I
                                                                  905R88102
                      FINAL SUPPLEMENTAL ENVIRONMENTAL IMPACT STATEMENT

           Wastewater Treatment Facilities  for the Columbus, Ohio Metropolitan Area




                                      Prepared by the

                        United States Environmental Protection Agency

                                          Region V

                                     Chicago, Illinois


                                           and
         Science Applications
         International Corporation

         McLean, Virginia
With
Triad Engineering
Incorporated

Milwaukee, Wisconsin
                                        August 1988
                                                            Approved by:
                                                            Valdas V. Adamkus
                                                            Regional Administrator

-------
        APPENDIX A

   BRIEFING PAPER NO. 1
WASTEWATER FLOWS AND LOADS

-------
 I
 I
 I
 •         BRIEFING  PAPER NO.  1
 I
           WASTEWATER FLOWS  AND LOADS
 I
 •         Supplemental Environmental Impact Statement
 B         USEPA Contract  No, 68-04-5035, D.O.  No. 40
 I         Columbus Ohio "Waste-water Treatment Facilities
 I
 I
 I
 I
 I
 •         Prepared  By:
 I         SCIENCE  APPLICATIONS INTERNATIONAL CORPORATION
 |          TRIAD ENGINEERING  INCORPORATED
I
I

-------
                          WASTEWATER FLOWS AND LOADS
1.   TERMS AND DEFINITIONS

2.   AVAILABLE DATA

3.   ANALYSIS OF AVAILABLE DATA

     3.1  General
     3.2  Dry Weather Flows
     3.3  Water Usage
     3.4  Wet Weather Flows

4.   EXISTING AND PROJECTED FLOWS AND LOADS

     4.1  Existing Wastewater Flows
          4.1.1  Existing Average Flows
                 4.1.1.1  Infiltration
                 4.1.1.2  Industrial and Commercial Flows
                 4.1.1.3  Domestic Flows
          4.1.2  Maximum Hourly Flows
          4.1.3  Peak Process Flow
          4.1.4  Wet Weather Flows
     4.2  Existing Wastewater Loads
     4.3  Projected Flows and Loads

5.   FACILITY PLAN METHODOLOGY

     5.1  Dry Weather Wastewater Flows
     5.2  Design Average Daily Flows
     5.3  Design Wastewater Loads
     5.4  Industrial Flows and Loads
     5.5  Projected Flows and Loads
          5.5.1  Design Flows
          5.5.2  Design Loads

6.   COMPARISON OF BRIEFING PAPER AND FACILITY PLAN FLOWS AND LOADS

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                                 INTRODUCTION



     Under the direction of USEPA,  a series of briefing papers are being

prepared addressing key issues in the development of the Supplemental

Environmental Impact Statement for the Columbus,  Ohio,  Wastewater Treatment

Facilities.  The  briefing  papers  form the basis of discussions between Triad

and USEPA  to resolve important issues.  The following paragraphs present the

background of the facility planning process, a description of the briefing

papers, and the purpose of this paper on flows and  loads.



FACILITY PLANNING PROCESS


     At the time this paper was prepared (March-August  1987) the city of

Columbus was proceeding to implement improvements at the Jackson Pike and

Southerly Wastewater Treatment Plants to comply with more stringent effluent

standards which must be met by July 1,  1988.    These improvements were based

on the consolidation of wastewater treatment operations at the Southerly

plant.   This  one-plant  alternative  is  a  change  from  the two-plant operation

proposed by the city in the  1970's and  evaluated  in the 1979 EIS.



     The development and documentation of wastewater treatment process and

sludge management alternatives for the Columbus metropolitan area has been an

extended and iterative  process.    The design and  construction of various

system components have progressed, because of the 1988  deadline,  while

planning issues continue  to be resolved.  As  a  result,  numerous  documents have

been prepared which occasionally revise a previously established course of

direction.



     The concurrent resolution of planning issues and implementation of

various project components has made preparation of the  EIS more  difficult

because final facility  plan recommendations are not available in a single

document.
                                     A-l

-------
BRIEFING PAPERS
     To facilitate preparation of the EIS,  a series of briefing papers are
being developed.   The purpose  of  the  briefing  papers  is  to  allow USEPA to
review the work of the EIS consultant and to identify supplemental information
necessary for the preparation of  the  EIS.   Six briefing  papers are being
prepared as follows:

     •  Flows and Loads
     •  Sludge Management
     •  CSO
     •  Process Selection
     •  One Plant vs. Two Plant (Alternative Analysis)
     •  O&M and Capital Costs

     The specific focus of each briefing paper will be different.  However,
the general scope  of  the  papers will  adhere  to the  following format:

     •  Existing conditions will  be documented.
     •  Evaluations,  conclusions, and  recommendations of the facilities
        planning process will be reviewed using available documentation.
     •  Where appropriate, an independent evaluation of  the future situation
        and viable alternatives will  be prepared.
     •  The facility  plan and  EIS briefing paper conclusions will be compared.

     The briefing paper process is intended to:

     •  Prompt the resolution of  any data deficiencies.
     •  Clearly establish and  define  existing and future conditions.
     •  Identify the  final recommended plan which the city  desires to implement.
     •  Provide a data base of sufficient detail to allow preparation of the
        draft EIS.
                                     A-2

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
WASTEWATER FLOWS AND LOADS


     This briefing paper presents an independent evaluation of wastewater

flows and loads which is based on an analysis  of operating records  from the

Jackson Pike and Southerly Wastewater Treatment Plants and the 1985 Revised

Facility Plan Update.  The determination of wastewater flows and loads is a

key factor in the sizing of facilities,  the evaluation of treatment alterna-

tives,  and the evaluation of solids  management scenarios.  Design flows and

loads are presented for the 20-year  planning  period which ends in 2008.  This

document is divided into six  sections.



     •   Terms and definitions

     •   Available data

     •   Analysis of available data

     •   Existing and projected flows and  loads

     •   Facility plan methodology

     •   Comparison of facility plan  and  briefing paper flows and loads
                                     A-3

-------
                          1.   TERMS AND DEFINITIONS
BOD or Biochemical Oxygen Demand:  An index of Che amount of oxygen required
for the biological and chemical oxidation of the organic matter in a liquid.

Combined Sewer:  A sewer which transports both wastewater and storm or surface
water  in a single pipe.

Commercial/Industrial  Flow:  Wastewater  flows from commercial businesses and
industry.

Design Average Flow;  The 24-hour average flow which the upgraded and expanded
treatment facilities will be sized and designed to process.

Diurnal Peaking Factor;  The factor  applied  to the design average  flow to
account for the maximum flow rate occurring  at the wastewater treatment plant
over a given  24-hour period.  The  peaking factor  is calculated as  the maximum
hourly flow rate divided by the average hourly flow rate.

Domestic Flow;  Residential sewage flow.

*Dry Weather/No Bypass Flow Condition:  Dry weather days when there were no
reported raw or settled sewage bypasses at the Southerly WWTP and no recorded
hours of operation at the Whittier Street Storm Tanks.

*Dry Weather Flow Condition:  Any day when  precipitation does not occur on a
particular day or during the day immediately  preceding  it.

Effluent;  The flow out of a process.

High Groundwater Infiltration:  Infiltration to sewers that occurs during
periods of extended  wet weather when the level of the groundwater  is high.

Infiltration;   Water other than wastewater that enters  a sewerage system,
including sewer service connections,  from the ground  through such sources as
defective pipes,  pipe joints,  connections,  or manholes.   Infiltration does not
include, and  is distinquished from,  inflow.

Inflow or RainJLnduced Flow:  Water other than wastewater that enters a
sewerage system, including sewer service connections,  from sources such as
roof leaders,  cellar drains, yard drains, area drains,  foundation drains,
manhole covers,  cross connections  between storm sewers and sanitary sewers,
catch basins,  cooling towers,  storm  waters,  surface runoff,  street wash
waters, or drainage.  Inflow does not  include, and is  distinguished from,
infiltration.
                                      A-4

-------
  I
  I
  I
  I
  I
  I
  I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
Force Main:  A sewer conduit which is pressurized by pumping.


Influent:  Flow into a process.


Low Groundwater Infiltration;  Infiltration that occurs during periods of
extended dry weather when the level of the groundwater is low.


Sanitary Sewer:  A conduit intended to carry liquid and  water-carried wastes
from residences, commercial  buildings, industrial plants, and institutions
together with minor quantities of ground, storm, and surface waters that are
not admitted  intentionally.


Storm Sewer:  A sewer designed to carry only storm waters, surface run-off,
street wash waters, and drainage.


*Wet Weather  Flow  Conditions:  Any  day (or days) on which measurable precipi-
tation occurred and the  single  day  following any day on  which precipitation
occurred.  The day  following any  day on  which  precipitation occurred is
defined as wet weather due to the lag  in  the peak rain-induced  flow  which is
seen at the plants  as  a result of in-system travel time.  Defining the
following day as wet weather also accounts for the  effect of in-line storage
following extended periods of wet weather.


*These definitions  were developed for the analysis contained in this document.
 They are not standard definitions.
                                     A-5

-------
                              2.   AVAILABLE DATA

     The 1985 and 1986 operating  records  from Che Southerly and Jackson Pike
WWTPs were reviewed  to determine existing and projected design flows and
loads.   The following plant  records were  obtained  from  the  city of  Columbus
and Ohio EPA.

     •  Monthly Operating Reports for both plants from January 1985 through
        December 1986.
     •  Monthly Report of Operations  for  the Jackson Pike WWTP from January
        1985  through December 1985.
     •  Monthly Report of Operations  for  the Southerly  WWTP from January 1985
        through September 1986.
     •  Hours of operation of the Whittier Street Storm Tanks from  January
        1985  through December 1986.
     •  Hourly flow data for both plants  for February and September 1985.
     •  1985  monthly water consumption records for the  Columbus Area.
     •  1983  Industrial Pretreatment  Report - Malcolm Pirnie.
     •  Sewer lengths and sizes  for  the Columbus Sewer  System.

     The Monthly Operating Reports (MORs) are submitted to Ohio EPA in
accordance with the NPDES permits.  Influent  flow  and load  data were obtained
from these reports.  However, these  reports did  not contain precipitation
data.  This information was  obtained  from the Monthly Report of Operations
which is submitted to  the Ohio Department of Health.

     The Southerly MORs include data  on amounts  of raw  sewage bypassed and
settled  sewage bypassed as well as treated flow.  The Southerly plant has  a
method of treatment termed Blending of Flows. When incoming flows  increase to
the point where the biological portion of the plant begins to show  signs of
potential washout, the flow  to the biological part of the plant is  fixed.   The
increase in flow above this  fixed flow, but  less  than the capacity  of the
primary tanks, is bypassed around  the biological portion and  blended  with  the
final effluent, thus, receiving  only  primary treatment and chlorination.
                                      A-6

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
These flows  are  reported on  Che MORs  as  settled  sewage  bypassed.  If the

primary  treatment  facilities are operating at capacity,  then all excess flows

are bypassed directly to the Scioto River through a 108-inch diameter pipe

originating  in the screen building.   These  flows are reported on the MORs as

raw sewage  bypassed.  After  August of  1986, no blending of flows was recorded

in the MORs  for the Southerly WWTP, however, bypassing  was  still  reported.



     The Jackson Pike MORs provide flow  monitoring  data for the plant.

Jackson Pike does not blend as Southerly does,  nor do they bypass raw sewage.

The major diversion point for Jackson  Pike flows occurs at the Whittier Street

Storm Tanks  before the  flows even  reach  the  plant.  The tanks are capable of

acting as a holding system for the excess flows until the flow  in the

interceptor subsides and they can be  bled back into  the system and carried to

the Jackson Pike plant.   If  the flows exceed the capacity of the tanks, they

overflow to  the  Scioto  River.  Flows  can also be directly bypassed along side

the tanks,  through an emergency bypass,  to  the Scioto River.



     Flow monitoring did not take place at the  Whittier Street Storm Tanks

until November of  1986.   However, hours  of operation of the storm tanks were

recorded during 1985 and 1986 on the  Monthly Report of Operations.  The fact

that hours of operation were reported does not necessarily mean there was

bypassing or overflowing occurring at  the tanks.   It only means that the gates

were open and flows were being diverted  into  the  tanks.  In November of 1986,

the city  began monitoring the overflow  but not  the bypass.  Therefore, the

data is still incomplete with respect to determining the total volume of flow

entering the Scioto River at the  Whittier Street  facility.



     Hourly flow data was obtained for February and September of 1985 for both

plants.   These months represent the periods of minimum and maximum water

consumption  respectively.   This hourly flow data was used to determine a

diurnal peaking  factor which is calculated by dividing  the  peak hourly  flow by

the average hourly flow. This  diurnal peaking  factor  is multiplied by the

design average  flow to determine  a  peak hourly  flow  for  use  in sizing the  wet

stream treatment facilities.
                                     A-7

-------
     Dry weather flows were determined  through an analysis of 1985 and 1986
flow data.   However,  only 1986  flow data was  used  to determine wet  weather
flows.  An analysis of 1985 MORs showed that data  on  raw and settled sewage
bypasses at Southerly were not complete.  Up until August of 1985,  only a
bypass flow rate (MGD) was reported with no duration specified.    These
bypasses did not always occur 24 hours a day, therefore, these  rates could not
be converted to the volume bypassed during that day.   In August  of 1985,
monitoring of the duration of  the bypasses began which  provided  a  more
accurate determination of the volume of the bypasses.    Therefore,  the 1986
calendar year data  were used to estimate wet  weather  flows.

Wet weather total system  flow can not be determined solely based on the volume
of flow arriving at the Jackson Pike and Southerly WWTPs.  There are numerous
points of combined sewer  overflow throughout the Columbus Sewer  System. The
Jackson Pike service  area has several  regulator chambers and  overflow
structures in addition to the  Whittier Street Storm Standby Tanks  discussed
previously.  The Southerly service  area  includes an overflow structure at
Roads End and the Alum Creek Storm Standby Tank.   There is no comprehensive
flow monitoring data  available for the  regulators, overflows, and  storm tanks.
The city began monitoring the overflows at the Whittier Street  facility in
November of 1986.  However,  they  did not monitor  the  bypass  line at the
Whittier Street facility.  The city also began monitoring some  of  the  other
points of combined sewer  overflow;  but according to the  MORs, the  flow
monitoring equipment  malfunctioned  frequently which provided no data.   Thus,
the only flow data included  in  the  wet weather analysis,  other than plant flow
data, was  that which was   reported for  the Whittier Street overflow during
November and December.

     The Industrial Pretreatment Report  prepared by Malcolm Pirnie in  1983 was
used to estimate the  industrial and commercial flows.   This  report quoted
figures on industrial and commercial flows based on 1980 water and sewage
records.  Due to the  lack of  more recent quantification of industrial  and
commercial flows,  these  figures were updated for  this  document  using 1985
population figures.
                                     A-8

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
                        3.  ANALYSIS OF AVAILABLE DATA


     The following sections present an analysis of the wastewater flow data.

This analysis was developed independently of that presented in the facility

plan as a check of  the  assumptions and methodologies.    It  was prepared using

Monthly Operating Reports (MORs) for the Jackson Pike and Southerly  WWTPs and

precipitation data and water usage records for the city of Columbus.   Using

these records,  wet weather and dry weather flows were developed for each

plant.   Dry weather flows were compared to water consumption data to aid in

the interpretation of monthly flow variations.



3.1  GENERAL


     Jackson Pike and Southerly MORs and precipitation data for the 1985 and

1986 calendar years were used  to establish existing wastewater flows.   The

following sections will discuss existing wet weather and dry weather flows.



     In order to determine wet and dry weather flows,  each  daily  record was

categorized accordingly.  Wet  weather was defined as any day on which

measurable precipitation occurred and the  single  day  following the last day on

which precipitation occurred.   The day following one on which precipitation

occurred is defined in  this analysis  as  wet  weather due to  the lag in  the peak

rain induced flow which  is seen  at the  plants.  This  lag is a result of in-

line storage and in-system travel time.  The  remainder of the daily  records

were categorized as dry weather.  Weather conditions for 1985 and 1986 are

summarized in Table 3-1 using  these classifications.   There were  a total  of

144 days in 1985 and 130 days  in 1986 on which measurable precipitation

occurred.   Wet  weather days totaled 212 for  1985 and 197 for 1986.   There were

153 dry weather days for  1985 and  168 for 1986.
                                     A-9

-------
                    TABLE 3-1.  WEATHER CONDITION SUMMARY


            Precipitation*    Days of Measurable     Wet Weather   Dry Weather
Month          (inches)      Precipitation (Count)   Days (Count)  Days  (Count)

1985

January          1.26                  16                 21           10
February         1.67                  12                 17           11
March            3.78                  17                 24             7
April            0.56                  11                 17           13
May              4.96                  12                 17           14
June             1.41                  12                 20           10
July             6.88                   5                  7           24
August           2.34                  10                 16           15
September        1.18                   4                  6           24
October          1.98                  11                 18           13
November        10.67                  21                 28             2
December         1.81                  13                 21           10

TOTAL           38.50                 144                212           153

1986

January          1.54                  12                 16           15
February         2.96                  16                 22             6
March            2.61                  11                 17           14
April            1.31                  13                 21             9
May              2.47                  13                 19           12
June             5.53                  11                 17           13
July             3.60                   8                 13           18
August           1.61                   6                 11           20
September        3.44                   8                 13           17
October          4.16                   9                 13           18
November         3.00                  11                 18           12
December         2.81                  12                 17           14

TOTAL           35.04                 130                197           168
* Measured  at Port  Columbus  Airport
                                        A-10

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
3.2  DRY WEATHER FLOWS
     Construction Grants 1985,  which is the USEPA guide for the preparation of
facility plans,  recommends that design flows for  treatment works be  determined
based on existing base flow;  estimated future flows  from residential,  com-
mercial, institutional and industrial sources; and nonexcessive I/I.
Traditionally, base flows are established using dry  weather flow.

     Dry weather days were classified as indicated in  the previous section.
By definition, they include any day on which measurable precipitation does not
occur that day or during the  day immediately preceding it.  In applying this
definition to plant data, it  was found that bypasses occurred in the  system on
several days which would be categorized as  dry weather.  Bypasses  are
monitored at  the Southerly WWTP and reported in the  records as settled sewage
bypassed and raw sewage bypassed.   The Jackson  Pike WWTP does  not  bypass  at
the plant.   However,  when flows increase beyond plant capacity, the  gates are
opened at the Whittier Street Storm Tanks and flows  are diverted to  the tanks
before they reach the Jackson Pike  WWTP.   When the  gates are open at the
Whittier Street facility, it  is considered  to be  in  operation.   Flows  diverted
through the Whittier Street Storm Tanks  were not  monitored until November of
1986,  but the hours of operation of the  storm tanks  are reported on  the
Jackson Pike WWTP  records.  Days with reported hours  of operation  were
considered as bypass  days.

     Closer  examination of  the days  with reported  bypassing and storm tank
hours showed  that  the majority occurred after  an extended  wet weather period.
Those that did  not follow an extended wet  weather  period were assumed  to be
related to operational problems at  the  plant.  Therefore,  in  establishing dry
weather flows, only dry  weather/no bypass days were considered.

     Using the classification of dry weather/no bypass, monthly average flows
were determined for the  1985 and 1986 calendar year.   These  flows  are
presented in Table 3-2.   In evaluating  these  flows,  the 1985  and 1986  averages
for each plant were very  close.  The maximum  and minimum combined values  both
                                     A-ll

-------
    TABLE 3-2.   DRY WEATHER/NO BYPASS MONTHLY AVERAGE FLOWS (MGD)
Month
1985
January
February
March
April
May
June
July
August
September
October
November
December
TOTAL
AVERAGE
1986
January
February
March
April
May
June
July
August
September
October
November
December
TOTAL
AVERAGE
Count

7
5
2
3
8
9
19
11
24
11
0
5
104


10
0
1
8
11
10
13
20
17
10
2
8
110

Jackson Pike

75.86
79.20
82.00
81.30
83.88
78.89
80.47
75.18
73.38
72.31
ND
81.62

77.27

78.53
ND
80.73
82.52
76.66
80.33
81.32
77.13
75.87
78.08
70.30
79.06

78.33
Southerly

56.44
60.74
60.55
58.92
60.88
55.14
58.40
51.85
50.64
52.57
ND
61.54

55.40

56.23
ND
62.50
57.69
48.21
58.34
55.79
55.74
55.18
53.25
54.30
60.98

55.52
Combined

132.30
1 39 . 94
142.55
140.22
144.76
134.03
138.87
127.03
124.02
124.88
ND
143.16

132.67

134.76
ND
143.23
140.21
124.87
138.67
137.12
132.87
131.05
131.33
124.60
140.04

133.85
ND - No dry weather/no bypass days
                                     A-12

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
occurred in 1985.   The dry weather combined maximum monthly average of 145  MGD

occurred in May of 1985,  and  the dry weather combined minimum monthly average

of 124 MGD occurred in September of 1985.



     The maximum  monthly average dry weather  flow of 145 MGD which occurred in

May is considered to represent a  high  groundwater condition due to the large

amount of precipitation  and extended wet weather periods in this month.  It

had the second highest monthly precipitation  for 1985 of 3.92 inches.   The

highest occurred in November, but there were  no dry weather/no bypass days  in

November.  May had eight  dry  weather/no bypass days  which occurred during two

4-day  periods.



     The minimum  monthly average dry weather  flow of 124 MGD which occurred in

September of 1985 is considered to represent a low  groundwater condition due

to the extended dry period  which occurred during that month.  September of

1985 had 24  dry weather/no  bypass days  which  occurred during one 22-day period

and one 2-day period.  This was  the  highest  number of dry weather/no bypass

days recorded in one month  for the  24  month (1985 and 1986)  data base  that  was

evaluated.



     The 1985 flows closely approximate the 1986  dry weather/no bypass maximum

monthly average of 143 MGD  and the minimum monthly average  of 125 MGD.
                                                  >•


3.3  WATER USAGE


     Information on water usage for  the Columbus area was obtained from the

Columbus Division of Water  -  1985 Annual Report.  These flows  were evaluated

to gain further insight  into  the  groundwater  condition.  The total amount of

water pumped to residential,  commercial, and  industrial customers in the

Columbus area during the  1985 calendar year was 44 billion gallons.  Using

the 1985 population figure  of 870,000  people,  developed by  Ohio Data Users

Center, the  water usage   figure was converted  to 139 gallons per capita per  day

(gpcd).
                                     A-13

-------
     The L985 average dry weather/no bypass flow of 136 HGD from Table 3-2 can

be converted Co 156 gpcd using the population figure for 1985 of 870,000
persons.  The per capita water pumpage (139 gpcd) value is 17 gpcd or approxi-
mately  12 percent less than the wastewater  (156  gpcd) value.   This 17 gpcd

difference may be the result of high infiltration in the sewer system, not all
the sewer customers being water customers, or a result of illegal connections
to the sewer system.


     Table 3-3 compares the monthly average water pumped to the Columbus area

vs.  monthly average  dry  weather/no bypass  wastewater flows.   The  table shows  a
higher wastewater flow  than water  pumpage  for the spring months.  This could

be due to more sewer customers than water customers as discussed in the

previous paragraph.   However,  it  could  also be  a  result of a  greater  amount of

infiltration from a high groundwater condition.  September, on the other hand,

which had 24 dry weather/no bypass days had an average water  pumpage figure

17.72  MGD greater than the  wastewater  figure.   The  high  water purapage figure

could be attributed to lawn sprinkling due to the extended dry period.  The

low wastewater flow indicates that less infiltration is entering the system,

which is a result of a low groundwater condition.


              TABLE 3-3  1985  WATER PUMPAGE VS.  WASTEWATER FLOW
          Month

        January
        February
        March
        April
        May
        June
        July
        August
        September
        October
        November
        December
Average Water
Pumped (MGD)

   111.23
   108.32
   109.65
   115.60
   120.33
   128.53
   127.15
   130.66
   141.74
   124.88
   117.23
   116.46
Average Dry Weather/
No Bypass Flow (MGD)

      132.30
      139.94
      142.55
      140.25
      144.75
      134.03
      138.87
      127.03
      124.02
      124.88
        ND
      143.16
                                     A-14

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
 I
I
3.4  WET WEATHER FLOWS


     A limited data base was reviewed with respect to wet weather flows.

Insufficient data was available to quantify the total wet weather  flow  for  the

entire Columbus system.



     The only data evaluated in determining wet weather  flows  was  that  which

was reported from monitoring flows arriving at the plants  from January  through

December 1986 and data reported from  monitoring overflows  at the Whittier

Street Storm Tanks during November and  December of 1986.  The  flow data

collected at the Southerly and  Jackson  Pike  plants is the  only data that was

collected for an entire year.   Flow data was  collected at  the  overflow  located

at the Whittier Street Storm Tanks during November and December of 1986.

However, no flow data was gathered from the  bypass at Whittier Street.  From

October through December of 1986,  flow  monitoring was performed at various

other overflows and regulators  within the  Columbus combined sewer  system.

However, it was never performed at all  the points  of  combined  sewer overflow

during the same month, and according  to the MORs,  the flow  monitoring

equipment malfunctioned  frequently.



     Wet weather days  were  categorized as discussed  in Section 3.1.  Wet

weather being defined as any day on which  measurable  precipitation occurs and

the single day immediately following  any day  on which measurable precipitation

occurs.



     A total flow  was calculated  for each wet weather day  during  1986.   This

total flow includes the following:



     •  Southerly treated sewage.

     •  Southerly settled sewage bypassed.

     •  Southerly raw sewage bypassed.

     •  Jackson Pike treated sewage.

     •  Whittier Street sewage  overflow volumes (November and  December only).
                                     A-15

-------
     Table 3-4 shows Che average  and  maximum daily wet weather flows for
January through December of 1986.  As shown  in the table, the maximum  wet
weather flow  of 309.52 MGD occurred in March.  The actual day was March 14,
1986.   On March 12, the reported precipitation was 0.71 inches and 0.51 inches
was reported for March 13.  It must be  remembered that this flow only
includes the flow arriving at the plants.  It does not include any bypassing
that may have occurred at the numerous  points of  combined sewer overflow
throughout the system.

     Wet weather flows are discussed in more detail in the CSO briefing paper.

                      TABLE 3-4.   WET WEATHER FLOW DATA
                         Wet Weather
Maximum
Month
January
February
March
April
May
June
July
August
September
October
November
December
Total
1986
Days (Count)
16
22
17
21
19
17
13
11
13
13
18
17
197

Daily Average (MGD)
165.31
298.62
309.52
155.02
160.98
227.60
184.29
158.61
165.23
266.00
223.73
294.24

309.52
Average (MGD)
147.73
183.01
181.62
143.68
137.46
152.45
154.75
137.00
147.27
161.47
149.80
178.64

156.24
                                     A-16

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                  4.  EXISTING AND PROJECTED FLOWS AND LOADS



     This chapter describes the development of average daily and peak hourly

flow rates and daily  loadings  of TSS (total suspended solids) and BOD which

are used to evaluate facility planning  recommendations.  The following sections

present the existing flows and loads developed for the Columbus WWTPs from an

independent analysis of the 1985 and 1986 plant data, as well as projected

flows and  loads  for the 2008  design year.



     An analysis of existing conditions established the current average day

flows.  This current condition is subsequently dissagregated into domestic,

infiltration, industrial,  and commercial flows.  A diurnal peaking factor and

a process peaking factor are established to project peak flow rates which will

be used in sizing some of  the  WWTP unit processes.  Wet  weather flows are

discussed briefly with a more  detailed  discussion  included  in  the  CSO  packet.



     The analysis also includes a review of existing influent BOD and TSS

loads.    BOD and  TSS loads are used to  determine sizings for WWTP unit

processes and to aid in the selection of the alternative treatment processes.



     Wastewater  flows and  loads  are  projected  for  the design year  (2008) using

existing per capita flows  and  loads and 2008 population projections.



4.1  EXISTING WASTEWATER FLOWS


     This section presents the existing average flow,  maximum hourly flow,

peak process flow,  and wet weather flow as determined from analysis of

available data.



4.1.1  Existing Average Flows


     According to USEPA guidelines,  WWTP design flows are determined based on

existing dry weather flows and non-excessive I/I.  As discussed in Section

3.2,  dry  weather flows were determined based on a dry weather/no bypass

condition.   Therefore,  the existing average  flow  was  determined through an
                                     A-17

-------
analysis of dry weather/no bypass flows.  The 1985 and 1986 combined Jackson
Pike and Southerly maximum monthly average dry weather/no bypass  flow from
Table 3-2 was selected.  This combined flow of  144.76 MGD occurred in May of
1985 and was based on 83.88 MGD for Jackson Pike and 60.88 MGD for Southerly.

     In subsequent paragraphs,  this flow of 145 MGD is further broken down into
infiltration, industrial, commercial, and domestic  flows.   In  Section 4.3,
population projections are used to increase this flow for  the  design  year.

4.1.1.1  Infiltration
     No current infiltration/inflow report was available for the  Columbus
sewer system; therefore,  wastewater flow, water use, and precipitation data
were evaluated to estimate  infiltration.

     The maximum monthly average dry weather/no bypass flow of 145 MGD
occurred in May of 1985.   The data base  consists of  two 4-day periods of dry
weather/no bypass conditions.  This month, which had 3.92 inches  of
precipitation, had the second highest monthly rainfall recorded during  1985.
Therefore,  it is  safe to  assume  that May would represent  a high groundwater
condition resulting in increased infiltration. November had the highest
precipitation with 10.67  inches, but  there  were  no dry weather/no bypass days
during  that month.

     September of 1985 had the  lowest combined monthly average dry weather/no
bypass  flow of 124.02 MGD for the  1985 and 1986 calendar years; and it had
24-dry  weather/no bypass days which occurred  in one 2-day period and one
22-day  period.  Due  to the extended dry  weather period,  it is assumed to
represent a low groundwater  condition.  Water usage figures presented in
Section 3.3  reinforce  May and September  as representing high and  low
groundwater conditions.  The difference  of 20.74 MGD between the  high
groundwater month (May) and the  low groundwater month (September) represents
that portion of the total infiltration which  is attributable to a high
groundwater condition.  However, this is only a portion of the  total  amount of
                                      A-18

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
infiltration occurring since there is also some infiltration occurring during

low groundwater  conditions.  Therefore, the amount of infiltration occurring

during low groundwater conditions  must be  determined and added  to  the 20.74

MGD in order to  establish a total  infiltration  rate.



     In the absence of a current infiltration/inflow report other methods of

estimating infiltration must be used.  A common method involves  using monthly

water records  to establish the domestic, commercial, and industrial portion of

the wastewater flow.  The remainder of the  wastewater flow is then assumed to

be infiltration.



     Since September 1985 has been established as a low groundwater month,

water usage rates from this month will be used.  As reported in  Table 3-3, the

September 1985 water pumpage  rate  is 141.74 MGD.  Literature states that

approximately 60 to 80 percent of  water becomes wastewater.  The 20 to 40

percent which is lost includes water consumed by commercial and  manufacturing

establishments and water used for street washing,  lawn sprinkling,  and

extinguishing fires.   It  also  includes water used by residences  that are not

connected to  the  sewer  system  as well as some leakage from water mains and

service pipes.   If it is  assumed that 70 percent of the  water becomes

wastewater, then the return flow for September would be 99.22 MGD.   Referring

to Table 3-3,  the wastewater flow  for  September  is  124.02 MGD.  The difference

between the actual wastewater flow (124.02) and the expected wastewater  flow

(99.22) is 24.80 MGD.   This value  is  assumed  to represent  the  amount of

infiltration occurring during  a low groundwater condition.  Thus,  the total

infiltration occurring during  high groundwater conditions is obtained by

adding 20.74  MGD to 24.80  MGD.   This total  infiltration figure of 45.54  MGD,

converts to 52 gpcd.



     It must be remembered that 52 gpcd is only a rough  estimate of

infiltration.   It is not known if all of the water customers  are sewer

customers or if all the sewer  customers are water customers.   Some  sewer

customers may have their  own private wells.  In addition,  the  consumptive use

of the brewery and the other industries is  unknown.
                                     A-19

-------
     It is,  however,  considered  to be  a non-excessive  infiltration  rate  when
compared to infiltration rates in the  USEPA document entitled Facility
Planning - 1981 Construction Grants Programs.   This  document states that 2000
to 3000 gpd/inch-diaraeter mile is considered a non-excessive infiltration rate
for sewer systems with  lengths  greater than 100,000 feet.   The Columbus  Sewer
System has a total length of 9,975,000  feet  which converts  to an  estimated
32,930 inch-diameter miles.  Multiplying the inch-diameter miles by 2000
gpd/inch-diaraeter mile  results  in  66 MGD or 76  gpcd.  Therefore,  52 gpcd of
infiltration would be considered non-excessive.

     The Revised Facility Plan Update  uses  a peak infiltration rate of 72
gpcd.   Divided  between the two plants,  it  is 82 gpcd for Jackson  Pike and 58
gpcd for Southerly.   Assuming more detailed  information  was available to
establish this  number for the facility plan and considering 72 gpcd is also a
non-excessive infiltration rate  according to the USEPA document,  it will be
used in this briefing paper as the existing infiltration rate.  It  converts to
22.1 MGD for Southerly and 40.1  MGD for Jackson  Pike,  totaling 62.2  MGD  for
the entire Columbus  Sewer System.  This number  will be held constant through-
out the planning  period.

4.1.1.2  Industrial and Commercial Flows
                                                   »•
     Current information on industrial and  commercial wastewater flows was not
available.   Therefore, estimates were  made  by  updating those values presented
in the Columbus Industrial Pretreatment Program Report as prepared by Burgess
and Niple.   The Burgess  and Niple values were  updated proportional  to the
increase in population from  1980 to 1985 since  they were based on 1980 water
consumption records.   The 1985 Estimates of industrial and commercial flows
are presented  in Table  4-1.
                                      A-20

-------
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1


TABLE 4-1. INDUSTRIAL AND COMMERCIAL FLOW ESTIMATES



1980 1980 1985 1985
1980 Industrial Commercial 1985 Industrial Commercial
Population Flow (MGD) Flow (MGD) Population Flow (MGD) Flow (MGD)
Jackson Pike 472,503 8.7 4.3 489,000 9.0 4.
Southerly 368,228 6.7 3.1 381,000 6.9 3.
TOTAL 840,731 15.4 7.4 870,000 15.9 7.
The analysis of variations in the dry weather/no bypass flows between
weekdays and weekends gives an indication of the magnitude of the industrial
and commercial flows. Table 4-2 presents a summary of the weekly flow
variations for the two plants.
TABLE 4-2. 1985 DRY WEATHER/NO BYPASS WEEKLY FLOW VARIATIONS (MGD)
Jackson Pike Southerly TOTAL
Weekday 78.71 55.37 134.08
Weekend 73.80 54.92 128.72
Difference 4.91 0.45 5.36
% Difference 6.2 0.8 4.0
From Weekday
Referring to Table 4-1, it can be seen that the total commercial and
industrial flow for Jackson Pike in 1985 is 13.5 MGD. Relating this to the
4.91 MGD difference in flow between weekdays and the weekend, suggests that
approximately 35 percent of the flow from commercial and industrial sources
in the Jackson Pike service area is from sources which operate on a weekday
schedule. Southerly, on the other hand, with 10.1 MGD industrial and
commercial flow, appears to have only 4 percent of its industrial and
commercial contributing flow sources operating on a weekday schedule.



A-21
5
2
7













-------
4.1.1.3  Domestic Flows
     Domestic flows were estimated simply by subtracting infiltration,
industrial,  and  commercial flows from the maximum dry weather/no bypass  flow
of 145 MGD.   The Jackson Pike domestic flow is  30.4 MGD and Southerly is 28.8
MGD.   Table  4-3  presents  the  breakdown of  the  existing flow for each plant  and
the two plants combined.

                       TABLE 4-3.  1985 ESTIMATED FLOWS

                            Jackson Pike       Southerly        Total
Design Average                   84              61             145
Flow (MGD)
     •  Infiltration           40.1            22.1            62.2
     •  Industrial              9.0             6.9            15.9
     •  Commercial              4.5             3.2             7.7
     •  Domestic               30.4            28.8            59.2

     4.1.2  Maximum Hourly Flow
     Just as demand for water fluctuates on an hourly basis,  so do  wastewater
flow rates.   Fluctuations observed  in wastewater flow rates tend to follow  a
diurnal pattern.  (See Figure 4-1.)   Minimum  flow usually occurs in the  early
morning hours when water use is low.   The  flow rates start to increase at
approximately 6 a.m.  when people are  going to  work, and they reach a peak
value around 12 noon.   The flow rate usually drops off in the early afternoon,
and a second peak occurs  in  the early evening  hours  between 6 p.m.  and  9 p.m.
In general,  where extraneous flows are excluded from the  sewer system, the
wastewater  flow-rate  curves  will closely follow  water-use  curves.  However,
the wastewater curves will be displaced by a  time period  corresponding to the
travel time in the sewers.
                                      A-22

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
 I
I
               d     -  0-
                                             m
                                             o
                                             »—i
                                             E-
                                             ^3
                                             §
                                             3
                                      u
                                      u
                                      e.
                                      U
                                      U
                                      c
                                      o
                                      u

                                      o
                                      •3
                                      c

                        
                                      C
                                      u
                                      u
                                      o
                                      I/I

                                      c
                                      a
                                      u

-------
     Diurnal curves are also affected by Che size of the community.  Large
communities with more industrial and commercial flows tend to have flatter
curves due to industries that operate on a 24-hour schedule,  stores and
restaurants that are open 24 hours a day,  and to the expansiveness of the
collection systems.  These 24-hour operating schedules also result in more
people working second and third shift, thus altering normal flow  patterns.
Longer travel times in the collection system dampen peak flows observed at the
WWTP.

     An existing average  flow  of  145  MGD  was determined in Section 4.1.1.
This flow was determined from average dry weather flows and it is generally
used in the design of wastewater facilities to determine quantities of
chemicals needed,  O&M costs, labor, and  energy requirements.   However,  the
peak hourly flow must be  used  for  hydraulic sizing of pumps.  Therefore, a
diurnal peaking factor must be  determined  and applied to the design average
flow to provide a peak hourly design flow.

     Figure 4-2 presents  wastewater flow rate curves for the Jackson Pike and
Southerly plants compiled from September  1985 dry weather/no bypass days.  The
diurnal peaking factor was determined for  the  Jackson Pike  and  Southerly WWTPs
through an analysis of hourly wastewater flows for February and September
1985.  These  two months represent minimum  and  maximum water consumption,
respectively for 1985.  The  1985 months  were chosen since the existing average
flow occurred in May of 1985.  Diurnal peaking factors were calculated by
dividing the maximum hourly  flow  by the average hourly flow for each dry
weather/no bypass day during February and September.  These values are listed
in Tables 4-4 and 4-5.

     The maximum  diurnal  peaking  factor seen  at Jackson Pike during this
period was 1.40,  and at Southerly  it  was  1.51.  Jackson Pike's value of  1.40
occurred several times and was selected as  the diurnal peaking factor for
Jackson Pike.  Southerly's  maximum value of 1.51,  however,  was considered to
be excessive.  It  occurred,  only once, on  September  21 when the average hourly
flow was at a low of 45 MGD.  The  next peaking factor in the series was  1.37
                                     A-24

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
                      CXJ

                           on
                           O
H
D
0
                                z
                                O
                      co
(QDW) 3lViJ MOU

         A-25
                                         W
                                         O
                                         I
                                         §«
                                         59


-------
TABLE 4-4.  HOURLY FLOW DATA SOUTHERLY WWTP
Average Hourly Peak Hourly
Date
2/4/85
2/8/85
2/9/85
2/16/85
2/17/85
2/18/85
2/19/85
2/20/85
2/26/85
2/27/85
2/28/85
9/1/85
9/2/85
9/3/85
9/4/85
9/5/85
9/6/85
9/7/85
9/8/85
9/9/85
9/10/85
9/11/85
9/12/85
9/13/85
9/14/85
9/15/85
9/16/85
9/17/85
9/18/85
9/19/85
9/20/85
9/21/85
9/22/85
9/28/85
9/29/85
* Peaking
Flow (MGD)
56.2
55.8
54.6
69.5
67.6
69.0
71.2
75.0
87.1
81.3
82.5
48.8
49.4
53.1
52.2
52.5
50.4
51.3
49.0
50.7
53.2
53.9
51.9
40.1
54.0
49.8
52.0
51.2
52.2
51.8
49.9
45.0
51.3
51.3
50.3
Factor = Peak
Flow (MGD)
58.0
63.0
65.0
83.0
79.0
78.0
78.0
81.0
97.0
90.0
85.0
56.0
62.0
62.0
59.0
57.0
57.0
60.0
56.0
54.0
57.0
61.0
59.0
55.0
64.0
57.0
64.0
55.0
58.0
57.0
58.0
68.0
68.0
62.0
59.0
Hourly Flow
Peaking*
Factor
1.03
1.13
1.19
1.19
1.17
1.13
1.10
1.08
1.11
1.11
1.03
1.14
1.26
1.17
1.13
1.09
1.13
1.17
1.14
1.07
1.07
1.13
1.14
1.37
1.19
1.14
1.23
1.07
1.11
1.10
1.16
1.51
1.33
1.21
1.17

Weather
Condition
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY

  Average Hourly Flow
                A-26

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
I
I
TABLE 4-5.  HOURLY FLOW DATA JACKSON PIKE WWTP

Date
2/4/85
2/8/85
2/9/85
2/16/85
2/17/85
2/18/85
2/19/85
2/20/85
2/26/85
2/27/85
2/28/85
9/1/85
9/2/85
9/3/85
9/4/85
9/5/85
9/6/85
9/7/85
9/8/85
9/9/85
9/10/85
9/11/85
9/12/85
9/13/85
9/14/85
9/16/85
9/17/85
9/18/85
9/19/85
9/20/85
9/21/85
9/22/85
9/28/85
9/29/85
* Peaking
Average Hourly
Flow (MGD)
76.0
73.0
69.0
91.0
87.0
92.0
91.0
92.0
98.0
95.0
99.0
69.3
72.0
76.9
81.0
81.3
79.5
75.6
72.1
78.7
79.0
76.0
71.7
73.7
69.3
72.4
72.3
73.4
72.9
72.6
70.0
67.0
70.6
68.5
Factor = Peak Hourly
Peak Hourly
Flow (MGD)
94.0
96.0
89.0
102.0
106.0
102.0
98.0
103.0
106.0
104.0
102.0
86.0
89.0
104.0
96.0
95.0
94.0
94.0
92.0
96.0
92.0
90.0
85.0
86.9
96.8
88.0
96.0
92.0
88.8
89.0
90.0
94.0
99.0
83.0
Flow
Peaking*
Factor
1.24
1.32
1.29
1.12
1.22
1.11
1.08
1.12
1.08
1.09
1.03
1.24
1.24
1.35
1.19
1.17
1.18
1.24
1.28
1.22
1.16
1.18
1.19
1.18
1.40
1.22
1.33
1.25
1.22
1.23
1.29
1.40
1.40
1.21

Weather
Condition
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY
DRY

   Average Hourly Flow
                 A-27

-------
which is more representative of the maximum diurnal peaking factor seen at the
Southerly plant.   Thus,  1.4  was  chosen  as  a  representative  diurnal  peaking
factor for both plants.

4.1.3  Peak Process Flow
     A peak process flow must be developed for use in sizing the  various  wet
stream processes.   This  flow establishes the maximum process capability of the
wet stream treatment  facilities.  Flows greater than the peak process  flow
will cause the treatment facilities to  operate beyond their intended design
criteria.   Sustained operation above  the peak  process  flow  may result  in  a
violation of permit limits.

     The peak process  flow is most reliably established through an analysis of
existing flow.  This approach was not possible  in  the  Columbus system  due to
the nature of the  flow record.   As  discussed  in Section 2,  the flow records
for the  two Columbus plants  provided  limited  information regarding  the amount
of sewage bypassed. As  a  result a  reliable  record of the total flow arriving
is not available.   Furthermore,  peak wastewater flows normally include some
combined sewage.  A combined sewage overflow study, which will define  a CSO
control strategy,  is currently  being  prepared  by the city.   The impact of the
CSO recommendation on the  wastewater treatment  facilities will be evaluated at
the conclusion of  that study.

     In the 1979 EIS,  the  following empirical  formula was utilized to  develop
a peak process flow, due to  the absence of a comprehensive  flow record.

     Peak Process  Flow = 1.95 (Average  Daily Flow) °'95

     Lacking flow information which would substantiate a peak process  flow,
the 1979 EIS formula provides a reasonable method  for developing a peak
process flow.  Based on  the  2008 average design  flow  of  154  MGD,  the formula
yields a peak process flow of 233 MGD.   This corresponds to a process  peaking
factor of  1.5.
                                      A-28

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 i
 I
 I
 I
 I
 I
I
 I
I
     The 1.5 process peaking factor was evaluated relative to the 1986
available flow data  to assess  the  extent of  its  range.  The 1986 flow record
includes flows treated  at Jackson Pike and  Southerly and also the flows which
are bypassed at Southerly.  The flow  record does  not  include  flows which were
bypassed at Whittier Street or any other combined sewer overflows.  The  1986
average flow of  the  two plants was 145 MGD.   Applying  the  1.5 process peaking
factor to this average flow yields a  peak process flow of 218 MGD.  Comparing
this flow with the 1986 record indicated that the daily flow rate of 218 MGD
was exceeded only nine days during the year  or approximately  2.5 percent of
the time.   In light of these few exceedances,  the  1.5 process peaking factor
established by the 1979 EIS provides a reasonable approach to establish a peak
process flow.

4.1.4  Wet Weather Flow
     The maximum monitored wet weather flow  as determined  from  1986 records
and discussed in Section 3.4  is 309.52  MGD.   This flow occurred on March  14.
It includes 95.57 MGD for  the Jackson Pike  WWTP  and  213.95 MGD for the
Southerly WWTP.  The Southerly flow can be broken down into 78.05 MGD
receiving complete treatment,  30.30 MGD receiving primary treatment and
chlorination, and 105.60  being  bypassed directly  to the Scioto River.   Note
that this maximum wet  weather  flow only includes flow  that arrives at the
treatment plants.  Any flow being bypassed  at the various  points of combined
sewer overflow is not  included.

4.2  WASTEWATER LOADS
     Monthly average influent  TSS  (total suspended solids) and BOD (biochemical
oxygen demand) loads  were  determined for all  weather conditions.  These loads
are presented in Tables 4-6 and 4-7.

     The sampling point at Jackson Pike for TSS  and BOD concentrations is
located at  the grit  chambers on the  O.S.I.S.   Therefore, the  samples  do not
represent the flow  from  the  Big Run Interceptor.  The O.S.I.S. carries in
                                     A-29

-------
TABLE 4-6.  1985 AVERAGE BOD LOADS (Ib/day)

Month
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul.
Aug.
Sep.
OcC.
Nov.
Dec.
ANNUAL
Jackson Pike
BOD
118,466
109,094
104,532
97,918
97,831
109,632
94,384
93,591
88,619
104,161
96,483
92,466
100,702
Southerly
BOD
91,187
82,506
82,819
87,777
89,108
85,513
84,649
86,073
98,992
105,446
76,140
76,992
87,258
Total
BOD
209,653
191,600
187,351
185,695
186,939
195,145
179,033
179,664
187,611
209,607
172,623
169,458
187,960
                      A-30

-------

TABLE 4-7.  1985 and 1986 MONTHLY AVERAGE TSS LOADS (Ib/day)
      Jackson Pike
Southerly
Total
Month
Jan.
Feb.
Mar.
Apr.
May
Jun.
Jul.
Aug.
Sep.
Oct.
Nov.
Dec.
ANNUAL
1985 TSS
120,331
121,223
136,509
110,170
136,038
158,045
153,317
126,033
114,192
121,086
148,916
105,969
129,347
1986 TSS
115,923
120,583
129,050
124,532
133,613
139,516
113,282
108,853
129,688
139,653
112,099
104,965
122,665
1985 TSS
99,391
108,739
107,085
106,911
108,516
99,145
105,571
91,308
95,424
93,693
99,165
97,948
101,042
1986 TSS
87,633
91,508
94,313
92,109
89,700
95,078
93,421
100,996
101,437
100,830
88,952
86,313
93,535
1985 TSS
219,722
229,962
243,594
217,081
244,554
257,190
258,888
217,341
209,616
214,779
248,081
203,917
230,389
1986 TSS
203,556
212,091
223,363
216,641
223,313
234,594
206,703
209,849
231,125
240,483
201,051
191,278
216,200
                            A-31

-------
approximately 65  to 70 percent of  the  flow  at  Jackson Pike.  Plant staff
believe that the  flow arriving through the O.S.I.S.  contains the majority of
the industrial flow in the Jackson Pike service  area.  Samples  taken from the
O.S.I.S.  have always been used to establish  waste loads for the total flow  to
Jackson Pike.  The Southerly flow  is sampled between the screens  and the grit
chambers.  Thus,  the  samples are representative of 100 percent of the flow
entering the Southerly plant.

     Only 1985 data were used  to determine  existing  BOD  loads  because there
were insufficient data available for 1986.  There were only 304 days of
reported BOD values for Jackson Pike in 1986.  There were  341  days of data  for
Jackson Pike in 1985.   Southerly reported BOD values on 362 days  in 1986 and
364 days in 1985.

     The 1985 annual average BOD load  for Jackson Pike, as presented in Table
4-6, is 100,702 Ib/day.   The maximum monthly average load is  118,466 Ib/day,
and it occurred in January.   The ratio of maximum monthly average to the
annual average results in a  peaking  factor of  1.2.

     The 1985 annual average BOD load  for Southerly, as shown  in  Table 4-6, is
87,258 Ib/day.  The maximum monthly average load, which  occurred  in October,
is 105,446 Ib/day. The peaking factor, as determined by  dividing  the maximum
monthly average by  the annual  average, is  1.2.

     1985 and 1986 data were used  to establish TSS loads  for Jackson Pike and
Southerly.   Jackson Pike  had 365 and 363 days of TSS data  for  1985 and  1986,
respectively.  There were 364 days of TSS data reported  for Southerly for both
years.

     The average TSS  load was  obtained by computing  the  average of the annual
averages for 1985 and 1986.   The Southerly  1985  and  1986  average  is 97,289
Ib/day; and Jackson Pike is 126,006 Ib/day.   Peaking factors were established
for each year in the  same manner as was used for BOD loads.  The  peaking
factors for Jackson Pike are  1.2 and 1.1 for 1985 and  1986, respectively.  The
                                    A-32

-------
 I
 I
 I
 I
 I
 I
 I
I
 I
I
I
I
higher value of  1.2 was chosen as  Che Jackson  Pike TSS  peaking  factor.   The

Southerly TSS peaking factors are  1.1 for both 1985 and  1986.  Table 4-8

summarizes  the  1985  and  1986  average and peak BOD and TSS loads.



                 TABLE 4-8.   1985  AND 1986 BOD AND TSS  LOADS
BOD LOADS
     •  Average (Ib/day)
                (lb/capita day)


     •  Peak (Ib/day)


     •  Peaking Factor
TSS LOADS
     •  Average (Ib/day)
                (lb/capita day)


     •  Peak (Ib/day)


     •  Peaking Factor
                                      Jackson Pike    Southerly
100,702
  0.206


118,466


    1.2
POPULATION
126,006
  0.258


151,207


    1.2



489,000
 87,258
  0.229


105,446


    1.2
 97,289
  0.255


107,018


    1.1



381,000
 Total





187,960
  0.216


223,912


    1.1
223,295
  0.257


251,925


    1.1



870,000
                                                A-33

-------
     A summary of the 1985 population figures and historic  wastewater flows
and loads is presented in Table 4-9.   These quantities were used as a basis
for projecting flows  and  loads  to  the design  year.

                       TABLE 4-9.  1985 FLOWS AND LOADS

                                Jackson Pike      Southerly      TOTAL
     Total Flow
     Ave. (MGD)                     84               61           145
             •  Infiltration      40.1             22.1          62.2
             •  Industrial         9.0              6.9          15.9
             •  Commercial         4.5              3.2           7.7
             •  Domestic          30.4             28.8          59.2
     BOD Load (Ib/day)         118,500          105,400       223,900
     TSS Load (Ib/day)         151,200          107,000       258,200
     Population                489,000          381,000       870,000

4.3  PROJECTED FLOWS AND LOADS
     This next section presents flows and  loads projected  to the 2008 design
year.

     Table 4-10 presents the flows of Table 4—9 in per capita/connection
form.  These data further reinforce the figures presented in Table 4-9 since
they represent reasonable values in agreement with the literature.

     Holding infiltration and industrial flows constant and using the existing
per capita commercial and domestic flows  (Table  4-10) and  the population
projections for 1988 and 2008,  wastewater  flows were projected for 1988 and
2008.
                                      A-34

-------
I
I
—                          TABLE 4-10.  1985 PER CAPITA/CONNECTION FLOWS


                                              Jackson Pike      Southerly      TOTAL

          Per Capita
          Domestic Wastewater Flow (gpcd)         62.2             75.6         68.1
•        Per Capita
          Commercial Wastewater Flow (gpcd)        9.2              8.4          8.9
•        Per Capita
          Industrial Wastewater Flow (gpcd)       18.4             18.1         18.2

I          Per Capita
          Industrial, Commercial, and
          Domestic Wastewater Flow (gpcd)         89.8            102.1         95.2
•        Per Capita
          Infiltration (gpcd)                      . 82               58           72

          Per Connection
          Commercial Wastewater Flows'
          (gal/connection day)                      ND               ND         816.7

•        Per Connection
™        Industrial Wastewater Flows
          (gal/connection day)                      ND               ND        62,109

I
          1985 Per Capita
          Water Pumped
          Industrial, Commercial,  and
          Domestic (gpcd)                          ND               ND         139.1
I
          1985 (Industrial,  Commercial,  and
•        Domestic) Water Pumped to Wastewater
I
          Discharge Factor                         ND               ND          .976

            SOURCE:  City of Columbus,  Division of Sewerage and Drainage,  December 1986

I

I

I

I

I
                                                A-35

-------
     There was insufficient information  available  to disaggregate the existing
industrial loads and the expected future industrial loads from  the  total.
Therefore, the existing total per capita BOD and TSS loads from Table 4-8 were
multiplied by the population projections and the respective peaking factors to
obtain the 1988 and 2008 projected loads.  In doing so,  growth of industrial
contributions is proportional to residential growth.

     Table 4-11 presents the 1988 projected population,  flows, and loads for
each plant; and Table 4-12 presents  the projected design average flows and
loads for  the 2008 design year.
                                      A-36

-------
1
1
1



1

1
1

1

1
1



1


1

1

1
1
1

1


TABLE 4-11. 1988
Jackson Pike
Total Flow
Ave. (MGD) 84.8
• Infiltration 40.1
• Industrial 9.0
• Commercial 4.6
• Domestic 31.1
BOD Load (Ib/day) 123,400
TSS Load (Ib/day) 154,500
Population 499,000

TABLE 4-12. 2008
Jackson Pike
Total Flow
Ave. (MGD) 87.9
• Infiltration 40.1
• Industrial 9.0
• Commercial 5.0
• Domestic 33.8
BOD Load (Ib/day) 134,600
TSS load (Ib/day) 168,600
Population 544,600



A-37



PROJECTIONS
Southerly

61.7
22.1
6.9
3.3
29.4
106,900
109,100
389,000

PROJECTIONS
Southerly

66.0
22.1
6.9
3.7
33.3
121,300
123,800
441,400








TOTAL

146.5
62.2
15.9
7.9
60.5
230,300
263,600
888,000


TOTAL

153.9
62.2
15.9
8.7
67.1
255,900
292,400
986,000






-------
                        5.  FACILITY PLAN METHODOLOGY

     The following sections summarize the design wastewater flows and loads
proposed in the facility plan and the methodology used in their development.
The Revised Facility Plan Update (RFPU)  and  the  General  Engineering Report and
Basis of Design (GERBOD) were used to prepare this discussion.   The following
sections include:

     •  Dry Weather Wastewater Flows
     •  Design Average Daily Flows
     •  Design Loads
     •  Industrial Flows and Loads
     •  Projected  Design Flows and Loads

     The facility  plan developed existing dry weather  wastewater flows to
approximate a low groundwater condition.  These  flows  were projected to the
2015 design year.   Then average  daily flows  were developed to approximate
average infiltration under a high  groundwater condition and  these  flows  were
projected to the 2015 design year.   The 2015 average  daily  flows approxi-
mating average infiltration under a high groundwater condition were selected
as the design average flows  for use in alternative development in the facility
plan.

     Existing waste loads were  determined and projected to the 2015 design
year.  Two scenarios were developed for  future additional  flows and loads from
undocumented industrial growth.  However, since neither of these scenarios was
included in the design average flows and loads,  it appears that a decision was
made not to plan  for  future  undocumented  industrial growth.  The last section
summarizes  the facility plan's selected  design wastewater flows and  loads.

5.1  DRY WEATHER WASTEWATER FLOWS
     The Revised Facility Plan Update developed  dry weather flows to
approximate low infiltration under low groundwater conditions.   Monthly
                                      A-38

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
I
I
Operating Reports (MORs) were used to  determine  dry  weather wastewater flows

through three different methods.



     •  An average daily flow was  derived  for the 1984 dry  months (July and
        August).


     •  A 50th percentile flow was determined from 40 randomly selected dry
        weather days between 1982  and 1984.


     •  An average daily flow was  derived  for the dry months of 1979
        through 1984.



     The GERBOD states that flows  of  74 MGD  for Jackson Pike and 53 MGD for

Southerly were determined from the first method listed above using 1984 dry

months.  The report states that the flows developed by the other two methods

closely approximate these flows, but  a direct comparison  is not provided.



     The 1983 population for each WWTP service area was selected as the

population value to be used for calculation of gallons per capita per day

(gpcd)  flow factors.  The  1983  Southerly  population  was  determined to be

356,901 and the Jackson Pike population was determined to  be 470,979.

Calculated gpcd flow factors are 149  gpcd for Southerly and 157 gpcd for

Jackson Pike  which results in a  system-wide average of 153 gpcd.   By

comparison, a system-wide  average  of  152 gpcd was calculated in the Original

Facility Plan based upon 1975 flow data.



