Bailey, and R. Stewart. 1989. Estimation of cloud frequency at high elevation forests in
the eastern United States.  In: Proceedings of the Symposium on the Role of Clouds in Atmospheric
Chemistry and Global Climate, Jan. 29-Feb. 3, 1989, American Meteorological Society, Anaheim, CA.
pp. 110-113.
                                               D-3
-------
Markus, M. and B. Bailey. 1989. Cloud frequency determination at high elevations using an optical
detector. In: Proceedings of the Conference on Agricultural and Forest Meteorology, March 7-10, 1989,
American Meteorological Society, Charleston, SC. pp.35-37.

Mclntyre, B.M., M.A Scholl, and J.T. Sigmon. 1990. A quantitative description of a deciduous forest
canopy using a photographic technique.  Accepted by Forest Science.

Meagher J. R, Olszyna K. J., Weatherford F.  P. and Mohnen V. A. (1990)  The availability of H2O2 and
Oj for aqueous phase oxidation of SC«2:  the  question of linearity. Atmos. Environ, (in press).

Meagher, J.F., K.J. Olszyna, P.P. Weatherford and V.A  Mohnen. The availability of H2O2 and Oj for
aqueous phase oxidation of SO2--The question of linearity. Accepted for publication in Atmospheric
Environment.

Mohnen, V.A. 1987. Airborne and ground-based cloud collectors: an overview, pp. 44-46. In: Proceedings
of the Sixth Symposium on Meteorological Observations and  Instrumentation, Jan. 12-16, New Orleans,
LA American Meteorological Society. Boston, MA.

Mohnen, V.A. 1988. The mountain cloud chemistry program. In: G.D. Hertel [Tech. Coordin.] Effects of
Atmospheric Pollutants on the Spruce-Fir Forests of the Eastern United States and the Federal Republic
of Germany. Proceedings of the United States/Federal Republic of Germany Research Symposium, Oct.
19-23, 1987, Burlington,  VT. General Technical Report  NE 120, USDA-Forest Service, Northeast Forest
Experiment Station, Broomall, PA 543

Mohnen, V.A 1989.  Mountain Cloud Chemistry Project - Wet, Dry and Cloud Water Deposition,
EPA/600/53-89/009. U.S.EPA Atmospheric Research and Exposure Assessment Laboratory, RTP, N.C.
27711.  77 pp.

Mohnen, V.A. 1989.  Exposure of Forests to  Air Pollutants, Clouds, Precipitation, and Climatic
Variables.   EPA/60/53-89-003.  U.S. EPA, Atmospheric  Research and Exposure Assessment Laboratory,
RTP, N.C. 27711.  190 pp.

Mohnen, V.A, R.L. Bradow, D. Landsberg, J. Healey, and B.H. Bailey. 1987. Overview of the EPA
mountain cloud chemistry program, pp.47-50.  In: Proceedings of the Sixth Symposium on Meteorological
Observations and Instrumentation, Jan. 12-16, New Orleans, LA American Meteorological Society.
Boston, MA.

Mohnen, V.A, K.  Leonard, and B.H. Bailey.  1987. Cap  cloud frequency and chemistry at Whiteface
mountain, pp. 51-54. In: Proceedings of the Sixth Symposium on Meteorological Observations and
Instrumentation, Jan. 12-16,  1987. New Orleans, LA. American Meteorological Society.  Boston, MA.

Mohnen, V.A 1984 Project Director. EPA/NSF Workshop on atmospheric deposition and its impact on
high elevation mountain forest systems, Albany, NY, April 5-7.  ASRC/SUNY Pub. No. 981.

Mohnen, V.A. and J.A Kadlecek. Special Problems in  Atmospheric Exposure.  Part I - Interception;
Part II - Throughfall.  MCCP Background Report to Site  Directors, November. ASRC/SUNY Pub.
No.  1113.

Mohnen, V.A  Mountain Cloud Chemistry.   Presented  at the Symposium on Cloud Chemistry and Acid
Precipitation, 20 August (XIX General Assembly of the IUGG) Vancouver,  Canada.  Expand Abstracts
V.3, p. 855, No. Ml 1-9.
                                              D-4
-------
Mohnen, V.A. 1988 The Challenge of Acid Rain.  Scientific American. Vol. 259, No. 2, August, 30-38.

Mohnen, V.A.  Mountain Cloud Chemistry Project (MCCP).  Presented at European Fog Workshop,
University Frankfurt.  In:  Chemistry and Physics of Fogwater Collection.  W. Jaeschke and K.H. Enderle
(Eds.).  Proceedings, Workshop Frankfurt am Main, 16-17 December 1986. BPT-Bericht 6/88, 97-121.

Mohnen, V.A  Air pollutant distribution patterns:  elevational gradients/local chemistry.  Proceedings,
15th International  Meeting for Specialists in  Air Pollution Effects on Forest Ecosystems-Air Pollution
and Forest Decline; Interlaken, Switzerland, October 2-8.

Mohnen, V.A.  Exposure of  forests to air pollutants, clouds, precipitation and climatic variables.  A
preliminary assessment-1987.  Annual report submitted to EPA under contract no. CR813934-01-2.
September.

Mohnen, V.A.  Quality assurance and quality control procedures for air quality measurements from
airborne platforms. Paper presented at VDI-Workshop "Fluggestutzte Messungen von
Luftverunreinigungen", Vol. 9,  157-166, VDI-Kommission Dusseldorf, Germany.  (Workshop on October
13-14, 1988, Trier).

Mohnen, V.A 1989 Elevational gradients/local  chemistry.  In: Biologic Markers of Air-Pollution Stress
and Damage in Forests.  National Academy Press. Washington, DC, 47-56 (Reviewed).

Mohnen, V.A and J.A Kadlecek.  Cloud Chemistry Research at Whiteface Mountain.  Tellus. 418:79-91.

Mohnen, V.A  Acid rain in  perspective.  The Nelson A Rockefeller Institute of Government Working
Paper Series (Spring, 1989).

Mohnen, V.A  Mountain  Cloud Chemistry Project Wet, Dry and Cloud Water Deposition.
EPA/600/3-89/009.

Mohnen, V.A and AS. Lefohn. Temporal development of the pollution load of reactive trace gases in
forested areas in the United States. Invited paper presented  at the International Congress on Forest
Decline Research:  State of Knowledge and Perspectives, Friedrichshafen, Lake Constance, Federal
Republic of Germany, October 2-6. Published  in Proceedings, 1990.

Mohnen, V.A  Acid Rain. Paper presented  at the International Power Technology Conference &
Exhibition, Chicago, IL, October 31, 1989.

Mohnen, V.A  Acid rain and  urban atmospheric pollution in North America.   Presented at The Royal
Institute of International Affairs, Chatham House Conference, Fourth International  Energy Conference
"Environmental Challenges: The Energy Response, London, December 4-5, 1989.

Mueller S. F. (1990)  Estimating cloud water deposition to subalpine spruce-fir forests  - I:  modifications
to an existing model (accepted by Atmospheric Environment).

Mueller S. F. and  Imhoff R. E. (1989)  Inferring cloud deposition to a forest canopy using  a passive
cloudwater collector.  Geophvs. Res. Let. 16:683-686.

Mueller S. F. and  Weatherford F. P. (1988)  Chemical deposition to a high elevation red spruce forest.
Water. Air, and Soil Pollution 38:345-363.
                                               D-5
-------
Mueller S. R, Joslin J. D. and Wolfe M. H. (1990) Estimating cloud water deposition to subalpine
spruce-fir forests - II:  model testing (accepted by Atmospheric Environment).

Mueller, S.R and RP. Weatherford. 1988. Chemical deposition to a high elevation red spruce forest.
Water. Air. Soil Pollut. 38:345-363.

Murphy, C.E. and J.T. Sigmon. In press. Dry deposition of sulfur and nitrogen oxide gases to forest
vegetation. In: Advances in  Environmental Sciences. Springer-Verlag:New York.

Murthy, A. and Aneja, V.P., 1990.  "Deposition and Interaction of nitrogen containing pollutants to a
high elevation forest canopy", in preparation for submission to Atmospheric Environment.

Olszyna K. J., Meagher J. R and Bailey E. M. (1988)  Gas-phase, cloud and rain-water measurements of
hydrogen peroxide at a high-elevation site.  Atmos. Environ. 22(8):1699-1706.

Reisinger L. M. (1989)  Particles sampled in the  southern Appalachian Mountains.  Water. Air, and Soil
Pollut. (in press).

Reisinger L. M. and Imhoff R. E. (1989) Analysis of summertime cloud water measurements made in a
southern Appalachian spruce forest.  Water. Air, and Soil Pollut. 45:1-15.

Reisinger L. M., Olszyna K. J. and Hetrick T. L.  (1989)  Comparison of enhanced and routine methods
for measuring ambient low-level sulfur  dioxide. JAPCA. 39:981-983.

Robarge, W.P., R.I. Bruck,  and E.B. Cowling. 1988. Throughfall and stemflow measurements at Mt.
Mitchell, North Carolina during the summer of 1986: a preliminary report, pp. 111-116. In: G.D. Hertel
[Tech. Coordin.]  Effects of Atmospheric Pollutants on the Spruce-Fir Forests of the Eastern United
States and the Federal Republic of Germany. Proceedings of the United States/Federal Republic of
Germany Research Symposium, Oct. 19-23, 1987, Burlington, VT.  General Technical Report NE 120,
USDA-Forest Service, Northeastern Forest Experiment Station, Broomall, PA 543 pp.

Saxena, V.K. 1987. Mountain cloud chemistry project at Mt. Mitchell, NC: strategies and highlights.
Trans. Amer. Geophys. Union (EOS). 68:270.

Saxena, V.K., E.B. Cowling, R.E. Stogner, and R.V. Crum. 1986. 1986:Cloud  chemistry measurements at
Mt. Mitchell, North Carolina, pp. 91-94. In: Preprints for the  Cloud Physics Conference, American
Meteorological Society, Boston, MA.

Saxena, V.K. and N.-H.  Lin. 1988. Cloud capture by a mountain top forest: acidity and dosage duration.
In: Preprint for the 10th International  Cloud Physics Conference, Aug. 15-20, 1988, Bad Homburg,
Federal Republic of Germany. 1:232-234.  Available from: Deutscher Wetterdienst, Frankfurter Strasse
135, D06050 Offenbach  Am Main, FRG.

Saxena, V.K. and N.-H.  Lin. 1988. Relative importance of dry, wet and cloud capture mechanisms for
acidic deposition, pp. 237-247. In: S. Hocheiser and R.K.N.  Jayenty, eds. Measurements of Toxic and
Related Air Pollutants, EPA/Air Pollution Control Association, Symposium, May 1-4, Raleigh, NC.

Saxena, V.K. and N.-H.  Lin. 1989. Contribution of acidic deposition on high  elevation forest canopy to
hydrologic cycle, pp. 193-202.  In: Proceedings of  the International  Association Hydrological Science, 3rd.
Science Assembly, May  10-19, Baltimore, MD.
                                               D-6
-------
Saxena, V.K., N.-H.Lin, and T.P.Defelice. In Press. Hydrological and chemical input to fir trees from rain
and clouds during a 1-month study at Clingmans Peak, NC.  Atmospheric Environment. 23:, 5pp., 1990.

Saxena, V.K. and R.E. Stogner. 1987. Wet deposition on forest canopy at Mt. Mitchell, North Carolina,
pp. 189-194. In: Measurements of Toxic and Related Air Pollutants, Air Pollution Control Association,
Pittsburgh, PA.

Saxena, V.K., R.E. Stogner, AH. Hendler, T.P. DeFelice, and R.J.-Y. Yeh.  1989. Monitoring the
chemical climate of the Mt. Mitchell State Park for evaluating its impact on forest decline. Tellus.
418:92-109.

Saxena, V.K. and R.J.-Y. Yeh. 1989. Acidic cloud immersion: a possible cause of forest decline.  In: Fifth
World Meteorological Organization Scientific Conference on Weather Modification and Applied  Cloud
Physics, March 27-31, 1989, Guangzhou, China.

Saxena, V.K. and R.J.-Y Yeh. 1989. Temporal variability in cloud water acidity: physio-chemical
characteristics  of atmospheric aerosols and windfield. J. Aerosol Sci. 19:1207-1210.

Sigmon, J.T. and M.J. Estes. 1988.  Relationship between  cloud chemistry  and meteorology in central
Virginia: A preliminary study. Preprints of the 81st Annual Meeting of Air Pollution Control
Association.

Sigmon, J.T., F.S. Gilliam, and M.E. Partin.  In press. Precipitation and throughfall chemistry for  a
montane  hardwood forest ecosystem: potential contributions  from cloud water. Can. J. Forest Res.

Sjostedt, D.W. 1987. A characterization of the nocturnal low-level jet in the Carolinas. Masters Thesis.
Department of Environmental Sciences. University of Virginia.

Smalley, D.J. and.B.H. Bailey. 1987.  Meteorological/climate aspects of mountain cloud chemistry
monitoring. In:  Proceedings of the Sixth Symposium on Meteorological  Observations and
Instrumentation, Jan.  12-16, New Orleans, LA American Meteorological Society, Boston, MA pp.
47-50.

Stogner, R.E.  1989. Acidic deposition to Mt. Mitchell, North  Carolina area forests resulting  from direct
cloud droplet interception.  Master's Thesis. Department of Marine, Earth,  and Atmospheric Science.
North Carolina State  University. 124 pp.

