United States          Office of Marine         EPA Region 1
                Environmental Protection     and Estuarine Protection     Boston MA
                Agency             Washington DC 20460

                Water              EPA 503/4-88-001    .    September 1988



«>EPA        Bacteriological Monitoring  in


                Buttermilk Bay

-------
         BACTERIOLOGICAL  MONITORING
                IN BUTTERMILK  BAY
              George R. Heufelder
           Barnstable County Health
                      &
           Environmental Department

                     BBP-88-03
     The Buzzards Bay Project is sponsored by The
US Environmental Protection Agency and The Massachusetts
      Executive Office of Environmental Affairs

-------
                  THE BUZZARDS BAY PROJECT

                  US Environmental Protection Agency     Massachusetts Executive Office of
                  WQP-2100                     Environmental Affairs
                  John F. Kennedy Federal Building       100 Cambridge Street
                  Boston, MA 02203               Boston, MA 02202
                             FOREWORD
In 1984, Buzzards  Bay was one of four estuaries in the country
chosen to be part  of the National Estuary Program.  The Buzzards
Bay Project was  initiated in 1985 to protect water quality and
the health of  living resources in the bay by identifying resource
management problems,  investigating the causes of these problems,
and recommending actions that will protect valuable resources
from further environmental degradation.  This multi-year project,
jointly managed  by United States Environmental Protection Agency
and the Massachusetts Executive Office of Environmental Affairs,
utilizes the efforts of local, state, and federal agencies, the
academic community and local interest groups in developing a
Master Plan that will ensure an acceptable and sustainable level
of environmental quality for Buzzards Bay.

The Buzzards Bay Project is focusing on three priority problems:
closure of shellfish beds, contamination of fish and shellfish by
toxic metals and organic compounds, and high nutrient input and
the potential  pollutant effects.  By early 1990, the Buzzards Bay
Project will develop a Comprehensive Conservation and Management
Plan to address  the Project's overall objectives: to develop
recommendations  for regional water quality management that are
based on sound information, to define the regulatory and
management structure necessary to implement the recommendations,
and to educate and involve the public in formulating and
implementing these recommendations.

The Buzzards Bay Project has funded a variety of tasks that are
intended to improve our understanding of the input, fate and
effects of contaminants in coastal waters.  The Project will
identify and evaluate historic information as well as generate
new data to fill information gaps.  The results of these Project
tasks are published in this Technical Series on Buzzards Bay.

-------
This report represents the technical results of an investigation
funded by the Buzzards Bay Project.  The results and conclusions
contained herein are those of the author(s).  These conclusions
have been reviewed by competent outside reviewers and found to be
reasonable and legitimate based on the available data.  The
Management Committee of the Buzzards Bay Project accepts this
report as technically sound and complete.  The conclusions do not
necessarily represent the recommendations  of the Buzzards Bay
Project.  Final recommendations for resource management actions
will be based upon the results of  this and other investigations.
 David  Fierra,  Chairman,  Management Committee
 Environmental  Protection Agency


 Thomas Bigford
 National Oceanic and Atmospheric Administration

 Steve  Bliven
 Massachusetts Office of Coastal Zone Management

 Leigh Bridges
 Massachusetts Division of Marine Fisheries

 Jack Clarke
 Cape Cod Planning and Economic Development Commission

 Richard Delaney
 Massachusetts Office of Coastal Zone Management

 Meriel Hardin
 Massachusetts Department of Environmental Quality
 Engineering

 Dr. Russell Isaac
 Massachusetts Division of Water Pollution Control

 Dr. Susan Peterson
 President, Coalition for Buzzards Bay

 Dr. Don Phelps
 Environmental Protection Agency

 Ted Pratt
 Chairman, Buzzards  Bay Citizens Advisory Committee

 Stephen Smith
 Southeast Regional  Planning and Economic Development District

 Bruce Tripp
 Massachusetts Executive  Office of Environmental Affairs

-------
                       TABLE  OF  CONTENTS

                                                        Page

LIST OF FIGURES 	   iii

LIST OF TABLES 	    V

ACKNOWLEDGEMENTS 	    vi

EXECUTIVE SUMMARY 	    1

INTRODUCTION 	    5

STUDY AREA 	    5

APPROACH AND METHODOLOGY 	    7

RESULTS

    STORMWATER 	   12

       Impacts from stormwater discharges and
       overland runoff 	   14

       Release from protected reservoirs 	   24

       Impacts from surcharge with contaminated
       groundwater 	   27

     SEPTIC SYSTEMS 	   29

     WILDLIFE 	   33

     BOATS AND MARINAS	   39

     FRESHWATER INPUTS 	   41

     POINT DISCHARGES	   43

     FURTHER CONSIDERATIONS 	   43

USE OF ALTERNATE BACTERIAL INDICATORS 	   52

     Escherichia coli 	   53

     Clostridium perfringens 	   55

    FECAL STREPTOCOCCUS 	   60

    ENTEROCOCCUS 	   60

-------
                    TABLE OF CONTENTS (CONT.)


VIRUSES - A LITERATURE REVIEW OF PERTINENT ISSUES....   63

     Entrainment of viruses in groundwater 	   65

     Predicting the movement of viruses in
     groundwater entering Buttermilk Bay 	   70


SURVIVAL OF VIRUSES IN MARINE SYSTEMS 	   71

GENERAL SUMMARY AND CONCLUSIONS 	   73

APPENDIX I 	   75

APPENDIX II 	   82

REFERENCES 	   85
                                ii

-------
                         LIST OF FIGURES
Figure                                                   Page

   1       Buttermilk Bay Study Area 	   6

   2       Location of Stormdrains and Approximate
           Areas of Service 	   8

   3       Location of Routine Bacteriological
           Monitoring Stations 	  11

   4       Fecal Coliform Densities Observed at Selected
           Sampling Stations During a Joint FDA -DEQE
           Survey of Buttermilk Bay in July 1985 	  13

   5       Geometric Mean Fecal Coliform Densities at
           Stormdrains Entering Buttermilk  Bay  During
           1986 (Six Drains Combined) 	  15

   6       Fecal Coliform Densities Observed at Selected
           Stormdrains in Buttermilk Bay During 1986  ....  16

   7       Summary of Rainfall Events in the
           Buttermilk Bay Area During 1986  	  20

   8       Location of Groundwater Sampling Stations
           in Buttermilk Bay Area 	  28

   9       Septic System Survey Results - System I	  31

  10       Septic System Survey Results - System II 	  32

  11       Average Monthly Waterfowl Census for
           Buttermilk Bay in 1986 	  36

  12       Locations of Marinas Sampled During  1986
           Survey 	  40

  13       Fecal Coliform Die-off With Time - Results
           of Preliminary Experiments in Buttermilk
           Bay to Determine the Effect of Solar
           Radiation on Fecal Coliform 	  44

  14       Comparison of U-V Light Absorbance Among
           Selected Stations in Buttermilk  Bay  	  47

  15       Comparison of Sediment-Protected Fecal
           Coliform vs. Light Exposed Samples 	  49

  16       Growth of Fecal Coliform in the  Presence of
           Indigenous Algae at 20 C tinder Laboratory
           Conditions	'.	  51
                               111

-------
                   LIST OF FIGURES (CONTINUED)
Figure                                                   Page

  17       Relationship Between Fecal Coliform  and
           E. coli Densities in Samples Collected
           at Routine Sampling Stations in Buttermilk
           Bay During 1986 Survey	   54

  18       Relationship Between Fecal Coliform and
           Clostridium perfringens Densities in
           Samples Collected at Routine Sampling
           Stations in Buttermilk Bay During 1986
           Survey 	   57

  19       Relationship Between Fecal Coliform and
           Fecal Streptococci Densities in Samples
           Collected at Routine Sampling Stations
           in Buttermilk Bay During 1986 Survey	   59

  20       Relationship Between Fecal Coliform and
           Enterococci Densities in Samples
           Collected at Routine Sampling Stations
           in Buttermilk Bay During 1986 Survey 	   62
                                IV

-------
                         LIST OP TABLES

Table                                                     Page

   1    Overlying Water - Suspended Sediment Fecal
        Coliform Densities From Selected Sites
        in Buttermilk Bay 	       25

  2     Fecal Coliform Densities in Rainwater Percolate
        of Decaying Eelgrass Found at Selected Sites in
        Buttermilk Bay	       27

  3     Summary of Groundwater Analyses in Buttermilk
        Bay 	       33

  4     Summary of Fecal Coliform and E. coli
        Densities at Routing Sampling Stations in
        Buttermilk Bay in 1986 	       35

  5     Summary of Fecal Coliform and E. coli
        Densities from a Fish Market DTs~charge Pipe
        Entering Buttermilk Bay on Selected Dates
        in 1986	       43

  6     Summary of Clostridia perfringens Densities
        at Routine Sampling Stations in Buttermilk  Bay
        in 1986 	       56

  7     Summary of Fecal Streptococci Densities at
        Routine Sampling Stations in Buttermilk Bay
        in 1986 	       58

  8     Summary of Enterococci Densities at Routine
        Sampling Stations in Buttermilk Bay in 1986 ...       61

-------
                       ACKNOWLEDGEMENTS


     The  completion of this report required the behind-the-scenes
assistance of  a  number of individuals whose help was  greatly
appreciated.  Quality assurance and bacteriological analyses were
aided to a large degree  by the  effort of  Donna  McCaffery,  our
staff bacteriologist.  Her  technical review and  comments on  the
manuscript were also appreciated.  Administrative assistance and
support of Stetson Hall,  County  Health Officer, and  Norma Jean
Peck  provided  for an  efficiently  run   field  investigation.
Laboratory Analysts  Sue Williams,  Susan Rask and Laurie Canning
are  thanked  for their  help during the often-unfortuitous
occurrences  of rain  events.

    Joseph Costa of Boston University  Marine Program is thanked
for his computer assistance.  Craig Fish of Boston University,
Geology Department  is  thanked  for providing water parcel
trajectory maps which helped in  understanding the possible fate
of contaminants from stormwater discharge pipes.  A  special
thanks is extended to the staff of the Southeast Region of the
Massachusetts Department  of Environmental  Engineering, Shellfish
Sanitation  Branch  for  their  assistance in  obtaining  and
interpreting historical bacteriological data from the study area
and technical assistance throughout the study.
                              vi

-------
                       EXECUTIVE SUMMARY

   Water quality in Buttermilk  Bay,  southeastern  Massachusetts
was historically  perceived as very good until  March,  1984 when
the entire area of  the bay was closed to shellfishing.  A year-
long  study  of  the  bacteriological quality of the bay and all
potential sources  of  contamination was  initiated  in  September,
1985.  Fecal coliform,  the currently-accepted index of  water
quality, was used in the study.  In  nearly all cases, further
differentiation of the fecal coliform group indicated that the
organisms  were Escherichia coli, a normal inhabitant in the
intestinal  tract  of warm-blooded  animals.   Investigations
centered around  six  possible  sources  of  contamination:
stormwater,  septic  systems, wildlife, marinas,  freshwater  inputs
and point discharges.                       '

Stormwater

     Our investigations suggest that stormwater discharges are
the most important factor causing the periodic reclassification
of the  area as unsuitable for  shellfish harvesting.   Maximum
contamination levels in Buttermilk Bay coincide  with rain events,
a  period  in which  samples  are taken for  the purpose  of
classification of  the  area  under presently-accepted practice
(sampling during "worst  hydrographic conditions").  The  level of
fecal coliform contamination at discharge points is shown to be
related to  three  main  factors.   The  extent of residential
development is shown  to  be positively related to  fecal  coliform
levels, with highest  levels observed  at the western shore where
housing density approximates 20  units per ha (8 units per  acre).
Frequency of rain events was negatively correlated with fecal
coliform densities at discharge points,  presumably due to a
shorter period available for  fecal  material to accumulate on
surfaces drained.   In agreement with  National  Urban Runoff
Program studies,  a seasonal effect  on coliform densities at
discharge points was  noted with minimum values observed  during
winter  sampling.  To place  coliform loadings  from discharge pipes
in perspective,  the amount  of water required  to  dilute the
discharge to 14 FC/1QO ml was calculated and compared with the
tidal prism  (2.5 x 106 cu.  m).  Between 18%,  during colder  months
sampled, and 440  % ,during  warmer months, of the tidal  prism
would be required for this dilution.  A  survey of each surface
drainage area  during dry periods  suggests that  the source  of the
fecal coliform during discharge events is not  sanitary wastes.
Domestic dogs and wildlife  are implicated as  over 100 dogs
inhabit areas served by drains.  Use of  in-situ measured fecal
loadings from dog wastes indicate that the predicted volume of
dog waste necessary  to account for the overall geometric mean
value of fecal coliform in discharged  water could be accounted
for by a 2-3 day accumulation.

    In  addition to direct  discharges,  storm events cause a
release of  fecal indicators  from  sediments and other  protected
reservoirs  into the  water column. The extent of this release
could not be quantified,  however  measurements of  fecal coliform

-------
from these protected reservoirs  indicate that the effect may be
substantial. Scavengers and waterfowl, particularly Canada geese,
are  strongly  implicated as  being the source  of high  fecal
coliform (>1/000 FC/gram) in strand line  deposits.

Septic Systems

     Groundwater  sampling in  the Buttermilk  Bay  watershed
presented  conflicting  indications  regarding  fecal  coliform
entrainment.   General sampling indicated groundwater
contamination at  distances  at  least  35  m from  the nearest
suspected  source,  however a more intensive sampling scheme
performed near two  septic systems indicated limited  mobility of
fecal  indicators  in this area's  soil. While the entrainment of
bacterial  indicators  appears  to be  limited, the  issue of
pathogenic virus entrainment remains unresolved after the present
study.  A review  of pertinent published studies  regarding  virus
entrainment is presented and suggests a lateral entrainment of
viruses to at least 67.05 m (220 ft)  in soil  types similar  to the
study  area.  It is concluded that,  regarding the entrainment of
pathogenic organisms in  groundwater,  a  major public health
threat, the viruses, is probably not adequately assessed with the
indicator system used.

Wildlife

     A waterfowl survey was conducted weekly to determine the use
of  Buttermilk Bay by migrating waterfowl  and to determine the
resulting bacteriological impact.   Following techniques employed
by  Hussong et al.  (1979), theoretical loadings were compared with
actual field values.  Predicted estimates generally  coincided
with field measurements and indicate  that, except in certain
areas,  the  use of the bay by waterfowl  had  fairly minimal direct
impact on water quality.   A long-term cumulative impact on  water
quality due to the fecal  deposits  in the beach  areas, however,  is
indicated due to the maintenance of fecal material in a  protected
area.   These wastes accumulate and can be released  in a  slow
diffuse pattern, and can  result in considerable local degradation
of  water quality.   The release of these fecal  indicators during
certain hydrographic and meteorological events (ie. high tides
 and rainfall events) can  expect to significantly  impact  the water
quality,  the extent of  which will be determined by a  number  of
 variables to include circulation, water temperature, etc..

 Marinas

    Utilizing two  sampling approaches, no measurable  impact on
 the bacteriological quality of water was observed as a  result  of
 marina operation.   These results  should  be  interpreted  with
 caution however,  since the nature of the suspected wastes would
 necessitate  extremely  fortuitous circumstances  in  order  to
 determine  the  actual impact of an overall operation.  In addition,
 the marinas studied should be  considered atypical due to inherent
 restrictions on boat size in  one case, and  the presence of pump
 out facilities in  the  other case.   It is concluded  that the

-------
extent of  impact  of  marina  operations will be determined by the
level of convenience and cost  associated with the proper handing
of sanitary wastes at each facility.  Studies documenting the
effect of  marinas are reviewed herein and  indicate that in more
"normal11  situations,  the impact of  marina operations can be
significantly adverse.

Freshwater Inputs

     The review of historical data  as  well as investigations
presented  herein  confirm  that  the  freshwater  drainage  into
Buttermilk Bay  is a consistent source  of fecal coliform.  In the
case  of  Red Brook,  no point sources were located.   It is
concluded  that two possibilities exist  for the consistently
higher fecal  coliform densities observed.   Although limited
groundwater sampling  failed  to  reveal  this as  an input, it is
possible   that septic  plumes  are  entering  the  brook  at
Undiscovered locations.  In addition, investigations reviewed
herein and small scale experiments suggest  that the extensive
marsh area near  Red Brook, and  possibly  the marshes and bogs
within the watersheds  of other  freshwater  inputs to  Buttermilk
Bay are sources of "natural" fecal coliform  (those fecal coliform
surviving  and possibly  multiplying outside  a warm-blooded animal
host).  It is  certain  that  each  of the marshes surrounding and
draining  into  Buttermilk  Bay contain  a  variety of wildlife
species which  additionally will act as a diffuse but  significant
source.

Point Discharges

     Only   one  point  discharge  was  discovered (not including
stormdrains) in the present  study.  This  consisted of  a discharge
pipe  located  at a local  fish market.   The location of this
discharge  in Cohasset Narrows probably minimizes its impact,
however during certain tidal stages  (incoming tide)  the impact
may be quite substantial and extend more into the bay.
Other Considerations

     Small scale  experiments  verified that solar radiation is a
prime determinant of fecal coliform die-off  in surface layers of
Buttermilk Bay.  Comparison  of bay water sampled at different
locations  suggests that in areas  coincident with higher nutrient
influents,  the  ultraviolet light  penetration  is attenuated which
may  cause an  increased persistence of  fecal  coliform  or a
modification of the die-off rate in  these areas. In  addition to
modifying the mortality  rate  of fecal  coliform,  nutrient
additions  may  allow  for  maintenance and limited growth of fecal
coliform and  pathogens as implied by  laboratory experiments. In
addition  to  the  direct utilization of  effluent   nutrients,
laboratory investigations  using water and  algae collected in
Buttermilk Bay suggest  that  algal growth, which results from
nutrient  inputs,   may  additionally supply  fecal indicators, and
likely pathogens, with complex nutrients,  resulting in fecal
indicator growth.   These initial  small-scale experiments
collectively suggest a link between  nutrient enrichment through

-------
on-site  sewage  disposal  practices and  bacteriological
      1   10n °f Buttermilk BaY  which  should  be  researched

-------
                         INTRODUCTION

     The United States Census Bureau (Carter, 1980) indicates a
strong preference of people to live in coastal regions.  If the
present  trends  continue,  by the year  1990, 75  percent  of the
population of  the  United States will  live within 50  miles of
tidal waters  and the  Great Lakes.  This demographic shift toward
coastal environments  will undoubtedly result  in the  inability of
many coastal  areas to absorb  the human wastes  generated,  and will
prevent  these areas from being used for many of  the activities
generally associated with  good water quality  (ie. shellfish
harvesting  and  swimming).  In  southeastern Massachusetts,  where
unprecedented growth  has  occurred in the last decade,  there has
been a 28 % increase  in the  number  of areas closed to shellfish
harvesting  between  the years 1983  and 1985,  and in  1984,
formerly-pristine  areas  such  as the Westport,  North and South
Rivers were closed  due to bacterial contamination  (Massachusetts
Division of  Marine  Fisheries, 1985).

     The issues involved  in  bacterial  contamination of surface
waters are complex  and cover both point- and  nonpoint sources of
contamination.   For the most part,  nonpoint sources  of pollution
have played the  major  role in affecting  the majority of shellfish
area  closures  in recent years.  Nonpoint pollution refers to
certain  categories of natural sources of wastes or wastes from
activity of  man which are dispersed or  diffused (Furfari,  1979).
While  many of  the  issues and questions regarding point source
pollution  problems  can  be  easily addressed,   the nature of
nonpoint source pollution  (being  dispersed and  diffuse) has
precluded many  successful attempts at abatement.  Paramount to
controlling  nonpoint  source  pollution is an understanding of the
mechanisms by which nonpoint sources affect the nearshore marine
resource in each situation.  While  many advances have been made
in understanding nonpoint pollution, there is general agreement
that many of the mechanisms resulting  in coastal  degradation of
water quality are poorly understood.  The purpose of the study
described  herein  is to  clarify  the  sources  and mechanisms
involved in  the contamination  of Buttermilk Bay,  southeastern
Massachusetts   with  fecal  coliform,   the generally-accepted
indicator of  water  quality  for  purposes  of shellfish harvesting
and recreational contact (swimming,  wading etc.).


                          STUDY AREA

    Buttermilk  Bay  is a small  embayment in the northern section
of Buzzards  Bay,  southeastern  Massachusetts (Figure 1.). Our
study area  included Little Buttermilk Bay ( 0.43 km2 at mean low
tide) and Buttermilk Bay proper  ( 1.71 km2 at  mean low tide).  In
addition, freshwater surface  inputs and their watersheds were
investigated  for coliform sources.

    Land use  within the watershed  of  Buttermilk  Bay ranges  from
intense  residential   use  (0.05  ha -  0.115 acre lots) on the

-------
o\
      Fig. 1   Study area - Buttermilk Bay, southeastern
              Massachusetts

-------
western and northern shores  to  lower density  residential use on
the eastern shore.   In addition,  stormwater  inputs  from a
commercial area along  Cranberry Highway (Routes  6-28) impact  the
bay along  the  southern shoreline.   There are over  20 direct
stormwater discharges  into the  bay  for a  combined drainage from
impervious  surfaces  of approximately 80,620 sq.  m  (867,800
square  feet )   (Figure  2).  Descriptions  of  all known direct
discharge  pipes with associated drainage areas are presented in
Appendix I.

    Until 1983,  the area of Buttermilk Bay was perceived as
having good  water quality for both shellfish harvesting and other
recreational uses.  In October of that year,  a  portion of  the bay
near  "Byron's Landing"  in Wareham  was  reclassified  as
"Prohibited" under MGL  Chapter 130, Section  74 A.   Following a
review  of the existing  data, the entire area of Buttermilk and
Little Buttermilk Bay was closed to  shellfish harvesting  on March
13f 1984.   Although the problem of high coliform counts  was
perceived as being recent to the  1984 closure,  historic data
indicate that high coliform and fecal coliform were evident as
far back as 1973.  Data collected by the Southeastern Regional
Office  of  D.E.Q.E.  indicate  continued sporadic contamination
problems from  1973  to date.  Certain areas in the bay were opened
for seasonal harvest of shellfish in October, 1986  when it was
determined  that the data supported  this type of classification.
In July,  1985  a  resurvey  of  the  area was  conducted  by  the
Southeastern Office of D.E.Q.E. and  the  United  States Public
Health  Service, Food and Drug Administration.   The results of
this study are  unavailable at this point in time.


                   APPROACH AND METHODOLOGY

    Investigations  reported herein centered around the following
six possible sources of  fecal coliform in  the  bay which  serve as
an outline for  presentation:

                1) Stormwater
                2) Septic Systems
                3) Wildlife  (to include waterfowl)
                4) Boats and Marinas
                5) Freshwater Inputs
                6) Point Source Discharge

    In addition to source delineation,  certain aspects of  the
ecology of  the  indicator organism were investigated,  particularly
in relation to possible links with nutrient enrichment.  These
later studies,  although small  in  scale,  point to the need  for
further study to determine possible  links between eutrophication
and the ecology of  pathogens  in  embayments.


Analytical  Procedures

     Fecal  coliform densities were  determined using  two methods.


                             7

-------
CD
                                               Buttermilk
                                                  Bay
                                                                                     Approximate drainage areas
                                                                                             0.5 KM
       Fig. 2 Locations of stormwater discharges in Buttermilk Bay, southeastern Massachusetts
              with approximate  drainage areas served.

-------
The primary method  used for comparability with data collected by
D.E.Q.E. was the "Modified A-l"  test  (APHA, 1985). A 5-tube,  3,
4 or 5  dilution series (5-dilution series was used for stormwater
analyses)  was performed with results obtained  from a standard
MPN table.  In addition to fecal coliform density determination,
the presence of Escherichia  coli  was determined by adding 4-
Methylumbelliferyl-beta-D-glucuronide (MUG) to the nutrient
broth.   E. coli produces glucuronidase,  an enzyme that hydrolyzes
MUG to yTeld a fluorogenic product.  The presence of E. coli was
determined  by observing for fluorescence using a  hand-held  long-
wave ultraviolet  light following the determination of  fecal
coliform. Tests for fecal coliform using  MFC Agar (Difco) were
also performed for comparison  with A-l methodology as well as for
the purpose  of  supplemental sampling where freshwater  fecal
coliform densities  were to  be determined.   In addition  to  fecal
coliform  tests, in  certain situations,  enterococci,   fecal
streptococci  and  Cj^os^rj^di^um  p_erfri^ngens  densities  were
determined in order  to evaluate theTr" utiTIty  in determining
contamination sources. Enterococci media was prepared after Levin
et al.  (1975)  and Cabelli  (personal communication).  Streptococci
were determined using Difco  media and incubated in accordance
with Standard Methods (APHA, 1985).   Media and  methods  for the
determination of Clostridium  perfringens  spores were performed
after Bisson  and Cabelli  (1979).

