PRELIMINARY  RESULTS
          OF THE
NATIONWIDE  URBAN  RUNOFF PROGRAM
UJ
              EXECUTIVE SUMMARY
             Water Planning Division
         U.S. Environmental Protection Agency
             Washington, D.C. 20460

-------
         PRELIMINARY RESULTS

               OF THE

   NATIONWIDE URBAN RUNOFF PROGRAM
            March 1, 1982
          EXECUTIVE SUMMARY
       WATER PLANNING DIVISION
U.S.  ENVIRONMENTAL PROTECTION AGENCY
       WASHINGTON, D.C.   20460

-------
                         DISCLAIMER

     This report  has  been  reviewed by the U.S. Environmental
Protection  Agency and  approved  for release.   Approval  does
not  signify   that   the  contents  necessarily  reflect  any
policies  or decisions  of  the U.S.  Environmental  Protection
Agency  or  any of  its  offices,  grantees,   contractors,  or
subcontractors.

-------
                              EXECUTIVE SUMMARY
GOALS AND OBJECTIVES OF THE NURP PROGRAM

The  overall  goal  of  NURP  is to  develop information that  will  help provide
local  decision makers,  states,   EPA,  and  other  interested  parties with  a
rational basis  for  determining  whether or not urban  runoff is causing water
quality problems and, in the event that it is, for postulating realistic con-
trol options  and  developing water quality management plans,  consistent with
local needs, that will lead to implementation of least cost solutions.   It is
also hoped that this information base will be used to help make the best pos-
sible  policy  decision  on   Federal,  State,   and  local  involvement  in  urban
stormwatar runoff and its control.

As a water  quality  management planning effort, the primary NURP objective is
to develop water  quality management plans that will be appropriate, accepta-
ble, affordable,  and  focused on local needs.  This requires information that
will allow determination of:

     •   The extent to which urban runoff is  a contributor to water qual-
        ity  problems as  compared  to other  sources and/or  background
        loads,

     •   The  effectiveness   of  control  short of  treatment in  reducing
        water quality problems where they exist,  and

     •   Whether best  management practices  for control  of  urban runoff
        are cost effective  in comparison to alternative options.

The  overall intent  is to initiate programs  at the  local  level for preventa-
tive measures for the control of pollution from urban runoff.   The Water Qua-
lity Management program established demonstration projects to help facilitate
this effort and to  ensure  that all pollution sources were being addressed in
relation to the extent of their contribution of pollution.  Through implemen-
tation of these locally derived programs to prevent and to control urban run-
off pollution, one can minimize future high cost corrective actions.

At  the  national  level,  a  primary  NURP  objective  is to  assess  urban  runoff
on a nationwide scale and attempt to determine:

     •   The nature  of urban runoff  problems  where  significant problems
        have been identified,

     •   the causes  of these  problems (e.g.,  sources,  transport modes,
        impacts),

-------
     •  the severity of  these  problems,  based on consideration of bene-
        ficial uses and water quality standards, and

     •  opportunities  for  controlling urban  runoff problems,  including
        descriptions of  control  measures,  their  effectiveness,  costs,
        and strategies for broad-scale implementation.

Detailed NURP objectives at the national  level are:

     •  to develop  the information  base  required to assess urban runoff
        problems,

     •  to  examine the  adequacy of  current dry-weather  water  quality
        standards when used to judge the significance of storm-dominated
        pollution problems,

     •  to  provide technical  support to  related programs within  EPA,
        other federal  agencies,  and program  participants  at  the state,
        areawide, and  local level, and

     •  to develop  the information base  required  to  identify,  assess,
        and implement effective controls.

NURP PROGRAM DESIGN

A  key  element of  the  NURP program is twenty-eight  prototype  projects being
conducted across the nation.   The locations of these  projects  are  indicated
in Table 1.   NURP  is  also closely following the urban runoff efforts of cer-
tain projects being conducted by others (e.g., USGS), and these non-NURP pro-
jects are also listed  in Table 1.