Stogner, R.E. and V.K. Saxena. 1988. Acidic cloud-forest canopy interactions: Mt. Mitchell, NC. Environ.
Pollut. 53:456-458.

Thorne, P.O., Lovett,  G.M. and Reiners, W.A (1982).  Experimental determination of droplet impaction
on canopy components of balsam fir.  Journal of Applied Meteorology 21, 1413-1416.

Valente R. J.,  Mallant R.  K. A, McLaren S. E. and Schemenauer R. S. (1989)  Field intercomparison of
ground-based cloud physics instruments  at Whitetop Mountain, Virginia.  J. Atmos. and Oceanic Tech.
6:396-406.

Vong, R.J., Bailey, B.R., Markus, M.J. and Mohnen, V.A (1989).  Factors governing cloud water
composition in the Appalachian Mountains  (submitted to Tellus BV
                                               D-7
-------
Vong, R.J., B.H. Bailey, M.J. Markus and V.A Mohnen.  Variation in cloud water chemistry with
synoptic type at five mountain sites in the eastern USA. EOS, Vol.  70(43):1007 (October 24,  1989),
Abstract #A11B-13 1130H. Accepted for publication in Tellus (1990).

Yeh, R.J.-Y.  1988. Measurements of cloud water acidity and windfield for evaluating cloud-canopy
interactions in Mt. Mitchell state park. Master's Thesis. Department of Marine, Earth, and
Atmospheric  Science, North Carolina State University.  165 pp.

Yeh, R.J.-Y.  and V.K. Saxena. 1988. Sulfur and nitrogen emissions along the path of the airmass and
cloud water acidity at Mt. Mitchell, North Carolina, pp. 15. In: 1988 Annual meeting of the American
Association for Aerosol Research, Oct. 10-14, 1988, Chapel Hill, NC.
                                                D-8
-------
    APPENDIX E




ADDITIONAL FIGURES
-------
Average Annual Low «2500 m)  Cloud  Amounts  (%)  in
Eastern North America,  For Land  Areas  Only,  1971 -
1981, (from Warren et al,, 1986).   The Time-Averaged
Cloud Amount is the Product of Frequency  of  Occurrence
and Amount-When-Present.  Values  Represent Stratus/
Stratocumulus, Cumulus  and Cumulonimbus Cloud Types.
                  20-29%
30-397.
40-497.
                   Figure 4-9
                         E-l
-------
Average Cloud  Base  Height  (m) in Eastern
North America  for Stratus/Stratocumulus
Clouds, the Dominant  Low Cloud Type, For
Land Areas Only, 1971-81 (From Warren et
al., 1986)
             600-690 m   700-790 m
800-890 m
               Figure 4-10
                     E-2
-------
._ oS CD
.5? CD CO
sz w CD
O CO
03 JQ CO
— -i— •
0. § Q
CL o
< o •§
• -- o
O Q
jC ^ I
*" W 0-
i- CO LU
LL 2
r~
c
r^ \~Z  "0 O 0 O O























1

o o o o o o o
o
o


00
o


1^"*
o
>>
•1— »
CD —

J3
0

0

CD
' d 15
CD
DC
CO
o
CM
O
O
O
d
^"-^•F oooooooooo
~C*= OCX)CD^CMOOOCD-«i-CM
CO.O-Q CMT-T-T-T-T-
CD O) 3
DC OJ 0
OC o £

 g>
"CD
X
       E-3
-------
        Probability of Cloud by Height and Season
            at Moosilauke, NH, (1985-1987)
     All
-•- DJF   -•- MAM  -o- JJA    -a- SON
30%
25%
20%
15%
10%
o
1
'c
/
to
/
'o
-------
        Probability of Cloud by Height and Season
             at Shenendoah, W, (1985-1987)
  — All     ••-  DJF   ••- MAM  -o- JJA    -a-  5ON
30%
25%
20%
15%
10%
 5%
 0%
         f
                           CCQ
                  m*
CCCCCD-
     0
                                  «!•€•••
                                           '«•%•««•
+
         500        1000       1500
               HEIGHT (m)
         2000
                      Figure 4-14

                          E-5
-------
   Probability of Cloud by Height and Season
   at Whiteface Mountain, NY, (1985-1987)
— All ••-
30%
25%
20%
15%
10%
5%
0%

DJF -•- MAM -o- JJA -n- SON





•

Ocoo
jjpcc
3
EEccccccccca 	
$ ?•.
	 •£ J ' 'I " j) ' ' '* 	
/ ** Var * I^SS^
	 rf^ •V^^\l--nrtDI%--c
/*// •' « j
^ 1 S
•ft \
«r i
H 	 	 	 •. 	 	 	
i 	 j 	 i


g 	
SS^:
	 "s!
1 	 ,
'
5
"i
0
500       1000       1500
      HEIGHT  (m)
2000
                 Figure 4-15

                     E-6
-------
Low-level Cloud Frequency Departure (%) From
    1965-85 Normals For 8 NWS Airports
           986 Field Season
           Figure 4-17
                   E-7
-------
Low-level Cloud Frequency Departure (%) From
    1965-85 Normals For 8 NWS Airports
           1987 Field Season
           Figure  4-18
                   E-8
-------
Low-level Cloud Frequency Departure (%) From
    1965-85 Normals For 8 NWS Airports
             988 Field Season
           Figure 4-19
                   E-9
-------
2
DC
   co
    •
   to
   CO
       CO
       cc
s
u_  ^ Q

1  < O
Zi  O o
O  2 CO
O  £ DC


O  < LL

LU
O
cc
               o
               LU
               CC
ID
g

o
                    0^
                    o
                    o
                          CD
                   \
                         X
                              x
                          \
                                              \
                                     \
               LU

               LU
                       O    O   O   O    O   o    O
                       O    O   O   O    O   O    O
                       •^t    CO   CM   T-    O   C35    CO
                                                                          O
                                                            03

                                                           "o



                                                           oo
                                                            II
                                                           •^

                                                            e
                                                            e

                                                            C/5

                                                            0)
                                                           x:


                                                           to



                                                            0)

                                                            cr
                                                            CD
                                                                           c
                                                                           o
                                                                           03
                                                                           Q.

                                                                           £

                                 E-10
-------
o

LU
z>
O     O5
LU     DC
OC     D
"-  So O
Z  co I
Q  CD Q
h-  °0 Z)
92Q
LJJ  •== LJ

>  Q- Z
£00

^  LU <
2  I- LU
2  = 05
2
       0
               O
               LU
               cr
               LL
               O

               h-

               >
               LU

               LU
                         CO
                                   X
                                        \
                                              \
                                                   N
                       CD   If)
                                       O   O


                                       CO   CJ
 co


 o
s^

"CO
                                                                         CO
                                                                         OJ
                                                                          E
                                                                          E

                                                                          CO

                                                                          0)
                                                                         .c


                                                                         "S
                                                                          o
                                                                          c
                                                                          cr
                                                                          Q)
 C
 O

"o
 C3
 Q.

 E
                                                                         _0
                                                                          o
                                                                          o

                                   E-ll
-------
         s
         S
         I"
         S  -'
,----7^
Figure 4-23  Cloud droplet size  distributions from Whiteface Mountain, NY
          calculated by integrating over individual size distributions
          and grouping the distributions into three categories by liquid
          water content (LWC  < 0.5 g/m3, 0.5 < LWC < 1.0 g/m3, and LWC >
          1.0 g/m3), and clear air (Mohnen, 1987).
                                E-12
-------
DROUGHT SEVERITY ( Long Term Palmer) OVER THE EASTERN U.S.
                    MID-SUMMER 1986
   MOIST
   NORMAL
   MODERATE
   SEVERE
   EXTREME
                                                Date of Analysis: July 26
Source: NOAA Weekly Weather and Crop Bulletin, July 29, 1986.
                       Figure 4-25
                          E-13
-------
DROUGHT SEVERITY ( Long Term Palmer) OVER THE EASTERN U.S.
                   MID-SUMMER 1987
     MOIST
     NORMAL
     MODERATE
     SEVERE
     EXTREME
                                                Date of Analysis: August 1
Source: NOAA Weekly Weather and Crop Bulletin, August 4, 1987.
                        Figure 4-26
                             E-14
-------
DROUGHT SEVERITY ( Long Term Palmer) OVER THE EASTERN U.S.
                   MID-SUMMER 1988
    MOIST
    NORMAL
    MODERATE
    SEVERE
    EXTREME
                                                 Date of Analysis: July 30
Source: NOAA Weekly Weather and Crop Bulletin, August 2, 1988.
                       Figure 4-27
                            E-15
-------
      Midwest
                                          Forested Areas

                                          Q 0%1o20%
                                          EH 21% 10 40%
                                            41% to 60%
                                            61% to 60%
                                            8t%to 100%
                                       South
                                                  Mid-Atlantic
                                                   Forested Areas

                                                  D 0% lo 24%
                                                  §3 25% to 49%
                                                  E3 50% lo 74%
                                                  • 75% to 100%
Source:  A.S.L.  & Associates
          Helena, MT

Figure 4-29  Percent forest  cover  by county (Regions:   Midwest,  Northeast,
           South, and Mid-Atlantic) .
                                       E-16
-------
Pacific  Northwest
Rocky Mountain
                West
                                                           Forested Areas

                                                          D 0%to20%
                                                          EJ 21% to 40%
                                                          El 41% to 60%
                                                            61% to 60%
                                                            61% to 100%
  Source:  A.S.L. & Associates, Helena/ MT
     Figure 4-30 Percent forest cover by county (Regions:
               Rocky Mountain, and West).
              Pacific Northwest,
                                        E-17
-------
Source:  EPA (1989)
             0.30
                  CONCENTRATION. PPM
              0.05-
              0.00
                      1978  1979 1980 1981 1982  1983  1984  1985  1986 1987
        Boxplot comparisons  of trends in annual second highest  daily maximum
        1-hour ozone concentration  at 274 sites,  1978-1987.
             0.18

             0.16-

             0.14

             0.12

             0.10

             O.OB

             0.06

             0.04

             0.02

             0.00
                  CONCENTRATION, PPM
• NAMS SfTIS (98)    • ALL_S(TES_(274)_
                      1978  1979  1980  1981  1982 1983  1984  1985  1986 1987
        National  trend in the composite average of the second  highest  maximum
        1-hour ozone concentration at both NAMS and all  sites with 95 percent
        confidence  intervals,  1978-1987.
                                      Fiuure 4-31
                                       E-18
-------
Source:   EPA (1989)

                CONCENTRATION, PPM
           0.040
           0.013-


           0.030


           0.023


           0.020


           0.015


           O.OtO


           0.003


           0.000
                                                               547 SUES
••——-NAAOS-
                   1978  1979  1980  1981  1982  1983 1984 1985  1986 1987

       Boxplot  comparisons  of  trends  in  annual  mean  sulfur  dioxide
       concentrations at 347 sites,  1978-1987.
           O.OJ5
           o.oso
            0.025-
            0.020-
            0.013-
            0.010-
            0.003-
                CONCENTRATJON, PPM
            0.000
                  •NAAQS'
   • NAMS STCS (105)     « AajyjE5_(34
                    —i	i	1	1	1	1	—i	1	1	1—
                    1978   1979  19BO  1981  1982  1983  1984  1985  1986  1987
      National trend in the composite average of the  annual average  sulfur
      dioxide concentration  at  both NAMS  and  all  sites  with 95  percent
      confidence  intervals, 1978-1987.

                                     Figure 4-32
                                         E-19
-------
Source:  EPA (1989)
     0.07
          CONCENTRATION, PPM
     0.06-


     0.05-


     0.04-


     0.03-


     0.02-


     0.01-
     0.00
                                                      84 SITES
                                      	NAAQS
             1978 1979 1980 1981 1982 1983  1984  1985  1986 1987
Boxplot  comparisons  of  trends  in  annual  mean  nitrogen  dioxide
concentrations  at 84  sites,  1978-1987.
      0.06
          CONCENTRATION, PPM
      0.05-
      0.04-
      0.03-
      0.02-
      0.01 -
      0.00
                                •NAAQS'
                              i	1	J
NAMS SfTES (U)    o ALLSniSj8_4)_
              1978 1979 1980 1981 1982 1983  1984  1985  1986 1987
 National   trend  in   the   composite   average  of  nitrogen  dioxide
 concentration at both  NAMS  and all  sites with 95 percent confidence
 intervals,  .1978-1987.
                              Figure 4-33
                                 E-20
-------
Source: EPA (1989)
     c
     o
     X
     O
     z
         35
         30-
         25-
     I   20"
     1

     I   ,,H
10-
          5-
           1979
                             g:;:;| Transportation

                                 PU*' Combustion

                                 Industrial Processes

                                 Solid Waste and Misc.
                              888
                              ilijg

                              iH

           - ^iyffisfr^^
         1980
1981    1982    1983    1984
1985
1986
               National Trend in Nitrogen Oxide  Emissions,  1979-1986.
      c
      o
      O
      O
         35-
         30-
                                                            1986
         National Trend  in Volatile Organic Compound Emissions,  1979-1986.