    Six stormdrains representative of various land usages were
chosen for monitoring during eight storm events.  Samples were
taken by immersing a  sterile sampling bottle in the discharge
flow  from  the pipes at  discharge points  for  the  drains
designated  as  Electric Ave  (Code #3)., State  (Code  #27),
Jefferson Shores (Code #26),  Red Brook Drain (Code #  16), and
Puritan Ave (Code #8)  (APPENDIX  1). Due to  the normal occurrence
of a tide which submerged the drain at the discharge point, the
Wychunus drain  (Code  #17) was  sampled at  the distal-most road
drainage basin. A rain gauge was placed at  the mouth of Red Brook
and served  as a  indicator  as to  time  of sampling.

    During  the initial two storms (5/22/86  and 7/2/86),  sampling
was conducted during  the  "first flush"  period (within 30  minutes
from the beginning of the rain event), and following at  least
0.64 cm  (0.25  inches) of precipitation.    Since there was no
consistent first flush effect  as described by Whipple  et  al.
(1983), it was decided to that the first  flush  sampling effort
would be discontinued and effort would be  realigned  to include
additional rain events (a total of eight).  Subsequent sampling
at discharge points on 7/21, 7/30, 8/11,  10/2,  11/5, and  12/19
was performed as soon after 0.64 cm of  precipitation as possible.

    For the purpose of estimating the fecal  coliform loading from
stormdrains,  the  volume  of runoff   entering  the bay from
impervious surfaces was estimated.   The impervious surface area
serviced by each stormdrain  was first  calculated by walking  the
area with an odometer calibrated to  the nearest linear  foot and
measuring the width of each road section.   Stormwater volume was
then calculated  using  the  formula:

-------
                          V = A x P(100)

where V = the volume of stormwater from  each drain (cu.  meters),
A = the impervious surface serviced by each drain ( sq. meters)
and  P = the amount of precipitation in cm. The geometric mean
fecal  coliform densities of  the six drains  combined (fecal
coliform/100 ml x 104  ) were used to calculate the fecal  coliform
loadings for each rain event monitored.

    Eleven routine  sampling stations  throughout  the bay (Figure
3.) were chosen for monthly monitoring from September, 1985-
October, 1986.   Samples were taken during the  mid phase of the
outgoing  tide  with the  prerequisite of having no significant
rainfall during the prior  72 hrs.  Samples were taken by wading
or  means  of boat  where  the bottom depth was at least  1 meter.
The sampling distance from the  surface of the water was  30-40 cm
with the exception of samples  taken near the mouth of Red Brook
where  a stratified sampling design was used. At this station, 1
sample was taken approximately  15  cm  from the surface and a
second sample  was taken 60-90 cm from  the surface.   Mid-bay
sampling  stations were omitted during winter months due to
navigational difficulties associated with ice cover.

    Groundwater samples were  taken using a shallow well-point
sampler with  a  slotted  well point.   Using  a vacuum pump,
approximately one  liter of  water was evacuated from the  well  into
a  sacrificial bottle.   Following this  procedure,  a sterile bottle
was connected  to  the device  and a  sample was withdrawn and
analyzed.

    Sediment  samples were  taken using  two methods.   For a
comparison of  sediment vs. overlying water vs.  shellfish meat,
approximately 1 m2 of bottom  was  disturbed and a  sample was taken
within the turbid water boil.  Following MPN analyses, a
determination of  the  amount of sediment  suspended per  100 mis of
sample was made  by filtering a  well shaken  sample  through a
weighed filter for the determination  of sediment weight and
 subsequent  translation  of  results  into number  of  fecal
 coliform/gram of  sediment.  For analyses  of bottom sediments  near
marinas sediment samples  were  taken using a ponar dredge .  Upon
bringing the  sample  to  the  surface  and depositing  it  in a
 sampling  tray,  approximately the  top  2 millimeters  were scraped
 off with a sterile tongue depressor  and  placed  in a solution of
 sterile phosphate  buffered water for  analyses.  Determination of
 fecal coliform per gram  of sediment was made  using the method
 previously described.

     All samples were stored in ice following collection and  were
 analyzed within six hours of  collection.

     To determine the use of the bay by  waterfowl, eight shore
 observation  sites were  chosen  which allowed  for  a   complete
 waterfowl census of  the bay.  Waterfowl counts were generally
 conducted between  0700 h and  1000  h  on  a  weekly basis.
                              10

-------
                                                                LITTLE
                                                              BUTTERMILK
                                                                BAY
Indian
Heights
       Jefferson
        Shores
Fig. 3  Location  of routine bacteriological
and nutrient  sampling in Buttermilk  Bay,
southeastern  Massachusetts, September  1985-
October 1986.
* Station 5.9  sampled at two depths

-------
Waterfowl were separated  into  the categories of  ducks, geese and
swans for the purpose of estimating  fecal coliform loadings.

     Viability  of various animal feces deposited  in the area was
approximated using two methods.  Initially, the location of waste
deposits were  marked and subsampled at various  intervals.  One
gram samples were placed  in 99 mis of sterile buffered water and
samples  were  processed using  the  MPN  technique  previously
described.  Fecal coliform content in organisms per gram were
thus determined.  In addition, during autumn, 1986, fresh goose
feces was placed in the strand line  in anchored nylon mesh bags
(365 micron mesh) and subsampled at selected  intervals.
                            RESULTS

STORMWATER

    The presence of large numbers of bacteriological  indicators
and pathogenic organisms  in stormwater runoff  is  well  documented
(Olivieri et al. 1977). As shown  in Long Island,  New York, the
impact of  this type  of bacteriological contamination  in
recreational-use areas  is  often  substantial  (Koppelman and
Tanenbaum 1982).  In what is perhaps the most  comparable portion
of  the National Urban  Runoff  Program  (NURP), these investigators
have  concluded  that,  in Long  Island,  stormwater  runoff  is
"generally  the single most  significant  source of pollution,
especially  bacterial pollution,  affecting  the  fresh surface
waters and  nearshore marine environment".  Runoff  to estuarine
waters of Nassau County, New York has been  implicated as the
primary  source of  bacterial  loading to all but  one of the nine
embayments.

    Although  the mechanisms  responsible  for  increased  fecal
coliform levels in  receiving  waters  following rain events are not
fully understood,  the negative affect of  stormwater runoff on
Buttermilk Bay is  unquestionable.  A summary of  selected  "open
water" stations sampled prior to,  during and following a rain
event (FDA-DEQE,  1985) clearly shows increased fecal coliform
levels in response  to  rainfall (Fig.  4).  This phenomena has been
observed elsewhere (Gerba and Shaiberger  1973, Hill and Grimes
 1984,  Schillinger and  Gannon 1985, Koppelman and Tanenbaum  1982)
and is the primary reason for a number of areas  in Massachusetts
being classified  as unacceptable  for  shellfish harvesting
 (D.E.Q.E,   S.E.R.O. - personal communication).  Stormwater
impacts on receiving  waters can be classified into three main
categories which are discussed separately:

       1.) Impacts from direct stormwater discharges and overland
          runoff.

       2.) Release of coliform from protected reservoirs.

       3.) Surcharge of receiving waters with contaminated
          groundwater.

                              12

-------
3.0-n
                                                                                          74FC/100  MLS
          7/8   7/8  7/9
          AM    RM   AM
7/9
PM
7/11   7/12

    DATE
7/15  7/16 7/16  7/17
      AM   PM    AM
7/17  7/18 7/19
 PM
      Figure 4.   Fecal  coliform densities observed at selected sampling stations during a joint
                 FDA -  DEQE  survey of Buttermilk Bay, southeastern Massachusetts, July 1985.

-------
Impacts  from stormwater discharges and overland  runoff

    The  area around Buttermilk Bay  is  serviced by  at  least  20
drains  discharging directly into  the surface  waters. The
impervious surfaces  (primarily roadways)  amount  to approximately
80,620 sq. m  (867,800  sq. ft). Following  a  2.54 cm (one-inch)
rain, approximately 2050 cu.  m of stormwater is directed to the
bay from man-made conduits. Using geometric mean fecal  coliform
densities  observed  during 1985 (Figure 5), volumes of  water
between 4.6 x 105  cu.  m and  1.1 x 10'  cu. m would be  required
to dilute  the incoming stormwater runoff from  impervious surfaces
to a level of  14  fecal  coliform/ 100 ml following a  2.54 cm rain
(these values are only  used  as a reference).    To  place  these
coliform loading  values in perspective,  assuming that the  tidal
prism  for Buttermilk  Bay is 2.5  x 106   cu.  m, the  incoming
stormwater entering the bay  from impervious surfaces  during a
2.54 cm rain event would require between 18 %   and 440 % of the
tidal prism  for dilution to  the  14 FC/  100  ml standard.  In
reality, using this  simplistic approach  to  predict the number  of
tidal  exchange  volumes necessary to dilute the  incoming
stormwater to acceptable level is  likely invalid.  Simple water-
parcel  transport  computer  models supplied  by  the Boston
University Geology  Department (APPENDIX II) indicate  that the
extent of  penetration of contamination from  drains near the  mouth
of Buttermilk Bay (such as the Electric Avenue Drain) and  hence
the  ultimate  passage of contamination back out of  the bay,  is
highly dependent  on  the stage  of  tide during  which the
contamination is  introduced.

    In all  instances  where  fecal  coliform were isolated from
stormwater, further  differentiation  using MUG indicated that the
fecal coliform involved was  E._ coli.   Reference  throughout this
section, however,  is retained   as "fecal coliform" for clarity  in
comparison with  the standard used  in classifying shellfish
harvesting areas.

    Geometric mean fecal coliform densities (all drains combined)
observed  in Buttermilk  Bay drains (Figure  5) generally compare
with values observed elsewhere ( Whipple  et al.  1983).  While the
variability  of  the data is considerable (Figure 6),   some
generalization can be  made relative  to coliform  loading and land
use.   In  general,  fecal  coliform densities  observed at the
"State" drain, typical of commercial-use land, were lower than
those observed at drains  servicing   residential  areas (all  other
drains  sampled).  Among the  remaining drains  monitored  which
serviced residential areas, the highest  fecal coliform densities
were observed at the "Wychunus" and "Red Brook" drains  compared
with the  "Electric Ave.",  "Jefferson  Shores" and "Puritan Ave."
Concomitantly,  the areas serviced by  the Wychunus and Red Brook
drains  are more intensively developed (ca.  20 units/ha or 8
dwelling  units/  acre)  compared  with  areas  serviced by the

-------
                           ALL DRAINS COMBIMED
7-t
6-
   $••
   L
L  I
0  F
G  0 *
10B
   M


E  0^
C  0
A
L  M
   L 2
   S
 1-
                                                                            Townhouse/Garden Apts.  (8-22 DU/Acre)
                                                                            Suburban Shopping Center


                                                                            Large lot residential  (.1-2  DU/Acre)
                                                                            Medium Density Residential
                                                                            (2.0 -8.0 DU/Acre).

                                                                            DU = dwelling units
                                                                     Source
     5/32      7/2      7/31      7/30     8/11

                                  RAIN EVENTS
                                                        10/2
11/5
12/19
                                                                                   Occoquan/Four Mile Run Nonpoint
                                                                                   Source Correlation Study - Final
                                                                                   Report  Northern Virginia Planning
                                                                                   District Commission and Virginia
                                                                                   Polytechnic Institute and State
                                                                                   University. 1978.
                                                                                   (as cited in Whipple et al. 1983)
  Figure 5. Geometric mean fecal coliform densities  in stormwater entering Buttermilk Bay,  southeastern
            Massachusetts from six selected drains.  Fecal coliform were further differentiated  and
            identified as Escherichia coll.

-------
                               PURITAN AVENUE
    5-
  C
  0
  L 4-
L I
0 F
6 0
 10 R
F  1
E  0
C  0
A
L  M
   L
   S
         5/22
7/2
7/21     7/30      B/ll     10/2

         RAIN EVENTS


     ELECTRIC AVENUE
11/5
IS/19
    5-
   C
   0
   L 4-
L  I
0  F

°»S

   7"
F  1
E  0
C  0
A   z.

   L
   S
     1-
         5/22      7/2      7/21     7/30     B/ll      10/2     11/5     12/18
                                   RAIN EVENTS


       Figure 6  Fecal coliform densities observed at selected stormwater
                 discharge points in Buttermilk Bay during 1986.  All
                 fecal coliform were further differentiated as Escherichia
                 coll.
                                     16

-------
                                  STATE DRAIN
    4-

  C
  0
  L
L I
0 F 3
G 0
 10 R
  II
A
L  II
   L
   S
I
         5/22      7/2     7/21     7/30      B/ll     10/2     11/S     IE/19
                                    RAIN EVENTS
                               JEFFERSON SHORES
5-
C
0
L 4-
L I
0 F
6 0
IQg
7"
F 1
E 0
C 0
A 2
L M2
L
S
•








1



•



i












    1-
        5/28      7/2      7/21



        Figure 6  - continued
                          7/30      e/11
                           RAIN EVENTS
10/2
ii/a
12/19
                                     17

-------
                               RED BROOK DRAIN
    5-
  C
  0
  L 4
L I
0 F
G 0
 10 j{

  M3-J
E  0
C  0
A
L  M
   L
   S
2-
        > 5.2
              >  5.2
         5/22      7/2     7/21     7/30      8/11      10/2     11/5      12/10

                                    RAIN EVENTS
                               WYCHUNUS  AVENUE
     8-1
     5-
   C
   0
   L 4
 L I
 0 F
 6 0
  10R
 F  1
 E  0
 C  0
 A
 L  M
    L
    S
 2-
      1-
          5/22      7/2      7/21


          Figure 6 - continued
                               7/30     8/11      10/2     11/5     12/19
                                RAIN EVENTS
                                      18

-------
remaining  "residential" drains  which  are less  intensively
developed « 20 units/ha). These findings are supported by the
Northern Virginia Planning Commission  and Virginia Polytechnic
Institute and State University (as reported in Whipple et al.
1983) and data  summarized by the North  Carolina  Division of
Environmental Management (1986) and  suggest that the degree of
residential development  directly  affects  fecal coliform
contamination within areas  serviced by surface drains (Figure 6).

    In addition to intensity of land use in  areas serviced by a
drainage system, two additional factors were identified which
also affect the fecal coliform loading of surface drains.  In
five of the six drains  sampled on August  11,  1986,  coliform
densities had decreased compared with the prior sampling date
(July 30).  Meteorological data for the prior ten days  indicated
that  total rain in excess of 3.5 cm   had  fallen on three dates
during this  period (Figure  7).  These  data  suggest that the time
required  for accumulation of  fecal coliform source material  (pet,
rodent, bird feces etc.) was considerably less than the other
dates sampled,  resulting in decreased fecal coliform densities
during the August 11  rain event.  Converse  to the situation where
numerous rain  events  in succession prevent the excessive
accumulation of source  material  in  areas serviced by surface
drains, samples collected on October 2,  1986 at all six drains
sampled showed  higher fecal coliform densities than during either
of  the  two prior rain  events  (July 30 and August  11).   An
examination  of  meteorological data for September indicates that
only one significant  rainfall had occurred during the entire
month (1.7 cm on September 16- Figure 7) prior  to the October 2
sampling date.  In addition,  rainfall for the entire month of
September did not exceed one inch. These  data suggest that the
fecal coliform  densities at  discharge points for surface drains
will depend, in  part,   on the interval of time since the  previous
rain.

     In addition  to  the the  frequency of  rain events,  data
collected during a December 19 rain event suggest that  decreased
air temperatures may  serve to  reduce the fecal  coliform  loading
from surface drains.   Although there were two substantial rain
events within two weeks  prior to sampling  (1.6 cm on December 10
and  1.3  cm  on December  12),  we  feel that  the degree of
attenuation on this date  exceeded what would be expected  from the
single  attenuating  effect  of  preceding  rain events.   For
comparison,  data collected during two July rain events  indicate
fecal coliform densities  more than one order of magnitude greater
than those of  December  19,  this despite a  greater amount of
rainfall (Figure 7)  on dates more approximate  to those of the
December 19  rain event.

    Intensity  of  rainfall  was  not  shown  to  correlate  with
geometric mean   fecal coliform  densities observed at stormdrains,
however, confounding  influences  imparted by seasonal effects,
and the frequency  and  proximity of previous  rainfalls   may
preclude a  definite  statement  regarding the  effect  of  this
factor. Fecal coliform densities during the July 21 rain event,


                             19

-------
ro
o
3.0,

2.0-
INCHES
53
M 1.0.
J
CK
Sj





I






ll






1







i
JAN




































1












III





FEE


* = Date








Mi







J
MAR







on which stormwater sampling








.1










II
APR







*

II.. L 1







I ,
1 MAY
1 986



















*




1 It
JM

























was performed



*



*
I
1
JUL












*


L







In








L.
AUG I SEP










*
I




























i
OCT











*



|





J



















NOV








*

















DEC


            Figure 7 Summary of precipitation near theBittermilk Bay study area Southeastern Massachusetts,
                     as recorded  (in inches) at the Cranberry Experimental Station, Wareham, Massachusetts.
                     * indicates  dates on which samples were collected at selected stormwater discharge
                      points in the bay.

-------
when the rainfall was 0.1 cm/h (as determined by the rain gauge
at the mouth of Red Brook) were higher than those observed on May
22 and July 2 when the rate of rainfall was 0.48 and 0.69 cm/h
respectively. Conversely there  were other dates such as August  11
(0.1 cm/h)  and  December  19 (0.2  cm/h)  on which  lower
precipitation  rates  resulted  in  lower geometric  mean  fecal
coliform densities in discharge water.

    Substantial seasonal differences  in  fecal coliform counts  in
stormdrains  have been  observed elsewhere (EPA 1983).  NURP
studies report that coliform  densities in  urban runoff  during
warmer  periods of the year are approximately 20 times  greater
than runoff during during colder periods.  As in our study, the
observed seasonal  differences could  not  be related to comparable
seasonal variations in land use.  Two  drains chosen (Electric
Ave. and Puritan)  have comparable  occupancy of  dwellings
throughout the year, yet still exhibited seasonal variation  in
fecal coliform  densities.
     Despite numerous studies nationwide under the National Urban
Runoff Program and others (Olivieri et al.  1977 and Van Donsel et
al.  1967), there  still remain many unanswered  questions regarding
the actual source of fecal coliform in runoff.  Although in the
Buttermilk Bay  area,  the possibility of direct connection of
sanitary  waste   lines  with  stormwater  conduits  can  not be
completely discounted, two  observations strongly suggest that if
sanitary waste does make its way into the collection system it is
by  less overt means  (ie.  surcharging of drainage  basins by
adjacent  septic systems).   Firstly,  all  storm  drains were
observed during peak land use periods (June -  August),  which were
concurrently dry  periods, for the presence of  running water. The
inspection was systematic  and involved all surface drains from
the distal to  proximate portions of the surface drainage areas.
With  few exceptions, which  were  verified by  bacteriological
examination not  to be due to sanitary  waste,  all drainage basins
remained  dry  during dry weather.   In  addition, all collection
pipes generally exhibited little or no  flow within approximately
two hours following  the  cessation of rain.  If direct discharges
of sanitary wastes were present,  we would expect at least some
continual flow into the collection system;  this phenomena was not
observed.

     Comparison  of  fecal coliform  values observed in our area
with those reported by  Olivieri et al. (1977) gives support to
our  contention that  direct input of  sanitary wastes via
subterranean pipes  would not be necessary  to  account  for the
fecal coliform values observed from discharge pipes in our study
area.  In this author's  report of two drainage  basins in which
"no sewage overflows are known to be  located", geometric mean
densities  (MPN)  of  8..S  x  104  Fecal Coliform/ 100  mis ("Bush
Street")  and 6.9  x 103 fecal coliform/  100 mis ("Northwood") were
reported.  These  levels  of  fecal coliform contamination are quite
comparable to  those  levels  observed in our study area (Figure 5).
Further, this  author's summary of stormwater  samples  taken in  a


                             21

-------
North  Carolina stormwater drainage  basins indicate that,
independent  of  sanitary  wastes  from  sewage systems,  fecal
coliform contamination of rainwater can occur quite quickly after
contact  with the  earth's surface  .   Fecal coliform values
observed at a roof  downspout  (trees overhanging) approximated 740
fecal coliform/ 100 mis, and  levels approximating  25,000 - 58,000
fecal coliform/ 100 mis of sample were  observed from overland
flow through a residential area (Betson and Buckingham 1969 as
cited  in Olivieri et al. 1977).  Collectively,  these studies
suggest that  although fecal coliform contamination at  stormwater
discharge points in Buttermilk Bay may appear  substantial, these
levels  do not necessarily   indicate  the presence  of direct
connections between subsurface on-site systems and stormwater
collection  systems or the  input of  sanitary  wastes  via
subsurface surcharging of  drainage basins with sanitary wastes.
It is still possible,  however, that some  septic systems installed
prior to present regulations  included a connecting overflow which
functions only during rain  events,  which remained beyond our
detection.

    Although no in-depth studies  have been performed to detail
sources of fecal coliform in stormwater  runoff  in areas where
sanitary connections  are not suspected,  certainly  animals
residing in the surface watershed are implicated.  In excess of
100  dogs live  within the surface watershed of Buttermilk Bay
(based  on  July 1986 survey).  In-situ  investigations of  the
viability of dog feces in July- August 1986 indicated that at 1
week, 2 weeks and  30 days, the viabilities of dog  feces deposited
near the beach  in overgrown areas were  not  significantly
different.   In each instance the  MPN fecal coliform densities
were 106 fecal coliform/gram  of feces.  Thus,  the cumulative
affect  of just dog wastes  can be quite substantial.  To place the
possible effect  of dog  wastes in perspective, an  overall
geometric mean fecal coliform concentration  in  stormwater  (all
dates,  all  drains combined) was  calculated and the amount of
canine  fecal waste necessary  to account for  this  value  was
calculated.  Considering an overall geometric mean fecal  coliform
concentration of 7,349 PC/ 100 mis  and the resulting  input of
fecal  coliform  to the bay of 1.5  x 1011  during  a  2.54 cm  (1
inch) rain,  the amount of dog wastes necessary  to  account for the
loading would approximate 150 kg (331 Ibs.).  It is conceivable
that this amount of waste  could be deposited by the resident dog
population in 2-3   days,  assuming  a 454 gram/dog/day deposit from
the  estimated 100 dogs  residing in  the  areas  serviced by
stormdrains.

    The ability to predict fecal coliform  loadings from domestic
dogs, using these  values  is precluded by  the complexities of the
factors affecting  the actual  transport of viable organisms to the
bay. The extent to which adjacent animal populations  impact a
water  body are  controlled by  at least two factors:  the
probability of the  wastes  becoming entrained  in stormwater
entering the bay,  and the survival characteristic  of the waste
following deposition. Regarding  the probability  that animal
wastes  are  entrained in  stormwater,  the  primary intermediating
                              22

-------
factor is the percent of impervious area in the  drainage basin
(paved surfaces) as well as the drainage characteristics of the
remaining "natural" area (primarily lawn and  landscaped sites of
residential areas). Residential properties serviced by stormdrain
systems surrounding the bay for the  most part have their drainage
patterns modified in such a way as to increase the probability
that surface-deposited   wastes  will be entrained in  stormwater
entering the bay.  These  practices include landscaping which
slopes from  the house  to the  street,  a sloping of driveways
toward the street, and in some  instances the diversion of rooftop
drainage into the paved roadways.   While these  practices  have
beneficial purpose and  prevent  the  flooding of properties,  they
concurrently allow  for a more  expedient pathway of  surface-
deposited wastes into the receiving  waters.

     Another  essential  element determining the impact  of  animal
populations  on  adjacent  water  bodies  is  the  survival
characteristic  of the  wastes following  deposit. The  extended
survival of fecal indicators observed in the present  study and
elsewhere  (Temple et  al.  1980, Edmonds 1976), suggests  the
possibility  of  substantial cumulative  effects  which must  be
considered in attempts to predict the fecal coliform loading  from
domestic dogs and other  wildlife inhabiting the surface drainage
basin of Buttermilk Bay ( rodents, birds, rabbits etc.).    The
factors controlling the  survival rate of  indicator  organisms  in
surface-deposited  wastes are solar radiation  (Bell 1976,  Bell et
al. 1976, Van Donsel et al. 1967), temperature  (Boyd  and  Boyd
1962, Edmonds 1976, Van Donsel et al. 1967,  Weiser and Osterud
1945 )  and moisture (Brown et al. 1979, Weiser and Osterud 1945).
While for the most part  these factors  are  beyond man's influence,
the drainage  system  itself does provide  a man-made  environment
for  the  protection of  deposited wastes as well  as  possible
habitat for additional wildlife such as rodents.