     Utilizing population projections  developed  in the Revised  Facility Plan

Update (presented  in Table  5-1) and the gpcd  flow  factors  discussed  above,  dry

weather flows were  projected for  each plant  for the years 1988, 2000, and

2015.   Table 5-2 presents these flows.
                                      A-39

-------
               TABLE 5-1.  FACILITY PLAN POPULATION PROJECTIONS
           Service Area                        Year
Jackson Pike
Southerly
TOTAL
1988
487,644
382,783
870,427
2000
531,366
420,495
951,861
2015
573,052
459,992
1,033,044
           SOURCE:   Revised Facility Plan Update - URS Dal ton 1985

                TABLE 5-2.   PROJECTED DRY WEATHER FLOWS (MGD)

           Service  Area                        Year
                               1988            2000            2015
           Jackson Pike          77              83              90
           Southerly             58              63              69
           TOTAL                135             146             159
           SOURCE:  Revised Facility Plan Update - URS Dalton 1985
5.2  DESIGN AVERAGE DAILY FLOWS
     The Revised Facility Plan Update developed design average daily flows to
approximate average infiltration under high  groundwater conditions.    Flow
values were obtained from Monthly  Operating  Reports  (MOR).  Forty-five days
were randomly selected from the years 1982 through 1985 based  on the following
criteria:

     •  .Weekdays only
     •  No significant rainfall on the sample day
     •  No significant rainfall for 24 hours prior to the sample day
     •  No reported bypassing

     The 50th percentile flow values were derived from probability plots
for each WWTP.  The Jackson  Pike flow was determined to be 84  MGD, and the
Southerly flow was  determined to be  59 MGD.
                                      A-40

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
I
     The WWTP  service  area  population  figures  for  1984  of 475,909  for Jackson

Pike and 363,097 for Southerly were used to convert the above flows to gpcd

flow factors.  Jackson Pike was calculated at  177 gpcd and Southerly was

calculated at 162 gpcd.   Table 5-3  shows the  breakdown of the flow factors as

presented in the RFPU.



                 TABLE 5-3.  DESIGN AVERAGE FLOW FACTORS  (gpcd)



                      Jackson Pike         Southerly             Total
Flow Component        Service Area        Service Area        Service Area


Domestic
Industrial
Infiltration              	                  	                  	


Total                    177                 162                 171


Source:  Revised Facility Plan Update - URS Dalton 1985


     Using population projections  (Table 5-1) and total gpcd flow  factors

developed for each WWTP, design flows were projected  for  the years 1988, 2000,

and 2015.  These flows are  shown  in Table  5-4.



                    TABLE 5-4.  DESIGN AVERAGE FLOWS  (MGD)
                                               Year
80
15
82
80
24
58
80
19
72
Service Area
Jackson Pike
Southerly
TOTAL
1988
86
63
149
2000'
94
68
162
2015
101
75
176
                 SOURCE:  Revised Facility Plan Update - URS Dalton 1985



5.3  DESIGN LOADS


     Existing wasteloads were determined by randomly selecting 80 days from

MORs for each WWTP based on the following criteria:



     •  40 days during the low infiltration season - July, August, September
                                     A-41

-------
     •  40 days during the low temperature season - December, January,
        February
     •  Years 1982, 1983, 1984
     •  No significant rainfall on the selected day.
     •  No significant rain for 24 hours prior to the selected day
     •  No bypassing
     •  Weekdays only

     The parameters selected for analysis were BODc, TKN, phosphorus,
suspended solids, and flow.

     Probability plots were constructed from  these  calculated loads for both
the low infiltration season and the low temperature season.  The 80th
percentile loads were chosen from  the  low infiltration plots.  These loads
were used in calculating projected loads for  1988,  2000, and  2015.
Adjustments were made for  the Anheuser-Busch Brewery.  It was assumed that the
existing loads  and  flow  from the brewery are the following:

     •  BOD  = 35,260 Ib/day
     •  SS     13,400 Ib/day
     •  Flow =3.13 MGD

     For future projections it was assumed that the brewery would increase
to its maximum  monthly  average BODc  limit  of 45,000 Ib/day.

     Table 5-5 presents  the Revised Facility Plan Update's  design wasteloads
for the Jackson Pike and Southerly WWTP.
                                      A-42

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 i
 I
 I
I
I
I
I
                   TABLE 5-5.  DESIGN WASTELOADS (Ib/day)
Jackson Pike
Design
Year
1988
2000
2015
1988
2000
2015
BOD5
127,150
137,060
148,620
117,060
123,730
131,740
Suspended
Solids
145,780
157,130
170,390
180,020
116,450
126,550
TKN
16,700
18,000
19,520
14,570
15,760
17,260
Total
Phosphorus
5,459
5,884
6,380
4,595
4,991
5,467
Southerly
SOURCE:  Revised Facility Plan Update - URS Dalton 1985



5.4  INDUSTRIAL FLOWS AND LOADS


     The Revised Facility Plan Update presented tables which included

additional flow and loading allowances for future,  undocumented industrial

growth.  This growth was only assumed  to  affect the Southerly  plant.  No

reason was provided for this assumption.   Flow projections for both dry

weather and design average  flow were  increased by 2 MGD and 4 MGD after 1988

to account for the  possibility  of  undocumented growth in the industrial

sector.  Table 5-6  presents dry weather flow projections with the additional

flow allowances of 2.0 MGD  and 4.0 MGD for undocumented industrial growth.

Table 5-7 presents  design average  flow projections  with the same flow

allowances.
 TABLE  5-6.   DRY WEATHER FLOW PROJECTIONS FOR UNDOCUMENTED INDUSTRIAL GROWTH
                        (2 MGD/4 MGD FLOW ALLOWANCES)
                Service Area


                Jackson Pike
                Southerly
                TOTAL
                SOURCE:   Revised Facility Plan Update - URS Dalton 1985
Year
1988
77/77
58/58
135/135
2000
83/83
65/67
148/150
2015
90/90
71/73
161/163
                                     A-43

-------
TABLE 5-7.  DESIGN AVERAGE FLOW PROJECTIONS FOR UNDOCUMENTED INDUSTRIAL GROWTH
                        (2 MGD/4 MGD FLOW ALLOWANCES)
                                              Year
                Service Area        1988      2000      2015
                Jackson Pike       86/86     94/94    101/101
                Southerly          63/63     70/72     77/79
                TOTAL             149/149   164/166   178/180
                SOURCE:  Revised Facility Plan Update - URS Dalton 1985

     The increase in industrial flow could also increase the wasteload
projection.   Therefore,  the facility planners  also revised their wasteload
projections to reflect the 2/4 MGD  flow  increases.  Since the industrial flow
increase was only added to the Southerly  WWTP, the increase in wasteloads will
only affect the Southerly plant.   Table 5-8 presents these revised wasteload
projections for  Southerly.

              TABLE 5-8.  DESIGN WASTELOAD PROJECTIONS (Ib/day)
                 ADJUSTED FOR UNDOCUMENTED INDUSTRIAL GROWTH
                                SOUTHERLY WWTP
Design
Year
2 MGD Allowance
1988
2000
2015
4 MGD Allowance 1988
2000
2015
BOD,;
117,060
138,730
146,740
117,060
163,730
171,740
Suspended
Solids
108,820
122,170
132,230
108,020
131,610
141,710
TKN
14,510
16,510
18,010
14,510
17,760
19,260
Total
Phosphorus
4,595
5,066
5,542
4,595
5,191
5,667
SOURCE:  Revised Facility Plan Update - URS Dalton 1985

     The flows and loads which included an allowance for undocumented
industrial growth were  not used as a basis for alternative development, which
assumes no industrial growth during the planning period.
                                      A-44

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
5.5  PROJECTED DESIGN FLOWS AND LOADS


     This section summarizes  the  facility  plan's projected design flows and

loads.



5.5.1  Design Flows


     The Jackson Pike and Southerly projected 2015 design average daily flows

of 101 MGD and 75 MGD, respectively, were chosen by the facility plan as the

basis for development.   These  flows were presented in Section 5.2.



     Since the Revised Facility Plan Update chose  the one-plant alternative,

the design flows for Jackson Pike and Southerly were combined  resulting in an

average daily design flow of 176 MGD  to  be  treated at Southerly.  Peak process

design was then  calculated as 300 MGD by multiplying the design average flow

by a peaking factor of 1.7.  The peak  process  flow  of 300  MGD is used to size

the wet stream  treatment facilities.  There is no  supportable information in

the facility plan on how the peaking  factor was derived.  There is reference

to the fact that anything greater than 1.7  would adversely affect process

efficiency under average flow conditions.  In subsequent correspondence and a

clarifying telephone conversation with the city's  consultant, it was

determined that 1.7  was based  on a hydraulic constriction  between the existing

primary clarifiers and aeration basins at the Southerly WWTP.  The consultant

indicated that each  existing train  is  limited  to an average to peak flow ratio

of 44 MGD to 75 MGD.  The 44 MGD average flow is based on mass loading to the

aeration basins,  and the  75  MGD peak flow is based on the  hydraulic capacity

of the existing conduits between the primary clarifiers and  the aeration

tanks.   In light of  the fact that  the  CSO study  is  incomplete and that

analyses of wet weather  data was limited,  the 1.7 peaking  factor was

considered appropriate by the city and their consultant.



     An additional 130 MGD for CSO control is added on to the peak process

flow of 300 MGD and  this  total  flow of 430  MGD is  considered as the peak

hydraulic flow.    There are  conflicting  statements  in the  facility plan

regarding which treatment processes will be sized  for 430 MGD.  In Chapter 2,
                                      A-45

-------
it is stated that flows between 300 MGD and 430 MGD will be processed through
primary settling and chlorination.  In Chapter 12, it appears that only the
influent pumps and bar racks and screens are being sized for 430 MGD.

5.5.2  Design Loads
     Section 5.3  presented design  loads determined in the facility plan
through analysis  of Jackson Pike and Southerly plant records.   In addition to
these loads, the  facility plan presents loads contributed by the additional
flows conveyed to the plant during peak hydraulic flow  conditions.  These
additional flows are considered to be diverted from Whittier Street.
Table 5-9 presents the design loadings for the year 2015 including the loads
from Whittier Street flows.

                    TABLE 5-9.  DESIGN LOADINGS (Ib/day)

    Parameter     Southerly     Jackson Pike     Whittier St.      Total
      BODr         131,740        148,620           10,000        290,360
      TSS          126,550        170,390           20,000        316,940
      TKN           17,260         19,520            1,300         38,080
      P              5,467          6,380              400         12,247
    SOURCE:  Revised Facility Plan Update - URS Dalton 1985
                                                   /
     Since the Revised Facility Plan Update chose the one-plant alternative,
the loads listed in the Total column in the previous table were used as the
basis for development of alternatives.
                                      A-46

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
      6.  COMPARISON OF BRIEFING PAPER AND FACILITY PLAN FLOWS AND LOADS



     This section summarizes the design flows and  loads developed in  the

facility plan and those developed by this briefing paper.  Table 6-1  provides

a comparison between the  two.



               TABLE 6-1.  COMPARISON OF DESIGN FLOWS AND LOADS
                                          Facility Plan
Briefing Paper
149
2015
176
1.7
300
290,360
316,940
147
2008
154
1.5
231
255,900
292,400
  1988 Projected Average Flow (MGD)
  Design Year
  Design Average Flow (MGD)
  Process Peaking Factor
  Peak Process Flow (MGD)
  Design BOD Load (Ib/day)
  Design TSS Load (Ib/day)
     The 1988 projected average flows are very close, being  149 MGD in the

facility plan and 147 for this briefing paper.  The projected design average

flows of 176 MGD  for  the  facility  plan  and  154 MGD  for  this  briefing paper

vary by 22 MGD (14 percent) due to different design years that result in a

difference in population projections.  The facility plan flows are based on

the year 2015 and the briefing paper flows  are based on a 2008 design year.

For purposes of comparison, the facility plan design, average flow was brought

back to the year 2008.  Using the 2008 population projections developed for

the EIS and the gpcd flow figures used  in the facility  plan, the facility plan

flows for 2008, presented in Table  6-2,  are  96 MGD for Jackson Pike and 72 MGD

for Southerly.  This  total flow of 168 MGD  for both plants is 14 MGD (9

percent) higher than the briefing paper  flow of 154 MGD.
                                      A-47

-------
          TABLE 6-2.  FACILITY PLAN POPULATION AND FLOW PROJECTIONS
Flow Values (gpcd)

1988
     • Population
     • Flow (MGD)
2000
2015
       Population
       Flow (MGD)
2008
     • Population*
     • Flow (MGD)**
       Population
       Flow (MGD)
                         Jackson Pike
    177
487,644
     86
531,366
     94
544,600
     96
573,052
    101
Southerly

    162
382,783
     63
420,495
     68
441,400
     72
459,992
     75
                                 TOTAL
      171
  870,427
      149
  951,861
      162
  986,000
      168
1,033,044
      176
Source:  Revised Facility Plan Update - 1985

 *EIS population projections.
**Developed for comparison with briefing paper 2008 design flow.
     Thus, even if the design years are the same and the population projec-

tions are the same, the design flows still differ slightly.  This difference

is because the flow projections made in the briefing paper were developed by

holding the infiltration and industrial portions of the flow constant and

increasing only the commercial and domestic flows proportional to the popula-

tion increase, whereas the flows  in the facility plan were developed by

increasing all of the flow, including  infiltration and industrial,

proportional to the population increase.   Projected increases in infiltration

do not appear justified if the population  increase  is  located within the

existing service area. The facility plan  does  not document why an increase in

infiltration should be planned for.  Projected industrial increases should be

based on documented industrial growth  by  existing industries and/or policy
                                     A-48

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
decisions by the municipality  Co  plan  for  future undocumented  growth.

Furthermore, such industrial growth should be an identifiable  part of  the

total design loads since capital cost  recovery for the added capacity  must be

addressed.



     The projected peak process flows  are  231 MGD  and 300 MGD  for  the  briefing

paper and facility plan, respectively.   These flows differ significantly due

to differences in design average flows and different peaking factors.  The

reasons for the different  design average  flows were discussed  in the previous

paragraphs.   The peaking factor is  1.5 for the briefing  paper  and  1.7  for the

facility plan.   The 1.5  peaking factor  for  the briefing  papers  is consistent

with the  peaking  factor used  in the  original EIS.  The  facility  plan's peaking

factor of 1.7 is based on  the  maximum hydraulic capability of  the conduits

between the primary clarifiers  and aeration  basins in the existing trains at

the Southerly WWTP.



     A breakdown of the briefing paper and facility plan projected BOD and TSS

loads is presented in Table 6-3.  The differences in loads are  partially due

to differences  in design years and also due to the inclusion of loads from

Whittier Street.  For comparison purposes,  Table 6-4 presents  the facility

plan loads brought back to 2008 without the Whittier Street  loads.  These

loads were decreased to the year 2008 using EIS population projections and

load factors developed  in Section 4.2 of  this document.   In comparing  the

briefing paper loads to the 2008 facility plan loads,  it was found that the

loads are within 5 percent of each other.   Therefore,  the 2008  facility plan

loads will be used as the basis for further EIS evaluations.
                                      A-49

-------
   TABLE 6-3.  COMPARISON OF FACILITY PLAN AND BRIEFING PAPER DESIGN LOADS
Design Year

Jackson Pike
     •  BOD (Ib/day)
     •  TSS (Ib/day)

Southerly
     •  BOD (Ib/day)
     •  TSS (Ib/day)

Whittier Street
     •  BOD (Ib/day)
     •  TSS (Ib/day)
TOTAL
        BOD (Ib/day)
     •  TSS (Ib/day)
                Facility Plan
                     2015
                   148,620
                   170,390
                   131,740
                   126,550
                    10,000
                    20,000
                   290,360
                   319,940
                 Briefing Paper
                      2008
                    134,600
                    168,600
                    121,300
                    123,800
                    255,900
                    292,400
                       TABLE 6-4.  2008 PROJECTED LOADS
Jackson Pike
     •  BOD (Ib/day)
     •  TSS (Ib/day)

Southerly
     •  BOD (Ib/day)
     •  TSS (Ib/day)
Total
     •  BOD (Ib/day)
     •  TSS (Ib/day)
                       Facility Plan
141,600
161,600
126,600
121,300
268,200
282,900
                             Percent Difference
                  EIS        From Facility Plan
134,600             -4.9
168,600             +4.3
121,300             -4.2
123,800             +2.1
255,900             -4.5
292,400             +3.2
                                      A-50

-------
  I
  I
  I
  I
  I
  I
  I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
Summary



     Table 6-5 summarizes  Che  2008  flows  and  loads which  will  be  used as a

basis for further EIS analysis.  The 2008 average flows developed in this

briefing paper will be utilized.  They were developed based on 1985 and 1986

plant records, industrial flow data from the 1983 Industrial Pretreatment

Report,  and the facility plan  infiltration values.  A process peaking factor

of 1.5  is  applied  to  this average flow to obtain the peak process flow.   The

facility plan BOD and TSS loads brought back to 2008, without the Whittier

Street loads,  will be utilized as the design  loads.  Further documentation is

required and has been requested to verify the industrial flows and

infiltration value.



                TABLE 6-5.  2008 PROPOSED EIS FLOWS AND LOADS


                          Jackson Pike         Southerly         Total


Average Flow (MGD)               88                  66              154


Peak Process Flow (MGD)         132                  99              231


BOD Load (Ib/day)            141,600             126,600          268,200


TSS Load (Ib/day)            161,600             121,300          282,900
                                      A-51

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
         APPENDIX B


    BRIEFING PAPER NO. 2
SOLIDS HANDLING ALTERNATIVES

-------
I
I
I
•         BRIEFING  PAPER NO.  2
I
I
          SOLIDS HANDLING  ALTERNATIVES
I
•         Supplemental Environmental Impact Statement
-         USEPA Contract No. 68-04-5035, D.O.  No. 40
I         Columbus Ohio Waste-water Treatment Facilities
I
I
I
I
I
•         Prepared By.
I         SCIENCE  APPLICATIONS INTERNATIONAL CORPORATION
|         TRIAD ENGINEERING  INCORPORATED
I
I

-------
                         SOLIDS HANDLING ALTERNATIVES
1.    EXISTING SLUDGE MANAGEMENT SYSTEMS
     1.1  Jackson Pike
     1.2  Southerly

2.    DEVELOPMENT OF SLUDGE MANAGEMENT ALTERNATIVES
     2.1  Jackson Pike Sludge Management Alternatives (Two-Plant Scenario)
          2.1.1  Jackson Pike Sludge Management Alternative JP-A
          2.1.2  Jackson Pike Sludge Management Alternative JP-B
          2.1.3  Jackson Pike Sludge Management Alternative JP-C

     2.2  Southerly Sludge Management Alternatives (Two-Plant Scenario)
          2.2.1  Southerly Sludge Management Alternative SO-A
          2.2.2  Southerly Sludge Management Alternative SO-B
          2.2.3  Southerly Sludge Management Alternative SO-C
          2.2.4  Southerly Sludge Management Alternative SO-D
          2.2.5  Southerly Sludge Management Alternative SO-E
          2.2.6  Southerly Sludge Management Alternative SO-F

     2.3  Southerly Sludge Management Alternatives (One-Plant Scenario)

3.    EVALUATION OF SLUDGE MANAGEMENT ALTERNATIVES
     3.1  Cost Effectiveness of Sludge Management Alternatives
     3.2  Sludge Dewatering
     3.3  Planned System Redundancy
     3.4  Ultimate Disposal Plan
          3.4.1  Distribution and Marketing of Composted Sludge
          3.4.2  Land Application of Digested, Dewatered Sludge
          3.4.3  Landfilling of Incinerated Dewatered Sludge

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
I
                                 INTRODUCTION



     Under Che direction of USEPA,  a series of briefing papers are being

prepared addressing key issues in Che development of the Supplemental

Environmental Impact Statement for the Columbus,  Ohio,  Wastewater Treatment

Facilities.  The  briefing  papers  form the basis of discussions between Triad

Engineering and USEPA to resolve these key issues.  The following paragraphs

present the background of the facility planning process, a description of the

briefing papers,  and Che purpose of this paper on solids handling alterna-

tives.



FACILITY PLANNING PROCESS


     At the time this paper was  prepared (March-August  1987)  Che city of

Columbus was proceeding to implement improvemenCs aC Che Jackson Pike and

Southerly WasCewater TreaCmenC Plants  to comply with more stringent effluent

standards which must be met by July 1,  1988.    These improvements were based

on the consolidacion of wasCewaCer treatment operations at Che SouCherly

plant.  This  one-plant  alternative  is  a change  from  the two-plant operation

proposed by Che cicy in Che  1970's and  evaluated  in the  1979  EIS.



     The development and documentation of a wastewater treatment process and

sludge management alternatives for the Columbus metropolitan  area has been an

extended and iterative process.    The design and construction of various

system componenCs have  progressed, because of Che 1988 deadline,  while

planning issues continue Co be resolved.  As  a  resulc,  numerous  documenCs have

been prepared which occasionally revise a previously esCablished course of

direction.



     The concurrent  resolution of planning issues and  implementation of

various project components has made preparation of the  EIS  more  difficult

because final facilicy plan recommendacions are noC available in a single

document.
                                      B-l

-------
BRIEFING PAPERS
     To facilitate preparation of the EIS,  a series of briefing papers are
being developed.   The  purpose  of  the  briefing  papers  is  to  allow USEPA to
review the work of the EIS consultant and to identify supplemental information
necessary for the preparation  of  the  EIS.    Six briefing papers are being
prepared as follows:

     •  Flows and Loads
     •  Sludge Management
     •  CSO
     •  Process Selection
     •  One Plant vs.  Two Plant (Alternative Analysis)
     •  O&M and Capital Costs

     The specific focus of each briefing paper will be different.  However,
the general scope of the  papers will  adhere to the  following format:

     •  Existing conditions will  be documented.
     •  Evaluations, conclusions,  and  recommendations of the facilities
        planning process will be reviewed  using available documentation.
     •  Where appropriate, an  independent evaluation of the future situation
        and viable alternatives will  be prepared.
     •  The facility plan and  EIS briefing paper conclusions will be compared.

     The briefing paper process is intended to:

     •  Prompt the resolution  of any data deficiencies.
     •  Clearly establish and  define existing and future conditions.
     •  Identify the final recommended plan which the city desires to implement.
     •  Provide a data base of sufficient detail to allow preparation of the
        draft EIS.
                                      B-2

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
SLUDGE MANAGEMENT ALTERNATIVES


     This briefing paper reviews  Che  facilities planning process and

subsequent efforts by the city relative to the development and adoption of a

sludge management  alternative.  The briefing paper is divided into four

sections as follows:



Section 1 - Existing Sludge Management System.


     Section 1 defines the current sludge processing and disposal practices of

     the Jackson Pike and Southerly plants.   It establishes a foundation from

     which potentially viable sludge management alternatives can be

     identified.



Section 2 - Development of Sludge Management Alternatives


     In Section 2 potentially viable  sludge  management alternatives are

     identified and developed sufficiently to allow a comparative evaluation.



Section 3 - Evaluation of Sludge Management Alternatives


     Section 3 evaluates the sludge management alternatives that were

     developed.  The alternatives are evaluated with respect to the analysis

     of the briefing paper and in light of the recommendations of the

     facilities planning process and  subsequent planning and preliminary

     design documents.



Section 4 - Planning Issues to be Resolved


     In Section 4 the issues that developed through this analysis are

     highlighted to facilitate discussion and  resolution.



The primary sources of information utilized in preparing this briefing paper

included:



     •  Revised Facilities Plan Update, September 30,  1985

     •  General Engineering Report and Basis of Design,  January 1,  1986

     •  Preliminary Design Evaluation of Sludge Dewatering, December 12,  1986
                                     B-3

-------
                   1.   EXISTING SLUDGE MANAGEMENT SYSTEMS

l.l  JACKSON PIKE

     Figure 1 presents Che sludge processing and disposal schematic currently
in operation at Jackson Pike.  The sludge processing operations include:

     •  Primary sludge (PS) thickening in primary clarifiers
     •  Centrifuge thickening of waste activated sludge (WAS)
     •  Thickened sludge storage and blending (i.e. PS and WAS)
     •  Stabilization by anaerobic digestion or thermal conditioning
     •  Centrifuge dewatering

Dewatered sludge is disposed of in one of the following ways:

     •  Dewatered sludge is incinerated and  the ash product is  ultimately
        landfilled.
     •  Dewatered sludge is land applied in an agricultural  reuse program.

     The Jackson Pike facility currently produces 230-250 wet tons per day of
dewatered sludge at a cake solids concentration of about 17  percent.  On a dry
weight basis approximately 50 dry tons per day (dtpd) of dewatered solids are
produced for ultimate disposal.  Based on recent operating records,  approxi-
mately 50 percent of the dewatered sludge is incinerated and 50 percent is
land  applied.

     Table 1 identifies and describes the existing sludge management
facilities at Jackson Pike.  The facility has provisions for short-term
storage of both PS and WAS outside of  the main liquid processing stream.  The
centrifuges for thickening of WAS were originally installed in 1975-76 and are
estimated to have a remaining  useful  life of approximately 10  years.
Thickened sludge storage and blending is accomplished using a secondary
digester.  The anaerobic digestion facilities consist of eight  primary
digesters constructed in 1937, and eight secondary digesters constructed in
                                      B-4

-------
1
1
1
1 -a
• 0£j-
1-
|§c
°£
LJ^
^5
1
AEREATIDN
PRIMARY
CLARIFICATION
1
PRELIMINARY
TREATMENT
1
1 I
1

EFFLUENT

S
^ « |
« ^2 ^u^
Z £ LJ LJ LJ X
i— i QL v; ^ L3 fP
^ ° ^~ O ^ U ^ z
CO i -^ *— ' t^) —^i.i
^a LJI< i_j_J
>x UH-> i— c^m
— — —
j