Individual project  work  plans  were  designed both  to respond  to  local issues
and water quality  management needs  and to contribute to selected elements of
the  broader  information  base needed to address transferability  and national
scale perspectives.  This was accomplished by active participation by EPA (at
both the  headquarters  and  regional  levels) i-n the  development  of individual
project work  plans  and the use of a  matrix of overall input requirements to
ensure effective  representation of an array of  pertinent factors  including
regional  effects,  climatological  conditions,  land use types,  receiving water
bodies, water uses  and problems, and control approaches.  Since project can-
didates were  at  different  stages as a result of  prior 208 efforts  and since
local  match  constraints  precluded any  individual  project from  completely
covering all  aspects of  concern, the matrix approach was also used  to assure
that no information gaps or deficiencies would exist at the completion of all
planned NURP efforts.

One of the key monitoring and evaluation tools is the project progress report
(quarterly) in which each  individual  project sets forth its progress against
the plan for the covered period, problems encountered,  how overcome, resource
expenditures for the reporting period and to date (and compared to plan), and
specific  activities proposed for the next period.  Copies are  reviewed and
commented upon by  the  designated regional project officer, headquarters pro-
ject  officer,  and  consultant  team  member.    These  comments  are consolidated

-------
             TABLE 1.   NATIONWIDE URBAN RUNOFF PROJECT LOCATIONS


                                NURP PROJECTS

 1.   Duram, New Hampshire                                           NH1
 2.   Lake Quinsigamond, Massachusetts                               MAI
 3.   Mystic River, Massachusetts                                    MA2
 4.   Long Island, New York                                          NY1
 5.   Lake George, New York                                          NY2
 6.   Irondequoit Bay,  New York                                      NY3
 7.   Metro Washington, D.C.                                          DC1
 8.   Baltimore, Maryland                                            MD1
 9.   Myrtle Beach, South Carolina                                   SCI
10.   Winston-Sal em, North Carolina                                  NCI
11.   Tampa, Florida                                                 FL1
12.   Knoxville, Tennessee                                           TNI
13.   Lansing, Michigan                                              Mil
14.   Oakland County, Michigan                                       MI2
15.   Ann Arbor, Michigan                                            MIS
16.   Champaign-Urbana, Illinois                                     IL1
17.   Chicago, Illinois                                              IL2
18.   Milwaukee, Wisconsin                                           WI1
19.   Austin, Texas                                                  TX1
20.   Little Rock, Arkansas                                          AR1
21.   Kansas City, Kansas                                            KS1
22.   Denver, Colorado                                               C01
23.   Salt Lake City, Utah                                           UT1
24.   Rapid City, South Dakota                                       SD1
25.   Castro Valley, California                                      CA1
26.   Fresno, California                                             CA2
27.   Bellevue,. Washington                                           WAI
28.   Eugene, Oregon                                                 OR1

                              NON-NURP PROJECTS

29.   Minneapolis, Minnesota                                         MN1
30.   Des Moines, Iowa                                               IA1
31.   Topeka, Kansas                                                 KS2
32.   Reno, Nevada                                                   NV1
33.   Salem, Oregon                                                  OR2
34.   Dallas, Texas                                                  TX2

-------
and returned to  the  individual  project with  requests  for clarification, re-
solution, etc.  as  appropriate.   Comments also include evaluation of data and
its  analysis,   deficiency  identification,  and  constructive  comments  for
offices.

Ancillary oversite by  others  has been structured into the program.   The USGS
established a  joint  Advisory  Technical  Planning Committee through the inter-
agency  agreement  with EPA.  This  joint  Federal  committee is made  up  of re-
presentatives  of the  USGS (Reston and Bay St. Louis offices),  EPA  (OWPO and
ORD),  the  Federal  Highway  Administration,  the Corps  of Engineers, and the
Soil Conservation  Service.   To  assist this joint  committee,  designated non-
federal   observers  have  been  assigned,  including representatives  of  the
Chicago  Metropolitan Sanitary  District, Purdue  University,  Illinois  State
Water Survey,  and the  City Engineer of  Portland, Oregon.   This  provides a
balance of perspectives  and helps assure that all  concerns  are represented.
They meet  on  a  regular basis,  review progress, and  provide recommendations
which are communicated  to all  NURP projects.   This represents  a first level
of peer review for NURP.