                                Figure 4-34
                                        E-21
-------
Source: EPA (1989)
      SOM EMISSIONS, K)1 METRIC TONS/YEAR
    1978   1979   1980   1981   1982   1983   1984   1985  1986   1987
         National trend in sulfur oxide emissions, 1978-1987.
                            Figure 4-35
                                E-22
-------
                                                                           CD
                                                                          4-1
                                                                  (w       03
                                                                  0)  -O  rH
                                                                  x;   c   3
                                                                  00  CO   O
                                                                       ^
                                                                       0)
                                                                      J3
                                                                       60
                                                                      •H
                                                                      x;

                                                                       cu
                                                                  CO  -U
                                                                  OJ
                                                                  en  x:
                                                                  3  4-1
                                                                  o  o
                                                                  o  xi
                                                                  M-l
                                                                      C
                                                                  to o
                                                                  c
                                                                  •H  03
                                                                  jj  OJ
                                                                  x:  03
                                                                  oo 3
                                                                  •H  U
                                                                  a)  o
                                                                  5  H-I

                                                                  ^  60
                                                                  +  C

                                                                      U
                                                                  u-i  xi
                                                                  ON  60
                                                                  3  -H
                                                                      0)
                                                                  0)  3
                                                                  A
                                                                  H  /-N
                                                                  03  vO
                                                                  C  CM
                                                                  O  —i
                                                                  •H  5
                                                                  4J
                                                                  O  0)
                                                                  C  J3
                                                                  3  1-1
                                                                  tM
                                                                      0)
                                                                  00 i-l
                                                                  C  -H
                                                                  •H  XI

                                                                  ^  *
                                                                  00   «
                                                                  •H  Cfl
                                                                  (U  C
                                                                  3  O
                                                                      •H
                                                                  iH  4J
                                                                  nj  aj
                                                                  •a  t-i
                                                                  •H  4-1
                                                                  o  c
                                                                  G  o
                                                                  00 O
                                                                  •H  c
                                                                  U)  O
                                                                      o
                                                                  o
                                                                  3  a)
                                                                  •u  oo
                                                                      a)
                                                                  U-J  |_|
                                                                  O  tt)
                                                                      >
                                                                  CO  ctJ
                                                                  (II

                                                                  O.r-1
                                                                  B  M
                                                                  c^  3
                                                                  X  O
                                                                  PJ  J3
 0)
 >,
1—(
 to

 ta

 en
-H
.C
•r-l

T3
 QJ
 CO
 3

 en
 n)


 oo
 C
•H
4J

 60
•H
 0)


x£>
CM
 01   X
A   •>  «J t-<
•H  T3  0)
 I-l  -H J
 3   O
 O   B
A   60  ••
    •H  CU
 M   0)  O
 QJ       l-i
 3   QJ  3
 O  J3  O
i-H  4-1 t/3
E-23
-------
2
Ed
             ROWLAND  FOREST  -- APR 16 -  OCT  15
       FREQUENCY OF CONSECUTIVE  HOURS OF OZONE  GREATER
          THAN  70  PPB  BETWEEN  THE  HOURS 7AM  AND 6PM
45 -
40 -
35 -
30 -
25 -
20 -
15 -
10 -
.5 -
 0
                                       HORIZONTAL BAR: 1986
                                            HASH BAR: 1987
                                            SOLID BAR: 1988
                         3456
                      CONSECUTIVE HOURS
                           Figure 4-37
          MT. MOOSILAUKE SUMMIT --  APR  16  - OCT  15
        FREQUENCY OF CONSECUTIVE HOURS OF OZONE  GREATER
           THAN 70 PPB  BETWEEN THE  HOURS 7AM  AND 6PM


>-
z.
Ed

or
K
U,



ou -
45 -
40 -
35 -
30 -
-
25 -
-
20 -
15 -
10 -
5 -
—
HORIZONTAL BAR: 1986
HASH BAR: 1987
SOLID BAR: 1988








....
1 1 I [ 1 1 1
                          3456
                       CONSECUTIVE  HOURS
                            Figure 4-38
                                                 > =
                              E-24
-------
          WHITEFACE  MT.  SUMMIT -- APR  16 -  OCT  15
       FREQUENCY OF  CONSECUTIVE HOURS OF OZONE  GREATER
          THAN  70  PPB BETWEEN THE  HOURS 7AM AND 6PM
      50
      45 -
      40 -
      35 -
      30 -
      25 -
      20 -
      15 -
      10 -
       5 -
       0
                                  HORIZONTAL BAR:  1986
                                       HASH BAR:  1987
                                       SOLID BAR:  1988
                         3456
                      CONSECUTIVE HOURS
                            Figure 4-39
w
Gf
a
Cti
b.
         WHITEFACE MT.  SUBSITE  3 --  APR  16 -  OCT 15
        FREQUENCY  OF  CONSECUTIVE  HOURS OF  OZONE GREATER
          THAN 70 PPB BETWEEN THE  HOURS 7AM AND 6PM
      50
45 -
40 -
35 -
30 -
25 -
20 -
15 -
10 -
 5 -
 0
                                        HORIZONTAL  BAR: 1986
                                              HASH  BAR: 1987
                                             SOLID  BAR: 1988
                         3456
                       CONSECUTIVE  HOURS
                            Figure 4-40
                                                 >= 8
                              E-25
-------
        WHITEFACE  MT. SUBSITE 4  --  APR 16  - OCT 15
       FREQUENCY OF CONSECUTIVE HOURS OF OZONE  GREATER
          THAN 70 PPB  BETWEEN THE HOURS 7AM  AND 6PM



0
3
or
w
QJ
fa



JU —
45 -
40 -
35 -
30 -
25 -
-
20 -

15 -
10 -
5 -


HORIZONTAL BAR: 1986
HASH BAR: 1987
SOLID BAR: 1988







^^
1 1 1
IfiliiliJBallillllll
1 1 1 1 1 1 I
                         3456
                       CONSECUTIVE HOURS
                            Figure 4-41
                                            7    > =
w
Crf
fa
              HUNTINGTON,  NY  -- APR 16  - OCT  15
        FREQUENCY  OF CONSECUTIVE  HOURS  OF OZONE  GREATER
           THAN 70 PPB  BETWEEN  THE  HOURS  7AM AND 6PM
      50
45 -
40 -
35 -
30 -
25 -
20 -
15 -
10 -
 5 -
 0
                                         HORIZONTAL BAR:  1986
                                              HASH BAR:  1987
                                              SOLID BAR:  1988
                          345
                       CONSECUTIVE HOURS
                            Figure 4-42
                              E-26
-------
         HAMPSHIRE COUNTY, MA --  APR  16 -  OCT 15
       FREQUENCY OF CONSECUTIVE HOURS  OF OZONE GREATER
          THAN 70 PPB  BETWEEN  THE HOURS  7AM AND  6PM
uu —
45 -
40 -
35 -
30 ^
25 -
20 -
15 -
10 -
5 -

HORIZONTAL BAR: 1986
HASH BAR: 1987
SOLID BAR: 1988





•
•1 11 J 11 KB! 1 Hi 1






II 1 1
                         3456
                      CONSECUTIVE  HOURS
                           Figure 4-43
o-
            BEAVER COUNTY, PA  --  APR  16  - OCT 15
        FREQUENCY  OF  CONSECUTIVE HOURS OF OZONE GREATER
          THAN 70  PPB BETWEEN THE  HOURS 7AM  AND 6PM
                                        HORIZONTAL  BAR: 1986
                                             HASH  BAR: 1987
                                             SOLID  BAR: 1988
                         3     4      5
                       CONSECUTIVE  HOURS

                           Figure 4-44

                             E-27
-------
O"
w
Di
            SHENANDOAH  SUMMIT --  APR  16 -  OCT  15
        FREQUENCY OF CONSECUTIVE HOURS OF  OZONE GREATER
           THAN 70 PPB  BETWEEN THE  HOURS  7AM  AND  6PM
      50
45 -
40 -
35 -
30 -
25 -
20
15 -
10 -
 5 -
 0
                                         HORIZONTAL BAR:  1986
                                               HASH BAR:  1987
                                              SOLID BAR:  1988
                     I
                       1       I
                          3456
                        CONSECUTIVE  HOURS
                             Figure 4-45
           SHENANDOAH  SUBSITE  2 -- APR 16  - OCT  15
        FREQUENCY  OF CONSECUTIVE HOURS  OF OZONE  GREATER
           THAN 70 PPB  BETWEEN  THE HOURS 7AM  AND 6PM
-------
o
z
w
^
Of
w
          SHENANDOAH  SUBSITE  3  --  APR  16  -  OCT 15
        FREQUENCY  OF CONSECUTIVE  HOURS  OF  OZONE GREATER
          THAN 70 PPB  BETWEEN  THE  HOURS  7AM AND  6PM
      50
45 -
40 -
35 -
30 -
25 -
20 -
15 -
10 -
 5 -
 0
                                        HORIZONTAL BAR:  1986
                                              HASH BAR:  1987
                                             SOLID BAR:  1988
                    I     I     I     I
1
                          3456
                       CONSECUTIVE  HOURS
                            Figure 4-47
1
2
W
P
Of
w
              BIG  MEADOW, VA  -- APR 16  - OCT  15
        FREQUENCY OF CONSECUTIVE HOURS  OF OZONE  GREATER
           THAN 70 PPB  BETWEEN THE HOURS 7AM AND 6PM
                                         HORIZONTAL BAR: 1986
                                              HASH BAR: 1987
                                              SOLID BAR: 1988
                          3456
                       CONSECUTIVE HOURS
                            Figure 4-48
                              E-29
-------
H
D
O*
W
OS
              SAWMILL  RUN,  VA -- APR  16  -  OCT  15
        FREQUENCY  OF CONSECUTIVE HOURS  OF OZONE GREATER
           THAN  70 PPB BETWEEN THE  HOURS  7AM  AND  6PM
                                          HORIZONTAL BAR: 1986
                                                HASH BAR: 1987
                                               SOLID BAR: 1988
                           3456

                        CONSECUTIVE  HOURS

                              Figure 4-49
 CJ>
 2
 w
 Cti
 b.
     DICKEY  RIDGE  WARREN  COUNTY,  VA -- APR  16  -  OCT  15
         FREQUENCY OF  CONSECUTIVE HOURS OF OZONE GREATER
            THAN 70  PPB BETWEEN  THE  HOURS 7AM AND  6PM
                                           HORIZONTAL BAR:  1986
                                                 HASH BAR:  1987
                                                SOLID BAR:  1988
                           3456
                         CONSECUTIVE  HOURS

                              Figure 4-50
                                E-30
-------
2
Cd
S
Of
W
Qi
           WHITETOP  MT.  SUMMIT -- APR 16  - OCT 15
        FREQUENCY OF  CONSECUTIVE  HOURS  OF OZONE GREATER
           THAN  70  PPB BETWEEN  THE HOURS 7AM  AND 6PM
      50
45 -
40 -
35 -
30 -
25 -
20 -
15 -
10
 5 H
 0
                                  HORIZONTAL BAR: 1986
                                        HASH BAR: 1987
                                        SOLID BAR: 1988
                          3456
                        CONSECUTIVE HOURS
                             Figure 4-51
2
Cd
w
a:
              GILES COUNTY,  TN  --  APR  16  -  OCT  15
        FREQUENCY  OF CONSECUTIVE HOURS OF OZONE GREATER
           THAN  70 PPB  BETWEEN  THE HOURS 7AM AND  6PM
                                          HORIZONTAL  BAR. 1986
                                               HASH  BAR: 1987
                                               SOLID  BAR: 1988
                           3456
                        CONSECUTIVE HOURS
                             Figure 4-52
                                E-31
-------
        MARION  SMYTH  COUNTY, VA  --  APR  16 -  OCT  15
       FREQUENCY OF CONSECUTIVE  HOURS  OF OZONE  GREATER
          THAN 70  PPB BETWEEN THE HOURS 7AM  AND 6PM
      50
      45 -
      40 -
      35 -
      30 -
      9=1 -
                                   HORIZONTAL  BAR:  1986
                                        HASH  BAR:  1987
                                        SOLID  BAR:  1988
                          3456
                        CONSECUTIVE HOURS
                             Figure 4-53
w
ft!
UH
           MT.  MITCHELL  SUMMIT  -- APR  16  -  OCT  15
        FREQUENCY  OF  CONSECUTIVE HOURS OF OZONE  GREATER
           THAN  70 PPB  BETWEEN THE  HOURS  7AM AND 6PM
      50
45 -
40 -
35 -
30 -
25 -
20 -
15 -
10 -
 5
 0
                                         HORIZONTAL  BAR: 1986
                                               HASH  BAR: 1987
                                               SOLID  BAR: 1988
                           3456
                        CONSECUTIVE  HOURS
                              Figure 4-54
                                E-32
-------
                                                                              4
Figure 4-55:  Ozone exposure - total season, Apr 15 - Oct 15, (Daylight hours 7AM - 6PM)
                                       E-33
-------
CO
w
CO
o
o
CO

CO
                                w
                                E-
                                h—1
                                00
                                                               H
                                                               z
                                                               o
                                      o <: o
                                      K W co

                                      I  I  I

                                      WWW
                                      CO CO CO
                                      000
                                      Q Q Q
                                      o o o
                                      CO CO CO
                                      CO CO CO

                                      CD r- co
                                      00 00 CO
                                      05 05 05
                                               vD
                                                I
                                                                        r-.