1
|


<


i


•
>-. 1°
tK^Z

-------
CO


M

H
o as
< o
H W
33
z o
w
CU
00
CO
o
4J
co

U-l
o

^
4-1
•r4
CJ
(0
Q.
CO
U
O
S
LO
^O
o
•
0
C
CO
CO
Cu
CU
0)
73
u
0
O
1
OO

X

4J
O
o
•r-t
4J
•r4
73
C
0
o
^x-
CO
CU
•r4
4->
•H
•— *
• r4
CJ
CO

o
O en
U-l 4-1
1 -r4
-* C
-t 3

X >>

4J 73
O C
O CO
14-| 4J
1 CO
00

'o
CU 3
C 4-1

00
C
73
1— 1
0
as


•l-l
4J
CJ
co
co
CU
CJ
o
CM
^J

CU
CO
E£
cu
oo
CO
V4
o
4J
eo

U-l
O

O
22

1— 1






p
CO

4-1
o
o
U-l
1
m
CN
•
m
CN

•«
•
CO
• H
73

4J
O
O
U-l
1
in
co

cu
c
O







00
C
•l-l
73
i— l
O
S3

CU
00
73
3
1—4
CO

^
J^
CO
e

PJ
4-1
• 1-4
C
3
04
PS
O
o »-«
-a- , cg-w-
CO 73
73^, U-l
~~~, eo O
co C
C O u
O 4-1 CO
4-1 O
O CJ
0 0
in CN 4-1
r-l 1 T-l
1 O C
o r- 3
CO
rH C 0)
O 4-1
4J -H CO
l-l 4-1 S
O CO 'r4
CU CJ X
CO -r4 O
C rH t,
co Cu Cu
lj Cu Cu
fr> «


















a
O
•H
4J
CO
l-l
CU
CU
o

4-1
CJ
CO
l_l
4-1
C
o





73
C
0)

4J
^1
0
CU
en
C C
CO O
r4 *r4
H 4J
CO
cu u
00'--'
73 r-4
3 Cu
r-4 CU
CO <3



0 days/yr
dependent
cropping patterns
VD CO 73
CN ,* C3
C\J (Q
C cu
O Cu l-i
•H Ol

CO Cfl 4J
0 C CO
•r4 O CU
r-l CO 3
Cu CO
Cu cu C
< CO O







^
^
"v^
CJ 0

0 O
O 0
0 O
CN O
1— 1


01
CU 00
OO CO
CO CU
Ol l-i
l-l O
CJ CO
CO
CU
73 r-l
01 43
rl CO
•H •— 1
3 'r4
o* trj
01 >
CU <







CO
01
4-1
• i-4
CO

Ci
o
• r4
4J
CO
CJ
• r4
r-4
CU
CU
^



                                   B-6

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
1950.  Current practice utilizes two of the secondary digesters as short-term
sludge holding facilities.  One digester is used as a mixing and blend tank as
identified previously, the second provides for storage of digested sludge
prior to dewatering.  Based on information furnished by the city, the
structural integrity of the digesters is adequate; however, the mechanical
components have reached their  useful  life.  The thermal conditioning units
have performed better  than those at Southerly and  have been maintained in good
operating condition, however, some process piping and mechanical
rehabilitation of the system is warranted.  The centrifuge dewatering
equipment is less than 10 years old and has been rated as adequate for future
use.  The multiple hearth incinerators were rebuilt  in 1978-79.  The units are
estimated to have 15-20 years  of remaining useful  service.

     The existing land application program is accomplished through contract
operations.   A local contractor is responsible for transport and spreading of
the sludge.   The  application  program is operated approximately 260 days per
year, 5-6 days per week,  applying 70 to 200 wet tons per day,  17 percent
solids,  of dewatered cake depending on seasonal demand.  In 1985, approxi-
mately 5800 dry tons of sludge  were land  applied,  in 1986  approximately 6800
dry tons of sludge were applied.  Dewatered sludge is normally removed from
the Jackson Pike  site on a uniform basis to either land application under
favorable weather conditions or to storage sites located near the application
sites.   The  city  also has  utilized  the Jackson Pike ash lagoons for temporary
short-term storage of dewatered cake.  The city is pleased with the
performance of the application program.  They believe land application has
been satisfactorily  received by the community and  tha't continuation of the
program should be included as  part of any future planning.  Adequate
application acreage  appears available within a reasonable haul distance from
the treatment facility.  Short-term storage of dewatered sludge has recently
been difficult and the city should address this problem if land application is
part of the future sludge management  alternative.

Currently,  incinerator ash is slurried and pumped  to ash lagoons.  The lagoons
are periodically  dredged with  the ash taken to a landfill.  Plant staff have
                                     B-7

-------
indicated that private landfill operators have declined  to accept  the ash.
Consequently, the only repository  for the ash has been the city-owned
landfill.

1.2  SOUTHERLY
     Figure 2 presents the sludge  processing and disposal schematic currently
in operation at Southerly.   The sludge processing operations include:

     •  Primary sludge (PS) thickening in primary clarifiers
     •  Centrifuge thickening of waste activated sludge
     •  Thickened sludge  blending  (i.e.,  PS and WAS)
     •  Centrifuge dewatering

     Dewatered sludge is  disposed  of in one of the following ways:

     •  Dewatered sludge  is incinerated and  the ash product  is  ultimately
        landfilled.
     •  Dewatered sludge  is hauled to the composting facility and distributed
        as a soil conditioner.

     The Southerly facility currently produces 350-400 wet tons per day of
dewatered sludge at a cake solids  concentration of about 17 percent.  On a dry
weight basis, approximately 64 dry tons per day (dtpd)  of dewatered solids are
produced for ultimate disposal.  Based on recent operating records,
approximately 70 percent of the dewatered sludge is incinerated and the
remaining 30 percent is composted.

     Table 2 identifies and describes the existing sludge management
facilities at Southerly.   Primary  sludge is thickened to approximately 4.5
percent in the primary clarifiers.  The thickening of WAS by solid  bowl
centrifuges was installed and operating in the latter part of 1986.  The PS
and WAS is directed  to separate tanks in a Sludge Control Building where it is
mixed prior to being pumped to the dewatering  centrifuges.
                                      B-8

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
(_(
«t



P)
u
n
CO
CO
(— 1

2
a
P
a
u.


0
z
z
UJ
u
X

^
a
u
a
z



_j
a:
Ul
X
< *""
N^
a
z
z
a
»— i
i— i
PI
z
a
u
/'V
PI
UJ
a
PJ
z
PP
^
                           P)
  UJ
>i
Oil
                                     B-9

-------

































to
w
r-l
H
I— I
rJ
I— I
U
< 32
Cn O

H -
55 CO
W =
o S
< s
2 J
< 0
S 0

u •-
CJ d,
O 8-4
5 s
rJ 5
to
>•
C3 J
ss 3
M U]
H 32
to H
r-l S
J*J O
u co


.
CM

Cx3
,J
3
H

















S~!
in
8~?
—1 CO
4J <«;
•H tO S
B TJ
3 -H TJ
>*
4-1
• H
O
CO
a
CO
0
••v.i-' CO
SOB
Cu CO CU
00 ^
TJ 0
O  .E
CM fri H
















ON
to 1-1
tu=«=
4-) 00 r-r
• H ^-^ 3 4J CO
E to to H-i 0 E
3 CO ,y •!-* co O
••>. E l-i lj •!•)
E
O
• H
4-1
• i-l
•o
B
O
O

to
CO
•i-4
4J
• H
i— 1
• H
CJ
CO
fn
Cl4 TJ 10 4-1 4J 4-1
co cu u B B cd
to D O U
O 3 E 0 CJ CO
o o cu
ON CO »H ' f— 4 •• O
•-I l-» 4J 3 OO
ON CO O OO ^\
CO) 1—4 ^ <"t ,-4
U 4J r-4
<0 TJ E T) 4-4
•r4 B CJ — 4 ••->
BOB O O 4J
3 TJ O to l-i to
BO (X- >,
H CO U 1
3 J3 CO 3 cu 4J
O «5 — ' S Cu PU 55








•-< >i
CO U
4-> >> CO
O l-i T)
4J aj e
S o
Cu -i-l O
CJ U CO
CX to
o
O CJ O
o^x s

CM 00 <±
r» • •
ON 
















§ g
co to

4J •- 4J
• • O to O
to o u o m
U 14-1 CU <4-l VO
CU 1 4-1 | ON
4j m to m i— i
U) CM CO CM
> • M • B
l-i « CO CO O
to .H T3 -H •*
0 T3 E T3 4-1
•r-l O CJ
I-l 4J CJ 4J 3
Cu O cu O u
O to O 4J
I-l 14-1 l«-| O)
3 1 O 1 B
O in 3 in o
Cu 00 H 00 U






















4J
•i-l
E
3
^^.
B
Cu
oo

o
0
CM














o
oo
ON
1— 1

T3
CU
E
O
T3
E
tfl
US
cfl
«x
to  .
B X
cu O
O M
Cu
r-l Q.
3 cfl
O
JO r-4
CO
•0 B
•H o
•-4 -H
O 4J
CO CO
^j
X cu
•^ Cu
CO O





to
E
0
4-1

4-1
a>
3

O
0
CO
**^.
^
O

O
O
^j-

M-l
O

CU
0
3
r—4
O



























B
E
•H
jQ

CO
00
CO
U
o
4-1
U)

co
E
O

• i-l
l-i
4J
T3
E
3
O
r4

to
to
r^
•1-1
0

r~-

<4-l
0

to
CJ
E
to
4-1
to
• I-l
•a

i— i
3
to
32

















X
3
cn-^
B to
0 u
4-> J=

4J VO
to m
B
in o
CM -H
4J
tPi co
u
to cu
^ Cu
o o
3
U 14-1
4-1 O

oo to
1 U
-,
cfl <0
TJ 00
	 T3
to 3
E r-4
O to
4-1
E
4J O
CU
3 >->
B
O co
O T3
CM E
1 to
O O.
CM CO
r-4 t)










0
cu
4-1
tn
>-.
to

CO
i— i
• •-i
Cu

o
• H
4J
CO
4-1
CO
T3
CO
4J
to
l_l
CO
CO

TJ
CU

E
0)
4-1
X
w




o
4-1
en
to l-i
0) cu
r-4 0
CO 3
to to
E
^ 0
r-4 0
3
X> 
3 -H
O r4
M a.
A
U TJ
B
r^ CO
CO
to cj
O -H
Cur-i
CO Ji
•r4 3
a cu



















^
t)
3
p
4-1

>N
-a
TJ
CU
>
o
0
CU
^1

4J
o
3
TJ
o
I-l
Pu









00
E
••-1
4-1
CO
• * 3
X 0)
0) B

>> >>
CO co
TJ TJ

to to
B E
O O
4J 4J

4-1 4J
CO CU
3 3

0 0
in vo
•-• CM










..»
to
4-1
• i-l
E

tn
J2 *-•
4J -H
l-l E
CO 3
CU
J2 J3
4-1
(U l-l
r-4 Cfl
Cu CO
••-" J3
4-1
r-l CU
3 •— i
S a
•H
00 4J
E r-4
•H 3
4J B
CO
•H 3
X cu
cu E

0 O
3 3
H H
























































Cu
3
1
4-1
I-l
CO
4-f
tn

B
•H

to
u
• I-l
E
a

3
0)
52

• *t
:*.
o

o
o
0

vO
r--

>•
4-1
• r4
CJ
tO
CUTJ
CO 0)
O TJ
CU
0) to
00 E
to
u tn
O CO
4J
01 TJ
CU
r-l C
CO CO
4-1 CO
O r-l
H 0































CO
E
O
O
oo
cfl
1— 1

o
£

to
E
O
TJ'H
CO 4-J
TJ ffl
CO M
cu q)
E Ou
1 O
en
eg u
u
E m
cfl U
4J
E B
O O
O
TJ
4) -C
r-l 00
1-4 3
•r4 O
U-l U
TJ J2
B 4J
cfl
r-l tO
• i-l
J2 tn
en to
<: ^3
























I— 1
rH
• l-l
14-1
TJ
E
CO
r-4

TJ
01
E
3
0
1
>•»
4-1
• i-l
CJ
















tn
(0
(0
u
O
1^4
P-i


C
0
• H
4-1
Cfl
U
O
rV CO
^
• H^x
00
TJ E
 B
r-l CO
O J
oo
3
14-1
• H
r4
4-1
B
0)
CJ
 B
 O
•H
 4-1
 CO
 0)
 60
 0)
 CO
 E

60
E
•H
C
O
•H
U
•H
T)
B
O
U

CO
g
0)
J=
H
00
c
•H
1-1
0)
4J
CO
3
CU
Q

tu
00
3
•i-4
M
JJ
E
0)
0
Storage

o>
00
•a
3
t—4
CO

•o
4)
Wi
(U
U
CO
3
a)
Q
osting
a.
B
0
o

o
4J

4J
U
o
a.

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
     The anaerobic digestion facilties have not been operational since 1978,

but are currently undergoing rehabilitation.   The  anaerobic digestion system

at Southerly consists of four primary and two  secondary digesters.  The

thermal conditioning units have not operated at Southerly  for almost  10 years.

Chloride stress corrosion which led to equipment deterioration and continuous

mechanical problems caused the abandonment of  the thermal  conditioning units.



     The existing sludge dewatering facility at the Southerly plants consists

of six centrifuges which produce a final cake  solids of 16-20 percent.

Dewatered sludge is disposed through incineration or composting.  There are

two eight-hearth incinerators at Southerly which are each capable of burning

150 wet tons per day at 20-percent solids.  Two new incinerators are in the

final stages of construction and start-up.  The new  incinerators will each

have a disposal capacity of 260 wet tons per day based on approximately 20

percent solids in the dewatered cake.



     Dewatered sludge which  is not incinerated is  hauled by city-owned

vehicles to the composting facility where it is mixed, dried, screened, and

cured as a soil conditioner.  The composting  facility  can  normally accept  30

percent of the volume of solids  produced  by  the Southerly   plant with peak

capacity of 50 percent under ideal  conditions.   These  ideal conditions relate

to dry weather and cake solids concentration.



Currently,  the ash is placed in on-site ash lagoons.   These ash  lagoons are

periodically cleaned  with the ash  being removed to the city-owned landfill.
                                     B-ll

-------
              2.   DEVELOPMENT OF SLUDGE MANAGEMENT ALTERNATIVES

     Preliminary evaluations necessary to  establish a  foundation  for  the
preparation of the EIS required that  alternative  sludge management  schemes be
identified and developed.  The sludge management  alternatives were  formulated
in light of several goals and objectives.   These  goals and  objectives  included
the following:

     •  The sludge management alternatives must consist of  processing  and
        disposal  options  that will  provide  for  environmentally  sound
        processing and ultimate disposal of sludge.
     •  The alternative must provide  a reliable means  for future  processing
        and disposal.
     •  The alternatives  should offer some flexibility allowing the city  to
        modify the processing and disposal  methods to  relieve pressures
        created by equipment failures or temporary loss of  the ultimate
        disposal  methods.

     The alternatives developed should consider,  to the extent  possible,
optimizing the reuse of the  existing  facilities thus minimizing implementation
costs.

     This preliminary evaluation identified alternatives for the  two-plant
scenario, where Jackson Pike and Southerly would  be operated independently,
and for the one-plant scenario,  where Southerly is expanded to  handle  the
projected flows and loads and the Jackson Pike facility in  abandoned.  Under
the two-plant scenario, three alternative sludge  management schemes were
identified for Jackson Pike,  and six  sludge management alternatives were
identified for Southerly.  For the one-plant scenario  (i.e.,  the  consolidation
of wastewater treatment at Southerly) the  sludge  management alternatives  which
were  identified for the Southerly two-plant scenario were considered appropri-
ate to  evaluate.

     The alternatives which were identified were first subjectively screened
to eliminate those alternatives which did  not  adequately address  future goals
                                     B-12

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
 I
I
I
I
I
I
and objectives.   Alternatives  which advanced from the subjective evaluation

were then developed in greater detail through performance of a solids balance,

identification of required facilities and appropriate facilities sizes,  and

development of a cost estimate  for each  alternative.  Sizing criteria used

were consistent  with  current  engineering practice.  The cost estimates

prepared during the facilities planning process for required facilities  were

reviewed in detail.  For the most part, these estimates were considered

reasonable and reflective of facilities planning work.  The cost estimates

developed in this briefing paper, revised and modified the facilities planning

estimates as appropriate to account for the difference between the

alternatives developed herein and the facilities plan alternatives.  In areas

where the facilities planning estimate was not adequately supported, this

evaluation adjusted the estimates appropriately.



2.1  JACKSON PIKE SLUDGE MANAGEMENT ALTERNATIVES (TWO-PLANT SCENARIO)


     Three potential sludge management alternatives were identified for the

Jackson Pike WWTP.   Each alternative  is discussed separately in the following

paragraphs.



2.1.1  Jackson Pike Sludge Management Alternative JP-A


     Figure 3 presents the sludge managment schematic  for alternative JP-A.

The alternative would involve the following sludge processes:



     •  Gravity thickening of PS

     •  Centrifuge  thickening of WAS

     •  Thickened sludge storage and  blending

     •  Stabilization by anaerobic digestion

     •  Centrifuge  dewatering



     Dewatered digested sludge would be  land applied in an agricultural  reuse

program.
                                      B-13

-------
   UI
   Z)
   _J
   U.
   u.
   Ul
LJ
a
                                                                                                         CO
Uin
^Z
H~
U.Z
•rJUl
£v^
zH"
LJX^
OH->
1 -
R
U
HICKEN
LUDGE
h- H
Qi
a
h—
l»
\
«
Z
UJ
_J
fp
- 1
a
                                                                                                  l
                                                                                               ZH
                                               B-14

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
 I
I
     Based on the subjective review of this management alternative, it was

eliminated from  further  consideration.  Relying  strictly on  land application,

for ultimate disposal of  the projected sludge quantities, lacks the

flexibility critical to maintaining a successful disposal program.  This lack

of flexibility would require an  increased degree of conservatism in design and

implementation to ensure  plant performance during an interruption of the

disposal process.  Furthermore,  the seasonal nature of the agricultural

application program would require  substantial sludge storage facilities.

Normally, such storage facilities experience community relation difficulties

associated with  aesthetics  and odors.



2.1.2  Jackson Pike Sludge Management Alternative JP-B


     Figure 4 presents  the sludge management schematic for alternative JP-B.

This alternative would consist of the following sludge processes:



     •  Gravity  thickening of PS

     •  Centrifuge thickening of WAS

     •  Thickened sludge storage and blending

     •  Stabilization by anaerobic digestion

     •  Centrifuge dewatering

     •  Incineration



     Dewatered sludge would be disposed of as follows:



     •  50 percent of the dewatered sludge would be incinerated and the ash
        product  landfilled.


     •  50 percent of the dewatered sludge would be land  applied.



     The 50:50  ratio  is approximately  consistent  with current Jackson Pike

disposal practices.   In  this brief  analysis,  a comprehensive  review of alter-

nate ratios  to  determine  an optimum was not performed.   Since land  application

is not a limiting factor and the incinerators at Jackson Pike require  some

rehabilitation,  a split  equal to current  practices appears appropriate.
                                     B-15

-------
                                                                                        O
                                                                                        O
                                                                                        PQ

                                                                                  O
h-

u
                                     B-16

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
     Subjective screening of JP-B indicated that the alternative adequately

addressed the goals and objectives.   Therefore,  it  was  developed for a more

detailed evaluation.  Table 3 describes the facilities required and presents

the estimated costs to implement  JP-B.



2.1.3  Jackson Pike Sludge Management Alternative JP-C


     Figure 5 presents the sludge management  schematic  for  alternative  JP-C.

This alternative would consist of the following sludge processes.



     •  Gravity thickening of PS

     •  Centrifuge thickening of WAS

     •  Thickened sludge storage and blending

     •  Stabilization by anaerobic digestion

     •  Stabilization by thermal conditioning

     •  Centrifuge dewatering

     •  Incineration



     Dewatered sludge would be disposed of as follows:



     •  50 percent of the dewatered  sludge would be incinerated and the ash
        product  landfilled.


     •  50 percent of the dewatered  sludge would be land applied.



     As previously discussed,  the 50:50 disposal ratio  is consistent with

current practice.   The stabilization processes would each handle 50 percent of

the thickened sludges produced under normal operating conditions.  The

dewatered, thermally conditioned sludge would be incinerated while the

dewatered, digested sludge would  be  land  applied.



     Sludge management alternative JP-C was also determined by the subjective

screening to merit more detailed  consideration.  Table  4 describes the

facilities required and presents the estimated costs to implement JP-C.
                                     B-17

-------
                                   TABLE 3
                  JACKSON PIKE SLUDGE MANAGEMENT ALTERNATIVE
                          JP-B (Two-Plant Scenario)
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping              $1,967,000
     Modify two (2)digesters; 85-foot dia. x 10-foot SWD

Centrifuge Thickening WAS                                      $4,500,000
     Two (2) existing; 500 gpm
     One (1) new; 500 gpm

Thickened Sludge Storage/Blend
     Existing Facilities Reused

Anaerobic Digestion                                            $9,170,000
     Six (6) existing; 85-foot dia. x 23.5-foot SWD

Centrifuge Dewatering                                          $  490,000
     Six (6) existing; 1200 Ib/hr

Incineration
     Two (2) existing, 7 hearth, 200 wet ton/day @ 20% solids  $3,600,000

Landfill
     Contract operations included with O&M

Land Application
     Contract operations included with O&M                         —

                         Capital Cost                         $19,727,000

                         Annual Operation and                 $ 3,070,000
                           Maintenance Cost

                         Present Worth (JP-B Two-Plant)       $41,827,000
                                         B-18

-------

                                                                                         IT)
                                                                                                   OS
                                                                                                   _JU
                                                                                                     _J
                                                                                                     I-H
                                                                                                     j±.

                                                                                                     §
LJ
                                             B-19

-------
                                   TABLE 4
                  JACKSON PIKE SLUDGE MANAGEMENT ALTERNATIVE
                          JP-C (Two-Plant Scenario)
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping              $1,967,000
     Modify two (2) digesters; 65-foot dia. x 10-foot SWD

Centrifuge Thickening WAS                                      $4,500,000
     Two (2) existing; 500 gpm
     One (1) new; 500 gpm

Thickened Sludge Storage/Blend
     Existing Facilities Reused

Anaerobic Digestion                                            $7,750,000
     Six (6) existing; 85-foot dia. x 23.5-foot SWD

Thermal Conditioning
     Two (2) existing; 200 gpm                                 $3,000,000

Centrifuge Dewatering
     Six (6) existing; 1200 Ib/hr                              $  490,000

Incineration
     Two (2) existing; 7 hearth, 200 wet ton/day @ 20% solids  $3,600,000

Landfill
     Contract operations included with O&M

Land Application
     Contract operations included with O&M

                         Capital Cost                         $21,307,000

                         Annual Operations and                $ 3,770,000
                           Maintenance Cost

                         Present Worth (JP-C Two-Plant)       $48,597,000
                                       B-20

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
2.2  SOUTHERLY SLUDGE MANAGEMENT ALTERNATIVES  (TWO-PLANT SCENARIO)


     Six potential sludge management alternatives were  identified for  the

Southerly WWTP.  Each alternative  is discussed separately in the following

paragraphs.



2.2.1  Southerly  Sludge  Management Alternative SQ-A


     Southerly sludge management alternative SO-A is graphically depicted by

the schematic presented  in Figure  6.  Alternative SO-A would utilize the

following sludge processes:



     •  Gravity thickening of PS

     •  Centrifuge thickening of WAS

     •  Thickened sludge storage and blending

     •  Stabilization by anaerobic digestion

     •  Centrifuge dewatering

     •  Incineration



     Dewatered digested sludge would be incinerated and landfilled.



     Alternative SO-A was eliminated from further consideration for two basic

reasons.  First,  the alternative proposes  to abandon the existing compost

operations.   Such  a  move would forfeit  the substantial  investment the city has

placed in the relatively new  facilities and would substitute disposal of all

of the sludge product by landfilling in lieu of the  current  practice which

reuses a portion of the sludge as soil  conditioner.   Second, alternative SO-A

lacks the flexibility needed  to  allow the  city to modify disposal operations

subject to equipment failures or external pressures such as public dissatis-

faction or regulatory requirements.



2.2.2   Southerly Sludge Management Alternative SO-B

                                        i
     Figure 7 presents  the  sludge management schematic  for  alternative  SO-B.

The alternative would feature the following sludge processes:
                                     B-21

-------
                                                   3.
                                                   p gl
                                                   .-J*"1
                                                   w
Ul
U,
Ul
                                                   o
                                                   w
s_n
•^ >— i
QH-
<<
00
z !~*
au.
o £
Ul°£
"3
0
z
AERATIC




J
j


|

a
i— <
>- H
QC <

-------
1
1

1

I»°
£p
s*
QU
|Z*-«
3s
LJ°?
COfj
1
1
-r
ID
5
o:
t.i
1 *

1
1 r-
• "— '
^ S_ 1—
^^ p^
c* — '
1 ^^
1 d
1
1 |g
1^
IZK-
»— i rf
-Ju
LJ^
£»-
1

• I
1

i
(-
z
Ul
_J
u.
1 .
u«
u









i













,


i








•1MI















.^Ifc^^
—P^


























o
z
i=S
lo
CCX^
OFO.






1






h-
-7
LJ
ll








•••
























•••














Uu
v^z
U.Z ^
^u<
p^^^
^










'
ux
oi-


^-\
z
a
>-H
h-
uj <^L3z ^^a
1 rfljZl zSg U_IH-
rF-cnoD  n^n








u^ 1
o 5
pS
fey
o^t
K- "^
Z>
U Ul
a«
^^ 1
^^-|





B-23
 w
 o
 CO


 PQ

 13

 Z
a n

H£

 a
 a

-------
     •  Gravity Chickening of PS
     •  Centrifuge thickening of WAS
     •  Thickened sludge storage and blending
     •  Centrifuge dewatering
     •  Composting

     Ultimate sludge disposal would be accomplished through the marketing and
distribution of compost as a soil conditioner.

     The subjective evaluation eliminated alternative SO-B from further
consideration.   Flexibility to alter disposal operations  was the critical
factor in the evaluation.   Composting the entire volume of dewatered sludge
would mean a 2-3 fold increase in compost product over current conditions.  If
Southerly were operated in a one-plant scenario, 5-6 times the current compost
product would be produced.   An aggressive  and successful  marketing  program
would be mandatory to  locate and maintain sufficient receptors for the
compost.  The long-term reliability of an alternative which relies solely on
distribution of compost was not considered adequate to merit more detailed
development and  evaluation.

2.2.3  Southerly Sludge Management Alternative SO-C
     The sludge management schematic for alternative SO-C is presented in
Figure 8.  Southerly  sludge management alternative SO-C would consist of the
following sludge processes:

     •  Gravity thickening of PS
     •  Centrifuge thickening of WAS
     •  Thickened sludge storage and blending
     •  Stabilization by anaerobic digestion
     •  Centrifuge dewatering
     •  Composting
     •  Incineration
                                      B-24

-------
 I
 I
 I
 I
 I
 I
 I
I
I
I
 I
 I
 I
I
I
                                                                                                         00
                                                                                                             n to

                                                                                                             H!
                                                                                                               n
                                                                                                               o
                                                                                                            au.
                                                                                                            h§
                                                                                                              <
                LJ
                                                      B-25

-------
     Dewatered sludge would be disposed of as follows:
     •  75 percent of Che dewatered sludge would be incinerated,  and the ash
        product would be  landfilled.
     •  25 percent of the dewatered sludge would be composted and the compost
        would be distributed as a soil conditioner.
     The 75:25 ratio is approximately consistent with current Southerly
disposal practices.   The digestion facilities  would be sized to process that
portion of the sludge that would be incinerated.  The portion of the sludge
that would be composted would not receive stabilization prior to dewatering.

     Alternative SO-C represents current practice at Southerly when the
digestion facilities are operational.   Therefore, subjective screening
concluded that the alternative merits more detailed development and
evaluation.   Table 5 describes the facilities  required and presents the
estimated costs to implement  SO-C.

2.2.4  Southerly Sludge Management Alternative SO-D
     Southerly sludge management  alternative SO-D is graphically depicted by
the schematic presented in Figure 9.  Alternative SO-D would utilize the
following sludge processes.

     •  Gravity thickening of PS
     •  Centrifuge thickening of WAS
     •  Thickened sludge storage and blending
     •  Stabilization by anaerobic digestion
     •  Centrifuge dewatering
     •  Composting
     •  Incineration

     Ultimate disposal of  the sludge would be accomplished  through one of the
following disposal options.
                                      B-26

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
                                  TABLE 5
                   SOUTHERLY SLUDGE MANAGEMENT ALTERNATIVE
                          SO-C (Two-Plant Scenario)
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping               $2,520,000
     Four (4) existing; 45-foot dia. x 17-foot SWD


Centrifuge Thickening WAS                                      $2,000,000
     Four (4) existing; 250 gpm, 1250 Ib/hr
     One (1) new; 250 gpm, 1250 Ib/hr


Thickened Sludge Storage/Blend
     Existing Facilities Reused


Anaerobic Digestion                                            $4,280,000
     Six (6) existing; 85-foot dia. x 25.25-foot SWD


Centrifuge Dewatering                                          $5,120,000
     Six (6) existing; 1000 Ib/hr
     Two (2) new; 1000 Ib/hr


Dewatered Sludge Storage                                       $1,300,000
     One (1) new; 400 cy plus material handling


Composting
     Existing Facilities; 120 wet ton/day @ 20% solids


Incineration
     Two (2) new; 8 hearth, 260 wet ton/day @ 20% solids


Landfill
     Contract operations included with O&M


                         Capital Cost                        $15,220,000


                         Annual Operation and                $ 3,260,000
                           Maintenance Cost


                         Present Worth (SO-C Two-Plant)      $39,080,000
                                     B-27

-------
LJ
U.
U.
UJ
UJ
                                         B-28

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
     •  25 percent of Che sludge would be dewatered, composted, and distri-
        buted as a soil  conditioner.


     •  25 percent of the sludge would be digested, dewatered, and land
        applied.


     •  50 percent of the sludge would be digested, dewatered, incinerated,
        and landfilled.
     Alternative SO-D meets the goals and objectives of the subjective

screening.   The alternative offers continuation of the existing incineration

and composting processes at Southerly and introduces land application as a

disposal process.   The city has indicated there is adequate acreage suitable

for land application within an economically  feasible distance of  the  plant.

Alternative SO-D was advanced for further development and evaluation.  Table 6

describes the required facilities and presents the estimated costs to

implement SO-D.



2.2.5  Southerly Sludge Management Alternative SO-E


     Figure 10 presents the sludge management schematic for Alternative SO-E.

Southerly sludge management alternative SO-E would consist of the following

sludge processes:



     •  Gravity thickening PS

     •  Centrifuge thickening of WAS

     •  Thickened  sludge storage and blending

     •  Stabilization by anaerobic digestion

     •  Centrifuge dewatering

     •  Composting



     Dewatered sludge would be disposed of as follows:



     •  50  percent would be composted and distributed as a soil conditioner.
        Sludge sent to compost would not go  through the  digestion process.


     •  50  percent would be land applied as  a fertilizer to agricultural
        acreage within a reasonable distance from the plant.
                                     B-29

-------
                                  TABLE 6
                   SOUTHERLY SLUDGE MANAGEMENT ALTERNATIVE
                          SO-D (Two-Plant Scenario)
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping              $2,520,000
     Four (4) existing; 45-foot dia. x 17-foot SWD

Centrifuge Thickening WAS                                     $2,000,000
     Four (4) existing; 250 gpm, 1250 Ib/hr
     One (1) new; 250 gpm, 1250 Ib/hr

Thickened Sludge Storage/Blend
     Existing Facilities Reused

Anaerobic Digestion                                           $4,280,000
     Six (6) existing; 85-foot dia. x 25.25-foot SWD

Centrifuge Dewatering                                         $5,120,000
     Six (6) existing; 1000 Ib/hr
     Two (2) new; 1000 Ib/hr

Dewatered Sludge Storage                                      $1,300,000
     One (1) new; 400 cy plus material handling

Composting
     Existing Facilities; 120 wet ton/day @ 20% solids

Incineration
     Two (2) new; 8 hearth, 260 wet ton/day @ 20% solids

Landfill
     Contract operations included with O&M                        —

Land Application
     Contract operations included with O&M                        —

                         Capital Cost                        $15,220,000

                         Annual Operation and                $ 3,340,000
                           Maintenance Cost

                         Present Worth (SO-D Two-Plant)      $39,680,000
                                      B-30

-------
1
1


1

1 si
%<•
P>0
• Zf-«
1 S£
• s<
o
1
1
Is
£
0£
UJ
1 <


1
IS
1— 1
s- 1-
PRIMAR'
CLARIFICA
1
|5fc

W uuj
I oo


l
H
z
UJ
•^
^J
u.
»— 1


'

B-31

      a
"3
      Q.
      a.

-------
     Based on Che subjective evaluation alternative SO-E was eliminated from
further consideration.   The reliability of utilizing only compost distribution
and land application as ultimate disposal  options  did not appear  reasonable.
The plant currently practices incineration and relies heavily on incineration
and landfil-1 to dispose of sludge.   Furthermore,  it is critical that the plant
have a disposal method that  is completely  within their  control,  i.e.,  not
influenced by sludge quality, weather, market demand, public perception or
other external  pressures.

2.2.6  Southerly Sludge Management  Alternative SO-F
     Figure 11 presents the sludge  management schematic  for  Alternative  SO-F.
Ths sludge management system would  consist of the  following processes:

     •  Gravity thickening PS
     •  Centrifuge thickening WAS
     •  Thickened sludge storage and blending
     •  Centrifuge dewatering
     •  Composting
     •  Incineration

     Ultimate disposal of  the sludge would be accomplished through one of the
following disposal options.

     •  50 percent would be composted and distributed as a soil conditioner.

     •  50 percent would be incinerated and landfilled.

     Alternative SO-F is similar to alternative  SO-C with the exception that
digestion is not provided.   The  evaluation of  alternative SO-F  was prompted
due to the fact that digestion prior to incineration has normally not proven
to be cost-effective.  Although  digestion diminishes the amount of solids to
be handled in subsequent processes, the heat content of digested sludge is
significantly reduced.   Furthermore, digested sludge tends to be more
                                      B-32

-------
1
1


1


• >.§
£p
<<
• *r
1 s£
• ui9?
/w ^»
°_j
1 °
1
• S
• a
I 5
a
1 . 1
1 *

1
1 °
• •— •
• v. i—
^^ i
a<

-------
difficult Co dewater than combined  raw sludges.  These factors cause digested
sludge to be more difficult,  and consequently more expensive on a unit basis
(i.e. dollars  per dry  ton),  than raw sludges to incinerate.  Since the
Southerly plant has a portion of the required digestion facilities and
adequate incineration facilities in place, the cost effectiveness of digestion
prior to incineration is less dependent on capital cost than an evaluation
where these  facilities  are not  in place.

     Table 7 describes  the  required  facilities and presents the estimated
costs to implement SO-F.

2.3  SOUTHERLY SLUDGE MANAGEMENT ALTERNATIVES (ONE-PLANT SCENARIO)
     The three sludge  management alternatives that were advanced from the
subjective screening phase for the Southerly two-plant scenario are considered
viable for Southerly one-plant scenario.  These three alternatives were
previously identified as SO-C, SO-D, and SO-F.  The remaining three
alternatives, which were identified for the two plant scenario, are not
considered viable for the one-plant scenario for the same reasons previously
discussed.

     The sludge management schematics  for alternatives SO-C, SO-D, and SO-F
have been presented in Figures 8, 9, and  11 respectively.   Table 8 identifies
the required facilities and presents the estimated cost to implement sludge
management  alternative  SO-C under a one-plant  scenario.  Table 9  presents the
facilities and estimated costs  to implement  SO-D  under a  one-plant  scenario.
Table 10 presents the  facilities and estimated costs  to implement SO-F under a
one-plant scenario.
                                      B-34

-------
I
I
I
I
                                             TABLE 7
                              SOUTHERLY SLUDGE MANAGEMENT ALTERNATIVE
                                     SO-F (Two-Plant Scenario)
                                   Facilities and Estimated Costs
          •Gravity Thickening PS plus Dilution Water Pumping             $2,520,000
               Four (4)  existing;  45-foot dia. x 17-foot SWD

_        Centrifuge Thickening WAS                                    $2,000,000
•             Four (4)  existing;  250 gpm,  1250 Ib/hr
•             One (1)  new;  250 gpm, 1250 Ib/hr

•        Thickened Sludge Storage/Blend
               Existing Facilities Reused

          Centrifuge Dewatering                                        $8,750,000
               Six (6)  existing;  1000 Ib/hr
               Four (4) new;  1000 Ib/hr
               One (1)  new;  400 cy plus material handling
 I
 •        Dewatered  Sludge  Storage                                      $1,300,000


 I

 I
          Composting
               Existing Facilities;  120 wet ton/day @ 20% solids

          Incineration
               Two (2) new;  8 hearth,  260 wet ton/day @ 20% solids

          Landfill
               Contract operations included with O&M
                                   Capital Cost                       $14,570,000
                                     Maintenance Cost

                                   Present Worth (SO-C Two-Plant)     $42,770,000
                                    •Annual  Operation and                $ 3,940,000
                                      M*3-Jr»f-ai"ianr»a  ("*/•» o f-

I

I

I

I

I

I
                                               B-35

-------
                                  TABLE 8
                   SOUTHERLY SLUDGE MANAGEMENT ALTERNATIVE
                          SO-C (One-Plant Scenario)
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping              $5,070,000
     Four (4) existing; 45-foot dia. x 17-foot SWD
     Two (2) new; 85-foot dia. x 10-foot SWD

Centrifuge Thickening WAS                                     $5,600,000
     Four (4) existing; 250 gpm, 1250 Ib/hr
     Four (4) new; 250 gpm, 1250 Ib/hr

Thickened Sludge Storage/Blend
     Existing Facilities Reused

Anaerobic Digestion                                          $11,460,000
     Six (6) existing; 85-foot dia. x 25.25-foot SWD
     Four (4) new; 85-foot dia. x 25.25-foot SWD

Centrifuge Dewatering                                        $21,040,000
     Six (6) existing; 1000 Ib/hr
     Nine (9) new; 1000 Ib/hr

Dewatered Sludge Storage
     One (1) new; 400 cy plus material handling               $1,300,000

Composting
     Existing Facilities; 120 wet ton/day @ 20% solids

Incineration
     Two (2) new; 8 hearth, 260 wet ton/day @ 20% solids
     Rehabilitate existing                                    $1,300,000

Landfill
     Contract operations included with O&M                        —

                         Capital Cost                        $45,770,000

                         Annual Operation and                $ 6,080,000
                           Maintenance Cost

                         Present Worth (SO-C One-Plant)      $89,590,000
                                      B-36

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
                                  TABLE 9
                   SOUTHERLY SLUDGE MANAGEMENT ALTERNATIVE
                          SO-D (One-Plant Scenario)
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping              $5,070,000
     Four (4) existing; 45-foot dia. x 17-foot SWD
     Two (2) new; 85-foot dia. x 10-foot SWD


Centrifuge Thickening WAS                                     $5,600,000
     Four (4) existing; 250 gpra, 1250 Ib/hr
     Four (4) new; 250 gpm, 1250 Ib/hr


Thickened Sludge Storage/Blend
     Existing Facilities Reused


Anaerobic Digestion                                          $11,460,000
     Six (6) existing; 85-foot dia. x 25.25 foot SWD
     Four (4) new; 85-foot dia. x 25.25 foot SWD


Centrifuge Dewatering                                        $21,040,000
     Six (6) existing; 1000 Ib/hr
     Nine (9) new; 1000 Ib/hr


Dewatered Sludge Storage
     One (1) new, 400 cy plus material handling               $1,300,000


Composting
     Existing Facilities; 120 wet ton/day @ 20% solids


Incineration
     Two (2) new; 8 hearth, 260 wet ton/day @ 20% solids
     Rehabilitate existing                                    $1,300,000


Landfill
     Contract operations included with O&M                        —


Land Application
     Contract operations included with O&M                        —


                         Capital Cost                        $45,770,000


                         Annual Operation and                $ 6,230,000
                           Maintenance Cost


                         Present Worth (SO-D One-Plant)      $90,710,000
                                     B-37

-------
                                  TABLE 10
                   SOUTHERLY SLUDGE MANAGEMENT ALTERNATIVE
                          SO-F (One-Plant Scenario)
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping             $5,070,000
     Four (4) existing; 45-foot dia. x 17-foot SWD
     Two (2) new; 85-foot dia. x 10-foot SWD

Centrifuge Thickening WAS                                     $5,600,000
     Four (4) existing; 250 gpm, 1250 Ib/hr
     Four (4) new; 250 gpm, 1250 Ib/hr

Thickened Sludge Storage/Blend
     Existing Facilities Reused

Centrifuge Dewatering                                        $27,430,000
     Six (6) existing; 1000 Ib/hr
     Fourteen (14) new; 1000 Ib/hr

Dewatered Sludge Storage
     One (1) new; 400 cy plus material handling               $1,300,000

Composting
     Existing Facilities;  120 wet ton/day @ 20% solids

Incineration
     Two (2) new; 8 hearth, 260 wet ton/day @ 20% solids
     Rehabilitate existing                                    $1,300,000

Landfill
     Contract operations included with O&M                        —

                         Capital Cost                        $40,700,000

                         Annual Operation and                $ 7,110,000
                           Maintenance Cost

                         Present Worth (SO-F One-Plant)      $92,440,000
                                      B-38

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
               3.  EVALUATION OF SLUDGE MANAGEMENT ALTERNATIVES



     Sludge management alternatives were evaluated based on cost-

effectiveness, dewatering, system redundancy, and ultimate disposal.  Facility

planning information is included for each of the criteria.  The results of the

evaluation are discussed in the following sections.



3.1  COST EFFECTIVENESS OF SLUDGE MANAGEMENT ALTERNATIVES


     Table 11 presents the potential sludge management alternatives and the

associated present worth of each.  These alternatives and the present worth

costs will be utilized in a subsequent briefing paper to assess the cost

effectiveness of the one-plant and two-plant scenarios.



     Alternative JP-B, which provides for digestion, dewatering, and a 50:50

split of the sludge to land application and incineration and landfill, is the

cost-effective sludge management scheme at Jackson Pike.  This alternative is

approximately 16 percent less costly than JP-C which proposes to retain the

thermal conditioning units for processing a portion of the sludge.



     The lowest present worth of the Southerly two-plant alternatives is

exhibited by SO-C.  Practically speaking, however, present worth of SO-D is

considered equal to that of SO-C.  At this level of planning analysis the 1.5

percent present worth difference is not a significant factor in selection of

an alternative.  In light of this fact, SO-D is the recommended sludge

management alternative for the Southerly two-plant scenario.  Alternative SO-D

offers more flexibility in that three disposal methods are utilized (i.e.,

marketing of a compost product, land application of dewatered, digested

sludge, and landfilling of incinerator ash.)  SO-C on the other hand utilizes

only two of the disposal options, not providing for land application.



     Alternative SO-F was developed to evaluate the cost effectiveness of

digestion prior to incineration.  SO-F proposes dewatering of raw sludge with

approximately 50 percent of the dewatered cake incinerated.  Alternative SO-F
                                      B-39

-------
              TABLE 11
     PRESENT WORTH COMPARISON
OF SLUDGE MANAGEMENT ALTERNATIVES
SCENARIO
TWO-PLANT



ONE-PLANT
ALTERNATIVE
JP-B
JP-C
SO-C
SO-D
SO-F
so-c
SO-D
SO-F
PRESENT WORTH
$ 41,827,000
$ 48,597,000
$ 39,080,000
$ 39,680,000
$ 42,777,000
$ 89,590,000
$ 90,710,000
$ 92,440,000
RECOMMENDED
ALTERNATIVES


JP-B + SD-D
$ 81,507,000

SO-D
$ 90,710,000
 ALTERNATIVE PROCESS / DISPOSAL INDEX

DIGESTION
THERMAL CONDITIONING
DEWATERING
INCINERATION
COMPOST
LAND APPLICATION
LANDFILL
REFERENCE
JP-B
•

•
•

•
•
FIG, 4
JP-C
•
•
•
•

•
•
FIG. 5
so-c
•

•
•
•

•
FIG. 8
SO-D
•

•
•
•
•
•
FIG. 9
SO-F


•
•
•

•
FIG. 11
               B-40

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
differs from  SO-C only  in  Chat  the digestion  provided  in  SO-C  is not included
in SO-F.  From Table  11  it can  be seen that alternative SO-F  (i.e., incinera-
tion without  digestion) exhibits a present worth approximately 9 percent
higher  than alternative  SO-C.

     Digestion prior  to incineration has proven to be  cost effective in this
case primarily due  to the  sunken  capital invested  in  the  Southerly facilities.
Southerly has six existing anaerobic digesters and four multiple hearth
incinerators  in place (i.e., two existing and  two in startup).   The new
incinerators are equipped with a waste heat recovery system which reclaims
waste heat from the incinerators to meet digestion and building heat require-
ments.  The waste heat  recovery system allows for  the  digester gas produced to
be used as a  fuel for the  incinerators, thus substantially reducing the
supplemental  fuel requirements  of the  incinerators.

     The existing digestion, incineration,  and waste heat recovery facilities
conservatively represent 20-25 million dollars of  sunken  capital.  If these
facilities were not in place,  the  required  additional capital costs would be
sufficient to show  incineration of raw sludge  to be more  cost  effective than
digestion prior to  incineration.

     Under the one-plant scenario, sludge management alternatives SO-C and
SO-D represent the  lowest  present worth options.  Again digestion prior to
incineration (SO-C)  is a lower cost alternative than digestion of raw sludge
(SO-F).  However,  the  difference between SO-C and SO-F has been diminished to
approximately 3 percent.  This smaller difference is due to the fact that four
new digesters are required in the one-plant scenario.  The cost of the four
new digesters weakens the  impact  of  the sunken capital in the  cost effective-
ness analysis and thus lowers the present worth difference.

     As with  the Southerly two-plant scenario, alternative SO-D is  recommended
as the preferrable  Southerly one-plant  sludge  management scheme.   For the
reasons previously discussed,  SO—D provides  the city  with three reliable
disposal paths and adequate flexibility.
                                     B-41

-------
3.2  SLUDGE DEWATERING
     The RFPU evaluated the sludge dewatering component of the various
alternatives in light  of the  existing  centrifuge  equipment currently in
service at Southerly.   The  design  criteria  for  the dewatering  centrifuges  was
revised a number of times  over the course of the planning and design.
Facilities planning documents indicate the centrifuges will be rated at
120 gpm with a  feed solids of 5 percent, or approximately 3000 Ibs/hr.
Subsequently,  the GERBOD revised  the  design criteria for the centrifuges to
1000 Ibs/hr.   Based on a feed solids of  4  percent,  the GERBOD  assumed a
dewatered cake of 20-21 percent could be produced.   The GERBOD further
indicated that the successful,  efficient operation of  the  dewatering process
is critical to the overall cost of sludge  processing and disposal.   The GERBOD
noted that increasing  the  solids  content of the dewatered  cake reduces
incinerator fuel consumption and  subsequent handling costs,  and  increases  the
efficiency of  downstream processes.  The GERBOD concluded  recommending that
alternative dewatering equipment (specifically belt presses  and diaphragm
plate and frame (DPF)  presses)  be  fully evaluated to optimize  the  sludge
processing scheme.

     As a result of the GERBOD's  recommendations,  pilot scale  testing of
dewatering equipment was conducted.   The pilot  testing and subsequent
dewatering evaluations were documented in the Preliminary Design Evaluation of
Sludge Dewatering, December 12, 1986.  The evaluation  acknowledges  that the
tests were carried out under less than optimum conditions.  Tests  were
performed on unthickened,  undigested  sludge of  indeterminate composition.   The
proportions of primary and waste activated sludge fed to the dewatering
devices could only be approximated.

     The dewatering evaluation selected  the diaphragm plate and frame press as
the optimum dewatering alternative.   The evaluation  recommended  installation
of four DPF presses in Project 88 and the  future  installation  of  an additional
five DPF presses—to provide a total  of  nine presses.   The six (6)  existing
                                     B-42

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
dewatering centrifuges  will become standby units after the Project 88 improve-

ments and will eventually be abandoned as the treatment plant project proceeds

and the  remaining  DPF  presses  are  installed.



     The dewatering evaluation also recommended that the diaphragm plate and

frame presses be located  in the existing thermal conditioning building with

appropriate modifications to that structure.   The estimated cost for  implemen-

ting the DPF recommendation, presented in the dewatering evaluation,  was

approximately $22,000,000   The cost estimate previously presented in the

facilities plan and utilized in the cost-effective evaluation for implementing

the centrifuge dewatering alternative was approximately  $12,000,000.   Both of

these estimates are based on  the one-plant scenario.



     In the evaluation of dewatering alternatives the capacity of the

centrifuges was again revised.  Based on the interpretation of pilot test

results, the capacity of the centrifuges was established at 700-750  Ib/hr.  As

a result, 17 centrifuges  (i.e.,  14  operating,  3  standby)  were needed  to

dewater approximately  240,000 Ibs/day of sludge.



     Following the pilot testing, one of the existing centrifuges was modified

and upgraded to allow a full-scale  test.  The feed  sludges  used were still not

representative of the future anaerobically digested sludge.  Review of the

data from this full-scale demonstration indicates  that  the  modified centrifuge

could process in excess of  1000  Ibs/hr  (i.e.,  1300-1700 Ibs/hr) on various

blend ratios of the existing sludge.



     Due to the sunken investment in the six existing centrifuges, the

established  design  capacity  of  modified  units is  important  to the selection of

the optimum dewatering  alternative.
                                     B-43

-------
     Based on the data contained in the Preliminary Design Evaluation of
Sludge Dewatering, use of less than 1000 Ibs/hr as the  rated capacity of the
modified centrifuge seemed unusually conservative.  Consequently,  an
independent cost-effective evaluation of dewatering  was  performed  using
1000 Ibs/hr as the design capacity.  This  independent analysis was  performed
based on the Southerly one-plant scenario.   The results  of this  analysis are
summarized  in Table 12.  Since the effectiveness of the  dewatering devices
impact downstream processing units, the operational costs  of incineration and
ash disposal have been included in the cost-effective  analysis.  The
centrifuge dewatering option at $40,800,000  exhibits a 7 percent lower present
worth than  the DPF option  at  a  present  worth  of  $43,600,000.

     As  a result of the higher rated capacity  of  the centrifuges,  fewer units
would be required. Fifteen units,  12 operating and  3 standby would be
adequate.   Assuming the thermal  conditioning building was  the  logical location
for the  dewatering facility,  a smaller expansion of that structure  would be
necessary.    Fewer centrifuges and  associated  equipment  and less building
expansion result in Che estimated cost  of  centrifuge dewatering approximately
equal to that of the DPF press option  (i.e., $22,000,000).

     The operation and maintenance costs associated  with the two dewatering
alternatives are reasonably consistent  with  those developed in the  Preliminary
Design Evaluation of  Sludge Dewatering. The DPF  presses are approximately
45 percent more expensive to operate and maintain primarily due  to  higher
labor costs and higher chemical costs.

     The centrifuges  will provide a 20 percent cake solids concentration,
whereas the DPF presses will  provide a  25 percent cake solids concentration.
Consequently, the operating cost of incineration  is  approximately  80 percent
higher for the centrifuge dewatered sludge.  The  supplemental  fuel  required  to
burn off the additional water is the major  reason for this difference.  From
Table 12, it can be seen that  operational  costs for the  incineration process
are $500,000 higher under the  centrifuge dewatering alternative.    The
Preliminary Design Evalution of Sludge  Dewatering identified a $750,000
                                     B-44

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                        TABLE 12
              PRESENT  WORTH COMPARISON
             OF DEWATERING ALTERNATIVES
PROCESS
DEWATERING
INCINERATION
ASH DISPOSAL
TOTAL
PRESENT
WORTH
CENTRIFUGE _
ALTERNATIVE
CAPITAL
$ 21,040,000
$ 0
$ 0
$ 21,040,000
0 & M
$ 1,300,000
$ 1,100,000
$ 330,000
$ 2,730,000
$ 40,800,000
DPF PRESS ,
ALTERNATIVE
CAPITAL
$ 21,920,000
$ 0
$ 0
$ 21,920,000
0 & M
$ 1,910,000
$ 600,000
$ 490,000
$ 3,000,000
$ 43,600,000
BASED DN 15  CENTRIFUGES <12 OPERATING AND 3 STANDBY),  RATED
CAPACITY DF  1,000 LBS / HR / UNIT, PRODUCING CAKE SOLIDS
CONCENTRATION  OF  20-PERCENT.
BASED DN 9 DPF PRESSES <7 OPERATING AND 2 STANDBY),  RATED
CAPACITY OF 35,000 LBS  / DAY / UNIT EXCLUDING PRECOAT SOLIDS,
PRODUCING CAKE SOLIDS CONCENTRATION OF  25-PERCENT.
                       B-45

-------
difference in incinerator operating costs, also with the centrifuge dewatering
alternative being higher.


     The facilities planning documents and the Preliminary Design Evaluation

of Sludge Dewatering do not address the ultimate  disposal  of  incinerator  ash.
As described in Section 1 of this briefing paper, currently ash is stored in

on-site ash lagoons and periodically removed to a landfill site.  Due to  the

fact that a substantial quantity of inert solids are added to the sludge  under

the DPF press alternative, a larger quantity of ash is produced.  Consequently,
the costs associated with ash disposal will be higher  for  the DPF press

alternative.   For purposes of this analysis,  an ash disposal cost of $15  per
cubic yard was utilized.  Based on this unit cost and the projected ash

quantities, ash disposal under the DPF press alternative will be $160,000

(i.e., approximately  50  percent)  more  costly than ash disposal for the

centrifuge dewatering alternative.


     The briefing paper analysis of dewatering alternatives reached a

different conclusion than the  Preliminary Design Evaluation of Sludge

Dewatering for several  reasons.  These reasons are briefly discussed below.


     •  Use of the higher capacity rating for the centrifuges in the briefing
        paper analysis,  resulted in lower capital costs for the centrifuge
        alternative.

     •  Although the Preliminary Design Evaluation of Sludge Dewtering
        projected higher operating costs for the DPF presses than the
        centrifuges (i.e., approximately  15  percent  higher),  this difference
        was increased to 45 percent in the briefing paper analysis.

     •  The supplemental  fuel required by the incineration process was higher
        for both alternatives in the briefing paper analysis.   This is due to
        the fact that heat value of digested sludge was taken as 8000 BTU/lb
        of volatile solids.   In the Preliminary Design Evaluation the heat
        value of digested sludge was taken as 10,000 BTU/lb of volatile
        solids.   This difference in sludge heat value necessitated that
        supplemental fuel be added to  the DPF press alternative in the
        briefing paper analysis, whereas in  the Preliminary Design Evaluation
        the DPF press dewatered  solids required no supplemental  fuel.

        The two analyses also utilized different unit costs for supplemental
        fuel (No.  2 fuel oil).   The Preliminary Design Evaluation used  $1.05
                                     B-46

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
I
I
        per gallon.  The  briefing  paper  analysis used $0.85 per gallon based
        on telephone conversations with  fuel  suppliers.  Reviewing the 1985
        Operating Report  for the Division of Sewerage  and Drainage indicated
        that Southerly was  purchasing  fuel oil  at  a  cost  of $0.66  per  gallon.


        The net  impact  of both of these differences (i.e., heat value of
        digested sludge and cost of fuel oil) was  that the briefing paper
        analysis estimated  less of an  economic  advantage  for  the DPF press
        alternative in  the  incineration  process.


     •  Lastly,  the briefing paper analysis  included  a cost for ash disposal,
        whereas  the preliminary design evaluation did  not.  Since more ash is
        produced with the DPF  presses, this slightly favored  the centrifuge
        alternative in  this cost-effective analysis.
     From the above analyses  it  is evident that  the selection of a dewatering

alternative is sensitive to the capacity criteria established for the devices

being evaluated and the final sludge cake solids concentrations these units

can produce.  In light of this sensitivity,  it appears reasonable to conduct

testing programs on sludges similar  to that which will be processed in the

future (i.e., in  this  case  anaerobically  digested)  to  provide a representative

picture of probable equipment performance.  If such testing is not possible

for whatever reason, selection of conservative design criteria appears

justified for the initial project phase.   However,  the six existing modified

centrifuges should be  evaluated with anaerobically digested sludge prior to

abandoning these units and implementing  the final project phases.



3.3  PLANNED SYSTEM REDUNDANCY


     The facilities planning documents recommended a sludge  management

alternative which provided redundancy in accordance with Table 13.  The table

is based on a Southerly one-plant scenario.



     The recommended alternative calls for 22  percent  of  the dewatered sludge

to be composted under normal  conditions.   The compost facility has a capacity

to handle as much as 55 percent of the sludge under ideal conditions.   Ideal

conditions relate to total solids content of dewatered sludge, favorable

weather conditions for composting, and adequate demand for the compost
                                     B-47

-------
                TABLE 13.  SLUDGE MANAGEMENT SYSTEM REDUNDANCY
                      FOR FACILITIES PLAN RECOMMENDATION
     Process/Disposal
Average
Annual
Maximum
Capacity
                                              Values as a percentage of
                                              annual sludge production
     Compost/Sales & Distribution

     Digest/Land Apply

     Digest/Incinerate/Landfill

     Lime Stabilization/Land Apply
22
19a
59a
	 c
55
67
80b
	 c
a  Digestion to handle total of  sludge  incinerated  and  land applied,  i.e.,
   approximately 80 percent of average annual sludge production.

   Incineration to provide complete redundancy for either composting or land
   application processes.

c  Lime stabilization is proposed by the facilities plan as a backup process,
   however, the sizing criteria and the need for these facilities are not
   clear.
                                      B-48

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
product.  The compost facility is planned to operate in a range of 120 wet ton
per day on the low side,  up to more than 240 wet ton per day at maximum.

     Approximately 20 percent  of  the average annual sludge production would be
digested, dewatered, and land  applied.   The current appplication program has
been successful, and the city  anticipates that the demand for the dewatered
sludge product will remain.   The extent to which  land application can
function as a disposal option  is subject to several factors including weather
conditions and cropping patterns.  Consequently,  the amount of sludge which
will be  land  applied  is  expected  to  vary substantially  throughout  the year.
The program will operate from  virtually no land application when factors
preclude application to a maximum of 60-70  percent of the average annual
sludge production (on a daily basis)  being  land applied  during favorable
application circumstances.

     Approximately 60 percent  of  the annual sludge production would be
incinerated.   The planning documents indicate  the  incineration facility,
however, would  be sized  to handle a maximum of as  much  as 80 percent of the
sludge production.   This  additional 20-percent  would function as a valuable
backup for either the composting or the land application disposal option.  In
the event either of these options are unable to process  and dispose of their
planned portion of the average annual sludge production, incineration and
landfilling would be available to alleviate the problem.  The incineration/
landfill option would be expected to routinely backup the land appplication
option for reasons previously discussed. The composting option would be
expected to perform more consistently  than  land application.  If incineration
were needed as a backup to  composting,  it should be on a scheduled basis at a
time when  land  application could  reasonably be expected to provide disposal
for a minimum of 20 percent of the sludge production.

     In addition to the three  processing and disposal options discussed above,
the facilities plan  recommends that lime stabilization and land application be
provided as a backup to other  ultimate  disposal options.  Figure 5.3 of  the
RFPU indicates that  the lime  facilities would  be  sized only to backup the
                                     B-49

-------
compose process, however, the details are not adequate to determine what is
proposed.   Furthermore,  the  need for this additional redundancy has not been
justified.   Recent correspondence with  the  city  (i.e.,  May 29,  1987, URS
Dalton Responses to May  12,  1987 Comments) indicates the  recommendation of
lime facilities has not been finalized completely and  is being reevaluated due
to the cost of  these  facilities.

     In this briefing paper  analysis the redundancy issue was considered
relative to the existing and new incineration facilities at Southerly.  The
two new incinerators  at Southerly will be capable of incinerating approximate-
ly 525 wet tons per day of dewatered cake at 20  percent  solids.  If a
dewatered cake solids of 25 percent can be realized, these two units would be
capable of incinerating  approximately 560 wet tons per day.   The  two existing
incinerators,  which according to the planning documents will be rehabilitated
under a one-plant option,  are  capable of incinerating 320 wet tons per day at
20 percent solids.   Again,  if a 25 percent cake solids concentration can be
obtained,  these units would  be capable of 350 wet ton  per day.   For purposes
of comparison, the total dewatered  sludge cake production of Southerly under a
one-plant scenario assuming  all sludge  was directed  to incineration would be
approximately 510 wet tons per  day  at 20  percent solids and 410 wet tons per
day at 25 percent solids.    With one new  (larger) incinerator out of service,
the remaining three incinerators can handle 15 percent more sludge at 20
percent solids than the  one-plant option can produce.  With  a dewatered cake
concentration of 25 percent  solids,  these three incinerators could process
more than 50 percent  more sludge  than the one-plant  option can generate.

     Based on the above  analysis, the incineration process offers sufficient
redundancy to allow processing and disposal  of  all sludge produced even when
the composting and land application options are  inoperative and one of the
larger incinerators is out of service.   In light  of the redundancy inherent in
the incineration process,  the need  for greater redundancy does  not appear
justified.
                                     B-50

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
3.4  ULTIMATE DISPOSAL PLAN


     The facilities plan proposes three basic methods for ultimate

use/disposal of the wastewater  sludges.  They are:



     •  Distribution and Marketing of Composted Sludge

     •  Land Application of Digested, Dewatered Sludge

     •  Landfilling of Incinerated, Dewatered Sludge



     The plan, however, does not offer many details relating  to  the operation,

costs, and planned reliability associated with these options.  The following

paragraphs briefly present the current understanding of the use/disposal

options.



3.4.1  Distribution and Marketing of Composted Sludge


     Dewatered, undigested sludge is transported by the city in trucks to the

composting facility.  The city operates and maintains the compost facility

which most recently has been processing approximately 120 wet tons/day of

dewatered sludge from  Southerly.  Conversations with city personnel have

indicated that the composting facility costs approximately  $1,200,000 per year

to operate.   The 1985 Operating  Report  published by the Division of Sewerage

and Drainage shows the 1985 operating budget for the compost facility to be

$2,000,000.   Based  on these costs and the total  production of the composting

facility, a unit cost of $26-40  per wet ton of sludge composted  is estimated.



Currently compost is disposed of through mine reclamation projects, bulk and

residential package sales, and nursery and institution use.   The city has an

active marketing program and anticipates that future demand will be adequate

to dispose of the compost  produced.



     The composting facility has been cited as a source of odors by the

community.  The city believes  that  most of  the odor problems can be attributed

to the moisture content in the raw  sludge and problems  with  the composting

equipment.   The  city  anticipates  that  future  operations  will increase
                                      B-51

-------
dewatered solids concentration and reduce the potential for odors from the
facility.

3.4.2  Land Application of Digested, Dewatered Sludge
     The current land application program originates from the Jackson Pike
plant.  Currently land application is conducted on a contract basis and it  is
expected that this practice will continue in the future.  Based on
conversations with city personnel, the current cost of  land appplication  is
approximately $12 per wet ton of sludge applied.  The contractor is
responsible for transport and spreading the sludge and  for remote sludge
storage if necessary.

     Based on the earlier EIS (1979), adequate suitable acreage for sludge
application is available within a reasonable distance of the plant site.  The
current program is subject to substantial variation in  peak and off peak
application rates due to weather and crop constraints.  Recent discussions
have indicated that remote sludge storage has become limited as farmers have
decided to accept the sludge only if it is spread immediately.  The ash
lagoons at Jackson Pike have been utilized to provide temporary storage and
relieve the pressure this situation has created.

     The future land application program should be planned and administered by
the city in such a way as to ensure the reliability of  the agricultural use of
sludge.

3.4.3  Landfilling of Incinerated Dewatered Sludge
     At both Jackson Pike and Southerly, incinerator ash is temporarily stored
in on-site ash lagoons.  The lagoons are cleaned on an  as-needed basis with
the ash being transported and deposited in a landfill by a contractor.  The
cost of ultimate disposal of ash has not been  identified.  These costs vary
significantly depending on local availability  of landfills, transport
distances, and composition of the ash.  Deposit charges alone  can  range
between $5-20 per cubic yard and may be substantially higher depending on
local  conditions and ash quality.
                                       B-52

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     Conversations with the city have indicated that only the city-owned

landfill is accepting the incinerator ash.  Details relating to the projected

useful life of this landfill are not contained in the facilities plan.  For

the EIS to review the reliability of the  landfill disposal option, the  city

must furnish planning information documenting the steps being taken to  ensure

a suitable disposal site will be available.
                                      B-53

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
         APPENDIX C
    BRIEFING PAPER NO.  3
BIOLOGICAL PROCESS SELECTION

-------
I
I
I
           BRIEFING PAPER NO. 3
I
           BIOLOGICAL PROCESS SELECTION
I
I          Supplemental Environmental Impact Statement
           USEPA Contract No.  68-04-5035,  D.O.  No.  40
|          Columbus Ohio Waste-water  Treatment  Facilities
I
I
I
I
I
•          Prepared By:
|          SCIENCE  APPLICATIONS  INTERNATIONAL  CORPORATION
•          TRIAD  ENGINEERING INCORPORATED
I
I

-------
                         BIOLOGICAL PROCESS SELECTION
1.   TERMS AND DEFINITIONS


2.   BRIEFING PAPER ASSUMPTIONS


3.   PROCESS EVALUATION

     3.L  Process Description
          3.1.1  Design Criteria
                 3.1.1.1  Aeration Basins
                 3.1.1.2  Trickling Filters
                 3.1.1.3  Clarifiers
          3.1.2  Recommended Sizing
                 3.1.2.1  Southerly Two-Plant Semi-Aerobic
                 3.1.2.2  Southerly Two-Plant Trickling Filter/Activated Sludge
                 3.1.2.3  Jackson Pike Two-Plant Semi-Aerobic
                 3.1.2.4  Jackson Pike Two-Plant Trickling Filter/Activated Sludge
                 3.1.2.5  Southerly One-Plant Semi-Aerobic
                 3.1.2.6  Southerly One-Plant Trickling Filter/Activated Sludge

     3.2  Technical Evaluation
          3.2.1  Reliability
          3.2.2  Flexibility

     3.3  Environmental Evaluation

     3.4  Cost Evaluation


4.   COMPARISON OF BRIEFING PAPER AND FACILITY PLAN CONCLUSIONS

     4.1  Process Selection

     4.2  Clarifier Utilization

     4.3  One-Plant versus Two-Plants

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                                 INTRODUCTION



     Under the direction of USEPA,  a series of briefing papers are being

prepared addressing key issues in the development of the Supplemental

Environmental Impact Statement for the Columbus,  Ohio,  Wastewater Treatment

Facilities.   The  briefing  papers  form the basis of discussions between USEPA