OUTPUTS TRANSFERABLE TO LOCAL GOVERNMENTS

The following  are things which  NURP,  upon completion, will  be  able  to pro-
vide to local  and State  governments  and which will be  useful  input  for the
development and or  implementation of the urban runoff element of water qual-
ity management plans:

     A.   Data  base  on concentrations and loads from  urban  runoff,  with
         transferability  factors.   Given that  the monitoring  of  urban
         runoff is  costly  and  time  consuming, and probably  beyond the
         capability  of most  State and  local  governments to  do effec-
         tively,  the  NURP data  base will provide  better estimates  than
         a limited local or State program.

     B.   Methods of  analysis  which  are  being developed  will  be useful
         to others  (especially  where State and local  governments do not
         have  capability for alternate  approaches).  Together  with the
         data base and appropriate local  information,  these will provide
         a basis for:   1) evaluating effects  of urban nonpoint sources,
         and 2) comparing tradeoffs  between point  and nonpoint sources.

     C.   28 Case Studies which will collectively provide useful informa-
         tion  to  other local non-NURP projects  on: monitoring, problem
         identification,  controls and  evaluation, implementation,  and
         institutional aspects such as financing and management.

     D.   Other areas  will  include:  information on  which pollutants are
         of more  concern;  conditions  under  which  urban  runoff will  be
         of more or  less  concern; effectiveness and costs of management
         practices;  and  preliminary data  on  toxic pollutants  in  urban
         runoff.

-------
CONCLUSIONS

It must be  stressed that this is a  preliminary  report.   A final NURP report
will be published in 1983.  The conclusions given below are based on data ac-
quired between initial program start-up in 1978 through mid-1981 and analyses
performed on those data by both individual NURP projects and EPA.

At  this  stage, data  acquisition  is  incomplete;  most  of  the  individual  pro-
jects still  have  active  monitoring  programs.   Analysis and interpretation of
data  is  even less complete,  due  to  the inherent time  lag  required for lab-
oratory analysis and entry into the data management system.

Partial data analysis has been accomplished, and preliminary conclusions have
been drawn  in  some  of the objective areas  - both  at  the  local level and the
national  scale.   In  addition,  certain  screening analyses  have been performed
making use  of  such  data  as are now available.   These analyses provide a use-
ful start to an assessment of the significance of urban runoff as a contribu-
tor to water quality problems.

While some  of the  preliminary (national scale)  conclusions  which  are  pre-
sented below are believed  to be sound,  particular analyses  were  performed
more to illustrate  approaches and procedures than to suggest actual results,
and these  analyses  are  so  identified.   Lack  of data at this  time has  pre-
cluded many  important areas from even being addressed as yet.

Even  for  those  analytical  conclusions  which  are  thought  to be  sound,  the
limited data base on  which they were  based does  not warrant suggesting that
they are all scientifically substantiated at this time.

A  number of local  projects have drawn  tentative  conclusions  on a variety of
aspects related  to  the overall objectives.   All  are  useful  and informative,
and thought  likely  to hold up under further analysis, but like the national
scale activities,  are based  on  incomplete data  and  have  not been reviewed
thoroughly  enough to  suggest  that they are all scientifically substantiated.

NURP  fully  expects  that  the  important conclusions which will  ultimately be
drawn at  both local  and national levels will be scientifically defensible.
NURP is not there yet, as a  careful reading  of  this preliminary report will
confirm.

Consistent with  the  structure of the overall  report,  conclusions  are organ-
ized  in  accordance with  the  three  basic categories  of loads,  water quality
effects,  and controls.
Loads.
     1.  Due to the highly variable nature of urban runoff, a large data
         base  significantly  improves the  confidence  in any conclusions
         drawn.   NURP's  emphasis  on data  acquisition  has,  even at this
         interim stage, produced a data base which significantly exceeds
         the total  pool  of  data  on  urban runoff available  from other
         sources.    Comparisons  have been  made with  other sources which
         extend the utility of the NURP data base.

-------
Urban runoff event  mean  concentrations of five pollutants ana-
lyzed to date  have  been  demonstrated to have underlying proba-
bility  distributions  which  are  log  normal.   The  statistical
parameters (mean and coefficient of variation) have been deter-
mined for  COD, suspended  solids,  total phosphorus,  lead,  and
copper.    Using the statistical  properties  of  the  fundamental
probability distribution,  it is possible  to do the following:

(a)  Define urban storm runoff concentrations in a concise
     way, and facilitate comparisons with other sources of
     data.