                                                                        cn

                                                                        3
                                                                        O
 60
.0
O
4-1
O
O
 I
         O
         m
o
CO
 c
 o
 tn
 cs
 QJ
 tn
                                                                        0)
                                                                        1-1
                                                                        p
                                                                        w
                                                                        o
                                                                        ex
OJ
c
o
N
O


3
             (jq-uidd)  3SOQ 1VNOSV3S
                                                                        E
                                    E-34
-------
o
o






W
t t
GO
£S
« 05
O |
1— 1
Sg
s!
2
W (72
*§
o oo
0 
-------





CO
CO
CD
V./r/
T
o

n~\
C*J
r^-
O)
T —




LL
c
o
f f
-3

rzz


• —
O

CO
o
>« /
^rj

LL
^>
^>






c
c

. 1
1

	 4

,- J
1

, 1
1 T

. _L
•~ r

4
r




, 	 .

.. j
i

, 4.
1

i- 4-
1

i 4
\










i- 4
T
1 1 1 1 1 1
3 O O O O O- O
D -«t 
                                             CD
(Aqdd) uojjBJiueouoo €O
               E-36
-------




00
CO

1
1
E-;
S
•^
CO


^
ACE MOUNTA]
Ct,
w
c_
w
>— <



o





r-
0}
CO 1
» f-
0 J °
H CO
« " M
CL
£* z; o
2 * D
SO2 CONCE
FROM 199O S>
MAY-87 THRO
1
"^
*^

















i

vvv
\A
N^

^W

^^^^^^^^^S
• | . | , , , | , | , , , | , | , | .
CO
_°°
_co
*
w >
_w 0 -
w Q •
H
O <"
«
-2 S:
_ 2N ^ CO
— * <™ -7 VO
1 ^ M . 
-------





co
05

1
1
CO
SUBSITE

E-^
*g
**-*

W
-------





CO
«— <
|

1
E-;
S
C/Q
Z
f-
O
S
ex,
o
9~*
E-
1



O





.n-'-'N
r-
03
 0
CENTRATl
SAMPLE
ROUGH 3<
2 o as
0 ^ H
° O
CM 03
O ^ I
M 2 i
Q* ^^
u. S
I
















£
L
K
K

^

^^^N

^^^^^^^^^N
. i . i . i . i i , . i . , . i . , i
CO
_CO
CO
w
rf-
W
u
W Q *
"W o^
H
O <"
~(M
_00
i-t
_<° 
-------





*t^
r*-
oo
O5
i— <
1 W
W
1 W D
^ * J
E-1 o <
>"— < - - *^
S3 H *
s ^ »
r^ « j
^ t* 0*
w 2 S
^ w <
•-3 o w
•-J ^
w cS ®
ffi 
u
_W Q®
w o!
H
O w
— — *^*>
w n
co °»
-™ cu
»»»•
hS a- y. %
5- o 4
_(, l-l
-2 H o
< In
-2 H- g S>
< • ^ r^
^ N Z. rr,
o - w
^ O
»-y
^
-CO 0
z- 0
-CD 5°
— ^

— W

SSTdWVS JO
         E-40
-------
                      15-30     50-70     90-110    130-150    170-190     210-
                  5-15     »50      70-90    110-130    190-170    190-210
                       Hydrogen PeroxWe Concentration (uM)
        Dole we lo« Summn
Figure 4-67  Frequency of  occurrence for H202 concentrations in cloud water
           collected at  the Whiteface, NY  and Whitetop,  VA MCCP sites.
                                      E-41
-------
a,
a.
O
CS1
O
100
 90 -
 80 -
 70 -
 60 -
 50 -
 40 -
 30 -
 20 -
 10
  0
                            ROWLAND FOREST
                      WARM  SEASONS 1986 -  1988
              MEAN OZONE  VS.  36-HOUR  BACK  TRAJECTORY
                             (7278 SAMPLES)
                MAXIMUM OZONE VALUE FOR EACH SECTOR BY YEAR
              36     33     37     27      32      35      64      34
              48     48     54     64      69      76      69      61
              41     65     57     58      79      98      90      75
             346
                    360    425
                                  361    1131  2067
                                  1886   702
                  I       I       I       I
             0-45   45-90  90-135  135-180 180-225 225-270  270-315  315-360
                  MEAN: 28.3           STANDARD DEVIATION: 14.5
                      36-HOUR BACK TRAJECTORY
                                Figure 5-1
                                                 1986
                                                 1987
                                                 1988
 n,
 OH
W
2:
o
tsi
o
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
  0
                         MT.  MITCHELL  SUMMIT
                       WARM  SEASONS  1986 -  1988
              MEAN OZONE VS.  36-HOUR  BACK TRAJECTORY
                             (8684  SAMPLES)
                MAXIMUM OZONE VALUE  FOR EACH SECTOR BY YEAR
              94     98     91     93     92     111     97
              90     86     78     83     91      93     85
              143     116     123     84     116    115    132
             1051
                                                       1629
1035   558
              560
                     980
                            1680
                                          100
                                          105
                                          145
                                                             1191

                         1      T   	I	I	I
             0-45   45-90  90-135  135-180 180-225  225-270 270-315 315-360
                  MEAN: 54.9            STANDARD DEVIATION: 18.9
                      36-HOUR BACK TRAJECTORY
                                Figure 5-2
                                   E-42
1986
1987
1988
-------
.a
 c-
Cd
o
CO
               MEAN
                  MT.  MOOSILAUKE  SUMMIT
                 WARM SEASONS  1986  -  1988
               OZONE VS. 36-HOUR BACK TRAJECTORY
                       (5602  SAMPLES)
100
 90
 80
 70
 60
 5.0
 40
 30
 20
 10
  0 •
                 MAXIMUM OZONE  VALUE FOR EACH SECTOR BY YEAR
               48      31      44     44      48      61      57
               63      74      82     94      88      102     98
               68      60      71     74      109     127     113
                                                  45
                                                  67
                                                  56
                                   199
321
       332
                            293
                                          758
                                                 1212

1702
                                                               785
                  T       I       I       \       \I
              0-45   45-90   90-135  135-180 180-225 225-270 270-315  315-360
                  MEAN: 47.5           STANDARD  DEVIATION: 17.8
                      36-HOUR BACK  TRAJECTORY
                                 Figure 5-3
               1986
               1987
               1988
,£>
a,
a,
DJ
z.
o
[S!
O
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
  0
                          SHENANDOAH  SUMMIT
                       WARM SEASONS  1986  - 1988
              MEAN  OZONE  VS. 36-HOUR BACK  TRAJECTORY
                             (7944  SAMPLES)
                MAXIMUM OZONE VALUE FOR EACH  SECTOR BY YEAR
              34     32     29     31     45     48     49
              98     83     82     75     81     90     99
              81     77     79     96     87     135     132
583
       474
                            697
                                   1850
                                         2382
                           248
                                   224
                                                  38
                                                  90
                                                  140
                                                 1486
                  I      I        \      \       1       I        I
             0-45   45-90  90-135 135-180 180-225 225-270 270-315 315-360
                  MEAN: 43.7           STANDARD  DEVIATION:  19.7
                      36-HOUR  BACK TRAJECTORY
                                 Figure 5-4
                                    E-43
              1986
              1987
             ' 1988
-------
ft
a.
2
O
C-J
O
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
  0
                        WHITEFACE  MT.  SUMMIT
                      WARM  SEASONS 1986 -  1988
              MEAN OZONE VS.  36-HOUR  BACK  TRAJECTORY
                             (6548 SAMPLES)
                MAXIMUM OZONE VALUE FOR EACH SECTOR BY YEAR
              57     69     64     69     81      86     70
              66     57     72     83     97      104     85
              79     58     53     75     102     135    133
                                         623
                                                1960
                                  324
325
              166
       23 1
                                                 54
                                                 61
                                                 66
                                I       I       I       I       I
             0-45   45-90  90-135  135-180 180-225 225-270 270-315 315-360
                  MEAN: 48.5           STANDARD DEVIATION: 18.1
                      36-HOUR BACK TRAJECTORY
                                Figure 5-5
1986
1987
1988
w
•z.
O
IS!
O
 100
 90
 80
 70
 60
 50
 40
 30
 20
 10
  0
                         WHITETOP  MT.  SUMMIT
                       WARM SEASONS  1986  - 1988
              MEAN  OZONE VS. 36-HOUR BACK TRAJECTORY
                             (10319  SAMPLES)
                 MAXIMUM OZONE VALUE FOR EACH SECTOR BY YEAR
               91      82     74     88     96     110     88
              102    106     93     98     104     111     124
              163    112     103     86     104     105     109
              917
                     736
                                         1058
              359
                     533
                                                2450
                                         2399
                                                 103
                                                 111
                                                 144

                                                 1867

                   I       I       I       I      I       I       T
              0-45   45-90   90-135 135-180 180-225 225-270 270-315 315-360
                  MEAN: 55.4           STANDARD  DEVIATION:  22.1
                      36-HOUR BACK  TRAJECTORY
                                 Figure 5-6

                                   E-44
 1986
 1987
 1988
-------
O
2
Cd
Z>
or
cd
K
Ex.
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
  0
                           ROWLAND FOREST
                       WARM  SEASONS 1986 - 1988
                 FREQUENCY OF OZONE CONCENTRATION VS.
                       36-HOUR BACK TRAJECTORY
                           (OZONE >= 70  ppb)
             (68 out of 7278 samples)
                  III!      I       1       I
             0-45   45-90  90-135  135-180 180-225 225-270 270-315 315-360
                     36-HOUR BACK  TRAJECTORY
                               Figure  5-7
>-
o
2
Cd
Cd
Pi
fc,
100
 90
 80
 70
 60
 50
 40
 30
 2Q
 10
 0
                        MT.  MITCHELL  SUMMIT
                       WARM  SEASONS 1986 - 1988
                 FREQUENCY OF  OZONE CONCENTRATION  VS.
                       36-HOUR BACK TRAJECTORY
                           (OZONE >= 70 ppb)
             (1585 out of 8684 samples)
                               I       I      I       1       I
             0-45   45-90  90-135  135-180 180-225 225-270 270-315 315-360
                     36-HOUR BACK  TRAJECTORY
                               Figure 5-8
                                E-45
-------
01
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
 0
                        MT. MITCHELL SUMMIT
                       WARM SEASONS  1906  -  1988
                 FREQUENCY OF  OZONE  CONCENTRATION
                       36-HOUR BACK TRAJECTORY
                          (OZONE >=  100 ppb)
                                             VS.
             (215 out of 8684 samples)
                 r     r      i      i       \^     r      i
             0-45   45-90  90-135  135-180 180-225 225-270 270-315 315-360
                     36-HOUR BACK TRAJECTORY
                               Figure 5-9
CJ>
oz
CL.
100
 90 -
 80 -
 70 -
 60 -
 50 -
 40 -
 30 -
 20 -
 10 -
  0
                      MT.  MOOSILAUKE  SUMMIT
                       WARM SEASONS  1986  - 1988
                 FREQUENCY OF OZONE  CONCENTRATION VS.
                       36-HOUR  BACK TRAJECTORY
                           (OZONE >=  70 ppb)
             (638 out of 5602 samples)
                  I     T      I
             0-45   45-90  90-135  135-180 180-225 225-270 270-315 315-360
                     36-HOUR BACK TRAJECTORY
                               Figure 5-10

                                E-46
-------
o
2
w
3
Of
w
ca
100
 90 -
 80 -
 70 -
 60 -
 50 -
 40 -
 30 -
 20 -
 10 -
 0
                       MT. MOOSILAUKE  SUMMIT
                       WARM SEASONS  1986 - 1988
                 FREQUENCY OF  OZONE  CONCENTRATION  VS.
                       36-HOUR BACK TRAJECTORY
                          (OZONE  >=  100  ppb)
             (66 out of 5602 samples)
                  I      I       1
             0-45   45-90  90-135  135-180 180-235 225-270 270-315 315-360
CJ>
•z,
or
ta
ca
                     36-HOUR BACK  TRAJECTORY
                               Figure 5-11
                         SHENANDOAH  SUMMIT
                       WARM SEASONS  1986  - 1988
                 FREQUENCY OF  OZONE  CONCENTRATION VS.
                       36-HOUR BACK TRAJECTORY
                           (OZONE >=  70 ppb)
             (806 out of 7944 samples)
             0-45   45-90  90-135  135-180 180-225 225-270 270-315 315-360
                     36-HOUR BACK TRAJECTORY
                               Figure 5-12
                                E-47
-------
Cd
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
 0
                        SHENANDOAH SUMMIT
                       WARM SEASONS 1986 -  1986
                FREQUENCY OF OZONE CONCENTRATION VS.
                       36-HOUR  BACK  TRAJECTORY
                          (OZONE >= 100 ppb)
             (47 out of 7944 samples)
                 I       I      I       I
             0-45   45-90  90-135 135-180 180-225 225-270 270-315 315-360

                    36-HOUR BACK TRAJECTORY
                              Figure 5-13
o
•z.
100
 90
 80
 70
 60
 50
 40
 30
 20
 10
  0
                       WHITEFACE  MT.  SUMMIT
                       WARM SEASONS  1986 -  1988
                FREQUENCY OF OZONE  CONCENTRATION VS.
                       36-HOUR  BACK TRAJECTORY
                           (OZONE >=  70  ppb)
             (840 out  of 6548  samples)
                 [       \       I      I       I      T      I
             0-45   45-90  90-135 135-180 180-225 225-270 270-315 315-360
                     36-HOUR BACK TRAJECTORY
                               Figure 5-14
                                 E-48
-------
w
^>
=  100 ppb)
1 UU —
90 -
80 -
70 -
60 -
50 -
40 -
30 -
20 -
10 -