and their consultants to resolve these key issues.  The following paragraphs

present the background of the facility planning process, a description of the

briefing papers,  and the purpose of this paper on biological process

selection.



FACILITY PLANNING PROCESS


     At the time this paper was  prepared  (March-July 1987) the city of

Columbus was proceeding to implement improvements at the Jackson Pike and

Southerly Wastewater Treatment Plants to comply with more stringent effluent

standards which must be met by July 1,  1988.    These improvements were based

on the consolidation of wastewater treatment operations at the Southerly

plant.  This  one-plant  alternative  is  a  change  from  the two-plant operation

proposed by the city in  the  1970's and  evaluated  in the 1979  EIS.



     The development and documentation of wastewater treatment process and

sludge management  alternatives for the Columbus metropolitan  area has been an

extended and iterative process.    The design and construction of various

system components  have  progressed, because of the 1988  deadline,  while

planning issues continue to be resolved.  As  a  result,  numerous  documents have

been prepared which occasionally revise a previously established course of

direction.



     The concurrent resolution of planning issues and implementation of

various project components has made preparation of the  EIS  more difficult

because final facility plan recommendations are not available in a single

document.
                                   C-l

-------
BRIEF-ING PAPERS
     To facilitate preparation of the EIS,  a series of briefing papers are
being developed.   The purpose of the briefing papers is to allow USEPA to
review the work of the EIS consultant and to identify supplemental information
necessary for the preparation of the EIS.   Six briefing papers are being
prepared as follows:

     •  Flows and Loads
     •  Sludge Management
     •  Process Selection
     •  CSO
     •  One Plant vs. Two Plant (Alternative Analysis)
     •  O&M and Capital Costs

     The specific focus of each briefing paper will be different.  However,
the general scope of  the  papers will adhere to the following format:

     •  Existing conditions will be documented.
     •  Evaluations,  conclusions,  and recommendations of the facilities
        planning process will  be reviewed  using available documentation.
     •  Where appropriate, an  independent evaluation of the future situation
        and viable alternatives will be  prepared.
     •  The facility plan and EIS briefing paper conclusions will be  compared.

     The briefing paper process is  intended  to:

     •  Prompt the resolution of any data deficiencies.
     •  Clearly establish and define existing and future conditions.
     «  Identify the final recommended plan which the city desires to implement.
                                    C-2

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     •  Provide a data base of sufficient detail to allow preparation of the
        draft EIS.
BIOLOGICAL PROCESS SELECTION


     This Briefing Paper presents an evaluation of three different biological

processes selected by the city of Columbus, for use at the Jackson Pike and

Southerly Wastewater Treatment  Plants  (WWTP).  The scope  of this report is to

review data made available by the city's consultant,  identify issues and data

gaps,  aid USEPA in the decision  making process, and focus on future data needs

so that a complete and thorough  Environmental  Impact  Statement may be

prepared.



     The biological processes to be  evaluated include the semi-aerobic process

(SA),  conventional  activated  sludge  process (AS),  and trickling filters

followed by activated  sludge  (TF/AS).  Data provided by the city's consultant

was evaluated against Ten State Standards, USEPA Design Criteria Documents,

and established literature values for critical design conditions for each of

the selected processes.  The  process evaluation includes a process descrip-

tion,  a review of the  technical criteria from  each of the process trains

including reliability, flexibility,  performance,  expandability and turndown,

and environmental impacts.  Capital  costs are also evaluated based on the

selection of system sizing and components necessary to meet Ohio EPA effluent

discharge standards.   The final  section of the report deals with conclusions

and recommendations based on  data available to date.
                                    C-3

-------
                          1.  TERMS AND DEFINITIONS


     An evaluation of municipal biological treatment processes requires a
fundamental knowledge of terminology used by design engineers.  The following

key words used in this briefing paper are defined to assist the reader in

understanding the key issues raised during process evaluation.


Semi-Aerobic  The semi-aerobic process is a modified activated sludge system
which contains an initial anaerobic/anoxic conditioning stage consisting of a
mixture of aerobic return activated sludge and raw primary effluent followed
by aerobic treatment.  This process uses an anaerobic selector zone to control
bulking sludge.

Anaerobic - A biological treatment process that occurs in the absence of
oxygen.  This process contains bacteria that can survive only in the absence
of any dissolved oxygen.  These bacteria are known as obligate anaerobes.  The
anaerobic section of the semi-aerobic process is critical in providing a
selector mechanism against those bacteria which cause bulking in a municipal
waste treatment plant.

Anoxic - A condition of low dissolved oxygen (less than 0.3 mg/1) or a
condition in which the only source of oxygen is mineral bound oxygen such as
nitrates.  Anoxic denitrification is a process by which nitric oxygen is
converted biologically into nitrogen gas in the absence of dissolved oxygen.
In the semi-aerobic process, the anoxic zone may change from anaerobic to
anoxic depending on the level and concentration of nitrates in the wastewater.

Biological Phosphorus Removal - (Also called Bio-P Removal) A process by which
phosphorus associated with biological cells, is precipitated from the
wastewater and contributes to the sludge of a biological treatment system.
The semi-aerobic process results in biological phosphorus removal.  The city's
consultant estimates that excess phosphorus removal results in approximately
4.5 milligrams additional sludge per milligram of phosphorus removed from the
mixed liquor suspended solids.  The mechanism which triggers removal is not
well understood; however, in plants where a phosphorus effluent limitation is
in effect, biological phosphorus removal is an additional benefit. Where
biological phosphorus removal cannot be triggered, physical-chemical
phosphorus removal must be employed.  In all cases, the removal of phosphorus
from the wastewater increases the sludge yield from the biological treatment
train.

Bulking Sludge/Rising Sludge - A bulking sludge is one which shows poor
settleability as measured by the sludge volume index (SVI Test).  The cause of
a bulking sludge is generally filamentous algae or bacteria.  The microbe
responsible for bulking at the Southerly wastewater treatment plant has been
identified as the cyanobacterium Schizothrax calcicola (Phorraidium).  Because
                                    C-4

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
of its poor settling characteristics a bulking sludge will cause BOD and  total
suspended solids violations due to the loss of particulates over the weirs of
the secondary clarifier.  High SVI numbers are indicative of a bulking  sludge.

A rising sludge is one in which the sludge blanket of the secondary clarifiers
floats to the surface, once again causing TSS and BOD violations.  Rising
sludges are frequently caused by biological activity in the clarifier
resulting in the release of micro gas bubbles which attach to the sludge
particles.  One of the most frequent causes of a rising sludge is denitrifica-
tion in the secondary clarifiers.  The denitrification process releases
nitrogen gas and carbon dioxide which causes the sludge to float.  No degree
of increased clarifier sizing or decreasing the clarifier surface overflow
rate will compensate for a rising sludge.  The cause of the denitrification in
the secondary clarifiers must be eliminated for the wastewater treatment  plant
to meet standards.

Carbonaceous BOD Removal - This is the biological conversion of carbonaceous
organic matter in wastewater to cell tissue and various gases and by-products.
In the conversion it is assumed that nitrogen present in the various compounds
is converted to ammonia.  High carbonaceous BOD values will result in effluent
violations.

Denitrification - The biological process by which nitrate is converted  into
nitrogen and other gaseous end products.  When denitrification occurs in  the
secondary clarifiers the result is a rising sludge and effluent violations.

F/M Ratio - The food to mass ratio.  This is a ratio of food substrate  (BOD)
to biological mass (MLSS) which is used as a control parameter for determining
the organic loading rate to a biological treatment system.  A high F/M  ratio
means that oxygen uptake rates will be high, biological metabolic rates will
be high, and in the absence of excess oxygen, obligate aerobic bacteria will
be removed.  A low F/M ratio generally results in high dissolved oxygen
concentrations and may result in the selection of bulking bacteria in a
municipal wastewater treatment system.  In the semi-aerobic process high  F/M
ratios are intentionally maintained in the first bay of the aeration tank in
order to maintain anaerobic or anoxic conditions necessary to select against
bulking bacteria.

Mixed Liquor Suspended Solids - (MLSS) The mixed liquor suspended solids  or
mixed liquor volatile suspended solids are a measure of the amount of biomass
present in the aeration system.  For most conventional activated sludge
systems, this concentration is approximately 1,200 to 3,000 milligrams per
liter (mg/1).

Nitrification - The two-stage biological process by which ammonia or total
kjeldahl (TKN) nitrogen is first converted to nitrite then to nitrate.
Nitrification is the necessary first step in the nitrification/denitrification
cycle.  The goal is to convert ammonia into nitrates and ultimately into
gaseous end products.
                                    05

-------
Over-pumping - This is the process by which the sludge inventory in the
secondary clarifiers is held to a rainimim in order to place the bulk of the
biomass back in the aeration system.  Over-pumping of the clarifier sludge is
necessary when there is the potential for denitrification to occur in the
secondary clarifiers or where mixed liquor suspended solids in the aeration
basins must be held at a high concentration.  A well-designed clarifier will
permit over-pumping on a routine basis by eliminating rat-holing, the
phenomenon by which water channels through the sludge blanket leaving behind
the solids.  Channeling is minimized by providing slow agitation and hydraulic
scouring devices in the sludge pumping system.  These devices are used in
circular clarifiers.  Rectangular clarifiers generally use a chain and flight
mechanism which drags the sludge down to a sludge sump located at the
discharge end of the rectangular clarifier.  The chain and flight sludge
mechanism is generally inefficient where over-pumping is required.

Surface Overflow Rate - (SOR) The surface overflow rate is one of the critical
design parameters for sizing a clarifier.  The dimensions for the surface
overflow rate parameter are gallons per day square foot (gpd/d. ft^) of
clarifier surface.  High surface overflow rates generally result in loss of
solids from the secondary clarifiers.  Ten States Standards cites a surface
overflow rate of 1,200 gallons per day per square foot of clarifier surface
area as a good design maximum for conventional activated sludge processes.
However, due to the fact that sludges produced from nitrification processes
are generally poor settling,  Ten States recommends a surface overflow rate of
800 gallons per day per square foot of surface area for nitrifying sludges.
For this briefing paper the general range of 700 to 1000 gallons per day per
square foot was selected as a conservative design criteria.

Sludge Volume Index - (SVI) The sludge volume index is expressed as the volume
in mils per gram of waste activated sludge after the mixed liquor has been
allowed to settle for 30 minutes under quiescent conditions.  A low SVI is
indicative of a well-flocculated, poor-settling sludge.  A high SVI is
indicative of a bulking, dispersed poor-settling sludge.  Sludges with SVIs in
the range of 50 to 100 exibit excellent settling characteristics.  Sludges
with SVIs in the range of 100 to 150 are generally transitional sludges with
fair to good settling characteristics.  Sludges with SVIs in the range of  150
to 200 are characterized as bulking sludges as indicated by the poor settling
characteristics in the secondary clarifier and poor dewatering characteristics
in the sludge handling process.
                                     C-6

-------
I
I
I
I
I
I
I
I
i
i
i
i
i
i
i
i
i
i
i
                        2.  BRIEFING PAPER ASSUMPTIONS

     The analysis contained in this briefing paper is based on  two key
factors.  These factors include wastewater flows and loads and  NPDES permit
limits for the Southerly and Jackson Pike Wastewater Treatment  Plants.
Alternative process trains were conservatively selected to meet applicable
1988 7-day and 30-day discharge limits.

     A separate briefing paper, prepared for the EIS, documents the
development of wastewater flows and loads.  Table 2-1 presents  the EIS  flows
and loads.

     The average design flow for Jackson Pike will be held at 70 MGD, and the
peak design flow will held at 100 MGD.  This results in an additional 18 MGD
at average flow and 32 MGD at peak flow being diverted to Southerly.  Section
3.1.2.3 discusses the reasons for limiting the flows at Jackson Pike.   Table
2-2 presents the actual flows and loadings which would be processed by  each
plant.  These flows and loadings are used to determine facility sizes in
Section 3.1.2.

     Tables 2-3 and 2-4 provide a summary of the permit limitations for the
Jackson Pike and Southerly WWTPs.  These were taken from Ohio EPA Permit No.
4PFOOOOO*GD (Jackson Pike) and 4PF00001*HD (Southerly).  As noted on the
attached tables, the effluent characteristics are segregated by time of year
as well as by 30-day and 7-day limits.  In addition to the concentration
limits, a mass loading limit based on an effluent loading of 60 MGD from
Jackson Pike and 120 MGD from Southerly are included.  Table 2-5 is an
estimate of the one-plant permit limitations.  These are based  on the
assumption that the water quality impacts to the Scioto River will be the
limiting factor in the event the Southerly wastewater treatment plant is
expanded.  Therefore, the concentration limitations were derived from those
assigned to the Southerly treatment plant.  Mass loading limits were derived
by adding the flows of the Jackson Pike and Southerly plant together and
converting them to a mass basis.
                                    C-7

-------
     TABLE 2-1.  2008 PROJECTED FLOWS LOADS
Tributary to Jackson Pike

        BOD (Ib/day)                    141,600
        TSS (Ib/day)                    161,600
        TKN (Ib/day)                     18,532
        Average Flow (MGD)                   88
        Peak Flow (MGD)                     132
Tributary to Southerly

        BOD (Ib/day)                    126,600
        TSS (Ib/day)                    121,300
        TKN (Ib/day)                     16,570
        Average Flow (MGD)                   66
        Peak Flow (MGD)                      99
Total From Planning Area

        BOD (Ib/day)                    268,200
        TSS (Ib/day)                    282,900
        TKN (Ib/day)                     35,102
        Average Flow  (MGD)                   154
        Peak Flow  (MGD)                      231
                      C-8

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
      TABLE 2-2.  ACTUAL FLOWS AND LOADS TO BE TREATED AT EACH FACILITY
Jackson Pike
     •  Flow (MGD)
     •  CBOD5 (Ib/day)
     •  TSS (Ib/day)
     •  TKN (Ib/day)


Southerly
     •  Flow (MGD)
     •  CBOD5 (Ib/day)
     •  TSS (Ib/day)
     •  TKN (Ib/day)
                                          Average
     70
112,600
128,500
 14,740
     84
155,600
154,400
 20,360
                         Peak
    100
107,300
122,400
 14,040
    131
160,900
160,500
 21,060
                                    C-9

-------
                TABLE 2-3.  PERMIT LIMITATIONS - JACKSON PIKE
Effluent Characteristics

     C-BOD5
          (June-Oct)
          (Nov-Apr)
          (May)

     Suspended Solids
          (June-Oct)
          (Nov-Apr)
          (May)

     Ammonia (N)
          (June-Oct)
          (Nov-Apr)
          (May)
 Concentration
     (mg/1)
 30-day  7-day
 8.0
20.0
13.0
16.0
30.0
26.0
 1.0
 5.0
 2.5
12.0
30.0
19.5
24.0
45.0
39.0
 1.5
 7.5
 3.75
               Mass Loading
                  (Ibs/d)
              30-day    7-ds
 3,995    5,993
 9,988   14,980
 6,492    9,737
 7,990   11,986
14,980   22,471
12,984   19,474
   499
 2,497
 1,247
  748
3,744
1,872
* Mass limits based on 60 MGD effluent loading
                                     C-10

-------
1
1
1


1




1



I

1
1
1
1
1
1
1
1
1
1


TABLE 2-4. PERMIT LIMITATIONS -
Concentration
(mg/1)
Effluent Characteristics 30-day 7-day
C-BOD5
(June-Oct) 8.0 12.0
(Nov-Apr) 25.0 40.0
(May) 13.0 19.5
Suspended Solids
(June-Oct) 16.0 24.0
(Nov-Apr) 30.0 45.0
(May) 26.0 39.0
Ammonia (N)
(June-Oct) 1.0 1.5
(Nov-Apr) 5.0 7.5
(May) 2.0 3.0

* Mass limits based on 120 MGD effluent loading








C-ll



SOUTHERLY
Mass Loading*
(Ibs/d)
30-day 7-day
7,990 11,985
24,968 39,950
12,984 19,474

15,979 23,969
29,962 44,942
25,967 38,951

999 1,498
4,994 7,491
1,998 2,996













-------
            TABLE 2-5.  ESTIMATED ONE PLANT PERMIT LIMITATIONS


                                Concentration           Mass Loading"
                                    (mg/1)                 (Ibs/d)
Effluent Characteristics        30-day  7-day          30-day    7-day

     C-BOD5
          (June-Oct)             8.0    12.0           12,010   18,014
          (Nov-Apr)             25.0    40.0           37,530   60,048
          (May)                 13.0    19.5           19,516   29,273

     Suspended Solids
          (June-Oct)            16.0    24.0           24,019   36,028
          (Nov-Apr)             30.0    45.0           45,036   67,554
          (May)                 26.0    39.0           39,031   58,547

     Ammonia (N)
          (June-Oct)             1.0     1.5           1,501    2,252
          (Nov-Apr)              5.0     7.5           7,506   11,259
          (May)                  2.0     3.0           3,002    4,507


* Mass limits based on 180 MGD effluent loading

1.   No one-plant permit presently exists.  Mass loadings were derived  from
     180 MGD flow and Southerly concentration limits.
                                    C-12

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     For the purposes of this briefing paper, it will be assumed that a

treatment train would be deficient if it would be unable to meet either the

30-day or 7-day concentration limit or the 30-day or 7-day mass loading limit.

It is understood that mass limits can be modified if the new loading does not

negatively impact receiving water quality.  Temperature considerations as they

impact such variables as nitrification rates were evaluated based on the most

stringent occurrence of those temperatures.  For example, nitrification was

evaluated utilizing a sewage temperature of 12°C to meet a May ammonia limit.
                                    C-13

-------
                            3.  PROCESS EVALUATION
     This section describes and evaluates the biological process alternatives
proposed in the facility plan.  The alternatives are evaluated based on
technical criteria, environmental criteria, and system  costs.

3.1  PROCESS DESCRIPTION
     The semi-aerobic (SA) and the trickling  filter/activated  sludge (TF/AS)
biological processes were evaluated for these alternatives:

     •  Southerly Two-Plant
     •  Jackson Pike Two-Plant
     •  Southerly One-Plant

     The semi-aerobic process is a modified form of the activated sludge
process.  The process consists of a non-aerated reaction zone ahead an aerated
activated sludge zone.  The non-aerated zone may be anoxic (nitrates are
present), anaerobic (no  nitrates  or oxygen present),  or a combination  of  both.
The purpose of the anaerobic zone is to function as a selector mechanism
providing an environment which discourages proliferation of filamentous
organisms and thereby controls bulking sludge.  The anaerobic zone may change
to anoxic depending on the level and concentration of nitrates in the
wastewater.  Denitrification  occurs in the anoxic  zone.  Denitrification  is a
process by which nitrates are converted into  nitrogen  gas.

     The only physical differences between the semi-aerobic process and the
conventional activated sludge process  is  the addition of an internal mixed
liquor  recycle loop and two  baffles in the  first bay  of the  aeration  tanks.
The internal recycle  loop is used to bring nitrates back to the anoxic zone
and thus cause denitrification to take place.  The baffles are incorporated
into the design to prevent backmixing from the aerated  zone to the anaerobic
zone.
                                    C-14

-------
 I
 I
 I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
     In reviewing full-scale operational data from the Southerly plant as well
as an evaluation of nitrification rates at both Southerly and Jackson Pike, it
is evident that the semi-aerobic process proposed by the city is in effect
similar to the conventional activated sludge  process  with the exception of the
internal mixed liquor recycle loop and the addition of two baffles in the
first bay of the aeration tanks.   Given these exceptions,  a conventional
activated sludge system can be operated as a semi-aerobic process simply by
reducing the amount of aeration provided in the first bay of the system.  If
one takes Chis reasoning one step further and adds an internal recycle pumping
system (estimated cost $10,000 per aeration tank),  the  result is a semi-
aerobic process minus two  23x15 foot concrete baffles.   For this reason, it
was assumed that the semi-aerobic process and the activated sludge process
were in effect identical and would be  evaluated on that basis.

     The trickling filter/activated  sludge process is comprised of roughing
trickling filters followed by aeration tanks.  The trickling filters are
designed to remove 40 percent  of  the  BODr.  They  function  in  the same manner
as the anaerobic/anoxic zone of the semi-aerobic process in that they select
for non-filamentous bacteria.   The aeration tanks  that  follow the filter
remove the remaining BODr  and provide  the  required nitrification.  An internal
recycle loop can be provided back to the trickling filters to initiate
denitrification there.

     Slightly reduced aeration tank capacity and aeration energy is required
since the trickling filter  has  the ability to dampen peak biological loads and
thus minimize the amount of aeration time needed to achieve complete
biological oxidation.

     The following sections present design criteria and  recommended process
sizing for each alternative.
                                    C-15

-------
3.1.1  Design Criteria
     The biological process design criteria are listed in Table 3-1.  These
criteria were derived from pilot data provided by the city's consultant, the
Ten States Standards (Recommended Standards for Sewage Works, 1978 Edition,
Health Education Service Incorporated, Albany, New York 12224), and USEPA
criteria (Innovative and Alternative Technology Manual, EPA-430/9-78-009,
1978).  The range of acceptable operating conditions given in Table 3-1
defines the critical regions for the aeration, trickling filter, and  final
clarification processes.  In the absence of more extensive full-scale piloting
data, it is assumed that violation of these criteria would result in
inadequate treatment of the wastewater received at Jackson Pike or Southerly
which would result in effluent violations.

3.1.1.1  Aeration Basins
     The aeration process listing in Table 3-1 includes evaluation criteria
for the hydraulic retention time in the aeration basin, F/M  ratios in the
first bay as well as the overall F/M ratio of the aeration basin, design mixed
liquor suspended solids concentrations, minimum solids retention times, and a
recommended ratio of oxygen uptake rates to dissolved oxygen (OUR/DO).

     A minimum hydraulic retention time in the aeration basin of 4.5  hours is
limited to the final 7 bays of the plug flow reactor for the semi-aerobic
process.  This datura was taken from the Southerly SBR Nitrification Study
conducted by the city's consultant (January 1987).  The hydraulic residence
time in the Project 20 full-scale semi-aerobic pilot study conducted  at the
Southerly waste treatment plant  typically ranged from 5 to 8 hours.   The use
of a shorter residence time in the aeration basin for the trickling filter
process is based on the fact that a roughing  filter has the  ability to  dampen
or attenuate peak biological loads, thus minimizing the amount  of aeration
time required  to achieve complete biological  oxidation.
                                     C-16

-------
1
1





1

1
•

1





1

1

1

1
^M



1
1
1
1
1
1


TABLE 3-1. BIOLOGICAL PROCESS DESIGN CRITERIA

Process Parameter Range
Aeration Hydraulic Retention Time 4.75-SA, AS
(HRT, hrs) Minimum 3-TF
F/M First Bay 5
Overall 0.13-0.17
MLSS (mg/1)
Southerly 3500
Jackson Pike 2500
Solids Retention Time (days)
Southerly 9.9
Jackson Pike 8.7

OUR/D.O. 250-500

Roughing Trickling Hydraulic Loading Rate 1400-4600
Filter (gpd/ft2)
Organic Loading Rate 100-500
(Ib BOD/d.1000 ft3)
Clarifiers Surface Overflow Rates 400-1000
(gpd/ft2)
Solids Loading Rates 20-50
(Ib/d. ft2)
Southerly SBR Nitrification Study, Orris Albertson, URS Dalton
2"The Control of Bulking Sludges", JWPCF, April 1987




Source
SBR Report1







City of Columbus
Comment Letter

Control of Bulking
Sludge2
USEPA3 >4

USEPA3
USEPA3
USEPA3

, January 1987

^"Innovative and Alternative Technology Manual", EPA 430/9-78-009, 1978
^"Wastewater Treatment Plant Design", WPCF, 1977
^No MLSS are provided for the combined plant option because the
studies have not been run for that wastewater blend.




C-17

nitrification








-------
     The selection of an F/M ratio of 5 in the first bay of the semi-aerobic
system is based on correspondence with Mr. Orris E. Albertson, Process
Consultant to the city's consultant.  Mr. Albertson also stated in an article
published in the April 1987 Journal of the Water Pollution Control Federation
that the maintenance of a high F/M ratio in the initial contact basin of a
semi-aerobic system was required to maintain the anaerobic and anoxic
conditions necessary to select against bulking bacteria.  This high F/M ratio
would be realized in both the semi-aerobic and activated sludge options.  It
is assumed that the trickling filter option would greatly reduce this F/M
ratio due to the attenuating effect the upstream roughing filter would have on
carbonaceous BOD loadings.  An overall aeration basin F/M value of 0.13 to
0.17 would be consistent for a well operated nitrifying activated sludge
system.

     The mixed liquor suspended solids concentrations of 3,500 mg/1 for
the Southerly plant and 2,500 mg/1 for the Jackson Pike plant were
derived from SBR studies conducted by the city's consultant.  It is assumed
that mixed liquor concentrations of the same magnitude would be required for a
conventional activated sludge system.  The primary reason for the higher mixed
liquor suspended solids in the Southerly aeration basin is the low nitrifica-
tion rates observed at that plant.  Increasing the MLSS to 3,500 mg/1 allows
nitrification to proceed with fewer aeration basins than would be required at
2,500 mg/1.  The Jackson Pike WWTP experiences nitrification rates well within
the range of most sewage treatment facilities.

     The cause of lower nitrification rates at the Southerly plant is most
likely due to toxicity of some non-conventional pollutants present in the
Southerly raw wastewater.  Nitrification rates for the Jackson Pike wastewater
treatment system are well within the range of nitrification rates realized in
North American municipal treatment facilities.
                                    C-18

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     The significance of the oxygen uptake rate to dissolved oxygen ratio
(OUR/DO) has been cited by Mr. Orris Albertson as necessary for the control of
bulking sludge organisms in municipal treatment facilities.  In his paper, Mr.
Albertson indicates that "The best control of bulking sludges is provided by
both reactor compartmentalization and by DO control in each of the
compartments.  Often in practice, either will provide the necessary SVI
control; but the maximum control will be available when both a high F/M
gradient is present and DO control as a function of time for each compartment
is provided."  Mr. Albertson further states that "Regardless of whether the
initial contact zone is aerated or unaerated a sufficiently high OUR/DO ratio
will ensure both SVI control and enhance phosphorus removal.  The suggested
minimum OUR/DO ratio is greater than 250 to 1 and preferably as high as 500 to
1."  Under conditions of high F/M ratios, Mr. Albertson contends that the
biological cell will uptake organic material and release soluble phosphates
given that the DO gradiant across the slime layer of the cell is less than
0.5 mg per liter.  Under endogenous conditions, such as occur in the final
zones of the aeration basin, soluble phosphate uptake occurs as well as the
release of endogenous decay products resulting in a well-flocculated mixed
liquor leaving the aeration basin.  It should be noted that these conditions
can be achieved in a conventional activated sludge aeration basin by the use
of compartmentalization and reducing the blower capacity in the initial stages
of the aeration tanks.  This condition is further enhanced by increasing the
mixed liquor suspended solids and providing for an internal mixed liquor
recycle loop.

3.1.1.2  Trickling Filters
     The design criteria for roughing trickling filters which are followed by
activated sludge systems is considerably higher than those for trickling
filters followed by clarification. Hydraulic loading rates ranging from 1,400
to 4,600 gallons per day per square foot of surface area are considered good
design criteria.  Organic loading rates in the range of 100 to 500 pounds of
BOD per day per 1,000 cubic feet volume would also provide adequate capacity
for a roughing trickling filter.  The trickling filters, when operated in this
                                    C-19

-------
condition, act as the initial zone or anaerobic/anoxic zone of the aeration
basin under the semi-aerobic or activated sludge options.  The roughing
trickling filters would reduce the volume of aeration basin required and
effectively assist in control of sludge bulking.

3.1.1.3  Clarifiers
     Given the fact that the three previously selected biological treatment
processes (semi-aerobic, conventional activated sludge, and trickling  filter
followed by activated sludge) all can act as effective selectors against
bulking organisms, it was assumed that SVIs would generally be in the  range of
70 to 150.  Given this SVI range, there are two critical design factors which
must be considered when selecting and sizing final clarifiers.  These  are
surface overflow rates (gallons per day per square foot surface area)  and
solids or floor loading rates (pounds of suspended solids per day per  square
foot).  The city's consultant has selected conservative surface overflow rates
for their final clarifiers.  These are generally in the range of 470 for
average flows and 800 for sustained peak flows.  Mr. Richard Brenner,  USEPA
Cincinnati, indicated that conservative design criteria for average  flow rates
would be  in the range of 500 to 550 with peak sustained surface overflow
loading rates set at 900 to 950.  For the purposes of  this evaluation, a. range
of 400 for average flow and 1,000 for sustained peak flow will be used.

     The  city's consultant selected solids or floor loading rates for  their
clarifiers in the range of 18 to 23 pounds per day per square foot for
average flows and 29 to 36 pounds per day per square foot for peak flows.  A
solids loading criteria of 20 to 50 pounds per day per square foot is  cited in
the USEPA Innovative and Alternative Technology Manual.  Rectangular
clarifiers should generally be sized on the lower end  of this solids loading
rate.  Circular clarifiers with hydraulically assisted sludge removal  devices
can easily accommodate the higher solids loading rates without causing sludge
channeling or solids entrainraent.  However, as pointed out by the city's
consultant, SVIs  are also a limiting factor in determining an acceptable
solids loading rate.  Therefore, the Daigger and Roper Clarification Tank
Design and Operation Diagrams will also be used in this evaluation.
                                     C-20

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
3.1.2  Recommended Sizing


     Based on the previously stated process design criteria and  the  2008

projected flows and loads given in Table 2-1 of this briefing  paper,  two

biological treatment trains (i.e. semi-aerobic and trickling filter/activated

sludge) were evaluated for the following alternatives:



     •  Southerly Two-plant

     •  Jackson Pike Two-plant

     •  Southerly One-plant



     A critical assumption in this evaluation is  that  the projected  flows  and

loads given in Table 2-1 will permit the plants to treat all anticipated dry

weather flows plus some additional inflow and infiltration during wet weather

events to a peak design flow of 100 MGD for Jackson Pike, 131  MGD for

Southerly two-plant, and 231 MGD for a Southerly  one-plant alternative.



     BOD and total suspended solids loadings developed  for this  briefing paper

were similar to those presented in the facility plan.   Total kjeldahl nitrogen

and total phosphorus loadings presented in the facility plan were used  in  this

process evaluation (Table 2.1).  It was further assumed that the increase  in

flow due to the application of a 1.5 peaking factor would have little or no

effect on the mass daily loading of BOD, total suspended solids, or  nitrogen.



     Tables 3-2 through 3-7 document the results  of this briefing paper

analysis relative to the sizing and performance of the  various biological

treatment processes for Southerly and Jackson Pike.  All previously  stated

design and loading criteria were used to derive the data presented in these

tables.
                                     C-21

-------
3.1.2.1  Southerly Two-Plant Semi-Aerobic
     Table 3-2 is a summary of the Southerly two-plant semi-aerobic or
activated sludge options.  The existing six aeration basins in the west train
would be utilized with the addition of an internal recirculation pump and
baffles for the semi-aerobic option.  Four of the existing center train
aeration basins would be used with the addition of two new 26 foot by 900 foot
by 15 foot sidewall depth aeration basins.  Given these conditions, average
and peak aeration times fall well within the design parameters cited in Table
3-1.  In terms of final clarification, the existing clarifiers would be
replaced with six new 190-foot diameter circular clarifiers fitted with
hydraulic sludge removal devices, flocculation chambers, and associated piping
and an internal mixed liquor recycle system.  The addition of these six
190-foot diameter units place the clarifiers well within the critically
designed surface overflow rates of 400 to 800 gallons per day per square foot
of surface area established by the city's consultant.  The solids loading
rates based on a mixed liquor suspended solids of 3500 mg per liter fall well
within process evaluation criteria.  Under peak hydraulic loading conditions
the solids loading rates would exceed design criteria established in the
facility plan.  Given the fact that circular clarifiers will be used in this
application, it is unlikely that a peak loading of 38 pounds per day per
square foot would overload the proposed clarifiers.  The Daigger and Roper
Diagram shows that the clarifiers could operate efficiently up to an SVI of
165 ml/g.

3.1.2.2  Southerly Two-Plant Trickling Filter
     The critical design data for the Southerly two-plant trickling filter
option is presented  in Table 3-3.  Any evaluation of trickling filters  at the
Southerly plant requires an understanding of the existing plant layout  and
related logistical problems.  There is inadequate space between the existing
primary clarifiers and aeration basins to install the proposed 110-foot
diameter trickling filters.  Due to this  limitation  it was assumed  that the
trickling filters would be located  in an  area remote to those processes and
that primary effluent would be pumped to  the trickling  filter and discharged
from the trickling filter to the influent end of the aeration basin by  gravity

                                    C-22

-------
1
1
1


1



1




1

1
^v

1
1


1
1
1
1
1
1
1


TABLE 3-2. SOUTHERLY PROCESS DESCRIPTION - TWO
West Train Center
Flow (Design)
Average (MGD) 42 42
Peak (MGD) 65.5 65
Aeration
Tankage
New — 2@26'x900
Existing 6@26 ' x900 'x!5 ' SWD 4@26'x900
HRT (hrs)
Average 9.00 9
Peak 5.77 5

Clarification
Tankage
New 6(3190' dia.x!5' SWD
Existing
Surface Overflow
Rate (gpd/ft2)
Average 490
Peak 770
Solids Loading
Rate (Ib/d. ft2)
Average 25
Peak 38





C-23



PLANT SEMI-AEROBIC, AS
Train Total

84
.5 131

'xl5'SWD 2
'xlS'SWD 10

.00
.77



6
"














-------
         TABLE 3-3.  SOUTHERLY PROCESS DESCRIPTION - TWO PLANT TF/AS
Flow (Design)
     Average (MGD)
     Peak (MGD)
West Train

    46
    71
Center Train

    38
    60
Total

  84
 131
Trickling Filters

  Filters           2@110 '0x22'ht,
  Hydraulic Loading
  Rate (gpd/ft2)
     Average           2420
     Peak              3740

  Organic Loading Rate
  (Ib.BOD/d. 1000 ft3)
     Average            160
     Peak               160

Aeration
  Tankage
    New                  —
    Existing              5
  HRT (hrs)
    Average
    Peak
  6.85
  4.43
                   2@110'0x22'ht.
                         2000
                         3160
                          130
                          130
  6.63
  4.20
Clarification

  Tankage
     New
     Existing

  Surface Overflow Rate
  (gpd/ft2)
     Average            490
     Peak               770
  Solids Loading Rate
  (Ib/d. ft2)
     Average
     Peak
    25
    38
        6@190'dia.xl5'SWD
                                    C-24

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
conduits.  These logistical problems while not insurmountable were taken into
consideration when evaluating the overall effect of the trickling filter
process as discussed in Section 4 of this briefing paper.

     As indicated in Table 3-3, four 110-foot diameter by 22-foot high high-
rate roughing trickling filters were sized for the Southerly two-plant option.
Two trickling filters would service each of the existing treatment trains.
Hydraulic loading rates of 3,740 and 3,160 gallons per day per square foot of
trickling filter surface area are well within the design criteria limit of
4,600 gallons per day per square foot.  The organic loading rates of 130 and
160 pounds BOD per day per 1,000 cubic feet of trickling filter volume are
well within the 100 to 500 range.  Following the trickling filters, five of
the six existing aeration basins in the west train and four existing basins in
the center train would be used for aeration capacity.  Although the hydraulic
retention times are considerably less than thoses cited for the semi-aerobic
or activated sludge systems, it is considered adequate for aeration following
roughing trickling filters.

     Final clarification consists of six 190-foot diameter clarifiers with
resulting surface overflow rates in the range of 490 gallons per day per
square foot under average conditions and 770 gallons per day per square foot
under peak conditions.  Solids loading rates range from 25 pounds per day per
square foot at average flow to 38 pounds per day per square foot at peak flow.

3.1.2.3  Jackson Pike Two-Plant Semi-Aerobic
     Table 3-4 summarizes the design criteria for the Jackson Pike semi-
aerobic and activated sludge process trains.  The assumption used throughout
this briefing paper is that the Jackson Pike plant is hydraulically limited to
100 MGD.  This assumption is based on information from the city and their
consultant.  The average flow to Jackson Pike was limited to 70 MGD based on
the capacity of the existing aeration tanks.  Average flows in excess of
70 MGD and peak flows in excess of 100 MGD will be diverted to the Southerly
WWTP.
                                    C-25

-------
      TABLE 3-4.  JACKSON PIKE PROCESS DESCRIPTION - SEMI-AEROBIC, AS
Flow (Design)

     Average (MGD)
     Peak (MGD)

Aeration

  Tankage
     New
     Existing

  HRT (hrs)
     Average
     Peak

Clarification
                       A-Train
     42
     60
6(?26lx9001xl5lSWD
      9.00
      6.30
  Tankage
     New                   —
     Existing   8@153'x60'x!2.5'SWD

  Surface Overflow Rate
     (gpd/ft2)
     Average             570
     Peak                820
                          B-Train
     28
     40
                        Total
 70
100
4(326'x900'xl5'SWD
      9.00
      6.30
 10
  Solids Loading Rate
      (Ibs/d.  ft*)
     Average
     Peak
     20
     29
                    2@153'x60lxl2.5'SWD
                    4
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     The semi-aerobic system would utilize 6 existing aeration basins in the A
train.  In the B train the 4 existing aeration basins would be utilized.  A
flow split of 60 percent to the A Train and 40 percent to the B Train would be
employed.  Under both conditions, hydraulic retention times are well within
the limits established in the evaluation criteria.  These criteria were
established during the SBR and piloting studies utilizing Jackson Pike primary
effluent.  As previously stated, the nitrification rates in the Jackson Pike
studies have been approximately 300 percent higher than those reported for
Southerly.  Given these conditions, the hydraulic retention time cited in
Table 3-4 is considered adequate when operating at a MLSS of 2500 mg/1.

     In evaluating final clarification for Jackson Pike, the selected option
includes rehabilitating the existing 12 clarifier units and adding 2 new
rectangular clarifiers (153-foot by 60-foot by 12.5-foot sidewall depth) to
the B-train.  The addition of 2 new rectangular clarifiers would provide
Jackson Pike with a combined surface overflow area of 128,000 square feet.
The facility plan recommended demolishing the existing clarifiers and
installing four new 200-foot diameter circular clarifiers.  This would provide
the facility with 126,000 square feet of final clarifier surface area.
Surface overflow rates and solids loading rates would be essentially identical
for the rectangular clarifiers versus the new circular clarifiers.  A
discussion of final clarifier utilization for both Southerly and Jackson Pike
is presented in Section 4.2 of this briefing paper.

3.1.2.4  Jackson Pike Two-Plant Trickling Filter/Activated Sludge
     The trickling filter/activated sludge option design criteria for the
Jackson Pike WWTP is summarized in Table 3-5.  The design criteria for
aeration and final clarification are essentially the same as those described
under the Jackson Pike serai-aerobic and activated sludge options.  Two new
110-foot diameter by 22-foot high and two new 90-foot diameter by 22-foot high
trickling filters would be added to the process treatment trains.  Critical
design conditions in terms of hydraulic loading and organic loading are well
within criteria cited in Table 3-1.  One limitation which impacts the
selection of trickling filters for Jackson Pike is space. While the evaluation

                                    027

-------
             TABLE 3-5.  JACKSON PIKE PROCESS DESCRIPTION - TF/AS
Flow (Design)

     Average (MGD)
     Peak (MGD)

Trickling Filters
                        A-Train
42
60
  Filters            2@110'0x22'ht.
  Hydraulic Loading
  Rate (gpd/ft2)
     Average           2210
     Peak              3160

  Organic Loading Rate
  (Ib. BOD/d.1000 ft3)
     Average            120
     Peak               120

Aeration

  Tankage
     New
     Existing             6

  HRT (hrs)
     Average              9.00
     Peak                 6.30

Clarification
                      B-Train
28
40
                  2(390'0x22'ht.
                      2200
                      3140
                       120
                       120
                         9.00
                         6.30
                   Total
 70
100
                                             10
  Tankage
     New
     Existing    8@153'x60'xl2.5'SWD

  Surface Overflow Rate
     (gpd/ft2)
     Average             570
     Peak                820
   Solids Loading Rate
      (Ib/d.  ft2)
      Average
      Peak
 20
 29
              2@153'x60'xl2.5'SWD
              4@153'x60'xl2.5'SWD
                        510
                        730
 18
 26
                      2
                     12
                                    C-28

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
of critical design criteria indicate that four units would be adequate,
limited available area would make siting difficult.

3.1.2.5  Southerly One-Plant Semi-Aerobic
     The critical design criteria for a one-plant semi-aerobic option are
presented in Table 3-6.  Under the one-plant option, the existing center and
west train would handle an average flow of 88 MGD with a peak flow of 132 MGD.
With the increased hydraulic loading, it will be necessary to construct a new
east train capable of handling an average of 66  MGD with peak sustained loads
of 99 MGD.  This would include use of ten existing aeration basins at the
center and west trains with the construction of two new 26-foot by 900-foot by
15-foot sidewall depth basins on the center train.  The new east train would
contain nine 26-foot by 900-foot by 15-foot sidewall depth aeration basins.

     Final clarification would include six new 200-foot diameter circular
clarifiers for the center and west train and four new 205-foot diameter
circular clarifiers for the east train for a combined facility clarifier
surface area of 320,360 square feet.  Given the amount of clarifier capacity,
both the surface overflow rates and solids loading rates are within design
criteria.

3.1.2.6  Southerly One-Plant Trickling Filter/Activated Sludge
     The trickling filter/activated sludge option for a Southerly one-plant
operation is presented in Table 3-7.  The  trickling filters would consist of
four 115-foot diameter units for the center and west trains and two 115-foot
diameter units for the east train.  Under sustained peak hydraulic loading,
the hydraulic loading criteria is within the 4,600 gallons per day per square
foot of surface area for all trains.  Organic loadings are within the range of
100 to 500 pounds of BOD per day per thousand cubic feet of reactor volume.
All other aeration and clarification criteria fall within the critical design
criteria for the one-plant option.
                                    C-29

-------
 TABLE 3-6.  SOUTHERLY PROCESS DESCRIPTION - ONE-PLANT SEMI-AEROBIC AND AS
                  Center and West Train    East Train (New)
Flow (Design)

     Average(MGD)
     Peak(MGD)

Aeration

  Tankage
     New
     Existing

  HRT (hr)
     Average
     Peak

Clarification
          88
         132
           66
           99
  2@26'x900'xl5'SWD
 10(326' x900'xl5'SWD
           8.59
           5.73
  Tankage
     New
     Existing
6(3200'0x15'SWD
  Surface Overflow Rate
  (gpd/ft2)
     Average             470
     Peak                700

  Solids Loading Rate
  (pounds/d.ft2)
     Average              23
     Peak                 35
   9@26'x900'xl5'SWD
            8.58
            5.72
4@205'0xl5'SWD
                               500
                               750
                                25
                                37
                                                   Total
154
231
 11
 10
 10
                                    C-30

-------
1
1
1


1



1

1

1

1
Ml

1

1


1

1

1




1
1


TABLE 3-7. SOUTHERLY PROCESS DESCRIPTION - ONE-PLANT TF/AS
West Train Center Train East Train
Flow (Design)
Average (MGD) 50 42 62
Peak (MO)) 75 63 93
Trickling Filter
Filters (New) 2@1 15 ' 0x22 ' ht . 2@115'0x22'ht. 2@115'0x22'ht.
Hydraulic Loading
Rate (gpd/ft2)
Average 2410 2020 2990
Peak 3610 3030 4480
Organic
(Ib.BOD/d. 1000ft3)
Average 150 120 180
Peak 150 120 180
Aeration
Tankage
New 6@900'x26'xl5'SWD
Existing 5@900'x26'xl5'SWD 4@900 ' x26 ' x!5 ' SWD
HRT (hrs)
Average 6.29 6.00 6.10
Peak 4.19 4.00 4.06
West and Center Train East Train
Clarification
Tankage
New 6(3200' 0x15 'SWD 4@200 ' 0x15 ' SWD
Existing
Surface Overflow Rate
(gpd/ft2)
Average 490 500
Peak 730 750
Solids Loading Rate
(pounds/d.ft2)
Average 24 25
Peak 36 37
C-31




Total

154
231

6








6
9





10











-------
     Final clarification would include six new 200-foot diameter clarifiers
for the west and center trains and four new 200-foot diameter units for the
east train.
                                                                                        I

                                                                                        I

                                                                                        I

                                                                                        I
3.2  TECHNICAL EVALUATION
     The previously described treatment options were evaluated in terms of              I
their reliability and flexibility.  Reliability is measured in terms of
potential loss of treatment system components as well as the impact of                  I
toxicity on the biological treatment process.  System flexibility is discussed
in terras of response to mass loadings as well as upsets within the system.              •

3.2.1  Reliability                                                                      —
     A summary of system reliability for the biological process options is              ™
presented in Table 3-8.  The semi-aerobic and activated sludge processes are            ^
evaluated with respect to aeration basin hydraulic retention time.  Trickling           |
filters are evaluated with respect to organic and hydraulic loading rates.
Final clarification is evaluated with respect to surface overflow rates and
solids loading criteria.

     The analysis of system reliability considered that one of the system
components was out of service.  The components remaining in-service would be
required to process the influent  flow.  Table 3-8 presents the impact on the
process design criteria of processing average and peak  flow through the system
with one unit out of service.  Under conditions where system components were
separated into two parallel treatment trains, or in  the case of Southerly one-
plant where there are three parallel treatment trains,  the worst case scenario
was represented by the loss of one essential component  in each of the parallel
trains.  The reliability evaluation may also be interpreted as a surge  in
hydraulic or mass loadings, where all units are operative, due to brief
intervals of raw wastewater flows or loads above those  projected in Table 2-1.

     The system reliability data  for the Jackson Pike semi-aerobic and
trickling filter/activated sludge alternatives is contained in the first  two
columns of Table 3-8.  For the aeration basin capacity, it was assumed  that

                                    C-32

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                       TABLE 3-8.  SYSTEM RELIABILITY


PARAMETER
Aeration
Hydraulic Retention
Time (hrs)
Average
Peak
Trickling Filter
Hydraulic Loading
Rate (gpd/ft2)
Average
Peak
Organic Loading
Rate (Ib BOD/d/1000ft3)
Average
Peak
Clarification
Surface Overflow
Rate (gpd/ft2)
Average
Peak
Solids Loading
Rate (lb/d/ft2)
Average
Peak
Jackson Pike
(2-Plant)
SA TF/AS



6.75 6.75
4.75 4.75



3150
4500

180
180



610 610
930 930


22 22
33 33
Southerly
(2-Plant)
SA TF/AS



7.50 5.00
4.80 3.15



2950
4590

190
190



590 590
920 920


29 29
46 46
Southerly
(1-Plant)
SA TF/AS



7.15 4.50
4.76 3.00



2970
4450

235
235



670 660
1000 990


33 33
50 49
NOTE:  This table assumes one component (aeration basin, trickling filter,
       or clarifier) is removed from service for repair or maintenance.  The
       resulting impact on the process design criteria is identified.
                                    C-33

-------
                                                                                        I
                                                                                        I
one of the four aeration basins in the B train was removed from service for an          •
extended period of time leaving three functional basins in the B train; and             |
one of the six basins in the A train was removed leaving five basins available
in the A train.  The efficiency of either train should not be affected since            I
the hydraulic retention times are still within design criteria listed in Table
                                                                                        I
     The impact of removing one of the 110-foot diameter trickling filters from
service would cause the hydraulic loading rate to increase to 4,500 gallons
                                                                                        I
per day per square foot of surface area under peak flow conditions.  This does          —
not exceed the design criteria of 1,400 to 4,600.  The organic loading rate             •
would increase to 180 pounds of BOD per day per 1000 cubic feet of filter.
This is still within the design range of 100 to 500.                                    I

     The removal of one of the clarifiers in each train would not have a                •
significant impact on the solids overflow or solids loading design criteria.            ™
                                     C-34
                                                                                        I
     Columns three and four of Table 3-8 show the impact of a loss of process
components at Southerly under a two-plant option.  Removal of one of the  six
aeration basins in either the west train or the center train would not cause            •
the minimum hydraulic retention time to be violated.  Removal of one of the
four trickling filters would not violate either the established maximum                 B
hydraulic or organic loading rates.

     Removal of one of the six circular clarifiers would not result  in a                •
violation of the established design criteria.  The solids loading rates would
be high (46 Ib/d/ft^); however, this should not be a problem for circular              I
clarifiers if the SVI is not excessively high.

     Columns 5 and 6 in Table 3-8 presents the system reliability evaluation
for the Southerly one-plant option.  One of the aeration basins in each of  the
three  trains would be removed from service due to maintenance or mechanical
failure.  For the semi-aerobic and TF/AS options, this would result  in                  _
hydraulic retention times in the aeration basins within specified design                |p
                                                                                        I
                                                                                        I
                                                                                        I

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
criteria.  Removal of one of the six trickling filters would not result in
violations of the established maximum hydraulic or organic loading rates.

     In terras of clarifier capacity it was assumed that one of the circular
clarifiers would be removed from the west and center section and one from the
east section.  Under these conditions, the surface overflow rate as well as
the solids loading rate under peak hydraulic loadings would approach the
critical limits of the design criteria; however, they would not violate them.
Once again, this should not be a problem for circular clarifiers.

     In summary, all of the components under each alternative would be capable
of operating within the specified design criteria in the event that a unit was
removed from operation.

     The second measure of system reliability is its ability to respond to
system upsets or toxicity problems.  The semi-aerobic process provides
excellent capabilities to adjust to high ammonia loadings.  Ammonia
concentrations will be monitored in the number 6 bay in each of the aeration
basins.  Once ammonia concentrations above 2 mg/1 are found, the aeration in
Bay 2 will be activated as well as a general D.O. increase which will
enhance the nitrification rate.  If this is not adequate to reduce NH^N to
1.0 mg/1, then the internal recycle pump will be shut down to increase the
real detention time.  The internal recycle pump reduces nitrification capacity
due to the volume used for denitrification.

     The roughing trickling filter acts as an anaerobic/anoxic aeration bay in
the semi-aerobic process.  The filter reduces BOD loadings to the aeration
basins and effectively aids in the control of sludge bulking.  Effluent
recycling from the aeration basins back to the trickling filters acts in much
the same way as the internal recycle of the semi-aerobic process.  Aeration
basin effluent recycling would also cause denitrification to occur within the
trickling filters.  Denitrification is vital during the summer months to
prevent a rising sludge in the final clarifiers.  One significant limitation
of the trickling filter in cold climates is the tendency to ice.  Under these
                                    C-35

-------
conditions, loss of bio-mass as well as reduced process efficiency reduce the
effectiveness of the trickling filters.

     Toxic effects will have a similar impact on the semi-aerobic and
trickling filter processes due to the common bacterial organisms used in
nitrification and denitrification.  One source of toxicity to a nitrification
system can be slug loads of ammonia or TKN.  Table 3-9 is a summary of Project
20 operating data for the month of February 1987.  Project 20 is not being run
in a manner exactly similar to that proposed for the semi-aerobic process.
Nevertheless, its performance is indicative of the ability of the semi-aerobic
process to nitrify under winter conditions.  The data in Table 3-9 includes
primary effluent ammonia concentrations for the west and center treatment
trains, final effluent ammonia concentrations for aeration basins 1 and 2
which represent the semi-aerobic process, and ammonia concentrations from the
remaining aeration basins which were operated in a conventional activated
sludge mode with reduced aeration in the initial bays of each basin.

     The data indicate that both the semi-aerobic and activated sludge process
can meet 7-day and 30-day ammonia limits under cold weather conditions.
However, it should be noted that periodic peak loadings of ammonia such as
occurred on February 4, February 22, and February 24, resulted in bleedthrough
of high ammonia concentrations to the  final effluent.  Generally, it appears
that ammonia concentrations in the primary effluent in excess of 25 mg/1 would
result in violations of the 7-day and  30-day permit if they were sustained.
It is apparent from the Project 20 data that slug loads of ammonia will cause
effluent violations for both  the semi-aerobic and activated sludge processes.
The trickling filter/activated sludge  option would respond in a similar
fashion allowing bleedthrough of high  influent ammonia loads.  The source of
the high nitrogen load is most likely  one or more industries within the
service area.  As a result  it is recommended that the sources of the high
ammonia loads be identified and be limited in the amount of TKN they are
allowed to discharge.  Without such control, it would be impossible to
consistently meet the  1988  effluent limits for ammonia.
                                    C-36

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
               TABLE 3-9.  AMMONIA BREAK-THROUGH -  SOUTHERLY
                            (Ammonia as N, mg/1)
     Date
     February

         1
         2
         3
         4
         5
         8
         9
        10
        11
        12
        15
        16
        17
        19
        22
        23
        24
        25
Primary Effluent
 West

  17
  17
  18
  24
  20
  17
  18
  15
  15

  17
  16
  21
  20
  30
  19
  24
  15
Center

  15
  14
  16
  24
  19
  15
  17
  15
  14
  14
  21
  20
  30
  18
  25
  15
SOURCE:  Contract 20 Operational Data
                  Final Effluent

1 & 2
(Semi-Aerobic)


















30-day
7-day
1.1
0.7
0.9
5.3
1.3
0.1
0.1
0.1
0.2
0.1
0.1
0.6
1.2
4.6
7.3
8.7
1.8
0.4
1.9
3.5
all others
(AS)
2.5
2.5
3.7
7.4
3.4
3.1
2.8
1.4
0.7
0.1
—
1.6
1.5
5.6
8.7
10.5
7.1
2.6
3.6
5.4
                                    C-37

-------
     The data in Table 3-10 summarizes reported pollutant concentrations in
the Jackson Pike and Southerly influent and presents inhibition levels of
these pollutants for various biological processes.  Influent concentrations of
copper and zinc at the Columbus plants may be found at levels which can
inhibit the nitrification process.  Copper and zinc could act as inhibitory
pollutants if the influent concentrations shown in Table 3-10 are carried
through the primary effluent and enter the biological treatment process.  The
city of Columbus must consider controlling the level of inhibitory industrial
pollutants to prevent system upsets.  An aggresive and well-monitored
industrial pretreatment program would be necessary to ensure the nitrification
process is protected from inhibitory and/or toxic effects of industrial
discharges.

3.2.2  Flexibility
     System flexibility is defined as the ability of the system to expand or to
turn-down (respond to reduced flows or loads) its biological processes.  It
will be necessary for the city of Columbus to control slug loads of ammonia
and TKN no matter which biological option or treatment plant option is
selected.  Impacts can also be manifested in terms of loss of load.  At the
present time, it is estimated that 35 to 45 percent of the BOD loading to
the Southerly plant originates with the Anheuser-Busch Brewery.  The impacts
of losing this BOD loading are most directly felt in the first bay of the
semi-aerobic system.  Mr. Albertson has indicated that in order to control
bulking, an OUR/DO ratio of at least 250-1 must be maintained.  Under current
design conditions, the OUR/DO ratio is approximately 500-1.  Given the loss of
all brewery waste for a sustained period, it can be assumed  that a critical
OUR/DO ratio can be maintained.   If the brewery wastes are the primary source
of the historical bulking problems at Southerly,  the plant could operate in a
semi-aerobic or conventional activated sludge mode with little or no problems.

     The second advantage of the  semi-aerobic process in terms of responding
to periodic upsets is what Mr. Albertson has described as sludge memory.  Most
activated sludge systems which have biological phosphorus removal capabilities
are able to respond  in a linear fashion to organic  loading upsets based on

                                    C-38

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
"8
   w
  r~*


  1?
I
       &
          CM
&
rJ *
                   s&
                     i
              ft
                                             M ED
                                      88   °
                                       *•  •* o o
                                      o o  ~ •
                                      i—i o i—i in
                                      M 3 H CQ
                                             o

                                             CM
                                      I—

                                      CM
                           ) ^
              o   r** o -ct o ui
              in   r-i in -. VJ  _ _
         CO *J AJ ^^ -i-l
         aa^^^s
         O r— in vo Is* oo
         o \o oo oo oo i—i
       1
       6
       O
                                        C-39
                                             O
                                              •
                                             o

                                             o
                                              •

                                             S
                                             CM

                                             O
                                                            O  -   Q O
                                                            -^ p^    O  **
                                                            CM CM    r- CM
                                                                   mm   o
                                                                    II    in

                                                                   uS O   Q
                                                                   CM in   oo
                                                       M 3 H M
                                                            0)
                                                           •o
                                                                          H M
                                                                          m  -
                                                                          CM" CM
                                                                   800  ~i
                                                                    ~ •> m i
                                                                CM rH CM CM I
                                                                   CM O
                                                  S
                                                                   O
                                                                   m
                                                           CM
                                                           CM

                                                                   3^    
-------
                                                                                       I
                                                                                       I
sludge age.  Assume the sludge age is maintained at 9 days for a 2-day period          •
and the primary source of organic loading is removed from the system.  The             |
impact on the effluent would be comparable to the ratio of 2-9 or
approximately 22 percent loss of system efficiency.  Under these conditions,           I
the system would recover rapidly once the source of organic loading is placed
back into the system.  The disadvantage of this type of activated sludge               I
(i.e., one which demonstrates biological phosphorus removal), is that the
sludge yield in terms of pounds of sludge produced per pound of BOD destroyed          •
is quite high.  This is due to the fact that the elemental phosphorus                  I
percipitated from the system contributes to the total sludge volume.  (Sentence deleted
                                                                                       I
3.3  ENVIRONMENTAL CRITERIA
     One purpose for evaluating treatment alternatives and options is                   •
ultimately to ensure that the treatment plants meet their environmental
limits.  Meeting these limits is predicated on a combination of conservative            I
design criteria, projection of hydraulic and pollutant loading rates, and
pilot testing to demonstrate system strengths and weaknesses under real-world           •
conditions.   To date, pilot testing in Columbus has utilized a sequencing              *
batch reactor (SBR), and most testing has been at the Southerly plant.  In
reviewing the work done to date, additional information needs to be gathered
on the impacts of blending Jackson Pike and Southerly primary effluent to
determine if nitrification rates can be sustained.                                      •
                                                                                        I
     It will also be necessary to limit the mass loading of TKN to the                  •
Southerly waste treatment plant in order for the nitrification process to be
effective.  Periodic high loadings of TKN have resulted in the bleedthrough of
ammonia from the primary effluent during the Project 20 pilot demonstration.
Unless these loads of TKN are controlled, all three biological processes would
be subject to ammonia bleedthrough resulting in violation of the permit                 •
ammonia concentration and mass-loading limits.

     Meeting total suspended solids and BOD limits is primarily a function of
clarifier efficiency.  Soluble BOD is rapidly removed in the aeration basin.            •
That portion of the BOD associated with the particulates in the wastewater as           •

                                    C-40
                                                                                        I

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
well as the suspended solids which escape from the clarifier, would cause BOD
or suspended solids violations.  Controlling suspended solids violations is
based on controlling the SVI of both Jackson Pike and Southerly biological
treatment systems.

     All three processes have the ability to select against  filamentous
organisms which cause bulking.  The semi-aerobic and activated sludge  systems,
as demonstrated by Project 20 data, could reduce SVIs and keep ammonia
concentrations well within permit limits given the absence of slug primary
effluent ammonia loadings.  Operating data for the Southerly waste treatment
plant from 1983 through 1986, indicate SVIs in the range of  75 to 181  are
possible.  (Sentence deleted)

     Denitrification is equally important during the summer months.
Denitrification will prevent the formation of a rising sludge in the final
clarifiers.  No amount of clarifier upsizing or clarifier configuration
modification can prevent a violation during episodes of rising sludges.  It
is, therefore, necessary that the denitrifiers complete the  chemical reaction,
converting the nitrates into nitrogen and carbon dioxide, in the aeration
basin.  This is accomplished by overpumping the secondary clarifiers,
maintaining a minimum sludge blanket in those clarifiers, and holding  the
mixed liquor suspended solids in the aeration basin to 3500 mg/1 (Southerly
plant).  Denitrification also has the side benefit of eliminating nitrites and
nitrates from the plant effluent.

     At the present time there is no nitrate or nitrite standard in the Ohio
EPA permit limitations written for the Jackson Pike and Southerly plants.
However, removing these pollutants from the effluent wastewater would  result
in the removal of pollutants from the receiving waters and subsequently any
groundwaters which are recharged from the surface waters.  Denitrification is
considered a benefit, not only in terras of removing unwanted pollutants from
the surface waters and the groundwaters of the state, but also in terms of
limiting the occurrence of rising sludges in the secondary clarifiers.
                                     C-41

-------
     Another secondary benefit from the semi-aerobic process would be that it
is a biological phosphorus removal system.  Although phosphorus removal
increases the volume of sludge to be treated by both Southerly and Jackson
Pike, it also results in the removal of a nutrient pollutant from the surface
water and groundwater.

     A negative environmental impact of odor and pests in the form of flies  is
associated with the trickling filter/activated sludge option.  Trickling
filters have been cited in odor complaints particularly under conditions of
high organic loadings such as will be employed in the roughing filters
proposed for Jackson Pike and Southerly.  In addition, fly larvae and flies
have been known to breed on these filter media resulting in nuisance
complaints.  Attempts to control odors and flies by covering the trickling
filters results in the installation of a drafting system to allow adequate air
to pass through the filter media.  This would add cost to the system and may
result in reduced efficiency during the summer months.

3.4  COSTS
     The biological treatment train cost components (shown in Table 3-11)
include trickling filters, aeration basins, aeration system blowers, blower
housing, diffusers, (and internal recycle loops in the case of semi-aerobic