(b)  Separate  the  "random" components  of  the variability
     in examining whether there are real, significant dif-
     ferences  in  urban runoff  quality  based on location,
     land use,  or any of the other factors which have been
     suggested  in  the past  to influence  runoff quality.

(c)  Make  statements  concerning the  frequency  with which
     concentrations  of  any   particular  magnitude  will
     occur.

(d)  Interpret data from other sources which report ranges,
     and apply an improved perspective to such information.

(e)  Examine water  quality impacts and  evaluate such ef-
     fects  in  terms of potential  for significant threats
     to beneficial  use.

Some  misleading  conclusions  that might  be  drawn  with
respect to  contaminant concentrations in urban runoff are
illustrated  in Figure 1.   In  Figure  l(a)  the  total  NURP
pilot data  set statistics  are presented by  zone,  or geo-
graphic  region.   Indicated   are   the maximum  likelihood
estimators for the mean,  the median (the best indicator of
central   tendency),   and  the  upper  decile  (that  portion
represented by  only  the  upper ten percent of the events).
In Figure l(b)  the  same  data sets are reproduced with the
lower decile  (that portion represented  by the  lower ten
percent  of  the  events),   which  is   of  little  interest
insofar as  water quality  management  is concerned, and the
upper  decile  (a  range  of   relatively  rare  occurrence)
omitted.  Although the copper data most dramatically illu-
strate the  result,  all  constituents examined indicate the
misleading  impression  that  simple  range  statements  can
give.   If  one  focuses  on  the median values, the change in
perception  of  concentration   magnitudes   is  even  more
striking.

Figure 2 provides an illustrative example of the rationale
which will  be  employed by  NURP to  evaluate  the extent to
which pollutants discharged  by intermittent, variable ur-
ban  runoff  events  may impair  assigned  beneficial  uses of
receiving  waters.   Applying the  analysis  methods adopted

-------
X
01
X
Ol
O
P
cc
I-
111
o
O
o
111
   2000
o>
cc
LU
O
o
o
z
111
I—
UJ
          1000
          LEGEND
                    MEDIAN
                    MEAN
                    UPPER DECILE
                    ZONE 1
                    ZONE 2
                    ZONE 3
                    ZONE
                    ZONE 5
                                    (a)
   2000
   1500
          1000
           SOO
                                             NOTE: RANGES ARE 10% TO 90% DECILE
                                        TP
                                    (b)
Figure 1.    Event  Mean  Concentration  Statistics  for  NURP  Pilot  Data
                    (824 site events) by Geographic Zone
                                                                              Set

-------
o
LU
O


<
HI
GC

CO
 o
 UJ
,o

 o
 o
 UJ
 cc

 CO
               50% MORTALITY

             THRESHOLD BIOLOGICAL EFFECTS
                                                                 DRINKING WATER-



                                                                EPA 24 HR 200 TH
           0.01
                                              YEARS

                                              (a)
        119/1
       10,000 -
il
i
-J DJ
CO >
        1000 -
            VA>>>50% MORTALITY v
            "W//////////////X///
              THRESHOLD BIOLOGICAL EFFECTS
                                                                EPA MAX (200 TH)
(DRINKING WATER) •


EPA 24 HR (200 TH)
                                                                 LEAD

                                                           EFFECT OF URO CONTROL
     100 -
           0.01
                                              YEARS

                                              (b)
     Figure  2.    Mean  Recurrence  Intervals  for  Lead Concentrations

-------
         by  the  NURP  program  to  the  statistical  characteristics  of
         pollutants  in  urban  runoff  developed  from  monitoring  data
         acquired by  the 28 individual NURP projects,  estimates  can be
         made of  the  frequency with which  various  concentration  levels
         in  the water  body will  occur as a  long  term  average.   This
         concentration/frequency   distribution   (see   Figure 2-a)   is
         compared with various  receiving water  target concentrations,
         for example those that relate to the specific beneficial  use of
         the water  body,  or  those which  reflect specific  criteria or
         different  degrees of  use  impairment.   The  procedure  may  be
         extended to illustrate  the  benefit to be expected  in terms of
         water use for different degrees of control  of urban runoff (see
         Figure 2-b).