    The effect of  temperature on the survival  of  fecal coliforms
noted by various  authors  may provide insight to the  apparent
dramatic  decrease  in   stormwater contamination  observed  in
December as well as decreases in  stormwater loading during winter
months observed elsewhere (Faust  et al.  1975).   While  frozen
ground generally decreases infiltration and would be   expected
to   increase stormwater  loading  (Crane  et al. 1983), field
experiments conducted in the present study,  as  well  as
investigation elsewhere  (Van  Donsel  et al. 1967,  Weiser and
Osterud  1945)  suggests  that  numerous  freeze-thaw cycles
significantly alter the  survival of surface-deposited wastes and
hence would attenuate cumulative effects  of waste deposit during
winter months.  In the  present  study,  fecal  coliform  in goose
feces lost viability after 4-5 days on the beach area  in winter
months,  whereas  survival exceeding  20  days was indicated in goose
wastes during warmer months.  Although  not  quantified, rapid
mortality in winter  months appeared also to be  related to the
moisture content of the  fecal pellet,  with moister  pellets
exhibiting more rapid die off  during  the freeze  thaw cycles.
This would support the conclusions  of  Weiser and Osterlud (1945)
and Kibbey et al.  (1978) who hypothesized that extracellular ice

                              23

-------
damage was the  mechanism involved in mortality of  JS^ coli at
freezing  temperatures.

    Thus, it appears plausible  that  the  fecal  coliform levels
observed  at  discharge points in Buttermilk Bay are  the result of
a resident  pet and  wildlife animal population,  although no
conclusions could be drawn regarding which of these  was the major
contributor.   The  effects  of  the  resident  animal  population are
emphasized  by an  extensive drainage  system  which  opens
considerable  surface deposit area to direct discharge to  the  bay.
This effect  is further  accented  by  land  use practices necessary
for the prevention of excessive flooding  of residential areas.
These contentions  appear  to  be substantiated by study elsewhere
under the National Urban Runoff Program and others which found
that  intensity of land use  in  a  surrounding watershed  has a
negative  impact on stormwater quality, often unrelated to actual
input  of human sanitary  wastes.  Although it is  known that
failing subsurface sewage  disposal systems  (failing to the point
where septage is on top  of the ground) have occurred in  the past
(  Carl  Wakefield,  Wareham  Board  of  Health - personal
communication) and that  these  wastes may contribute sporadically
to the overland or pipe discharge, in general public awareness
and the  nuisance  aspect of odor associated with these system
failures compels their immediate repair.


Release of fecal colifora from protected reservoirs

    Within Buttermilk  Bay both the sediments and decaying eel
grass and other debris remaining after the tide  ebbs (henceforth
referred to  as  "wrack") have been determined to  act as protected
reservoirs  and accumulators of fecal coliform.   The second major
category of   stormwater  impact to this  marine  system is the
dislodging  or  elution  of fecal coliform contained in  these
protected reservoirs,  which  causes  their return to the water
column and a compromising  of the water quality in the area.

    The accumulation of fecal coliform in marine sediments has
been  observed by many investigators  (Rittenburg et al. 1958,
Sayler et al. 1975,  Van Donsel  and Geldreich 1971, Volterra et
al. 1985, Gerba and McLeod 1976,  Erkenbrecher 1981, LaLiberte and
Grimes 1982), and was supported by preliminary observations in
our study area. Sediment-overlying water  samples  from  selected
cites in Buttermilk Bay (Table 1)  taken by disrupting  the
sediments in a 1 sq. m and  sampling in  the boil suggest two
conclusions.   Foremost,  it appears that certain  areas of the bay
contain  sediments  with  the capacity to accumulate  fecal
indicators,  and secondly,  these fecal coliform can return  to  the
water column by physically disrupting the  sediments.

    Qualitative observations at the time  of sampling suggest  that
accumulation of fecal coliform in the  sediments of Buttermilk  Bay
was related to  two factors:  the  organic  content of the sediment,
and the proximity of the area  to a contamination source. Samples
taken at Electric Ave. (public beach),  Sewell Cove and Gibbs

-------
Narrows (in the  cut) showed very little organic  matter or "fluff"
following disruption of the bottom;  these  samples concurrently
showed  little difference  between the  overlying water before
disruption of  the  sediments and after.  Samples in the  Red  Brook
area,  the area near Hideaway Creek  and Station 4 contained  light
fluffy sediments and concurrently  contained higher fecal  coliform
concentrations in  the  boil  (turbid  water sampled after  sediment
disruption).  Proximity to a  source  as an important factor  in
determining the  fecal  coliform content of sediments as was
demonstrated by  the Gibbs Narrows samples of  8/29/86 and 9/11/86.
Although  these  samples were  taken at  approximately the same
locations, it  was  noted  that on 9/11/86 a large flock of seagulls
was feeding on exposed clams in the area during the time prior to
sampling.   This  site  had  an  intermediate  degree  of organic
material in the  sediment.
TABLE  1.  Overlying water - suspended sediment fecal coliform
densities   (MPN  fecal coliform/ 100  ml  sample)  from selected
sites  around the  perimeter of  Buttermilk  Bay,  southeastern
Massachusetts.   All  fecal  coliform  were  further  differentiated
and identified as E.  coli.

Sample Date    Location      Fecal  Coliform    Fecal Coliform
                               per  100 ml      per 100 ml
                             Overlying Water   Water + Sediment

3/05/86        Hideaway Creek     <2              33 *
3/05/86        Red Brook  (Tidal
               pool)               13            350 *
3/05/86        Electric Ave.      <2              <2
3/05/86        Sewell Cove        <2              <2
3/05/86        Gibbs Narrows      <2               5
3/05/86        Station 4          11              79 *
8/29/86        Gibbs Narrows       5               2
8/29/86        Red Brook  Mouth    49              79
8/29/86        Miller Cove         4              21 *
9/11/86        Gibbs Narrows       8             240 *
9/11/86        Red Brook  Mouth    27             110 *
9/11/86        Miller Cove         8               2

* Sediment + water sample value outside the 95% confidence limit
of the water value  as defined by a standard MPN table.


     The organic content of  the suspending medium  has been shown
to be an important factor in prolonging the survival of  various
enteric organisms  in  marine (Orlob 1956, Vaccaro et al. 1950,  Won
and Ross 1973), freshwater (Sinclair and Alexander 1984),  and
terrestrial systems (Mailman and Litsky 1951, Tate  1978).   Some
investigators,  notably Gerba and  McLeod (1976), Hendricks  (1970)
and Hendricks and  Morrison  (1967) have demonstrated the ability
of enteric organisms, including E.  coli, to utilize  nutrients
extracted from sediments  for growth.  Thus is appears that, not


                              25

-------
only  do sediments  have the  potential for  protecting and
accumulating  fecal  coliforms,  but they may also support their
growth in proportion  to  the available nutrients.  This mechanism
gives an added implication to  the  input of nutrients from  on-site
subsurface septic waste disposal.  While this practice  may not
result in the  actual input of enteric organisms to the bay (see
section  on Septic  Systems)/  the input of  nutrients from this
practice which was evidenced (Valiela et  al. 1987),  may result in
the multiplication of enteric organisms,  to include pathogens, in
the receiving  sediments.

   The accumulation  of  fecal coliform in sediments  may result in
their resuspension during rainfall events  (Schillinger and Gannon
1985,  Roper and Marshall 1974).  Although  the  mechanism is
unclear, mechanical disruption of the  sediments  at discharge
points for stormwater as well as a more generalized changing of
the adsorptive capacity  of sediments for bacteria in response to
a sudden influx  of freshwater is suspected.  The ability of fecal
coliform to become  resuspended  following physical disruption has
been demonstrated in our study and is supported  by  Grimes (1975).
This  may have additional implication for hydraulic clamming
operations  in  areas  where  the sediments are laden with excessive
fecal indicators.  A single set  of samples taken in October, 1968
amidst a hydraulic  clamming  operation in Gibbs Narrows indicated
that  in  this  case,  the sediments were  not accumulating fecal
coliform.

    In addition  to  sediments,  deposits left with  the receding
tide  (wrack)  which  were primarily dead and  decaying ellgrass
(Zostera  marina) were examined for  fecal  coliform.   Large
deposits approximating 5-10  kg per meter of  beach were common
throughout the summer  months. Subsampling one  gram portions of
eel grass by placing  it  in sterile buffered  water and performing
tests for fecal  coliform (MPN) showed an extremely high degree of
variability ranging  from below  the detectable  limit  to  > 24,000
FC/100 gram.  No obvious correlations with area  were noted with
the exception of very  high  (>1,000 FC/gram)  values which were
generally noted in areas where waterfowl,  primarily  Canada geese,
had been observed in days prior  to sampling.  During the rain
event of November 6, 1986 subsamples of approximately 2  kg were
taken from  six sites (concurrent  with stormwater drain sampling)
and placed in a sterile tray.  The samples were then squeezed to
extract the rainwater they had collected and the resulting sample
was  cultured with MPN methodology.  The results  are  presented  in
Table 2.

   These results suggest that  in addition to  serving as  a
protective reservoir for fecal coliform, decaying eelgrass and
other matter in the wrack line  serve  as  a  diffuse source of
bacteria following the  percolation of  rainwater through  it.   It
is  likely  that  during the  following  inundating tide,  fecal
coliform will be released to the  bay from this source,  possibly
resulting  in  severe  degradation of water quality in addition  to
that imparted by  the  existing stormdrains.  Although the
hypothesis has not been tested, it is likely  that even under  dry


                              26

-------
conditions,  the  inundation of contaminated wrack results in some
degradation  of water quality in  the  immediate area.


Table  2. Fecal  coliform densities in rainwater  extract from
decaying eelgrass  found at selected locations  in Buttermilk Bay,
southeastern Massachusetts,  1986.  Numbers are expressed as fecal
coliform per 100 mis of extract.

Location               Fecal coliform/ 100 ml extract

Electric Avenue  Beach              13,000
Mouth of Red Brook                  2,300
Approximately 30 m from mouth
    of  Red Brook                 >16,000
Hideaway Cove                    >16,000
Little  Buttermilk Bay, at Old
  Head  of Bay Rd.                 >16,000
Puritan Avenue  (near stormdrain)   16,000


    Although the source of  fecal coliform in dead and decaying
eelgrass is  not  definitely know^n,  wastes  from  both waterfowl and
scavenging  animals  is  strongly suspected.  In conjunction with
weekly bird  census collection discussed  later,  weekly counts of
animal  scats on  selected  sections of  beach  indicated  on  an
average of   2 dog  scats per  100 m of shoreline  and an abundance
of Canada goose  feces (during their  presence  in  winter months and
late summer).  This fecal matter when rolled into the decaying
eelgrass by water motion on the incoming tide  is  apparently
afforded a  protected site where fecal indicators can survive for
extended periods and possibly multiply.

    Due to  the  variability of the data,  no attempt has been made
to determine the  actual  fecal  coliform loading  from  dead and
decaying wrack,  however it does appear  that in certain situations
the effect may  be  considerable.


Impacts from surcharge with contaminated groundwater

    Although not specifically studied within  the confines of the
present study,  local increases in  groundwater flow  velocities
associated with rainfall, must  be addressed  as to  their possible
implications to  bacterial contamination within the bay.  Sampling
groundwater at  selected sites  around Buttermilk  Bay (Figure 8,
Table  3)  indicates  that groundwater  entering  the bay  does
occasionally entrain  fecal  indicator organisms,  although the
sources  of  these organisms were  not determined.   Since the
mortality  of fecal coliform  released  from  any source  is a
function of  time, all  factors  affecting the residence time of
fecal coliform in groundwater prior  to the  release  into the
surface water must  be  considered as important in defining the
overall  impact of contaminated  groundwater  to  the  bay.   The
increased entrainment distance  of fecal indicator  organisms


                             27

-------
       Indian
       Heights
            Jefferson
            Shores
                                      *'*;•&''->v• ^$]C^  W^g^J-;
                                      •.' '*!i ^';^\^^"$^XmorvBourps^ra^

                                      •feAi^of.VO&•.;-%' k^j^^^ •/'dem^iJi
                                                                      B
Fig. 8   A)  Approximate  locations of  groundwater  sampling stations
         in  Buttermilk Bay,  southeastern Massachusetts 1986.
         B)  Reference map to indicate relative development.  Source
         USGS, 1979.

                                   28

-------
following  rainfall has been demonstrated elsewhere  (Bouwer et al.
1974 and Hagerdon et al.  1978, Bitton and Gerba  1984).

   The mechanism by which  rainfall expedites the transport of
bacterial  organisms in groundwater is likely twofold.  A primary
consideration in  bacterial entrainment,  as well as  virus
entrainment  discussed later,  is the  distance  of  travel of
bacteria through an unsaturated vadose  zone. Saturated conditions
are generally more  conducive  to  bacterial  entrainment to
groundwater  (Moore and Beehier 1984).  Under certain  recharge or
rainfall events,  the areas beneath the  most probable source of
fecal  coliform (septic systems)  may  approach conditions of
saturated flow in some  situations.   If this becomes the case
increased numbers of fecal coliform reaching the groundwater,
coupled with increased  lateral  flow  velocity  of  groundwater,
could result in more  enteric organisms reaching the breakout
point in the bay.   In addition to the  passive role of  moving
more fecal organisms in a  shorter time frame, the  flooding  of a
subsurface system  with rainwater, which would  have  a different
ionic characteristic than septic  effluent, may cause  the elution
or  desorption  of  bacteria from the  soil matrix as has been
observed to occur  with viruses  (Lance et al.  1976 and Landry et
al. 1979).

     Thus  it appears from studies elsewhere that transport of
fecal coliform and other  enteric organisms  in groundwater is
facilitated by recharge events.  This phenomena may, in part,
account -for  some of the water  degradation observed  in Buttermilk
Bay  following rain events, however results discussed later in
this report (see Septic Systems) suggests that this impact is
comparatively minimal.

SEPTIC SYSTEMS

     Evidence  that   septic   system  effluent   affects  the
bacteriological quality of  groundwater, often for considerable
distances, has been presented from a  variety of studies conducted
elsewhere.  Studies summarized by Hagedorn (1984) indicate the
entrainment  of various enteric organisms for distances of  0.6-830
m (2-2723  ft) and survival times to  27  weeks.  The major factor
affecting the entrainment distance  of enteric organisms is the
soil type. Factors  most important in determining the  survival of
enteric organisms  in  subsurface soil  systems are  defined as
moisture  content, moisture holding capacity, temperature, pH,
organic matter and competition/antagonism from soil flora  (Gerba
and Bitton 1984).

     In general,  these  investigators conclude  that the greatest
survival  of enteric organisms in  soil systems  occurs in
conditions of moist soils  with  good moisture-holding capability
at low temperatures in more alkaline soils  (pH above  5) that are
devoid of  normal flora.  In addition,  there is evidence that  when
sufficient  amounts  of organic  matter  are  present, increased
survival and possible regrowth can occur.
                             29

-------
     Groundwater sampling in the Buttermilk Bay watershed during
1986 presented conflicting  indications regarding the entrainment
of bacterial indicators in groundwater entering the bay.  The
sporadic occurrence  of  fecal indicators  from selected sites
around the bay (Table 3, Fig  8.) gave initial indication  that
organisms can be  entrained in lateral  flow for at least the
distance  from the point of discovery to the  nearest septic system
(ca. 35 m).  In some situations, however, where  septic systems
were less than  5  m upgradient  from the sample withdrawal point,
no fecal  indicators were isolated from  samples taken.  Further
indication  of very  limited  mobility of fecal indicators  in  soil
came from  an in-depth study of a septic  system located  in  a
periodically-saturated flow  situation and a  septic system located
located  within  70  cm  of  groundwater.   The  first septic  system
was chosen due to  the fact that during  a  portion  of the tidal
cycle,  the zone  surrounding a portion of  the system  was
saturated,  and thus would be most conducive  to indicator organism
entrainment in groundwater.  The system would be classified as
"operating" even though at the time of  our  study  it was under
administrative  orders  for  repair due  to  the  potential for
contamination of the bay. The  fluid level in  the  second leaching
element of this  system was   near the  top of  the  element
suggesting that aging of the  system had  significantly impaired
its leaching ability.  Nutrient analyses (Valiela et al. 1987)
were used to confirm  the sampling  locations as being within the
boundries of a septic plume.

    Sampling  at this  first site on  two dates  (Table 3, Figure  9)
indicates very efficient removal  of fecal indicators by  soil
(>99.99  % removal) within  7  m of lateral  flow, with  further
indication that the major removal takes place within 2  m. The
limited  entrainment  of fecal  indicators  has  been reported
elsewhere utilizing various soil types.   Using soils of 80%,  41%
and 7.6  % sand, Brown et al. (1979) observed that, on only a few
occasions were fecal indicators isolated  from effluent  120 cm
below  septic  leaching lines.   Under  saturated  conditions,
Hagedorn et al. (1978)  indicated that  wells  3,000 cm downgradient
from a fecal  coliform source did not show the  entrainment of  this
indicator.   In  what is  perhaps  the most  comparable study design
to  the area of Buttermilk Bay, Vaughn  et al. (1983) only rarely
detected coliform at lateral distances greater than  1.52 m from  a
leaching pool  having  coliform  densities  of 10 5   - 108
organisms/ 100 ml.  Collectively,  these studies suggest that sand,
even of  very low  clay  content is  an efficient restrictor  of
bacterial indicator entrainment.

     The second septic  system  investigated  was located on the
western  shore of Buttermilk  Bay and again only  showed  limited
entrainment of  indicator bacteria (Figure 10).


     In light of data  indicating both very limited entrainment of
bacterial indicators as  well as entrainment to distances of  35  m
or  more,  we can only conclude  that site specific information on a
more substantial number of septic systems in the area would be

                              30

-------
                 SEPTIC   SYSTEM    SURVEY
                         approx. mean high-water table
                             . water table at time of sample
            SCALE (le et )
                            /-^-Sampling location and observed value (or:

                            Fecal  Conform/WO mis
                    p 3 6,0 00
                        approx. mean high-water table
                        appj-ox. water tab^e_at_time <^f_sa_mpl_e_
o
-------
                   SURFACE ELEVATION
                                                RETAINING WALL
      o » o. e> a.
      o o o a f,
      fl e  »» o »
      o   c- e> t>
       t> O o t> o
                         GROUNDWATER ELEVATION
                        V
                                     < 2
        O < 2

        O 22
        O < 2

        O < 2          O < 2
Sampling location and observed value for:
O < 2
r* < 2
                                                  SCALE (feet)
               Fecal  Conform / 10O mis
                    SURFACE ELEVATION
                   V
                                       RETAINING WALL





o » a o a-
o «> o a t.
0 to e> ^ ^
B * O o o o



	 . 	 • 	 ^_____
GROUNDWATER ELEVATION
V
o 
-------
necessary  in  order to assess the overall  effect that on-site
subsurface  sewage disposal practices  have on the area as a  whole.
It is our  general  conclusion that  the direct contribution of
fecal bacterial indicators  from subsurface systems  via the
groundwater route  is fairly minimal.  In  contrast,  however, the
effect of  on-site sewage disposal  in  relation  to  virus and
nutrient contribution in groundwater, discussed later in this
report,  likely has the  more  substantial negative  impact on
surface water  quality  in  the bay not  suggested by the traditional
fecal indicator system. Study is presently underway which will
sample a larger  number of septic systems, which should refine our
assessment  of  impact from this type of  source.
Table 3.  Summary of groundwater analyses performed in Buttermilk
Bay, southeastern Massachusetts,  1986.   Samples  were taken near
the high-water mark  using a shallow-well point sampler at a depth
of approximately 1  m. All fecal coliform densities reported as
number per 100 mis of  sample.  For Locations  refer to Figure  8.
                                                 Log10
Location   Date  Sample   High Value   Number of   Mean   Nearest
                  size    observed    positive    value  suspect
                         FC/100 ml     tests            source
   A
   A
   B
   C
   D
   A
   H
   I
   B
   E
   F
   D
   F
   G
   B
5/20
6/9
6/23
6/24
6/24
6/25
6/25
6/25
8/13
8/13
8/13
8/13
9/3
9/22
V
9
9
8
8
3
4
4
4
8
1
2
2
6
12
15
 130
 110
   7
  11
 279
   0
   3
   3
   0
>160
 3
 2
 4
 0
 0
 1
 0
 0
 4
 0
 1
 1
 0
 0
13
14
15
3
0
0
11
0
0
31
0
3
3
0
0
ca.
ca.
ca.
ca.
ca.
ca.
ca.
ca.
ca.
ca.
ca.
ca.
ca.
5-5
25
25
35
25
10
25
25
15
35
35
1
5
1
0
m
m
m
m
m
m
m
m
m
m
m
m
m
m
ca.  35  m
WILDLIFE

     The most obvious source of  fecal coliform in the Buttermilk
Bay  system is  the  indigenous  wildlife.   Fecal  coliform are
present  in the intestinal  tract  of  a  variety of  warm-blooded
animals  (Geldreich et al.  1962) which inhabit the area,  and  have
also been reported to occur in some fishes (Geldreich and Clark
1962).  The implications of this source are substantial since
fecal waste from wildlife often enters the  system unabated by any
land treatment. In general,  fecal coliform loading  from wildlife
has two main components:  direct fecal deposits resulting from use
of the bay by aquatic-oriented species and fecal  matter which
results  from  terrestrial  deposition,  which subsequently  	u~
                                              washes
                               33

-------
into the bay with  stormwater.  This latter component could be
expected to receive  some treatment in its overland passage to the
bay,  however,  there is evidence that some land-deposited wastes
contribute  to  and  maintain a protected reservoir  of fecal
coliform  which acts as  a diffuse  source  under  certain
hydrographic conditions  (see "Release  of  fecal coliform  from
protected reservoirs").

      Following techniques employed by Hussong et al.  (1979),
theoretical  values for coliform inputs from waterfowl on  a 24-hr
basis were  calculated based on weekly waterfowl census  data
collected  from  December,   1985 to November,  1986  (Figurell).  Per
capita,  per  diem  estimates of 1Q7   fecal coliform per goose, 10y
fecal coliform per swan and 109  fecal coliform per duck  were
used based on Hussong et al.(1979) and Koppelman and Tanenbaum
(1982).  Estimated daily  fecal coliform loadings  varied f^0111 8 X
10' during  the month of June to a maximum of 3.1  x 10iu fecal
coliform during January when  maximum use of the bay by  waterfowl
was observed.   To place these values in  perspective, a projected
resultant  fecal coliform  density  in receiving water  was
calculated.  To perform this calculation, certain assumptions
were made  to include  the availability of two tidal  prism  volumes
(2.5  x  106  cu.  m  x 2  or 5.0 x  106  cu.  m)  for  dilution.
Simplistically,  this volume  represents the dilution   volume
available  during  the 24-hr period (2 tidal cycles),  which  is  used
in conjunction with the  fecal coliform load per bird per 24-hr
period.   Based on  these calculations, at no  time would the
predicted  fecal coliform  density in a uniformly-mixed bay exceed
2 FC/100 ml.

     Although  the assumption of  uniform  mixing is inappropriate,
generally our field  measurements  of  fecal coliform densities
concurrent with waterfowl usage of  the  area  (Table 4)  coincided
closely with  the  predicted estimates  for  more   "open  water"
stations  (Stations 1,  16,  10.3, 15,  4a and 14).   As might be
expected,  fecal coliform  densities tended  to be  higher at
locations  which concurrently tended to  concentrate  waterfowl in
less exchanged  situations (Station 3.5,12a,  6H2 and Red Brook
Stations  5.9S and  5.9B).  Although some of the fecal coliform
observed at these later stations may have been related to the
increased presence of waterfowl, it should be noted that other
effects were also responsible,  as  evidenced by  their continued
tendency  to exhibit  higher fecal coliform densities  than  open
water stations  during times of little  or no  waterfowl usage (ie.
June and July).