(84 out of 6548 samples)








Y/////^

%%&





%%%'
mz
•0%%,







Wj^
%£/>>?
1 1 1 1 1 1 1 •
             0-45   45-90  90-135  135-180 180-225 225-270 270-315 315-360



                     36-HOUR BACK TRAJECTORY


                               Figure 5-15
=  70 ppb)
             (2511 out of 10319 samples)
                        1       I      I       I       I      I

             0-45   45-90  90-135 135-180 180-225 225-270 270-315 315-360
                     36-HOUR BACK TRAJECTORY


                               Figure 5-16

                                E-49
-------
or
Cd
K
Ci-
ts
100
 90 -
 80 -
 70 -
 60 -
 50 -
 40 -
 30 -
 20 -
 10 -
  0
                       WHITETOP  MT.  SUMMIT
                      WARM  SEASONS i9ae  - 1988
                FREQUENCY OF OZONE CONCENTRATION VS.
                       36-HOUR BACK TRAJECTORY
                          (OZONE  >= 100 ppb)
             140 out of 10319 samples)
                        ii       i      r^     i      i
            0-45   45-90   90-135  135-180 180-225 225-270 270-315 315-360
                    36-HOUR BACK  TRAJECTORY
                              Figure 5-17
                               E-50
-------
                       WHITEFACE  MT.  SUMMIT
                     WARM SEASONS 1986 -  1988
              MEAN  H202  VS. 36-HOUR  BACK TRAJECTORY
                            (697  SAMPLES)



— ~
r* -
c~

c
C-J


S


iu -
9 -

8 -
7 -

6 -
-

4 -
3 -
2 -
-
1 -


UAXIUUU H202 VALUE FOR EACH SECTOR BY YEAR
ND ND ND ND ND ND ND ND
2.2 1.2 1.1 2.5 4.3 4.2 3.1 6.5
ND ND 1.2 0.9 4.3 1.6 1.5 1.4








91 136
54 246 96
.,,,,,,. %%$^ %%&'
\^/^/^( V/S/f/ZA W/fXA \%%Z& '%%%' '%%%• \%Z%& \%%%$.
1 1 1 1 1 1 1

1986
1987
1986










                 MEAN: 0.8            STANDARD DEVIATION: 0.7
                    36-HOUR  BACK  TRAJECTORY
                         ND:  No Data  Available
                              Figure 5-18
.0
c.
c.
CM
o
-------
.3
C.
C.
CM
O
CO
W
s
10
 9
 8
 7
 6
 5
 4
 3
 2
 1
 0
                            ROWLAND  FOREST
                       WARM SEASONS  1986  -  1988
               MEAN  S02  VS.  36-HOUR BACK TRAJECTORY
                             (2426  SAMPLES)
UAX1UUM S02 VALUE FOR EACH SECTOR BY YEAR
ND ND ND ND ND ND ND ND
3.4 2.7 3.1 1.6
2 7 1.8 2 3
71 77 53 59
1 1 1 1
7 8 3.7 2.5
7 6.3 5.4 6.1
431 674 ,_.,

HH HH ^^
1 1 1
1986
1987
1986

             0-45   45-90   90-135 135-180  180-325 225-370  270-315  315-360
                   MEAN:  1              STANDARD  DEVIATION:  1
                      36-HOUR BACK  TRAJECTORY
                           ND:  No Data  Available
                                 Figure 5-20
o
CO
10
 9
 8
 7
 6
 5
 4
 3
 2
 1
 0
                         MT.  MITCHELL  SUMMIT
                       WARM SEASONS  1986  -  1988
                MEAN  S02  VS.  36-HOUR 'BACK TRAJECTORY
                             (3623  SAMPLES)-
                  MAXIMUM
              ND     ND
              58     26
              24 7    7.8
S02 VALUE FOR
   ND     ND
   12.6    10.9
   12.9     5.8
EACH SECTOR BY  YEAR
    ND      ND      ND
    10      8.8      5.6
    10     13.9     13.9
              549
                                                        577
                     365
                            343    297
ND
8 4
 14
                                                               429
                   I       I       I       I       I       II
              0-45   45-90  90-135  135-180  180-325 325-370 270-315 315-360
                   MEAN: 2.6             STANDARD  DEVIATION: 2
                      36-HOUR  BACK  TRAJECTORY
                           ND: No  Data  Available
                                 Figure 5-21

                                     E-52
1986
1987
1988
-------
.0
 C.'
 C.
OJ
O
1/3
                         WHITEFACE  MT.  SUMMIT
                       WARM SEASONS 1986 -  1988
                MEAN S02  VS.  36-HOUR BACK  TRAJECTORY
                             (4664 SAMPLES)
1 U -
9 -

8 -
7 -
6 -
5 -
4 -
3 -

•


MAXIMUM S02 VALUE FOR EACH SECTOR BY
3 1.8 1.8
1.4 1.7 1.1
6 0.7 0.6







235 I5fi 10g
I 1 1
3.3
6.9
1.6






184
HH












1
8.8
13.1
13.1





496
Up













1
10.1
20.5
16.3





1394
HP


YEAR











1
14.2
11.5
10.4






1575
i^
1 9
4
4.3






51 ^
1 O
T

1986
1987
1988









              0-45   45-90  90-135  135-180 180-225 225-270 270-315 315-360
                   MEAN: 1.2            STANDARD DEVIATION: 1.8
                      36-HOUR BACK  TRAJECTORY
                           ND: No  Data  Available
                                 Figure 5-22
CM
O
OJ
10
 9
 8
 7
 6
 5
 4
 3
 2
 1
 0
                         WHITETOP  MT.  SUMMIT
                       WARM SEASONS  1986  -  1988
                MEAN  S02  VS.  36-HOUR BACK  TRAJECTORY
                             (7483 SAMPLES)
                  MAXIMUM S02 VALUE FOR EACH SECTOR  BY YEAR
              ND     ND     ND     ND      ND     ND      ND
              18.5    10.5    2.5     15.5    21.5    16.5    21.5
              30 5    18.5    9.5     3.5     21.5    14.5    30.5
              699
 ND
47.5
20.5
                                                               1462
                     534                  805   1596
                            259    426
1986
1987
1988
             0-45   45-90   90-135 135-180 180-225 225-270 270-315  315-360
                  MEAN:  8.2            STANDARD  DEVIATION:  2.4
                      36-HOUR BACK TRAJECTORY
                           ND:  No Data Available
                                Figure 5-23

                                     E-53
-------
                                        REFERENCES
Anlauf, K.G., H.A. Weibe, and P. Fellin.  1986. Characterization of several integrative sampling
methods for nitric acid, sulphur dioxide and atmospheric particles. J. Air Pollut. Control Assoc. 36:715-
723.

Ashmore,  M.R.,  1988. A comparison of indices that describe the relationship between exposure to ozone
and reduction in the yield of agricultural crops [Comments on article by Lefohn et al. (1988)].  Atmos.
Environ., 6:695-696.

Atlas, D. and S.  Bartnoff, 1953.  Cloud visibility, radar reflectivity and drop-size distribution.  J. Meteor.,
10:143-148.

Bache, D.H., 1979a.  Particle transport within plant canopies - I. A framework for analysis. Atmos.
Environ., 13:1257-1262.

Bache, D.H., 1979b.  Paniculate  transport within plant canopies - II. Prediction of deposition velocities.
Atmos. Environ., 13:1681-1687.

Bache, D.H., 1984.  Prediction of the bulk deposition velocity and concentration profiles within plant
canopies.  Atmos. Environ.,  18:2517-2519.

Bailey, B.H. and M.J. Markus.  1987. Cloud climatology of the Appalachian region.  Prepared for the
MCCP by Associated Weather Services, Inc., Albany, NY.

Bailey, B., M. Markus, and P. Samson. 1989.  Appalachian cloud climatology based on real-time
nephanalysis archive data.  Prepared for the MCCP by Associated Weather Services, Inc., Albany, NY.

Baldocchi, D.D., B.B. Hicks  and  P. Camera, 1987.  A canopy stomatal resistance model for gaseous
deposition to vegetated surfaces.  Atmos. Environ., 21:91-101.

Baldocchi, D.D., D.R. Matt,  R.T. McMillen and B.A. Hutchison, 1985. Evapotranspiration from an
oak-hickory forest.  In: Advances in Evapotranspiration, ASAE Publication 14-85, St. Joseph, MI, pp.
414-422.

Baumgardner, D., 1983:  An Analysis and Comparison of Five Water Droplet Measuring Instruments.  J.
Clim. and Appl.  Meteor., 22.

Baumgartner, A. 1958a.  Zur Hohenabhangigkeit von Regen- und Nebelniedersch-lag am Grossen
Falkenstein (Bayerischer Wald), pp. 529-534.  In: J. Tome, ed. UGGI Comptes  Rendues et Rapport -
Assemblee generale de Toronto 1957 Gentbrugge.

Baumgartner, A. 1958b Nebel lund Niederschlag als Standortsfaktoren am Grossen Falkenstein
(Bayerischer Wald).  Forstw. CBL.  77:257-320.

Bergman, K.H., C.F. Ropelewski and M.S.  Halpert, 1986.  The record Southeast drought of 1986.
Weatherwise, 39:262-266.

Best, AC, "Drop-size distribution in fog and cloud," Q. Jl. R. met. Soc. 77: 418-426, 1951.
-------
Blyth, A.M., A.M.I. Chittenden, and J. Latham, 1984:  An optical device for the measurement of liquid
water content in clouds.  Q.J. Roy.  Meteor. Soc., 110, 53-64.

Bowman, L.D. and R.F. Horak.  1972.  A continuous ultraviolet absorption ozone photometer.  Anal.
Instrum. 10:103-108.

Box, G.E.P. et al., 1978.  Statistics for Experimenters.  New York-John Wiley.

Bredemeire, M. 1988.  Forest canopy transformation of atmospheric dposition. Water, Air, Soil Pollut.
40:121-138.

Brutseart, W.P., 1975.  The roughness length for water vapor, sensible heat, and other scalars.  J. Atmos.
Sci., 32:2028-2031.

Burrows, F.J. and F.L. Milthorpe, 1976.  Stomatal conductance in the control of gas exchange.  In:
Kozlowski, T.T. (Ed.), Water Deficits and Plant Growth, Vol. IV, Academic Press, New York.

Byers, H.R. 1965.  Cloud droplet spectra and growth, pp. 141-172.  In: Elements of Cloud Physics.  The
University of Chicago Press, Chicago.

Calvert, J.G., A. Lazrus, G.L. Kok, E.G. Heikes, J.G. Walega, J. Lind and C.A. Cantrell, 1985:  Chemical
mechanisms of acid generation in the troposphere. Nature, 317, 27-35.
Castillo (1979) - Tl

Chylek, P., 1978:  Extinction and liquid water content of fogs.  J. Atmos. Sci., 35, 296-300.

Cook, E.R., M.A. Kablack and G.C. Jacoby, 1988.  The 1986 drought in the southeastern United States:
How rare an event was it?  J. Geophys.  Res.

Dasch, J.M. 1988.  Hydrological  and chemical inputs to fir trees from rain and clouds during al-month
study at Clingmans Peak, NC. Atmos.   Environ. 22(10):2255-2262.

Daum, P.H., T.J. Kelly, S.E. Schwartz, and L. Newman. 1984. Measurement  of the chemical composition
of strataform clouds.  Atmos. Environ.  18:2671-2684.

Davidson, C.I., J.M. Miller and M.A. Pleskow, 1982.  The influence of surface structure on predicted
particle dry deposition to natural grass canopies.  Water,  Air, and Soil Pollut., 18:25-43.

Davies, C.N., 1968: The entry of aerosols into sampling tubes and leads. Brit. J. Appl. Phys., Ser 2.,
Vol. 1,  921-932.

DeFelice,  T.P.  1988.   Personal communication.  Unpublished data summary of cloud droplet size spectra
of clouds sampled in  the vicinity of Mt. Mitchell, NC, during 1987-1988.  North Carolina State
University, Raleigh, NC.

Dollard, G.J.,  M.H. Unsworth, and M.J. Harve.  1983.  Pollutant transfer in upland regions by  occult
precipitation.  Nature 302:241-243.

Dye, J.E. and  D. Baumgardner.  1984.  Evaluation of the forward scatter spectrometer probe.  Part I:
Electronic and optical studies. J. Atmos. and Oceanic Technol. 1:329-344.

Eaton, J.S., G.E. Likens and F.H. Bormann, 1978. The input of gaseous and paniculate sulfur to a
forest ecosystem.   Tellus, 30:546-541.
-------
Eidmann, F.E. 1959.  Interception in beech and spruce stands: Results of investigation over a number of
years in Rothaer Mountains (Sauerland).  Publication No. 48, Association Internationale D'Hydrologie
Scientifique, Gentbrugge, pp. 5-25.

Enderle, K.H. and W. Jaeschke.  1988a. Problems of fog sampling.  J. Aerosol Sci.  18:793-795.