     3.   It will  be possible,  when additional data have  been analyzed,
         to  develop  quantitative  expressions  which  will  express  dif-
         ferences in urban  runoff  quality as a function of factors such
         as  land  use,  geographical  location,  season  of  the year,  etc.
         The  procedure  adopted  (examining  subpopulations)  appears  to
         provide  a viable  technique  for so doing.  Indications of reg-
         ional,  seasonal,  and land use effects presented in Chapter 5,
         should be considered  tentative,  because  the  data base for per-
         forming  such  comparisons  is still sparse.   NURP expects,  how-
         ever,  that some of the differences suggested  will be confirmed.
         For example,   Denver  and  other locations  with similar geography
         will,  because of  climate  and soil characteristics, continue to
         exhibit  higher  concentrations  of suspended  solids  and  other
         contaminants  closely associated with them.

     4.   About  one-half  of the  substances  on EPA's  priority pollutant
         list sometimes occur in urban runoff.   Heavy  metals (especially
         lead,   zinc,  and copper)  are  much more  prevalent than  organic
         priority  pollutants.   Some  of  the  metals  are   present  often
         enough and in high  enough  concentrations  as  to be considered
         threats to beneficial uses.

     5.   Organic priority pollutants  appear to be  generally present only
         in  certain  urban  runoff  discharges.  Where  present, and  the
         beneficial  use  is  potable water supply,  there is a significant
         concern.   Stream dilution and other mitigating factors may pre-
         vent any real  problem from  developing, but  communities  with
         water  supply  intakes in  close proximity to  urban runoff dis-
         charges are encouraged to check for the relatively limited num-
         ber of  pertinent priority pollutants which have been detected
         in urban runoff.

     6.   Total   suspended  solids   concentrations  in  urban  runoff  from
         available NURP  data appear  to be lower in  general  than  sug-
         gested by other studies.

Water Quality Effects.
Conclusions on  the receiving  water impacts of urban runoff will  be presented
for each designated beneficial use for which it is believed that  urban runoff

-------
will have a  negative  effect on water quality  sufficient to cause impairment
or denial of the use.

Fresh water aquatic life.

     1.  The effect  of  oxygen  demanding substances  in  urban runoff on
         dissolved oxygen  levels in  receiving water  has  not been exa-
         mined  yet.   This  analysis  will be  done when  sufficient data
         become available.

     2.  The potential  for  suspended  solids to have a significant dele-
         terious effect has not yet been assessed.   Thus  analysis will
         be done when sufficient data are available.

     3.  Organic priority   pollutants do not  appear  to pose  a general
         threat to  fresh water  aquatic life.   This  is based  upon  the
         limited available  data  on  the  frequency  with which  they  are
         found  in urban runoff discharges and the  measured end of pipe
         concentrations  relative to published toxic criteria.

     4.  At  this  time  heavy  metals  appear  to be the  urban runoff con-
         taminants that have  the greatest potential  for impacts  on  the
         aquatic life  beneficial  use.   This conclusion  is  based  on  the
         fact  that  a number  of heavy  metals  are consistently found in
         urban runoff in high concentrations relative to suggested toxic
         limits.

     5.  It  appears  reasonable to conclude that  some  degree of impair-
         ment  of  this  benefical  use  is  probably  common in  many  water
         bodies which  receive  urban  runoff.   With some exceptions, how-
         ever,   the degree of  any impairment that exists does not appear
         to  be of  sufficient  magnitude to generate  serious local con-
         cern.   This conclusion is based on the following:

         •  The screening  analysis  suggests that local  conditions
            can be  expected where the  magnitude and  frequency of
            intermittent concentrations of  some  metals  would have
            adverse effects.   Conversely, there  will  be local con-
            ditions  where   adverse  effects  are  very unlikely  or
            extremely  rare  occurrences.  The  governing  local cir-
            cumstances  are  storm flow  characteristics and average
            rainfall intensity,  both  of which  vary with geographic
            locality.