     Thus, in  general, it appears that  in well-exchanged,  open
waters of  Buttermilk  Bay, predictive estimates of fecal coliform
loading from waterfowl approximate  observed values and  indicate
little  compromise of the  water  quality.  However,  since the
distribution of waterfowl in the bay is not uniform,  and since
the flushing of the bay with tidal  water also  fails to exhibit  a
uniform pattern  throughout the bay, local degradation of water
quality due, in part, to concentration of waterfowl in sheltered
shoreline areas should be expected. In  contrast to the apparent

-------
Table. *f .  Summary of fecal coliform and Escherichia coli  densities at routine monitoring
stations  in Buttermilk Bay, southeastern Massachusetts, March  -  October  1986.   For
locations of stations see Figure 3.   All densities  expressed as number of organisms / 100
mis. Note that analyses from January to October, 1986 differentiated E. coli  from among
other fecal coliforms. E. coli  densities are expressed parenthetically beneath the fecal
coliform densities.
                                        STATION

Date       1

09/05/85   3

09/25/85  10

12/16/85   7

01/13/86   2

02/19/86  11


03/05/86  <2

3/17/86   <2

4/15/86   <2

05/08/86  <2

6/04/86   2

6/19/86   2

7/24/86   <2

8/27/86   <2

10/01/86  14
                 (2)   (5)
                                         (above 6H2 at bog 230 FC/lOOml"
ND= No data
* Analyses performed was a membrane filtration  method and  values observed were within the
95% confidence limit of the MPN technique employed on  the routine sample.   These  data
suggest that on these dates  that there was no significant upstream-downstream effect.
16
<10
ND
2
8
23
(18)
10.3
1
27
<2
27
<2

15
<10
20
<2
<2
2

12a
ND
ND
2
<2
2

6H2
ND
ND
<2
<2
23
(8)
5.9S
88
70
8
<2
30
(18)
5.9B
13
<10
4
5
8

(Upstream of 6H2 at
<2
<2
<2

2

<2
5
<2
4

4
(2)
<2
<2
<2

13

<2
2
<2
5

13
(5)
<2
<2
<2

<2

<2
2
<2
4

2

ND
2
<2

<2

180
79
34
240

540

ND
<2
<2

110

130
49
>2400
920
(above
170

ND
2
49

79

22
240
49
170
6H2 at bog
240
(130)
ND
2
<2

<2

22
49
70
20
1490
50

4a
<10
<10
<2
<2
41

bog - <2
11
<2
8
(5)
8
(5)
<2
<2
2
•A1
FC/100ml
7
(2),
3.5
7
<10
<2
11
19

FC/100
ND
11
5
(2)
13

17
5
2
8
(5)
4
(2)
14
ND
ND
<2
ND
ND

ml)
ND
ND
ND

ND

ND
<2
ND
14

2


-------
   30
-------
minimal effects  of  fecal matter directly  deposited into  the bay,
there is increasing evidence  that  the shoreline deposits provide
the opportunity of fecal wastes to exhibit cumulative effects.
The  extent of  this  cumulative effect  is related  to  the
association of  the wastes  with protective  elements  such  as
decaying eelgrass and other debris, temperature and moisture
content.  The implications of fecal matter  accumulation  and
extended survival of fecal indicators directly relates to  the
probability of  their  reintroduction to the water of the bay.
Again,  it appears  that  hydrographic conditions  are the main
determinant  of this reintroduction.   If  terrestrially-deposited
wastes  are  stranded above  the high tide  mark  for a  period
exceeding the major  mortality period of fecal  coliform,  then
their reintroduction  has little significance.   If,  however,
spring tides in  conjunction with rainfall cause fecal coliform
dissociation in conjunction with inundation of the wastes,  we
would expect diffusion of the wastes from the source proportional
to its volume and resultant degradation of water quality.   In
situ  experiments to determine the viability  of goose  feces
(See APPROACH AND METHODOLOGY) suggest that during the  spring and
autumn,  fecal coliform maintain their  viability essentially
unaltered for 2- 3 weeks.  During winter months a viability of 1-
10 days  is  likely.

     Our investigations  of fecal  coliform inputs  to Buttermilk
Bay from wildlife,  particularly waterfowl,  are perhaps the  most
controversial aspect of the overall investigation.  Due to the
assumption that large concentrations of  waterfowl, occupying
strategic areas with respect to shellfish or recreational-use
areas,  can result  in  severe degradation of  water  quality,  a
number of municipalities have  instituted bans  on the  artificial
feeding  of  waterfowl.  These  regulations have  drawn  varying
degrees of criticism from various groups  which generally contend
that the regulations are not warranted and constitute an undue
measure which  deprives the public of  the  pleasure of feeding the
birds and  observing them  at  close range,   and  deprives  the
waterfowl of an available food source to sustain them  through
times when food is scarce .  The purpose  of this portion of  our
investigation was  to  clarify the role  of  wildlife,  primarily
Waterfowl, in contributing  to  the fecal coliform loading into
Buttermilk Bay.   Although it  is evident  from  fecal coliform
loading  rates of various  waterfowl  that the  potential  for
immediate as  well as cumulative impacts  on water quality exists,
it was not within the scope of work  or our ability to determine
what percentage of the observed waterfowl were present due to
artificial  feeding  practices.    It  is  evident  from  our
investigations that waterfowl concentrations do negatively impact
the water quality in receiving  waters. Therefore, any effort to
curb the practice of artificially encouraging increased residence
time of waterfowl would appear  to  be  justified.

     In addition to the  controversy surrounding the feeding of
waterfowl,  there  has been increasing concern  whether  the impact
of  waterfowl waste  should  have public  health significance
attached to it.  This issue can be dissected  into  two basic


                             37

-------
questions which are as follows:

      1) Do waterfowl harbor pathogens transmissible  to humans ?
 and
      2) Are there currently available cost-effective methods
         for the determination of waterfowl vs.  human wastes ?

Both  of  these  questions have  been periodically embroiled in
controversy. Numerous  investigators have indicated the presence
of  Salmonella in avian feces ( Berg and Anderson 1972, Faddoul
and Fellows 1966,  Mitchell and Ridgwell 1971) however,  other
investigators  (Hussong  et  al.  1979)  have suggested  that this
pathogen  is  not  ubiquitous  in waterfowl.   These  later
investigators additionally  failed  to  isolate Shigella spp. from
migratory waterfowl in Chesapeake  Bay,  however enteropathogenic
E._ coli was isolated  in  a limited  number of waterfowl from this
area.   In an attempt to determine the  incidence of Campylobacter
jepuni in waterfowl,  Hill  and Grimes (1984)  failed to isolate
this pathogen from cecal,  water and sediment samples  in a flyway
in Wisconsin, however the  literature review  presented therein
indicated that this pathogen had been isolated from waterfowl
elsewhere.   Thus  it appears  that certain bacterial pathogens are
carried in  waterfowl,  and hence  the  possibility of  transferring
the pathogen to the receiving water via the fecal route appears-
to exist.  It should be noted that there  are a number of unknowns
regarding  the  transmittance of  human  pathogens via avian wastes
which have  not been  investigated.   In light of the  lack  of
information,  it appears prudent  to  assume t.hat fecal  wastes
deposited from  waterfowl do have public  health significance.
Since specific  information  does  not exist  to  characterize these
wastes, application of the presently accepted standards .for fecal
coliform indicators appears justified.

    If it was concluded  that fecal wastes from waterfowl should
be considered less of a public  health  risk,  there  would still
remain the  question of source differentiation.  With investigation
by  Geldreich and Kenner (1969)  there has been popular use of the
fecal  coliform/fecal streptococcus  ratio  to attempt  source
differentiation.  In general,  if  the ratio is greater than 4, the
source is generally considered human.  If  the ratio is less than
0.7  then the higher fecal streptococcus values  indicate an animal
source. Underlying the  employment of this ratio is  the assumption
that  the sample  is withdrawn at  a point that would  not allow
differential die off the  indicators.  Numerous investigators have
made  comment  regarding the differential  die off of  these
indicators in  various  situations  (Ostrolenk  et al.  1947,
Van Donsel  et al. 1967, Davenport  et al. 1976 and Borrego et al.
1983).  In general, it is popularly believe that sources may be
differentiated if sampled within 24 hours of the source.

     The question of  whether  the  fecal   coliform/fecal
streptococcus ratio can be used to differentiate waterfowl wastes
was  specifically addressed by Hussong  et al.  (1979).   Data
presented by these  investigators using wild waterfowl migrating
through the  Chesapeake Bay area suggests that  it  is  not  possible


                             38

-------
to separate  avian  fecal contamination from human  wastes based on
this ratio.  These  investigators suggested that  diet  was the
determining  factor  in  controlling  the  fecal  coliform/fecal
streptococcus ratio.   Wild feeding ducks exhibited ratios typical
of human enteric flora.   It thus appears  that  should waterfowl
waste be determined  to have  less public health  significance, the
presently-accepted means for differentiation has been shown to be
unreliable.  Although typing  of fecal  streptococci  for  the
purpose  of  determining  more  definitely  the source  of
contamination is possible,  to date, cost-effective timely methods
have not been developed and incorporated into  routine sampling
schemes.

BOATS AND MARINAS

   The  effect  of  boating  activities  and  marinas  on  the
bacteriological quality of  water can  be segregated into  two
general categories,  the  actual  input  of sanitary wastes  and the
secondary effects of resuspension of coliform-laden sediments
into the water  column.   Both of these  effects are,  by  nature,
extremely variable and intermittent making assessment of  impacts
on  a study  area difficult.   While it is generally  believed that
a release of sanitary wastes in the vicinity of shellfishing or
recreational  waters  is an  unacceptable practice,   circumstances
can often come into play which compel  a  marine  craft owner to
disregard or occasionally violate this common sense intuition.
In general it can  be assumed that the actual input of sanitary
wastes by marine craft into the  nearshore marine environment near
shellfish harvesting or  recreational waters will bear direct
relationship  to  the  level of  convenience or  cost associated with
disposing of the wastes properly.  The majority  of marinas do not
have convenient  means of disposing of  sanitary wastes.

     In an attempt  to assess the  impact of  marine  craft/marinas
on the bacteriological quality of the water in Buttermilk Bay,
two  sampling designs  were  employed.    Since  fecal coliform
generally survive for  longer  periods in  sediment,  and  thus
sediments might  act  to integrate water  quality over an expanded
time frame  (as  opposed to  a grab  sample  of  overlying water)
sediment samples were taken at various locations  in the Bourne
Marina (Fig. 12).  Sediment  and  overlying water  samples failed to
indicate any  measurable impact  on the bacteriological quality of
the water or the sediments by marina operation.  In addition to
this study,  a sampling team responding to a complaint of "feces
in the water" near Fries' Marina sampled upstream of the  tidal
flow, various points below the boats in the marina and downstream
of the  tidal flow.   Again,  these samples failed to show any
measurable impact of marina operation  or marine craft discharge.

    In  interpreting  these results it  should  be  noted that by its
nature  waste  disposal  from  marine craft  would occur  only
intermittently  and in a  covert manner.   Present regulations
prohibit  this  practice  in  coastal  waters of  Massachusetts.
Therefore to actually observe this practice and measure its
effect  would necessitate extremely  fortuitous  circumstances as
                             39

-------
Indian
Heights
                                                                    LITTLE
                                                                  BUTTERMILK
                                                                     BAI
                                                Fig. 12 Locations of marinas sampled during
                                                       1986 in Buttermilk Bay, southeastern
                                                       Massachusetts.
                                              BOURNE
                                                         ARINA

-------
have been reported to occur in other portions of Buzzards Bay
(Beskinis - Personal  communication).  A second consideration to be
made prior to drawing conclusions regarding marina operation from
the present study is the nature of the  two  marinas studied.
Fries'  marina  primarily services small day-trip boats, the size
of which is  restricted  by  the  height  of the Route  6  bridge
restricting ocean access. The Bourne Marina is atypical due to
the pump-out  facilities which are not common among marinas and
which would  allow owners to properly dispose of sanitary  wastes.
It  is  concluded  that,  in  general,  marina operation will
periodically result  in the bacteriological degradation of the
surrounding water in direct relationship to the inconvenience
imposed by boat operators to dispose of their wastes by proper
means.   Other  investigators,  notably  Faust (1978)  and  Garreis et
al. (1979) have  observed  significant measurable impact due to
recreational  marine craft,  to include increase in  indicator
organisms.

     In relation to the impact of  marine craft operation and the
resulting  turbulence and  resuspension of coliform-laden
sediments,  the  impact in Buttermilk Bay would appear  to be
minimal.  Most  marina operations are of  an  adequate depth to
ensure  that  propeller  turbulence does not  cause  resuspension of
sediments.

FRESHWATER INPUTS

    There are  five  significant  freshwater  surface inputs to
Buttermilk and Little Buttermilk  Bay  which have  continual flow
throughout the year .  Generally, these provide  continual inputs
of  fecal coliform,   the  importance of each dependent  on  its
relative  flow.

    Red Brook enters  Buttermilk  Bay at its  eastern shore and
comprises the most substantial surface freshwater input to the
Bay. Its headwaters  are  White  Island Pond in Plymouth County.
The primary  portion  of  its surface watershed  is relatively
undeveloped, comprised mainly of a  series of cranberry bogs.  A
survey of the area  indicated  no direct discharges  with  the
exception of runoff from Head of the Bay Road.  Fecal coliform
densities taken at the point  where Red Brook passes under Head of
the Bay Road have shown continually high (>14 FC/100  ml) values
as least as early as 1973.  Samples taken by regulating officials
on  10/25/73  and 11/02/73  indicated densities  of  280 Total
Coliform/36  Fecal Coliform per 100 ml and 750 Total Coliform/36
Fecal Coliform per  100 ml sample respectively (D.E.Q.E.  - data
files).  The sources within the watershed of  this  brook are
evidently too diffuse for  identification.   A  comparison of
surface  samples  taken  near  the mouth of Red Brook  with
concurrently-taken lower strata  samples (Table 4)  indicates   that
a surface plume emanates from Red Brook on the surface, as would
be  expected,  with the  freshwater  flow.   Thus,  the  extent
laterally in the bay to which the effect  of Red  Brook could be
increasing the fecal coliform densities,  will  be highly dependent
on the mixing and dilution  characteristics  encountered when

-------
entering  the  bay.

   We believe that there are at least  two  factors leading to the
contamination of this brook with fecal coliform.   Initially a
considerable  wildlife  population as evidenced  by  animal tracks
and droppings probably provides a considerable fecal  input  which
enters the brook  essentially untreated.  To accent the effect of
animal droppings, the floodplain of this  brook is periodically
flooded due  to tidal action,  thus washing the surface load of
contaminants into  the  brook. Several  other  factors likely
contribute to generally-higher fecal  coliform densities in Red
Brook  to include the  protective nature  of  the  Marsh  (from
destructive  effect of light),  the  nutrient availability, and the
moisture  retaining  characteristic of  the soils.  Many studies
have shown that  the  natural fecal  coliform background levels are
high and extremely variable in  many freshwater wetlands (Kadlec
and Tilton 1979).

    Two additional freshwater surface  inputs,  one from cranberry
bogs to the north of Head of the Bay Road near  Hideaway Village
and one on the eastern shore  of Little Buttermilk Bay  entering
near  the  end of Little Bay Lane in Bourne  both have  similar
characteristics  of  fecal  coliform contamination as interpreted
from data of D.E.Q.E.- F.D.A..  Again,  we  believe  that  the
productive watershed  operates much in  the same way as described
for  Red  Brook,  however,  these flows are  generally  less
substantial  than Red Brook and have a more  local impact.   The
flow  in both of  these streams  can be  restricted by artificial
means to accommodate the needs of cranberry  culture.  In one
instance, the flow of Hideaway  Creek was negligible  due to this
practice.  A restriction of flow and consequent flooding of the
bogs  may  alter the resultant  fecal coliform loading,  but this
effect could not  be measured in the present study.

    In the course of the  present study, a  small stream  with
headwaters  in Goat Pasture Pond in Bourne was investigated for
fecal coliform inputs.  The relative  seasonal changes in fecal
coliform inputs  generally followed similar trend as the stream in
the  vicinity of Hideaway Creek.  No overt  discharges  were
observed in the  watershed  of  this stream-pond system however it
was noted that this area was used as a nesting area for waterfowl
which may, in part,  account for the values observed.

    A small stream entering Little Buttermilk Bay via a culvert
under Old Head  of the  Bay Road was periodically measured for
fecal coliform.   The surface drainage  area of this small stream
contains a corral in which at least one horse  was  observed during
each  monthly sampling.   This,  in  part,   may explain  the
sporadically high coliform densities observed at the point of
discharge.  No other overt  discharges were observed.

    In summary,  it was evident that all freshwater surface inputs
to  the  Buttermilk  Bay System contain fecal  coliform  on a
continual basis.  In the case of Red Brook, the volume input is
substantial  and  may  be the cause  of unacceptable fecal coliform

-------
densities  along  the  path of travel  (generally along the western
shore).  Input from the  Hideaway Creek stream enters the bay in a
restricted cove  with  considerable sediment build up.   This,
coupled  with  a high  nutrient load are likely the  reasons for the
consistent local  degradation of  water quality.  In addition, a
joint DEQE-FDA  study during 1985  found  significant  increases in
total and  fecal  coliform below the  Hideaway Village development
which may  suggest  contribution by  adjacent septic  systems via the
groundwater. The  remaining  three  freshwater inlets  would be
expected to  exhibit  even more  localized effect due  to the lower
volumes  of flow.

POINT DISCHARGES

    Only one  point discharge  into Buttermilk Bay was discovered
in  the  course  of our  investigations.,  the 10  inch  diameter
concrete  pipe  discharges  from  lobster  holding  tanks  of  the
adjacent  fish market.  Samples taken from  this  discharge on 5
dates (Table  5) showed  fecal coliform densities ranging from 172
- > 16,000 FC/100  mis.  Sampling adjacent  water on  15 July, 1986
indicated that within  ca. 15 m in all directions from the pipe,
levels were reduced from 1300 FC/100 mis at point  of discharge to
13-33 FC/100 mis.   It is difficult  to determine  whether values
observed in adjacent water were  totally  the result of the pipe
discharge.   A  substantial  rodent population  living  in  the
adjacent rip-rap could also be a fecal coliform source.


Table 5.  Fecal coliform  densities of  discharge water  from
concrete pipe adjacent to fish market.

   Date                      Fecal Coliform/100 ml sample

6/9/86                         172
6/11/86                         >  16,000
7/8/86                         >  1,600
7/15/86                         1300
7/17/86                         240


FURTHER  CONSIDERATIONS

     In  the  process of  interpreting  fecal coliform  data and
estimating fecal coliform loading  from various sources,  it became
apparent that small scale experiments  were necessary to verify
the applicability  of published studies  in reference to  mortality
rates of  indicator  organisms.   An  intimate knowledge  of all
factors affecting mortality of fecal coliforms is paramount to
determining  the  significance of organism density  measured  in  the
field.  The purpose  of this portion  of our  study  was to
determine,  in  a preliminary way,  whether factors existed in
Buttermilk Bay  which would invalidate comparison  with other
studies  (Rittenberg  et  al.1958, Borrego  et al. 1983, Gerba et al.
1977, Bellair et al.  1977 and  others).

-------
   3-0-
£
o
VJ
IS
u
QJ
   2.0-
   1.0—
                                                            A  Fresh  water Exposed

                                                            A  Fresh  water Control

                                                            O  Saltwater Exposed

                                                            •  Salt water Control
                                                          -r = 0.92
                                                                                    -r = 0.97
                20
40        60

    time
 \
80
              100

(minutes )
120
140
160
         Figure 13  Fecal coliform die-off with  time in freshwater and saltwater  samples immersed in
                  1-2 mm. of water in Buttermilk Bay, July 17, 1986.

-------
   Upon entering an estuary or receiving water,  coliform bacteria
are  influenced by a  number  of  factors which  tend to  limit
population  growth and induce mortality.   A typical mortality
curve for Escherichia  coli, indicates up to  90% mortality in 3-5
days (Mitchell 1968, Canale et al.  1973).  A number of factors,
summarized  below, have shown to act in a regulatory manner to
coliforra in  marine environments:

     1) Solar Radiation (light)

     2) Temperature

     3) State of association with sediments

     4) Availability of nutrients

    While certain aspects of these  factors  have  been  previously
discussed and  will not  be  reviewed here,  what follows  is  a
summary of small  scale experiments designed to explore very site-
specific possibilities regarding  the ecology and  hence mortality
factors  of  fecal coliform  in  Buttermilk  Bay. In  addition,
pertinent  literature  is presented regarding  the   effect  of
temperature  on  fecal  coliform survival  in order  to fill in gaps
of understanding regarding  season effects on  fecal coliform
growth.

The effect of solar radiation

     Perhaps the  most  influential factor exhibiting adverse
effect on coliform organisms is solar radiation.  Kapuscinski and
Mitchell  (1983) relating their  in  vitro mortality rates for a
pure culture of E. coli exposed to  sunlight (T90 < 4.9  h)  with
those  of FoxwortEy and Keeling  (1969)  (T90 < 5.5 h)  suggested
that solar radiation is the principal determinant of microbial
mortality in seawater.  While some  reviews  have minimized the
effect  of  solar  radiation  (Hendricks  1978)  or neglected  it
altogether  (Mitchell  1968), other authors (Fujioka et al. 1981,
Bellair et al.  1977 and Garaeson and Saxon 1967)  have indicated
substantial bactericidal  effects  of sunlight  penetrating  to
depths of 4m.   Kapuscinski and Mitchell  (1983)  reviewing studies
of Webb and  Baker  (1979) and others  indicated that wavelengths of
solar radiation shown  to  induce  the  highest  mortality  of E. coli
are reduced by only 10 fold  at  ca.  2  m in enriched seawater.
Preliminary  experiments  in  our study area conducted during sunny
days with both fresh  and salt water show close  agreement with
work  of Fujioka  et  al.  (1981),  and  indicate   a substantial
bactericidal effect  of sunlight  for  fecal coliform  (Figure  13).
These data strongly  implicate  sunlight as a major factor in the
decay  of  fecal  coliform  populations  in  openwater areas  of
Buttermilk Bay.   This  contention is supported by  Chamberlain and
Mitchell (1978) who provided convincing argument based on work of
other authors  (notably Gameson and  Gould 1975 and Foxworthy and
Kneeling 1969) that the variability in coliform decay rates in
seawater  can  be  primarily attributed  to  the  variability  of
surface light  intensity  and other factors influencing the depth

-------
profile  of  light intensity and  bacterial concentration.

     Preliminary investigation at selected sites in Buttermilk
Bay indicate that at certain locations, notably Hideaway Creek
and Red Brook,  where fecal coliform densities are generally at
unacceptable levels for shellfish harvesting,  the ultraviolet
light penetration  is  attenuated in comparison to open-water
stations (Figure 14).  Although the reason for this attenuation
is not  known,  it  is  correlated with a  comparatively  higher
nutrient content (Valiela et  al.  1987).  In  addition  to nutrients,
natural  products  of  plant decomposition  (tannins  etc.)  or
extracellular products of algae may be  providing  the ultraviolet
light absorbing  compounds.  Excretion of  such compounds  from
marine  algae  has  been reported  (Craigie  and McLachlan 1964).
These data  initially suggest  that nutrient  addition may alter the
ultraviolet  absorbing  characteristic  of  receiving  water  in
certain situations which may  result in an attenuation  of the
fecal  indicator mortality.   This  may  explain  in  part,  the
generally higher fecal coliform densities at these two sites.

     Our initial experiments  suggest that,  in areas of  high
nutrient input,  the ultraviolet  light  absorbance  characteristic
of the receiving  water is modified  to allow greater survival of
indicator organisms, and likely pathogens as well.  This  adds to
the  implication of nutrient  inputs  from septic systems  in
affecting  the overall microbiological quality of the receiving
water.

The effect of temperature

    While  it  is generally  believed that there is increased
survival of fecal  coliform at  lower  temperatures and conversely
that mortality  increases with temperature (Vasconcelos and Swartz
1976) work by other investigators, notably Hendrichs (1972) and
Won  and Ross (1973) suggests  that  this relationship is  highly
dependent on ambient nutrient  levels.  Hendricks  (1972)  observed
that below a sewage effluent, maximum growth of enteric bacteria
was observed at 30  C. Won and Ross (1973) observed that mortality
rate decreased  at 22 C depending  on the concentration of nutrient
to include extract from autoclaved feces.  Vaccaro et al. (1950)
found that nutrient addition  extended the survival time  of E^_
coli threefold at higher temperatures compared  with raw  water. It
thus appears  that nutrient  rich  water may exert a  second
attenuating effect on  the mortality of fecal  coliform by
providing the nutrients  for cell maintenance and  growth.  We are
unsure at what temperature  nutrient  enrichment has no effect on
fecal coliform growth, however Shaw et al. (1971) has suggested
that the minimal temperature  for growth in glucose minimal medium
was  between 7.5 and 7.8  C.


The effect of state of_ association with sediments

     Many investigators  have indicated that bacterial  survival is
enhanced by association with  sediments,  however the mechanism for


                              1*6

-------
                            MAXIMUM CELLULAR DAMAGE
200
Figure
                    250              300
                Wavelength  (millimicrons)
350
          Companian of ultraviolet light absorption among selected
          stations in Buttermilk  Bay.  Open water  stations were
          represented by station  1, 3.5, 14, 16, 10.3 and Station 15,
          (Figure 3)  Hideaway Creek was sampled at Station GH2.

-------
this protection  is not fully understood  (See "release of coliform
from protected reservoirs"). An experiment conducted on  August
11,  1986 in which  contaminated water samples were buried beneath
approximately l-2mm  of sediment  in plastic bags confirms that at
least  one of the  protective  mechanisms  of  sediment  is  the
screening  of  light  (Figure 15).  Samples  concurrently  placed
beneath 1-2 mm of water experienced complete fecal  coliform  die
off  within  2 h, whereas samples buried  in  the  sediment
experienced significantly less die  off over the  same time  period.
Variables  such  as dissolved  oxygen and temperature remained
constant over the  2  h period.