Enderle, K.H. and W. Jaeschke.  1988b. An intercomparison study dealing with chemistry and physics of
fogwater collection, pp. 7-74. In: W. Jaeschke and K.H. Enderle eds. Chemistry and physics of fogwater
collection.  PBT Bericht, Munich West Germany.

Evens, L.S. 1982.  Biological effects of acidity in precipitation on vegetation:  A review.  Environ.  Exp.
Bot. 22:155-169.

Falconer, R.E. and P.D. Falconer. 1980. Determination of cloudwater acidity at a mountain observatory
in the Adirondack Mountains of New York State.  J. Geophys. Res. 85C:7465-7470.

Federal Register. 1979.  Ambient air monitoring reference and equivalent methods.  40 53:952-986.

Federal Register. 1980.  Ambient air monitoring reference and equivalent methods.  45: 18474.

Fowler, D., J.N. Cape, I.D. Leith, T.W. Choularton, M.J. Gay, and A. Jones. 1988.  The influence of
altitude on rainfall composition at Great Dunn Fell, Atmos. Environ. 22(7):1355-1371.

Fuchs, N.A. 1964.  Mechanics of Aerosols. Pergamon Press, New York.

Fuzzi, S., R.A. Castillo, J.E. Jiusto, and G.G. Lala.  1984.  Chemical composition of radiation fog water
at Albany, NY and its relationship to fog microphysics. J. Geophys. Res. 89:7159-7164.

Fuzzi, S., O. Giordano, G.  Nardini, M.C. Facchini, S. McLaren, E. McLaren, and M. Mariotti. 1988.
Heterogeneous  processes in the Po Valley radiation fog.  J. Geophys. Res. 93(D9):11,141-11,151.

Fye, F.K. 1978. The AFGWC automated cloud analysis model. Tech. Memo. 78-002, US  AirForce,
Offutt Air Force Base, NE.

Galloway, J.N. and D.M. Whelpdale, 1980.  An atmospheric sulfur budget for eastern North America.
Atmos. Environ., 14:409-417.

Garratt, J.R. and B.B. Hicks, 1973.  Momentum, heat, and water vapor transfer to and from natural and
artificial surfaces.  Quart. J. Roy. Meteor.  Soc., 99:680-687.

Geladi, P. and B.R. Kowalski, 1985.  Partial least squares regression:  a tutorial.  Anal. Chim  Acta,
185:1-32.

Gerber, H.  1984.  Liquid water content of fogs and hazes from visible light scattering. J.  Clim Appl.
Meteor. 23:1247-1252.

Gerber, H.  1988.  Measurement of suspended paniculate volume and farinfrared extinction coefficient
with a new laser-diffraction instrument.  In Proceedings of 19th Annual Meeting of the Fine Particle
Society, July 19-22, Santa Clara, CA.

Gertler, AW. and R.L. Steele.  1980. Experimental verification of the linear  relationship between IR
extinction and liquid water content of clouds. J. Appl. Meteor. 19:1314-1317.
-------
Goodell, B.C. 1952. Watershed management aspects of thinned young lodgepole pine stands. J. Forestry
50(5):374-378.

Gordon, C., R Hovanec and W. Stern. 1984.  Analyses of monthly mean cloudiness and their effect on
GCM diagnosed radiative fluxes. J. Atmos. Sci.

Grunow, J. 1953.  Kritische Nebelfrostststudien.  Arch. Met. Geophys. Bioklim.  8:389-419.

Grunow, J. 1957.  Vergleichende  Messungen des Nebelniederschlags. UGGI Assoc. Int. Hydr. Sci.
Assemblee, Toronto.

Grunow, J. 1958.  Kritik des Nebelfangers. Zur Methode der Messung des Nebelniederschlags.  Techn.
Mitt. d. Instrumentenwesens DWD N.F. 60-68.

Hering, S.V., D.L.  Blumenthal, R.L.  Brewer, A. Gertlen,  M. Hoffman, J.A. Kadlecek, and K. Pettas.
1987.  Field intercomparison of five  types of fogwater collectors.  Environ. Sci. Technol. 21(7):654-663.

Hicks, B.B., D.D. Baldocchi, R.P.  Hosker Jr., B.A. Hutchinson, D.R. Matt, R.T. McMillen and L.C.
Satterfield, 1985.  On  the use of monitored air concentrations to infer dry deposition.  U.S. Department
of Commerce,  NOAA Tech. Memo.  ERL ARL-141.

Hicks, B.B., D.D. Baldocchi, T.P. Meyers, R.P. Hosker and  D.R. Matt, 1987. Deriving dry deposition
velocities from measured quantities.  Water, Air,  and Soil Poll., 36:311-330.

Hicks, B.B., M.L. Wesely and J.L. Durham, 1980.  Critique of methods to measure dry deposition:
Workshop summary.  EPA-600/9-80-050.

Hogsett, W.E., D.T. Tingey and E.H. Lee, 1988.  Exposure  indices:  concepts for development and
evaluation of their use.  In:  W.W. Heck, O.C. Taylor and D.T. Lee (Eds.), Assessment of Crop Loss
from Air Pollutants. Elsevier Applied Science Publishing, London, U.K., 107-138.

Hornbeck, J.W., and R.B. Smith.  1985.  Documentation  of red spruce decline.  Canadian J. Forest Res.
1199-1201.

Hosker, R.P. Jr. and S.E. Lindberg,  1982.  Review, atmospheric deposition and plant assimilation of
gases and particles. Atmos. Environ.,  16:889-910.

Houghton, H.G.  1955.  On the chemical composition of fog and  cloud water. J. Meteorol. 12:355-357.

Houghton, H.G. and Radford, W.H. 1938. On the measurement of drop-site and liquid water content in
fogs and clouds. Pap. Phys. Oceanog. Meteor. Vol. 6, Masachusetts Institute of Technology, 31 pp.

Hughes, N. and A. Henderson-Sellers.  1985.  Global 3D-Nephanalysis of total cloud amount: Climatol-
ogy for 1979. J. Climate Appl. Meteor. 24:669-686.

Hughes, N. and A. Henderson-Sellers.  1986.  Layered cloud amounts for January and July 1979  for 3D-
Hephanalysis. J. Climate Appl. Meteor. 25:118-132.

Huygen, C.  1963.   The sampling  of sulfur dioxide in air with impregnated filter paper. Anal. Chim.
Acta 28:349.

Imhoff, R.E. and J.E.  Malo. Submitted.  An estimate of cloud impaction frequency along the Ap-
palachians.
-------
Jacob, D.J., J.M. Waldman, J.W. Munger, and M.R. Hoffman. 1985a.  Chemical composition of fogwater
collected along the California coast.  Environ. Sci. Technol. 19(8):730-736.

Jacob, D.J., J.M. Waldman, J.W. Munger, and M.R. Hoffman. 1985b.  An instrument to collect fogwater
for chemical analysis.  Rev.Sci. Instru. 56:1291-1293.

Jacob, D.J., J.W. Munger, J.M. Waldman, and M.R. Hoffmann.  1986.  The H^CVHNOj-NH? system
at high humidities and in fogs: 1. Spatial and temporal patterns  in the San Joaquin Valley of California.
J. Geophys. Res. 91 (Dl): 1073-1088.

Jarvis, P.O., 1976.  The interpretation of the variations in leaf water potential and stomatal conductance
found in canopies in the field. Phil. Trans. Roy. Soc., London, B 273:593-610.

Jarvis, P.O., 1971.  The estimation  of resistances to carbon dioxide transfer.  In: Sestak, Z., Catsky, J.,
and P. G. Jarvis (eds), Plant photosynthetic production: Manual  of methods, Junk, The Hague,  pp.
566-631.

Jarvis, P.G., G.B. James and J.J.  Landsberg, 1976. Coniferous forest. In: Vegetation and the
Atmosphere, Vol. II, J.L. Monteith (Ed.), Academic Press, New  York, pp. 171-236.

Jiusto, J.E., 1965. Cloud Particle Sampling - Dept. of Meteor., Penn State University, Report No. 6,
University Park. Penn.

Jiusto, J.E., and C.G.  Lala, 1982: Liquid water content in radiation fogs.  Preprints Conf. on Cloud
Physics, Chicago, IL, Amer.
Meteor. Soc., 300-302.

Johnson, A.H., and T.A. Siccama.  "Acid deposition and  forest decline." Env. Sci. & Tech.  17:  294-306,
1983.

Joslin, J.D. Mueller, S.F. and Wolfe, M.H. (1990). The use of artifical and living collectors in the testing
of models of cloud water deposition to forest canopies (submitted to Atmospheric Environment).

Joslin, J.D., C. McDuffie, and P.P.  Brewer.  1988. Acidic cloud  water and cation  loss from  red  spruce
foliage.  Water, Air, Soil Pollut. 39:355-363.

Junge, C. 1963.  Air Chemistry and Radioactivity. Academic Press,  New York.

Kadlecek, J.A, S. McLaren, V. Hohnen, B. Mossl, A. Kadlecek,  and N. Camarota.  1985. Winter
cloudwater chemistry studies: Whiteface Mountain 1983-84.  Report to Empire State Electric Energy
Res. Corporation, EP82-10, Atmospheric Science Res. Center, State University  of New York at Albany,
Albany,  NY.

Kadlecek, J.A., T. Arons, I. Cohen, S. McLaren, and R. Pratt. 1988.  Wintertime trace gas and
cloudwater chemistry studies. Report to Empire State Electric Energy Research Corporation, Atmos.
Sci. Res. Center, SUNY-Albany, NY.

Kadlecek, J.A. 1989.  Special Problems in Atmospheric Exposure, Appendix II.  ASRC/State University
of New York at Albany, Publication No. 1113, Albany, NY.

Kahl, J.D. and P.J. Samson, 1988.  Trajectory sensitivity to rawindsonde data resolution.  Atmos.
Environ., 2(7):1291-1299.
-------
Kaupenjohann, M., B.U. Schneider, R. Hantschel, W. Zech, and R. Horn.  1988. Sulfuric acid rain
treatment of Picea abies: Effects on nutrient solution, throughfall chemistry, and tree nutrition.  Z.
Pflanzenernahr.  Bodenk.  151:123-126.

Kelly, T.J., P.M. Daum, and S.E. Schwartz.  1985.  Measurements of peroxides in cloudwater and rain. J.
Geophys. Res. 90:7861-7871.

Kelly, J.M. and R.C. Strickland.  1986.  Throughfall and plant nutrient concentration response to
simulated acid rain treatment. Water, Air, Soil Pollut. 29:219-231.

King, W.D., D.A. Parkin, and R.J.  Handsworth.  1978.  A hot-wire liquid water  device having fully
calculable response characteristics.  J. Appl. Meteorol.  17:1809-1813.

Kins, L., F.X. Meixner, D.H. Ehhalt, K.P.  Muller, J. Slanina, J.J. Moels,  N. Beltz, W. Jaeschke, and K.J.
Rumpel.  1986.  Scavenging of trace  substances by cloud precipitation.  Preprints Symposia  "Acid  Rain:
I.  Sources and atmospheric processes."  Am. Chem. Soc., PETR Div., 31(2):457-463.

Kittredge, J.  1948.  In:  Interception and  stemflow.  In: Forest Influences. Chapt.  XI. McGraw-
Hill,NY, 99-114, 394 pp.

Kleindienst, T.A., P.B. Shepson, D.N. Hodges, C.M. Nero,  R.R. Arms, P.K. Dasgupta, H. Hwang, G.L.
Kok, J.A. Lind, A.L. Lazrus, G.I. Mackay, L.K. Mayne, and H.I. Schiff.   1988.  Comparison of techniques
for measurement of ambient levels of hydrogen peroxide. Environ. Sci. Technol. 22:53-61.

Kley, D., H. Geiss, T. Heil and C.  Holzapfel, 1989.  Concentration, trend and distribution of reactive
trace gases in rural  Germany.  Paper presented at the International Congress on Forest Decline
Research: State of Knowledge and Perspectives, Friedrichshafen, 2-6 October.

Knollenberg, R.G. 1981.  Techniques for probing cloud microstructure, pp. 15-89. In: P.V.  Hobbs and
A. Deepak, eds. Clouds, Their Formation, Optical Properties, and Effects.  Academic Press, London.

Kohlmaier, G.H., E.O. Sire, U. Fischbach, H. Broehl, and M. Ploechl. 1986.  Modes of interaction of air
pollutants with spruce-fir forests: Summary and new mechanisms, pp.  In: W. Jaeschke, ed.  Chemistry of
Multiphase Atmospheric Systems.  NATO ASI Series, Vol G6.  Springer-Verlag, Berlin.

Kroll, G. and P. Winkler.  1989.  Trace substance input to coniferous forests via cloud interception.  In:
H.W. Georgii, ed. Mechanisms  and Effects of Pollutant-Transfer into Forests.  Kluwer Academic
Publishing.  Dordrecht, Holland.

Krovetz, D.O.,  M.A Reiter, J.T. Sigmon and L.S. Gilliam, 1988. Assembly and field testing of a
ground-based presence of cloud detector.  J. Atmos. Oceanic Technol., 5:579-581.

Krovetz, D.O.,  M.A Reiter, and J.T. Sigmon.  1989. A model for  estimating cloud water deposition to a
hardwood forest.  Atmos. Environ.

Krupa,  S.V. and R.N. Kickert, 1987.  An analysis of numerical models of air pollutant exposure and
vegetation response.  Environ. Pollut., 44:127-158.