systems), and clarification processes.  Table 3-11 provides a comparison
between the costs from the Revised Facility Plan Update and the briefing paper
costs.

     In general, costs developed for this briefing paper are lower than those
presented in the facility plan due to lower projected average and peak flows.
In general, the higher facility plan costs at Southerly for the two-plant
alternative are due to the fact that a new east train was required.  The lower
projected flows used  in the briefing paper analysis did not require a new  east
train for the two-plant option.
                                     C-42

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I

































C.O
H
CO
0
o

i—3
•<
H
CM
<
0

fa
0
z
0
CO
r-l
{*

Pw
2
O
0


rH
1— 1
1
CO

W
r4
PQ
^
EH





































Cd
U
H ^
r-3 en
M t-l
fa «
O r-l
Z O
I—I *O
,_J
^ 14-4
CJ O
Crf en
H C
O
• *iH
CO r-l
> i— 1
• I-l
u e
M
CQ (3
O -H
Of
U en
•O 4J
1 en
M O
S o

CO












a
w
z

03
s
0
0








60
C
•H M
<4-l (U
a) a
•i-l CO
r-l P 1
«

CO
->
•UJ
•r-l ti
rH CO
•r-l r— 1
CJ pn
CO
fa

CO





C
•r-l r-l
LM cu
^
4J
•r-l C

U CM
cO
fa

CO
•^
'"1*»^
fa
H


^
CO








^-4
jj
Pfj
W

H
ri3
0
CO









60
C
•r-l U
14-1 a)
a) o.
•H CO

03

CO
<^
^^,
fa
H



•*,
4J
• r-l G
rH CO
•r-l i— 1
0 CM
cfl
fa

CO
^
^>^
fa
H








1
1





0
•
1— <
r-l


1
1



O
•
r-l
rH



1
1




r**
•
CO
CM



1
1

in
•
O



in
•
o
^j-


00
•




in
•
0
CM



in
•
0
CM


CO
•
vD
CM


00
•
co
m
•
r*^
oo


O

CM
00


m
•
r-l
rH


^J>
•
in
0
rH

rH
•
rH
-<}•


CM
•
1 — 1
co




^D
•
^
-d-

V0
•
CO
co


r~.
•
00
co



oo
•
CM
co


CJS
•
< — 1
r~~


in
•
oo



•
en
S
o
i— i
M-l

tl
aj
4-t
CJ
a)
0
r-l
Cu

V-i
a>
3
O
I — 1
o
4J

eu
3
•a

en
4J
en
0
O

C
cO
i— i
Cu

>-,
•r-l
r— 1
• H
O
CO
fa
C
CO
i-1
4-1

p
OJ
^
o
rH

>>
1 — 1

C
cu
4J
en
• i-i
en
C
o
0
CD
4J
g
D
c
0
a
6
0
o

^j
en
O
CJ
i— i
• r4
fa

60

• i-l
rH
J*!
O
• r-l
i-l
H





C
0
• I-l
4J
CO
rJ
0)

c
o
•I-l
4J
CO
O
• H
U-l
•iH
^-1
CO
i— I
CJ








^4
•^
H
0
H
    C-43
                                              CO
                                              o
                                              a

                                              1-1
                                              
-------
     Under the combined plant option, the lower costs associated with the
briefing paper estimates are due to the requirement for fewer treatment
facilities based on lower flows.

     In most cases, the serai-aerobic option was less costly than the trickling
filter/activated sludge option.  This is due to the fact that a significant
portion of the required aeration capacity already exists.

     The briefing paper analysis also reviewed the clarifier evaluation
prepared during the facility planning process.  In this clarifier evaluation,
it was assumed that the Southerly One-Plant Alternative would be implemented.
Three clarifier configurations were developed.  These included:

     •  Alternative 1:  Construction of 12 new 200-foot diameter clarifiers,
     •  Alternative 2:  Constructing 6 new 200-foot diameter clarifiers  for  the
        east train, using all existing clarifiers for the center train,  and
        constructing three 200-foot diameter clarifiers for the west train.
     •  Alternative 3:  Constructing 6 new 200-foot diameter clarifiers  for
        the east train; using existing rectangular clarifiers and adding 2 new
        175-foot diameter clarifiers for the center train; and using the
        existing rectangular clarifiers and adding 2 new 175-foot diameter
        clarifiers to the west  train.

     Alternative 3 was discarded as being unworkable in terras of hydraulic
limitations.  Alternatives 1 and 2 were evaluated with a cost of $43,194,000
for Alternative 1, and $40,126,000 for Alternative 2.  The RFPU study
concluded that the two alternatives exhibit similar present worth costs.
Consequently, due  to the advantages of circular clarifiers, it was  recommended
that the existing  clarifiers be demolished and new 200-foot clarifiers be
installed.  These  advantages include:

     •  Easier flow splitting and control of  flow to each clarifier.
     •  Reduction  in the number of telescoping sludge valves to be  controlled.
     •  Ability to provide flocculation within the clarifier.
     •  Less potential risk  for shortcircuiting.
     •  Automatic  scum removal  for the entire surface.
     •  Less complicated construction phasing.
                                    C-44

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     In addition to these advantages the capability to rapidly return sludge

to the aeration basin is a distinct advantage of circular clarifiers.  Due to

the low nitrification rates at Southerly, the briefing paper evaluation

concurs with the advantages of circular clarifiers and recommends their

installation at Southerly.
                                    C-45

-------
       4.   COMPARISON OF BRIEFING PAPER AND FACILITY PLAN CONCLUSIONS
4.1  PROCESS SELECTION
     The briefing paper analysis concurs with the facility plan in its
selection of the semi-aerobic process as the preferred biological process.
The process is superior to the trickling filter/activated sludge process due
to its ability to provide nutrient removal and the flexibility of process
control it affords operators.  As  previously stated,  the  semi-aerobic  system
is essentially the same as the conventional  activated sludge  system and could
easily be operated in the conventional activated sludge mode if necessary.
Although the trickling filter/activated sludge option is  considered reliable,
it does exhibit the disadvantages of producing nuisance odors and pests, and
requires additional space to implement.

     Process selection is also predicated on the assumption that Columbus will
implement and enforce a rigid  industrial  preteatment  program  which will limit
the concentration of toxic pollutants and slug loads  of ammonia.  Pilot data
have indicated that slug loads of  ammonia or TKN will pass through the primary
clarifiers and may result in  ammonia bleedthrough from the aeration basins.
These influent conditions must be controlled to ensure that any biological
process will perform effectively and meet permit limitations.

4.2  CLARIFIER UTILIZATION
     The briefing paper evaluation agreed with  the facility plan
recommendation  to demolish  the existing rectangular clarifiers at Southerly
and replace them with  new circular  clarifiers.  Due to the lower flows and
loads utilized in the briefing paper analysis,  less facilities  are recommended
in the briefing paper for both the Southerly one-plant and Southerly two-plant
alternatives.

     Contrary to the RFPU, the briefing paper recommends retaining the
existing retangular clarifiers at Jackson Pike.  The  arguments for the
selection of circular clarifiers at  Southerly, primarily high mixed liquor
                                    C-46

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
suspended solids,  the need for overpumping,  and  low nitrification rates do not

apply to the Jackson Pike waste treatment facility.  The 12 existing

rectangular clarifiers at Jackson Pike should be rehabilitated,  and 2 addi-

tional rectangular clarifiers should be constructed to provide adequate final

effluent clarification capacity.



4.3  ONE-PLANT VS. TWO-PLANT


     The decision to utilize a combined one-plant option versus  a two-plant

option must be based on process reliability as well as  cost factors.  The data

presented in this briefing paper shows the biological treatment process for

the two-plant option is less costly.  The  unknown factor at this point is the

effect of nitrification rates on blending Jackson Pike and Southerly primary

effluent.  In  the  absence  of this  data,  it is  speculative  to recommend a one-

plant versus two-plant option based on process considerations alone.
                                    C-47

-------

      APPENDIX D

 BRIEFING PAPER NO. 4
O&M AND CAPITAL COSTS

-------
 I
 I
 I
 I         BRIEFING PAPER  NO. 4
 I
 I
           0 & M AND CAPITAL COSTS

           Supplemental Environmental Impact Statement
           IUSEPA Contract No.  68-04-5035,  D.O. No.  40
           Columbus Ohio Waste-water Treatment  Facilities
 I
 I
 I
 I
 I
           Prepared By:
 •         SCIENCE APPLICATIONS INTERNATIONAL  CORPORATION
I         TRIAD  ENGINEERING INCORPORATED
I
I

-------
                            O&M AND CAPITAL COSTS
1.   DEVELOPMENT OF BRIEFING PAPER COSTS

     1.1  CAPITAL COSTS
     1.2  O&M COSTS
2.   FACILITY PLAN COSTS
3.   COMPARISON OF BRIEFING PAPER AND FACILITY PLAN COSTS

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                                 INTRODUCTION



     Under the direction of USEPA,  a series  of briefing papers  are being

prepared addressing key issues in the development of the Supplemental

Environmental Impact Statement for the Columbus,  Ohio, Wastewater  Treatment

Facilities.   The  briefing papers  form the  basis  of  discussions  between Triad

Engineering and USEPA to resolve  important issues.   The  following  paragraphs

present the background of the facility planning process, a description of  the

briefing papers,  and the purpose  of this paper on costs.



FACILITY PLANNING PROCESS


     At the time this paper was prepared  (March-July 1987) the  city of

Columbus was proceeding to  implement improvements at the Jackson Pike  and

Southerly Wastewater Treatment Plants to comply with more stringent effluent

standards which must be met by July  1, 1988.   These  improvements  were based

on the consolidation of wastewater treatment operations at the  Southerly

plant.  This  one-plant  alternative  is a  change from  the  two-plant  operation

proposed by the  city in the  1970's and evaluated in the 1979 EIS.



     The development and documentation of  wastewater treatment  process and

sludge management  alternatives for the Columbus metropolitan area has  been an

extended and iterative process.    The design  and construction of various

system components  have progressed, because of the 1988  deadline, while

planning issues continue to be resolved.   As  a result, numerous  documents  have

been prepared which occasionally  revise  a previously established course of

direction.



     The concurrent resolution of planning issues and implementation of

various project components  has made preparation  of  the EIS more difficult

because final facility plan recommendations  are not available in a single

document.
                                      D-l

-------
BRIEFING PAPERS
     To facilitate preparation of the EIS,  a series of briefing papers are
being developed.   The purpose  of  the  briefing  papers  is  to  allow USEPA to
review the work of the EIS consultant and to identify supplemental information
necessary for the preparation  of  the  EIS.   Six briefing papers  are being
prepared as follows:

     •  Flows and Loads
     •  Sludge Management
     •  CSO
     •  Process Selection
     •  One Plant vs. Two Plant (Alternative Analysis)
     •  O&M and Capital Costs

     The specific focus of each briefing paper will be different.  However,
the general scope of  the  papers will  adhere to the  following format:

     •  Existing conditions will  be documented.
     •  Evaluations,  conclusions,  and recommendations of the facilities
        planning process will be reviewed  using available documentation.
     •  Where appropriate, an independent evaluation of the future situation
        and viable alternatives will  be prepared.
     •  The facility plan and  EIS briefing paper conclusions will be compared.

     The briefing paper process is intended to:

     •  Prompt the resolution of any data deficiencies.
     •  Clearly establish and define existing and future conditions.
     •  Identify the final recommended plan which the city desires to implement.
     •  Provide a data base of sufficient detail to allow preparation of the
        draft EIS.
                                      D-2

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 It
 II
 II
 R
I
I
I
O&M AND CAPITAL COSTS


     This briefing paper presents capital and operation and maintenance costs

associated with the Southerly One-Plant and Jackson Pike and Southerly Two-

Plant alternatives.  Topics addressed include:



     •  Development of briefing paper capital and O&M costs

     •  Facility plan capital and O&M costs

     •  Comparison of briefing paper and facility plan costs



     The briefing paper cost analysis is based on the 2008 design flows and

loads which were presented in the Wastewater Flows  and  Loads Briefing Paper.

The facility plan costs are taken from  the  1985  Revised Facility Plan Update.

These costs were developed for a 30-year planning period ending in 2015.
                                      D-3

-------
                   1.  DEVELOPMENT OF BRIEFING PAPER COSTS
     This section presents briefing paper capital and O&M costs for the
Southerly One-Plant, Southerly Two-Plant, and Jackson Pike Two-Plant
alternatives.  Due to the differences in design flows between the facility
plan and the briefing papers, an independent cost analysis was prepared based
on the 2008 flows and loads developed in the Wastewater Flows and Loads
Briefing Paper.  Table 1-1 presents the flows and loads which were used as a
basis for developing these costs.  The Process Selection Briefing Paper
recommended the semi-aerobic process.  Therefore, O&M and capital costs were
developed assuming semi-aerobic as the biological process being employed.
Solids handling costs are consistent with facilities recommended in the Solids
Handling Briefing Paper.

                TABLE 1-1.  BRIEFING PAPER FLOWS AND LOADS
                             Jackson Pike      Southerly
Average Flow (MGD)                  70               84
Peak Process Flow (MGD)            100              131
BOD Load (Ib/day)              112,600          155,600
TSS Load (Ib/day)              128,500          154,400
NOTE:  Average flows in excess of 70 MGD and peak process flows in excess of
       100 MGD at Jackson Pike will be diverted to Southerly under the two-
       plant alternative.

1.1  CAPITAL COSTS
     Detailed cost estimates prepared during the facilities planning process
by the Turner Construction Company were utilized in preparing the construction
costs for this briefing paper.  These cost estimates were reviewed in detail
and adjusted as appropriate to account for differences in the briefing paper
design flows and unit process sizing.  Table 1-2 presents the construction
costs for the Southerly One-Plant, Southerly Two-Plant, and Jackson Pike Two-
Plant alternatives.
                                      D-4

-------
I
I
                             TABLE  1-2.   BRIEFING  PAPER CAPITAL  COSTS


•                                        Southerly         Southerly        Jackson Pike
           Cost Component                 (One-Plant)        (Two-Plant)        (Two-Plant)

           (Site Work                     $ 22,932,000       $  11,448,000       $  1,550,000
           Miscellaneous Buildings          5,232,000          4,857,000          1,857,000
           Plumbing/HVAC                    5,875,000          5,875,000          4,337,000
           IHeadworks                       14,300,000            —              8,271,000
           Preaeration                      5,905,000          1,533,000          3,750,000
           Primary  Settling                13,590,000          4,717,000          7,372,000
m         Aeration                       46,533,000         12,284,000         22,502,000
•         Final  Settling                  35,462,000         20,521,000          8,691,000
»         Chlorination                     4,000,000          2,500,000          2,000,000
           Effluent Pumping                 6,270,000            —              4,340,000
I           Outfall  Line                     3,000,000            —                700,000
           Gravity  Thickening               5,070,000          2,520,000          1,967,000
           Digestion                       11,460,000          4,280,000          9,170,000
—         Centrifuge Thickening            5,600,000          2,000,000          4,500,000
•         Centrifuge Dewatering           21,040,000          5,120,000            490,000
*         Dewatered Sludge  Storage         1,300,000          1,300,000
           Incineration                     1,300,000              —             3,600,000
           •Sludge Conveyor System               —                 —             5,000,000
           Instrumentation & Control       10,070,000          4,799,000          6,995,000
           Electrical Distribution          1,896,000          1,896,000            607,000
I           Jackson  Pike Rehabilitation     13,564,000
           Interconnector South             4,982,000
           Interconnector North             5,048,000


I


I


I


I


I


I


I



I
                                               	—             5,048,000


TOTAL CONSTRUCTION COSTS     $244,429,000      $ 85,650,000      $102,747,000


Contingency (15%)              36,664,000        12,848,000        15,412,000
Land                              200,000           200,000
Salvage Value (PW)           - 12,582,000      -  4,644,000      -  5,137,000

CAPITAL PRESENT WORTH        $268,711,000      $ 94,054,000      $113,022,000
                                      D-5

-------
1.2  O&M COSTS
     Operation and maintenance costs were developed for each plant alter-
native.   The costs are presented in Table 1-3.   The basis  of these costs are
described in the following paragraphs.

               TABLE 1-3.  BRIEFING PAPER ANNUAL O&M COSTS ($)

                        Southerly       Southerly       Jackson Pike
                        (1-Plant)       (2-Plant)        (2-Plant)
Labor                   4,050,000       2,850,000        2,880,000
Material & Supply       4,446,000       2,250,000        2,412,000
Chemicals               1,197,000         580,000          658,000
Energy                  4,800,000       2,425,000        2,425,000
Land Application          712,000         342,000          712,000
Composting              1,314,000       1,314,000
Ash Disposal              330,000          60,000          170,000
TOTAL                  16,849,000       9,821,000        9,257,000
NOTE:  These costs are based on 2008 design flows and loads.

     Labor costs for operation and maintenance were determined by evaluating
information on the number of employees currently employed at the treatment
facilities and their  respective  salaries.  An average annual salary (including
benefits) of $30,000  per  employee was established  for future cost  projections.
The projected number  of  workers  for  each alternative  is 135 for the Southerly
One-Plant, 95 for the Southerly Two-Plant,  and 96 for the  Jackson Pike Two-
Plant.

     Typically, annual material  and  supply costs are estimated as a percentage
of total construction costs.   However, in this situation,  with a portion of
the facilities already in place, doing so may underestimate the  actual  cost.
Therefore, cost curves were used to determine the construction costs for each
plant alternative as a new facility.  One percent of this  cost was estimated
as the annual material and supply  cost.
                                      D-6

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     Chemical costs were determined  for three processes:  chlorination,
centrifuge thickening,  and centrifuge dewatering.  A current chlorine cost of
                              »
$200 per ton and a polymer cost of $1 per pound were used for these estimates.

     Costs for electrical  energy were estimated  based on costs  documented in
the 1985 Operating Report  prepared by the City of Columbus Division of
Sewerage and Drainage.   These costs were adjusted to account for the
following:

     •  An increase in power costs from $0.04 to $0.05 per kilowatt-hour
     •  An increase in average flow from 145 MGD to 154 MGD
     •  Additional oxygen  requirements for nitrification

     Fuel cost estimates were determined based on the assumption that the
future solids handling scheme would  include digestion and dewatering to a
minimum  cake  solids content of 22 percent.   Under this  assumption,  enough
sewage gas is produced to meet  the fuel requirements of the incinerators and
the digesters.  Additional  fuel  cost  estimates  for heating and  service were
based on costs documented in the 1985 Operating Report prepared by the City of
Columbus Division of Sewerage and  Drainage.  Fuel costs for heating and
service were estimated as $200,000 per year for each plant under the two-plant
alternative and $350,000 per year for the Southerly One-Plant Alternative.
The total energy cost  for  the Jackson Pike WWTP and the Southerly WWTP in 1985
was $4.5  million.   In  comparing this cost to the 2008 projected cost of $4.7
million, it must be remembered that the following factors differ between the
two costs.

     •  In 1985 the Southerly digesters were not operating.
     •  Dewatered cake  solids at both plants averaged only 17 percent in 1985.
     •  Power costs in 1985 were $0.04 per kilowatt-hour.
     •  There is a 6 percent increase in average flow from 1985 to  2008.
                                     D-7

-------
     Land application is a contract operation.   Based on past contract costs
from the city, it has a unit cost of $15 per wet ton.

     Operation and maintenance costs for the compost facility were estimated
based on historical O&M costs.  A unit cost of  $30 per wet ton was used.  This
cost includes materials, supplies, energy, and  labor.

     Ash disposal, which includes hauling and landfilling,  was estimated at a
cost of $15 per cubic yard.

     The total present worth O&M  cost  for  the combined Jackson Pike and
Southerly Two-Plant option is $189,940,000.  This cost is 13 percent higher
than the Southerly  One-Plant  cost of $168,200,000.

     Table 1-4 presents the present worth of the O&M costs for each plant
alternative.

                   TABLE 1-4.  BRIEFING PAPER O&M COSTS  ($)
                                                                                       I
                                                                                       I
Annual O&M
1988-1992
1993-1997
1998-2002
2003-2007
Total
Southerly
[1-Plant]
16,047,000
16,247,500
16,448,000
16,648,500

Southerly
[2-Plant]
8,848,000
9,091,000
9,334,000
9,577,000

Jackson Pik
[2-Plant]
9,257,000
9,257,000
9,257,000
9,257,000

Present Worth
(1988)
168,200,000
94,140,000
95,800,000
                                      D-8

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                           2.  FACILITY PLAN COSTS


     Costs were presented in the Revised Facility Plan Update  (RFPU)

for the Southerly One-Plant,  Southerly Two-Plant, and Jackson Pike Two-Plant

alternatives.   These  costs are  for  facilities  which were sized based on the

flows and  loads presented in Table  2-1.



                  TABLE 2-1.   FACILITY PLAN FLOWS AND LOADS



                     Jackson Pike    Southerly    Whittier Street    Total


Average Flow (MGD)         101             75             —             176
Peak Process Flow (MGD)    172            128             —             300
CSO (MGD)                   -              -              130            130
BOD Load (Ib/day)       148,620        131,740          10,000        290,360
TSS Load (Ib/day)       170,390        126,550          20,000        316,940


NOTE:   Flows at Jackson Pike in excess of 100 MGD will be diverted to
       Southerly under the two-plant alternative.  The additional flow of 130
       MGD of CSO will be transported to Southerly under either alternative.


     These flows and loads differ from those  used  in  the briefing  paper.

Table 2-2 presents the Revised Facility Plan Update capital costs associated

with these flows and Table 2-3 presents  the  RFPU O&M costs for the one-plant

and two-plant alternatives.   The O&M costs associated with wet stream

treatment and solids handling were increased throughout the planning period to

account for increases in flows and loads. The O&M costs for headworks,

administration, Whittier Street facilities,  and the Jackson Pike diversion

chamber were held constant throughout the planning period.



     The RFPU O&M costs also  include an amount allocated to "Other Capital

Costs".   These  costs  were originally estimated with capital costs.   The costs

are for rehabilitation or replacement of  existing equipment.  Table 2-4 shows a

breakdown of these costs.
                                     D-9

-------
                   TABLE 2-2.  FACILITY PLAN CAPITAL COSTS
                                                                                       I
                                                                                       I
Cost Component

Site Work
Miscellaneous Buildings
Plumbing/HVAC
Headworks
Storm Bypass
Stormwater Tanks
Preaeration
Primary Settling
Aeration
Final Settling
Effluent Filters
Chlorination
Effluent Pumping
Outfall Line
Gravity Thickening
Digestion
Centrifuge Thickening
Thermal Conditioning
Centrifuge Dewatering
Incineration/Ash Lagoon
Lime Stabilization
Instrumentation & Control
Electrical Distribution
Jackson Pike Rehabilitation
Whittier Storm Tanks
Whittier to Jackson Pipe
Grit to Flow Diversion Pipe
Interconnector North
Interconnector South
Miscellaneous

TOTAL CONSTRUCTION COSTS

Engineering Fees
Land Acquistion
Process License Fees
Salvage Value

CAPITAL PRESENT WORTH
Southerly
(One-Plant)
$ 24,817,490
6,314,840
100,000
26,278,310
2,316,950
5,506,390
8,802,450
15,734,910
63,605,700
41,812,710
50,066,830
6,489,190
9,221,730
2,491,210
5,866,755
9,833,400
7,766,720
—
11,943,104
2,546,770
1,200,000
11,439,090
2,097,610
15,000,000
7,465,180
3,782,300
4,738,940
5,727,010
5,509,780
—
Southerly
(Two-Plant)
$ 21,120,060
5,856,230
100,000
19,536,520
2,316,950
5,506,390
6,516,490
13,020,930
34,661,730
33,848,295
29,682,210
4,367,810
122,340
—
4,781,660
5,913,650
5,120,895
—
6,721,880
2,546,770
1,200,000
8,697,710
2,097,610
—
7,465,180
3,782,300
4,738,940
6,279,510
5,509,780
—
Jackson Pike
(Two-Plant)
$ 8,056,750
3,514,900
3,071,870
10,163,620
—
—
4,148,080
5,819,100
23,856,500
9,832,890
25,060,510
3,218,280
7,321,300
796,280
7,272,140
9,377,250
6,917,630
3,030,260
517,580
3,975,830
—
8,331,650
607,160
—
—
—
—
—
—
5,000,000
$358,475,369

  42,592,748
     200,000
   8,000,000
  -5,431,000

$403,837,117
$241,511,840

  26,629,941
     200,000
   4,000,000
  -3,774.000

$268,567,781
$149,889,580

  15,962,812

   4,000,000
  -2.059,000

$167,793,000
                                      D-10

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                  TABLE 2-3.  FACILITY PLAN O&M COSTS ($/YR)
WETS IDE
1988-1990
1991-1995
1996-2000
2001-2005
2006-2010
2011-2015
SOLIDS
1988-1989
1990-1999
2000-2015
WHITTIER STREET
2-PLANT
SOUTHERLY
4,642,259
5,694,560
6,078,667
6,466,235
7,030,124
7,694,711
6,109,400
6,380,500
6,511,500

2-PLANT
JACKSON PIKE
3,455,820
4,225,624
4,589,694
4,976,569
5,625,354
6,416,125
4,887,200
5,090,000
5,192,400

2-PLANT
TOTAL
8,098,079
9,920,184
10,668,361
11,442,804
12,655,478
14,110,836
10,996,600
11,470,500
11,703,900

1 -PLANT
SOUTHERLY
6,478,414
7,953,897
8,408,234
8,929,529
9,581,484
10,451,592
8,100,200
8,426,500
8,956,400

1995-2015


JP DIVERSION CHAMBER


1988-2015


HEADWORKS


1988-1089

1990-2015            687,400


ADMINISTRATIVE


1988-2015            400,000


OTHER CAPITAL


1986-2000            915,995


TOTAL
Present Worth
   (1985)
127,734,009
                     47,900
                    136,000
                  1,074,660
                  1,074,660
                    400,000
                    366,066
104,785,539
                     47,900
                    136,000
                  1,074,660
                  1,762,060
                    800,000
                  1,282,061
232,519,548
                     47,900
                    136,000
                  1,219,500
                    500,000
                  1,141,329
176,166,114
                                      D-ll

-------
                TABLE 2-4.  FACILITY PLAN OTHER CAPITAL COSTS


                                                         COST ($)
SOUTHERLY (ONE-PLANT)
  Preaeration
       Replace Diffusers                                  193,000
       Replace Flushing Equipment                          26,000
       Replace Blowers                                     88,000
       Replace Cross Collectors                           111,000
  Pr raary Settling
       Replace Flights, Chains, and Cross Collectors    1,800,000
       Replace Skimming Equipment                          90,000
       Weir Replacement                                   100,000
  Digester Renovation                                   3,600,000
  Centrifuges - Automatic Backdrives                      300,000
  HVAC Renovation                                       8,747,500
  Jackson Pike Sewer Maintenance Yard                     975,440
  Incineration                                          1,300,000

TOTAL                                                  17,330,940

SOUTHERLY (TWO-PLANT)
  Preaeration
    •  Replace Diffusers                                  193,000
    •  Replace Flushing Equipment                          26,000
    •  Replace Blowers                                     88,000
    •  Replace Cross Collectors                           111,000
  Digester Renovation                                   3,600,000
  HVAC Renovation                                       8,747,500
  Jackson Pike Sewer Maintenance Yard                     975,440

TOTAL                                                  13,740,940

JACKSON PIKE (TWO-PLANT)
  Miscellaneous Building Renovation                       257,000
  Primary Building Renovation                             168,000
  Primary Tanks
       Not Filling Tanks                                  230,000
       Replace Flights, Chains, and Cross-Collectors      991,000
       Replace Skimming Equipment                         133,000
       Replace Weirs                                      214,000
       Replace Sluice Gates                               266,000
  Ae ation
       Replace Sluice Gates                               932,000
       Renovate Control Building                          612,000
  HVAC Renovation                                       1,918,000

TOTAL                                                   5,721,000
                                      D-12

-------
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
           3.  COMPARISON OF BRIEFING PAPER AND FACILITY PLAN COSTS
     Tables 3-1 and 3-2 present cost comparisons between the one-  and  two-plant
alternatives for the briefing paper and facility plan, respectively.   The
facility plan shows the two-plant alternative being 15 percent more costly  than
the one-plant.  However, the briefing paper shows the one-plant as 10  percent
more costly than the two-plant.  This difference between the facility  plan  and
the briefing paper is primarily a result of differences in design  flows.  At
the briefing paper's lower flows, a new east train, headworks, and expanded
Interconnector Sewer are not required under a two-plant alternative.   At the
facility plan flows, these facilities are required under either alternative.
However, they vary in size being larger for the one-plant.

     A direct cost comparison between the briefing paper and facility  plan
costs is not possible for several reasons.  There is a difference  in design
flows, costing methods, equipment (the facility plan's recommended CSO
facilities), and planning periods.

     The difference in flows between the facility plan is 22 MGD for average
flow and 69 MGD for peak flow.  This difference affects the costs  for  the
two-plant alternative more than the one-plant alternative.

     The method used in the facility plan for O&M costs also caused
differences in the capital costs between the briefing paper and the facility
plan.  As discussed in Section 2, the facility plan shifted some
rehabilitation costs from capital to O&M.  The following processes were
affected by this shift:

     •  HVAC renovation
     •  Preaeration
     •  Primary Settling
     •  Aeration
     •  Centrifuge Dewatering
     •  Incineration
     •  Digestion
                                      D-13

-------
      TABLE 3-1.  PRESENT WORTH OF BRIEFING PAPER CAPITAL AND O&M COSTS


                                Capital            O&M             Total

One-Plant [Southerly]         268,711,000      168,200,000      436,911,000

Two-Plant [So. and JP]        207,076,000      189,940,000      397,016,000

Difference From One-Plant     -61,635,000      +21,740,000      -39,895,000

Percent Difference                -30              +13              -10
NOTE:  These costs are based on a 2008 average flow of 154 MGD and a peak  flow
       of 231 MGD.  Present worth costs are in 1988 dollars.
                                      D-14

-------
 I
 I
 I
 I
 I
 i
 i
 i
 i
 i
 i
 i
 i
 i
 i
 i
 i
i
       TABLE 3-2.  PRESENT WORTH OF FACILITY PLAN CAPITAL AND O&M COSTS




                                Capital            O&M             Total


One-Plant [Southerly]         403,837,000      176,166,000      580,003,000


Two-Plant [So. and JP]        436,361,000      232,519,000      668,880,000


Difference From One-Plant     +32,524,000      +56,353,000      +88,877,000


Percent Difference                 +8              +32              +15
NOTE:  These costs are based on a 2015 average flow of 176 MGD and a peak flow
       of 300 MGD.  Costs are included for an additional  130 MGD of CSO
       facilities.  Present worth costs are in 1985 dollars.
                                      D-15

-------
     A. portion of the costs were for routine maintenance costs such as
painting and roof repairs.   These costs are covered under annual  maintenance
expenditures.   However,  some of the costs were for major equipment  renovation.
For example, the $6 million renovation of the existing digesters,  which have
not operated since 1980, was placed under the O&M costs as "Other Capital
Costs".   It was felt that these costs should remain in the capital costs.  The
city has indicated that  the digesters  are  currently undergoing renovation.
Therefore,  the briefing paper capital costs include these costs.   The cost of
renovating the HVAC and plumbing systems was placed under O&M costs in the
facility plan because the systems will be  replaced as they become inoperable.
The briefing paper included this cost as capital expenditures in the future
brought back to a present worth  amount.

     An additional factor in the cost difference between the briefing paper
and the facility plan costs is the recommended  CSO facilities.  Triad has not
included costs for facilities to control CSO as the city has.  Triad recom-
mends that a CSO study be completed prior to any recommmendations on CSO
facilities.  The following facilities are included in the facility plan costs
for CSO:

     •  Storm bypass
     •  Stormwater tanks at Southerly
     •  Whittier storm tanks
     •  OSIS Relief Sewer from Whitter Street to the flow diversion chamber.

     The costs for the CSO facilities are common to both a one and two-plant
alternative.  Therefore, they do not have a significant impact on the
comparison between the costs of one-plant vs. two-plants.

     Effluent filters are no longer  needed  at either plant due to changes in
the permit  limits.  Therefore, they were not included in the briefing paper
costs.
                                      D-16

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     The final factor-which causes a difference between the briefing paper and

facility plan costs is the planning period.  The faciltiy plan has a 30-year

planning period beginning in 1985 and ending in 2015.  The briefing paper

planning period extends 20 years, from 1988 to 2008.  This affects the flow

projections, which in turn affect the capital costs.  But more importantly, it

affects O&M costs.  The facility plan has 30 years of annual O&M  costs.  These

costs are presented in 1985 dollars.   The briefing paper only has 20 years of

O&M costs presented in 1988 dollars.
                                      D-17

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
       APPENDIX E
  BRIEFING PAPER NO. 5
COMBINED SEWER OVERFLOWS

-------
BRIEFING PAPER NO. 5
COMBINED SEWER OVERFLOWS
Supplemental Environineiital Impact Statement
USEPA Contract No.  68-04-5035,  D.O. No.  40
Columbus Ohio Wastewater  Treatment  Facilities
Prepared By:

SCIENCE APPLICATIONS INTERNATIONAL CORPORATION

TRIAD ENGINEERING INCORPORATED

-------
                           COMBINED SEWER OVERFLOW
1.   TERMS AND CONDITIONS
2.   AVAILABLE DATA
3.   CSO ANALYSIS
     3.1  Traditional Approach
     3.2  RFPU Approach:  Review and Critique
     3.3  Briefing Paper Analysis

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 i
 i
 i
 i
 i
 i
 i
i
i
i
                                 INTRODUCTION

     Under the direction of USEPA,  a series of briefing papers are being
prepared addressing key issues in the development of the Supplemental
Environmental Impact Statement for the Columbus,  Ohio,  Wastewater Treatment
Facilities.  The briefing  papers  form the basis of discussions between Triad
Engineering and USEPA to resolve important issues.  The following paragraphs
present the background of the facility planning process, a description of the
briefing papers, and the purpose  of  this  paper  on  Combined  Sewer Overflow
(CSO).

FACILITY PLANNING PROCESS
     At the time this paper was  prepared  (June-August  1987) the city of
Columbus was proceeding to implement improvements at the Jackson Pike and
Southerly Wastewater Treatment Plants to comply with more stringent effluent
standards which must be met by July 1,  1988.  These  improvements were based on
the consolidation of wastewater  treatment operations at  the Southerly plant.
This one-plant alternative is  a change  from the two-plant operation proposed
by the city in  the  1970's  and evaluated  in the 1979 EIS.

     The development and documentation of wastewater treatment process and
sludge management alternatives for the Columbus metropolitan area has been an
extended and iterative process.   The design and construction of various  system
components have progressed, because of the 1988 deadline,  while planning
issues continue to be resolved.   As  a result, numerous  documents have been
prepared which occasionally revise a previously established course of
direction.

     The concurrent  resolution of planning issues  and implementation of
various project components has made preparation of the  EIS  more difficult
because final facility plan recommendations are not available  in a single
document.
                                     E-l

-------
BRIEFING PAPERS
     To facilitate preparation of the EIS,  a series of briefing papers are
being developed.   The purpose of the briefing papers is to allow USEPA to
review the work of the EIS consultant and to identify supplemental information
necessary for the preparation of the EIS.   Six briefing papers are being
prepared as follows:

     •  Flows and Loads
     •  Sludge Management
     •  Process Selection
     •  O&M and Capital Costs
     •  CSO (Combined Sewer Overflows)
     •  One Plant vs. Two Plant (Alternative Analysis)

     The specific focus of each briefing paper will be different.  However,
the general scope of  the papers will adhere to the following format:

     •  Existing conditions will be documented.
     •  Evaluations,  conclusions,  and recommendations of the facilities
        planning process will be reviewed using available documentation.
     •  Where appropriate, an independent evaluation of the future situation
        and viable alternatives will be prepared.
     •  The facility  plan and EIS briefing paper conclusions will be compared.

     The briefing paper process is intended to:

     •  Prompt the resolution of any data deficiencies.
     •  Clearly establish and define existing and future conditions.
     •  Identify the  final recommended plan which  the city desires to
        implement.
     •  Provide a data base of sufficient detail to allow preparation of the
        draft EIS.
                                      E-2

-------
COMBINED SEWER OVERFLOWS
     This briefing paper presents an independent evaluation of the status of
the Combined Sewer Overflow (CSO) problem in the city of Columbus.  The
Supplemental Environmental Impact Statement being prepared will address only
Phases 1 and 2 of the city's  facility  planning  process.   The  city  has advised
USEPA that these two phases do not contain provisions for CSO.  Normal
facility planning processes incorporate CSO into the plan prior to developing
design flows for wastewater treatment  facilities.  However, the city of
Columbus intends to conduct a  detailed CSO analysis  after a majority of the
wastewater  treatment  facilities  are in place.

     The  purpose of this briefing paper is to  describe  the traditional
approach to the problem of CSO analysis and in this light, review and critique
the approach used in the 1985  Revised  Facility  Plan Update (RFPU).   While the
lack of data does not allow a comprehensive analysis of  the CSO problem,  some
general calculations are provided in comparison to those in the RFPU.
                                     E-3

-------
                          1.  TERMS AND DEFINITIONS

     The following terras and definitions are contained in Appendix A of the

current USEPA Region V NPDES Permit Strategy for Combined Sewer Systems.  This

list is reprinted in its entirety and thus,  not  all of the terras are referred

to in this briefing paper.   These terms are  used throughout the discussions in
this briefing paper.


Best Management Practices (BMPS); - means schedules of activities,
prohibitions of  practices, maintenance procedures,  and other  management
practices to prevent or reduce  the  pollution of  "waters  of the  United  States."
BMPS also includes treatment requirements, operating procedures, and practices
to control plant site runoff, spillage or leaks,  sludge or waste disposal, or
drainage from raw  material  storage.  (40 CFR 122.2).

Bypass: - the intentional  diversion of waste streams  from any portion of a
treatment facility.  (40 CFR 122.41(m)(4)).  "Treatment  Facility"  means
"Treatment Works" as defined below.

Combined Sewer: - a sewer that  is designed as a sanitary sewer and a storm
sewer.   (40 CFR 35.2005(b)(ll)).   (This  is distinguished from a  sanitary  sewer
to which inflow sources prohibited  by the sewer use ordinance have been
connected).

Complete Waste Treatment  System; -  a complete waste treatment system consists
of all  the treatment works necessary to meet the  requirements of title III of
the Act, involving:   (i) the  transport of wastewater  from individual homes or
buildings to a plant or  facility where  treatment  of the  wastewater is
accomplished; (ii) the  treatment of the wastewater to remove pollutants; and
(iii) the ultimate diposal,  including recycling or reuse, of the treated
wastewater  and residues which result  from the treatment  process (40 CFR
35.2005(b)(12)).   (the catch  basins  and overflow points are part of the
complete waste treatment  system in a combined  sewer system.   Also see  No. 17,
below.)

Dry Weather Flow; - flows that  are  not attributable to rainfall or snowmelt,
and include  infiltration.

Excessive Infiltration:  - the quantity of infiltration which can be
economically eliminated from a sewer system as determined in a cost-
effectiveness analysis that compares the costs for correction of the
infiltration conditions  to the total costs for  transportation and treatment of
the infiltration.  (40 CFR 35.2005(b)(16)).

Excessive Inflow; - the  quantity of inflow which can be economically
eliminated  from a sewer system  as determined by  a  cost effectiveness analysis
that compares the costs  for  correcting the inflow conditions  to the total
costs for transportation and treatment of the inflow (normally determined in
                                     E-4

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 i
 I
 I
I
conjunction with  the  determination of excessive  infiltration).  (40 CFR
35.2005(b)(16)).

Infiltration; - water other than wastewater that enters a sewer system
(including sewer service connections and foundation drains) from the ground
through such means as defective pipes, pipe joints, connections, or manholes.
Infiltration does not include, and is distinguished from,  inflow.  (40 CFR
35.2005(b)(20)).

Inflow; - water other than wastewater  that enters a sewer  system (including
sewer  service  connections) from sources such as, but not limited to, roof
leaders, cellar drains,  yard drains, area drains, drains from springs and
swampy areas,  manhole covers, cross connections  between storm sewers and
sanitary sewers, catch basins, cooling towers, storm waters, • surface runoff,
street wash waters, or drainage.   Inflow does not include,  and is
distinguished  from,  infiltration (40 CFR 35.2005(b)(21)).

Nonexcessive Infiltration; - The quantity  of  flow which is less than 120
gallons per capita per day (domestic base flow and infiltration) or the
quantity of infiltration which cannot be economically and effectively
eliminated from a sewer  system as determined in a  cost  effectiveness analysis.
(40 CFR 35.2005(b)(28)).

Nonexcessive Inflow; - The maximum total flow rate during storm events which
does not result in chronic operational problems  related to hydraulic
overloading of the treatment works or  which does not result  in a total flow of
more than 275 gallons  per capita  per day (domestic  base  flow plus infiltration
plus inflow).   Chronic operational problems may include surcharging, backups,
bypasses,  and overflows.   (40 CFR 35.2005(b)(29)).

Operational Plan;  - The  objective of the operational plan is to reduce the
total loading of pollutants  entering  the receiving stream  from  the  complete
waste  treatment system.  This plan, tailored to the local government's
complete waste treatment system,  will include mechanisms and specific
procedures to  ensure:

     a.  the collection and  treatment systems are operated  to maximize
         treatment;
     b.  all dry weather flows are treated to the level specified in their
         permit;
     c.  storm water entry into the sewerage  system is  regulated;
     d.  the sewerage  system hydraulic and storage capacity is  identified and
         fully utilized  during wet weather with eventual treatment of stored
         flows;
     e.  the greatest  quantity of  wet  weather  flows receive maximum possible
         treatment;
     f.  the sewerage  system is adequately maintained to ensure optimum
         operational capability.

Overflow;  - the uncontrolled  diversion  of waste streams  from a  combined sewer
system which occurs during wet weather when flows exceed conveyance capacity.
                                     E-5

-------
Sanitary Sewer; - a conduit: intended Co carry liquid and  water-carried wastes
from residences, commercial buildings, industrial plants, and institutions
together with minor quantities of ground, storm, and surface waters that are
not admitted intentionally.  (40 CFR 35.2005(b)(37)).

Sewer Use Ordinance: - that ordinance or other legally binding document
enacted to prohibit any new connections  from inflow  sources  into  the  sewer
system and require that new sanitary sewers and connections thereto are
properly designed and  constructed.   Such ordinance  shall  further  require that
all wastewater  introduced  into the sewer system does not contain toxics or
other pollutants in amount or concentration that endanger public  safety and
physical integrity of  the  sewer  system,  pump stations,  or wastewater  treatment
facilities,  cause violation of effluent  limitations  or water quality
standards, or preclude the selection of  the most cost-effective alternate for
wastewater treatment  and  sludge disposal.   (40  CFR  35.2130).

Storm Sewer: — A sewer designed to carry only storm waters, surface run-off,
street wash  waters,  and  drainage.  (40  CFR 35.2005(b)(47)).

Treatment Works; - Any devices and systems  for the storage,  treatment,
recycling, and reclamation of  municipal  sewage,  domestic  sewage,  or liquid
industrial wastes used to  implement  section 201 of  the Act, or necessary to
recycle or reuse water at  the  most  economical cost  over the design life of the
works.  These include  intercepting  sewers,  outfall sewers,  sewage collection
systems, individual systems, pumping, power, and other equipment  and
alterations thereof; elements essential to provide  a reliable recycled supply
such as standby treatment  and  clear water facilities;  and any works,  including
acquisition of the land that will be an integral part of the treatment process
or is used for ultimate disposal of residues resulting from such  treatment
(including land for composting sludge,  temporary storage  of such  compost, and
land used for the storage  of treated  wastewater  in land treatment systems
before land application);  or any other  methods or system for preventing,
abating, reducing, storing, treating, separating, or disposing  of municipal
waste or industrial waste, including waste in combined storm water and
sanitary sewer systems.  (40 CFR 35.2005(b)(48)).
                                     E-6

-------
                              2.  AVAILABLE DATA
     The operating records for both  the  Jaekson  Pike  and  Southerly  Wastewater
Treatment Plants include information regarding the major overflows for each
sewerage system, specifically, the Whittier Street Storm Tanks for the Jackson
Pike sewerage system and  the  bypass at  the plant for the Southerly sewage
system.  Additional reports which were reviewed for this analysis include the
fol lowing:

     •  Revised Facilities Plan Update (RFPU)  prepared  by URS Dalton  September
        30, 1985.
     •  General Engineering Report and  Basis of Design (GERBOD) prepared by
        URS Dalton January 31, 1986.
     •  Combined Sewer Overflow Monitoring Report prepared by Malcolm Pirnie,
        Inc. January 2, 1979.
     •  CSO Progress Report prepared by  Malcolm Pirnie,  Inc.  July 28, 1983.
     •  Central Scioto  River Mainstem Comprehensive Water Quality Report
        prepared by Ohio EPA September 30, 1986.
     •  Use of Combined Sewer Overflow Analysis in the  September 30,  1985,
        Revised Facilities Plan Update prepared by the  city of Columbus  March
        23, 1987.

     While the RFPU contains one page of conclusions reached in the CSO
analysis,  there is no information provided describing the analysis itself.
The GERBOD provides greater detail on the analysis performed, which references
data provided in the CSO Montioring Report and the CSO Progress Report.   The
final report cited was issued in response to questions  by USEPA - Region V in
regard to the CSO analysis referenced in the RFPU.
                                     E-7

-------
                               3.   CSO ANALYSIS

3.1  TRADITIONAL APPROACH
     The traditional approach to a combined sewer system analysis used in this
briefing paper is outlined  in a recent USEPA -  Region  V  document entitled
"Technical  Guidance for Use in the Development  of  a Combined Sewer  System
Operational Plan" published in September 1986.  The recommended tasks  in the
development of a stormwater management program are:

     •  Establishment of objectives
     •  Development of a data base
     •  Understanding the operation and response  of  the  combined sewer system
     •  Identification of drainage areas
     •  Hydraulic analysis
     •  Review of meteorological data
     •  Monitoring of flows and collection of samples
     •  Selection of mathematical  models
     •  Discussion of CSO control  alternatives

     The following paragraphs discuss these aspects  in greater detail.

     The establishment of objectives  is a project specific task which  leads
directly into the development of a data base.

     The development of a data base allows a municipality to become
knowledgeable about its combined sewer system,  including operation,
maintenance, and response to different meteorological conditions.  Data is
obtained through detailed interviews with sewer,  public  works, and  engineering
personnel.   The data required for  such a data base includes, but is not
limited  to:  geographical,  geological, topographical,  and hydrologic data;
known physical condition of  the sewer system, manholes,  and all appurtenances;
age, length, materials, sizes, slopes, and depths  of sewers; maintenance
practices;  problems and system failures; treatment plant flow  records  and
                                     E-8

-------
 I
 I
 I
 I
 I
 I
 I
 I
 1
 I
 I
 I
 I
 I
 I
I
I
I
I
charts; pumping station flow records, location of overflows and associated
operating experience and records;  identification  of sewer  system problem
areas; available combined sewer system maps; groundwater levels for all
seasons, with correlation to rainfall;  quality of local  receiving waters and
required effluent or water quality standards; and existing ordinances
governing inflow connections to sewers and enforcement programs and policies,
as well as estimates of the extent and significance of such inflow
connections.

     Detailed maps of the municipalities sewerage system should be up to date
and the combined,  sanitary and  storm sewer systems should be clearly defined.
The maps should show sewer sizes,  slopes,  direction of flow,  manhole
locations, and other major sewer system elements such as regulating or control
structures and overflows.   This information will allow a general understanding
of the operation and response of the  sewerage system.  These  maps and data
will also allow the identification of drainage areas  within the sewerage
system and thus establish  key hydraulic  locations where  flows  can be monitored
and gaged.

     The hydraulic analysis proceeds from the collection of data in regard to
the sewerage system itself.  This  analyis  allows  the  determination of the
hydraulic capacities of the sewers.   The portion  of the  sewer system capacity
available for carrying  stormwater  runoff is a function of  the total hydraulic
capacity of the sewerage system as determined by:  the pipe size, slope and
material of construction,  the quantitiy  of flows; and  the  level  to which a
particular sewer can surcharge  without causing an overflow,  basement flooding,
or other damage.   The most  important  parameter which  may be determined in the
hydraulic analysis is that of  the  time of concentration, which is used in the
computation of the peak stormwater runoff rate.   Since most  rainfall events
are of short duration,  the peak rate  of  runoff is of primary  importance,  with
the total volume secondary.

     The monitoring of flows and collection of samples should proceed only
after the hydraulic analysis and data collection  efforts previously described
                                     E-9

-------
have been completed.   At key locations  in the  sewerage system and  at  the major
overflow points, as previously determined,  flows should be monitored  along
with collection of rainfall data.   Further,  it  is  desirable  that samples be
collected at the overflows  at  short  time  intervals during an overflow in order
to associate pollutant concentrations  with the overflow. The rain gauges
should be located throughout the sewer  system  service  area  in order to
characterize the rainfall in terms of duration, intensity,  and volume.  Thus,
in the knowledge provided by the data  base and hydraulic analysis  that all
flows are accounted for, the volume of overflow and  mass loading of pollutants
discharged to the receiving waters  may  be computed.  Then the intensity
duration relationships of the  observed  rainfalls may be compared to historical
records to relate the observed overflows to a  recurrence interval.   Thus,
statistical relationships may be developed which will  relate rainfall to
overflow volumes  and quality.

     The CSO analysis may then be taken a step further in complexity  with the
use of a mathematical  model.   Mathematical  stormwater  models are capable of
predicting the volume of stormwater discharge  and its constituent pollutants.
These models typically consist of  two elements - a runoff element  that
simulates the washoff of pollutants by rainfall on the watershed,  and a
transport element that simulates the movement  of those pollutants  in  the sewer
system  and their  eventual discharge from it.   Data from flow monitoring and
sampling efforts may be used to calibrate such  a model after  which  the model
is an invaluable tool in evaluating the effectiveness  of various control
alternatives and to identify optimum solutions.

3.2  FACILITIES PLAN APPROACH:  REVIEW AND CRITIQUE
     Very little of the data necessary  for  a traditional CSO analysis  was
presented in the RFPU.   Existing monitoring data was utilized from two support
documents:  The Combined Sewer Overflow Report of January 1979,  and the CSO
Progress Report of 1983.
                                     E-10

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
     While  Che quality of the monitoring data used is unknown, its value is

of some question due to the  poorly developed data base regarding  the sewerage

system.  No maps of the sewerage system were presented or developed, thus, the

drainage areas for each overflow or each system  (combined, sanitary, or storm)

were not clearly defined.  While  the CSO Progress Report (1983) did include a

discussion of computer modeling of the sewerage system with reference to the

Whittier Street Storm Tanks using the SWMM model, the input data  was not

presented and the results were not discussed or  presented.   The Combined

Sewer Overflow Report  (1979) includes a one year record of  overflow monitoring

data for nineteen overflow sites,  but the  completeness of this record is

subject to some question.  Further,  due to  the  lack  of data in regard to how

the sewerage system responds to wet weather conditions, whether or not all

major overflows  are accounted for with the monitoring data is unknown.  The

fact that both support  documents and the RFPU overlooked the Renick Run

overflow supports this contention.



     The discussion of the Combined Sewer Overflow analysis in the RFPU makes

four points:



     1. The CSO analysis  consisted of ten  rainfall (overflow)  events from 1979
        and 1982.


     2. The statistical analysis performed on these ten events showed that the
        80th percentile storm could  be controlled at  a cost of $42 million.


     3. The environmental impacts of the existing combined  sewer  overflows
        were shown to be  insignificant according to documentation in the Draft
        OEPA Central Scioto  River Water Quality Report and the city river
        sampling results  as  reported in the monthly  operating  reports (MORs).


     4. The city would  meet  its  NPDES permit requirements regarding the sewer
        system overflows and would continue to closely observe the Scioto
        River, the Olentangy  River,  and Alum  Creek in  order to mitigate any
        adverse environmental impacts due  to overflows.


     There is no other  detail provided on  the CSO analysis  in  the RFPU other

than to make these four points.  It  was  not  until four  months  later that the

analytical  methodology for the CSO analysis was presented in the  GERBOD

(January 31,  1986) and  later  in  the  report  titled  "Use of Combined  Sewer
                                     E-ll

-------
Overflow Analysis" in the September 30, 1985,  Revised Facilities Plan Update
(RFPU) which was submitted to USEPA - Region V March 23, 1987,  by the city.

     While it was stated  in  the RFPU that the  environmental impacts of the
existing combined sewer overflows were shown  to be insignificant according to
documentation in the draft OEPA Central Scioto River Water Quality Report
(CWQR),  a review of this report did not substantiate this statement.  In fact,
information in the CWQR suggests that the environmental impacts of the
existing CSOs are significant.   On  page 195  the CWQR states that "combined
sewer overflows, and as previously discussed,  plant bypasses also contributed
significant loadings of BODc, NH-j-N,  TSS,  and  other substances  to the Central
Scioto River Mainstem".  Further, page 317 states, "Reductions  in the
magnitude and frequency of combined sewer overflow discharges is needed to
improve aquatic community function, alleviate aesthetic problems, and reduce
risks to human body contact  recreation  in  the  segment between Greenlawn Dam
and the Jackson Pike WWTP".

     The combined sewer overflow analysis presented in the RFPU considered
only Whittier Street overflows and  neglected all others including the bypass
flows at Southerly.   The city analyzed  ten events  which were selected from a
larger data set of twenty-six events.  The ten events  selected  were those that
had both quality and quantity data.   Data  for  the  events not selected was not
presented.   Thus, due  to  the  manner  in  which the ten events were selected,
whether or not they can be considered representative of flows at the Whittier
Street facility is questionable, and  whether or not they can be considered to
be representative of combined sewer overflows  from the entire sewerage system
is more questionable.

     The statistical analysis performed for the RFPU consisted of plotting the
ten events  on probability  paper  using a simple Weibull plotting position
calculation.   Using this method, m is  the rank of  the event  [highest (1) to
lowest (10)] and n is the number of events (10).   The  plotting  position thus
calculated  refers to the probability or return period that is associated with
each of the observed events.   The use of  this  method is illustrated in Table
                                     E-12

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
1, where Che calculated plotting positions and associated return period for

each of the ten events analyzed are shown.  Thus, for event number 1 of

10/8/77, a  plotting  position  of 0.364 is calculated.  This number refers to

the fact that 36.4 percent of  the  observed overflow volumes are less than that

of event number 1, while  63.6 percent of  the observed overflow volumes are

greater that that of event number  1.  Thus, if the data set is established and

representative, projections may be made on the overflow volumes.  This method

further defines the recurrence interval (in years) as the inverse of the

calculated plotting position.   Thus,  the recurrence interval for event number

1 with a volume of 44.1 would be 1/0.364 or 2.75  years.   Thus  an overflow of

this magnitude could be expected to occur every  2.75 years.  Note,  however,

that the data base from which this  projection is made consists of ten hand

picked events from a period of time of  about  one year.  While the objective of

this method is to make such projections  with a limited amount of data, the

questions still remain as  to how representative  these ten events are and what

about the other overflows  in the  system?  The calculated probabilities of the

ten events may have easily been checked using the rainfall data for each event

and associating a recurrence  interval  with the rainfall  intensity which

induced the overflows based on a histrocial record  of rainfall  for the area.

However, this check was not performed.



     Thus,  the 80th percentile overflow was shown to be 50 million gallons.

In addition,  the  80th percentile overflow  at Renick  Run was estimated  at  12  MG

by taking a simple proportion of the 80th percentile flow to the hydraulic

capacity of the pipes converging at each overflow.  Thus,  a total  volume of 62

million gallons was recommended  in the RFPU for  storage or treatment and the

cost associated with  control  at  this  level was estimated at $42 million.



3.3  BRIEFING PAPER ANALYSIS


     Lack of flow data does not  allow an independent comprehensive

analysis of the CSO problem.   Therefore,  the  following analysis  is  presented

only to provide data  for comparison with  the  figures in the RFPU.  The

combined sewer overflow volumes may be estimated using a procedure outlined  in
                                     E-13

-------
TABLE 1
           Plotting Position (P)


Total CSO
Rank (ra)
Event Date Volume (MG) Highest to lowest
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
10/8/77
10/26/77
8/6/78
8/19/78
8/29/78
8/30/78
9/16/78
8/4/82
8/25/82
9/14/82
44.1
3.1
26.1
10.4
6.1
73.3
16.2
44.9
51.9
12.8
4
10
5
8
9
1
6
3
2
7
m/(n+l)
n=10
0.364
0.909
0.455
0.727
0.818
0.091
0.545
0.273
0.182
0.636
Recurrence
Interval (yr)
    (1/P)
                                          ,75
                                          .10
                                          .20
                                          .38
                                          ,22
                                        10.99
                                          ,83
                                          ,66
                                          .49
                                         1.57
   E-14

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
reference 1.  This method first estimates the percent imperviousness of the

area by the following equation:
    Percent Imperviousness -1-9.6 PD<°-573 ~ °'0391 Io8 PD)


    Where PD = Population Density (Persons Per Acre)


     The population density for the combined sewer  area  for the city of

Columbus was cited as  15.78 persons per acre in reference 2.   While the
                                                                   o
combined sewer  service area is known  to have decreased from 18.4 mi   to the
                           *y
present estimate of 11.1  mi  (as  per city officials),  the population density

for this area may be assumed  to have remained about the  same.   Therefore, 16

persons per acre will be  assumed,  thus:



                     I =  9.6 (16)(0-573 ~ 0-0391 log 16)

                       =  41.3%


     Next the runoff coefficient  (CR)  weighted  between pervious and impervious

areas is estimated as follows:



                           CR - 0.15 + 0.75 (1/100)
                              - 0.15 + 0.75 (41.3/100) -  0.460


     The area weighted depression storage (DS)  is  then estimated assuming

0.0625 inches for impervious areas and 0.25 inch for pervious  areas.



                          DS - 0.25 - 0.1875 (1/100)
                             = 0.25 - 0.1875 (41.3/100) = 0.327 in.


     Finally,  the annual  runoff (AR) is  estimated  for  the CSO  area  in  terms of

inches per year over the  given area.



                         AR =  (CR) P-5.234 (DS)0-5957


     Where P » Annual precipitation, in/yr = 37.01 in/yr


               AR - (0.46X37.01  in. )-(5. 234X0. 327  in.0'5951)
                  - 14.33 inches  per year
                                     E-15

-------
                                                                          7
     Since the existing combined sewer service area is known  to be  11.1 mi ,
the estimated annual volume of runoff from this area  may be calculated as
follows:

11.1 mi2 x 14.33 in. x 1 ft. x 52802 ft.2 x 7.481 Gal = 2,760  x 106  gallons per year
                      12 in.     1 mT21 ft.
     A summary of  the 1985  and  1986  precipitation record for the city of
Columbus is provided in Table 2.  This table shows the  number  of days  in which
precipitation was  recorded  for each year broken down by depth.  In order to
relate the previously calculated annual volume  of  runoff  from  the  combined
sewer area to a rainfall day basis, an average value  for 1985  and  1986 of 58
significant days of rainfall may be assumed.  A "significant"  rainfall may be
defined as all days  when greater than 0.15 inches of rainfall were  recorded.
This number is reasonable  since  the depression  storage  for  the  combined sewer
area was  previously  calculated  at  0,327 inches.  Thus on a per-significant-
rainfall-day basis, a daily  volume  of runoff from  the combined  sewer area may
be calculated as follows:

             2,760 x 106 Gallons _•_ 58 Significant Rainfall Days
                          Year    *             Year
                        = 48 x 106 Gallons
                                   Signficant  Rainfall Day
     In addition to this flow, however,  is inflow from the separate  sewer area
which must be estimated.  Since the extent of the inflow problem in the
separate sewer area is unknown,  it will be assumed to be at the point of being
nonexcessive (refer to Section 1:  Terms and Definitions).    The  construction
grants program defines a nonexcessive inflow value of 275 gallons  per capita
per day (gpcd)  as  the maximum allowable  total  daily flow during a storm.
Thus, knowing  that  the average dry weather flow  is  167  gpcd  (ref. Briefing
Paper No.  1),  a  maximum allowable  inflow volume of 108 gpcd can be assumed for
this area.  The  population for the  separate sewer  area  may be  estimated by
subtracting the product of  the assumed combined sewer  population density (16
                                                 *\
persons/acre) by the combined sewer area (11.1  mi  =  7104 acres) from the
total service area population (870,000 persons).   Thus  the  population served
                                     E-16

-------
           TABLE 2.  PRECIPITATION
1985
Total
(Inches)
P
R
E
C
I
P
I
T
A
T
I
0
N

D
E
P
T
H

R
A
N
G
E









0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.00
.05
.10
.15
.20
.25
.30
.35
.40
.45
.50
.55
.60
.65
.70
.75
.80
.85
.90
.95
.00
.05
.10
.15
.20
.25
.30
.35
.40
.45
.50
.55
.60.
.65
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
to
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
.05
.10
.15
.20
.25
.30
.35
.40
.45
.50
.55
.60
.65
.70
.75
.80
.85
.90
.95
.00
.05
.10
.15
.20
.25
.30
.35
.40
.45
.50
.55
.60
.65
.70
Days
43
27
18
8
7
5
4
3
7
1
2
2
0
5
2
2
0
0
4
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
Depth
(In.)
0
1
2
1
1
1
1
1
2
0
1
1

3
1
1


3


1






1





.95
.82
.21
.33
.48
.34
.28 .
.13
.93
.47
.02
.13

.33
.42
.53


.69


.07






.41





1986
Total
Days
40
19
10
9
9
6
7
3
2
2
3
0
1
1
4
3
1
0
3
2
0
0
0
0
1
1
1
0
1
0
0
0
0
1
Depth
(In.)
0
1
1
1
1
1
2
1
0
0
1

0
0
2
2
0

2
1




1
1
1






1
.83
.24
.17
.48
.93
.65
.24
.1
.86
.95
.55

.63
.7
.89
.32
.82

.75
.93




.21
.29
.31






.69
ALL              143    31.92       130    35.04
                     E-17

-------
by the separate sewer area is about 756,336 persons.   Therefore,  the allowable
inflow from the separate sewer area may be estimated as:

                108 gpcd x 756,336 persons = 82 x 106 Gallons
                                                        Day
     The 108 gpcd  figure used above as the  inflow  from the  separate  sewer area
may be put into perspective in the following manner.  The daily volume
calculated above may be expected to occur on "significant"  rainfall  days,  as
previously defined.  The total "significant" rainfall average for 1985 and
1986 is shown to be 29.37 inches.   Dividing this  figure  by  the  58 significant
rainfall days per year  (used previously) results in an average of 0.51 inches
of rainfall per significant rainfall  day.   This rainfall depth of 0.51 inches
over the entire separate sewer area equates to a volume of 1,320 x 10
gallons.   Thus,  the calculated inflow  from the separate  sewer area using the
275 gpcd maximum  allowable daily  total  flow accounts for about  6.2 percent of
the total rainfall as inflow.

     Therefore,  the total combined  sewer and inflow volume  which must be dealt
with for each signficant rainfall  day  is estimated as follows:

                Combined sewer area runoff and inflow 47.7  MG
                Separate sewer area inflow            81.7  MG
                                              Total  129.4  MG

     Several points should  be  made  in  regard to this  figure.  While  the method
used for the calculation of the annual runoff volume from the combined sewer
area is a general one,  it certainly is what would be considered a "first order
approximation" using an EPA approved  procedure.  Secondly,  since the extent of
the inflow problem in the separate sewer area has not been investigated or
defined, the estimate used must be assumed to be reasonable.  It is  important
to note that this  brief  analysis shows that the inflow problem  from the
separate sewer  service  area on a volume basis  could be greater than the runoff
and inflow from the combined sewer area.  Further,  note  that for the inflow
from the separate sewer service area  to be  equal in volume  to the runoff and
                                     E-18

-------
  I
  I
  I
  I
  I
  I
  I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
inflow from  Che combined  sewer service area Chat only 3.6 percent of Che total