         •  Absence of  evidence from NURP  projects confirming im-
            pairment  of this  beneficial use  which  is  considered
            significant  at  the local  level.   Bellevue, Washington,
            with a  high degree of local  awareness  and interest in
            protecting  a fishery,  has  concluded that chemical con-
            stituents  in urban runoff  create  no  problem for their
            sport fishery.   Glen Ellyn  (NIPC) documents high metals
            concentrations which undoubtedly strongly influence the
                                    10

-------
            nature of the fishery which is present, but local opin-
            ion  does not  characterize  this  as  a problem,  i.e.,
            local  residents  do  not consider the  degree  of use im-
            pairment to -be significant.

         •  In an  EPA examination  of  over 10,000 fish  kills that
            occurred from 1970 to 1979, only 265 were attributed to
            runoff of any kind,  and less than one-half of that num-
            ber were associated with urban runoff.

     6.   Indications to date do not provide any basis for believing that
         a  denial  of  this  benefical  use  is   a  widespread,  general
         occurrence.

Marine aquatic life.

     1.   Adverse  effects  of  urban runoff  in marine  waters will  be a
         highly  specific  local  situation.   It  is not  a beneficial use
         generally threatened by urban runoff, though specific instances
         where it  is  impaired or denied can be of significant local and
         even regional importance.

     2.   A significant  impact of  urban  runoff on  shellfish harvesting
         has  been well  documented by  the  Long  Island,  New York NURP
         project.

     3.   Coliform  organisms  present  in  urban  runoff  are  the  primary
         pollutant indicator of concern.

Recreation.

     1.   Pollutants  in urban runoff  which pose  the most  significant
         threat  to  this  beneficial use are  pathogenic  organisms  (as
         possibly  indicated  by  the presence of coliform bacteria), nu-
         trients (particularly phosphorus), and suspended solids.

     2.   Coliform  bacteria  are  present  at  high  levels  in urban  runoff
         and  affect  all  types of water  bodies  — streams,  lakes,  bays,
         estuaries,  and  oceans.   Whether the potential  for  use impair-
         ment is realized depends principally upon physical factors such
         as the location of the discharge relative to swimming areas and
         the  degree  of dilution and  dispersal.    In  this  sense,  actual
         use  impairment  will be  a local  condition.   This  problem has
         been identified by  a number  of projects; however, data are not
         yet available for characterizing the extent and severity of use
         impairment.

     3.   Water bodies  particularly s.usceptible to  recreational  use im-
         pairment  by  nutrients  are  lakes.   In the cases where at least
         some  information  is presently  available (Lake  Quinsigamond,
         Irondequoit Bay, Lake  Austin,  Lake George) it is apparent that
         the degree  of  beneficial  use  impairment varies widely, as does
         the  significance  of the  urban runoff contribution.   Detailed
                                    11

-------
         analysis which  will  permit establishing  the  relative signifi-
         cance of urban runoff compared with other sources is not avail-
         able  at this  time except  for  Lake  Quinsigamond  but can  be
         expected to be  very  site specific.   At Lake Quinsigamond urban
         runoff  was  found to  be  a significant  source of  the critical
         nutrient phosphorus,  such that a proposed water quality manage-
         ment  plan will  include the objective of  reducing  urban runoff
         loads.   By  contrast,  very  preliminary  indications  are  that
         nutrients  in   urban   runoff  are  not  causing any  appreciable
         negative impact  on Lake  Austin and controls do not appear war-
         ranted,  pending  results of  analysis  of  long  term effects.

     4.   Adverse impacts  on recreational  use caused by suspended solids
         have not yet been evaluated.

Public water supply.

     1.   Ground  water  aquifers do  not  appear to  be threatened by pol-
         lutants in urban runoff caused to percolate through the ground.
         This  finding  is based on  only one NURP  project (Long Island)
         which  has specifically examined this issue and  must be consi-
         dered in that light.

     2.   Lakes and impoundments that receive urban runoff and also serve
         as water  supply sources are  being addressed  in NURP, but  at
         this  time results  available  are too limited to warrant drawing
         preliminary conclusions of a general nature.