Lang, A.R.G.,  1987.  Simplified estimate of leaf area index from transmittance of the sun's  beam. Agric.
For. Meteorol., 41:179-186.

Lazrus, A.L., H.W.  Baynton, and J.P. Lodge,  Jr. 1970.  Trace constituents in oceanic cloud  water  and
their origin.  Tellus 22:106-113.
-------
Lazrus, AL., P.L. Haagenson, G.L. Kok, B.J. Huebert, C.W. Kreitzberg, G.E. Likens, V.A Mohnen,
W.E. Wilson, and J.W. Winchester.  1983.  Atmos. Environ. 17(3):581-591.

Lazrus, AL., Kok, G.L., Gitlin S.N., Lind, J.A and McLaren S.E.  1985. Automated fluorimetric
method for hydrogen peroxide in atmospheric precipitation.  Analyt. Chem. 57:917-922

Lazrus, AL., G.L. Kok, J.A Lind, S.N. Gitlin, B.G. Heikes, and R.E. Shelter. 1986.  Automated
fluorometric method for hydrogen peroxide in air.  Anal.  Chem. 58:594-597.

Lee, E.H., D.T. Tingey and W.E. Hogsett, 1988.  Evaluation of ozone exposure indices in
exposure-response modeling.  Environ. Pollut., 53:43-62.

Lefohn, AS.  and H.M. Benedict, 1982.  Development of a mathematical index that describes ozone
concentration, frequency, and duration.  Atmos. Environ.,  16:2529-2532.

Lefohn, AS.  and KM. Benedict, 1985.  Exposure considerations associated with characterizing ozone
ambient air quality monitoring data.  In: Evaluation of the Scientific Basis for Ozone/Oxidants Standards.
Air Pollution Control Association, Pittsburgh, PA 17-31.

Lefohn, AS.  and C.K. Jones, 1986.  The characterization of ozone and  sulfur dioxide air quality data for
assessing possible vegetation effects. JAPCA, 36:1123-1129.

Lefohn, AS.  and J.E. Pinkerton, 1988.  High resolution characterization of ozone data for sites located
in forested areas of the United States. JAPCA, 38:1504-1511.

Lefohn, AS.  and V.C. Runeckles, 1987.  Establishing a standard to protect vegetation - ozone
exposure/dose considerations.  Atmos. Environ., 21:561-568

Lefohn, AS., S.V. Krupa and D. Winstanley, 1989a.  Surface Ozone Exposures Measured at Clean
Locations Around the World.  Environ.  Pollut., (In Press).

Lefohn, AS., J.A  Laurence and  R.J. Kohut,  1988. A comparison of indices that describe the
relationship between exposure to ozone  and reduction in the yield of agricultural crops. Atmos.
Environ., 22:1229-1240.

Lefohn, AS., V.C. Runeckles, S.V. Krupa  and D.S. Shadwick, 1989b. Important Considerations for
Establishing a secondary ozone standard to protect vegetation.   JAPCA 39:1039-1045.

Lefohn, AS., D.S. Shadwick and V.A Mohnen, 1989c. The characterization of ozone and sulfur dioxide
concentrations at a select set of  high-elevation sites in the eastern United States.  Environ.  Pollut.
(Submitted).

Leonardi, S. and W. Flueckiger.  1987.  Short-term canopy interactions of beech trees: Mineral ion
leaching and absorption during rainfall.  Tree Physiol.  3:137-145.

Lin, N.H. and V.K. Saxena.  1989.  Acidic deposition by dry, wet,  and cloud capture mechanisms in a
coniferous forest.  In: Measurements of Toxic and Related Air Pollutants, U.S. Environmental Protection
Agency/Air and Waste Management Association Symposium, Raleigh, NC.

Lindberg, S.E., G.M. Lovett, and K.J. Meiwes. 1987.  Deposition  and forest canopy interactions of
airborne nitrate, pp. 117-130.  In: T.C. Hutchinson and K.M. Meema, eds.   Effects of Atmospheric
Pollutants on Forests, Wetlands  and Agricultural Ecosystems.  Springer-Verlag, Berlin.
-------
Lindberg, S.E., D. Silsbee, D.A Schaefer, J.G. Owens, and W. Petty.  1988.  A comparison of at-
mospheric exposure conditions at high-elevation forests in the southern Appalachian Mountain  range, pp.
321-344.  In: M.H. Unsworth and D. Fowler, eds.  Acid Deposition at High Elevation Sites.  NATO ASI
Series C: Vol. 252.

Lindberg, S.E., D.W. Johnson, eds.  1989.  1988 Annual report of the Integrated Forest Study.
ORNL/TM-11121 Draft, U.S. Department of Energy contract DE-ACO5-840R21400, report to Electric
Power Research  Institute (RP2621), Oak Ridge National Laboratory, Oak Ridge, TN.

Linke, F.  1916.  Niederschlagsmessungen unter Baumen.  Met.  Zschr.  33:140-141.

Lovett, G.M., W.A, Reinert, and R.K. Olson.  1982.  Cloud droplet deposition in subalpine balsam fir
forests:  Hydrological and chemical inputs. Science.  218:1303-1304.

Lovett, G.M., 1984.  Rates and mechanisms  of cloud water deposition to a subalpine balsam fir forest.
Atmos. Environ. 18:361-371.

Lovett, G.M., and W.A Reiners. "Canopy structure and cloud water deposition in subalpine coniferous
forests," Tellus 38B:  319-327, 1986.

Lovett, G.M. 1988b.  A comparison of methods for estimating cloud water deposition to a New
Hampshire (U.S.A>) Subalpine Forest, pp. 309-320.  In: M.H. Unsworth and D.  Fowler, eds. Acid
Deposition at High Elevation Sites, NATO ASI Series C: Vol. 252.

Mallant, R.K.AM., J. Salanina, G.  Masuch, and A  Kettrup.  1985.  Experiments on H^-containing  fog
exposures on young trees. Presented  at the  U.S.-Dutch Symposium on Aerosols, Williamsburg, VA.

Mallant, R.K.AM.  1988.  A fog chamber and wind tunnel facility for calibration of cloud water
collectors, pp. 479-490.  In: M.H. Unsworthand D. Fowler, eds.  Acid  Deposition Processes at High
Elevation Sites, NATO ASI Series  C, Vo. 252, Dordrecht, Holland.

Mallant, R.K.AM. and G.P.A Kos, 1989. An optical device for the detection of clouds and fog.
Submitted to Aerosol Science and Technology, in press.

Mallant, R.K.AM., K. Elsholz, and B. Bailey.  1989.  Fog detection  with a low-cost, forward-scattering
device.  In: Proceedings of the Symposium on the Role of Clouds in Atmospheric Chemistry and Global
Climate.  Anaheim, CA, Amer. Meteorological Society.

Markus, M. and  B. Bailey.  1989.  Cloud frequency  determination at  high elevations using an optical
cloud detector.   In:  Proceedings of the 19th Conference on Agriculture and Forest Meteorology,
Charleston, SC, American Meteorological Society.
Markus et al., 1989

Markus, M. and B. Bailey, and R. Stewart.  1989.  Estimation of cloud frequency at high elevation
forests in the eastern United States.  In: Proceedings  of Symposium on the Role of Clouds in
Atmospheric Chemistry and Global Climate.  Anaheim, CA American Meteorological Society.

Mason, B.J. 1971.  The Physics of Clouds.  Clarendon Press, Oxford, 671 pp.

Matt, D.R., R.T. McMillen, J.D. Womack and B.B.  Hicks, 1987. A comparison of estimated and
measured SO2 deposition velocities.  Water,  Air,  and  Soil Poll., 36:331-347.

Matt, D.R. and J.D. Womack, 1988.  Atmosphere-surface energy exchange of ozone to a spruce conifer
forest: Measured and inferred.  Quadrennial  Ozone Symposium-88, Gottingen, FRG, August 4-6.
-------
Mountain Cloud Chemistry Project.  Quality Assurance Plan for the Mountain Cloud Chemistry Project,
(revision 5), 1988.

MCCP Standard Operating Procedures Manual, 1989.  W.S. Fleming and Associates, Albany, NY.

McCune, D.C. and T.L. Lauver.  1986.  Experimental modeling of the interaction o wet and dry
deposition on conifers. In: S.D. Lee, T. Schneider, L.D. Grant, and PJ. Verkerk, eds. Aerosols. Lewis
Publishing, Chelsea, MI.

McLaughlin, S.B. 1985. Effects of air pollution on forests. J. Air Pollut.  Control Assoc. 35:516-34.

Merceret, Francis J. and T.L. Shricker, 1975:  A new hot-wire liquid cloud water meter.  J. Appl.
Meteor., 14, 319-326.

Milne, R., A. Crossley, and M.H. Unsworth.  1988.  Physics of cloudwater deposition and evaporation at
Castlelow, S.E.  Scotland,  pp. 299-307.  In: M.H.  Unsworth and D. Fowler, eds. Acid Deposition at High
Elevation Sites, NATO ASI Series C, Vol. 252, Kluwer Publishing, Boston.

Mohnen, V.A. 1987.  Droplet size spectra from Whiteface Mtn. Report to MCCP management, State
University of New York at Albany, Albany, NY.

Mohnen, V.A. 1988a.  Mountain Cloud Chemistry Project: Wet, dry, and cloud water deposition. E.P.A
Contract No. CR-813934-01-2, U.S. Environmental Protection Agency AREAL, Research Triangle Park,
NC.

Mohnen, V.A, 1988b.  Chemical deposition to high elevation-fir forests in the eastern United States: A
preliminary assessment 1987.  EPA Grant No. EPA813-934010, U.S. Environmental Protection Agency
AREAL, Research Triangle Park, NC.

Mohnen, V.A. 1988c.  Exposure of forests to air pollutants, clouds, precipitation and climatic variables.
EPA Contract No. CR-813934-01-2, U.S. Environmental Protection Agency AREAL, Research Triangle
Park,  NC.

Mohnen, V.A. and J.A. Kadlecek.  1989.  Cloud chemistry research at Whiteface Mountain.  Tellus
41B:79-91.

Monteith, J.L.,  1965.  Evaporation and environment.  Symp. Soc. Exp. Biol., 19: 205-234.

Morgan, J.J. 1982. Factors governing the pH, availability of H+, and oxidation capacity of rain, pp. 17-
40.  In: E.D. Goldberg, ed. Atmospheric Chemistry.  Springer-Verlag, NY.

Mrose, H. 1966. Measurements of pH and chemical analyses of rain-, snow-, and fog-water.  Tellus
8:266-270.

Mueller, S.F. (1990).  Estimating cloud water deposition to subalpine spruce-fir forests - I:  modifications
to an  existing model (under revision-submitted to Atmospheric Environment).

Mueller, S.F. and Weatherford, P.P. (1988).  Chemical deposition to a high elevation red spruce forest.
Water. Air, and Soil Pollution 38, 345-363.

Mueller, S.F., Joslin, J.D. and Wolfe, M.H. (1990).  Estimating cloud water deposition to subalpine
spruce-fir forests - II:  model testing (under revision-submitted to Atmospheric Environment).
-------
Mueller, S.F. and R.E. Imhoff.  1989.  Inferring cloud deposition to a forest canopy using a passive
cloudwater collector.  Geophys. Res. Lett.

Muir, P.S., K.A. Wade, B.H. Carter, T.V. Armentano, and R.A. Pribush.  1986.  Fog chemistry at an
urban midwestern site. J. Air Pollut. Control Assoc.  36(12): 1359-1361.

Munger, J.W., D.J. Jacob, J.M. Waldman, and M.R. Hoffmann.  1983.  Fogwater chemistry in an urban
atmosphere. J. Geophys. Res. 88(C9):5109-5121.

Murphy, C.E. Jr., T.R. Sinclair and K.R. Knoerr, 1977.  An assessment of the use of forests as sinks for
the removal of atmospheric sulfur dioxide. Environ.. Quality, 6:388-396.

National Oceanic and Atmospheric Administration.  1975.  Local climatological data, annual summary
with comparative data, Mount Washington Observatory, Gorham, New Hampshire.  National Oceanic
and Atmospheric Administration, National Climatic Center, Asheville, NC.

Nicholson, S. and J.T. Scott.  1969.  Slope aspect variation in the vascular plant species composition in
the treeless community near the summit of Whiteface Mountain, NY, pp. 147-160.  In: J.G. Holway and
J.T. Scott, eds.  Vegetation-environment relations at Whiteface Mountain in the Adirondacks.
Atmospheric Sciences Research Center, State University of New York at Albany, Albany, NY.
                         H

Noone, K.J. 1988.  Solute concentration is size dependent.  J. Geophys. Res. 93(D8):9477-9482.

Oddie, B.C.V.  1962.  J. Royal Meteorol. Soc. 80:535-538.

Ogren, J. and H. Rodhe. 1986.  Measurements of the chemical composition of cloud water at a clean
air site in Central Scandinavia.  Tellus 383:190-196.

Okabe, H., P.L. Splitstone, and J.J. Ball.  1973. Ambient and source 862 detectors based on a
fluorescence method.  J. Air Pollut. Control  Assoc. 23:514-516.

Okita, T.J. 1968.  Concentration of sulfate and other inorganic materials in  fog and cloudwater and in
aerosol. J.  Meteorol. Soc. Japan (46(2):120-126.