significant rainfall would have Co be accounted for as inflow.   It  must also

be noted that this simple volumetric analysis only considers average

conditions, i.e. 0.51 inches of  rainfall  per significant  rainfall  day.   Note

from the precipitation record of Table 2 that an average of 20 days each year

were recorded with precipitation greater  than this  amount with  a maximum daily

total of 1.41 inches for 1985 and 1.69 inches for 1986.   It  must also be

recognized that this analysis  is only volumetric and does not account for the

maximum  rate of runoff or rain-induced inflow.   This  maximum rate would be of

primary importance in the selection of control  alternatives  or design of

facilities.  This  parameter,  however, can only be determined through a

detailed hydraulic analysis of the sewerage system.
                                     E-19

-------
                                  REFERENCES
1.    Heaney,  J.  P.,  et. al. SCorm Water Management  Model:   Level 1 -
     Preliminary Screening Procedures.  USEPA Report No. EPA-600/2-76-275.
     NTIS No. PB 259 916.   October 1976.

2.    Heaney,  J.  P., et. al. Nationwide Evaluation of Combined Sewer  Overflows
     and Urban Stormwater Discharges, Volume II:  Cost  Assessment and Impacts.
     USEPA Report No. EPA-600/2-77-064.   NTIS No.  PB 266 005.   March 1977.
                                     E-20

-------

       APPENDIX F

  BRIEFING PAPER NO.  6
ONE-PLANT VS. TWO-PLANT

-------
I
I
I
•          BRIEFING PAPER NO.  6
I
           ONE-PLANT VERSUS TWO-PLANTS
I
•          Supplemental  Environmental Impact Statement
           USEPA Contract No.  68-04-5035,  D.O.  No.  40
I          Columbus Ohio Waste-water Treatment Facilities
I
I
I
I
I
•          Prepared By:
|          SCIENCE  APPLICATIONS INTERNATIONAL CORPORATION
|          TRIAD  ENGINEERING INCORPORATED
I
I

-------
                          ONE-PLANT VERSUS TWO-PLANTS
1.   EXISTING FACILITIES
     1.1  Jackson Pike Wastewater Treatment Plant
          1.1.1  Major Interceptors
          1.1.2  Preliminary Treatment (O.S.I.S.  Flow)
          1.L.3  Major Treatment Processes

     1.2  Southerly Wastewater Treatment Plant
          L.2.1  Major Interceptors
          1.2.2  Interconnector Pump Station
          1.2.3  Major Treatment Processes
2.   IDENTIFICATION OF SYSTEM ALTERNATIVES
     2.1  No Action Alternative
     2.2  Upgrade the Existing Facilities
     2.3  Eliminate Jackson Pike,  Upgrade and Expand Southerly
3.   DEVELOPMENT AND EVALUATION OF SYSTEM ALTERNATIVE COMPONENTS
     3.1  Interconnector/Headworks
          3.1.1  One-Plant System Alternative
          3.1.2  Two-Plant System Alternative

     3.2  Wet Stream Treatment
          3.2.1  One-Plant System Alternative
                 3.2.1.1  Primary Treatment
                 3.2.1.2  Secondary Treatment
                 3.2.1.3  Post Treatment
          3.2.2  Two-Plant System Alternative
                 3.2.2.1  Primary Treatment
                 3.2.2.2  Secondary Treatment
                 3.2.2.3  Post Treatment

     3.3  Solids Handling and Disposal
          3.3.1  One-Plant System Alternative
          3.3.2  Two-Plant System Alternative
4.   EVALUATION OF SYSTEM ALTERNATIVES

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
                                 INTRODUCTION



     Under the direction of USEPA,  a series of briefing papers are being

prepared addressing key issues in the development of the Supplemental

Environmental Impact Statement for the Columbus,  Ohio,  Wastewater Treatment

Facilities.   The  briefing  papers  form the  basis  of discussions between USEPA

and their consultant to resolve important issues.  The  following paragraphs

present the background of the facility planning process, a description of the

briefing papers,  and the purpose of this paper on one-plant versus two-plants.



FACILITY PLANNING PROCESS


     At the time this paper was  prepared (July-August 1987) the city of

Columbus was proceeding to implement improvements at the Jackson Pike and

Southerly Wastewater Treatment Plants to comply with more stringent effluent

standards which must be met by July 1,  1988.   These improvements were based

on the consolidation of wastewater treatment operations at the Southerly

plant.  This  one-plant  alternative  is  a  change from  the two-plant operation

proposed by the city in the  1970's and  evaluated in the 1979 EIS.



     The development and documentation of  wastewater treatment process and

sludge management alternatives for the Columbus metropolitan area has been  an

extended and iterative process.    The design and  construction  of various

system components have progressed, because of the 1988  deadline,  while

planning issues continue  to be resolved. As  a result,  numerous  documents have

been prepared which occasionally revise a previously established course of

direction.
                    \


     The concurrent resolution of planning issues and implementation of

various project components has made preparation  of  the  EIS  more difficult

because final facility plan recommendations are not available in a single

document.
                                     F-l

-------
BRIEFING PAPERS -
     To facilitate preparation of the EIS,  a series of briefing papers are
being developed.   The  purpose of the  briefing papers  is  to allow USEPA to
review the work of the EIS consultant and to identify supplemental information
necessary for the preparation of the  EIS.   Six briefing  papers are being
prepared as follows:

     •  Flows and Loads
     •  Sludge Management
     •  CSO
     •  Process Selection
     •  One Plant vs.  Two Plant (Alternative Analysis)
     •  O&M and Capital Costs

     The specific focus of each briefing paper will be different.  However,
the general scope  of the  papers  will  adhere to the  following format:

     •  Existing conditions will be documented.
     •  Evaluations, conclusions,  and  recommendations of the facilities
        planning process will be reviewed  using available documentation.
     •  Where appropriate, an independent evaluation of  the future situation
        and viable alternatives will  be prepared.
     •  The facility  plan and EIS briefing paper conclusions will be compared.

     The briefing paper process is intended  to:
                    \
     •  Prompt the resolution of any data deficiencies.
     •  Clearly establish and define  existing and future conditions.
     •  Identify the  final recommended plan which the city desires to implement.
     •  Provide a data base of sufficient detail to allow preparation of the
        draft EIS.
                                     F-2

-------
 I
 I
 I
 I
 I
 I
 II
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
ONE-PLANT VS. TWO-PLANTS


     This briefing paper evaluates the  comprehensive  wastewater management

alternatives in light of previous biological process, solids handling, and

cost analyses.   The briefing paper is divided into four sections as follows:



Section 1 - Existing Facilities


     This section discusses the facilities which existed at the Jackson Pike

and Southerly WWTPs prior to implementing construction  for  Project 88.



Section 2 - System Alternatives


     Section 2 provides a description of the one-plant and two-plant

alternatives.



Section 3 - Development and Evaluation of System Alternative Components


     Section 3  summarizes the facilities required for each process under the

one-plant and two-plant alternatives.  Costs are included for all facilities.



Section 4 - Evaluation of System Alternatives


     This section provides a technical  evaluation of  the one-plant and two-

plant alternatives based on present worth cost,  reliability, flexibility,

iraplementability, and operational ease.
                                     F-3

-------
                           1.   EXISTING FACILITIES

     This section describes  the Jackson  Pike  and  Southerly  Wastewater
                                                                                        I
                                                                                        I
                                                                                        I
Treatment Plants (WWTP).  Figure 1-1 shows the locations of the two treatment          I
plants and the Southwesterly Compost Facility within the planning area.
                                                                                       I
                                                                                       I
1.1  JACKSON PIKE WASTEWATER TREATMENT PLANT
     The Jackson Pike WWTP began operation in 1937.   The  plant  was modernized           •
and expanded in capacity  in the raid-fifties.  Currently  (prior to Project 88)           ™
there are two parallel flow trains for wet stream treatment consisting of
preaeration, primary settling, aeration, and final clarification.  The
original train is called Plant A and the newer train is  called Plant B.  The
two trains operate relatively independently of each other during liquid                 •
processing but share sludge handling  facilities.

1.1.1  Major Interceptors
     Wastewater arrives at the Jackson Pike plant via the 108-inch diameter             •
Olentangy-Scioto Interceptor  Sewer  (O.S.I.S.) and the 72-inch Big Run
Interceptor Sewer.   The maximum hydraulic  capability of  the plant  is 100 MGD.           •
Current average day flows are  approximately 84 MGD.   The  plant  accepts all the          ™
flow from  the  Big  Run Interceptor but limits its acceptance of  the  O.S.I.S.
flow so the hydraulic capability  of  the plant will not  be exceeded.   The major
diversion point for  the O.S.I.S. flows  is  at the  Whit tier Street  Storm Standby
Tanks.

     Seven miles of  150-inch and 156-inch diameter gravity sewer currently
                     \
exists between the Jackson Pike and Southerly treatment  plants.  It begins
3,000 feet  from the Jackson Pike WWTP and connects with  a pump station on the
west side of the Scioto River  near the Southerly WWTP.    In September of 1986,
USEPA provided funding for construction of the remaining 3000 feet of the
sewer (Figure 1-2).   This  will  complete  the  Interconnector  Sewer  between the
two plants.  Included in the  north end construction  will  be a diversion
chamber which  will connect the Interconnector Sewer with the O.S.I.S. north of
                                      F-4

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
JACKSON PIKE WWTP

SOUTHERLY WWTP
APPROXIMATE SCALE: 1  INCH - 4.12 MILES
SOUTHWESTERLY COMPOST FACILITY

PLANNING AREA  BOUNDARY  ... ,
                          FIGURE 1-1
                          PLANNING  AREA

-------
                                                    i    ,
                                                    \   \
                                                     V  -\
                                                     i   \
                                                      i »\
                                                      \ I \
                                                      \  o  \
                                                       .

                         JACKSON PIKE WASTEWATER TREATMENT PLANT
                          PROPOSED 150"

                          INTERCONNECTOR EXTENSION
                          & 8" SLUDGE LINE  EXTENSION
SOURCE:  REVISED FACILITY PLAN UPDATE
                              F-6
FIGURE 1-2

NORTH END  INTERCONNECTORl

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
I
Jackson Pike.  These improvements will  allow  Che  flow  to Jackson Pike to be

controlled by diverting excess flows  to Southerly.



1.1.2  Preliminary Treatment (O.S.I.S. Flow)


     Preliminary treatment  is provided  for flows entering Jackson Pike through

the O.S.I.S.  at a facility called  the  Sewer Maintenance Yard which is located

approximately one mile north of Jackson Pike.   These  preliminary treatment

facilities were  constructed in 1948.  They are rated at a capacity of 160 MGD

and provide preliminary  screening and grit removal for flows  in  the  O.S.I.S.

prior to their arrival at Jackson Pike.



1.1.3  Major Treatment Processes


     The Jackson Pike WWTP consists  of the following major treatment

processes:



     •  Preliminary Treatment

     •  Primary Treatment

     *  Secondary Treatment

     •  Disinfection

     •  Solids Handling

     •  Solids Disposal



     Figure 1-3 shows a flow schematic  of  the Jackson Pike WWTP.  Table 1-1

presents the equipment sizes and  the  capacities for each unit  process.



1.2  SOUTHERLY WASTEWATER TREATMENT PLANT


     The Southerly WWTP began operation  in 1967 with a  single  train.  In the

early seventies,  an  additional  wet stream train was added.  The original  train

is termed the Center Section.   The newer train is called the West Section.
                                      F-7

-------

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
H
r-l
o
o
z
H
CO
X
w
z
o
CO
^
o
r-J
«
<
H

















d
o
•H
4-1
• H
•a
d
O
o

CO
CU
•H
4-1
•rl
•H
CJ
CO



















CO
CO
CU
CJ
O
(X

co
C


u
to


d
CU
CU

o
co

4-1
to
CU


r.
en
60
a
• r-l
C

t-< CJ
CO
X! r-l
4-J CU
•rl CU
S rJ

























O
in
1— 1
s:
e
X-N
33 4J
Q M-I
H
in
4-1 •
M-I r-
CM
CM v-<
CO
v^ Q
CJ
rO 2£
f\
•
in r»
m CM

4-1 4J
CO CO

^0 *O
CU CU
CU CU
a, cu
CO CO
CU 4-1
r-l d
rO CO
CO 4J
•H to

CO O
^ o
0 0
3 .*
H H



















60
C
• r-l
Cu
B
3
(V,
^^
33
Q
H

4J
M-I

O
CO
s^x

o

•g

o


4J
CO

•a
CU
0)
cu
CO
u
c
CO
4J
tn
a
o
°
CU
d
o
\
























CU cu
B B
3 3
r-l r-l
0 O
> >
r-l r-l
CO CO
4-1 4-1
0 0
4J 4_l
m vo
C^ vO
• •
r-l O
CO CO
4-1 4-1
M-I M-I
m m
r-l I"H
X X

4-1 4-1
M-I M-I

vO vO
CM CM

X X

4J 4-1
M-I U-J

O CO
OO r-l
r-l r-l

4-1 4-1
CO CO

CO CO
r»j* ^/
G d
CO ca
4J 4J
CM CM

1 t

^J p-^

d C
a) co
r— 1 r-l
IX PH















C
0
• r4
4-1
CO
M
CU
CO
01























e

U
0
0
o

1— 1
CM
4-)
CO
**
T)
B
CO
C
O

4-1
CO
l^
0)
CO
0)
^4
P)

\4
o
M-I

CO
^i
CU
3t
O
1— 1
pq






















B S B

O O U
000
o o o
O O in

m co CM
r-l r-l
4-1 4-1 4-1
CO to CO
CM CM CO
















C
o

4-J
CO
M
CU
<£

CO CO
cu CU
rJ P
CO CO
CU CU
O 0
CO CO
M-i M-i
l-i M
3 3
co to
r-l r-l
CO CO
4-1 4-1
0 0
4-1 4-1
CO CO
0 0
0 0
o o
oo oo
g g
CO CO
4-1 U
M-I M-I
0 0
rH r-l
X X

4J 4J J2
MH MM CJ
CO
O O CU
OO 00
e
X X CU
60
4-1 4J
MM MM O
m
O O CM
in in
r-l r-l 4J
CO
4-1 4-1
CO CO CO
CU
co en S
X # 3
C C CU
CO to
4-J U (1)
60
*cf ^^* ^3
3
1 t r-l
to
^4 ro
CU
C C3 r-l
co ca cu
r-l i— 1 3
PU Pn H





C!
o
• r-l
4J
CO
O
• rl
M-I
• r4
rl
CO
r-l
O

^
U
CO
B
• rl
r4
CU O)
S B
3 3
r-l r-l
O O
i— 1 r-l
CO co
4J 4-J
0 0
4J 4J
O O
o m
r^ 0
CM r-l
CO Cfl
4J 4-1
m m
i— i i— i
X X

4-1 4J
MM M-I

vO vO
CM CM

X X

4-1 4-1
H _l ii»

o o
o o
ON ON

4-1 4-1
Cfl CO

co co
^ r^*
C C
cfl cfl

00  4-1
B co
cu
60 CU
d

60
CO
O
4-1
co
M-I
O
O
S
in
vO
o
o
d
•rl
en
cfl
-Q

CU
CU
CU
•O

4-1
O
O
M-I
1
OO

X

4J
O
O co
M-I 4-1
1 -r-l
•n^ C
rH 3
X^-
r"»
4J T3
o d
O cfl
MM 4-1
1 CO
00
0
(U 3
d 4J
O ^1




0)
60
•o
3
r-l
CO

•a
CU
4-1
CO
^
• rl
4-1
O
^4 60
d
CU -r*
4J -O
CO r-l
CO O
3 33
                                                          F-9

-------
f storage
o

o
f4.

r-l
U
• H
d
PU
a
o
0
»• r-4
• H tn
d TJ
3 -r-4
"" — r-*
a o
CX CO
00

O 4)
in 41
m [14
to
TJ
4>
0)
v
u
• r-l
^
H
r— 1
m
4-1 ^
0 I-i
H >% to
VJ TJ
CZ* CO d
cj a o
•H U
vO I-i 41
o 01 co
r-l
X S S
vO CO O
• •
0)
a
3
r-H
o
>
of storage
e>
52

0

r— 1
•H
C
3
"a1

00

O
o
CM
u
•tH
d
3
a, oo
BJ r*
I
O vO
O r-l
r— 1
s»* cu
-CO &
U CO
•r-i tn u
C T)
3 -H ^>.
•^. r— 1 
00 4-1
TJ ™
O 4) 3
O 4) CU
-H b. n
OO
CO 1
TJ vO
*••» r— 1
tn
d to
0 TJ
JJ *H
r-H
4-1 0
o> en
3
TJ
O Ol
P- 01
r-l Ct,
storage capacity 48,000 cy
d as needed
41
r-l d
tO tO
U 41
O -4
H O
ndfilled on an as-needed
through contract operation
tO
r-4 tO
• r-4
f-* (n
to to
< £>
ort 130-150 tons/day
ation 70-200 tons/day
iraate Unit Cost of
t ton
CX U X! 41
tO .rH O 3

0) CX CX r-H
M CX CX — I
H < < -st-
ation 260 days/yr
al peaks dependent
ther and cropping patterns
U d CO
•rH O 41
r-H W 3
CX CO
CX 41 d
< to o
                                                                                                                                                                                      I
con
CO
w
M
H
i—i
J
M
CJ
<
b

O
Z

H
CO
r-H
X
U

u
a
o
 i

w
en
C
o
 d
 o
CJ

 tn
 tu
 u
 CO
{Z4
        4J
        o
        o

        I
       in
       CM
 tO
•H
TJ

 u
 O
 O

 I
in
oo

 4)
 d
O
                             en
                             41
 3
 O


TJ
 O
 to

 O
 3
H
to
tn u
IH O
41 O
LJ '1 \
tn |
41 in
oo •
TJ CM
to w
a -4
•H 13
v^
CX U
o
4J O
XI *4-4
00 1
•H O
W r^
en
u
tn
CU
DO
•H
IH
CO
TJ
C
o
o
41
to

X
• rH
CO
to
o
o
i
m
CO
CM
CO

TJ

U
0
O
IIH
1
in
oo
CO
u
8
i
in
vn
CM
eo
u
O
o
MH
1
in
00

41
a
o
CM
ON
TJ
41
r-H
r-H
CO
4-1
to
C
•rH
en
0
u
O
tO
41
IH

O
3
H
en
Vj
O
u
o
41
O
u
00
r-H

TJ

                                                                                                                                                      •O  r-l
                                                                                                                                                       4)  X>
                                                                                                                                                       U   
                                                                                                                                                      ptf  
                              oo
                              3
                              C
                              0)
                             CJ
                                              c
                                              o
                                 01
                                 41
                                 00
                                 O
                                 • r-l
                                 JO
                                 O
                                 rJ
                                 0)
                                 0)
00
d
•H
TJ
,-H
O
m

0)
oo
TJ
3
r-l
co

TJ
41

en
60
•H
O



00
c

d
o
• rH
4J
•rH
TJ
d
0
0

r-4
cfl
e
(U
f-!
H


00
c

VJ
41
4J
CO
3
cu
O

cu
00
3
UH
•rH
4-1
d
4)
CJ
                                                                                          d
                                                                                          o
                                                                                  41
                                                                                  d
                                                                                 • rH
                                                                                  y
                                                                                  d
                                                                                 M
                                                                                               d
                                                                                               O
                                                                                               O
                                                                                               00
                                                                                                     x;
                                                                                                      CO
                                                                                                 •rH
                                                                                                 UH

                                                                                                 •g
                                                                                                  <0
                                                                                                 rJ
                                                                                                                           d
                                                                                                                           o
                                                                                                                           to
                                                                                                                           u
                                                                                                                                       CX
TJ
 d
 CO
                                                                                                                                               T)
                                                                                                                                                c
                                                                                                                                                to
                                                                                                                   I-l
                                                                                                                   o
                                                                                                                   cx
                                                                                                                   en
                                                                                                                   c   d
                                                                                                                   CO   O
                                                                                                                   IH  *H
                                                                                                                  H   4J
                                                                                                                       CO
                                                                                                                   41   u
                                                                                                                   00'H
                                                                                                                  TJ  -H
                                                                                                                   3   CX
                                                                                                                  r-H   CX
                                                                                                                  to  <;
                                                                                                                                                       CO
                                                                                                                                                       41
                                                                                                                                               CO

                                                                                                                                                C
                                                                                                                                                o
                                                                                                                                               •r-l
                                                                                                                                               4-1
                                                                                                                                                OJ
                                                                                                                                                U
                                                                                                                                                       CX
                                                                                                                                                       CX
                                                                              F-10

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
1.2.1  Major Interceptors

     Southerly receives approximately 50 to 60 MGD via the Big Walnut Sanitary
Outfall Sewer which serves the northeast, east, and southeast portions of
Columbus and Franklin County.  An additional 5 MGD of flow is carried to
Southerly by the Interconnector Sewer which serves  a  portion  of western
Columbus.  The Southerly WWTP only accepts the amount of flow that it can
successfully treat and bypasses the remaining flow.    Plant records  show that
bypassing occurs when treated flows  are as low as 54 MGD.  At other  times
treated flows can be as high as 90 MGD and no bypassing  is reported.  Excess
flow can be diverted to the Scioto River through a 108-inch diameter bypass
sewer at the plant's  influent regulator chamber.

1.2.2   Interconnector Pump Station
     The purpose of the Interconnector Pump Station is to pump flows from the
Interconnector across  the Scioto  River to the Southerly WWTP.  The Intercon-
nector Pump Station is located on the south end of the Interconnector near
Southerly (Figure 1-4).  Flows  from the  156-inch  Interconnector Sewer enter a
58-foot wide by 25-foot long by 16-foot deep chamber  to be distributed to
three channels containing coarse  bar  racks  and mechanically-cleaned  bar
screens.   Each channel  is  6 feet  wide by 30 feet long by 33 feet high.   Flows
from the screening channels enter a 20-foot wide by 66-foot long by  23-foot
high wet well and are  pumped by two 20 MGD and two 30 MGD extended shaft
centrifugal pumps through one 36-inch and  one 48-inch force main to  the
Southerly headworks.

1.2.3   Major Treatment Processes
     The Southerly WWTP consists  of the  following major treatment processes:

     •  Preliminary Treatment
     •  Primary Treatment
     •  Secondary Treatment
     •  Disinfection
                                     F-ll

-------
     •  Solids Handling
     •  Solids Disposal

     Figure 1-5 shows a flow schematic of the Southerly WWTP.   Table 1-2
provides sizings and capacities of individual unit treatment processes.
                                     F-12

-------
 I
 I
 I
 I
 I
 I
 i
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
I
I
 .
UJ
H-
2
UJ
o
                                               F-13

-------
CO
w
r-l
H
r-l
_)
rH
CJ

£

O
z
I—I
H
to
M
X
w

a,
>•

oc
u

H
O
O
to
CN
 I







d
o
•H
4J
•i-l
T)
ti
O
CJ

CO

0 0























0
S
0
r-
i— i
1*
O
4J H
CU
CU 4-*
u-4 cu
cu
OO U-l
•^ CM
-*
Q v_^
4J VO
cd
u
co to
CX
B co
3 CX
cx e
3
•a cx
CU
0) TJ
CX 0)
ai a)
ex
cu cn
•— i
XI 
l-i O
££















00
c
•I-l
cx
S
3
PVI
Q
H
4J

»—4
0)
H






















a) a>
e e
3 3
r-l i-4
O O
I— 1 r-l
CO co
4J 4-1
0 0
U 4-1
m oo
O in
r-4 i— 1
to fa
4J tO
U-l
4-1
m u-i
r-i
m
X r-l
" X
U-l
O U-l
o
O\ CO
X CM
4J X rC
U-l O
U CO
vO U-l 4)
CM
vO S
4J CM U^
CO CJ
4-1
co co O
X O
C CO O
cd ^1 •*
*J C 0
CO CM
*^ 4-1
U
1 vO cd

C 1 to
•H P
CO C 
M -H 3
H co O
P H Xi
CU
C in C
i-l
0 S Z














C
0
• r4
4J
cd
p
CU
<)
CO
CU
I-l
CO
CU
0
CO
U-4
I-l
3
CO
r-4
CO
4J
0
4J
to
o
CN
in
vO
CO
4J
U-l
m
•
CM
r-l
X
4J
U-l
o
r-.
i— 1
X
4-1
U-l
CX)

4-1
CO

in
^4
C
CQ
u

^°

1

c
•r4
CO
P
H
I-l
CU
C
CU
O
C
o
• r4
4-1
(0
CJ
• •-1
U-l
• H
CO
r-l
CJ

^
I-l
CO
•0
c
o
0
cu
CO
cfl
0)
CO
CU
CJ
CO
U-l
I-l
3
CO
r-4
cd
4J
O
4-1
CO
o
oo
00
r-
to
4-1
U-l
in
•
o
i— i
X
4-1
U-l
0
OO
r-4
X
u
U-l

CO
O >
0 -r4
O 4J
- CJ
C^l to
r-4 1
CU
CO 10
CO






















CO
o
4J
CO
C
• H
U
O
1— 1
n
CJ

^
CO
•a
•*--^
Xi
1— 1

o
o
_
CN
X
• r-l
to










d
o
• rH
4J
O
CU
U-l
d
• i-l
co
•r4
Q
cn
I-i
0
4-1
CO
I-l
O
CX
cfl
>
0)

^
CO
•o
--^,
Xi
I— 1

o
o
f^

OtT
X
•i-4
to






















to
4-1
U-l
r-
X
4J
U-l
o
CM
X
4-1
U-l
O
vO
CM
4-1
CO
C
•H
in
co
f>

4J
O
CO
4-1
C
0
o

CU
d
• •-4
V-l
O
^_l

0
cu
O























4J
• r4
d
3
^^
pLi
to

o
o
o\
•«
1— 1

<&>

in
4-1
• i-l
3

3
O
d
o
• H
U
CQ
4-1
O
rH
fe
M
•H
^

T)
CU
^
i— 1
O
CO
CO
•l-l
Q



y— \
in in
CO X:
C
-O cfl
D 4-1
co
3 d
0
OO -r^
l~ 4-1

H
                                                            F-14

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
                                   0
 d
 o
 CJ
CO

I—I
H
CJ

fa

O
55
M
H
CO
r-4
X
W

Oi
H
                       r-l CO
                    4-1     «£
                   •r-l  CO |3
                    d T3
                    3 -H T3
                   	r-l  CU
                    e  o  d
                    CX  co  cu
                    60     ^
                       •a  o
                   O  CU •f'
                   o  cu JT
                   CM Cu H
                 -I       >s
                 (d       M
                 4J   >%  Cfl
                 O   rH  T3
                 4J   CO   d
                     B   O
                 il  -H   O
                     U   CU
                     CX  ca

                    o  o
                    s  s
                CN  OO
                r--    •
                UJ
                 o

                 CU
                 e


4J
•r4
C
3

g
cx
60

0
o
CN

01
4J
•H CO
iw *O
3 -H
*"-, f— 4
0 0
CX co
60

O CU
O cu
r-4 fa
OO
1
vO
1— 1
*s.
CO
CJ

*G
CU
l-l
Q)
4-1
CO
3
CU
Q
CO
d
o
4-1
U
0)
o
o
fO
CJ
o
o


14-1
o

CU
B

r— 1
o

CU
'a
0
CU
CJ
d
CO
4J
CO

*rj

r-4
3
CO
23





CX
•H

4-1
•o
d
3
0

CO
T3
CO
d
o
4J

4-1
0)
3

0
0
CN
1
0
CN
r— I
sludge and weather

d
o

4j
d

*o
d
CO
CX
CU
-o
o
4J
CO
CO
I— 1
CO
ca
X
i—i
3
.43
•a
3
O
l-i
(-!
4-1

r-l
CO
CO
0
CX
CO
• H
a
ivate consumers
(-1
cx

•o
d
CO

CJ
• l-l
r— 1
ft
3
CX
day existing
— ^,
CO
d
o
4J

4-1
CU
3

0
m
r-l
3
cu
d
>-,
CO
•o

CO
d
o
4-1

4-1
CU
3

O
^O
CN
capacity of
CO
60
CO
l-i
O
4J
CO

1—4
CO
4J
0
H
eaned as needed
i— i
cj

• V.
^
U

0
o
o
•*
vO
r~-
d on an as-needed
CU
I— 1
1— 1
• f-l
t4-<
"O
d
cO
r— 1

J2
CO
<
ract operations
4J
d
O
CJ

^
r*

CO
• f-l
CO
CO
JZ
                                                                                           s
                                                                                           CU
TABLE 1-2. SOUTHERLY Wl
li ties /Condition
• r-l
CJ
CO
fa
solid bowl centrifuges
Project 88, Contract #19
yet fully operational
l-l 1
3 CU 4J
O t-> O
fa a, z;
primary digesters;
oot dia., 25.25-foot SWD
£j IM
3 1
o m
fa 00
secondary digesters;
oot dia., 25.25-foot SWD
truction date 1965
u-4 
O O T3
33d
H EH 3
CO
0
O
60
cO
r— 1

O
3
H
-owned landfill
r>\
LJ

O
            ca
            ca
            cu
            o
            O
                    60
                    d
                   •f-i
                    d
                    CU
 CU
 60

u-i
• H
 Vl /
                    CU
                   CJ
                d
                o
                                   CO
                                   CU
                                   60
rO
 o
 t-l
 CU
 CO
 d
                                        60
                                        d
                                                           d
                                                           o
                                                          •o
                                                           d
                                                           o
                                                          cj
                                        cu
                                       A
                                       H
                                                   60
                                                   d
                                    CO
                                    3
                                    CU
                                   o

                                    0)
                                    60
                                    3
                                                                      d
                                                                      cu


cu
60
•o
3
f— i
00

•o
0)
V4
CU
4-1
CO
3
cu
Q











CU
60
CO
jj
O
4-1
CO
osting
cx
B
0
CJ

0
4J

4J
U
O
CX
co
d
CO
U
H








60
d
•i-i
4-1
to
O
CX
e
o
CJ


1—4
CO
ca
O
CX
CO
•H
O

4-1
CO
O
CX

o
CJ






d
o
• H
4J
CO
l-l
CU
d
•H
0
d
M
 d
 o
 o
 60
 CO
_J
                                                                                                                                            •o
                                                                                                                                             d
                                                                            F-15

-------
                  2.   IDENTIFICATION OF SYSTEM ALTERNATIVES

     The current wastewater treatment facilities for the Columbus  metropolitan
area are the Jackson Pike and Southerly  Wastewater Treatment Plants (WWTP)
(See Figure 1-1.)  Upgrading and expansion of one  or  both  of these facilities
is required to meet federal effluent limitations.  Thus, the following three
wastewater system alternatives have been selected to be evaluated  for
preferred treatment.

     •  No action.
     •  Upgrade the existing facilities.
     •  Eliminate Jackson Pike, upgrade  and  expand Southerly.

     The following  sections discuss these three alternatives.

2.1  NO ACTION ALTERNATIVE
     The development of a no action alternative is consistent  with EPA
guidelines for preparing an EIS.   A no-action alternative  cannot be  eliminated
during a preliminary screening.  It must be  included in a  detailed evaluation
of alternatives. This  is  because  it serves  as  a baseline  when comparing  and
evaluating action alternatives.

     The no action alternative would involve normal maintenance but no
improvement to  the  existing facilities.   Failure to rehabilitate  and upgrade
the existing facilities will result in permit  violations.   This may result
in violations  of water  quality standards for receiving  waters  and  possible
public health problems in the Columbus metropolitan area.

     New NPDES permit limits have been established for  the Columbus  wastewater
treatment plants which they must  be in  compliance with by July 1, of 1988.  The
plants are currently operating under interim limits.  The Columbus  wastewater
treatment plants, without improvements,  cannot meet the new NPDES  permit
limits.  The new permits are more stringent  with respect  to CBOD^, TSS, and
                                     F-16

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
fecal coliform limits.  The permits also include a limit  for ammonia  and  a

minimum requirement for dissolved oxygen.  An inability to meet  permit  require-

ments may result in sanctions by OEPA and USEPA that could have  adverse social

and economic impacts in the facilities planning area.



2.2  UPGRADE THE EXISTING FACILITIES


     This alternative, which is consistent with current operation, was

evaluated by the city in the facility plan.  This alternative will be referred

to as the two-plant alternative.  In this alternative, the existing treatment

plant sites will be maintained.  Each plant will be rehabilitated and expanded

as necessary to provide advanced wastewater treatment on  site for wastewater

flows expected through the year 2008.  Due to site limitations and existing

hydraulic constraints at Jackson Pike, the city maintains that the wet stream

treatment capacity cannot be expanded.  However, the existing facilities  can

be upgraded to provide necessary treatment to meet proposed effluent  require-

ments.   Average flows in excess of 70 MGD and peak flows in excess of 100 MGD

at Jackson Pike would be diverted to Southerly via the Interconnector Sewer.

Figure 2-1 provides a flow schematic for the two-plant alternative.



2.3  ELIMINATE JACKSON PIKE, UPGRADE AND EXPAND SOUTHERLY


     This alternative was evaluated and recommended by the City  in the

facility plan.  Under this alternative, also called the one-plant alternative,

Jackson Pike would be phased out and all flows would be diverted to Southerly

via the Interconnector Sewer.  Expansion and rehabilitation of the existing

facilities at Southerly would be required.  Figure 2-2 provides  a flow

schematic for the one-plant alternative.
                                      F-17

-------
LJ
CK


^gi


O Z
LJ
  _ _
Q_ LuU_
                                                      u
                      IX) LA- Lu

                      ^  - .,
                                                    01=1
                                                    a a
                                                    co <=>


                                                    u  u
                                                  i—



                                                  a. L^LL.
                                                                                   cu °


                                                                                    ii ii
                                                                                 o_ ulu.
                                                                                   -> LJ
                                                                                    u
Z3 
-------
1
1

I

1

1


1
1



U
a:
MQ, CU
a:03*-*
$ " "
^^
1,1 ^ ^
S!u.c!
°>ui
->










1
1
a
IS
1
IS
2
***
2
1
1
1
1

1
1

<
^J
f ^ ^_Q ijl
•-• «jj (7*
£ II II
>• ^1 3
i^
r


\
i






zz
33
dM
- u ii
•"•* *B,^.
\ji-t t"'t t^^
°^<
>UJ

l— i
h-
<
-ZL
UIO
<|
bz y

















§o
i— i vD (T>
•
^ u a
u! a u.





















^ ^f NO
^C ^t* vD
02
H- II II
|g|
°13^
<£


11
^ II "
>_>>
^u.u.
d^

-------
       3.  DEVELOPMENT AND EVALUATION OF SYSTEM ALTERNATIVE COMPONENTS

     This section presents the  recommended process  components and the
facilities required to implement the one-plant and  two-plant system
alternatives.   The  components which will be discussed include the following:

     •  Interconnector/Headworks
     •  Wet Stream Treatment
     •  Solids Handling and Disposal

     The Interconnector component involves options for conveyance between the
two WWTPs.  Included in the headworks are the  coarse  bar  racks,  mechanically
cleaned bar screens,  aerated grit  chambers and pumps.  Wet  stream treatment
includes primary, secondary, and post treatment.   Solids  components include
thickening, processing, disposal,  and reuse processes.

     Secondary treatment and solids handling and disposal have been evaluated
in previous briefing papers. Therefore,  this  briefing paper will  summarize
the recommendations of  those papers.

     The Interconnector, headworks, primary treatment, and post treatment are
presented for the first time in this briefing paper.   An evaluation of
available options is contained  herein.    They  will be discussed in greater
detail than secondary treatment and  solids handling and disposal.

     Recommended  facility sizings  in this  paper are based on the flows and
loads developed in Briefing Paper No. 1.   Costs are consistent with those
costs presented in Briefing Paper No. 4.

3.1  INTERCONNECTOR/HEADWORKS
     The Interconnector and headworks alternatives are being discussed
together since they directly affect one another.
                                      F-20

-------
1
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     The 150-inch to 156-inch Interconnector Sewer runs in a north-south
direction between Jackson Pike and Southerly along the west side of  the Scioto
River.  The south end connects to the Interconnector Pump Station.   The
Interconnector Pump Station, with a firm capacity of 70 MGD, pumps the flow
across the Scioto River to Southerly through a 48-inch force main and a
36-inch force main.

     The north end of the Interconnector Sewer is incomplete.  However,
funding has been provided for its completion.  The remaining segment will be
constructed along the west and north side of Jackson Pike (Figure 1-2).  A
diversion chamber will be built connecting the Interconnector with the
O.S.I.S.  This will allow regulation of flows to Jackson Pike and diversion of
flows to Southerly.

     The existing Southerly headworks are rated at a capacity of 170 MGD.  The
headworks consist of coarse and fine screening, pumping, and aerated grit
removal.  The Jackson Pike headworks are rated at a capacity of 165  MGD.  They
consist of fine screening and pumping.  Preliminary treatment is provided for
flows entering Jackson Pike through the O.S.I.S. at the Sewer Maintenance
Yard.  These preliminary treatment facilities are rated at a capacity of
160 MGD and provide screening and grit removal for flows in the O.S.I.S. prior
to their arrival at Jackson Pike.

3.1.1  One-Plant System Alternative
     Under the one-plant system alternative, the Jackson Pike plant would be
phased out of service and all flows tributary to Jackson Pike would  be
conveyed to Southerly via the Interconnector Sewer.  In order to convey the
Jackson Pike flows to Southerly, the south end of the Interconnector and the
Southerly headworks capacity must be expanded.
                                     F-21

-------
     The Interconnected currently conveys approximately 5 MGD to Southerly
from a connection at Grove City.  Under the one-plant system alternative, it
would be required to convey an additional 132 MGD from Jackson Pike
(Figure 2-2).  This total flow exceeds the 70 MGD capacity of the existing
pump station and force mains.  Alternatives for expansion which were evaluated
by the city include the following:

     •  Option A - additional pumping facilities and force mains
     -»  Option B - extension of the 156-inch gravity Interconnector to
        Southerly
     Option A consists of increasing the current 70 MGD capacity to 150 MGD by
construction of a new pumping facility on the south side of the existing pump
station, and by constructing one new 48-inch and one new 36-inch force main
parallel to the existing force mains to the Southerly headworks.  The pump
station expansion will include the addition of three, 30 MGD submersible
centrifugal pumps and motors, three mechanical bar screens, and a screenings
conveyor system.

     Option B consists of extending the 156-inch Interconnector Sewer to the
Southerly WWTP.  Four 78-inch pipes would be used for the Scioto River
crossing to avoid the construction of a low head dam.

     Under the one-plant alternative, the existing Southerly headworks would
not be able to handle the combined peak flow of 231 MGD (i.e. 99 MGD from
Southerly and 132 MGD from Jackson Pike).

     The headworks options are affected by the Interconnector option selected.
The potential options available are:

     •  Option A-l - Expand existing headworks.
     •  Option B-l - Construct separate headworks for the Interconnector flows.
     •  Option B-2 - Construct new headworks for all flow.
                                        F-22

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 1
 I
 I
 I
 I
 I
 I
 I
I
     If Interconnector Alternative A is selected,  the flows from the Big

Walnut Interceptor and the Interconnector  would arrive at the plant at the

same elevation.  Therefore,  the existing headworks could be expanded to handle

all of the flow.   Expansion  would include  additional  pumps, screens, and grit

chambers.  This option will be known as Option A-l.



     If Interconnector Option B is  selected,  the gravity  sewer  will enter the

Southerly headworks approximately eight feet lower than the Big Walnut

Interceptor.   This results in the need  for separate headworks (Option B-l) for

the gravity Interconnector or completely new headworks  (Option B-2) to handle

the flows from both sewers.



     Option B-l consists of utilizing the  existing 170 MGD headworks at

Southerly for handling the flows  from the  Big Walnut  Interceptor and

constructing new  150 MGD  headworks  for handling the Interconnector flows.  The

new Interconnector headworks will be located adjacent to the existing

headworks.  They will include coarse bar racks, raw pumping,  followed by

mechanical screening and aerated grit removal; all designed for 150 MGD.

Mixing of the Interconnector and  Big Walnut flows  would follow aerated grit

removal.



     Option B-2 involves constructing completely new  headworks  which include a

mixing chamber, coarse bar racks, pumping, and aerated grit chambers.   The

flows from the Big Walnut Interceptor and  the Interconnector would combine in

a mixing chamber and be conveyed  through manually  cleaned bar racks.  The

combined flow will then enter a wet well to be pumped to mechanical bar

screens followed by aerated  grit  chambers.  The new headworks will be  designed

for a peak process flow of 231 MGD.  The combined  costs for the

Interconnector/headworks alternatives are  presented in Table 3-1.
                                     F-23

-------
               TABLE 3-1.  INTERCONNECTOR/HEADWORKS ALTERNATIVE
                                  PRESENT WORTH COSTS
Interconnector
Headworks
TOTAL
Option A/A-1
$14,058,000
$17,006,000
$31,064,000
Option B/B-1
 $4,432,000
$25,847,000
$30,279,000
Option B/B-2
 $4,432,000
$30,496,000
$34,928,000
     Option B/B-1 exhibics the lowest present worth  cost.  However, practically
speaking the present worth of A/A-1 is equal to B/B-1.  Reliability,  implemen-
tability, and ease of operation must also be considered when selecting the best
alternative.

     The gravity sewer options (B/B-1 and B/B-2) are more reliable than the
force main option (A/A-1) because there  is less  chance that the gravity sewer
will rupture.  Also,  gravity failure normally results in infiltration to the
conduit; while force mains exfiltrate to the environment.  In addition, the
gravity sewer does not rely on the operation of a pumping facility to function
properly.  Therefore,  it  would be easier to operate and maintain.  However,
separate headworks are needed for option B/B-1 which would require additional
operation and maintenance time.

     The force mains, on the other hand, may not require as deep of an
excavation as the gravity sewer; and therefore,  they would be easier to
implement.

     Based on the cost and reliability of Option B/B-1 (gravity), it is the
recommended Interconnector/headworks option  for  the  one-plant  alternative.

3.1.2  Two-Plant System Alternative
     The two-plant alternative does not require any  expansion of the
Interconnector or any additional headworks at the  Southerly WWTP.  New
headworks are required at the Jackson Pike WWTP.  The total present worth cost
of the headworks is  $14,170,000.
                                      F-24

-------
 I
 I
 I
 I
 1
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
3.2  WET STREAM TREATMENT


     Briefing Paper No. 3 - Process Selection presented a detailed evaluation

of secondary treatment alternatives and provided recommendations for secondary

treatment  facilities under each system  alternative.  In light of these

recommendations, this section will summarize the facilities required under

each system alternative for the following processes:



     •  Primary treatment

     •  Secondary treatment

     •  Post treatment



     Secondary treatment recommendations will be consistent with the

conclusion of the process selection breifing paper.   Primary treatment and

post treatment are being presented here for the  first time.



3.2.1  One-Plant System Alternative


     The one-plant alternative requires upgrading and expansion of the

Southerly plant to handle  ail  flows from the Jackson Pike and Southerly

service areas.   It was concluded in the process  selection briefing paper that

in addition to the two existing trains,  one  additional wet  stream treatment

train would be required at the Southerly WWTP.   Figure  2-2  shows how the flow

will be distributed between the three trains.



3.2.1.1  Primary Treatment


     The Southerly WWTP currently has primary treatment consisting of

preaeration and primary settling.   Preaeration of wastewater prior to primary

settling is done for odor control,  to prevent septicity,  and to improve

subsequent settling.   Little  or no BOD reduction occurs  in  the preaeration

tanks.   However,  preaeration  does  increase  the removal  of BOD and suspended

solids in the  primary tanks.   Primary settling should remove 25 Co 40 percent

of the influent BOD and 50 to  70 percent of the suspended solids.
                                     F-25

-------
     The Southerly WWTP currently has  four preaeration  tanks in each of the
Center and West Trains.  These preaeration tanks are adequate for providing
treatment for the flows in  these two  trains  under  the one-plant alternative.
However, an additional East Train is required under the one-plant alternative.
As presented in Figure 2-2,  this new East Train will provide treatment for an
average flow of 66 MGD and a peak process  flow of 99 MGD.   Assuming a
detention time of 30 minutes at average flow, four additional preaeration
tanks are required in the  new East Train.   These new  tanks are  the  same size
as the tanks in the  existing trains.

     The Southerly WWTP has  four primary  settling  tanks in each of the
existing Center and West Trains. These tanks have  adequate primary settling
capacity for the average and peak flows allocated  to these trains under the
one-plant alternative.   However,  additional  tanks are  required  for the new
East Train.   Assuming a surface loading rate of 1000 gallons per day per
square foot  at  average  flow  as  recommended by Ten States Standards,  66,000
square feet of surface area is required.   This surface area can be provided by
adding four new 150-foot diameter circular clarifiers.

3.2.1.2  Secondary Treatment
     The form of secondary treatment currently provided at the Southerly  WWTP
is conventional single-stage activated sludge.  This process includes
rectangular aeration tanks followed by rectangular secondary clarifiers.   The
plant was designed based on  NPDES permit limits of 30 rag/1  for CBOD^  and TSS.
The CBODc and TSS limits have become more stringent and an ammonia standard
has been added Co both permits.  As  a result of these changes,  the plants are
not capable of treating design flows to the more stringent permit limits.

     Through the course of the  facilities  planning  process for  the Columbus
wastewater  treatment facilities, other alternatives to the conventional
activated sludge process have been  proposed.   The 1979 EIS recommended a
trickling filter process followed by activated sludge for the Jackson Pike
plant.  The Facilities Plan Update  (FPU) and  Revised Facilities  Plan Update
                                     F-26

-------
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
recommended  a  semi-aerobic treatment process.  The Process Selection Briefing
Paper evaluated the semi-aerobic, trickling filter/activated sludge, and
single-stage activated sludge processes and recommended utilizing the semi-
aerobic process at both plants.

     The semi-aerobic process is a modified form of  the activated sludge
process.  The process consists of a non-aerated reaction zone ahead of an
aerated activated sludge zone.  The non-aerated zone may be anoxic  (nitrates
are present), anaerobic (no oxygen or nitrates  present),  or a  combination of
both.   The purpose of the anaerobic zone is to control bulking sludge.  The
anaerobic zone may change to anoxic depending on the level and concentration
of nitrates  in the wastewater.   In the  anoxic  zone denitrification  occurs.
Denitrification is a process by which nitrates are converted into nitrogen
gas.

     The only physical differences between the semi-aerobic process and the
conventional activated sludge process is an internal  mixed liquor recycle loop
and the addition of baffles to compartmentalize the  aeration tanks.  The
baffles are incorporated  into the design to prevent back-mixing from the
aerated zone to the anaerobic zone.   The internal  recycle loop is used to
bring nitrates back to the anoxic zone and thus cause denitrification to
occur.

     Under the one-plant  scenario,  the Southerly WWTP would be upgraded to
handle all flows from the Columbus service area.  The Southerly WWTP currently
has a West Train and a Center Train.   The West Train has six aeration  tanks
which are capable of treating an average design flow of 44 MGD.  The Center
Train has four aeration tanks which are capable of treating an average design
flow of 29 MGD.  These flows are based on the design  parameters of the semi-
aerobic process.

     The 2008 average design flow for the one-plant alternative is 154 MGD.
This will require an additional aeration basin capacity of 81 MGD.  This can
                                     F-27

-------
be provided by adding two tanks to the existing Center Train and by construct-
ing a new East Train consisting of nine aeration basins.  Figure 2-2 shows how
the flow is allocated to each train.

     The existing aeration basins will require some modifications to allow
them to be operated in the semi-aerobic mode.  Two baffles must be installed
in the first bay of each of the ten existing tanks and an internal mixed
liquor recycle loop must also be added to each tank.

     The existing rectangular clarifiers will be replaced by six new circular
clarifiers.  New circular clarifiers were recommended for the Southerly WWTP
due to the high mixed liquor concentration which must be maintained for
nitrification and the difficulty associated with settling a nitrified sludge.

     In addition to the six new secondary clarifiers for the existing Center
and West Trains, four new circular clarifiers are required for secondary
settling in the new East Train.

3.2.1.3  Post Treatment
     The current post treatment provided at the Southerly WWTP is
chlorination.  The Southerly WWTP has an earthen contact basin with internal
baffles.  This basin was designed as a temporary structure until a decision on
tertiary treatment could be finalized.  Since new regulations require
disinfection, Southerly needs permanent facilities.

     Southerly would require two new chlorine contact tanks sized at 81 feet
by 200 feet by 10 feet side water depth.  Dechlorination is also required  to
limit the chlorine residual in the effluent.  Post aeration will take place in
the final pass of the tanks to maintain a dissolved oxygen in the effluent of
7.0 mg/1.

     Table 3-4 summarizes the wet stream facilities required under the one-
plant alternative and the associated costs.
                                       F-28

-------
I
1
I
I
I
I
I
»


I

I

I

I

I

I

I

I

I

I
                                            TABLE  3-4
                                       WET  STREAM  TREATMENT
                                       (Southerly  One-Plant)
                                  Facilities  and Estimated Costs
          PREAERATION                                                       $ 5,905,000
               Eight existing tanks;  112.7  ft x  26  ft x  15.5  ft  SWD
               Four new tanks; 112.7  ft x 25.5 ft x 15.5  ft SWD
          PRIMARY SETTLING                                                   13,590,000
               Four existing tanks; 80  ft x  165  ft  x  10  ft  SWD
               Four existing tanks; 100  ft x 170  ft x 10  ft SWD
               Four new tanks; 150  ft dia. x 15  ft  SWD


          AERATION                                                           46,533,000
               Ten existing tanks; 26  ft x 900  ft x  15  ft  SWD
               Eleven new tanks; 26  ft x 900  ft x 15  ft SWD
          FINAL SETTLING                                                     35,462,000
               Demolish existing tanks
               Ten new tanks; 200 ft dia. x 15 ft SWD
          CHLORINATION/DECHLORINATION/POST AERATION                           3,000,000
               Two new tanks; 81 ft x 200 ft x 10 ft SWD
               including mixers, chlorinators, evaporators, and  sulfonators.
               Post Aeration takes place in the final pass of
               the chlorine contact tanks.
                                                     TOTAL CAPITAL COSTS   $104,490,000

                                                     ANNUAL O&M COSTS         5,224,000

                                                     TOTAL PRESENT WORTH   $144,504,000
                                                 F-29

-------
3.2.2  Two-Plant System Alternative
     The two plant alternative requires upgrading of both plants and minor
expansion of the Southerly plant.  No additional wet stream treatment trains
are required at either plant.  Flows are distributed to each of the plants as
shown in Figure 2-1.

3.2.2.1  Primary Treatment
     Under the two-plant alternative, the Southerly WWTP has adequate primary
settling and preaeration capacity.  However, upgrading of the existing
facilities is required.

     The Jackson Pike WWTP currently has two preaeration tanks in each of the
two trains, Plant A and Plant B.  The two tanks in Plant A provide 1.05 MG of
total volume.  The two tanks in Plant B provide 0.66 MG of total volume.
These tanks are capable of treating an average flow of 70 MGD.

     The Jackson Pike WWTP has four primary settling tanks in each existing
train, Plant A and Plant B.  These tanks are also adequate to treat an average
flow of 70 MGD.

3.2.2.2  Secondary Treatment
     The semi-aerobic process is recommended at both plants under the
two-plant alternative.

     Under the two-plant option the Southerly WWTP will be required to treat
an average flow of 84 MGD and a peak process flow of 131 MGD.  These flows
include 18 MGD under average conditions and 32 MGD under peak conditions being
diverted from Jackson Pike.  The Jackson Pike WWTP is limited to an average
flow of 70 MGD and a peak process flow of 100 MGD.
                                       F-30

-------
I
I
I
I
f
I
I
I
I
I
I
I
I
I
I
I
I
I
I
     In accordance with the evaluation presented  in  the Process  Selection
Briefing Paper, only two additional aeration basins  are required in  the  Center
Train at Southerly under the two-plant alternative.  Then  each train would
have six basins and could treat an average  flow of 42 MGD  and a  peak process
flow of 65.5 MGD (see Figure 2-1).

     The existing rectangular clarifiers should be demolished and  replaced
with six new circular clarifiers.

     The Jackson Pike WWTP is hydraulically limited  to a peak process flow  of
100 MGD.  Any peak flows in excess of this  flow would be diverted  to the
Southerly WWTP under a two-plant alternative.  An average  flow of  88 MGD was
projected for the 2008 design year.  However, in  evaluating  the  existing
facilities, the aeration facilities were found to be limited to  70 MGD average
flow.

     At an average flow of 70 MGD and a peak process flow  of 100 MGD,  the
existing aeration facilties at Jackson Pike have  adequate  capacity.   However,
extensive rehabilitation and the addition of baffles and an  internal mixed
liquor recycle system would be required to operate in the  semi-aerobic mode.

     The final clarifiers, on the other hand, are not sufficient to  treat a peak
process flow of 100 MGD.  Two additional rectangular clarifiers  would be necessary.

3.2.2.3  Post Treatment
     Under the two-plant alternative, the Jackson Pike and Southerly WWTPs
would require new chlorine contact tanks.  As discussed in the previous
section, Southerly has a temporary contact basin.  Jackson Pike  performs
disinfection by injection of  chlorine into the discharge  pipeline.   Under  the
two-plant alternative, Southerly would need two new  tanks  sized  at 150 feet by
64 feet by 10 feet side water depth.  Jackson Pike would need two  new tanks
sized at 100 feet by 70 feet by 10 feet side water depth.  Dechlorination
would also be employed.  Post aeration would take place in the final pass of
the tanks.
                                       F-31

-------
     Tables 3-5 and 3-6 present a summary of the required  wet stream treatment
facilities and the associated  costs  for  the  Southerly  WWTP and the Jackson
Pike WWTP, respectively.

3.3  SOLIDS HANDLING AND DISPOSAL
     This section summarizes the recommended solids handling and disposal
components for the one-plant system alternative and the  two-plant system
alternative.   These recommendations were  identified in the solids handling
briefing paper after a thorough evaluation of solids management options for
each plant.

3-3.1  One-Plant System Alternative
     The solids handling and disposal scheme identified for Southerly under
the one-plant system alternative is shown in Figure 3-1.  This handling and
disposal scheme includes the following sludge processes:

     •  Gravity thickening of primary sludge
     •  Centrifuge thickening of waste-activated sludge
     •  Thickened sludge storage and blending
     •  Stabilization by anaerobic digestion
     •  Centrifuge dewatering
     •  Composting
     •  Incineration
     •  Land Application

     Dewatered sludge would be disposed of as follows:

     •  25 percent would be composted and distributed as  a soil  conditioner.
        Sludge sent to compost would not go through the digestion process.
     •  25 percent would be land applied as a fertilizer to agricultural
        acreage within a reasonable distance from the plant.
     •  50 percent would be incinerated, and the ash product would  be  landfilled.
                                     F-32

-------
                                  TABLE 3-5
                             WET STREAM TREATMENT
                            (Southerly Two-Plant)
                        Facilities and Estimated Costs
PREAERATION                                                       $ 1,533,000
     Eight existing tanks; 112.7 ft x 26  ft x  15.5  ft  SWD


PRIMARY SETTLING                                                    4,717,000
     Four existing tanks; 80 ft x  165 ft  x 10  ft  SWD
     Four existing tanks; 100 ft x 170  ft x 10  ft SWD


AERATION                                                           12,284,000
    JL xvyii
     Ten existing tanks; 26 ft x 900 ft x  15  ft  SWD
     Two new tanks; 26 ft x 900 ft x 15 ft  SWD
FINAL SETTLING                                                     20,521,000
     Demolish existing tanks
     Six new tanks; 190 ft dia. x  15  ft  SWD
CHLORINATION/DECHLORINATION/POST AERATION                           1,800,000
     Two new tanks; 150 ft x 64 ft x 10  ft SWD
     including mixers, chlorinators, evaporators,  and  sulfonators.
     Post aeration takes place in the  final pass
     of the chlorine contact tanks.
                                           TOTAL  CAPITAL  COSTS    $40,855,000

                                           ANNUAL O&M  COSTS         2,382,000

                                           TOTAL  PRESENT  WORTH    $61,562,000
                                      F-33

-------
                                  TABLE 3-6
                             WET STREAM TREATMENT
                           (Jackson Pike Two-Plant)
                        Facilities and Estimated Costs
PREAERATION                                                       $  3,750,000
     Two existing tanks; 180 ft x 26 ft x 15 ft SWD
     Two existing tanks; 113 ft x 26 ft x 15 ft SWD
     Building renovation


PRIMARY SETTLING                                                    7,372,000
     Eight existing tanks; 150 ft x 80 ft x 10 ft SWD
     Control building renovation
AERATION                                                           22,502,000
     Twelve existing tanks; 900 ft x 26 ft x 15  ft SWD
     Control building renovation
FINAL SETTLING                                                      8,691,000
     Twelve existing tanks; 153 ft x 60 ft x  12.5  ft  SWD
     Two new tanks; 153 ft x 60 ft x 12.5 ft  SWD


CHLORINATION/DECHLORINATION/POST AERATION                           1,300,000
     Two new tanks; 100 ft x 70 ft x 10 ft SWD
     including mixers, chlorinators, evaporators,  and sulfonators.
     Post aeration takes place in the final pass of
     the chlorine contact  tanks.
                                           TOTAL  CAPITAL  COSTS    $43,615,000

                                           ANNUAL O&M  COSTS         2,648,000

                                           TOTAL  PRESENT  WORTH    $66,722,000
                                      F-34

-------
  I
  I
  I
  I
I

E

I

I
I

I

I
              UJ
              ZD
             ,-J
             :u.
             :2
                                                                                                           o
                                                                                                        U U
                                                                                                        ID I
                                                                                                        PI O
                                                                                                        ID (^
                                                                                                     >-    z
                                                                                                     _J«UJ
                                                                                                     QiLd 2:
                                                                                                     Ld C^ U
                                                                                                     ^a^3
                                                                                                     I- Q. <
                                                                                                     ID az
                                                    F-35

-------
     This alternative provides a great deal of flexibility for disposal.  It
offers continuation of the existing incineration and composting processes at
Southerly and introduces land application as a disposal process.   Table 3-7
presents the required sizing and  associated  costs  of the  sludge management
facilities for the one-plant system alternative.

3.3.2  Two Plant System Alternative
     The recommended  solids  handling and disposal scheme for Southerly under a
two-plant system alternative is the same as that for a one-plant  system
alternative.   This  scheme was previously described  in Figure 3-1.  Table 3-8
presents the sizing and  costs of  the  required  sludge management facilities for
Southerly under a two-plant system alternative.

     The recommended  solids handling and disposal scheme for Jackson Pike
under a two-plant  alternative  is presented  in  Figure  3-2.  This alternative
includes the following sludge processes:

     •  Gravity thickening of primary sludge
     •  Centrifuge thickening of waste-activated sludge
     •  Thickened sludge storage and blending
     •  Stabilization by anaerobic digestion
     •  Centrifuge dewatering
     •  Incineration
     •  Land Application

     Dewatered sludge would be disposed of as follows:

     •  50 percent would be incinerated and the ash product landfilled
     •  50 percent would be  land appplied as a fertilizer to agricultural
        acreage within a reasonable distance from the plant.
                                      F-36

-------
                                  TABLE 3-7
                    SOUTHERLY SLUDGE MANAGEMENT COMPONENTS
                         One-Plant System Alternative
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping              $5,070,000
     Four (4) existing; 45-foot dia. x 17-foot SWD
     Two (2) new; 85-foot dia. x 10-foot SWD

Centrifuge Thickening WAS                                      5,600,000
     Four (4) existing; 250 gpm, 1250 Ib/hr
     Four (4) new; 250 gpra, 1250 Ib/hr

Thickened Sludge Storage/Blend                                    —
     Existing Facilities Reused

Anaerobic Digestion                                           11,460,000
     Six (6) existing; 85-foot dia. x 25.25-foot SWD
     Four (4) new; 85-foot dia. x 25.25-foot SWD

Centrifuge Dewatering                                         21,040,000
     Six (6) existing; 1000 Ib/hr
     Nine (9) new; 1000 Ib/hr

Dewatered Sludge Storage
     One (I) new; 400 cy plus material handling                1,300,000

Composting
     Existing Facilities; 120 wet ton/day @ 20% solids

Incineration
     Two (2) new; 8 hearth, 260 wet ton/day @ 20% solids
     Rehabilitate existing                                     1,300,000

Landfill
     Contract operations included with O&M                        —
                    \
Land Application                                                  —
     Contract operations included with O&M

                         Capital Cost                        $45,770,000

                         Annual Operation and                  6,230,000
                           Maintenance Cost

                         Present Worth (One-Plant)           $90,710,000
                                      F-37

-------
                                  TABLE 3-8
                    SOUTHERLY SLUDGE MANAGEMENT COMPONENTS
                         Two-Plant System Alternative
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping               $2,520,000
     Four (4) existing; 45-foot dia. x 17-foot SWD

Centrifuge Thickening WAS                                       2,000,000
     Four (4) existing; 250 gpm, 1250 Ib/hr
     One (1) new; 250 gpra, 1250 Ib/hr

Thickened Sludge Storage/Blend
     Existing Facilities Reused

Anaerobic Digestion                                             4,280,000
     Six (6) existing; 85-foot dia. x 25.25-foot SWD

Centrifuge Dewatering                                           5,120,000
     Six (6) existing; 1000 Ib/hr
     Two (2) new; 1000 Ib/hr

Dewatered Sludge Storage                                        1,300,000
     One (1) new; 400 cy plus material handling

Composting
     Existing Facilities; 120 wet ton/day @ 20% solids

Incineration
     Two (2) new; 8 hearth, 260 wet ton/day @ 20% solids

Landfill
     Contract operations included with O&M

Land Application                                                  —
     Contract operations included with O&M
                    \
                         Capital Cost                        $15,220,000

                         Annual Operation and                  3,340,000
                           Maintenance Cost

                         Present Worth (Two-Plant)           $39,680,000
                                      F-38

-------
                               LJ
                               O
LJ
LJ
[NTRIFUGE
KCKENING
^S
{ICKENED
.UDGE
.END/STDR
     LJLJ
     LDX
     PIO

  u^,"
  CL   Z
(U    O LJ

ri»ZUj:E;
w a co LJ
     a LD
     a. <
                                                                           u
                                                                                0.
                                                                           -JO
                                                                             »—«

                                                                           P&
                                   F-39

-------
     The difference between this alternative and the alternative  recommended
for Southerly is that the Jackson Pike alternative does not include
composting.  The recommended 50:50 disposal ratio between land application and
incineration is approximately consistent with current Jackson Pike disposal
practices.   Table 3-9 provides a list of the required facilities  for Jackson
Pike and their associated costs.
                                      F-40

-------
                                  TABLE 3-9
                   JACKSON PIKE SLUDGE MANAGEMENT COMPONENTS
                         Two-Plant System Alternative
                        Facilities and Estimated Costs
Gravity Thickening PS plus Dilution Water Pumping             $1,967,000
     Modify two (2) digesters; 85-foot dia. x 10-foot SWD

Centrifuge Thickening WAS                                       4,500,000
     Two (2) existing; 500 gpm
     One (1) new; 500 gpm

Thickened Sludge Storage/Blend
     Existing Facilities Reused

Anaerobic Digestion                                             9,170,000
     Six (6) existing; 85-foot dia. x 23.5-foot SWD

Centrifuge Dewatering                                             490,000
     Six (6) existing; 1200 Ib/hr

Incineration
     Two (2) existing, 7 hearth, 200 wet ton/day @ 20% solids   3,600,000

Landfill
     Contract operations included with O&M

Land Application
     Contract operations included with O&M                         —

                         Capital Cost                         $19,727,000

                         Annual Operation and                   3,070,000
                           Maintenance Cost