    In addition  to protection from antagonistic factors  such as
light  ,  sediments  can  apparently  offer  a  media  for  growth
depending upon the nutrient availability.  Hendricks  and Morrison
(1967) found  that E^ cpli.  growth was  supported at 16 C  on
extract from bottom  sediment.  Gerba and McLeod (1976) observed
growth in fecal coliform from sediments  in areas receiving sewage
effluents as  well  as areas free from such inputs.

    There are two major implications of these findings  to  the
Buttermilk Bay system.  If,  as noted by Schillinger  and  Gannon
(1985),  a significant portion (16-47 %) of bacteria in  stormwater
runoff become associated with sediment,  the  nutrient level of  the
receiving sediments becomes an important factor  in estimating  the
significance  of  this portion of the stormwater bacterial loading.
For instance,  if  stormwater pipes  are located as to discharge in
an area receiving  substantial  nutrients through the  groundwater,
increased bacterial survival and  perhaps growth  of settled
bacteria increases the  significance of  the  stormwater discharge
by prolonging its negative impact.  In addition to stormwater,  all
other sources such as waterfowl feces,  groundwater inputs etc.
from which fecal coliform  settle out into the sediments may also
utilize  nutrients  in  sediments  and  hence  have  their  effect
prolonged.


Availability of  nutrients

    Numerous studies have  thus far been reviewed suggesting that
fecal coliform can utilize nutrients in  the  water column and  the
sediments. These studies will not be reviewed here.  Although the
origin  of  the  nutrients  has  not  been  documented,   the
possibilities include natural decomposition, septic system wastes
via groundwater,  nutrients from fecal wastes deposited directly
in the water and stormwater runoff.   What began as an  incidental
experiment in conjunction  with study of nutrient  enrichment  and
algal succession  (Costa - Boston University Marine Program) gives
initial indication of an additional source of  nutrients for fecal
coliform,  that of algal products.  As  part of an experimental
design to examine  algal growth in response  to nutrient  gradients,
acetate  strips were left in the bay  at  various locations in
respect to a nutrient source (Red  Brook) to assess algal  growth.
Concurrent with the collection of  the strips for chlorophyll
analyses, 10 cm x 1  cm strips were removed and  placed  in  sterile

-------
o

\
 E
 v*
 O
 TJ
 VJ
 QJ
 U,
 10
                   I
                   15
30
45
 I
60
90
                                                    sediment

                                                    control

                                                    exposed
120
ISO
                                      TIME    (minutes)
            Figure 15 - Fecal coliform die-off with time  in samples buried to a depth of 1-2 mm
                       of  sediment compared with samples immersed in 1-2 mm of water in
                       Buttermilk Bay, August 11, 1986.

-------
phosphate buffered water.  Concurrent water samples  were also
taken. The samples were taken to the lab and processed using MPN
methodology  previously described.   The samples  were then stored
at 20 C under subdued natural light (laboratory ambient light).
After 7 days, the density of fecal coliform was  again measured in
all samples. The results (Figure 16) showed that those samples
containing  the  acetate strips with algae growth experienced 1-3
log 10 increases in population of fecal coliform, while  those
water samples taken from the  same  areas without the algal  strip
cultures  experienced  at least  90  % mortality.   These  data
suggest  that  the  algae  were providing nutrients capable  of
supporting growth of fecal coliform. The excretion of organic
compounds by various  species of marine and  freshwater  algae has
been  reported (Hellebust  1965,    Larsson and   Hagstrom
1982,McFeters et al. 1978, Nalewajko et al. 1980).  These  later
investigators reported that the doubling time for Pseudomonas
fluorescens was reduced from  2-19  days to 2  hours  in  mixed
culture  with  Chlorella sp.   This  algal  genus  was  also  the
subject  of  study by  McFeters et al.  (1978) who observed that the
supernatant  from cultures supported growth of fecal coliform.  In
general  it appears  plausible from our investigations as  well as
published studies that at least  some species  of marine  algae
produce  extracellular products which  can support growth of  fecal
coliform.

     The  implications of this  "secondary"  effect of nutrient
enrichment in Buttermilk Bay  are uncertain.   It appears plausible
that in certain situations where water  exchange is limited and
nutrients concentrate in  amounts that support algal growth,  fecal
coliform  may derive nutrients  capable of  maintenance and possibly
growth.  Collectively,   these  issues of  the  implications  of
nutrient  enrichment and  its effect on the bacteriological quality
of the receiving water warrant further research.
                             50

-------
—  5.0
E
o
o
              4.0
           N.
           5
           Q
           U

           •—  2.0
           VJ
           0)
           Cl
           Q
                              WATER
                             • WATER + ALGAE
           -_  5.0-

           e

           o
           S  4.0.
              3.0H
           Q
           vj

           ~.  2.0-
           VJ
           01
           u.
           en
           o
                  WATER
                  • WATER + ALGAE
— 5.0-

€
            2 4.0.
            s
            V.
              3.0
            1J
            u
            Oj
           u.
            Cl
            Q
                            — WATER
                            -— WATER + ALGAE
                      DAY
                                 DAY
                                  7
Figure  16- Growth of fecal coliform in the  presence  of
             indiginenous  algae at 20°C under laboratory
             conditions.

-------
USE OF ALTERNATE BACTERIAL INDICATORS TO EVALUATE SOURCES AND
               DEGREE OF BACTERIAL CONTAMINATION

    Almost since  the  acceptance of the total and fecal coliform
standard, researchers  and  health officials have  searched for
alternate indicator  organisms  for the  classification  of
recreational waters  which would  more  accurately  reflect the
public health risks.   The  problems  associated  with  the
development  of an indicator  organism  system  fall  into  two
categories:

     1) Cost effective methods  for analysis  must be  developed,
and
     2) It must be demonstrated that under "average  conditions"
        the organism(s)  is(are) consistently  correlated with
        the occurrence of human pathogenic organisms  which
        produce an observable effect (ie.  waterbourne disease
        outbreak).

    The  search  for  alternative indicator  organisms  for
recreational waters basically stems from  the fact that it is not
possible to analyze water for the myriad of  pathogens which could
be present.  In addition,  methods for direct measurement of some
pathogens, notably the viruses,  have not been developed.

    With the  exception of the enterococci, which has gained some
acceptance as  an indicator of public  health risk in  contact
recreational waters, other bacteria,  although not  incorporated
into  a defined indicator organism scheme,  have  found  use as
diagnostic tools in evaluating contamination sources.   In the
course of the  present  study,  certain alternate bacterial
indicators were investigated  to  assess  their  utility  in
determining contamination sources.  Among those indicators chosen
was  Escherichia coli (a subset of the fecal coliform group),
Clostridium perfringens,  fecal streptococcus  , and a subgroup of
the  fecal streptococcus  group  which includes  strains  of
Streptococcus  faecalis and Streptococcus  faecium called the
enterococci.The  following summarizes  this author's experience
with each of  these  alternate  indicator organisms,  with an
analysis of how each  organism's density patterns related to the
presently-employed  standard  for  the evaluation  of  shellfish
harvesting areas,  the fecal  coliform group.   For regression
analyses,  data  were first Iog10 -transformed.

    In general, all  of the indicators investigated  have been
shown  in  various situations  to have  sanitary  significance.  The
difficulties  in applying  results from one study area to another
however, are  numerous since the variability in characteristics of
each study site are often substantial.   Generally, the shallow
closed embayment served by subsurface sewage systems has  not been
widely studied.  Thus,  it is questionable whether data collected
in the majority of published studies relating  to  sewage  outfalls
in more open,  exchanged areas is applicable  to our  study area in

                             52

-------
 Buttermilk  Bay.   At  minimum,  however,  it can be stated  that
 alternate indicators are useful  tools in  the whole  arsenal of
 public  health officials  which,  if interpreted  in light of
 specific  situations,   can  possibly   serve  to  delineate
 contamination  sources.


 Escherichia  coli

    One  of  the most  useful  alternate  indicator  organisms
 investigated  was  Escherichia coli.  This subset of the fecal
 coliform group is  generally considered more fecal specific, since
 the  fecal coliform system  includes the  genus Klebsiella, an
 organism which is not  fecal specific.  Although Klebsiella is
 infrequently  present in human feces, a  substantial number of
 extra-enteral  sources have been noted (Seidel  et al. 1977,  Dufour
 and Cabelli 1976). In addition to  the the existence of extra-
 enteral sources,  it  should  be  noted  that  there  is no
 epidemiological link implicating the transmission of  Klebsiella
 infections via the waterbourne route, further  negating its use as
 a  bacteriological standard  for  recreational water.   The
 persistence of use of the fecal coliform system in spite of the
 shortcomings of one of  its components appears  to be the result of
 historical  misconceptions  (Cabelli  et al.  1983)  and lack of
 definitive  investigations  relative to  alternative indicator
 systems.

    The usefulness of fecal  coliform differentiation in our study
 area was diagnostic. Since  there is indication from published
 studies  that  Klebsiella sp. can multiply outside  of  a warm-
 blooded  hosts^Tt was  beneficial  to rule  out extra-enteral
 sources  of a  non-fecal specific  "fecal coliform"  (Klebsiella)
 such as decomposing marsh situations which are common in the
 Buttermilk Bay watershed.

    In all but seven cases during  monitoring at routine  sampling
 stations,  fecal coliform  in samples were comprised entirely of
 Escherichia  coli (Table  4).   This  accounts  for the high degree of
 correlation  between fecal coliform and E.coli  densities  (r=  0.98)
 in routine samples (Fig. 17). In addition, all fecal coliform
 observed in  stormwater samples were  further differentiated as E.
 coli.

    Although, as stated,  E._ coli  is considered  more fecal
 specific, our investigations and  published studies reviewed
 elsewhere herein,  do suggest  the possibility of extra-enteral
 sources  of  E.  coli.   If  this is the case,  the usefulness of
 differentiating members of  the  fecal coliform system is reduced
 to determining whether Klebsiella is  a component of the fecal
 coliform densities being observed.  In the  event that Klebsiella
 is the major component in any situation, this information may
 redirect effort away from investigations of human sources, which
are often quite expensive and time-intensive.   In the event  that
E.  coli is the major component of the fecal coliform densities
observed, the indication  is  less  definitive  and leaves the


                             53

-------
o
o
     100
LU
      10
                                               Y = X ± .lit
                                               R =  .98
                                    D
                          DD  D
                            D
                                 10                   100

                                   Fecal  coliform  (#/100 ml)
           Figure 17 - Relationship between fecal coliform and E. coli
                      densities in samples collected at routine
                      sampling stations in Buttermilk Bay during
                      1986.  Data were log-,n transformed.

-------
possibility  of  either  contamination  of  sanitary  significance or
significant  contribution by extra-enteral  sources.


Clostridium  perfringens

    Despite  the fact that Clostridium perfringens is  consistently
associated with human  wastes(Akama  and Otani  1970,  Haenal 1970
and Drasar et al. 1975 as cited in Cabelli 1980),  it is presently
not incorporated  into  a  definite fecal  indicator system.
Although densities of this organism were generally low at routine
sampling stations in the present  study (Table 6), some general
patterns did emerge.  With  the exception of one sampling date,
Clostridium  perfringens showed highest densities at  Station 6H2
near the Hideaway  Village development.  This may suggest the
input  from septic system leachate in the area,  since  it has been
shown  elsewhere herein (See SEPTIC SYSTEMS)  that C._  perfringens
is found in  septic  system leachate  in  densities exceeding 103.
The poor correlation between this indicator and fecal coliforra
(r=.099) is somewhat expected (Figure 18). The  culture technique
used in this study  enumerates  the spore of  this  organism,  which
is resistive to adverse environmental  conditions.  It is thus a
conservative  tracer  which would indicate a  contamination source
long after the  fecal coliform experienced  die off.

    In addition to  station  6H2, the  occurrence of C._  perfringens
was consistent  at the mouth of Red Brook (Station 5.9)  and in the
inner  portion of Miller  Cove (Station 3.5).  While the occurrence
of £._  perfringens at Miller  Cove may  again indicate contamination
from groundwater sources (ie. septic systems), since 50% of the
groundwater  samples  taken in this area  on 6/23/86  were positive
for £._ perfringens,  there is no direct  evidence  for  this source
from groundwater samples taken  at  the mouth of Red Brook.

    Another possible source of this  indicator could be waterfowl
and pet fecal material deposited near or at  the shoreline.  A
survey of the wrack on July 25 at ten selected sites around the
bay indicated  £._ perfringens  at densities  ranging from <10
organisms/gram of  wrack (observed  at  one station)  to over 600
organisms/gram.  There was  no  correlation between C.  perfringens
and fecal coliform densities  in  the wrack at thTs  time.the
occurrence at 9 of the ten stations sampled however,  suggests
that the organism may be somewhat  ubiquitous in the strand or
wrack  material, with waterfowl and domestic animals as the  likely
source. Geldreich  (1977) for  instance  reported that  dog waste
contained  2.51 x 108 C^ perfringens per  gram  of feces.


    Thus  it appears that there are  two sources of C perfringens
entering the bay, surface  deposit  by  pets  and  waterfowl, and
groundwater  inputs  from septic systems.  While  the  groundwater
observations may point  to areas  of  septic effluent  inputs, the
lack of correlation with  the currently-used standard  (fecal
coliform) brings into question the role of this source  in the
classification  of the bay for shellfish harvesting.


                              55

-------
     Table. 6 .  Summary of Clostridium perfringens densities at routine monitoring  stations in
VJl
Os
Buttermilk Bay/ southeastern Massachusetts, March - October 1986. For locatior
stations see Figure 3. All densities expressed as number of organisms/100 mis.
STATION
Date 1
3/17/86 <10
4/15/86 <3
05/08/86 6
6/19/86 <3
7/24/86 <3
8/27/86 <10
10/01/86 <10
16 10.3 15
<10 <10 <10
<3 <3 <3
6 <3 <3
<3 3 <3
<3 <3 <3
<10 <10 <10
10 <10 <10
12a
20
7
<3
3
<3
<10
6H2
40
10
30
3
63
129
12
5.9S
10
7
6
3
<3
18
(upstream
5.9B 4a
<10 <10
7 <3
10 13
6 <3
<3 <3
9 <3
of 6H2 at bog)
3.5
40
3
33
<3
<3
<3
14
ND
ND
ND
ND
ND
ND
10
     ND= No data

-------
 O
 O
 CD
 cz
 CD
 CD
 C
 c_
 CD
 0.

 CD
 00
 O
I — 1
CJ
      100
L
      10
 D
 -
         -n
                                                                           a
                                   a
           a
D    a
                                     a
                     a     D
                                              D  a a
               DO  m
                .  .... i
              ODD
D  D D
D
D
                                10
                                          100
         Figure 18 -
                        Fecal  coliform   (#/100  ml)

         Relationship between fecal coliform and Clostridia perfringens
         densities in samples collected at routine sampling stations
         in Buttermilk Bay during 1986.  Data were Iog10  transformed.
         D —  in
                    R =  .10.
                                  57

-------
VJ1
CD
      Table.  7 .  Summary of fecal streptococci densities at routine monitoring   stations in
      Buttermilk Bay, southeastern Massachusetts,  March - October 1986.  For locations of
      stations see Figure  3.  All densities expressed as number of organisms  /100 mis.
                                             STATION
Date 1
3/17/86 <10
4/15/86 <3
6/19/86 <10
7/24/86 <10
8/27/86 <10
10/01/86 <10
16 10.3 15
<10 <10 <10
<3 <3 <3
<10 <10 <10
<10 <10 <10
10 20 10
<10 10 <10
12a
20
270
240
200
540
60
6H2
10
27
120
TNTC
600
950
5.9S
10
40
20
30
300
310
5.9B
10
3
30
290
60
10
                                                                        4a
                                                                        10
3.5
                                                                             <3
30

-------
o
o
 CD
 13
 U
 U
 O
 U
 o
4->
 Q.
 O)
 c_
4-J
 CO
 fD
 U
 
-------
FECAL STREPTOCOCCUS

    Within  the context of recreational  waters;>  the  fecal
streptococcus  group is generally used in a diagnostic manner to
determine sources of fecal contamination.  A discussion of the
use of the  fecal coliform/fecal streptococcus ratio is presented
elsewhere (see  WILDLIFE  section). This  indicator group generally
contains members with limited sanitary significance which has
restricted its use as a fecal indicator  (Geldreich  and Kenner
1969).    In our study area/  their occurrence was most consistent
at points of surface freshwater inputs  (Stations 12a,6H2, and
5.9- Table  7).  Their  significance could not be determined due to
studies  questioning  the utility of  the  fecal coliform/fecal
streptococcus ratio,  however there  was some correlation between
these two indicator groups (r= .70 - Figure 19).  Other studies
being conducted in Buzzards  Bay  (Larry  Gil,  DEQE,  Div.  W.ater
Pollution Control - personal communication) suggested  that a high
percentage  of colonies growing  on the K-F Strep media were false
positives  as  determined by confirmatory  tests  for  fecal
streptococcus.


ENTEROCOCCUS

     The use of this fecal  indicator group has received much
attention  in  recent years as an  indicator  of risk in  using
recreational  waters,  particularly  for  contact  purposes.  Cabelli
(1983) reports  a high correlation between enterococci densities
and gastrointestinal disorders  among bathers at New York  Beaches.
As a result of that author's findings,  some states have adopted
this standard  for use at bathing beaches.

    During the present  study, the occurrence of enterococci
paralleled  that of streptococcus as expected since  it  is a subset
of this group (Table  8).

    Two  main problems arise when attempting  to  interpret our
enterococci data. Firstly,  study by  Cabelli  (1983)  was performed
in  relation to situations  receiving large  point discharges.
Since these situations  typically  are in more  open, exchanged
areas, the  numerous factors present in a closed shallow embayment
receiving  smaller point  and non-point  discharges,  such as
Buttermilk  Bay,  have not been addressed.  In  addition to the
question of  applicability,  some  difficulty was experienced
counting the colonies due  to the growth of atypical colonies on
the media used.  This was particularly true for areas  where  there
were  bordering wetlands.    In  general  it was  observed that
enterococci  showed  little  correlation  with the fecal indicator
group (r= .46 - Figure 20). Due to the lack of  epidemiological
data  from comparable  sites, the sanitary significance of
enterococci  densities observed  could not be determined.
                             60

-------
Table.  8 . Summary of enterococci densities at routine monitoring  st
Bay, southeastern Massachusetts, March - October 1986.  For locations
Figure 3.  All enterococci densities expressed as number of organisms

                                         STATION
Date 1 16
3/17/86 <10 <10
4/15/86 <3 37
6/19/86 <10 <10
7/24/86 30 <10
8/27/86 70 <10
10/01/86 <10 <10
10.3
36
<3
<10
10
160
20
15
20
7
<10
<10
30
<10
12a
<10
3
70
50
550
30
6H2
<10
<3
30
TNTC
280
140
5.9S
82
53
<10
<10
80
310
5.9B
10
<3
10
40
110
500

-------
o
CD
     100  .
 CO
 13
 O
 C_>
 o
 o
 o
 c_
 cu
 c
LU
10  .
       1   .
                                               D
                                                       D



                                                       D
                                                                          D
               D   DD     DD
                                 10                   100

                                   Fecal  coliform   (#/100  ml)
           Figure 20
               - Relationship between fecal eoliform and  enterococci
                densities in samples collected at routine sampling
                stations in Buttermilk Bay during 1986.  Data were

                      transformed.
                                  62

-------
       VIRUSES - A LITERATURE REVIEW OF PERTINENT ISSUES


     While the issue of fecal coliform presence in  Buttermilk
Bay has been addressed  in the  course  of  the present study, it
should  be understood that  the presence of  an additional
biological component of domestic sewage,  the viruses, also has
public health  implications.    Due  to  the constraints  of  the
present study,  the impact from  viruses entering Buttermilk Bay
could not be evaluated  directly,  however, what  follows  is  a
comprehensive review  of published  studies relating  to virus
entrainment  in  groundwater and persistence in  the  nearshore
environment,  with  comment on the applicability of these studies'
results  to our area.

     More  than 110  different viruses  are known  to be excreted in
human feces (Goyal  1984).  While the number of viruses shed from
individuals  varies  .greatly depending  on the infection status of
the host,  10b to 107 virus particles per gram of feces has been
reported  by this author  as an average.  A  review  of  the
literature leaves  little doubt that human pathogens, notably
enteric  viruses   are  neither inactivated  nor  removed  in
conventional wastewater treatment systems (Dubois et  al 1979).
Since  on-site  subsurface  sewage disposal  units receive  no
chemical  treatment,  with  the  possible exception of  limited
disinfection  as  a  result of residual  chemicals  from laundry and
cleaning wastes, it is imperative to  ensure that system location
practices  allow  for adequate means of pathogen  removal by means
of  filtration,  adsorption  and retaining the  waste until
pathogenic organisms lose their ability to enter routes where
they will allow exposure to a human host in an  infective state.
In order to ensure this,  a sound  understanding of the process
and  factors involved in  pathogen travel and  survival  in
groundwater as well as the persistence  in the marine environment
is essential.  What follows is a discussion of these factors from
the published literature regarding viruses.

     At the outset, it should be understood that, to date, there
have been no comprehensive studies  documenting the impact of
bacteria  or  virus originating in  on-site subsurface  septic
facilities on adjacent  surface waters.   With  few exceptions,
therefore,  much of the  inferences made herein relative to  on-site
systems  originate from  data  collected  in association  with
alternate land disposal  techniques such  as rapid or slow sand
filtration beds,  septic  lagoons or irrigation  practices.  Since
collectively  these alternatives  as well  as  the on-site
alternative  have as their basic  component waste application into
soils, an  extrapolation  of data gathered with respect to pathogen
entrainment and  survival from  alternate methodology into  on-site
application  situations  appears  justified.

    With respect  to the  threat  of compromising  water quality of
adjacent surface waters with on-site septic  system practices, the
issue can be broken down into three  basic questions which will

                              63

-------
serve  as  the outline for the following  discussion and literature
summary.

    (1)  Do viral pathogens  have  the ability to become entrained
         in groundwater which recharges  surface water?

    (2) Can these pathogens  survive  for  extended periods  in
         groundwater?

    (3)  Can viral pathogens remain viable  for extended periods
         in the marine  environment ?

     A number  of obstacles  immediately present  themselves  when
trying to answer these basic questions,  particularly in reference
to viral  pathogens.  In  1983,  Vaughn  and  Landry  (1983) indicated
that the  field of  environmental virology  was in its "embryonic"
state.  To draw upon this same analogy,  it can be stated that due
to technical problems associated with  various aspects of virus
culture and monitoring the field to this date remains in its
infancy.

     Although  there is no epidemiological evidence linking on-
site subsurface  disposal practices with  disease  outbreaks  from
adjacent  recreational water  usage,  the  evidence of water-related
disease outbreaks involving  consumed water  (see Gerba et al  1985
for  a review of  the  literature  relating  to waterbourne
gastroenteritis  and  hepatitis)  is  considerable.  Thus,
investigation  into  the disease-transmission possibilities
involving recreational usage of  water receiving  septic leachate,
particularly in areas where  edible products such  as shellfish are
harvested  appear  quite warranted.  Within  the confines  of a
properly  operating  on-site  subsurface system,  groundwater
entrainment is the  only  vector by which pathogens  migrate   from
the disposal site.

     Groundwater  recharge enters adjacent surface  waters  by one
of two modes.   In most situations, recharge enters the surface
water laterally at some point beneath  the surface of the water.
At this point, biological contaminants  entrained in groundwater
must  "survive" the  passage  through the  sediment-water interface
or survive containment  in  those sediments.  It is  the lack of
understanding  of  the processes in this sediment-water interface
which calls into question the validity of applying information
collected  in  most  published  groundwater microbiology studies
toward  resulting  surface  water  pollution.   Simply  stated,
extended entrainment/survival of pathogens  in  groundwater may
have  reduced  implication  to surface  water  pollution if the
pathogens are unable  to "survive" the  passage into the surface
water.   For  this purpose,  the  later portion  of this report
summarizes the available data on the survival of certain
pathogens in marine  sediments.

     A second mode by which  biological  contamination  in
groundwater can reach  surface waters is commonly referred  to as
seepage  or  "breakout".   In this  situation,  hydrographic

-------
conditions  such as an ebbing tide, rain or snowmelt  recharge etc.
increase the  hydraulic head  or gradient  to  the  point where
groundwater  seeps or breaks  through the surface of the soil
adjacent to the  body of water  and  completes its path to the
surface  water  by way  of  an overland route.  During these
situations, biological  contaminants entrained must survive a
different, typically  more  variable  set of environmental
conditions,  during  its overland passage  to  be  considered a
possible threat to the public health.   For this purpose,
additional  environmental factors  and their effect on the survival
of various  pathogens  is also reviewed.