Olson, R.K., W.A. Reiners, C.S. Cronan, and G.E. Lang.  1981. The chemistry and flux of throughfall
and stemflow in subalpine balsam fir forests.  Holartic Ecol. 4:291-300.

Olszyna, K.J., J.F. Meagher,  and E.M. Bailey.  1988.  Gas phase, cloud and rain-water measurements of
hydrogen peroxide at a high elevation site.  Atmos. Environ. 22:1699-1706.

Paluch, I.R. 1988. A note on the variability of cloud droplet size spectra.

Parker, C.G. 1983.  Throughfall and stemflow in the  forest nutrient cycle. Advances in Ecological
Research.  In: A. Macfadyn and E.D. Ford, eds. Academic Press.

Parry, M.A.J. and W.  Day, 1988. A comparison of indices that describe the relationship between
exposure to ozone and reduction in the yield of agricultural crops [Comments on article by Lefohn et al.
(1988)].  Atmos. Environ., 22:2057-2058.

Penman, H.L.,  1948.  Natural evaporation from open water, bare soil and grass. Proc. Roy. Soc. London
(a) 193:120-145.

Petrenchuk, O.P. and V.M. Drozdova.  1966. On the chemical composition of cloudwater.  Tellus
18:280-286.
-------
Pinnick, R.G., D.M. Garvey, and L.D. Duncan.  1981.  Calibration of Knollenberg FSSP light scattering
counters for measurement of cloud droplets. J. Apl. Meteor. 20:1049-1057.

Pratt, R.W., I.D. Cohen, J.A. Kadlecek, and S.E. McLaren.  1986.  Meteorological analysis of Whiteface
Mountain winter cloud chemistry events, pp. 352-364.  In: J. Laznow and G. Stensland, ed. Meteorology
of Acid Deposition.  Air  Pollut. Control Association.

Pruppacher, H.R. and J.D. Klett.  1980.  Microphysics of clouds and precipitation, pp. 12-18. D. Reidel,
Dordrecht, Holland.

Raschke, K., 1956. Uber die Physikalishchen zwischen Warmeubergangszahl, Strahlungsaustausch,
Temperatur, und Transpiration eines Blattes.  Planta, 48:200-238.

Reuss, J.O., B.J. Cosby, and  R.F. Wright.  1987.  Chemical processes governing soil and water
acidification. Nature 329:27-32.

Reynolds, E.R.C. and L. Leyton. 1963.  Measurement and significance of throughfall in forest stands, pp.
127-141.  In: The Water Relations of Plants.  Blackwell Scientific Publications, London.
                                                                                      9
Robarge, W. 1989. Personal communication. Unpublished throughfall chemistry data from North
Carolina  State University, Raleigh, NC.

Rodenkirchen, H. 1986. Auswirkungen von saurer Beregnung und Kalkung auf die Vitalitaet,
Artenmaechtigkeit und Naehrstoff versorgung der  Bodenvegetation eines Fichtenbestandes.  Forstwiss.
Cbl. 105:338-350.

Rodriguez, J.L. and W.J.  Davies, 1982.  The effects of temperature and ABA on stomata of  Zea Mays L.
J. Expt. Botany, 33:977-987.

Rogers, R.R. 1976.  Droplet growth by condensation, pp. 69-85. In: A Short Course in Cloud Physics.
Pergamon Press, Oxford,  England.

Romer, F.G., J.W. Viljeer, L. vanden Beld, H.J. Slangowal, and A.A. Veldkamp. 1985.  The  chemical
composition of cloud and rainwater: Results of preliminary measurements from an aircraft.   Atmos.
Environ.  19(11):1847-1858.

Rubner, K. 1932.  Der  Nebelniederschlag im Wald und Seine Messung (1).  Thrandt. Forstl  jb.  83:121-
149.

Rubner, K. 1935.  Der  Nebelniederschlag im Wald und Seine Messung. Tharandt Forstl. Jb., 86:330-342.

Runeckles, V.A., 1988.  A comparison of indices that describe the relationship between exposure to
ozone and reduction  in the yield of agricultural crops [Comments on article by Lefohn et al. (1988)].
Atmos. Environ., 22:1241-1242.

Saxena, V.K. et al. 1989.  Monitoring the chemical climate of the Mt. Mitchell State Park for evaluation
of its impact on forest decline.  Tellus 418:92-109.

Schaefer, D.A., S.E. Lindberg, and G.M. Lovett.  1989.  Canopy acid balance and foliar cation exchange.
In: S.E. Lindberg, and D.W.  Johnson, eds. 1988 Annual report of the Integrated Forest Study.
ORNL/TM-11121 Draft, U.S. DOE contract  DE-AC05-840R21400, Report  to Electric Power Research
Institute (RP2621), Oak Ridge National  Laboratory, Oak Ridge, TN.
-------
Schemenauer, R.S. 1986.  Acidic deposition to forests:  The 1985 chemistry of high elevationfog (CHEF)
Project. Atmosphere-Ocean 24:303.

Schemenauer, R.S. 1986.  Acidic deposition to forests:  The 1985 chemistry of high elevationfog (CHEF)
Project. Atmosphere-Ocean 24:303.

Scherbatskoy, T. and Klein, R.M. 1983.  Response of spruce and birch foliage to leaching by acidic
mists. J. Environ. Quality 12:189-195.

Schulz, T. and P. Samson.  1988.  Nonprecipitating low cloud frequencies for central North America:
1982. J. Appl. Meteor. 27:427-440.

Scott, W.D., 1978. Atmos.  Environ. 12:917-921.

Scott, B.C. and N.S. Laulainen.  1979.  J. Appl. Meteorol. 18:138-147.

Seigneur, C. and P. Saxena.  1988. A theoretical investigation of sulfate formation  in clouds.  Atmos.
Environ.  11:101-115.

Shaw, R.W. Jr., R.K. Stevens, and xx Bowermaster.  1982. Measurement of atmospheric nitrate and
nitric acid: The denuder difference experiment.  Atmos. Environ. 16:845.

Shreffler, J.H., 1978.   Factors affecting dry deposition of SO2 on forests and grasslands.  Atmos.
Environ., 16:1497-1503.

Siccama, T.G., 1974.   Vegetation, soil, and climate on the Green Mountains of Vermont. Ecological
Monographs 44:325-349.

Sickles, J.E. 1987.  Sampling and analytical methods development for dry deposition monitoring.
Research Triangle Institute Report No. RTI/2830/00-15F.

Sigmon, J.T., F.S. Gilliam, and M.E. Partin.  1989. Precipitation and throughfall chemistry for a
montane hardwood ecosystem:  Potential contributions from cloud water.  Canadian J. Forest Res.

Sinclair, T.R., C.E. Murphy and K.R. Knoerr, 1976.  Development and evaluation of simplified models
for simulating canopy photosynthesis and transpiration.  J. App. Ecol.,  13:813-829.

Slinn, W.G.N., 1982.  Predictions for particle deposition to vegetative canopies.  Atmos. Environ.,
16:1785-1794.

Stevens, R.K., 1985.  Modern methods to measure air pollutants.  Proc. of 2nd U.S.-Dutch International
Symposium: Aerosols. May 19-24, Williamsburg, VA.

Stewart, J.B. and AS. Thorn, 1973.  Energy budget in a pine forest.  Quart. J. Roy. Meteor. Soc.,
99:154-170.

Stogner, R.E. and V.K. Saxena.  1988.  Acidic cloud-forest canopy interactions:  Mt. Mitchell, North
Carolina.  Environ. Pollut. 53:456-458.

Spyer-Duran, P.A 1968.  Comparative measurement of cloud liquid water using heated wire and cloud
replicating devices.  J. Appl. Meteor. 7:674-678.

ten Brink, H.M., S.E.  Schwartz, and D.H. Daum.  1987.   Efficient scavenging of aerosol sulfate by liquid-
water clouds.  Atmos. Environ. 21(9):2035-2052.
-------
Thome, P.O., Lovett, G.M. and Reiners, W.A. (1982).  Experimental determination of droplet impaction
on canopy components of balsam fir. Journal of Applied Meteorology 21, 1413-1416.

Tingey, D.T., 1984.  The effects of ozone on plants in the United States. In: P. Grennfelt (Ed.), Proc.
International Workshop on the Evaluation of the Effects of Photochemical Oxidants on  Human Health,
Agricultural Crops, Forestry, Materials and Visibility. Swedish Environmental Research  Institute,
Goetborg, Sweden.

Trautner, F.  1988.  Entwicklungund Anwendung von Mess systemen zur untersuchun der chemischen
und physikalischen Eigenschaften von Nebelwasser und dessen Deposition  auf Fichten.  Diplomarbeit,
Faculty ofBiology, Chemistry and Geosciences.  University Bayreuth, Bayreuth, West Germany.
Turner, D.P., D.T. Tingey, and W.E. Hogsett.  1988. Acid fog effects  on conifer seedlings.  In:
Proceedings of the 15th International Meeting for  Specialists in Air Pollution Effects on Forest
Ecosystems, "Air Pollution and forest Decline," October, Interlaken, Switzerland.

Unsworth, M.H. 1980.  The exchange of carbon dioxide and air pollutants  between vegetation and the
atmosphere.  In: Grace, J., Ford, E.D., and P.G. Jarvis (Eds.), Plants and their atmospheric environment,
Chapter 7,  Blackwell Sci.  Publ., Boston, pp. 111-138.

Unsworth, M.H. 1984.  Evaporation from forests in cloud enhances the effects of acid deposition.
Nature 312:262-264.

U.S. Dept.  of Agriculture, 1986.  Weekly Weather  and Crop Bulletin,  66:15-44.

U.S. Dept.  of Agriculture, 1987.  Weekly Weather  and Crop Bulletin,  74:16-44.

U.S. Dept.  of Agriculture, 1988.  Weekly Weather  and Crop Bulletin,  75:16-44.

U.S. Forest Service, 1989; Personal communication with staff of Northeastern Forest Experimental
Sation.

U.S. EPA,  1986. Air Quality Criteria for Ozone and Other Photochemical Oxidants. Vol. III.  EPA
600/8-84/020cF, U.S.  Environmental Protection Agency, U.S. Environmental Criteria and Assessment
Office, Research Triangle Park, NC.

U.S. EPA,  1988. Summary of Selected New Information on Effects of Ozone on Health and Vegetation:
Draft Supplement to Air  Quality Criteria  for Ozone and Other Photochemical Oxidants.
EPA-600/8-88/105A, U.S.  Environmental Protection Agency, Office of Health and  Environmental
Assessment, Washington,  D.C.

U.S. EPA,  1989. National Air Quality and Emissions Trends Report, 1987. EPA-450/4-89-001, U.S.
Environmental Protection Agency, Office of Air Quality Planning and Standards, Research Triangle Park,
NC.

USFS, Northeastern Forest Experimental  Station, personal communication

Valente, R., R.  Mallant, S. McLaren, R. Schemenauer and R. Stogner, 1989. Field intercomparison of
ground based cloud physics instruments at Whitetop Mountain, Virginia.  J. Atmos. and  Oceanic
Technol., 6:396-406.

Vong, R.J., Bailey, B.R., Markus, M.J. and Mohnen, V.A. (1989).  Factors  governing cloud water
composition in the Appalachian Mountains (submitted to TellusJB).
-------
Vong, R.J. et al., 1988a.  Source contributions to ambient aerosol calculated by discriminant Partial
Least Squares (PLS).  J. Chemometrics, 2:281-296.

Vong, R.J. et al., 1988b.  Changes in rainwater acidity associated with closure of a copper smelter.  J.
Geophys. Res., 93(D6):7169-7179.

Waggoner, P.E., 1969.  Environmental manipulation for higher yields. In: Physiological Aspects of Crop
Yield, J.D. Eastin, F. A.  Raskins, C.Y. Sullivan, and C. H. M. van Bavel (Eds.). ASA, Madison, WI,
pp. 343-374.

Waldman, J.M., J.W. Munger, D.J. Jacob, R.C. Flagan, J.J. Morgan, and M.R. Hoffman.  1982.  Chemical
composition of acid fog.  Science 218(12):677-679.

Waldman, J.M., and M.R. Hoffman.  1988.  Nutrient leaching from pine needles impacted by acidic
deposition.  Water, Air, Soil Pollut. 37:193-201.

Warren, S.G., C.J. Hahn, J. London, R.M. Chervin, and R.L. Jenne.  1986. Global distribution of total
cloud cover  and cloud type amounts over land.  DOE/ER/60086-H1.  U.S. Department of Energy.

Wattle, B.J., E.J. Mack, R.J. Pilie, and J.T. Hanley. 1984. The role of vegetation in the low-level water
budget in fog. Final Report (DAAG29-82-C-022).  ARVIN/CALSPAN, Buffalo, NY.

Weathers, K.C., G.E. Likens, F.H. Bormann, J.S. Eaton, W.B. Bowden, J.L. Anderson, D.A. Cass, J.N.
Galloway, W.C. Keene, KD. Kimball, P. Huth, and  D. Smiley.  1986.  A regional acidic cloud/fog water
event in the Eastern United States.  Nature 319:657-658.

Wesely, M.L. and B.B. Hicks, 1977.  Some factors that effect the deposition rates of sulfur dioxide and
similar gases on vegetation.  J. Air Pollut. Control Assoc., 27:1110-1116.

World Meteorological Organization.  1976.  Technical Regulation Standards and Recommended
Practices. 66 pp.
-------