                         Present Worth (Two-Plant)            $45,827,000
                                      F-41

-------
                    4.   EVALUATION OF SYSTEM ALTERNATIVES

     This section evaluates the one-plant and two-plant system alternatives
based on cost, reliability, fexibility, implementability, and operational
ease.

     Table 4-1 presents the capital, annual O&M, and total present worth costs
for the one-plant and two-plant system alternatives.

                     TABLE 4-1.  SYSTEM ALTERNATIVE COSTS
                                                                 Total
                                   Capital       Annual O&M   Present Worth
One-Plant [Southerly]              268,711,000   16,849,000   436,911,000
Two-Plant [So. and Jackson Pike]   207,076,000   19,078,000   397,016,000
Difference from One-Plant          -61,635,000   +2,229,000   -39,895,000
Percent Difference                 -30           +13          -10

     Details on the development of the costs in Table 4-1 are presented in
Briefing Paper No. 4 - Capital and O&M Costs.

     The two-plant system alternative exhibits a total present worth cost
approximately 10 percent lower than the one-plant alternative.

     Both the one-plant and two-plant alternatives are equal with respect to
their reliability in meeting final effluent limits.  However, the two-plant
would be more reliable with respect to shock loads.  Under the one-plant
alternative, a plant upset at  Southerly could result in a significant loss of
biological treatment capacity  and may cause a serious water quality problem.
However, if the shock and/or toxic load can only reach one of the two plants,
the impact may not be as severe.

     The two-plant alternative is judged more flexible than the one-plant
alternative.  With both facilities operational, the city would have more
flexibility to adapt to increased future flow, to meet more stringent effluent
limits, and to address combined sewer overflows.  The two-plant alternative
                                      F-42

-------
would leave more land available at Southerly for expansion.  The two-plant
alternative would improve and upgrade Jackson Pike to provide a solid  100 MGD
treatment capacity.  The two-plant alternative would allow for future
expansion of the Interconnector system to divert more flow to Southerly while
optimizing the use of the Jackson Pike facility.

     The two-plant alternative would be easier to implement since the majority
of the facilities already exist.  Most of the construction would consist of
rehabilitation of existing facilities.  No expansion of the conveyance system
between the plants is required under this alternative.

     The one-plant alternative would be easier to operate and maintain since
all facilities would be consolidated at one location.

     A recommendation on a system alternative cannot be made based solely on
this technical evaluation.  Environmental impacts must be considered prior to
making a recommendation.
                                      F-43

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
         APPENDIX G


   GRAPHS OF STORE! DATA

  FOR DO, BOD, AND AMMONIA
   FROM 1971-1986 AT SIX
STATIONS ON THE SCIOTO RIVER

 BETWEEN JACKSON PIKE WWTP
   AND CIRCLEVILLE, OHIO

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
5
«?
i
  5
(UU.O
   t
        §
        tu
        f
        in
          N
          «

                oni
                xr-
                   |
           «v
           atf-
           «M •
           or-
          a.«
          or- •
       ju>  ^n
^ .222«  88
  e  ocoe  xu>
1/>I-O>  KM  uiul
toevooz  o_>
                                                                                             u>
                                                                                             «
                                                                                             n
                                                                                             ao
                                                                                             at
                                                                                             00
                                                                                             at
                                                                                             oa    i
                                                                                               *
                                                                                               at
                                                                                              e-
                                                                                              at
                                                                                               01
                                                                                               r-
                                                                                               o»
                                                G-l

-------
j
          UJ


          P
          in
    &     1
    ••«     X

    £     I
uiot
f--»
^

f*
UJOO


25
Si
                  zta

                  c*
                  iv
                   ni
                  U)

                  3S
  CUU.J


  I
      J*    «*T  oca
      «-<     O»P-  «o
             8- •  a.*
             ™^*
00
ot
                                                                                             «
                                                                                              M
                                                                                              00   QA


                                                                                              S   2
                                                                                              
    o-»oo
                                                  G-2

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                                                                                   Is
                                                                                     r—
                                                                                   §-
*
             uir-
             i-nt

   r-
   ni
 Js     aF
 nw      ^ Y
 t5    . 23

 2°    * 2*

 •  ooc3* niut
 ^:_:E!t'r!  ••
                                                                         00
                                                                         O)
                                                                         ni
                                                                         r-
                                                                         o>
 UkOTOO
                                       G-3

-------
Jr

I
       «

       £
       VI


*      I
Y      «
(U      N

5      I
                  £8
                  zr->
                  min
                  2?
                    o
                    £
            g?;«
  Wtorf     ^

  SS     § 28
  »*-    «s 5 .

  • OOC^* OJlft


    OttOO X^»
  Wf-0*  K*"* UJU
  U»OTOOX SrS
                                                                                              U)
                                                                                              aa
                                                                                               r-    ^
                                                                                               a»    r^
                                                                                               -•    o»
                                                                                                            o
                                                                                                            00
                                                                                                            oo
                                                                                                            o ^r
                                                                                                            o  •
                                                                                                            vo 
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
,?
i
                woo
                u>
                tacn
                000
                z
         X
   I     >

   •23   ,
                  i
                                                                                                    o
                                                                                                    CO
                                                                                                    CO
                                                                                                    o -

   O  oeoo  xui
      ox
  88RSSS  ££
                                               G-5

-------
u>
m


i
                   WP-
                   t-(\l
                    Sv
                    in
               £5
                 •
    4C    UIU>  •*<*)
  • ooc5«  CUM
   > .i-iuii-i   «oi
                                                                                                                O
                                                                                                                co
                                                                                                                cx>
                                                                                                                o -
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
    s
    o»
  c
            
   *  •*MU|K«*    901
    o  OEOO  xin
    K ok  ^*^  uiuj
                                                                                                                             CO
                                                                                                                             O
                                                                                                                             O
                                                                                                       C i-H
                                                                                                       O
                                                                                                       •<-t z:
                                                                                                       •w OS
                                                                                                       a
                                                                                                       ifj
                                                                                                       CO
                                                                                                        en
                                                                                                        oo
                                                                                                        a>

                                                                                                         at
                                                                                                         r-
                                                                                                         i>-
                                                                                                         5
                                                                                                         in
                                                                                                         (^
                                                                                                         o»
                                                                                        ni
                                                                                        &
                                                           G-7

-------

 i
 in
 ia


ni i o
   i




i!
       *•
               £2
               uiw
               «
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
I
i
Jr
t-

        P
        VI
        1

        I
              a x
              58
Si
              28
          o»o»  «*
          — •  o.n
                                                                                  oo
                                                                                  O
                                                                                  O
                                                                                   « pa
                                                                                   ffl
                                                                        s
                                                                         ni
                                                                         09
                                                                         o>
  • o«5*
      biin —P
      5« niu
  "2$S2i Sri
  R8R88S ~*
                                       G-9

-------
J
           e
           
                                                                                                         5
                                                                                                                    o
                                                                                                                    o
                                                                                                                    ON

                                                                                                                    O
                                                                                                                    ^O

                                                                                                                     C
                                                                                                                     o
                                                                                                                       C/3
                                                                                                                         "
I
I
I
I
    O  OCOO  XJ>
                                                          G-10

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
                SI

                XOB
                « •
                cu>
            f-
                                                                                                «"    O«
                                                                                                ni
                                                                                                c^
                                                                                                a
!«
     otoo  x«>
                                                  G-ll

-------
*
                IS
 CM

 JS


 at

 in

  i >-<


 &IS
 MWUt

«u£2s
   &.
         i

  §
 ..-  II
   o»m  «»
   — •  a.n
   nu>    x
   «««^    X
 X x
 t-      •
 a.«r>    »
 ujnia    •»

r  S3
                                                                                      oo
                                                                                      Ol
                                                                                           U3
                                                                                           00
                                                                                           at
                                                                                      u>
                                                                                      r-
                                                                                      01
                                                                                      S!
                                                                                      01
        toe ,  _
        Ulljl  +*
   w  Ott=>*  (U
   • «*MyjH4   9




   pg«S22  2
          «
                                                                                                  O

                                                                                                  O
                                                                                                   c
                                                                                                   o
                                                                                                       I

                                                                                                       I

                                                                                                       I

                                                                                                       I
                                              G-12

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
i
i
                    zu>
                    uieo
Si
 I •
« X
U|f*1
a»r-
                      S
                      a.
    u>
     I > "

    si§
               <7>o  «ca
               ~>  •  0.
             £  -    o,
             o.«n    CM
             uinin    ni
             oc-  •
           -  ii
  •tt    at*  ut^
  •  OQC.?4I  IMU)
   . .>.U|m    »0»
                                                                                                                          13 5
                                                                                                           IB
                                                                                                           8
                                                                                                            ni
                                                                                                            to
                                                                                                            (71
                                                                                                            oo
                                                                                                            r>-
                                                                                                            (71
                                                                                                            P-
                                                                                                            (71
                                                          G-13

-------
I
I
                  urn

                    ~
     Z1A
     woa
                  « s,
                  Q^.

                   I •
                  OS.
                   ®r-  gco
 Ac^  c o    ^*
 -• •  a.m     o
           «  IS
                                                                                                       «
                                                                                                  ni
                                                                                                  s
    at    ac«  **
    Z    Ullfl  -«D
  •  eat5«  niut
   • •«-"«4i<-i    oat
  •aexSac    -
  M  5—I

                                                                                                                       I
                                                                                                                       I
                                                     G-14

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
               a*-*
        Or
        t-
        
 •  >MUJH   i»(7l
•acz5ac   ~
*4  O*^
  e  acoo XIA
  tM*4t^««l     «
R88I8S  "
                                                                                       at
                                                                                       (71
                                                                                       ru
                                                                                       oo
                                                                                       (71
                                                                                   01
                                                                                   i^
                                                                                   a
                                                                                   r-
                                                                                   r>-
                                                                                   01
                                                                                       ni
                                                                                       r-
                                                                                       o»
                                             G-15

-------
J
Jf
         s
         UJ
         £
   o    
                                                                                         00
                                                                                         r-
                                                                                         o>
                                                                                                       CM
                                                                                                       O
                   I
                   I
                   I
                   I
                                                                                                     C/D
    e  otoo  xu>

  Donjooi  ^j
    «M*4^«*^O  C»4
                                                 G-16

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
j
j?
    k«ee xi/>

 R88583 52
                                                                         ^CM
                                                                         §2
                                  G-17

-------
jr
             oa
             si
 S
       >
          82  I
•a
t*2    -  ZS
u»o    •
                                                                               eg
                                                                               15
                                                                               o»
                                                                                 ^•
                                                                                 o»
                                                                                            CM
                                                                                            O
                                                                                                  I
                                                                                                  I
                                                                                                  I
                                                                                                  I
    ocoo  xvi
                                          G-18

-------
         APPENDIX   H
TABLES OF ENDANGERED SPECIES

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE H-l.  ENDANGERED FAUNA SPECIES KNOWN TO OCCUR IN THE
         COLUMBUS FACILITIES PLANNING AREA, OHIO8
Species
Indiana bat
(Myotis sodalis)
Peregrin falcon
(Falco peregrinus)
Bald eagle (Haliaeetus
leucocephalus)
Kirtland's warbler
(Dendroica kirtlandii)
Upland sandpiper
(Bartramia longicauda)
Common tern (Sterna
Hirundo)
Four-toed salamander
(Hemidactyliumd scutatum)

Northern brook lamprey
(Icthyomyzon fossor)
Paddlefish (Polyodon
spathula)
Blacknose shiner
(Notropis heterolepis)
River redhorse
(Moxostoma carinatum)
State Federally
Endangered Endangered
X X
X X
X X
X X
X
X
X
X
X
X
X
Remarks
Habitat requirements are
not fully known.
Occurs as an uncommon
migrant .
Occurs as an uncommon
migrant .
Occurs as an uncommon
migrant .
May occur in suitable,
grassy habitat anywhere in
the country. Recent
records exist for Bolton
Field and Rickenbacker Air
Base.
Occurs as an uncommon
migrant .
Requires a bog-like
habitat. A recent record
exists for the northeastern
corner of the country.
Rare occurrence in Big
Walnut Creek and Big Run
(tributary of Olentangy
River) .
One specimen observed in
Scioto River below
Greenlawn Dam in 1976.
Population in Rocky Fork
Creek (tributary of Big
Walnut Creek, northeast
Franklin County).
Known population in Scioto
River and tributaries.
                           H-l

-------
          TABLE H-l.   ENDANGERED  FAUNA  SPECIES  KNOWN  TO  OCCUR IN THE
             COLUMBUS  FACILITIES  PLANNING  AREA,  OHIO3  (Continued)
                                                                                       I
                                                                                       I
        Species
          State     Federally
       Endangered  Endangered
Remarks
Slenderhead darter
(Percina phoxocephala)
Spotted darter
(Etheostoma Maculatum)
Lake Chubsucker
(Erimyzon succtta)

Shortnose gar
(Lepisosteus platostomius)

Mooneye
(Hiodon tergisus)0

Tippecanoe darter
(Ethestoma tippecanoe)
Scioto madtom
(Noturus trautmani)
d , e
Piping plover
(charadrius melodus)
                                Known population in Big
                                Walnut and Big Darby
                                Creeks.

                                Small population in
                                Olentangy River and Big
                                Walnut Creeks.
                                Collected just downstream
                                of FPA at Circleville.

                                Found only in Big Darby
                                Last seen at the Jackson
                                Pike Wastewater Treatment
                                plant in the 1940's.
'Source:  Ohio Department of Natural Resources 1986, unless otherwise noted.

bSource:  OEPA 1986a.

GSource:  Yoder 1987; Ohio Department of Natural Resources 1986.

 Source:  Cavender 1986.

"Source:  Multerer 1986.

£Source:  Huff 1988
                                      H-2

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
   TABLE H-2.   LIST OF STATE AND FEDERALLY ENDANGERED PLANT SPECIES IN OHIO
Selaginella rupestris, Rock Spikemoss
Isoetes engelmannii, Appalachian Quillwort
Botrychium lanceolatum, Triangle Grape-fern
Ophioglossum engelmannii, Limestone Adder's-tongue
Trichomanes bpschianum, Appalachian Filmy Fern
Polypodium polypodioides, Little Gray Polypody
Dryopteris clintoniana, (D. cristata var. clintoniana) Clinton's Wood Fern
Sparganium androcladum, Keeled Bur-reed
Sparganium chlprocarpum, Small Bur-reed
Potamogeton filiformis, Filiform Pondweed
Potamogeton gramineus, Grass-like Pondweed
Potamogeton hillii, Hill's Pondweed
Potamogeton praelongus, White-stem Pondweed
Potamogeton robbinsii, Robbin's Pondweed
Potamogeton tennesseensis, Tennessee Pondweed
Scheuchzerialialustris, Scheuchzeria
Sagittaria graminea, Grass-leaf Arrowhead
Cinna latifolia, Northern Wood-reed
Danthonia compressa, Flattened Wild Oat Grass
Digitaria filiformis, Slender Finger-grass
Glyceria acutiflora, Sharp-glumed Manna-grass
Koeleria macrantha (K. cristata), Junegrass
Melica nitens, Three-flowered Melic
Muhlenbergia cuspidata, Plains Muhlenbergia
Oryzopsis asperifolia, Large-leaved Mountain-rice
Panicum bicknellii, Bicknell's Panic-grass
Panicum boreale, Northern Panic-grass
Panicum leibergii, Leiberg's Panic-grass
Panicum villosissimum, Villous Panic-grass
Panicum yadkinense, Spotted Panic-grass
Poa wolfii, Wolf's Bluegrass
Schizachne purpurascens, False Melic
Carex aquatilis, Leafy Tussock Sedge
Carex arctata, Drooping Wood Sedge
Carex argyrantha, Silvery Sedge
Carex atherodesT Wheat Sedge
Carex bebbii, Bebb's Sedge
Carex cryptolepis (C. flava var. fertilis), Little Yellow Sedge
Carex debilis var. debilis, Weak Sedge
Carex decomposita, Cypress-knee Sedge
Carex folliculata, Long Sedge
Carex garberi, Garber's Sedge
Carex gravida, Heavy Sedge
Carex haydenii, Hayden's Sedge
Carex louisianica, Louisiana Sedge
Carex nigromarginata, Black-margined Sedge
Carex ormpstachya, Stiff Broad-leaved Sedge
Carex pallescens, Pale Sedge
Carex sprengelTI, Sprengel's Sedge
Carex striatuIaT Lined Sedge
Cyperus acuminatus, Pale Umbrella-sedge
                                      H-3

-------
   TABLE H-2.   LIST OF STATE AND FEDERALLY ENDANGERED PLANT SPECIES IN OHIO
                                  (Continued)


Cyperus dipsaciformis, Teasel-sedge
Rhynchospora globularis, Grass-like Beak-rush
Scirpus expansus, Woodland Bulrush
Scirpus smithii, Smith's Bulrush
Scirpus subterminalis, Swaying Rush
Wolffiella floridanaT Wolffiella
Juncus interior, Inland Rush
Clintonia borealis, Bluehead-lily
Lilium philadelphicum, Wood-lily
Melanthium virginicum, Bunchflower
Nothoscordum bivalve, False Garlic
Smilax pulverulenta, Downy Carrion-flower
Streptopus roseus, Rose Twisted-stalk
Iris brevTcaulis, Leafy Blue Flag
Iris verna, Dwarf Iris
Sisyrinchium atlanticum, Atlantic Blue-Eyed-grass
Sisyrinchium montanum, Northern Blue-eyed-grass
Arethusa bulbosa, Dragon's-mouth
Coeloglossum viride (Habenaria viridis), Long-bracted Orchid
Corallorhiza trifida, Early Coral-root
Corallorhiza wisteriana, Spring Coral-root
Cypripedium calceolus var. parviflorum, Small Yellow Lady's-slipper
Cypripedium candidum, White Lady's-slipper
Hexalectris spicata, Crested Coral-root
Platanthera blephariglottis (Habenaria blephariglottis)> White Fringed Orchid
Spiranthes romanzoffiana,Hooded Ladies'-tresses
Populus balsamifera, Balsam Poplar
Populus heterophylla, Swamp Cottonwood
Salix caroliniana, Carolina Willow
Salix pedicellaris, Bog Willow
Myrica pensylvanica, Bayberry
Ulmus thomasii, Rock Elm
Urtica chamaedryoides, Spring Nettle
Polygonum cilinode, Mountain Bindweed
Pplygonum ramosissimum, Bushy Knotweed
ChenopodTum leptophyllum (sensu Fernald 1950), Slender Goosefoot
Froelichia floridana, Cottonweed
Arenaria patula, Spreading Sandwort
Silene caroliniana var. wherryi, Wherry's Catchfly
Nuphar variegatum, Bullhead Lily
Aconitum noveboracense, Northern Monkshood
Aconitum uncinatum, Southern Monkshood
Actaea rubra, Red Baneberry
Ranunculus pusillus, Low spearwort
Trollius~Iaxus, Spreading Globe-flower
Magnolia macrophylla, Bigleaf Magnolia
Magnolia tripetala, Umbrella Magnolia
Arabis divaricarpa, Limestone Rock-cress
I
I
I
                                      H-4

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
   TABLE H-2.   LIST OF STATE AND FEDERALLY ENDANGERED PLANT SPECIES IN OHIO
                                  (Continued)


Arabis drummondii, Drummond's Rock-cress
Arabis patens, Spreading Rock-cress
Draba brachycarpa, Little Whitlow-grass
Draba cuneifolia, Wedge-leaf Whitlow-grass
Draba reptans, Carolina Whitlow-grass
Erysimum arkansanum (E. capitatum), Western Wall-flower
Drpsera intermedia, Spathulate-leaved Sundew
Ribes niissouriense, Missouri Gooseberry
Ribes rotundifolium, Appalachian Gooseberry
Ribes triste,  Swamp Red Currant
Amelanchier sanguinea, Rock Serviceberry
Palibarda repens, Robin-run-away
PotentilTa arguta, Tall Cinquefoil
Prunus nigra,  Canada Plum
Pyrus decora (Sorbus decora), Western Mountain-ash
Rubus setosus, Small Bristleberry
Astragalus neglectus, Cooper's Milk-vetch
Baptisia australis, Blue False Indigo
Desmodium illinoense, Prairie Tick-trefoil
Desmodium sessilifoTium, Sessile Tick-trefoil
Galactia volubilis, Milk-pea
Lathyrus venosus, Wild Pea
Oxalis montana (0. acetosella), White Wood-sorrel
Geranium bicknelTii^Bicknell's Crane's-bill
Polygala cruciata, Cross-leaved Milkwort
Polygala curtissii, Curtiss' Milkwort
Euphorbia serpens, Roundleaf Spurge
PhyllanThus caroliniensis, Carolina Leaf-flower
Paxistima canbyi,Cliff-green
Acer pensylvanicum, Striped Maple
Caenothus herbaceus (C. ovatus), Prairie Redroot
Hypericum denticuTatum, Coppery St. John's-wort
Hudsonia tomentosa, Beach-heather
Viola missouriensis, Missouri Violet
Viola nephrpphylla, Northern Bog Violet
Viola primulifolia, Primrose-leaved Violet
Viola tripartita yar. glaberrima (forma glaberrima), Wedge-leaf Violet
Viola walteri,"Walter's Violet
Arlia hispida, Bristly Sarsaparilla
Hydrocotyle umbellata, Navelwort
Ledum groenlandicum,~Labrador-tea
Rhododendron calendulaceum, Flame Azalea
Vaccinium myrtilloides, Velvet-leaf Blueberry
Vaccinium oxycoccos, Small Cranberry
Hottonia inflata, Featherfoil
Halesia"carolina, Silverbell
Styrax americanus, Snowbell
Gentiana puberulenta (G. puberula), Prairie Gentian
Gentiana saponaria, Soapwort Gentian
Cuscuta compacta, Sessile Dodder
Cynoglossum virginianum var. boreale (C. boreale), Northern Wild Comfrey
                                     H-5

-------
   TABLE H-2.   LIST OF STATE AND FEDERALLY ENDANGERED PLANT SPECIES IN OHIO
                                  (Continued)


Collinsonia verticillata (Micheliella verticillata),  Early Stoneroot
Monarda punctata, Dotted Horsemint
Trichostema dichotomum var. lineare (T. setaceum),  Narrow-leaved Bluecurls
Agalinis auriculata (Gerardia auriculata; Tomanthera auriculata), Ear-leaf
  Foxglove
Agalinis purpurea var. parviflora (A. paupercula var. pauperula and var.
                                   ~~                  -. be
  borealis; Gerardia paupercula var.  paupercula and var. borealis), Small
  Purple Foxglove
Agalinis skinneriana (Gerardia skinneriana), Skinner's Foxglove
Aureolaria pedicularia var. ambigens (Gerardia pedicularia var. ambigens),
  Prairie Fern-leaf False Foxglove
Orobanche ludoviciana, Louisiana Broom-rape
Utricularia cornuta, Horned Bladderwort
Plantagp cordata, Heart-leaf Plantain
Galium Tabradoricum, Bog Bedstraw
Galium palustre, Marsh Bedstraw
Symphoricarpos albus var. albus, Snowberry
Cirsium carolinianum, Carolina Thistle
Eupatorium hyssopifolium, Hyssop Thoroughwort
Heterotheca~graminifolia (Chrysopsis graminifolia), Silkgrass
Hieracium canadense, Canada Hawkweed
Hieracium longipilum, Long-bearded Hawkweed
Hymenoxys acaulis (Actinea herbacea), Lakeside Daisy
Prenanthes aspera, Rough Rattlesnake-root
Silphium laciniatum, Compass-plant
Solidago odora, Sweet Goldenrod
Verbesina occidentalis, Yellow Crownbeard
                                      H-6

-------
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
 I
I
          TABLE H-3.   ENDANGERED UNIONID MOLLUSCS KNOWN TO
               HAVE INHABITED THE SCIOTO RIVER SYSTEM
Scientific Name
Simpsonaias ambigua

Quadrula cylindrica

Quadrula metaneura

Quadrula nodulata

Fusconaia maculata

Plethobasus cyphyus

Pleurobema clava

Pleurobema cordatum

Cyprogenia stegaria

Potamilus laevissimus

Lampsilis teres

Lampsilis orbiculata

Lampsilis ovata
Common Name
Simpson's Shell

Cob Shell

Knobbed Rock Shell

Winged Pimpleback

Long-solid

Common Bullhead

Club Shell

Ohio Pigtoe

Ohio Fan Shell

Fragile Heel-Splitter

Yellow Sand Shell

Pink Mucket Pearly Mussel

Ridged Pocketbook
                                H-7

-------
    Table H-4.    Rare  or  endangered  fish  species  collected  during the  1979-1981
                    sampling   period   and/or   listed  as   occurring  in  the  central
                    Scioto   River  mainstem   study  area  by  the  Ohio   Department   of
                    Natural   Resources,    Heritage  Program.     (Source:      Ohio    EPA

                    1986a).
Spedtf
Lake chubsucker
Bluebreast darter
Goldeye
ODNR Status
Endangered
Threatened
Undete rained
OOHR Locations. Tear
Ctrclevllle Canal off Rd 100 (1974, 1981)
1) Scioto R at aouth of Deer Cr. (1961)
2) Scioto R near Clrclevllle (1962)
3) Scioto R. below Big Darby confluence (1963)
1) Scioto R. at Sreenlawn Ave. (1959)
2) Scioto R. dst. Big Darby Cr.
Ohio EPA collections, datts
Not collected
Not collected
KM 74.1 (1981)
Silver lanprey-


Shortnose gar


River redhorse
Endangered


Endangered


Endangered
Shorthead redhorse
Paddleflsh
                     Undetermined
Endangered
   confluence (1962)

Scioto R at CMIIIcotht (1964)


Scioto R. dst. Chillieothe  1973
1) Scioto R. ust. Big Darby Cr.  (1962)
2) Scioto R..dst. Dublin Rd. VTP dan (1979)
                  1) H. brevlceps - Scioto R.  at
                                          M. brevlceps - Sclo
                                          "CMIUcothe (1964)
                                       Scioto R. dst.  Breenlawn da* (1976)
Not collected
(only below CMIUcothe, 1979)

RN 118.8 (1981)
RM 70.7  (1980
RM 78.3  (1979
RM 102.0 (1979
RN 102.0 (1981
RM 104.8 (1979
RM 134.8 (1981)
RM 138.6 (1981)
                                         RM 138.6-70.7 (72 fish;
                                          1979-1981)
                                         Olentangy R. (1980)
                                         Btg Walnut Cr. (1980,1981)

                                         Not collected
                                                                                            I

                                                                                            I

                                                                                            I

                                                                                            I
                                                H-8

-------
       APPENDIX   I
 SITES  AND  STRUCTURES  IN
THE COLUMBUS AREA LISTED
ON THE NATIONAL REGISTER
   OF HISTORIC PLACES

-------
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
 I
                                  APPENDIX I


     The following sites/structures are listed on the National Register of
Historic Places.
Delaware County

Ashley, Building at 500 East High Street
  (Eastlake Houses of Ashley Thematic
  Resources) (11-25-80)

Building at 505 East High Street
  (Eastlake Houses of Ashley Thematic
  Resources) (11-25-80)

Building at 101 North Franklin Street
  (Eastlake Houses of Ashley Thematic
  Resources) (11-25-80)

Building at 223 West High Street
  (Eastlake Houses of Ashley Thematic
  Resources) (11-25-80)

Ashley vicinity, Sharp, Samuel, House
  (Sharp's Run), 7436 Horseshoe Rd.
  (07-29-82)

Delaware.  Delaware County Courthouse.
  N. Sandusky St. and Central Ave.
  (5-22-73) PH0034681

Delaware Public Library, 100 N. Sandusky
  St. (01-11-83)

Elliott Hall, Sturges Library, and
  Merrick Hall.  Ohio Wesleyan
  University Campus (4-23-73) PH0094480

Monnett Hall, Ohio Wesleyan University
  Campus at Elizabeth and Winter Sts.
  (6-23-75).

Sandusky Street Historic District.
  44 S. to 92 N. Sandusky, 47 E. to
  31 W.

St.  Mary's Church and Rectory, 82 E.
  William St. (5-23-80)

Van Deman, Henry, House, 6 Darlington
  Rd. (05-31-84)
                                     1-1
Delaware vicinity.  Greenwood Farms.
  S. of Delaware off U.S. 42 (4-17-79);
  79/07/23 079 0001773

Limestone Vale, 3490 Olentangy River Rd.
  (10-2-78)

Ufferman Site, N. of Delaware (7-24-74)
  PH0034711

Warren Tavern Complex.  U.S. 36
  (08/30/83)

Galena vicinity.  Curtiss, Marcus, Inn
  E. of Galena at 3860 Sunbury Rd.
  (12-12-76)

Keeler, Diadatus.  House, SE of Galena
  at 4567 Red Bank Rd. (2-2-79)
  80/01/10079 0006789

Spruce Run Earthworks.  About 3 mi. S.
  of Galena, (7-16-73) PH0034703

Harlem vicinity.  Cook, John, Farm, E.
  of Harlem at Miller Paul Rd. and
  Gorsuch Rd. (4-11-77)

Olive Green vicinity.  Chambers Road
  Covered Bridge, 1.5 mi. NE of Olive
  Green (11-21-74) PH0085049

Sunbury Tavern (Hopkins House), NW
  corner OH 37 and Galena Rd. (2-24-75)

Sunbury Township Hall, Town Sq.
  (2-20-75)

Sunbury vicinity, Center Inn, SE of
  Sunbury on OH 37 (01/11/83)

Westerville vicinity, Sharp, Stephen,
  House, N. of Westerville on Africa Rd.
  (09/30/82)

-------
Winter, and 9 E. to 17 W. William
  (12-17-82)

Worthington vicinity.  Highbank Park
  Works.  E. bank of Olentangy River
  (2-15-74) PH0112895
Fairfield County

Amanda, Barr House, 350 W. Main St.
  (11-26-80)

Amanda vicinity.  Allen, Lyman, House
  and Barn, NW of Amanda on OH 188
  (11-18-76)

Baltimore vicinity. Bright, John,
  Covered Bridge, 2.5 mi. SW of
  Baltimore over Poplar Creek (5-28-75)

Miller Farm, S of Baltimore on
  Pleasantville Rd. (5-22-75)

Musser, Henry, House, SE of Baltimore at
  7079 Millersport Rd. (5-5-78)

Pugh-Kittle House, 2140 Bickel Church
  Rd. (06-16-83)

Canal Winchester.  Loucks Covered
  Bridge, SE of Canal Winchester on SR
  207 (Diley Rd.) (10-8-76)

Carroll vicinity. Ety Enclosure, NE of
  Carroll (7-12-74) PH0034801

Ety Habitation Site, NE of Carroll
  (7_24-74) PH0034819

Carroll vicinity.  John Bright, No. 1
  Iron Bridge, 2 mi. (3.2 km) NE of
  Carroll on Havensport Rd. (9-20-78)

Lancaster.  Bush, Samuel, House, 1934
  Cold Spring Dr. (10-1-74) PH0034762

Lancaster Historic District, Roughly
  bounded by 5th Ave., Penn Central

Lancaster West Main Street Historic
  District, W. Main St. from Columbus to
  Broad St. (2-2-79); 80/01/10079
  0006790
                                        I
                                        I
                                        I
Medill, William, House, 319 N. High St.
  (3-30-78)

Sherman, John, Birthplace, 137 E. Main
  St. (10-15-66) PH0034845 NHL.

Square 13 Historic District, Roughly    I
  area along Broad and High Sts. between*
  Mulberry and Chestnut Sts. (7-24-72)
  PH0034851 HABS;G
                                        I
St. Peter's Evangelical Lutheran Church,
  Broad and Mulberry Sts. (4-16-79);    —
  79/07/23 079 0001775                  1

Lancaster, Tallmadge-Mithoff House, 720
  Lincoln Ave. (5-6-76)
                                        I
Lancaster vicinity. Chestnut Ridge Farm,
  3375 Cincinnati-Zanesville Rd., SW.
  (7-24-72) PH0034771                   •

Concord Hall, 1445 Cincinnati-Zanesville
  Rd., SW. (U.S. 22) (10-25-72)
  PH0034789
Reber, Valentine, House, W. of Lancaster
  at 8325 Lancaster-Circleville Rd.
  (OH 188) (7-30-75)
                                        I

                                        I
Willow Lane Farm (Nathaniel Wilson      —
  House), SW of Lancaster on U.S. 22    •
  (10-26-72) PH0034878                  •

Lithopolis vicinity.  Old Maid's Orchard*
  Mound, E. of Lithopolis (7-15-74)     •
  PH0034843

Lockville.  Lockville Canal Locks, Off  •
  Pickerington-Lockville Rd. (9-10-74)  •
  PH0085006

Pickerington vicinity, Dovel, J.H.,     •
  Farm, 660 N. Hill Rd. (03-15-82)      m
Hizey Covered Bridge, E. of Pickerington
  on SR 235 (10-8-76)
Stemen Road Covered Bridge, NE of       »
  Pickerington over Sycamore Creek,     •
  (4-20-79); 79/07/23 079 0001776       *

RR tracks, OH 33 and Tennant St.
  (08-11-83)
                                      1-2
                                        I

                                        I

                                        I

-------
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Rushville, Rushville Historic District,
  Bremen Ave., Main and Market Sts.
  (11-24-80)

Rushville vicinity.  Winegardner
  Village (7-30-74) PH0034886

Rock Mill. Rock Mill Covered Bridge,
  SR 41 (4-26-76)

Royalton.  Royalton House, Amanda
  Northern Rd. (7-30-75)

Sugar Grove vicinity.  Crawfis
  Institute, Crawfis and Old Sugar Grove
  Rds. (11-29-79); 80/01/10079 0006402

West Rushville, Ijams, Joseph, House,
  Broad and Main Sts. (06/16/83)

Franklin County

Bexley, Duncan, Robert P., House, 333 N.
  Parkview Ave. (08-23-84)

Jeffrey, Malcomn, House 358 N. Parkview
  (05-08-83)

Canal Winchester, Canal Winchester
  Methodist Church, S. Columbus and High
  Sts. (03-15-82)

Canal Winchester vicinity.  Bergstresser
  Covered Bridge, W. of OH 674 over
  Walnut Creek (5-3-74) PH0070181

Central College Multiple Resource Area.
  This area includes:  Westerville
  vicinity, Central College Presbyterian
  Church, Sunbury Rd.; Fairchild
  Building.

Central College vicinity.  Squire's Glen
  Farm, 6770 Sunbury Rd. (8-13-74)
  PH0070432

Columbus, American Insurance Union
  Citadel, 50 W. Broad St. (3-21-75)

Camp Chase Site, 2900 Sullivant Ave.
  (4-11-73) PH0112909

Broad Street United Methodist Church,
  501 E. Broad St. (11-26-80)
Columbia Building, 161-167 N. High St.,
  (08-12-83)

Capital University Historic District,
  E. Main St. and College Ave.
  (12-17-82)

Columbus Country Club Mound, 4831 E.
  Broad St., (2-15-74) PH0070211

Columbus Near East Side District,
  Roughly bounded by Parsons Ave., Broad
  and Main Sts., and the railroad
  tracks (5-19-78)

Columbus Savings and Trust Building
  (Atlat Building), 8 E. Long St.
  (9-15-77)

Columbus Transfer Company Warehouse,
  55 Nationwide Blvd. (02-24-83)

Drake, Elam, House, 2738 Ole Country
  Lane (4-6-78)

East Town Street Historic District,
  Roughly bounded by Grant and Franklin
  Aves., Lester Dr. and E. Rich St.
  (7-30-76)

Felton School, Leonard Ave. and N.
  Monroe St. (05-31-84)

Fort Hayes, Cleveland Ave. and 1-71
  (1-26-70) PH0070238

Franklin Park Conservatory, 1547 E.
  Broad St. (1-18-74) PH0070246

Franklinton Post Office (David Deardurf
  House), 72 S. Gift St. (3-20-73)
  PH0070254

German Village, Roughly bounded by
  Livingston Ave., Pear Alley, Nursery
  Lane, Blackberry Alley, and Lathrop
  St. (12-30-74) PH0044148

Great Southern Hotel and Theatre,
  S. High and E. Main Sts. (12-02-82)

Hamilton Park Historic District, Broad
  and Long Sts. (07-28-83)
                                      1-3

-------
Hanna House, 1021 E. Broad St.
  (4_19_79); 79/07/23 079 0001778

Harrison, Gen. William Henry,
  Headquarters (Jacob Oberdier House),
  570 W. Broad St. (12-15-72) PH0070271

Hayes and Or ton Halls, Ohio State
  University, The Oval (7-16-70)

Higgins, H.A., Building (Flatiron
  Building), 129 E. Naghten St.
  (8-27-79); 79-11-30 079 0005031

Holy Cross Church, Rectory and School,
  212 S. 5th St. (4-26-79) 79/07/23 079
  001779

Huntington, Franz, House, 81 N. Drexel
  Ave. (5-29-80)

Indianola Junior High School, 420 E.
  19th Ave. (6-30-80)

Jaeger Machine Company Office Building,
  550 W. Spring St. (06-16-83)

Jefferson Avenue Historic District,
  Roughly bounded by 1-71, E. Broad,
  llth, and Long Sts. (12-02-82)

Jones, W.H., Mansion, 731 E. Broad St.
  (10-2-78)

Krumm House, 975-979 S. High St.
  (09-30-82)

Long and Third Commercial Building,
  103-113 E. Long St. (07-01-82)

Near Northside Historic District, Off OH
  315 (6-4-80)

North Market Historic District, Roughly
  bounded by W. Goodale, Park, High,
  Front, and Vine Sts. (12-30-82)

Ohio Asylum for the Blind, 240 Parsons
  Ave.  (7-26-73) PH0070351

Ohio National Bank, 167 S. High St.
  (11-26-80)

Ohio Stadium, 404 W. 17th Ave. (3-22-74)
  PH0070360
                                      1-4
Ohio State Arsenal, 139 W. Main St.
  (7-18-74) PH0070378

Ohio Statehouse, SE corner of High and
  Broad Sts. (7-31-72) PH0070386 G.
  Ogers, Isaiah Saiah Rogers.

Ohio Theatre, 39 E. State St. (4-11-73)
  PH0070394 NHL; G.

Old Governor's Mansion (Ohio Archives
  Building, Charles H. Lindenberg
  House), 1234 E. Broad St. (6-5-72)
  PH0070408

Old Ohio Union, 154 W. 12th Ave.
  (4-20-79); 79/07/23 079 0001780

Old Port Columbus Airport Control Tower,
  420 E. 5th Ave. (7-26-79); 79-11-13
  079 0004392

Orton Memorial Laboratory, 1445 Summit
  St. (11-25-83)

Pierce, Elijah, Properties, 435 E. Long
  St. and 142-44 N. Everett Alley
  (08-03-83)
I
I
1
I
I
I
I
I
I
I
Pythian Temple and James Pythian
  Theater, 861-867 Mt. Vernon Ave.      m
  (11-25-83)                            •

Rankin Building, 22 W. Gay St.
  (03-10-82)
Rickenbacker, Capt. Edward V., House,
  1334 E. Livingston Ave. (5-11-76) NHL.
I

I
Schlee-Kemmler Building, 328 S. High St
  (12-02-82)

Second Presbyterian Church, 132 S. ThirdB
  St. (01-11-83)                        "
Seneca Hotel, 361 E. Broad St.
  (12-29-83)

Sessions Village, Both sides of  Sessions,
  Dr. (2/20/75)

Smith, Benjamin, House,  181 E. Broad
  St. (6/4/73) PH0070424
I

I

I

I

I

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
South High Street Commercial Grouping,
  Bounded by Pearl, Mound, Main, and
  High Sts. (12/29/83)

Sullivant, Lucas, Building, 714 W. Gay
  St. (3/20/73) PH0070441

Thurber, James, House, 77 Jefferson Ave.
  (11/8/79); 80/01/10079 0006403

Toledo and Ohio Central Railroad
  Station, 379 W. Broad St. (6/18/73)
  PH0070475 HAER; G.

Trinity Episcopal Church, 125 E. Broad
  St. (11/13/76)

U.S. Post Office and Courthouse (Old,
  Old Post Office), 121 E. State St.
  (4/11/73)

Valley Dale Ballroom, 1590 Sunbury Rd.
  (12/17/82)

Welsh Presbyterian Church, 315 E. Long
  St. (11/24/80)

Wyandotte Building, 21 W. Broad St.
  (2/23/72) PH0070491 HABS

York Lodge No. 583, 1276 N. High St.
  (07/19/84)

Columbus Vicinity

Agler-la Follette House, 2621 Sunbury
  Rd. (12/14/78)

Davis, Samuel, House, 4264 Dublin Rd.
  (2/15/74) PH0070220

Hartman Stock Farm Historic District,
  S. of Columbus on U.S. 23 (10/9/74)
  PH084999

Jackson Fort (12/10/74) PH0085251

McDannald Homestead, NE of Columbus at
  5847 Sunbury Rd. (2/17/78)

Noble, Jonathan, House, 5030 Westerville
  Rd. (SR 3) (12/3/75)

Dublin vicinity.  Davis, Anson, House,
  4900 Hayden Run Rd. (7/7/75)
                                     1-5
Holder-Wright Works (2/15/74) PH0070319

Sells, Benjamin, House, S. of Dublin at
  4586 Hayden Run Rd. (7/30/75)

Gahanna, Shepard Street School (Gahanna
  Nursing Home), 106 Short St.
  (11/29/79); 80/01/10079 0006404

Grove City, Gantz Homestead, 2233 Gantz
  Rd. (6/20/79); (10/23/79) 079 0002507

Groveport, Groveport Log Houses, Wirt
  Rd. (5/6/76)

Groveport Town Hall Historic Group, 628,
  632 Main and Main and Front Sts.
  (7/31/78)

Billiard vicinity.  Wesley Chapel, SE of
  Billiard at 3299 Dublin Rd. (2/27/79);
  79/07/13 079 0000620

Lockbourne vicinity, Herr, Christian S.,
  Bouse, N. of Lockbourne at 1451
  Rathmell Rd. (03/05/82)

Marble Cliff, Miller, J.F., House, 1600
  Roxbury Rd. (05/31/84)

Riverlea, Russell, Mark, House 5805 N.
  High St. (12/12/76)

Sunbury Rd.;  Presbyterian Parsonage,
  6972 Sunbury Rd.; Washburn, Rev.
  Ebenezer, Bouse, 7121 Sunbury Rd.
  (11/25/80)

Washington Township.  Washington
  Township Multiple Resource Area.  This
  area includes various properties at
  various locations.  Details available
  upon request. (4/11/79); 79/07/16 079
  0001090

Westerville,  Alkire House, 269 N. State
  St. (3/30/78)

Hanby, Benjamin, House, 160 W. Main St.
  (11/10/70)  PH0094501

Hart, Gideon, House, 7328 Hempstead Rd.
  (8/14/73) PH0070289

-------
Otterbein Mausoleum, W. Walnut St.
  (11/29/79); 80/01/10079 0006405

Towers Hall, Otterbein College, Main and
  Grove Sts., Otterbein College campus
  (3/4/71) PH0070459

Westerville High School, Vine Street
  School, 44 N. Vine St. (5/29/75)

Westerville vicinity.  Everal, John W.,
  Farm Buildings, 7610 Cleveland Ave.
  (9/18/75)

Osborn, Charles S., 5785 Cooper Rd.
  (3/28/77)

Worthington, Johnson, Orange, House,
  956 High St. (4/3/73) PH0070335

New England Lodge, 634 N. High St.
  (3/20/73) PH0070343

Snow, John, House, 41 W. New England
  Ave. (7/26/73) PH0071251

Worthington Manufacturing Company
  Boardinghouse, 25 Fox Lane (6/19/73)
  PH0112917

Worthington Multiple Resource Area.
  This area includes:  Adams, Demas,
  House, 721 High St.; Bishop-Noble
  House, 48 W. South St.; Brown, Sidney,
  House, 12 E. Strafford Ave.; Fay,
  Cyrus, House, 64 W. Granville Rd.;
  Gardner House, 80 W. Granville Rd.;
  Johnson, Orange, House, 956 High St.
  (previously listed in the National
  Register 4-3-73); Kilbourne House,
  679-681 High St.; Ladd-Mattoon House,
  73 E. North St.; New England Lodge,
  634 High St. (previously listed in  the
  National Register 3-20-73); Old
  Worthington Inn, New England and High
  Sts.; President's House, 38 Short  St.;
  Ripley House, 623 High St.; St. John's
  Episcopal Church, 700 High St.; Scott,
  Travis, House, 72 E. Granville Rd.;
  Sharon Township Town Hall, Granville
  Rd. and Hartford St.; Skeele, Capt.
  J.S., House, 700 Hartford St.; Snow,
  John, House, 41 W. New England Ave.
  (previously listed in the National
Licking County

Brownsville vicinity.  Flint Ridge
  (11/10/70) PH0070904

Croton vicinity.  Belle Hall Covered
  Bridge, E. of Croton on Dutch Cross
  Rd. (10/22/76)

Granville, Avery-Hunter House, 221 E.
  Broadway (12/27/79)

Buxton Inn, 313 E. Broadway (12/26/72)
  PH0070874
                                                                                      I
                                                                                      I
  Register 7-26-73); Topping, J.R.,
  House, 92 E. Granville Rd.; Park,
  Jonathan, House, 91 E. Granville Rd.; I
  Wilcox, Hiram, House 196 E. Granville §
  Rd.; Worthington Historical Society
  Museum, 50 W. New England Ave.;       ^
  Worthington Manufacturing Company     I
  Boarding House, 25 Fox Lane           ™
  (previously listed in the National
  Register 6-19-73); Worthington United
  Presbyterian Church, High St. and W.
  Granville Rd.; Worthington Village
  Green, Village Green; Wright, Horace, _
  House, 137 E. Granville Rd.; Wright,  •
  Potter, House, 174 E. New England Ave. •
  (4/17/80)
I
I

I

I

I

I

I
Granville Multiple Resource Area
  (Partial Inventory).  This area
  includes:  Granville, Granville
  Historic District, OH 37; Bancroft,
  A.A., House, N. Pearl St. and
  Washington Dr.; Carpenter, Wallace W., _
  House (The Castle) 323 Summit St.;    •
  Dustin Cabin, 597 N. Pearl St.; RogersM
  House, 304 N. Pearl St.; Rose, Capt.
  Levi, House 631 N. Pearl St.
  (11/28/80)
St. Lukes Episcopal Church, 111 E.
  Broadway St. (4/26/76)
I

I
Granville vicinity, Bryn Mawr (Fassett's
  Folly), 3758 Lancaster Rd., SW        m
  (03/29/83)                            •

McClune's Villa, 537 Jones Rd.
  (04/22/82)
I
                                      1-6
                                        I

-------
I
I
I
1
I
i
i
I
i
i
i
i
1
i
i
i
i
i
i
Stanbery, Edwin, Office, 1 mi (1.6 km)
  E. of Granville (11/30/78)

Heath, Ohio Canal Groundbreaking Site,
  OH 79 (5/24/73) PH0070963

Johnstown, Monroe Township Hall-Opera
  House, 1 S. Main St. (7/6/81)

Johnstown vicinity.  Lynnwood Farm, S.
  of Johnstown at 4986 Caswell Rd.
  (6/22/79); (10/23/79) 079 0002509

Newark.  Chapel Hill Cemetery Buildings,
  Cedar St., Chapel Hill Cemetery
  (4/13/77)

Courthouse Center, 35-37 S. Park PI. and
  jet. of S. Park and S. 2nd St.
  (11/29/79); 80/01/10079 0006411

Home Building Association Bank, 6 W.
  Main St. (7/2/73) PH0070912

Hull Place, 686 tf. Main St. (12/21/79)

Licking County Courthouse, Courthouse
  Sq. (3/20/73) PH0070921

McNamar-McLure-Miller, Residence, 124 W.
  Main St. (06/17/82)

Newark Earthworks, Mound Builders State
  Memorial (10/15/66) PH0070955 NHL.

Oakwood, 64-70 Penney Ave. (5/29/80)

Pennsylvania Railway Station, 25 E.
  Walnut St. (11/29/79); 80/01/10079
  0006412

Rhoads, Peter F., House, 74 Granville
  St. (11/28/80)

Sherwood-Davidson and Buckingham Houses,
  W. Main and 6th Sts. (11/10/77)

Shield's Block, 23-29 S. Park PI.
  (11/29/78)

Upham-Wright House, 342 Granville St.
  (6/22/79); (10/23/79) 079 0002510

West Side Planning Mill, 197 Maholm St.
  (01/21/83)
                                      1-7
Williams, Elias, House (Bolton House),
  565 Granville St. (4/16/79); 79/07/23
  079 0001786

Newark vicinity.  Upland Farm, N. of
  Newark off OH 657, (12/1/78)

Pataskala, Bethel Baptist Church
  (Pataskala MRA), Vine and Cedar Sts.
  (09/22/83)

Casterton House (Pataskata MRA), 105
  Broadway (09/22/83)

Elliot House (Pataskala MRA), 301 S.
  Main St. (11/14/83)

Kauber, Warren F., Funeral Home
  (Pataskala MRA), 289 S. Main St.
  (09/22/83)

Mead House (Wind Flower House)
  (Pataskala MRA), 245 S. Main St.
  (09/22/83)

Pataskala Banking Company (Pataskala
  MRA), 354 S. Main St. (09/22/83)

Pataskala Elementary School (Pataskala
  MRA), 396 S. High St. (09/22/83)

Pataskala Jail (Pataskala MRA), Main St.
  (09/22/83)

Pataskala Presbyterian Church (Pataskala
  MRA), Atkinson and Main Sts.
  (11/14/83)
Pataskala Town Hall (Pataskala MRA),
  Main St. (09/22/83)

Pataskala United Methodist Church
  (Pataskala MRA), 458 S. Main St.
  (09/22/83)

Madison County

Lafayette.  Red Brick Tavern, 1700
  Cumberland Rd. (9/5/75)

London.  Madison County Courthouse,
  Public Sq. (3/14/73) PH0094552

Swetland House, 147 E. High St.
  (01/11/83)
430

-------
Mount Sterling.  Mount Sterling Historic
  District, Both sides of London St.
  (10/1/74) PH0060801

Plain City vicinity.  Gary Village Site,
  SE of Plain City (5/13/75)

Somerford vicinity.  Wilson, Valentine,
  House, About 1 mi. N. or Somerford off
  1-70 (5/22/73) PH0060828

Pickavay County

Ashville, Ashville Depot, Madison and
  Cromley Sts. (2/25/80)

Circleville.  Anderson, William
  Marshall, House, 131 W. Union St.
  (11/29/79); 80/01/10079 0006419

Circleville Historic District, Main and
  Court Sts. (5/16/78)

Memorial Hall, 165 E. Main St.
  (11/21/80)

Morris House, 149 W. Union St. (8/3/79);
  79-11-13 079 0004400

Circleville vicinity, Horsey-Barthelmas
  Farm, W. of Circleville on OH 104
  (7-24-80)

Lawndale Farm Complex, 26476 Gay
  Dreisbach Rd. (04/19/84)

Mount Oval (Tolbert House), Off U.S. 23
  (7/25/74) PH0071293

Peters, Stevenson, House, OH 188
  (02/09/84)

Redlands,  1960 N. Court St. (05/14/82)

Kingston vicinity.  Bellevue, N. of
  Kingston on OH  159 (3/17/76)

Marcy vicinity.   Fridley-Oman Farm,
  W. of Marcy in  Slate Run Metropolitan
  Park  (12/6/75)

South Bloomfield  vicinity, Renick Farm,
  N. of Bloomfield on U.S. 23 (03/05/82)
Williamsport vicinity.  Bazore Mill,
  S. of Williamsport on OH 138 at Deer
  Creek (12/19/78)

Williamsport vicinity.  Shack, The, NW
  of Williamsport (5/23/74) PH0071307

The following properties have been
  determined to be eligible for
  inclusion in the National Register.

Fairfield County
I
I
I
I
I
Lancaster, U.S. Post Office—Lancaster   I
  (10/28/83)                             fl

Richland, R.F., Baker Bridge, Thornville •
  Rd. and Little Rush Creek; 78/11/13    •
  078 0055084                            "

Franklin County
Columbus, Barber Shop, 82-86 E. Town
  St. (1204.3)

Beggs Building, 21 E. State St.

Bldg. at 736-40 East Long Street
  (02/17/84)
I

I

I
Central National Bank Building, 152-166
  S. High St. (1204.3)                   •

Hartman Theater Building, 73-87 E. State
  St. (1204.3)
I

I
LaSalle Wine Store, 242-244 S. High St.
  (1204.3)

Owen, Jim, Real Estate, 232 S. High St.
  (1204.3)

Trailways, 246-254 S. High St. (1204.3)  I

1000-02 S. High Street  (63.3)

17-19 E. Stewart Avenue (63.3)           (

21-33 E. Stewart Avenue (63.3)

99 S. High Street  (63.3)
                                      1-8
I

I

I

I

-------
I
I
I
I
I
I
i
l
i
I
i
i
i
i
i
i
i
i
i
Licking County

Health.  Digiondomenico Site (Ohio LIC
  343-0.00).; 78/11/15 078 0050632

Pickavay County

Darby township, Orient Bridge, OH 762
  over Big Darby Creek (63.3)

The following sites/structures are
  pending inclusion to the National
  Register.

Franklin County

Broad Street Apartments, East Broad
  Street MRA, 880—886 E. Broad St.,
  86003404 11/04/86

Broad Street Christian Church, East
  Broad Street MRA, 1051 E. Broad St.,
  86003448, 11/04/86

Cambridge Arms, East Broad Street MRA,
  926 E. Broad St., 86003412, 11/04/86

Central Assurance Company, East Broad
  Street MRA, 741 E. Broad St. 86003421,
  11/04/86

East Broad Street Commercial Building,
  East Broad Street MRA, 747, 749, 751
  E. Broad St., 86003424, 11/04/86

East Broad Street Historic District,
  East Broad Street MRA, Along E. Broad
  St. between Monypenny and Ohio Aves.
  86003393, 11/04/86

East Broad Street Presbyterian Church,
  East Broad Street MRA, 760 E. Broad
  St., 86003397, 11/04/86

Garfield—Broad Apartments, East Broad
  Street MRA, 775 E. Broad St.,
  86003427, 11/04/86

Heyne—Zimmerman House, East Broad
  Street MRA, 973 E. Broad St.,
  86003450, 11/04/86

Hickok, Frank, House,  East Broad Street
  MRA, 955 & 957 E. Broad St., 86003444,
  11/04/86
                                     1-9
House at 753 East Broad Street, East
  Broad Street MRA,, 753 E. Broad
  Street, 86003425, 11/04/86

Jacobs, Felix A., House, 1421 Hamlet
  St., 86003434, 11/04/86

Johnson—Campbell House, East Broad
  Street MRA, 1203 E. Broad St.,
  86003414, 11/04/86

Joseph—Cherrington House, East Broad
  Street MRA, 785 E. Broad St.,
  86003429, 11/04/86

Kauffman, Linus E., House, East Broad
  Street MRA, 906 E. Broad St.,
  86003410, 11/04/86

Kaufman, Frank J., House, East Broad
  Street MRA, 1231 E. Broad St.,
  86003420, 11/04/86

Levy, Soloman, House, East Broad Street
  MRA, 929 E. Broad St., 86003427,
  11/04/86

Lovejoy, Carrie, House, East Broad
  Street MRA, 807 E. Broad St.,
  86003435, 11/04/86

Morris, C.F., House, East Broad Street
  MRA, 875 E. Broad St., 86003398,
  11/04/86

Frentiss, Frank, House, East Broad
  Street MRA, 706 E. Broad St.,
  86003396, 11/04/86

Prentiss—Tulford House, East Broad
  Street MRA, 1074 E. Broad St.,
  8603413, 11/04/86

Saint Paul's Episcopal Church, East
  Broad Street MRA, 787 E. Broad St.,
  86003430, 11/04/86

Schueller, Erwin W., House, East Broad
  Street MRA, 904 E. Broad St.,
  86003406, 11/04/86

Scofield—Saner House, East Broad Street
  MRA, 1031 E. Broad St., 86003447,
  11/04/86

-------
Sharp—Page House, East Broad Street
  MRA, 935 E. Broad St.

86003445

86003449

ARCHAEOLOGICAL SITES

Delaware County

Powell vicinity.  Highbanks Metropolitan
  Park Mounds I and II, E. of Powell on
  U.S. 23 (3/19/75)

Fairfield County

Canal Winchester vicinity.  Schaer,
  Theodore B., Mound, SE of Canal
  Winchester (6/20/75)

Carroll vicinity, Coon Hunters Mound
  (5/2/74) PH0034797

Pinkerington vicinity.  Fortner Mounds
  I, II.  NE of Pinkerington (7/12/74)
  PH0034827

Tarlton vicinity.  Tarlton Cross Mound,
  N. of Tarlton (11/10/70) PH0034860

Franklin County

Columbus.  Campbell Mound (11/10/70)
  PH0094498

COE Mound, W. of High St. (7/18/74)
  PH0070203

Columbus vicinity.  Hartley Mound, N. of
  Columbus (7/15/74) PH0070297

Galloway vicinity.  Galbreath, John
  Mound, W. of Galloway (7/15/74)
  PH0070262

Georgesville vicinity.  Cannon, Tom,
  Mound (5/2/74) PH0070190

Worthington vicinity.  Jeffers, H.P.,
  Mound (5/2/74) PH0070327
Licking County

Granville vicinity.  Alligator Effigy
  Mound (11/5/71) PH0070891

Homer.  Dixon Mound (Williams Mound)
  (6/4/73) PH0070882

Reynoldsburg vicinity.  ETNA Township
  Mounds I and II, E. of Reynoldsburg
  off 1-70 (9/5/75)

Utica vicinity.  McDaniel Mound (5/2/74)
  PH0070939

Melick Mound, S. of North Fork of
  Licking River (3/27/74) PH0070947

Madison County

West Jefferson vicinity.  Skunk Hill
  Mounds (7/30/74) PH0060810

Pickavay County

Circleville vicinity.  Arledge Mounds I
  and II (7/30/74) PH0071285

Luthor List Mound (10/16/74) PH0034291

Fox vicinity.  Clemmons, W.C., Mound
  (5/2/74) PH0071315

Tarlton vicinity.  Horn Mound (8/7/74)
  PH0034304

Williamsport vicinity.  Tick Ridge Mound
  District, NW of Williamsport (6/11/75)
                                     1-10
I
I
I
I
I
I
I
I
I
I
I
I
i
i
i
 i
 i
 i
 i

-------
      APPENDIX  J
ARCHAEOLOGIC BACKGROUND

-------
 I
 I
 I
 1
 I
 I
 I
 I
 I
 I
 I
I
I
I
I
1
I
I
I
                                  APPENDIX J
                           ARCHAEOLOGIC BACKGROUND

     The earliest evidence of human culture within the Scioto Drainage system
is evidenced by the Fluted Point Complex of the Palaeo-Indian Tradition, which
has been dated to between 18,000 and 10,000 years B.C.  This component, the
Fluted Point Complex, is represented primarily by the surface recovery of
isolated Fluted Points (projectile points) and other characteristic artifacts
of this manifestation.

     The Fluted Point Complex is followed by the Piano Complex of the Palaeo-
Indian Tradition, dating between 10,000 and 6,000 years B.C.  The Piano
Complex is documented in the Scioto Valley by a series of isolated surface
finds of characteristic projectile point types including Lanceolate Points,
Sawmill Stemmed Lanceolate Points, and Stringtown Spurred-Stemmed Lanceolate
Points.

     The known distribution of Piano Complex workshop sites centers in
Coshocton County in proximity to the outcrops of Upper Mercer Flint with a
secondary center in Licking County adjacent to the heavily utilized Flint
Ridge Flint.  These raw materials were used in the manufacture of the vast
majority of Lanceolate-style projectile points.  The distribution of excavated
sites and surface finds in this region would be along major stream valleys.

     The Archaic Developmental Stage spans the time interval from ca. 8,000 to
1,500 years B.C.  In part, the Piano Complex and the earliest manifestations
of the Archaic stage overlap in time.

     The Archaic Development Stage is evidenced by two cultural traditions
throughout the Scioto Drainage system:  the Appalachian Archaic Tradition
(8,000 to 3,500 years B.C.) and the Laurentian Archaic Tradition (3,500 to
1,500 years B.C.).

     The Kirk Phase of the Appalachian Archaic Tradition has been dated to
between 8,000 and 7,000 years B.C., while the St. Albans Phase dates to
between 7,000 and 6,100 years B.C.  The majority of sites are situated within
the low terraces of the major stream valleys—in environmental zones that have
been reconstructed as bottomland hardwood forests.
                                     J-l

-------
     The distribution of Kirk and St. Albans Phase components is rather well
known for the Scioto Drainage south of Circleville.  Stray surface finds of
Kirk Corner-Notched and St. Albans Bifurcated-Base projectile points were
recovered along both the east and west banks of the Scioto River in southern
Franklin County.  However, it was not possible to define either clusters of
artifact occurrence or to define sites on the basis of this analysis.  The
only evidence for Archaic Stage (6,000 and 3,500 years B.C.) occupation of the
Scioto Drainage has come from the surface recovery of several well-defined
projectile point types, either as isolated surface occurrences or from
occurrences in multicomponent surface manifestations.

     The Laurentian Tradition represents the most recent of the Archaic
Development Stage manifestations within Ohio.  The various components of the
tradition have been radiocarbon dated to between ca. 3,500 years B.C. and
prior to 1,500 years B.C.  In this region, one phase of the tradition has been
defined:  the Dunlap Phase of the Laurentian Tradition within the central and
lower Scioto Valley.  Sites occur as both open sites and within rock shelters
in the eastern portion of Ohio.  The majority of open sites are situated in
close proximity to the then-contemporary shorelines of water sources (lakes,
bogs, swamps, and streams).

     The Glacial Kame Manifestation represents a poorly understood series of
archaeological remains that are contemporaneous with the terminal portion of
the Laurentian Tradition.  The manifestation is known primarily from the
discovery of human burials that occur deep within shaft graves excavated into
glacial kames, usually elevated over adjacent stream valleys.  Sites of the
Glacial Kame Manifestation do occur in both Pickaway and Franklin Counties.

     The Scioto Tradition spans the time interval from 1,500 years B.C. to ca.
900 years A.D.  Three phases of the Scioto Tradition have been defined.  The
three major manifestations include the following:

     a.  Adena Phase (Early Woodlands), dating from  1,500 years B.C. to 1
         A.D./B.C.
         This phase represents the earliest manifestation of the Scioto
         Tradition within  the Scioto Drainage.  The  majority of manifestations
         have been dated  to between  1,500 years B.C. and 1 A.D./B.C., although
                                      J-2

-------
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
    components predating 1,500 years B.C. are known from both the Hocking
    River and Ohio River Valleys.  Only four Adena Phase burial mounds
    are known from Franklin and Pickaway Counties.

    The Adena Phase mortuary ceremonial manifestation is evidenced by one
    site occurring within a 10-kilometer radius of the Southerly WWTP.
    Typical burial mounds of the Adena Phase represent small structures
    covering less than 20 inhumations located on high terraces and/or
    bluffs overlooking major stream valleys.  The settlement pattern of
    the Adena Phase, known for limited information, consists of small
    villages or hamlets (2 to 10 structures) scattered along the low
    terraces and flood plain of the stream valleys.  One large habitation
    site—the Dominion Land Company Site in Franklin County—has been
    reported.

b.  Hopewellian Phase (Middle Woodland), dating from 150 years B.C. to
    650 years A.D.

    This phase of the Scioto Tradition has been dated to between 150 B.C.
    and 650 A.D.  The greatest concentration of sites occurs in the
    Scioto River Valley between Circleville and Portsmouth.  The concen-
    tration of Hopewellian earthworks occurs in the Scioto Valley and its
    tributaries south of Columbus.

    Within the central and lower Scioto Valley, Hopewellian hamlets
    appear to be composed of two to four structures (houses) situated on
    rises of the flood plain and first terrace of the Scioto River and
    the major tributary stream.

    Four Hopewellian Phase sites are known to be in the vicinity of the
    Southerly WWTP project area.

c.  Chesser Phase, Peters Phase, Cole Complex (Late Woodland),.dating
    from 650 to between 900 and 1,000 years A.D,

    The subsequent portion of the Scioto Tradition consists of a series
    of regionally defined phases:  the Peters Phase in the Hocking
    Valley, the Chesser Phase in the lower Scioto Valley, the Cole
    Complex in the upper Scioto Valley, and the Licklighter Phase in the
    Miami Valley.  These various Late Woodland phases occupy a time
    interval that has been radiocarbon dated to between 650 and 950 to
    1,000 years A.D.

    The Cole Phase (also known as Cole Complex) has been defined by Baby,
    Potter, and their co-workers for the upper portion of the Scioto
    Valley (Circleville to Columbus) and for the Darby Creek, Upper
    Scioto, and Olentangy Drainages.

    The terminal portion of the pre-European culture history of the
    Scioto Drainage is dominated by the Fort Ancient or Mississippian
    Tradition.
                                J-3

-------
Fort Ancient Tradition settlement patterns consist of large nucleated
villages, frequently oriented around vacant plazas or areas con-
taining platform or "temple" mounds and frequently defined by
palisades.  Villages are most frequently located in close proximity
to major streams and on rises within the flood plain, or on first
terraces of the stream valleys; frequently in close proximity to the
richest of the available soils.

A total of seven Late Woodland Mississippian manifestations are known
to occur in the vicinity of the Columbus Southerly project area.
Nearly all adjacent manifestations are known only from the results of
the phase II survey.

The terminal portion of the prehistoric sequence within the Scioto
Valley—the time interval from 1650 to 1680 until Anglo-European
settlement during the 1780s and 1790s—is poorly known from both the
archaeological and historical literature.  Only one site has been
excavated from this time interval—the Morrison Site from the Scioto
Valley south of Chillicothe.

In summary, the Scioto Drainage system has been used by a succession
of prehistoric cultures and prehistoric populations for over 18,000
years.  Many of these cultural manifestations are well represented
within the region.

A phase I and phase II survey of the Southerly WVTP in 1985 revealed
four prehistoric sites.
                             J-A

-------
            APPENDIX K
POPULATION PROJECTIONS AND METHODS

-------
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
                                  APPENDIX K
                      POPULATION PROJECTIONS AND METHODS

Introduction
     This appendix presents past population trends of  the overall  planning
area as well as the proposed interceptor areas.  The overall service  area
covers most of Franklin County, including all of the City of Columbus as well
as small portions of Delaware, Fairfield, and Licking  Counties.  This review
outlines baseline data used in evaluating population projections and  for
estimating the relative attractiveness for development of various  communities
within the planning area.

     Most of the available population projections have not been prepared for
small areas and the detailed information required for  accurate small  area
projections is not available.  The 1980 census provides  the baseline  for the
trend analysis used to prepare all of the regional population projections.
Because growth between 1970 and 1980 was less than expected and the growth
between 1980 and 1985 was greater than expected, the Ohio Department  of
Economic Development, which prepares the State population estimates at the
Ohio Data Users Center (ODUC), has revised its official  estimates  three times
since the 1980 U.S. Census.  The most recent estimate  was published and