ENTRAPMENT OF VIRUSES IB GROUNDWATER

    In their  review of recent  literature  Keswick and Gerba
(1980) reported that various  studies indicated that viruses  could
be entrained in groundwater to distances up to 408  m  from their
point of  application.  It should  be noted, however,  that  reports
of lateral entrainment of viruses in excess of 70 m are  rare in
the literature.  The majority of  investigations relative to  virus
entrainment and  survival in groundwater have been in regard to
land application of domestic wastes and have indicated limited
mobility  of viruses  in soils.  Duboise et al (1976) in laboratory
studies using soil columns  indicated that  viruses migrated at
least 10  cm.   In  Eustis fine sand,  Bitton et al (1984) noted that
virus migration following sludge  application  was less than 33 cm.
McConnell et al  (1984) reported that,  applying seeded river water
at an application rate of 0.2m/h  through  soils containing higher
than  98% sand,   reovirus were noted to be retained within a
distance of 1.22  m.   Gerba and  Lance (1978) applying primary
sewage effluent  to soil columns of loamy sand soil indicated that
only on one occasion  were viruses detected at the 250 cm. depth.
Laboratory  experiments of Dubois et al (1974) using sandy forest
soil indicated a 98.6% removal of Polio I and 99.6% removal of T
7  Virus  within  19.5  cm.   of sandy forest soil.   In other
laboratory  experiments Wang  et al  (1981) comparing  4  soil  types
and  using Polio I  and Echo I  viruses found that with the
exception of Rubicon Sand, in which there was a penetration of at
least 87 cm., other soils  tested removed  at least  98% of the
viruses within 67 cm.  While laboratory  experiments of Lance and
Gerba (1984) indicated virus movement of  160 cm. under saturated
conditions, this author noted  considerably less entrainment
distances (40  cm) when unsaturated conditions were tested.

     In apparent  contrast to these studies,  field investigations
indicate a more  extensive  entrainment of  viruses.   Mack  et al
(1972) isolated Poliovirus Type 2 from a 30.5 m-deep well  located
91.5 m from the edge of a wastewater drain field.  Wellings et
al.(1975)  describing  the penetration  and entrainment  of  viruses
following  application of  secondary  sewage to a cypress dome
indicated from a  single   isolation of virus that  lateral
entrainment was 38 m from the  source at a depth movement to 3  m.
Vaughn et al (1983) in what is likely  the most  pertinent
investigation  to  our  study area of Buttermilk Bay, investigating


                              65

-------
leachate from  septic tanks  in  a shallow  sandy soil aquifer
observed virus  to be entrained to aquifer depths of 18 m and
lateral movement to a distance of 67.05 m.   This author, citing
additional work by  himself  and  coauthors also noted lateral
entrainment  45.7  m  downgradient  from recharge  basins  receiving
effluent following  tertiary treatment. Further work cited by
these authors reported a single isolation of coxsackievirus from
an observation well 402 m downgradient  from  a sanitary landfill.
Using a tracer  virus (coliphage  f2),  Schaub and Sorber  (1977)
have demonstrated lateral entrainment of virus to 182.8 m at a
site used for rapid infiltration of wastewater.

     Not all field studies indicate extensive lateral movement of
viruses.  Brown et  al (1979) concluded  from  studies  using
coliphage  that  120  cm. was adequate to minimize the possibility
of groundwater  contamination.   Gilbert et al (1976) observed that
when secondary sewage effluent was applied to sandy-loam soil
99.99%  of the viruses  were removed within 9  m of  passage.

     There are many possible reasons for the  variability noted in
the  previously  cited  studies regarding the mobility of virus
within  the soil column.  Foremost/ it should be understood that
the  field of environmental virology, to include  sampling
methodology  and isolation techniques has undergone significant
changes in  the past two decades.  Thus, there are significant
differences in the detectability  limits in many of the studies
reviewed.  Perhaps more important, however,  in accounting for the
variability in  entrainment values are  differences among the
studies in the  many variables  affecting the movement of viruses
in  groundwater.   Keswick  and Gerba (1980) in  reviewing those
factors affecting  the entrainment and survival of viruses in
groundwater have listed numerous specific  factors under three
broad  categories  of hydrogeologica1, biological  and
meteorological.  While each  of the specific factors affecting
virus retention  in  soil are discussed under  separate headings in
the  following discussion, the  reader should realize that no one
factor  can  be singled out as the  most influencing factor in
survival or entrainment of viruses in groundwater.  In a given
situation,  it  is  likely  that many factors operate in complex
concert to cause an observed level  of entrainment or  survival.
In  addition,  it should  be understood  that soil systems are
dynamic systems  in  which  many  influencing  factors undergo
periodic changes which in turn alter survival and entrainment
characteristics  of  pathogens.


Factors Affecting  the Survival and Bntrainment of Viruses in
                         Groundwater

Temperature

     Similar  to what has been reported  for  bacterial indicators,
temperature  has been reported  to exert a primary  influence on  the
persistence  of viruses in groundwater. A review  of  investigations
in  this area  (Duboise et al. 1979)  generally  indicates an

                               66

-------
inverse relationship between virus survival and soil temperature.
Dubois et  al.  (1976)  found that  survival  at 4  C  was greatly
prolonged over  20 C. During investigation of sludge amended  soil,
Tierney et al.  (1977) found maximum survival  (  at  least  96  days)
during winter  compared  with survival of only  11 days during
summer months. Under  winter conditions Damgaard-Larsen et al.
(1977)  observed  the survival of coxsackievirus  to  persist for 23
weeks.   Yates et al.  (1985) found that poliovirus  underwent a 1-
log reduction  in  titer in 3-5 days  at 26 C as opposed to 28.8
days at "lower" temperatures.   Hurst et al. (1980 b) concluded
that temperature was  one  of the most  important factors  affecting
virus survival  in soil systems.


Moisture  Content of Soil

    Numerous studies summarized by Vaughn and  Landry (1983)
strongly support the  role of  soil moisture in virus survival.
These  studies  generally suggest  that losses  in  soil  moisture
result  in higher inactivation rates for viruses and  suggest that
on-site subsurface  sewage  disposal systems situated  where drying
cycles could be achieved will exhibit the more  optimal  virus
inactivation  as opposed to  systems  situated where soils
underlying the  leaching  facility  retains adequate moisture  on a
continuing basis (such as near or in the  saturated zone).


Distance  to Groundwater

     The likelihood that  viral pathogens will be entrained in
groundwater and carried significant  distances from the site of
deposition is directly related to  the  underlying conditions which
facilitate the  travel  of  the virus to the groundwater table. The
two factors integrally involved in this process are  distance and
time.  If  the velocity  of an effluent  is  kept  constant,  the
probability that  a virus particle will  reach the  groundwater
table is  negatively correlated with the depth of the vadose  zone.
In close conjunction with the soil depth, however, the rate of
recharge  is clearly shown to positively affect the penetration of
viruses into the vadose zone (Wang et al. 1981 and Vaughn et al.
1981).  In short, higher recharge rates generally cause further
movement  of viruses  toward the groundwater table, and in some
cases  the  virus may reach the  saturated zone.  Thus  a  lower
recharge  rate in some  situations  may determine whether viruses
become entrained in  groundwater or get retained in the vadose
zone.
     Due  to  the electrochemical nature of  virus particles,  the pH
of both  the  sewage effluent and  the  soil  microenvironment
strongly  influence  the  adsorption of  virus. Although an in-depth
discussion of  the  electrochemical double  layer (ECDL)  of viruses
and the various reactions of  this layer is outside the scope of

                              67

-------
this paper, it is sufficient to  say  that, in general, lower pH
ranges tended  to increase  adsorption  of viruses  in studies
reviewed.  Goyal  and  Gerba (1979) using nine different soils
concluded that soil pH  was the single  most  important factor
influencing virus adsorption to soil.   Burge  and Enkiri  (1978)
noted an  increased adsorption of a bacteriophage  at decreased
pH.  Schaub and  Sagik  (1975) additionally  observed increased
adsorption at lower pH  using 2  Encephalomyocarditis viruses.
Conducting a study using  34 minerals and soils,  however,  Moore et
al (1981) did not demonstrate significant correlation of polio
virus type  2 with pH alone.

     Since  each  virus  has  a  specific  surface charge
characteristic and since  the  pH and ionic  strength of the aqueous
environment has a key role in determining the overall strength of
the charge  (and hence its mobility in soil), it is likely virus
type (discussed later)  is an important factor in  determining the
effect pH has on virus  entrainment.  In addition to this two
factor interaction, Sobsey et al  (1980) indicated that soil type
can also  modify the correlation of pH and virus adsorption.  In
general,  however, it does appear that lower pH environments do
enhance adsorption.   This author  additionally reported that, even
in  soils with  low adsorptive  capability  (i.e.   sands)  an
enhancement of the limited ability can be achieved by lowering
the pH.

Type of Soil

     The  majority  studies  reviewed  dealing with  adsorptive
capacity of soils indicate  soil type  is extremely important in
governing  the  mobility  of  viruses  in  applied  effluent.   In
general,  one  of  the key  characteristics  of soil  determining its
adsorptive  capacity is  the clay content.   Sandy and organic soils
are poor adsorbers of virus  (hence would allow for more extensive
entrainment), whereas clay soils  are good adsorbers (Keswick and
Gerba 1980).  This condition is  supported in study by Sobsey et
al  (1980),  Schaub  and Sagik (1975)  and others.   Conversely,
however,  Schaub and  Sorber (1977)  demonstrated  extensive
migration (ca. 183 m) of viruses  in "silty, sandy  gravel".

     Two  additional factors reported by Vaughn and Landry  (1984)
that  are closely related  to overall clay  content are cation
exchange capacity  (CEC)  and specific surface area  (SSA).   In
general,  these two factors are positively correlated with clay
content.   Burge and Enkiri  (1978) specifically mentioned these
factors noting that adsorption  of a bacteriophage increased as
these two factors increased.   Findings of  Funderburg et al  (1981)
using  this  same  bacteriophage  as well as poliovirus  1 and
reovirus 3 correspond well  with findings of  Burge and Enkiri
(1978).    While  correlation    with   these   specific  soil
characteristics (CEC and SSA) and adsorption  was  not  observed by
Goyal and Gerba  (1979)  these authors do concur that soil type is
a major factor affecting adsorption  of viruses,'reporting that
clay in soil enhances adsorption.


                             68

-------
     In addition to clay components,  insoluble organic components
in soil  have  been shown  to  affect adsorption.  Natural  soil
organic  matter has been shown  to be a weak adsorber of viruses
(Moore et  al  1981).   Studies  conducted using  septic sludge,
however,  seems to indicate the  ability of this organic  matter to
retain viruses  in upper  soil  layers (Bitton et al.  1984  and
Damgaard-Larson  et al.(1977).   Pancorbo  (1981) reported  that
viruses had more affinity toward aerobically digested sludge than
anaerobically  digested sludge.


Dissolved Solids

     Many studies have been  published  relating to  the various
affects  of dissolved  solids  on virus adsorption to soils.   As
mentioned  previously,  those  materials affecting the pH of  the
system can have  appreciable affect by modifying the overall
surface  charge  of the virus particle  (see  pH), and  hence  its
mobility in groundwater.  In addition, there are some indications
in the literature, notably study of Schaub and Sagik (1975)  and
Carlson  et al (1968)  that the metallic cation concentration
positively  correlated  with virus adsorption.  The later authors
noted  that divalent cations affected  virus adsorption more
positively than  equal  molar strength solutions of monovalent
cations.  In subsequent study  Bitton et al (1975) determined that
trivalent  aluminum cations  were more effective in  enhancing
adsorption than  divalent  ions.  Lefler and  Kott (1974)  found
differential  adsorptive enhancement abilities of soil columns
dependent on the concentration of the divalent calcium  cation.

     Studies by both Schaub and Sagik (1975), Moore et  al.(1981)
and  Scheuerman et al  (1979)  suggest  that dissolved  organic
material  negatively affects virus  adsorption.  In an experiment
designed to  separate  out the affects of  the water-soluble
organics  from  the  insoluble components in "muck" soil,   the later
authors concluded that the "Humic  substances"  within  the muck
soil interfered  with  the adsorptive capacity.  Lo and Sproul
(1977) using dissolved proteinaceous organic  materials found
that not  only  did the  substances compete for adsorption sites on
silicate  minerals, but caused  viruses to desorb and become mobile
in the soil column.  It was apparently this type of  interaction
that Sobsey et al (1980) observed when he found nutrient broth to
be the most effective treatment to elute  viruses. These  findings
stand in  apparent conflict  with  other studies by Gerba  and Lance
(1978) who  found that  organics present in primary on  secondary
effluent did  not  affect  virus adsorption and  Goyal  and Gerba
(1979) who  found no consistent pattern of affect relative to %
organic  matter  in applied effluent. In addition to  the prior
mentioned surfactants have  been  shown to diminish the adsorptive
capacity  of soils  (Dizer et al.   1984).

     It thus  appears  that in  general an  increase  in  total
dissolved  solids enhances virus adsorption  with the possible
exception of  high molecular weight  organic   fractions  which  may
compete for binding sites  in  some situations.  Di and  tri-valent

                             69

-------
cations  show consistent enhancement of adsorption capabilities in
studies  reviewed.

Virus Type

    It is  generally believed that  adsorption of viruses is the
primary  mechanism for  removal in on-site  subsurface  wastewater
systems.  Since adsorption,  in part, is governed by the overall
electrochemical nature  of the  virus,  which  is  a unique
characteristic,  it  is expected that  different viruses would
exhibit  differential  entrainment potential  in similar  soil.
Numerous investigators  have  verified this contention (Landry et
al. 1979, Gerba et al. 1980, Schaub et al 1980, Funderburg et al.
1981, Goyal and  Gerba   1979, Hurst et al. 1980,  Lance et al.
1982). The overwhelming  conclusion which emerges following review
of these studies  is that, to  this date, a virus type which could
serve as a predictive model  for  all  virus  movement  in  soil
systems has not  yet been identified.


Meteorological Events

     Virus particles adsorbed onto soil beneath a  septic system
by in large are of little public health significance so as  long
as they are not  entrained in groundwater and moved from the site
to  proximity to  a  susceptible host.   Despite  the  enormous
adsorptive capacity of some soils, numerous studies have shown
that the virus-soil particle association is  not necessarily a
permanent one which endures until the virus is deactivated.  A
number of factors have been shown to  cause desorption or elution
of  the  viruses  from adsorption sites.

     The most common occurrence  which  results in the desorption
of viruses in wastewater systems is rainfall. Sobsey et  al (1980)
and Landry et al. (1979)  found that  considerable quantities of
retained viruses were washed from  soil columns  of  sandy and
organic materials by simulated rainfall.  This phenomena was not
observed  in columns  containing more clayey  soils.  Using
deionized water, Lance et al. (1976)  observed that  viruses  were
desorbed and moved more readily through  soil columns.   The
elution of viruses  from sludge disposal  sites with  rain was
observed by Damgaard-Larsen et al.  (1977).


Predicting the   Movement of Viruses  in  Groundwater  Entering
                        Buttermilk Bay

    It  is evident from  the previously  presented review that the
movement of viruses in groundwater is subject  to a myriad of
variables acting  in complex concert.  Although the site- specific
information necessary for prediction of virus entrainment in
Buttermilk Bay has not been developed,  a  review  of  the  surficial
geology of the Buttermilk  Bay study area (Moog,  1987  and Weiskel
1986) does suggest that soil  characteristics are quite comparable
to those encountered by Vaughn et al. (1983) in the Long Island,

                               70

-------
New York area. This author noted that, in a shallow sandy-soil
aquifer, similar to our area,  viruses were entrained  laterally to
distances  of   least 67.05 m  and  to depths in the aquifer of  18
m. Despite  the lack  of  more site-specific data ( soil CEC, ph,
organic content etc.) it appears  feasible to  expect  similar
entrainment in our area.

SURVIVAL OF VIRUSES IK MARINE SYSTEMS

     In the event that viruses are entrained for  distances which
would allow their passage into surface waters,  it then becomes
important  to have  an  understanding of  the  factors affecting the
survival in the  receiving water.  For this purpose, the following
is a review of literature regarding the survival of viruses in
marine systems.

     Many  factors  affecting  the  persistence of  human  enteric
viruses in the marine environment are   similar  to those factors
affecting  survival of enteric bacteria.   Among these factors,
association with sediments, temperature, and  solar radiation are
the most  important determining  the persistence in the marine
environment.   In addition,  some  authors (Akin et al.  1976 and
Fujioka et al.  1980) have suggested  the  presence  of  a virus-
deactivating compound in seawater.

Association with Sediments

     Numerous other authors  have reported the accumulation and
persistence of  viruses  in sediments  (Smith  1978,  Metcalf and
Stiles 1968, Landry et al. 1983, LaBelle and Gerba 1979 and  1982,
and others). In comparing  persistence rates between  raw seawater
and  sediments  Roa  et  al.   (1984) indicated   that Polio and
rotaviruses retained their infectivity  for 19 days in association
with sediment compared with a loss of infectivity  after 9 days
when  freely  suspended.   Similar   results  indicating  the
stabilization  of viruses associated with sediments were  obtained
by LaBelle and Gerba  (1980) and Liew and Gerba (1980) using  Polio
1 and Echo 1 viruses.  Generally,  therefore it can be stated that
viruses can accumulate in sediments  and  experience prolonged
persistence due  to  their protective effect.


Temperature

     There is  general  agreement that in  marine  environments,  as
in  groundwater,  virus  persistence  is  inversely  related  to
temperature.  Goyal  et al.  (1984)  suggested that  the  extended  (17
months) survival of  viruses at a sewage sludge ocean disposal
site was due not only to protective  sludge-sediment matrix  but
also  due  to  the  generally low (ca. 7 C)  temperatures.
Poliovirus and coxsackieviruses in unassociated  state were  found
to be more stable at lower temperatures (Obrien and Newman 1977).
In addition to unassociated and sediment-associated viruses,
viruses in shellfish, in particular oysters, have been shown to
persist for  120 days at  7C (Metcalf  and Stiles  1965).  A

                                 71

-------
comprehensive  review  of literature  in  regard  to  this  aspect  was
presented by Melnick  and  Gerba  (1980).
Solar Radiation

    The negative  effect of solar radiation in surface waters has
been noted by a  number of investigators (Bitton et al.  1979,
Kapuscinski and Mitchell 1983).  A number of factors govern the
photinactivation  process to  include  association  with sediments,
depth  in  the  water column,  and presence  of blue-green  algae
(Bitton et al.  1979).  Although this author suggested that  light
inactivation was less important at depths exceeding 6  inches,
Kapuscinski  and  Mitchell  (1983)  have  suggested  that  light
wavelengths exceeding  370  nm  can cause  mortality and that  these
wavelengths can penetrate to depths of 5 m and experience only a
10-fold reduction in intensity.
                             72

-------
GENERAL SUMMARY AND CONCLUSIONS

    The causes  for the  bacteriological  contamination of
Buttermilk Bay can be separated into two categories: continual
inputs and inputs related  to storm or rain events.  It  is the
later category  which plays  the  major  role in the classification
of Buttermilk Bay for shellfishing  purposes and  to a great extent
has determined the public's perception of water quality  in the
bay in general.  The primary sources of  fecal coliform  during
rain events are discharge pipes serving  an extensive drainage
area  in the surface watershed  of  the bay.  The  degree of
contamination  from  any one pipe is related to  the intensity of
land use within its  service area, periodicity of rain events and
season of year.  In general  the drain servicing the most
intensively  used  land during summer months in which rain  events
are infrequent will exhibit the highest contamination  levels
during rain events.  Although the sources of fecal coliform in
discharges could not be positively identified, a large pet and
wildlife population in the area is implicated.  Comparison of
this  study's values with  nationwide  observations  under the
National Urban Runoff Program suggests that fecal coliform  values
observed at discharges are not necessarily  caused by  direct
sanitary waste input to the discharge systems.  An intensive
survey of the  areas serviced by all surface drains failed to
disclose any cross connections with the drainage system.

    In addition to stormwater discharges, the release of coliform
from protected  reservoirs,  such  as  the strand line and sediments,
also contributes to the increased  fecal coliform levels  during
rain  events. Although the fecal  coliform  loading from these
factors could not be determined, data  presented do  indicate the
effect may,  in  certain situations,  be  substantial.

    During periods of little precipitation, water quality  in the
bay  is generally  good with  the  exceptions  of  areas  around
constant  inputs.    These  constant inputs  include  freshwater
streams as well as  one intermittent point discharge.   Regarding
the freshwater  inputs,  historic comparisons  showing  high fecal
coliform values  as far back as 1973  in  the case of Red Brook
suggest that encroaching development is  not entirely responsible
for values observed  just prior to and during  the recent shellfish
closures. Point  sources within this  watershed  and the four
remaining were not located,   suggesting fecal coliform of a
diffuse source, possibly natural in origin.  Nutrient levels as
well as protective characteristics  of  the drainage basins suggest
the  possibility that indicator organisms  are subject to
significant modification to its  mortality  rate.   The single point
source discharge located near a  local fish market can be expected
to impact  the  immediate  area, the  extent of which  will be
determined by the local mixing conditions.

    The effect of waterfowl use  on  the  bacteriological quality of
water was found  to  have two components.  Direct fecal deposit
into the bay was shown to have minimal  effect in most situations,
while  beach deposition was  shown to  have  the potential for


                             73

-------
significant cumulative effects.

    The total effect of septic systems  on the bacteriological
quality of Buttermilk Bay could not  assessed,   but three main
possibilities for impact were  suggested.  Initially/ there is
some  indication that  fecal  coliform can be  entrained  in
groundwater near  the bay to distances of  approximately 35 m.  In
addition to bacteria,  an extensive review  of published studies
indicates  that viral pathogens are entrained to greater distances
in  groundwater.   Using  published studies which are  most
applicable to our study area,  it is possible that  viral pathogens
can be entrained for lateral  distances of at least 67 m,  even in
situations where the  subsurface  septic system of  origin  is
properly  located  in accordance with present regulation.   The
variety of variables involved  in predicting  virus  movement in
groundwater,  however, precludes using this value  as  a definitive
guideline.

     Small scale  laboratory  and in-situ experiments suggest the
possibility that nutrient inputs, some  of which can be traced
to  septic systems,  affect  the bacteriological  quality of the
water additionally by altering  the mortality  rates of organisms
in  the receiving  waters.  Two major  mechanisms are  suggested to
include an altering  of  the ultraviolet light  penetration of the
water and  providing nutrients for maintenance  and  possibly growth
of indicator organisms and pathogens.

-------
                   APPENDIX I




DESCRIPTION OF DISCHARGES ENTERING BUTTERMILK BAY
                       75

-------
                          APPENDIX ±

         Descriptions of_  Discharges into Buttermilk Bay

Preliminary  notes;   All drains have been observed during dry
conditions.  With the exception of locations referenced as # 10
(Marsh Stream), # 11 (Groundwater recharge at Old Head of Bay), #
17  (Wychunus  St.  Drain),#  21  (Inner Miller's  Cove)  and # 29
(Lobster  tank and drains from fish  market),  all  drains  exhibited
no flow.  Flow observed at  #  17 and  #  19  are considered to be
from tidally influenced groundwater.  Additionally,  with the
exception  of  those  aforementioned  drains,  flow  at the other
stormdrains ceases  within approximately  2  hours following rain
events.

   The following  is a  summary of  the  drains  and discharges
located in  Buttermilk Bay.  Impervious surface drainage area was
estimated  based  on drainage  area walks using  a  Rolotape and
calculating area.

Reference Code: t 1

Location: East side of Cohasset Narrows draining highway.
Description: 45.7 cm (18 inch)  diameter  steel pipe discharging
above  high  water  line.
Approximate  Impervious-Surface Drainage  Area  in  Sq.M.: 8,046
                        (86,570 Sq.Ft.)
Characteristics  of  Drainage  Basin:   Commercial  (Craft Shop,
Diner,  Restaurant and Gas Station)

Reference Code:  t 2

Location: Electric Avenue Boat  Ramp.
Description:  Boat ramp with  bermed paved surface to direct
drainage.
Approximate  Impervious-Surface Drainage  Area  in Sq.M.:  2091
                        (22,500 Sq.Ft.)
Characteristics  of Drainage Basin: Residential, Boat ramp and
some beach  parking.

Reference Code: f 3

Location: Public Beach near Electric  Ave.
Description: Corrugated Pipe  with discharge below high water
line.
Approximate  Impervious-Surface Drainage  Area  in Sq.M.:  5980
                        (64,340 Sq.Ft.)
Characteristics of Drainage Basin: Residential, Beach Parking,
light commercial. Bermed to direct  drainage.

Reference Code: f 4

Location: Brom  Dutcher Rd.
Description:  Open Ditch  emptying at edge  of bay above hiah water
line.