verified in September of 1985.  Both the Mid-Ohio Regional Planning Commission
(MORPC) and Ohio Environmental Protection Agency (OEPA)  prepare population
estimates for small areas; that is counties, cities, and unincorporated areas.
These two agencies have not revised their population estimates to  reflect the
most recent ODUC projections.  Therefore, these small  area projections do not
reflect the most recent State-approved projections.  The Revised Facility Plan
Update (RFPU) considers these revisions, but does not  reflect the  region's
most recent growth trends.

Revised Facilities Plan Update Projections
     Population levels were forecast for the year 2015 in the RFPU.   Besides
the year 2015 population, used for planning purposes,  the population  in 1988
also was evaluated by this RFPU because the Clean Water Act Amendments mandate
compliance by all wastewater treatment facilities with NPDES permit limits by
                                     K-l

-------
July 1, 1988.  The following sources of population data and existing projec-

tions were  reviewed prior to development of RFPU projections:


     o  Environmental Impact Statement (EIS) for Wastewater Treatment
        Facilities for the Columbus Metropolitan Area (US EPA 1979);

     o  Design Finalization Overview Team Report (AWARE 1984);

     o  Facilities Plan Update Report (Malcolm Pirnie 1984);

     o  Growth Potential Report (City of Columbus 1984);

     o  Ohio Department of Development, Data Users Center, State and County
        Projections (June 1982);

     o  Ohio Department of Development, Data Users Center, Draft Final
        Population Projections (August 1985);

     o  Traffic Zone Projections - 1980 and 2010 (Mid-Ohio Regional Planning
        Commission 1983);

     o  Franklin County projections developed by the Design Parameters Team as
        a check against other projections;

     o  Miscellaneous Facilities Plan and Facilities Plan Segment documents
        pertaining to sewer service areas;

     o  Ohio Environmental Protection Agency, Office of the Planning
        Coordinator, Water Quality Management Plan Projections (1977, 1982).


As the preceding list indicates, numerous sources using various methodologies
were used to make population projections in the RFPU and  the Consolidated

Environmental Information Document (BID).  For the purpose of this EIS, the
population projections prepared by the Ohio Data Users Center (a division of
the Ohio Department of Development) and the OEPA were reviewed and adjusted to
reflect the overall service area for 1988, 2000, and 2008.


Ohio Data Users Center Projections

     The ODUC prepares the official population projections for the State of

Ohio.  ODUC bases its recent projections on the 1980 U.S. Census, and historic

trends for migration, births, and deaths.  The projections reviewed in this

EIS were revised in September 1985.  These projections were prepared on the

county and state level, for the years 1980 through 2010.  A preliminary

accuracy check recently was conducted by ODUC for their Franklin County
                                      K-2

-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
forecasts.  During this check, the population estimates were well within  the

accepted statistical confidence level with an error rate of less than  2

percent.



     Table K-l shows ODUC's population projections for the State of Ohio  and

the four counties in the service area for 1980 through 2010.  This table  shows

that the State of Ohio will decrease over the 30-year period while the

population in Franklin County as well as the other counties in  the service

area will increase.



     ODUC is responsible for certifying population projections  prepared within

the State.  In 1982, after several public hearings, ODUC certified OEPA's

Planning and Engineering Data Management System for Ohio (PEMSO) population

projections.  The OEPA prepared its projections for selected service areas on

the township, village, and county levels.  These projections were prepared

before the 1980 U.S. Census was released, and therefore are based on the  1970

U.S. Census and the growth trends exhibited in the area prior to 1980.

Although the methodology employed to make these projections is  sound,  the

growth between 1970 and 1980 was less than expected.  And the growth between

1980 and 1984 was larger than expected.



           TABLE K-l.   POPULATION PROJECTIONS FOR THE STATE OF  OHIO
                AND THE COUNTIES IN THE  COLUMBUS  SERVICE AREA

Ohio
Delaware
Fairfield
Franklin
Licking
1980
10,797,630
53,840
93,678
869,132
120,981
1990
10,681,863
61,709
98,655
924,592
127,390
2000
10,583,083
71,381
104,033
975,013
132,154
2010
10,398,338
81,164
107,577
1,026,008
136,765
Source:   (ODUC, 1985).
                                     K-3

-------
     Table K-2 lists the service area population by county as a percent of  the
total county population for the same time period.  This table indicates that
most of Franklin County's population (over 99%) will be located in the service
area.  Less than 1 percent of Fairfield's population and an average of 3
percent of Delaware and Licking Counties will be included in the service area.

Ohio Environmental Protection Agency Projections
     Table K.-3 compares OEPA's projections with ODUC's by county for 1980,
1985, and 2000.  This table indicates that ODUC currently assumes a slightly
higher growth rate for Franklin County than OEPA used in its earlier
forecasts.  ODUC'S projections are based on the 1980 U.S. Census and show a
higher 1980 population in Franklin County than OEPA.  In fact, OEPA's earlier
forecasts underestimate the 1980 Franklin County population by 56,000 persons.
OEPA uses higher growth rates for Fairfield, Delaware, and Licking Counties
than ODUC.  This results in higher overall population estimates by OEPA for
these counties.

     OEPA acknowledges that its 1982 PEMSO estimates may not reflect an
accurate picture of the service area population and has attempted to modify or
revise these estimates.  However, since the 1982 estimates are the only
numbers that have been certified by the State, OEPA cannot release the revised
version of these estimates.  The growth rates used for Franklin County are
similar for both OEPA and ODUC; this analysis will assume that if OEPA's 1980
population is adjusted to reflect the 1980 U.S. Census, then the two
projections will be more closely aligned.  This adjustment, referred to
hereafter as OEPA (adj.), is reflected in Table K-2 as part of the comparison
for Franklin County.  Since Fairfield, Delaware, and Licking Counties combined
comprise 1 percent of the total service area population in 1980 and 2 percent
of the total service area population in 2015, no adjustments were made for
these counties.
                                      K-4

-------











to

CzD
OS

M
M
Pi
to
to

«
*
—4






•a
01
o *-•
00 U
ON 0)
~H """l
O
U
CM






01
e r-
• H r^
•—* ~~*.
0) O
Ml P-»
(8 ON
CO — i






flj
0)
u x
^ *j
c

u X
.
C^ CO *^ CO
vO r-
CO CO

-4* co m CM
CO \O O O
**Q ^O CO "O
A A A m
n
in CN CN O
00 ON ^H ON
•*•»•• M
CN — -^ m
o o
00 00





ON '•O O in
oo -^ CN CN
CO ^O c^ r^
•»*«•» «
Q"» ^-H »^ «-4
OO ON
r^ r**








CN ON in »aO
vjO ^^ ^ in
O — i 00 O
ft ft M
p"*t ^^ ^^
OO OO
r^ P^>






1—4
(8
u
S 01 O
•H M 60 H
Ul i— < <8 C
3 .* 3 -^ (8
»Q C <8 £i Ol
S <8 •— ' O 14
3 M 0) -frf <8
-. Cb Q J ^2
O 3
CJ CO













































































-3- OO CN
00 ON
CO CO
*» »j
00 00



CO OO — I
in \o
CO CO
•* •*
r*» r^





00 OO v*3
00 O\
m m






f^ GO C^
o o
CO 00
A •%
"^ ^







vO oo -d-
-^ CN
O 0
* f.
*& *^










c '3 ON ON
CN CM
•* 4 CQ CQ
C &^ fM »O
08 3
O CO






a
•H
CN
~-4
"^
CN

"O
C
18

13
0)
U-4
•."4
Cb
c
• p4)

f*^
CO
9]
3
u
-4
•^
VOVO-
CN 00 CO ~J
•* P- 00 O
•* M f.
m co o
CN CO


vO in ~* CN
>^ CO CN O

ft *\ f,
CO CO ^^
CN CN




— • -* r~ CN
C^ QO C3 /^Q
CO \O O tPt
CN CO vO
CN CM





O ^ CN ^
— t CO O ^3"
CM vO vO -^
M M M
O\ CN CN
^H CN






\o r** co "*o
•sj OO CO vO
CN in CN o
m « *
r~ CN o
— I CN









P1^ ^O CO \O
CO CN -^ O
^"4 in vO co
M  -i-i e

.-i c >-> ^ oi
-J 18 -H U VJ
Ji U <8 -H CO
cj pE4 Cb i~3 J3
« 3
CO to

*
OA
A
,_4
^4
ON

in

00

CO
r-
00



OO
CN CN
•J3 2
in
00 U
3
0 0
^4
tJ X
.5 °
O U U
v0 Hi O
— ' U-l
* U
co 01 e
CO 4J 01
00 3 u
a en
B >>
O CO
O
>. c
O 01
in eg
—4 11 Ol
oo oo oo
< 18
in c
— < C 18
00 OS

u 18
U 4-1
Ol C8
u a
o
U 00
cu a
00 •*
CN —" '-i
— i 18 01
" U Ol
-3" C C
O> Ol --4
r- S 00
C C
o w

18 >
U C 00
o u c
H -4
O C
m •-< c
 U CO
H 3 S
O> O UJ
CO CO Cu
K-5

-------
             TABLE K-3.   COMPARISON OF POPULATION PROJECTIONS BY
                      COUNTY FOR THE COLUMBUS,  OHIO AREA
County
Source
Franklin
ODUC
OEPA
OEPA (adj.)**
Fairfield
ODUC
OEPA
Delaware
ODUC
OEPA
Licking
ODUC
OEPA
1980
869,132
812,670
869,132
93,678
82,401
53,840
54,779
120,981
125,943
1985
898,345
829,523
887,400
96,120
87,972
57,693
62,320
124,394
137,648
2000
975,013
906,903
971,700
104,033
106,180
71,381
87,810
132,154
162,791
2015*
1,048,000
974,900
1,044,600
109,000
128,500
86,500
123,500
138,000
187,000
 *This estimate is a simple extrapolation of OEPA
  on previous growth rates and rounded to 500.
**OEPA figures were increased to reflect the 1980
  complete explanation).
and ODUC projections based
U.S. Census (see text for
Comparisons
     Using the proportions shown in Table K-4 ODUC's county-wide population
estimates were adjusted to reflect the OEPA estimate of the service area
populations.  Table K-5 compares OEPA's PEMSO estimate with ODUC's estimates
adjusted to reflect the service area boundaries, and with the OEPA (adjusted)
estimates.  A straightline extrapolation was used to estimate the 1988 and
2008 populations.  When this table is compared with Table K-6 RFPU population
projections, it shows that ODUC and OEPA (adjusted) estimates are higher  than
the RFPU.  The difference between the two projections is less than 2 percent
                                      K-6

-------
which is within an acceptable range for statistical error.  This sets the
service area population at a high of 1,020,000 and a low of 950,400 in 2008.
This figure is at most 23,000 individuals greater than the revised service
area projections shown in Table K.-6.  These population figures will serve as a
baseline for estimating the growth that would be likely to occur without the
construction of the interceptors.

Revision of RFPU Projections
     The RFPU used the Mid-Ohio Regional Planning Commission's (MORPC) traffic
zone population projections as the initial data base for developing the
overall and subservice area population projections.  MORPC's traffic zone
system is based on a network of roadway intersections developed in the 1960's
and was updated between 1974 and 1980.  This network is based on land use and
traffic patterns and is able to predict population projections, changes in
land use and transportation needs.  These projections are a disaggregation of
the ODUC projections for Franklin County.  As a result of ODUC's 1985
revisions,  MORPC is revising its projections.  MORPC will increase its 2010
population projection from 941,341 to 1,027,341 (1,026,000 is the ODUC
estimate) for Franklin County.  This increases the forecast population in
Franklin County by 86,000 individuals by 2010.
                                     K-7

-------
              TABLE K-4.  POPULATION DISTRIBUTION BY COUNTY FOR
                          THE COLUMBUS SERVICE AREA
County/Service Area
Franklin
Total Population
Service Area Population
% of total
Fairfield
Total Population
Service Area Population
% of Total
Delaware
Total Population
Service Area Population
% of Total
Licking
Total Population
Service Area Population
% of Total
1980
812,670
810,351
99.7%
82,401
595
.77,
54,779
1,616
3.0%
125,943
3,253
2.6%
1985
829,523
826,868
99.7%
87,972
642
.7%
62,320
1,928
3.1%
137,648
3,722
2.7%
2000
906,903
902,816
99.5%
106,180
796
.7%
87,810
3,131
3.6%
162,791
5,206
3.2%
Source:  OEPA, PEMSO Projections, February 1982.
                                     K-8

-------
          TABLE K-5.   COMPARISON OF POPULATION PROJECTIONS BY COUNTY
                         FOR THE  COLUMBUS  SERVICE  AREA
Source
ODUC
OEPA
OEPA (adj.)
1988
925,900
848,600
902,200
Increase
Between
2000 2008 2000-2008
982,600 1,018,000 35,400
911,947 950,347 38,400
976,130 1,052,900 41,000
Source:  Interpolation of Tables K-l, K-2, K.-3, K-4.
                                     K-9

-------
        TABLE K-6.  REVISED FACILITY PLAN UPDATE POPULATION FORECASTS
                                              Service Population
Service Area                       1980       1988      2000      2015
Jackson Pike (1980 Bdry.)
Southerly (1980 Bdry.)
West Scioto (a)
Big Run
Darby Creek
Grove City
Minerva Park
Sunbury-Galena
Big Walnut
Blacklick
Groveport
Rickenbacker AFB
Rocky Fork
TOTAL
NOTES:
(a) A significant portion of
467,153
324,336
(b)
(b)
(b)
15,941
(b)
(c)
(b)
(b)
(b)
(b)
(b)
807,430

the Upper
487,644
336,633
(b)
(b)
(b)
16,601
2,063
(c)
(b)
21,904
3,436
2,146
(b)
870,427

Scioto West
500,294
360,834
31,072
(b)
(b)
17,490
2,187
(c)
(b)
31,034
3,499
2,146
3,305
951,861

Interceptor
511,035
372,344
42,564
(b)
(b)
22,571
2,265
(c)
(b)
35,091
3,542
2,146
3,601
995,159

presently is
     served by temporary pump stations and force mains that discharge to the
     Upper Scioto East Interceptor.  This service area population is included
     in the 1980 and 1988 service population of Jackson Pike.  By the year
     2000, this service population is deducted from the Jackson Pike service
     area and allocated to the West Scioto service area, reflecting
     construction of the Upper Scioto West Interceptor Sewer.

(b)  Area not served during projection period.

(c)  Area excluded from analysis.  No service planned.

Source:  URS Dalton 1986.
                                     K-10

-------
                APPENDIX  L
                   DRAFT
                CRITIQUE OF
       WATER QUALITY MODELING ISSUES
          FOR THE COLUMBUS, OHIO
SUPPLEMENTAL ENVIRONMENTAL IMPACT STATEMENT

-------
                        Draft


                    Critique of
         Water Quality Modeling Issues
             for the Columbus, Ohio
Supplemental Environmental  Impact Statement
                   August 31, 1987
                     Submitted To:
        U.S. Environmental Protection Agency, Region V
                230 South Dearborn Street
                 Chicago, Illinois 60604
                     Submitted By:
         Science Applications International Corporation
                  8400 Westpark Drive
                 McLean, Virginia 22102
           EPA Contract No. 68-04-5035; D.O. #040
             SAIC Project No. 2-813-06-193-40
                         L-1

-------
                                     1. Introduction
   The Columbus Supplemental EIS (SEIS) is being prepared to evaluate the current facilities
planning information for the City of Columbus. Key provisions of the current facilities plan for
Columbus include:

      o  Upgrading and expansion of the Southerly wastewater treatment plant (WWTP)
      o  Phase-out and ultimate abandonment of the Jackson Pike WWTP, and
      o  Re-routing of Jackson Pike flows to the Southerly WWTP.

Future facilities planning activities will address the issue of CSO control at the Whittier Street
storm tanks overflow. The two wastewater treatment plants operated by the City of Columbus
(Jackson Pike and Southerly) are projected to discharge almost"! 80 mgd of treated effluent to
the Scioto River by the year 2008.

   A simplified graphic of the locations of point source discharges and riverine features in the
Columbus area is provided below.
                                       River Mile 132
                            Whittier Street Pumping Station
                                            River Mile 129
                                       Jackson Pike POTV
                                             River Mile 127


                                          Southerly POTW
                                              River Mile 1 1 8


                                              River Mile 1 17
                                             L-2

-------
   Facilities planning activities for the City of Columbus, including future (but as yet
unspecified) modifications for CSO control at Whittier Street, involve decisions which will
directly impact water quality in the Scioto River. These decisions include:

      o  Locations of effluent discharge outfalls
      o  Quantity of treated effluent released from each outfall, and
      o  Extent of wastewater treatment prior to discharge.

Due to the Federal government's participation in the proposed project (through the USEPA §201
grant), and the potential for the proposed project to result in significant impacts on the
environment, an environmental review is required.  This review is necessitated in the USEPA
procedures for implementation of NEPA.

   Results of current and future  facilities planning activities will directly affect the three
major point sources (Whittier Street, Jackson  Pike and Southerly) which presently influence
water quality in a 30 mile stretch of the Scioto River, in the Columbus area.  Because of this
direct relationship between facilities planning decisions and the quality of the aquatic
environment, it is essential that these aspects of project impacts be carefully considered in the
SEIS.

   The most common approach to evaluating the water quality impacts related to a WWTP
effluent discharge is through application of a water quality model. The USEPA relies on models
to determine the need for upgrading wastewater treatment plants beyond secondary, and whether
Federal grant monies may be  used for such purposes. As an evaluative tool, the model provides a
mathematical simulation of the naturally-occuring physical/chemical processes which
biodegrade, or assimilate, wastes in the receiving water. Through such mathematical
representations, models assist managers in determining whether proposed improvements in
wastewater treatment processes will provide significant benefits to the water quality of the
receiving waters.

   Typically, initial  model development is followed by a process of model calibration and
verification, with site specific field data, to ensure that the model is faithfully characterizing
and reflecting natural conditions. At this point, individual variables (such as effluent quantity,
quality or total wasteload) can be selectively modified to predict and evaluate the impacts
(positive or negative) on water quality in the receiving water.  Permitted effluent  limits can
then be established for the discharger. Such limits reflect a quantification of the total excess
wasteload assimilative capacity of the receiving water which is allocated to the subject
discharger (generally, the total available assimilative capacity is not allocated to a single
discharger).

   As an aid to the establishment of wasteload allocations (WLA) and related permit discharge
limits, a water quality model was initially developed for the Scioto River, in the Columbus
vicinity, by the Ohio EPA (OEPA), using QUAL2 (a commonly used, reliable  framework).  Based
on this model, the OEPA proposed permit limits for the Jackson Pike and Southerly POTWs, in
the Comprehensive Water Quality Report (CWQR).

   The original QUAL2 model was later updated, by a consultant to the City (URS Dalton), and


                                             L-3

-------
transformed to a QUAL2E format, which is operable with PC hardware.  The updated model was
then used to derive alternate wasteload allocations and discharge limits, which were proposed by
the City to OEPA.  These alternate allocations and limits were accepted by OEPA and sent to USEPA
in an amended CWQR.  To date, the amended CWQR has not been approved by the USEPA, however
the discharge limits have been approved. The discharge limits in the amended CWQR are the
basis for the current facilities planning efforts and a key component in the SE1S evaluations.

   Therefore, in order to determine if the proposed project will significantly impact the quality
of the natural environment,  it is necessary to determine the accuracy of the wasteload
allocations and resultant discharge limits. This determination is made through examination of
the reasonableness of variables and assumptions on which the model was constructed, and
through assessment of the reliability of these variables and assumptions to represent natural
conditions.

   Preliminary evaluations conducted as part of the SEIS have questioned a variety of the
variables and assumptions  used in the QUAL2E model and resultant discharge limits. Although
these questions are currently unresolved, they include technical input variables to which the
model is especially sensitive. Collectively, these questions seriously undermine the reliability
of the current model.

   In the following sections, the questioned model input variables and assumptions are identified
and discussed.
                                             L-4

-------
                                  2. Technical Issues

   This section includes an identification of specific model input variables and assumptions
whose reliability is considered questionable, as a result of preliminary SEIS review efforts.

2.1    Existing Modeling.

   Two attempts were made at developing a water quality model for the Scioto River near
Columbus, Ohio. The first model was developed by the OEPA using the computer program QUAL2
(SEMCOG). This model was calibrated with water quality data collected during a July, 1982
"intensive survey", and verified with similar data collected in August, 1981. The second
modeling effort was conducted by the City of Columbus and its consultant, URS-Dalton. An
updated version of QUAL2, QUAL2E (enhanced), was used in the second effort.

   The major difference between the two efforts, other than the computer program employed,
was that URS incorporated a term to account for the production of oxygen by the  benthic
(attached) algae that staff members observed growing in the river during a field reconnaisance
survey in September, 1985. The OEPA had not included the effects of benthic algae or
phytoplankton in its earlier (QUAL2) model of the Scioto.

   Two major problems are associated with the existing water quality modeling.  The first
problem is that steady state modeling frameworks (ie; QUAL2  and QUAL2E) were applied to
stream conditions that were essentially not at steady state. The second major problem stems
from the use of a benthic photosynthesis oxygen production term. In constructing both models,
rate constants were derived through analysis of field data on physical/chemical parameters in
the river. However, comparison of the field data and the resulting calibration and  verification
plots has indicated that acceptable fits to DO data were not obtained in either study.

   Other problems include  inappropriate or incomplete consideration of ammonia data, other
nitrogen species, phytoplankton, and cross-sectional profiles.

2.2   Steady State Modeling Framework.

   Modeling frameworks such as QUAL2 and QUAL2E are normally applied to situations in which
none of the state variables (the concentrations of DO and other water quality constituents) or
"forcing functions" (effluent and boundary BOD, nitrogen loadings, etc.) vary at any given
location with respect to time, i.e., when the system is at "steady state". However, it is often
acceptable to apply steady  state models to certain non-steady state situations.  For example,
steady state models are often used to model estuaries. However, the model is constructed with
data gathered at high or low slack tides, and is therefore tidally-averaged.  In this  way, any
error introduced by using a steady state framework to model a dynamic, periodically varying
estuary is reduced.

   Similarly, in streams, when a state variable such as DO varies at specific locations due to
such dynamic processes as photosynthesis, error is introduced into the output of a steady state
model. If such time variation in the inputs or state variables occurs,  and steady state models are
the only practical tools available, care must be taken to consider the effects of any time variable


                                             L-5

-------
factors in order to minimize the model error resulting from the use of the steady state
framework. As in the preceding estuarine example, error minimization is usually accomplished
by time-averaging the data over the period of concern (e.g., periods of darkness, periods of
light).

   Non-steady state conditions occurred in the Scioto River near Columbus during the July,
1982 intensive survey. Streamflows were observed to decline steadily during this survey,
reflecting the effects of a moderate rainfall event which had occurred just prior to the start of
sampling. Therefore, background BOD and NOD loadings, and velocities and depths throughout the
system, were also declining.  More significantly, a diurnal DO variation of greater than 2 mg/L
was measured at several of the sampling stations (as indicated in Figure 6-7 of the Central
Scioto River Mainstem CWQR). At several of the most downstream stations sampled,
supersaturated DO conditions occurred during the early evening hours. Thus, true steady state
conditions were not realized in the Scioto during the July intensive survey.

   Steady state model error due to non-steady state effects, such as variable waste loading, can
be reduced by considering individual "plugs", or parcels, of water. In plug flow sampling,
unique parcels of water are followed and sampled as they move downstream, at intervals
according to the expected time of travel (determined from dye studies conducted concurrently, or
at similar streamflow). Each plug is then treated as a separate water quality sampling run,
from which a predicted profile can be generated using the corresponding inputs. However, no
data have been collected from the Scioto River from which such a plug flow model can be
developed.

   No steady state model, no matter how carefully developed, will allow an accurate prediction of
the DO time series as impacted by photosynthesis. QUAL2E may be run in the dynamic mode,
which will allow the development of a model to predict the time-variable effects of
phytoplankton  (but not periphyton algae) on the instream DO and nutrient profiles. While
inputs and forcing functions (i.e. effluent and background Streamflows, BOD and NOD loadings,
and DO concentrations)  must remain constant, the variation  in stream DO concentrations due to
diurnal variation in algal photosynthesis can be simulated. However, QUAL2E does not have the
capability to properly simulate benthic photosynthesis which, apparently, may be quite
significant in the Scioto River. To successfully accomplish the simulation of benthic or
planktonic algae over time, detailed knowledge of the algae nutrient uptake kinetics and
light-growth relationships are required. This knowledge is preferably gained from
site-specific studies which, in the present case, are lacking for the intensive survey periods.  In
an attempt to compensate for these deficiencies, literature information would have to be used as
initial values for most of the parameters.

    In both existing versions of the model, all of the observed DO data points for the four-day
survey were used in calibrating the model. However, these DO values were taken from samples
collected in both the morning and afternoon hours.  Simultaneously calibrating to both morning
and afternoon  DO observations resulted in the underprediction of the afternoon, and
overprediction of the morning, DO profiles. Only 11 of 50 observations were from the morning,
and none were from before 9:30 am. Since most of the DO values used for calibration were from
the afternoon hours, when DO will be at its highest level of the day, the resulting predicted DO
profile was skewed towards a higher level than it probably should have been. This procedure


                                             L-6

-------
disregards the night and early morning hours, when DO is usually at its lowest level. (Note that
the DO concentration measured at river mile 118.5 at 9:30 am on July 21,1982 was 3.9 mg/l,
which was the lowest value observed during the entire survey.)  In the URS model (QUAL2E),
inclusion of the benthic photosynthesis term resulted in an even higher mean DO profile than the
OEPA model (QUAL2), based on the same sampling data.

   As an alternative, it might be appropriate to segregate the water quality data with respect to
the date and time sampled, i.e. to "time-average" the data.  The use of a steady state model could
then  be better justified, perhaps by calibrating separately to morning (low DO) and afternoon
(high DO) observations, or by assigning weights to each observation so that a more realistic
picture of daily average water quality values could be obtained.  Actual mean parameter values
(e.g., Kd, Kn, etc.) could be better estimated in this way, and more accurate predicted profiles
could subsequently be generated.

2.3   Benthic Photosythesis Term in QUAL2E.

   The QUAL2E model is an improvement over the earlier QUAL2 model  in that it recognizes the
need to include the effects of algal periphyton on DO in the Scioto River. It is apparent from the
URS data that, at times, these attached algae can significantly impact the  observed DO profile in
the river. However, the URS model incorporates a negative sediment oxygen demand (ie; benthic
oxygen production).  There are numerous pitfalls associated with the use of a negative sediment
oxygen demand (SOD) term in the URS model.

   URS' experiment was conducted over a two day period beginning September 25,1985.
During this  period, it was generally sunny, but periods of clouds and rain occurred on the 26th.
During the periods of cloud cover on the 26th, a net consumption of oxygen was measured in the
DO chambers, which would be expected.  During the sunny periods, a net production of oxygen
was  observed. By plotting the change in DO in the chambers and bottles over time, URS derived
slopes, in mg/Umin, of the oxygen depletion curves. It is not stated whether these slopes
represent averages over the entire experimental period, or instantaneous maxima. However,
only  the results of the experiments on September 25, when a net production of oxygen was
occurring, were used to calculate the "overall" net 24 hour SOD of -1.74 g/m2/day (the
negative sign implies a net production of oxygen).

   The applicability of this SOD rate to a model calibrated with data collected three years prior
to these experiments must also be questioned.  Stream conditions, such as substrate composition
and   nutrient availability to adequately support benthic algae growth, can change in three
years, especially in a relatively small, wastewater-dominated stream such  as the Scioto.  Also,
the sunlight conditions that greatly influence the rate of oxygen production were much different
during the July 1982 study period. URS reports that sky cover ranged from 0% to 100%
during the survey period. However their SOD rate was derived from an experiment conducted
only  during bright sunshine. In addition, the experiments were conducted  at only one station,
located between the Jackson Pike and Southerly discharges.  This rate was applied to all of
reaches in the model, except in those reaches where the predicted DO greatly exceeded
observed values. In those cases, the SOD term was arbitrarily removed in order to "fit" the
observed DO data.
                                            L-7

-------
   Finally, there is also the difficulty in determining a "design condition" SOD rate for use in a
WLA model. No precedent or EPA guidance exists in applying a net benthic oxygen production
term for use in a WLA model.  Before using a term in the model which has little or no literature
justification, and which appears to have been used mainly to better fit the data representing
higher DO concentrations, additional data collection should be required to support its use.

2.4   Other Significant Problems.

   In addition to the major technical problems discussed in the preceding sections, several
additional problems are noted.  These additional problems are discussed in the following.

2.4.1     Phytoplankton Influence on DO Profile.

   Neither  model  accounted for the influence of phytoplankton on the DO profile  of the river.
This is especially apparent in the lower reaches of the Scioto (below river mile 109) where
significant populations of phytoplankton apparently caused supersaturated DO conditions. In
addition, elevated ultimate CBOD values were also observed beginning at river mile (RM) 109.
Chlorophyll a samples taken in September, 1982 during a diurnal DO study conducted by OEPA
indicate that this section of the river is probably impacted by an active phytoplankton
population.  (In bottle BOD tests, the presence of algae in  the samples can increase the  measured
ultimate BOD considerably over that which is traceable directly to wastewaters.)  However, the
OEPA CWQR mentions  that "algal simulations were  not performed. With the information
available, the QUAL-II model could not be accurately calibrated to the Scioto River."

   URS attempted to incorporate the effects of phytoplankton in their WLA analysis, and a
sensitivity analysis of the effects of phytoplankton on the DO profile under design wasteflow
conditions was conducted.  The analysis showed that an increase in DO of only 0.14 mg/L would
be expected if the phytoplankton population were to increase from 0 to 100 ug/L. This
contradicts the July and  September, 1982 data presented by OEPA, where significant increases
in afternoon DO were observed at the most downstream stations, correlated with high
chlorophyll a levels. URS provides no information in the report concerning values for algal
kinetics or cell stoichiometry parameters used in the  sensitivity analysis.

2.4.2     Nitrogen Species.

   The QUAL2E model does not appear to be successfully calibrated for NH3-N and NO2+NO3-N,
and organic N was not modeled.  Figures 6-5  and 6-6 of the CWQR appear  to indicate an
erroneous value for the nitrification rate coefficient. Observations for both ammonia and
nitrite-nitrate nitrogen are generally underpredicted.  This carries over, although to a
somewhat lesser  extent, to  the verification profiles for these variables given in Figures 6-9 and
6-10 of the CWQR. This could stem from not accounting for the  effects of algal uptake on
nutrients in the model.

   Organic nitrogen was not considered in either model. Organic N can hydrolyze to produce
ammonia, which can then be taken up by algae or oxidized by nitrifying bacteria. Inclusion of all
of the nitrogen species,  as well as the effect of algal  uptake, may result in a closer
correspondence between observed and predicted values.


                                             L-8

-------
2.4.3     Cross Sectional Profiles.

   Cross sectional profiles provided by the OEPA do not indicate significant variation in cross
sectional area or depth.  OEPA established 25 reaches based on times-of-travel, river cross
sections and flows observed during studies conducted in 1980-1981. Each of these reaches is
characterized by a unique power function for flow vs. velocity and flow vs. depth. However, none
of the information required to assess the flow-velocity or flow-depth power equations presented
in Table 6-2 (flows, times-of-travel, depths for each reach) is given in the report.

   Flow vs. velocity and flow vs. depth relationships affect the model's internal calculation of
reaeration, and the rate of transport of pollutants through the  system. Therefore, it is critical to
properly define these relationships to correctly predict the DO response to changes in flow,
especially when determining the WLA. However, predicted stream depths developed in the
QUAL2E model do not appear to correlate with actual field data. For example, if a flow of 150  cfs
is used, the depth equations for reaches 2 and 4 yield depths of 2.8 and 1.8 feet, while the
equation for reach 3 yields 0.7 feet. In contrast, based on observation of the profiles submitted
by OEPA, there do not appear to be any locations that were sampled that have a mean depth of less
than two feet.

   Since the shapes of the cross sectional profiles appear to  be relatively uniform, it may be
more appropriate to divide the study area into fewer physical  reaches, so  that less variation in
depths is obtained.  It is accepted modeling practice not to divide a stream system into any more
reaches than  is  necessary, especially if a general physical uniformity throughout the stream is
observed.
                                             L-9

-------
                      3. Conclusions and Recommendations.

   Based on the reviews conducted to date, it does not appear that an accurate and reliable
predictive model for use in assessing current and future environmental conditions has been
developed.  This is most likely due to limitations in the available data, and to the inability of
QUAL2E to simulate diurnal DO variations in the steady state mode.  Specific conclusions and
recommendations concerning the existing water quality models are listed below.

3.1    Algal Effects on DO.

   The contribution of dissolved oxygen by algae to the stream DO balance is not usually
accounted for in determining assimilative capacity. However, an attempt should be made to
factor the effects of algae on DO into the modeling for the Scioto River, using the  September,
1982 diurnal DO data.  Any assumptions on daytime oxygen production by algae  must be balanced
against the catastrophic effects that nighttime respiration of these cells can have on DO, and
subsequently on stream biota.

   A diurnal DO study was conducted in September, 1982. Some extremely low  DO values were
observed, and most values recorded were below the 5 mg/L DO standard. However, the number
of samples and their times of collection were not reported in the Draft CWQR. The appropriate
data for developing an accurate and reliable model of algae in the Scioto for the July, 1982
survey are apparently not available. Thus, it may be difficult to improve on either modeling
effort for that period, given the existing data set. However, it may be possible to use the diurnal
data collected during September, 1982 to formulate a model of the river which includes the
effects of algae on DO. This  data set should be analyzed for its potential use in model development.

3.2   Benthic Algae/SOD.

   The effects of benthic algae on SOD should not be incorporated into the WLA model until a
more complete set of data is available.  More studies similar to the URS SOD study should be
conducted before a term describing the  benthic production of oxygen is incorporated  into the
model.  Due to the existence of the Whittier St. CSO and periodic bypasses of the Southerly
WWTP, organic solids introduced into the river during storm events may settle out in the study
area and result, at times, in a benthic oxygen demand that exceeds the production of oxygen by
benthic algae. This needs to be considered in establishing a steady state net SOD term for use in
the model.

3.3   Non-steady State Modeling.

   The feasibility of using a non-steady state modeling framework should be explored. USEPA
has developed WASP, a multi-purpose dynamic modeling framework that can be used to simulate
the effects of phytoplankton on nutrients and DO in streams as well as other types of water
bodies.  The available body of data should be examined carefully to determine whether WASP, or
any other similar framework, may be a more appropriate tool for modeling the Scioto  River
than QUAL2E.

   Although WASP is more flexible than QUAL2E, it is also more complicated and, therefore,


                                            L-10

-------
more costly to develop and utilize.

3.4   Additional Data.

   The feasibility of collecting an additional set of data should be examined. The shortcomings of
the available data upon which to build a valid water quality model for the Scioto River have been
described by the OEPA and URS.  The basic problem is that water quality samples were collected
without regard to plug flow in the system. Due to the time variability of DO in the Scioto during
low flows, the assumption of steady state (which is crucial to the successful utilization of
QUAL2E as the modeling framework) is invalidated.

   Also, the depth and velocity vs. flow relationships were developed for flows that may be
exceeded as a result of plant expansion in the future.  New information collected at higher flows
in the Scioto would decrease the uncertainty in the results produced by these equations.

3.5   Evaluation of Alternatives.

   Because the existing models were developed under a two-discharge scenario, these models
should not be used to evaluate the one-plant alternative without further data collection and
modeling analysis. It is likely that,  under a one-plant scenario, the water quality impacts of the
Southerly plant will extend even farther downstream during 7Q10 flow events than presently
occurs.

   The existing models extend downstream only to RM 103, near Circleville.  Beyond RM 103,
there is no information - physical, hydrologic or chemical - on which to base model
development. Furthermore, there are other large industrial discharges below RM 100 whose
effluent limits may be impaired by the downstream relocation of the DO sag likely to occur due to
the combination of flows in a one-plant scenario. This data deficiency needs to be corrected
before a reliable model of this section of the river can be constructed.
                                             L-1 1

-------
             APPENDIX  M
U.S. ENVIRONMENTAL PROTECTION AGENCY,
  WATER  QUALITY  BRANCH, MEMORANDUM
   ON COLUMBUS WATER QUALITY MODEL

-------
                   UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
                                         REGION V
    DATE:  SEP 30  1987


 SUBJECT:   Columbus, Ohio Water  Quality  Modeling
                                 i  ^^**^~^
   FROM:   Kenneth  A.  Fenn>
           Water  Quality
     TO-               /"
           Todd A.  Cayef,  Chief
           Municipal  Facilities Branch
                     i1
           In  response  to your memorandum  of September  9,  1987, regarding the
           water  quality modeling  for  Columbus, Ohio, my staff has  reviewed the
           EIS  consultant's  (SAIC)  critique  of the modeling work  performed by
           OEPA and  later work completed by  URS-Dalton  for the City of Columbus.

           As you may know,  the  original OEPA two-plant modeling  analysis for
           Jackson Pike and  Southerly  was  reviewed by the  Eastern District
           Office, the  Planning  and  Standards Section and  our Permits Section.
           These  offices found the  original  QUAL  II analysis to be  sound.
          OEPA effort  resulted  in  the  following  limits:

          Plant            River  Mile        Flow       CBODt
          Jackson  Pike

          Southerly
127

118
110 MGD

 85 MGD
5.2 mg/1

5.0 mg/1
 NH-3-N

1.3 mg/1

1.5 mg/1
           Subsequently, the City  of Columbus  employed  URS-Dalton to model the
           Scioto  River  as  a means  of  confinning the State's  analysis  and for
           exploring a one-plant alternative.   URS-Dalton employed lower  flow
           discharge projections,  and  concluded that the  following limits would
           achieve  dissolved oxygen and  ammonia water quality standards:

           Plant

           Jackson  Pike

           Southerly

           Jackson  Pike

           Southerly

           As you  can see,  the results are comparable,  with a decrease in
           ammonia  that  allows slightly  higher CBODs values.  We would also
           point out that these  limits approach those achievable with  available
           technology,

                                        M-l
Time
Pre-1992
Pre-1992
1992-
1992-2015
Flow
60 MGD
90 MGD
Dec
156 MGD
CBODs
8.0 mg/1
8.0 mg/1
o m m i s s i
8.0 mg/1
NHyN
1.0 mg/1
1.0 mg/1
o n e d
1.0 mg/1
EPA FO8M 1320-6 (REV. 3-76)

-------
                                -2-
In terms of the critique by SAIC, we agree that model  calibration
and verification would be improved with further work  on stream
hydraulics and algal kinetics.  However,  we can make  this  same
statement regarding virtually any other water quality  model  in
Region V and perhaps the rest of the Country.  Furthermore,  it does
not always follow that a better calibration of existing conditions
would necessarily improve the prediction  of future conditions.
This is because future conditions will  be dramatically different
due largely to changes in hydraulics and  the control  of both point
and nonpoint sources of pollution.  For these reasons, professional
judgment is an overriding factor in developing and applying  a water
quality model.  The current model may have an error margin of
+1.0 mg/1 of dissolved oxygen.  Given the complexity  of the  Scioto
River in this area, we are not convinced  that future  modeling work
will either significantly reduce this error or significantly revise
the current effluent limitations.

We agree with the Environmental  Review  Branch that further modeling
of the one-plant vs. two-plant alternative seems counterproductive.
This is because reasonable estimates of the one vs. two-plant
approach are available to your staff.  We have endorsed the  two-plant
analysis developed by URS-Dalton which  is the basis for the  current
permit limits at each facility.

We also agree that OEPA may be able to  provide additional  professional
judgment if that is deemed necessary for the purpose  of the  EIS.
                           M-2

-------
            APPENDIX  N
THE INFRASTRUCTURE PROJECT 1985-1986
  FINAL REPORT:  EXECUTIVE SUMMARY

-------
           m
  DEVELOPMENT COMMITTEE FOR GREATER COLUMBUS
THE INFRASTRUCTURE PROJECT
          1985-1986
     FINAL REPORT
         5 DECEMBER 1986
              N-l

-------
y
                                                              "While the space directly beneath  a

                                                           building contains the systems required to

                                                           support its structure, the area under the


                                                           surface of the streets and sidewalks is filled

                                                           with the systems essential to support its


                                                           occupants. The basic systems, which we


                                                           call utilities, include water, sewage removal


                                                           and drainage, electricity, steam, gas, and

                                                           telephone communication."

                                                                         The  quotation  above  & the
                                                                        cover  illustration are  from
                                                                        UNDERGROUND  by  David
                                                                        Macaulay.  Copyright  ©  1976
                                                                        by  David  Macaulay.  Used by
                                                                        permission of Houghton Mlfflin
                                                                        Company.



                                                             ' 'The nation's infrastructure:  The physical

                                                           framework that supports & sustains virtually

                                                           all economic activity."
                                                                          Definition by the National
                                                                        Council on  Public  Works
                                                                        Improvement.
                                            N-2

-------
                       CONTENTS
Extracts from the "GREATER COLUMBUS INFRASTRUCTURE INVESTMENT
REQUIREMENTS AND FINANCING STRATEGY:  THE NEXT FIVE YEARS"
(the Final Report of the DCGC 1985-1986 Infrastructure Project)

                                        Pages
            EXECUTIVE SUMMARY          v to viii
            Section VI. NEXT STEPS:
            IMPLEMENTATION             104 - 107
INFRASTRUCTURE MAPPING & INFORMATION SYSTEM
NOTE:   The Final Report of the DCGC 1985-1986 Infrastructure
Project will be distributed to all agencies which were
participants in the Project.
                               N-3

-------
                            EXECUTIVE  SUMMARY
     The Development Committee for Greater Columbus undertook a major
study to assess the area's infrastructure condition and to develop
strategies for keeping its capital facilities well managed and
maintained.  This report represents the final product of two earlier
reports written by The Urban Institute in February and September 1986.
The first report, Greater Columbus Infrastructure Investment
Requirements (Feb. 1986), evaluates capital facility performance and
determines capital investment requirements and funding availability over
the next five years for area roads, bridges, water, and sewer systems.
The second report, Financing Greater Columbus's Infrastructure (Sep.
1986), provides a detailed analysis of the area's options for financing
its capital program.  The present report represents a final statement of
this Greater Columbus Investment Strategy.

GREATER COLUMBUS'S CAPITAL PLANT

     o   Two indicators of performance — a street maintenance
         effectiveness index and resurfacing cycles — suggest that
         several jurisdictions are falling behind in road repair.  The
         City of Columbus, in particular, falls below a sampl*1 of other
         large cities in road performance, reflecting a fluctuating
         program of maintenance and repair.

     o   Jurisdictions with high ratings generally show short
         resurfacing cycles and more systematic programs of street
         resurfacing.

     o   Based on several performance measures, bridges in  the county
         are generally in good condition.  Only 6 percent are
         structurally deficient, the potentially most serious bridge
         problem.  The older structures that fall largely under county
         responsibility, are in the poorest condition.

     o   Area water supply is adequate to the year 2000, provided
         additional sources are identified beginning in 1991.,

     o   Area water distribution systems show low main breaks relative
         to other cities.  The level of unaccounted-for-water, at  20
         percent, is higher than average, but it is not out of line with
         older cities.

     o   The major performance problem facing the Greater Columbus area
         in sanitary sewers is the need to upgrade the City of Columbus'
         sewage treatment plants to meet EPA requirements.
                                  N-4

-------
     o   Condition of area collection systems appears to be adequate.
         The City of Columbus and its suburbs fare well relative to
         other cities with respect to the number of main breaks and
         sewer line back-ups.

     o   As a relatively new area of infrastructure concern,  information
         is not readily available on the performance of area storm sewer
         facilities.

GREATER COLUMBUS'S CAPITAL INVESTMENT PATTERNS

     o   The majority o-f area improvements are slated for expansion (46
         percent) and upgrading (43 percent) of capital facilities.
         Only 11 percent of total investment dollars are targeted for
         rehabilitation of existing facilities.  The City of Columbus,
         in particular, should consider a more balanced division of
         resources to insure that existing facilities will be kept in
         good repair.

     o   Approximately half of projected investment requirements over
         the next five years can be met from available federal, state,
         and local resources.

     o   The City of Columbus is responsible for nearly half of the
         funding shortfall, not surprising in view of the city's major
         role as provider of area highway services, water supply, and
         sewage treatment.  Nearly two-thirds of suburban investment
         projects, however, are also unfunded.

     o   The area shows large projected investments of $454 million over
         the next five years for roads, but only 18 percent is slated
         for rehabilitation.

     o   Planned investment requirements for bridges over the next five
         years are small relative to other infrastructure areas.  The
         majority are for rehabilitation and upgrading.

     o   The City of Columbus system accounts for 90 percent of total
         planned water investments.  The majority of city investments
         are for supply improvements and system upgrading.  The majority
         of suburban needs are for rehabilitation and upgrading.

     o   Planned sanitary sewer investments over the next five years
         represent the second largest spending area.  The majority of
         improvements are for upgrading, to meet Environmental
         Protection Agency requirements, and for expansion.  Only a
         small fraction of planned spending is for rehabilitation.

     o   Planned storm sewer investments over the next five years are
         the smallest of all of the infrastructure areas.  However,
         since several studies are only now underway that could lead  to
         a higher needs estimate.


                                 N-5

-------
THE COST OF RENEWING GREATER COLUMBUS'S CAPITAL FACILITIES:  REASSESSING
THE AREA'S NEEDS

     'The Development Committee for Greater Columbus (DCGC) and the Mid-
Ohio Regional Planning Commission (MORPC) with guidance and input from
local area officials prioritized an initial list of capital projects
according to several criteria such as funding availability, health and
safety standards, and impact on the local community.  The effort
produced a list of priority projects and an estimate of the funding
shortfall expected for 1987-1991.

     o   Investment requirements across all infrastructure areas and
         jurisdictions were reduced from $1.05 billion (1986-1990) to
         $946 million (1987-91).  The funding shortfall declined by 9
         percent, from $500 million to $457 million.

     o   Funding shortfall as a percentage of total planned capital
         investment is 48 percent.

     o   Sanitary sewer projects represent 39 percent of the total
         investment; storm sewers 2 percent.  Capital spending for water
         systems are 8 percent of the total.  Road and bridge
         improvement expenditures comprise 47 percent and 5 percent of
         total investments, respectively.

     o   Sanitary sewers comprise about 35 percent of the total
         shortfall.  Road funding shortfalls account for 54 percent.

         Funding shortfalls in the water area amount to 2 percent;
         bridges 7 percent of the total shortfall.

     o   The shortfall as a percent of total investment requirements in
         each infrastructure area is greatest in bridges, at 67%,
         followed by: roads, 567.', sanitary sewers, 44%; storm sewers,
         31%; and water, at 11%.

     o   The City of Columbus comprises the bulk of total area
         shortfalls in roads, bridges, and sanitary sewers.  The growth
         suburbs account for most of area shortfalls in water systems
         and storm sewers.

FINANCING CAPITAL REQUIREMENTS

     o   Proposed financing mechanisms to support infrastructure
         requirements should be consistent with accepted public finance
         practice:  large scale investments should be debt-financed;
         debt issuance should be by jurisdictions with the greatest
         overall responsibility for area infrastructure; improvements
         should be paid for by those who directly benefit from the
         improvements.
                                  N-6

-------
o   The bulk of area road and bridge requirements could be
    supported with modest increases in user charges: $10 increases
    in license tag fees; a 2C per gallon local fuel tax; an
    extension of the county sales tax to gasoline.

o   Other road and bridge funds required could be raised through
    developer contributions, so that the increased capacity needed
    to service growth is supported by those who create added
    demand.

o   The Columbus area appears to have a strong claim on increased
    ODOT discretionary funds, which historically represent a
    smaller share than total road mileage responsibility.

o   Most jurisdictions with water and sewer funding shortfalls
    could support the required investment with increases over
    current rates.  The remaining jurisdictions likely will have to
    partially support investment requirements through general fund
    support.

o   The creation of a storm-water management district represents
    the best avenue for handling the area's flood and drainage
    investment requirements.

o   Institution of a comprehensive, automated, infrastructure
    mapping and information system, including inventory, condition,
    and investment data across jurisdictions and sectors, would
    encourage better infrastructure management, and ensure reduced
    long-range capital requirements through improved maintainance
    programming for existing infrastructure.
                            N-7

-------
                     VI.  NEXT STEPS:  MPLEMEHIATIOH








     The development of a comprehensive analysis of Greater Columbus's




 infrastructure and financing opportunities represents the first step in



 a capital stock investment and management strategy.  The main objective



 lies ahead: translating this strategy into concrete action.  This will




 require a coordinated effort to educate the public as to its importance,




 secure authorization from the General Assembly for key steps in the




 financing plan, cement, local cooperation regarding the choice and




 financing of capital priorities, and ensure the collection and automated



 storage of the information needed to wisely choose among capital




 projects, and between capital and maintenance expenditures.






Taxpayer/Voter Approval




     The financing plan contemplated in this report places on local




 highway and utility service users the responsibility for financing




 improvements to these same services.  In the area of roads and bridges,




increased fees and charges can support general obligation borrowing by




 the County to support area-wide investment.  This requires authorization




of general obligation bond issues by the Franklin County Commissioners,




and approval by county voters.




     The rate increases required to fund utility system investments do




not require general voter approval.  However, city  councils will  have to




support increases of the needed magnitudes to cover borrowing




requirements, or authorize the expenditure of general fund revenues to




meet extraordinary investment costs.  In contrast to the multi-




 jurisdictional approach needed to address area road and bridge




                                  N-8

-------
 Investment, increased funding for water and sewer investment is a




 decision  for each system, in each community.






 State Legislative Change




     The  first action required of the state is to modify the current




 limitation on the county vehicle license fee.  Instead of the current $5




 ceiling,  counties should be allowed to increase the license fee, either



 by a designated amount or according to locally perceived needs.




     State action also will be required to permit local imposition of




 fuel taxes.  Since the 2c increase in the gasoline tax is the major




 source of planned new revenues to support road and bridge investment,




 this approach should have a local priority.  In addition, the decline in




 fuel prices offers a window of opportunity to impose an a^itional  fee




 at a time when the Impact on consumers will be minimal.  Similarly,




 State approval will be required to permit extension of the  county sales




 tax to gasoline.  As this would not represent an  increase in the  tax,




 but an increase in the taxing base, resistance to this approach should




 be somewhat muted.
Local Goveraeat Cooperation



     Area governments already have demonstrated  willingness  to cooperate



in a coordinated infrastructure renewal  effort by  participating in the



DCGC's Greater Columbus Community Capital  Investment  Strategy effort.



However, many specific project priorities  remain to be negotiated, a



process requiring continuing good will and cooperation among all



governments in the County*
                               N-9

-------
      A  cooperative agreement will have to be negotiated between the City




 of  Columbus and Franklin County  to determine the specific road and




 bridge  projects to be financed from a County bond issue, and the



 sequencing of  repair work.  In addition, suburban jurisdictions will



 have  to  reach  their own accomodation with the County in slating local



 road  projects  for renewal.  The  DCGC has a vital role to play in




 sustaining the areawide cooperation that has developed during this first



 project  effort.




      The creation of a storm-water management district represents the




 best  way to finance areawide flood and drainage improvements.  The




 structure of such a district remains to be worked out among the




 prospective participants, for example, the rights and terms of entry and




withdrawal, the allocation of investment, and the type and level of




 service  charges.  Though difficult, this process will result in a secure




mechanism for storm-water funding.






Improved Coordination and Manageaent of Capital and Maintenance Spending




     The overall Community Capital Investment Strategy effort till now




has focused primarily on capital investment needs and funding




requirements.  The immediate thrust of this effort is to  remedy any




investment backlogs, and ensure  that the facilities needed to accomodate




new population and economic growth are in place.  However, the Columbus




area faces a unique opportunity  to reduce long-run capital investment




requirements.  The complex interrelationship between ongoing maintenance




spending and capital improvements requirements is long-recognized but




not always considered as a basis for action.  By acting now to improve




local infrastructure management, area jurisdictions can ensure that  the



                                 N-10

-------
existing, and planned, capital stock, is adequately maintained, thus




forestalling more costly future investments in infrastructure renewal.




The creation of a comprehensive, multi-jurisdictional, automated




geographic information and mapping system would be an important, indeed



critical, step in improved capital planning.




     Such a system would store and display inventory and condition




information, and combine water and sewer distribution system




information, street and bridge information, including traffic data,




zoning and land use data, demographic and economic information, and




virtually aay other information to allow an assessment of service demand




for any infrastructure link.  With the addition of maintenance and




repair history data, area infrastructure managers can plan for cost-




effective maintenance investment, to ensure the longest useful life of




any capital asset.  In addition, capital Investments across sectors can




be coordinated to ensure cost-effective repair and minimal disruption;




for example, through the sequencing of street repairs and water line




replacement.



     The cost recovery period of an investment in infrastructure




management systems can be quite short.  The City of San Jose  estimated




their system investment at $3.3 million, with annual  operating  costs of




$705,000.  The annual benefits expected were $995,000 for avoidance of




higher future replacement costs; $1,020,000 for increased maintenance




productivity; and $75,000 for reduced  costs of tort settlements and




insurance.  The initial capital investment would be repaid  in 2.5  years.
                                N-ll

-------
            INFRASTRUCTURE MAPPING & INFORMATION SYSTEM
     A  timely recommendation by the Urban Institute is for the
 "institution of a comprehensive, automated infrastructure mapping
 and  information system..."  The recommendation is considered
 timely  due to a great amount of interest locally and activity
 nationally in such systems.  Variously known as a "Geographic
 Information System"  (CIS in Chattanooga, Tennessee), a "Mapping
 and  Geographic Infrastructure System" (IMAGIS in Indianapolis),
 "Mapping Oriented Information System" (MOIS at American Electric
 Power), "Automated Mapping/Facilities Management" (AM/FM for the
 U.S. Air Force, Consolidated Gas Transmission Corporation of
 Clarksburg, West Virginia,...and others), or some combination,
 such as AM/FM-GIS, in Seattle, the systems are basically similar.
 For  sake of simplicity, here and until a better acronym is
 devised - we will refer to the system as AM/FM.

     An important part of  the DCGC Infrastructure Project has
been gathering information concerning AM/FM.  We have found a
wealth of experiences available for reference as the Greater
 Columbus community investigates developing an AM/FM system.

     Many agencies admit to having come to a realizat.'jn that
 they are having difficulties and high costs in their mapping and
 information systems.

     These were well summarized by Peoples Natural Gas Company
of Pittsburgh as due to:

          "redundancy of data due to decentralized divisions;
     update delays to complete a record; difficulties in
     researching data due  to the independent maintenance of
     varying documents; inadequacies for special applications
     such as network analysis; and expensive maintenance
     costs since the effort was very labor intensive."

     They continue by stating:

          "The Peoples Natural Gas Company feels that a
     corporate Facilities  Information Management System
     with computer graphics has enormous potential.  Consi-
     dering the changing needs of the gas industry for
     extensive record keeping, mapping and design, we must
     pursue means to improve the effort."

     Statements repeated often by utility companies and metro-
politan areas are that the high costs of AM/FM systems can be
made acceptable by forming coalitions and thereby sharing the
system and costs.  Repeatedly, the experienced voices in  this
new industry of AM/FM urge adequate planning and project
definition as being crucial to the success of an AM/FM project.

                              N-12

-------
     Some experiences have been of starting over,  or regrouping,
as stated in the following concerning Seattle:

          "The City of Seattle Joint Automated Mapping Project
     began as individual efforts by four separate  City agencies.
     The City's Budget Office, recognizing their commonality,
     initiated this joint project.  The City is currently con-
     ducting a unique pilot project intended to develop cost
     estimates of conversion of the City's land and utility
     facilities data base and to provide a better  understanding
     of AM/FM-GIS system capabilities and related implementation
     procedures and processes.  The pilot project  was developed
     by the City participating agencies jointly following an
     evolutionary process.  The process of development of a
     request for proposal for a consultant to complete the pilot
     began with an agreement on goals and objectives, included
     identification of candidate applications and development of
     a pilot project approach, and concluded with a memo of
     agreement, or charter, between the agencies assuring
     commitment of adequate resources (dollar and personnel) to
     the pilot.  Key to the success of this process and to the
     success of the project itself, is the ability of project
     management to (1) maintain management commitment to the
     project, (2) continue an open communicative,  synergistic
     decision process, and (3) assure sufficient resources in
     the leadership role.  With these factors, the cooperative,
     i.e., joint, nature of the project can be maintained through
     the pilot and into an implementation decision process.
     In summary, management commitment, an open process, and
     leadership support have forged a team effort from the
     initial set of individual agency, or turf, interests."

     In conclusion, we will report a statement from the Coachella
Valley Water District (California) Deputy Chief Engineer:

          "CAD is here to stay - AM/FM is on the way I

          When?  How soon?  No one has the answer today.  The
     only statement that can be made to a certainty  is that
     AM/FM is as inevitable to the District and to all similar
     public agencies as data processing was 15 or 20  years  ago."
                              N-13

-------
        APPENDIX 0

SEIS DISTRIBUTION LIST TO
PUBLIC GROUPS AND OFFICES

-------
Federal Agencies

U.S. Department of Agriculture
U.S. Department of Commerce
U.S. Department of Defense,
  Army Corps of Engineers
U.S. Department of Housing and Urban Development
U.S. Department of Health and Human Services,
  Public Health Service
U.S. Department of the Interior,
  Fish and Wildlife Service
  National Park Service
U.S. Department of Labor
U.S. Department of Transportation,
  Coast Guard
  Federal Highway Administration
Ohio Congressional Delegation,
  U.S. Senators
  U.S. Representatives

State of Ohio

Building Industry Association of Ohio
Office of the Governor
Ohio Office of Management and Budget
State Clearinghouse
Ohio Environmental Protection Agency
Ohio Department of Natural Resources
Ohio Department of Public Health
Ohio Department of Transportation
Ohio Department of Justice
Ohio Department of Economic and Commercial Development
Ohio Department of Energy
Ohio Water Development Authority
Ohio Department of Agriculture
Ohio Federation of Soil and Water Conservation Districts
Ohio Historic Preservation Office
Ohio Attorney General
Ohio Department of Parks and Recreation
Ohio Utilities Company
                                       0-1

-------
Local

Capital Square Commission
City of Bexley
City of Gahanna
City of Grandview
City of Grove City
City of Hilliard
City of Reynoldsburg
City of Upper Arlington
City of Worthington
Clinton Area Commission
Columbus Dispatch
Columbus Health Department
Columbus Industrial Association
Delaware County Regional Planning Commission
Fairfield County Regional Planning Commission
Franklin County Farm Bureau
German Village Commission
Greater Hilltop Area Commission
Hamilton Township
Italian Village Commission
Logan-Union-Champaign Regional Planning Commission
Madison County Regional Planning Commission
Mid-Ohio Health Planning Federation
Mid-Ohio Regional Planning Commission
Near East Area Commission
Northeast Area Commission
Pickaway County Regional Planning Commission
Public Library of Columbus and Franklin County
Rickenbacher Air Force Base
South Linden Area Commission
University Area Commission
Village of Brice
Village of Canal Winchester
Village of Dublin
Village of Galena
Village of Harrisburg
Village of Johnstown
Village of New Albany
Village of New Rome
Village of Obetz
Village of Orient
Village of Pataskala
Village of Plain City
Village of Riverlea
Village of Urbancrest
Village of Valleyview
Village of West Jefferson
                                     0-2

-------
Public Interest Groups

American Association of University Women Great Lakes Basin
  Task Force
Archaeological Society of Ohio
Audubon Society of Ohio
Citizens for a Better Environment
Citizens Advisory Council
Citizens for Good Planning
Columbus Board of Realtors
Environmental Clearinghouse, Inc.
Environmental Defense Fund
Franklin County Health Department
F.U.T.U.R.E.
Future Farmers of America
Greater Cleveland Growth Association
Izaak Walton League
League of Ohio Sportsmen
League of Women Voters of Ohio
Natural Wildlife Federation
Nature Conservancy of Ohio
Ohio Academy of Sciences
Ohio Air Quality Development Authority
Ohio Biological Survey
Ohio Chamber of Commerce
Ohio Conservation Foundation
Ohio Conservation Fund
Ohio Electric Utility Institute
Ohio Environmental Council
Ohio Environmental Health Association
Ohio League of Conservation Voters
Ohio Natural Areas Council
Ohio State University
Ohio Natural Heritage Program
Ohio Sierra Club
Ohio Soil and Water Conservation Commission
Ohio Water Pollution Control Conference
Ohio Water Resources Center
Sciota Bass Anglers
Water Pollution Control Federation
Water Resources Council
Wildlife Legislative Fund

Interested Citizens

Complete list available upon request.
                                       0-3

-------