                              76

-------
Approximate Impervious-Surface  Drainage  Area  in Sq.M.: 325  (3500
Sq.Ft.) plus approximately 372 Sq.M. (4,000 Sq.Ft.) semi-pervious
roadbed.
Characteristics of Drainage Basin:  Residential

Reference Code: f 5

Location: Brom Dutcher Rd.
Description:  Pipe corrugated-galvanized -installed 1986.
Approximate Impervious-Surface  Drainage Area  in Sq.M.:  325  (3,500
Sq.Ft.)
Characteristics of Drainage Basin:  Residential

Reference Code: t 6

Location: Vanderdonk Rd.
Description:  Concrete Pipe with approximately 6.1  M of  vegetative
buffer between discharge point and waterline.
Approximate Impervious-Surface Drainage Area  in  Sq.M.:  1905
                         (20,500 Sq.Ft.)
Characteristics of Drainage Basin:  Residential with the majority
of road bermed to direct drainage.

Reference Code: t 7

Location: End of Quamhasset Rd.
Description:  Small Collection Basin with Pipe and  at  least  30.5 M
vegetated buffer between discharge and bay.
Approximate Impervious-Surface  Drainage Area  in Sq.M.:  465  (5,000
Sq.Ft.)
Characteristics of Drainage Basin:  Residential

Reference Code: t 8

Location: Puritan Rd. near Quamhasset Rd.
Description: Two Corrugated  Pipes  discharging above high water
mark, but not accessible at high tide.
Approximate Impervious-Surface Drainage Area  in  Sq.M.:  9812
                        (105,572 Sq.Ft.)
Measured value was multiplied by 1.1 to estimate paved driveways
Characteristics of Drainage Basin:  Residential.   Much  of surface
is bermed to direct drainage.  Basins located  on  Puritan Ave
south of Erin Lane may be  leaching basins alone.

Reference Code: I 9

Location: Little Bay Lane                      .
Description:  Collection basin with  discharge  pipe into Spartina
patens  (short  chordgrass)  marsh.  Approximately 22.9 M of  buffer
exists between discharge point and bay.
Approximate Impervious-Surface Drainage Area  in  Sq.M.:  2/u»
                         (30,000 Sq.Ft.)
Characteristics of Drainage  Basin:  Residential.  Bermed surfaces
to direct drainage.

                               77

-------
Reference Code:  f  10

Location: End of Little Bay Lane
Description:  Termed "Marsh Stream"  by  D.E.Q.E.. Has Clapper Valve
and drains an old  bog and  marsh.
Approximate Impervious-Surface  Drainage Area in Sq.M. N/A
Characteristics  of Drainage Basin: Marsh with  stormdrain entering
in at Puritan Rd.

Reference Code:  f  11

Location: Old Head of Bay  Rd.
Description:  Stream resulting  from  groundwater recharge. Culvert
under Rd. discharges into bay at high water  mark.
Approximate Impervious-Surface  Drainage Area in Sq.M. N/A
Characteristics of Drainage Basin:  Private Property,  farmlike.
Two horses kept on property.

Reference Code:  I  12

Location: Corner of  Old Head of Bay and Head of  Bay Rd.
Description:  Pipe not  located  (although  a discharge  pipe
direction can be  seen  in the adjacent  drain).   May  be just a
road-cut discharge.
Approximate  Impervious-Surface Drainage  Area in  Sq.M.:  4294
                        (46,200 Sq.Ft.)
Characteristics of Drainage  Basin:  Not bermed,   steep slope.
Residential area.  Very pervious surface both sides  of road.

Reference Code:  f  13

Location: In  Hid-saway Village
Description:  Concrete Pipe emerging  from rip rap
Approximate  Impervious-Surface Drainage  Area in Sq.M.  Not
Determined but less than  93 (1,000  Sq.Ft.)  impervious.
Characteristics  of  Drainage  Basin:  Intensely  developed
residential.

Reference Code:  i  14

Location: Near the stream  behind hideaway village.
Description:  Road  berm  into Hideaway Village  Stream. Road bermed
on one side only.
Approximate  Impervious-Surface Drainage  Area in  Sq.M.:  1626
                        (17,500 Sq.Ft.)
Characteristics  of Drainage  Basin:  Residential  and agricultural.
Cranberry Bog stream which passes under Head of the Bay Road.
Thus drainage is primarily from bog.

Reference Code:  i  15

Location: Bayhead  Shores
Description:  Corrugated Pipe discharging  above the high water
mark.
Approximate  Impervious-Surface Drainage  Area in Sq.M.:  1413

                              78

-------
                         (15,200  Sq.Ft.)
Characteristics  of  Drainage Basin: Residential

Reference  Code:  I 16

Location:  Red Brook at  Head of  Bay Rd.
Description:  Included here are  two discharges, one concrete pipe
discharging above  the  water  line on the west side of the creek,
and a bermed discharge  on the east side.
Approximate Impervious-Surface  Drainage Area  in Sq.M.
             West  side pipe  -  8,643 (93,000 Sq.Ft.)
             East side  cut  - 3,764 (40,500 Sq.Ft.)
Characteristics  of  Drainage Basin: primarily residential with the
exception  of  the north side  of Head of the Bay Rd.  which is an
undeveloped parcel  of privately-owned and municipally-owned land.

Reference  Code:  f 17

Location:  Wychunus  Ave  in Wareham, Mouth of Red Brook
Description:  Corrugated pipe discharging  below  the  high water
mark.
Approximate Impervious-Surface Drainage  Area in Sq.M.:  10,855
                       (116,800  Sq.Ft.)
Characteristics of  Drainage  Basin:  Residential,  lot  sizes
approximately  465  Sq.M. (5,000  Sq.Ft)-  primarily  year-round
residences.

Reference  Code:  t 18

Location:  Town Landing  in Wareham
Description:  16 ft x  1  ft.  grate receiving runoff  from road.
Located on beach.
Approximate Impervious-Surface Drainage Area  in  Sq.M.:  1631
                         (17,650  Sq.Ft.)
Characteristics of  Drainage Basin:  residential. Lots
approximately 465 Sq.M.  (5,000  Sq.  Ft.).  Mix of seasonal  and
year-round homes (mostly year-round).

Reference  Code:  f 19

Location:  Between 55 and  51 Cleveland Way, Wareham.
Description:  38.1 cm (15 in)  diameter corrugated pipe  discharging
below the  high water  mark.
Approximate Impervious-Surface Drainage Area  in  Sq.M.:  1631
                         (17,550  Sq.Ft.)
Characteristics  of  Drainage Basin: Residential with approximately
465  Sq.M.  (5,000  Sq.Ft)  lots.   High percentage of  seasonal
cottages.   The drained area is split about evenly between paved
surface and semi-pervious hardpan.   This drain  has been observed
after rains and  appears to stop draining prior to other drains.

Reference  Code:  i 20

Location:  The  end of Chippewa Drive,  Wareham
Description:  38.1 cm (15 in)  diameter corrugated pipe  discharging

                              79

-------
generally below  the  sand surface, below the high water mark.
Approximate  Impervious-Surface Drainage  Area in  Sq.M.:  2035
                        (21,900 Sq.Ft.)
Characteristics  of Drainage Basin:  Residential with approximately
465  Sq.M.  (5000 Sq.Ft.) lots.  Higher percentage of  seasonal
cottages in the drainage basin. 725 Sq.M. (7800 Sq. Ft.) of basin
is composed of semipervious hardpan.

Reference Code:  §21

Location:  Discharge  into inner Miller's Cove
Description: Corrugated pipe discharging below high water  mark.
Approximate Impervious-Surface Drainage Area in Sq.M.:  279 (3,000
Sq.Ft.)
Characteristics  of Drainage Basin:  Residential
Reference Code:  §22

Location:  Between two end  houses on Jefferson Road.
Description:  25.4  cm (10  in)  diameter  cement  pipe  with
approximately 4.6 M  of grass buffer between point of discharge
and bay.
Approximate Impervious-Surface Drainage  Area in  Sq.M.:  1115
                        (12,000 Sq.Ft.)
Characteristics  of Drainage Basin:  Residential

Reference Code:  § 23

Location:  Between #  61 and # 63 Jefferson Rd.
Description:  15.2 cm (6 in) pipe  discharging above  high water
mark.
Approximate Impervious-Surface  Drainage  Area  in  Sq.M.: 537  (5775
Sq.Ft.)
Characteristics  of Drainage Basin:  Residential

Reference Code:  § 24

Location:  Beside  I 37 Jefferson Rd.
Description: 30.5 cm  (12 in) concrete pipe discharging below the
high water mark.
Approximate Impervious-Surface Drainage  Area in  Sq.M.:  1004
                        (10,800 Sq.Ft.)
Characteristics of  Drainage Basin: Residential.  372-744  Sq.M.
lots (4,000-8,000 Sq.Ft.)  lots.

Reference Code:  § 25

Location:  Near #  23  Jefferson Rd.
Description:  Corrugated 25.4  cm (10 in) pipe discharging below
the high water mark.
Approximate Impervious-Surface Drainage  Area in  Sq.M.:  1338
                        (14,400 Sq.Ft.)
Characteristics of  Drainage  Basin: Residential, 465-1208 Sq.M.


                              80

-------
(5,000-13,000 Sq. Ft. lots. Primarily year-round.

Reference Code:  f 26

Location:  Between f  15 and f 17
Description:  Two  corrugated  25.4  cm  (10  in)  diameter pipes
discharging below the high water mark.
Approximate  Impervious-Surface Drainage  Area in Sq.M.: 3399
                        (36,575 Sq.Ft.)
Characteristics of Drainage Basin:  Residential.  372-744 Sq.M.
(4,000-8,000 Sq. Ft.lots,  primarily  year-round residences.

Reference Code:  f 27

Location:  On Route  28 By Boatyard
Description:  45.7  cm (18 in)  concrete  pipe draining  highway.
Discharge at high water  mark.
Approximate  Impervious-Surface Drainage  Area in Sq.M.: 7026
                        (75,600 Sq.Ft.)
Characteristics of Drainage  Basin: Commercial  (Professional
Building,  Restaurant, Boatyard, Realty,  Hotel)


Reference Code:  f 28

Location:  By Captain Harris Fish Market
Description:  30.5 cm  (12  in) steel pipe draining highway.
Discharges below high water mark.
Approximate  Impervious-Surface  Drainage  Area  in  Sq.M.:4740
                        (51,000 Sq.Ft.)
Characteristics of Drainage  Basin: Commercial  (Fish Market,
Liquor Store,  Antique Shop, Bait and Tackle). Highway.

Reference Code:  t 29

Location:  By Captain Harris Fish Market
Description:  25.4 cm (10 in) concrete pipe
Approximate  Impervious-Surface Drainage Area  in Sq.M.  N/A  -
apparently ties in  to lobster  tank and floor drains.
Characteristics of  Drainage Basin:  see above
                                81

-------
                          APPENDIX  II

PARTICLE TRAJECTORY MODEL REPRESENTATION FOR SIMULATING THE FATE

  OF BACTERIOLOGICAL CONTAMINANTS ENTERING DISCHARGE POINTS IN

                        BUTTERMILK  BAY
Note:  These  particle trajectories are  based on entry  at  the
specified tidal  stages and were obtained using TEA, a  finite
element computer model developed by the Massachusetts Institute
of Technology,  Energy Laboratory.   They have  been provided
compliments  of Craig Fish, Boston University Geology Department.
                              82

-------
        TIDAL STAGE:  Approximately 3 hr after high tide
oo
uo
                        •::•>*:
           Red Brook  ..•.'.*
                        "
                                                                Little Buttermilk Bay
                                                            Queen Sewell Cove
                               ..

                              ^::'\  Water  Parcel Trajectory  Map
                                                 500
iooo m

      Cohassef Narrows

-------
      TIDAL STAGE: beginning of outgoing tide
      Red Brook
                                                       Little Buttermilk Bay
                                                    Queen Sewell Cove
                            Water Parcel Trajectory Map
                               o         500       1000 m
Cohasset Narrows

-------
                          REFERENCES

Akama, L.  and  S. Otani. 1970.  Clostridia perfrinqens as the  flora
intestine  of healthy persons. Japan J.  Med. Sci. Biol. 23:  161.

Akin, E. W., W. F.  Hill, Jr.,  G. B. Cline and W. H. Benton.  1976.
The loss of poliovirus 1 infectivity in marine waters.  Water
Res. 10:59-63.                                            	

APHA. 1985. Standard  Methods for the Examination of Water and
Wastewater. American  Public Health Association.  1268 pp.

Allen, M.  J.   1981.   Microbiology  of  ground water.  J.  Water
Pollut.  Control Fed.  53:1107-1109.                     	

Bell,  R.  G.  1976. Persistence  of fecal coliform  indicator
bacteria  on  alfalfa  irrigated  with  municipal  sewage lagoon
effluent.  J^ Environ.  Quality  1:39-42.

Bell, R. G. and J. B.  Bole.  1976.  Elimination of fecal coliform
bacteria  from reed canarygrass irrigated with  municipal sewage
lagoon effluent. J^ Environ. Quality 4:417-418.

Bellair,   J.  T.,G.  A.  Parr-Smith  and I.  G. Wallis.  1977.
Significance  of diurnal variations  in  fecal coliform die-off
rates in the design of ocean outfalls. J. Water Pollut. Control
Fed. 49:2022-2030.

Berg,  R.   W.   and A.  W.  Anderson.   1972.  Salmonellae  and
Edwardseiella  tarda  in gull feces:  a source  of  contamination of
fish processing plants. Appl.  Microbiol. 24:501-503.

Betson,  R. P. and  R. A.  Buckingham.  1969.  Fecal  coliform
concentrations in  stormwater.   Tennessee  Valley  Authority.
Knoxville,  Tennessee.

Bisson, J.  W.  and  V. J. Cabelli.  1980. Clostridium  perfringens as
a water pollution  indicator.J. Water Pollut.  Control Fed. 52:241-
248.                       ~~

Bisson,  J.  W. and  V.  J.  Cabelli.  1979.  Membrane  filter
enumeration method for Clostridium perfringens.  Appl. Environ.
Microbiol. 37:55-66.

Bitton, Gabriel.  1975.  Adsorption of viruses onto surfaces in
soil and water.  Water Res.  9:473-484.

Bitton, G. and C. P.  Gerba Eds.  1984. Groundwater Pollution
Microbiology.  John Wiley  and Sons, Inc. New  York.  377  pp.

Bitton, G.,  O. C. Pancorbo  and S.  R.  Farrah.   1984.  Virus
transport  and  survival after land application of  sewage sludge.
Appl.  Environ.  Microbiol. 47:905-909.

Bitton, G., R. Fraxedas and G. E. Gifford. 1979. Effect of  solar

-------
radiation on poliovirus:  preliminary experiments.  Water Res.
13:225-228.

Borrego,  J.  J.  and P.  Romero.   1985.   Coliphage  survival  in
seawater.    Water  Res.  19:557-562.

Borrego, J. J., F. Arrabal, A. de  Vicente, L. F. Gomez, P.
Romero.  1983.  Study  of microbial inactivation in the marine
environment.   J._ Water  Pollut. Control Fed.  55:297-302.

Bouwer,  J., J. C. Lance, and M. S. Riggs.  1974.  High-rate land
treatment II:  water quality  and  economic aspects  of  the  Flushing
Meadows  project.J._  Water Pollut. Control  Fed. 46:844-855.

Boyd. W. L.  and  J.  W. Boyd. 1962. Viability  of  thermophiles and
coliform  bacteria in arctic soils  and   water.  Can.  J.
Microbiol.  8:189-192.

Brown,  K.  W.,  H.  W. Wolf,  K.  C. Donnellly,  and J. F.  Slowey.
1979.   The  movement of  fecal coliforms and  coliphages  below
septic  lines.  J. Environ. Qual. 8:121-125.

Burge,  W. D. and N. K.  Enkiri. 1978.  Virus adsorption  by five
soils.   J. Environ. Qual. 7:73-76.

Cabelli,  V.  J. 1983.  Health  Effects  Criteria  For Marine
Recreational  Waters.  Health Effects Research Laboratory, U.S.
Envir. Protection Agency EPA-600/1-80-031. 98 p.

Cabelli, V.J., A.P. Dufour,  L.J. McCabe  and M.A.  Levin. 1983. A
marine  recreational water  quality criterion consistent with
indicator concepts and risk analysis. J. Water Pollut. Control
Fed. 55:1306-1314.

Carlucci,  A.  F.  and  D. Pramer.  1959.  Factors  affecting the
survival of bacteria in seawater.  J.Appl.  Microbiol.. 7:388-392.

Carter,  J. 1980.   Year of  Coast Address, statistics  from the
Bureau of the Census,  Washington, D.C.

Chamberlin, C. E.  and R. Mitchell. 1978.  pages  325-348 In: Water
Pollution Microbiology Volume 2. John Wiley and Sons,  Inc. New
York.

Coleman, R. N.,  J.  N. Campbell, F. D.  Cook and  D.  W.  S. Westlake.
 1974.  Urbanization and the  microbial  content of  the  North
Saskatchewan River. Appl.  Environ.  Microbiol.  27:93-101.
Craigie, J. S. and J. M. McLachlan.  1964. Excretion of colored
ultraviolet-absorbing substances by marine algae. Can. J. Bot.
42:23-33.                                          	

Crane,  S.  R.,J. A.  Moore, M. E. Grismer and  J.  R.  Miner. 1983.
Bacterial pollution from agricultural sources: A review.  Trans
ASAE. 26:858-866.                                         	


                              86

-------
Damgaard-Larsen,  Soren,  K.  0. Jensen and E. Lund and B. Nissen.
1977.  Survival and movement of enterovirus in connection with
land disposal  of sludges.  Water  Res. 11:503-508.

Davenport, C. V.,  E.  B. Sparrow,  and R.  C.  Gordon.  1976.  Fecal
indicator bacteria persistence under natural conditions in an
ice-covered river.  Appl.  Environ. Microbiol.   32:527-536.

Dizer, H., A.  Nasser, and J.  M.  Lopez.   1984.   Penetration of
different  human pathogenic  viruses into  sand  columns percolated
with distilled water,  groundwater, or wastewater.   Appl. Environ.
Microbiol.  47:409-415.                              	

Draser, B.S.   et al.  1975.  Clostridia  isolated from  feces. J.
Medical Microbiol.  9:63.

Duboise,   S.  M., B. E. Moore,  C. A.  Sorber,  and  B.  P, Sagik.
1979.  Viruses  in Soil Systems.   CRC Critical  Reviews in
Microbiology   245-285.

Duboise,  S. M.,  B.  E. Moore, and  B. P.  Sagik.  1976.  Poliovirus
survival and  movement in a sandy forest soil.  Appl.  Environ.
Microbiol. 31:536-543.

Dufour,  A.  P.  and V.  J. Cabelli.  1976.  Characteristics of
Klebsiella from textile finishing plant  effluents.  J_._ Water
Pollut. Control  Fed.  48:872

Edmonds,  Robert L.  1976. Survival  of coliform bacteria  in sewage
sludge applied to a forest  clearcut  and  potential  movement  into
groundwater.  Appl.  Environ.   Microbiol. 32:537-546.

Enzinger, R.   M.  and R.  C.  Cooper.  1976. Role  of  bacteria and
protozoa  in  the  removal of  Escherichia  coli  from estaurine
waters. Appl.  Environ.  Microbiol.  31:758-763.

Environmantal Protection Agency. 1983.  Results of  the  National
Urban Runoff Program,  Volume 1 - Final Report. Water  Planning
Division,  USEPA,  Washington  D. C.   NTIS PB84-185552.

Erkenbrecher,  Carl  W.  Jr. 1981. Sediment bacterial  indicators in
an urban shellfishing subestuary of the lower Chesapeake  Bay.
Appl.  Environ.  Microbiol. 42:484-492.

Faddoul,  G. P. and G.  W. Fellows.  1966. A five-year survey of the
incidence of  salmonellae in avaian species.  Avian Pis. 10:296-
304.

Farrah,,  S.  R., G. Bitton, E.  M.  Hoffmann,  0.  Lanni, 0.  C.
Pancorbo, M.  D.  Lutrick, and J. E. Bertrand.  1981.   Survival of
enteroviruses and coliform bacteria  in a  sludge  lagoon. Appl.
Environ.  Microbiol.   41:459-465.

Faust, M. A.,  A. E. Aotaky and M. T. Hargadon.  1975.  Effect  of
physical  parameters on the  in situ survival of Escherichia  coli

                              87

-------
MC-6 in an estuarine environment. Aggl^ a_nd Environ. Microbiol.
30:800-806.

Faust,  Maria  A. 1976. Coliform bacteria from diffuse sources as a
factor  in estuarine  pollution.  Water Res.  10:619-627.

Fujioka,  Roger S.,  Philip  C.  Loh and L.  Stephen Lau.   1980.
Survival  of human enteroviruses  in  the  Hawaiian  Ocean
environment:  evidence for virus-inactivating microorganisms.
Appl.  Environ.  Microbiol. 39:1105-1110.

Fujioka, R. S.,H.  H.  Hashimoto, E. B.  Siwak and R. H.  F.  Young.
1981. Effect of sunlight on survival  of  indicator bacteria  in
seawater. Appl. Environ. Microbiol. 41:690-696.

Funderburg,  S. W.,  B. E. Moore,  B.  P.  Sagik  and  C. A. Sorber.
1981.   Viral  transport  through  soil  columns under  conditions  of
saturated flow.  Water Res.  15:703-711.

Furfari,  S.  A.  1979. Training Course Manual  -  nonpoint  pollution
in shellfish sanitation.  Dept. Health Education and Welfare,
U.S.  Public Health Service,  Food and  Drug Administration.

Gallagher,  J. L.,  W.  J.  Pfeiffer and  L. R. Pomeroy.   1976.
Leaching and microbial utilization of dissolved organic  carbon
from leaves  of Spartina  alterniflora. Estuarine and Coastal
Marine Science  4:467-471.

Gameson, A. L.  H. and J. R. Saxon.  1967. Field studies on  effect
of daylight on  mortality of coliform bacteria.  Water Res.  1:279-
295.

Geldreich, E . E. and N. A. Clarke.  1966.  Bacterial pollution
indicators in the intestinal  tract of freshwater fish.   Appl.
Microbiol. 14:429-437.

Geldreich, E. E.,  R. H.  Bordner, C.  B.  Huff,  H.  F.  Clark  and P.
W. Kabler. 1962. Type distribution of  coliform bacteria  in the
feces  of  warm-blooded animals. J.  Water Pollut.  Control Fed.
34:295-301.

Geldreich,  E. E. and B.  A.  Kenner.  Concepts of  fecal streptococci
in stream pollution. J.  Water Pollut. Control  Fed.  41:336

Gerba,  C. P.  and and G.  Bitton.  1984. Microbiological Pollutants:
Their survival  and transport  pattern  to groundwater.  In  Bitton,
G. and  C. P.  Gerba Eds.  1984. Groundwater Pollution  Microbiology.
John Wiley and  Sons,  Inc. New York.

Gerba, Charles P. and J. McLeod.  1976. Effect of sediments on
survival of  Escherichia  coli  in marine waters. Appl. Environ.
Microbiol.  32(1):114-120                         -"	

Gerba,  C. P.  and G. E. Schaiberger. 1975. Effect of particulates
on virus survival in seawater. J._  Water Pollut.  Control Fed.

-------
47:93-103.

Gerba,  C.  P.,  S.  M. Goyal,  R.  L.  LaBelle, I. Cech and G. F.
Bogdan.  1979.  Failure  of  indicator bacteria  to reflect  the
occurrence  of  enteroviruses in marine  waters. American J. of
Public Health.  69:1116-1119.                    	

Gerba,  C.  P.,  and  J.  F.  McNabb.  1980.   Microbial  aspects of
groundwater pollution.  ASM News  47:326-329.

Gerba, C. P., C. Wallis,  J. L.  Melnick.  1975. Fate  of  wastewater
bacteria and viruses in soil.   J. of the  Irrigation and Drainage
Division.157-175.

Gerba,  C. P.,  and J. C.  Lance.   1978.  Poliovirus removal  from
primary and secondary  sewage  effluent by soil filtration.  Appl.
Environ. Microbiol.   36:247-251.

Gerba,  C.  P.,  J. B. Rose,and S.  N.  Singh. 1985   Waterborne
gastroenteritis  and viral hepatitis.  Crit. Rev. Env.  Control
15:213-236.

Gerba C. P., S. M. Goyal, E. M. Smith,  and J. L. Melnick.  1977.
Distribution of viral and bacterial pathogens in a coastal canal
community.   Marine Pollution  Bulletin 8:279-282.

Gerba,  C.  P. and  G.  Schaiberger.  1973.  Biscayne Bay:
bacteriological data interpretation. Florida Scientist 36:104-
109.

Gilbert, R. G., C. P. Gerba,  R. C.  Rice,  H.  Bouwer,  C.  Wallis,
and J.  L.  Melnick.   1976.   Virus and bacteria  removal  from
wastewater by land  treatment.   Appl.  Environ.  Microbiol..
32:333-338.

Goyal, S. M.  1984.  Viral pollution of the marine environment.
Crit. Rev,  in Env. Control  14:1-32.

Goyal,  S.  M., C.  P. Gerba  and  J.  L.  Melnick.  1979. Human
enteroviruses  in  oysters  and  their  overlying  waters.  Appl.
Environ. Microbiol. 37:572-581.

Goyal,  S. M.,  W.  N.  Adams,  M. L. O'Malley and D. W. Lear.  1984.
Human pathogenic  viruses at sewage  sludge disposal sites in the
middle atlantic region.   Appl. Environ.  Microbiol. 48:758-763.

Goyal, S. M.,  C. P. Gerba, J.  L. Melnick.  1978.   Prevalence of
human enteric  viruses in coastal  canal  communities.  J. Water
Pollut.  Control Fed.  October  1978:2247-2256.

Goyal, S. M. and  C.  P. Gerba.   1979.  Comparative adsorption of
human  enteroviruses,  simiam  rotavirus,  and  selected
bacteriophages  to  soils. Appl. Environ.  Microbiol.  38:241-247.

Grimes, D.  J.  1975.  Release  of sediment-bound fecal coliforms

                              89

-------
by dredging.   Appl.  Microbiol.  29:109-111.

Haenal, H.  1970.  Human normal  and abnormal  gastrointestinal
flora.  Amer. J._ Clin. Nutrition.  23: 433

Hagedorn,  C. D. Microbiological aspects  of  groundwater  pollution
due to septic tanks, pages 181-196 In Bitton,  G. and C. P. Gerba
Eds. 1984. Groundwater Pollution  Microbiology.  John Wiley and
Sons, Inc. New York.

Hagedorn, C.,  D. T  Hansen and G. H. Simonson. 1978. Survival
and movement of fecal indicator bacteria in soil under conditions
of saturated flow.  J_._ Environ. Quality.  7:55-59.

Hellebust, J.  A.. 1965. Excretion  of  some  organic compounds by
marine phytoplankton. Limnol. and Ocean.  10:192-206.

Hendricks,  C. W. 1970.  Enteric bacterial metabolism of stream
eluates.  Can.  J._ of_ Microbiol. 17:551-556.

Hendricks,  C. W. 1971.  Increased recovery of  Salmonellae from
bottom sediments versus surface  waters.  Appl.  Microbiol.21:379-
380.

Hendricks,  C.  W.  1972.  Enteric bacterial growth rates in river
water.  Appl.  Microbiol.  24:168-174.

Hendricks,  C. W. and S. M.  Morrison. 1967. Multiplication and
growth of selected enteric bacteria in clear mountain stream
water.  Water Resources.  1:567-576.

Hill,  G.   A.  and  D. J.  Grimes.  1984.  Seasonal  study  of a
freshwater lake  and migratory waterfowl  for  Campylobacter
jejuni.Can.  J._ Microbiol.  30:845-849.

Hill,  W.F.,Jr., E.W. Akin, W. H.  Benton and F.  E.  Hamblet.  1971.
Viral disinfection  of estuarine water by UV. J._ ASCE SA5:601-615.

Hood,  M. A. and G.  E. Ness.  1982.   Survival of Vibrio cholerae
and Escherichia coli in estuarine waters and sediments^Appl.
Environ.  Microbiol.  43:578-585.

Huntley,  B. E. et al.  1976. Klebsiella  densities  in waters
receiving wood pulp effluent.  J. Water  Pollut. Control  Fed.
48:1766-                         — ~	-

Hurst, C.  J.,  C.  P.  Gerba  and  I. Cech. 1980.  Effects of
environmental variables  and  soil  characteristics  on  virus
survival  in soil.  Appl.  Environ.  Microbiol. 40:1067-1079.

Hurst, C.  J.,  C. P.  Gerba,  J. C.  Lance,  and  R.  C.  Rice.  1980.
Survival  of enteroviruses in rapid-infiltration basins during the
land application  of  wastewater.   Appl.  Environ.  Microbiol.
40:192-200.                         ~"— 	 	
                              90

-------
Hussong, D., J. M.  Damare, R.  J.  Limpert,  W.  J. L. Sladen, R. M.
Weiner and R.  R.  Colwell.  1979. Microbial impact of Canada geese
(Branta  canadensis) and  whistling swans  (Cygnus columbianus
columbianus) on  aquatic ecosystems. Appl.  Environ.  Microbiol.
37:14-20.                                  	

Jones,  Galen.   1964.  Effect of chelating agents on the growth of
Escherichia coli  in seawater.  £._ of Bact. 87:483-499.

Kadlec,  R.  H.  and  D.  L.  Tilton. 1979. The  use  of freshwater
wetlands as a  tertiary wastewater treatment alternative.  CRC
Crit_. Rev.  Environ.  Control 185-212.

Kapuscinski, R.  B., and R.  Mitchell.   1983.   Sunlight-induced
mortality  of viruses and  Escherichia  coli  in coastal seawater.
Environ. Sci. Technol.  17:1-6.

Keswick,   B.  H.   and C.  P.  Gerba.   1980.    Viruses   in
Groundwater.Environ. Sci.  Technol.   14:1290-1297.

Kibbey, H.  J.,  C. Hagedorn, and E. L. McCoy.  1978.  Use of Fecal
Streptococci as Indicators of  Pollution in  Soil.  Appl. Environ.
Microbiol..  35:711-717.

Koppelman,  L. E.  and E.  Tanenbaum. 1982.  The Long  Island segment
of the  National Urban  Runoff Program.  Long Island  Planning
Board,  Hauppauge, New York.

LaBelle, R. L. and  C. P. Gerba.  1980.  Influence of  estuarine
sediment on  virus survival  under field  conditions.   Appl.
Environ.  Microbiol.  39:749-755.

LaBelle, R. L. and C.  P.  Gerba.  1982.  Investigations  into  the
protective  effect of estuarine  sediment on  virus  survival. Water
Res.  16:469-478.

Labelle, R. L., C.  P. Gerba, S. M. Goyal, J. L. Melnick, I. Cech
and  G.  F.  Bogdan.   1980.  Relationships between  environmental
factors, bacterial indicators  and the  occurrence of enteric
viruses  in  estuarine  sediments. Appl. Environ.  Microbiol.  39:
588-596.

LaBelle, Raymond L., and  C. P. Gerba.  1979.   Influence of  pH,
salinity and organic matter on the  adsorption  of enteric viruses
to estuarine sediment.   Appl.  Environ.  Microbiol.   38:93-101.

LaLiberte,  P.  and D. J. Grimes.   1982.  Survival  of Escherichia
coli  in lake bottom sediment.  Appl. Environ.  Microbiol.   43:623-
6287

Lance, J.  C. and C. P. Gerba.   1984.  Virus movement in soil
during saturated  and unsaturated  flow.  Appl.  Environ.  Microbiol.
47:335-337.

Lance, J. C.,   C. P. Gerba and S. S. Wang.    1982.  Comparative


                              91

-------
movement of different enteroviruses in soil columns.  J._ Environ.
Qual. 11:347-351.

Lance.  J.  C., C. P. Gerba,  and J.  L. Melnick.   1976.  Virus
movement in soil  columns  flooded with  secondary sewage effluent.
ADD!.  Environ.  Microbiol.  32:520-526.
Appl.  Environ.  Microbiol.  32:520
Landry, E. F.,  J.  M. Vaughn, T. J. Vicale  and R.  Mann.  1983.
Accumulation  of  sediment-associated viruses in shellfish.  Appl.
Environ.  Microbiol.   45:238-247.

Landry, E. F.,  J.  M. Vaughn, M. Z. Thomas and C.  A. Beckwith.
1979.  Adsorption  of  enteroviruses  to soil  cores and  their
subsequent  elution  by  artificial  rainwater.  Appl.  Environ.
Microbiol.  38:680-687.

Larsson, U.  and A.  Hagstrom.1982. Fractionated phytoplankton
primary production, exudate release and bacterial production in a
Baltic eutrophication gradient.  Marine Biology  67:57-70.

Lefler, E. and Y.  Kott.  1974. Virus retention and  survival in
sand. In J.  F.  Malina and B. P. Sagik (Eds.).  Virus Survival in
Water and Wastewater Systems. Univ. Texas, Austin, TX.

Levin, Ma. A.,  J.  R. Fischer and  V. J.  Cabelli.1975.  Membrane
filter technique for enumeration of enterococci in Marine Waters.
Appl. Environ.  Microbiol.  30:66-71.

Liew,  P.  F.  and C.  P.  Gerba.  1980. Ther mostabi 1 iza t ion of
enteroviruses by estuarine sediment. Appl. Environ. Microbiol.
40:305-308.

Lo, S.  H., and 0. J.  Sproul.  1977.   Polio-virus adsorption from
water onto silicate  minerals. Water Res. 11:653-658.

Lo,  S., J. Gilbert and F.  Hetrick.  1976. Stability  of  human
enteroviruses  in  estuarine and marine waters. Appl. Environ.
Microbiol.  32:245-249.

Mack,  W. N.,  M.  SA.  Yue-Shoung Lu., D.  B. Coohon.   1972.
Isolation  of poliomyelitis virus from a contaminated  well.
Health Services  Reports  87:271-274.

Mallmann,  W. L.,  and W. Litsky.   1951.  Survival  of  selected
enteric organisms  in various  types of soil.   American  J. of
Public Health 41:38-44.                          	

Marzpuk, Yosef, S. M. Goayal and C. P. Gerba. 1980.  Relationship
of viruses and  indicator  bacteria in water and wastewater of
Israel.  Water Res.  14:1585-1590.

Massachusetts Division of  Marine  Fisheries.  1985.  Massachusetts
Marine Fisheries  - Assessment  at Mid Decade  - Economic,
environmental  and  management  problems facing Massachusetts'
commercial and recreational  marine fisheries.  Dept.  of  Fisheries
                               92

-------
and Wildlife & Envir.  Law  Enforcement, Exec. Off.  Envir. Affairs.
Publ.  | 14224-65-500-10-85-C.R.

Matossian. A. M., and  G. A. Garabedian.  1967.  Virucidal action
of sea water.  American J._ of  Epidemiology.  85:1-8.

Matson, E. A., S. G. Hornor,and  J. D. Buck.  1978.  Pollution
Indicators and other Microorganisms  in River  Sediment. J. Water
Pollut.  Control Fed.    13-19.                          	

McCambridge, J.  and  T. A. McMeekin.1980.  Relative effects of
bacterial  and protozoan predators on  survival of Escherichia coli
in estaurine  water  samples. Appl. Environ.  Microbiol. 40:907-911.

McConnell, L.  K.,  R.  C. Sims, B.  B.  Barnett.   1984.   Reovirus
removal and  inactivation by  slow-rate sand  filtration.  Appl.
Environ.  Microbiol. 48:818-825.

McFeters, G. A.,  S. A.  Stuart,  and S. B. Olson.  1978.  Growth of
heterotrophic bacteria and   algal  extracellular  products  in
oligotrophic  waters.  Appl. Environ. Microbiol.  35:383-391.

Melnick,  J.  L.,  and C.  P.   Gerba.  1980.   The  Ecology  of
Enteroviruses in  Natural Waters.  Critical  Rev. Environ. Control.
10:65-93.

Metcalf,  T.  G. and W. C.  Stiles. The accumulation of enteric
viruses by the oyster Crassostrea virginica.  J._ of Infectious
Diseases.  115:68-76.

Metcalf,  T.  and W.  C. Stiles.  1968  Enteroviruses  within an
estuarine environment. American J. of Epidemiology.  83(3):379-
391.

Metcalf,  T.  G.  C. Wallis and J.  L.  Melnick.  1974.  Virus
enumeration  and public health assessments in polluted surface
water contributing  to  transmission of virus  in nature.  In; Malina
and B. P.  Sagik (eds.), Virus  Survival in water and Wastewater
Systems. Center for Research in Water Resources, Austin, Texas.

M«tcalf, T. G.,E.  Moulton and D. Eckerson. 1980. Improved method
and test strategy for  recovery of enteric viruses from shellfish.
Appl.  Environ.  Microbiol.  39:141-152.

Mitchell, Ralph.   1968.  Factors affecting the decline of non-
marine micro-organisms in  seawater.  Water Res.  2:535-543.

Mitchell,  R.,  R. Yankofsky  and H. W. Jannasch. 1967. Lysis of
Escherichia  coli by  marine  microorganisms. Nature  ,  London
215:891-8932.

Mitchell, T.  R.  and T.  Ridgwell.  1971. The  frequency of
salmonellae in wild ducks. £._  Med. Micro.  4:359-361.

Moog, Polly Lu. 1987.  The hydrogeology and  freshwater influx of


                              93

-------
Buttermilk  Bay,  Massachusetts with  regard to the circulation of
coliform and pollutants:  a  model study and development of methods
for general application.  M.S.  Thesis,  Department  of  Geology,
Boston University Graduate  School, Boston, Massachusetts.
166 p.

Moore, B. E., B. P.  Sagik,  and C. A.  Sorber.  Viral  transport to
ground water at a wastewater land  application site.   J.  Water
Pollut.  Control  Fed.  53:1492-1502.

Moore, James A. and G. R. Beehier. 1984. A study of the pollution
potential of  land-spread septage.  J. ojf Environ.  Health.
46(4):171-175.

Moore, R. S., D. H.  Taylor, L. S. Sturman, M. M. Reddy,  and G. W.
Fuhs.  1981.   Poliovirus  Adsorption by  34  Minerals and Soils.
Appl.  Environ.  Microbiol.  42:963-975.

Morris, R.  L., A. J.  Mearns and J. Kim. 1976.  Viruses-bacteria in
coastal  waters and shellfish.  Annual  Report of the  Southern
California  Coastal Waters Resource Project.

North Carolina Division of Environ. Management. 1986.  Proposed
regulations  to implement stormwater  controls -  Information
Package.

Nusbaum,  I. and R.  M. Carver. 1955.  Survival of  coliform
organisms in  Pacific Ocean coastal waters. Sewage and Industrial
Wastes.   27:1383-1390.

O'Brien,  R.  T.  and  J.  S. Newman.  1977.   Inactivation  of
Polioviruses  and Coxsackieviruses in  surface water.   Appl.
Environ.  Microbiol.  33:334-340.

Olivieri,  V.,  C.  W.  Kruse, K.  Kawata and J.  E.  Smith.  1977.
Microorganisms  in stormwater.  Municipal Environmental  Research
Laboratory, ORD,  USEPA, Rept.  No.  EPA-600/2-77-087.

Olsen, S.,  and  V. Lee. 1985. Rhode Island's  Salt Pond Region: A
Special  Area Management  Plan.   The Coastal Resources Center,
Graduate School  of Oceanography,  University  of  Rhode Island.  113
pp.

Orlob, G.  T. 1956.  Viability of  sewage  bacteria in sea water.
Sewage and Industrial Wastes. 28:1147-1167.

Ostrolenk, M.,  N Kramer and R. C. Cleverdon.  1947.  Comparative
studies of enterococci  and  Echerichia coli  as  indices  of
pollution. J. of Bact. 53;197-2(T3^         "

Pancorbo,  0. C.,  P. R. Scheuerman,  S. R. Farrah, and G. Bitton.
1981.  Effect of  sludge type on poliovirus association with and
recovery from  sludge  solids.  Can.  J^  Microbiol.  27:279-287.

Portnoy, B. L.,P. A. Mackowiak,  C.  T. Caraway,  J. A. Walker, T.

-------
W. McKinley  and  C. A. Klein.  1975. Oyster-associated hepatitis:
failure of shellfish  certification programs to prevent  outbreaks.
Appl.  Microbiol.  233:1065-1068.

Rao, V. C., K. M. Seidel, S. M. Goyal, T. G. Metcalf, and J. L.
Melnick.  1984.  Isolation of  enteroviruses from water ,
suspended solids, and sediments  from Galveston Bay:   survival of
soliovirus and rotavirus adsorbed to sediments.  Appl. Environ.
Microbiol.  48:404-409.

Reneau, R. B., and D. E.  Pettry.  1975.   Movements  of coliform
bacteria from septic  tank effluent through selected coastal plain
soils of Virginia.  J. Environ. Qual. 4:41-44.

Rittenburg, S. C., T Mittwer and D. Ivler.1958. Coliform bacteria
in sediments around three marine sewage outfalls.  Limnology  and
Oceanography. 3:101-108

Roper,  M.  M.f and K. C.   Marshall.   1974.  Modification of  the
interaction between Bscherichia coli and bacteriophage in saline
sediment.   Microbial  Ecology 1:1-13.

Roper,  M.  M. and  K. C. Marshall.   1978.  Effects of a clay
mineral on microbial predation and parasitism of  Escherichia
coli.  Microbial Ecology  4:279-289.

Roy,  D.,  M.  Tittlebaum, and J.  Meyer.   1981.  Microbiology:
detection,  occurrence, and  removal of viruses.  J_._ Water Pollut.
Control Fed.   53:1138-1142

Sayler,  G.  S.   ,J.  D.  Nelson Jr.,A.  Justice  and R.  R.
Colwell.1975. Distribution and  significance of fecal  indicator
organisms  in  the upper Chesapeake Bay.  Appl.  Environ.  Microbiol.
30:625-638.

Schaub, S. A., K. F. Kenyon, B.  Bledsoe,  and  R. E.  Thomas.  1980.
Evaluation of the  overland  runoff mode of  land  wastewater
treatment  for virus removal.  Appl.  Environ.  Microbiol.  39:127-
134.

Schaub,  S. A.,   and  C.  A.  Sorber.   1977.   Virus and bacteria
removal from wastewater by rapid  infiltration through soil.
Appl.  Environ.  Microbiol.  33:609-619.

Schaub, S.  A.,  and B.  P.  Sagik.   1975.   Association  of
enteroviruses with natural  and artificially  introduced colloidal
solids  in  water  and  infectivity of  solids-associated virions.
Appl.  Microbiol.   30:212-222.

Scheuerman,  G. Bitton, A. R. Overman, M. ASCE, and G. E. Gifford.
1979.   Transport  of  viruses  through  organic  soils and  sediments.
J._ o_f the Environ.  Engineering Division  629-641.

Schillinger, J.  E.  and J. J.  Gannon.  1985. Bacterial adsorption
and suspended particles in urban stormwater. J._ Water Pollut.


                                95

-------
Control  Fed.  57:384-389.

Shehata,  T.  E.,  and A.  G.  Marr.   1971.   Effect of  nutrient
concentration on the growth of Escherichia coli.  J. £f_ Bact.
48:210-216.

Sinclair, J.  L., and M.  Alexander.  1984.  Role of resistance to
starvation in bacterial  survival in sewage and lake water.  Appl.
Environ.  Microbiol.  48:410-415.

Smith,  E.  M.  1978.   Role  of sediment  in  persistence  of
enteroviruses  in the  estuarine environment.  Appl.  Environ.
Microbiol. 35:685-689.

Sobsy, M. D., C. H. Dean,  M. E. Knuckles and R.  A. Wagner.  1980.
Interactions  and  survival  of  enteric viruses in  soil  materials.
Appl.  Environ.  Microbiol.  40:92-101.

Stiles,  C. W., and H.  R. Crohurst.   1923.  The principles
underlying the movement of Bacillus coli in ground-water, with
resulting pollution of wellsTPublic Health Report 38:1350-1353.

Tate, R.  L.  III.   1978.  Cultural  and environmental  factors
affecting the longevity of Escherichia  coli  in  histosols.  Appl.
Environ.  Microbiol.  35:925-929.

Temple. K. L., A.  K. Camper, and G. A. McFeters.  1980.  Survival
of  two  enterobacteria  in  feces buried in soil under  field
conditions.   Appl. Environ. Microbiol.  40:794-797.

Tierney,  J. T.,  R.   Sullivan,  and  E.  P.  Larkin.   1977.
Persistence  of poliovirus 1  in soil and on vegetables grown in
soil  previously  flooded with inoculated sewage  sludge  or
effluent.Appl. Environ.  Microbiol.   33:109-113.

Tsai, Shiow-Chuan,  R.  D. Ellender, R.  A.  Johnson,  and F.  G.
Howell.   Elution of viruses from coastal  sediments.  1983.  Appl.
Environ.  Microbiol.  46:797-804.

Vaccaro,  R.  F., M. P.  Briggs,  C. L. Carey, and B. H. Ketchum.
1950.  Viability of Escherichia coli in  sea water.  American J.
of_ Public Health.   40:1257-1266.

Van  Donsel,  D.  J.,  E. E.  Geldreich  and N.  A.  Clarke.1967.
Seasonal variations in survival  of indicator bacteria  in soil and
their contribution to  storm-water pollution.  Appl.  Microbiol.
15:1362.                                          	

Van Donsel,  D. J. and  E.  E.  Geldreich.  1971.  Relationships of
Salmonellae  to fecal  coliforms in bottom  sediments.  Water
Resources 5:1079-1087.                                    	

Vasconcelos, G. J., and R. G. Swartz. 1976. Survival of bacteria
in seawater  using a diffusion chamber  apparatus  in situ.Appl.
Environ.  Microbiol.  31:913-920.                          ~"—

                             96

-------
Vaughn, James M. and Edward Landry. 1977. Virus Study - Report
Series 6.  Nassau-Suffolk Regional  Planning Board,  Hauppauge, New
York.

Vaughn, James  M. and  Edward Landry.  1983. Viruses  in soil and
groundwater.  In  G. Berg (Ed.) Viral Pollution  of the  Environment.
Chapter 9.  CRC Press. Boca Raton,  Florida.

Vaughn,  James M.   and Edward Landry.  1984.  Public  health
considerations  associated with molluscan aquaculture systems:
human viruses.  Aquaculture 39:299-315.

Vaughn,  James  M.,  E. F. Landry and T. Z.  McHarrell.  1983.
Entrainment of viruses from septic tank leach fields through a
shallow,  sandy soil  aquifer. Appl.  and  Environ.   Microbiol.
45(5):1474-1480.                                     	

Vaughn,  James  M.,  E. F.  Landry,  C.  A.  Beckwith and  T. Z.
McHarrell.  1981. Virus removal  during groundwater recharge:
effects of infiltration rate on adsorption of poliovirus to soil.
Appl. Environ.  Microbiol. 41:139-147.

Vaughn, James  M.,  E.  F.  Landry, T. J.  Vicale and  M. C.  Dahl.
1980. Isolation of  naturally  occurring enteroviruses  from a
variety of shellfish  species residing  in Long Island and New
Jersey marine embayments. J._ of Food Protection.  43:95-98.

Vaughn, James M., E. F. Landry, M. Z. Thomas,T. J.  Vicale and W.
F. Penello. 1979. Survey of human enteroviruses occurrence in
fresh and marine surface waters on Long Island.  Appl. Environ.
Microbiol.  38:290-296.

Verstraete, W.,  and  J. P. Voets.  1976.  Comparative study of E.
coli survival in two aquatic ecosystems.  Water Res..   10:129~::"
T3TT

Volterra,  L,  E.   Tosti,  A.  Vero,   and  G.  Izzo.   1985.
Microbiological pollution of  marine sediments  in the southern
stretch of the  Gulf of Naples.   Water,  Air  and Soil Pollution
26:175-184.

Wang, D-S, C. P. Gerba,  and J.  C. Lance.  1981.  Effect  of soil
permeability  on virus  removal  through soil  columns.  Appl.
Environ.  Microbiol.  42:83-88.

Weiser, R.  S.  and  C.  M. Osterud.  1945. The influence of the
freezing  temperature,  repeated fluctuations  of  temperature and
the  exposure to feezing temperatures on the mortality of
Escherichia  coli. J., Bact. 50:413-439.

Weiskel,  Peter K.  1986.  The  sediments of  the Buttermilk Bay
Drainage  Basin:  a field investigation.  Report submitted to the
Geology Department, Boston  University Graduate School, Boston
Massachusetts, December 1986. 28 p.

                              97

-------
Weiss, C. M. 1951.  Adsorption of E. coli of river and estuarine
silts.  Sewage and Industrial Wastes 23:227-237.

Wellings, F. M., A. L.  Lewis,  C.  W. Mountain, and  L.  M.  Stark.
1975.   Virus consideration  in  Land disposal of  sewage  effluents
and sludge.   Florida Scientist  38:203-207.

Wellings, F. M., A. L.  Lewis, C. W. Mountain,, and L.  V. Pierce.
1975.   Demonstration  of virus in groundwater after effluent
discharge onto  soil. Appl.  Microbiol.   29:751-757.

Whipple, W., C.  Randall,  N. Grigg, R.  Shubinski, T. Grizzard and
L. S. Tucker. 1983.  Stormwater  Management in Urbanizing  Areas.
Prentice-Hall, Inc.  Englewood Cliffs,  New Jersey. 234 pp.

Won, W. D. and H. Ross. 1973. Persistence  of virus and bacteria
in seawater.  J._  of Environ.  Engineering Division American Society
Civil  Engineers. 99:205-211.

Yates,  M. V.,   C.   P.  Gerba,  and L.  M. Kelley.  1985.  Virus
persistence in groundwater.  Appl. Environ. Microbiol.  49:778-
781.
                             98

-------