     3.   Rivers  or streams  which  serve  directly as water supply sources
         with  intakes  in  close proximity to urban runoff discharges are
         probably rare  nationwide,  but  some  do exist.   From information
         and preliminary analysis  developed to date, it is inferred that,
         where  water supply  intakes  are  in  close proximity  to  urban
         stormwater  discharges, a serious enough  potential  for adverse
         impacts  exists   that  careful  examination of  the  situation  is
         prudent.  The  principal  contaminants  of concern  suggested  by
         preliminary results  include  heavy metals, a  limited number  of
         the  organic priority  pollutants,  and coliform  bacteria.   Al-
         though  the  potential  problem is present,  there  are mitigating
         factors which  reduce   the  likelihood  that a  problem situation
         will  actually  exist at any given location.  The present status
         of our  analysis does  not  warrant  any  firm conclusions  to  be
         drawn with  regard to  the  prevalence of  actual  problem situa-
         tions of this type.  However,  based on the fact that local fac-
         tors  can  significantly  reduce the  potential,  and  the belief
         that  relatively  few water  supply intakes exist at stream loca-
         tions  dominated  by urban  runoff,  NURP believes that any pro-
         blems  that  do  exist   will  be  limited  rather than widespread
         occurrences.
                                    12

-------
Control effectiveness.

     1.  Some detention basins are quite effective at reducing pollutant
         concentrations in urban runoff, with reductions on the order of
         90 percent  or  better  reported.   High  removals of  suspended
         solids  and1 heavy  metals,  especially  copper,  are  noted,  with
         somewhat lower  removal efficiences  for phosphorus and chemical
         oxygen demand.

     2.  Other NURP  project detention  basins  do not appear  to  be per-
         forming very  well  in  terms of reducing effluent concentrations
         and may not be capable of providing a significant water quality
         improvement.  The  performance  of  some is quite variable, being
         fairly good for some  storm events (~ 50 percent reduction) and
         quite poor  for  others.   Some detention basins are consistently
         poor performers.

     3.  The design  features  of detention basins play  a  very important
         role  in their performance  in terms of  reducing  pollutant con-
         centrations.  Data suggest that  basins designed to merely con-
         trol  water  quantity  (drainage) problems may  not function well
         in  water quality terms.   The installation of  an arbitrary de-
         tention basin could even exacerbate water quality problems.  As
         more NURP  project data become available, it will  be possible to
         make  definitive recommendations  on detention  basin  design for
         water quality improvement.

     4.  Street  sweeping is effective for its  original purpose,  but it
         does  not appear to be universally effective  in  reducing urban
         runoff pollution.  Only one project out of six has been able to
         show  any water quality  improvement  as a  result of  even  very
         intensive  (daily) street sweeping.

     5.  The  effectiveness  of  street sweeping  as  a  water  quality BMP
         should  be  measured in terms of actual pollutant reduction in
         the  runoff,  not  determined theoretically  from estimates  of
         pounds  removed  from  street  surfaces  obtained  from examination
         of  street   sweeper hopper contents,  pounds  remaining  on  the
         street surface, etc.

     6.  Data  are insufficient  to allow firm projections  to  be  made as
         to  those conditions  under  which street sweeping might  be able
         to provide a water quality improvement, but preliminary indica-
         tions are  that  it might be  limited to semi-arid  parts  of the
         country with pronounced wet and dry seasons.

     7.  Data  indicate that under certain conditions street sweeping may
         actually increase  urban  runoff pollutant levels, but this con-
         clusion must  be considered  highly speculative until  more data
         become  available.  A possible mechanism  is that  sweeping can
         break up larger particles  making them available for washoff by
         smaller, less  intense  storms with less runoff volume for dilu-
         tion.  At  its conclusion NURP will be  able  to make definitive
         statements in this regard.
                                    13

-------
     8.   Recharge  Basins  appear  to  offer  promise  as  effective  and
         economical BMP's for reducing urban runoff pollution to surface
         waters, without  creating  a threat of adverse  impact  on ground
         waters.  This conclusion is tentative, and is based on prelimi-
         nary data from  monitoring  efforts on one device, and screening
         analyses  to  examine applicability.   Applicability can  be  ex-
         pected  to  vary on  the  basis of  site-specific conditions such
         as  soil  percolation rates,  depth to water  table, topography,
         proximity to water  supply wells,  and  the  like.   Preliminary
         indications are, however,  that recharge basins could have rela-
         tively broad applicability.

GENERAL DISCUSSION

In summary,  the possibility  of having water quality problems caused by urban
runoff varies  in type and degree.   Whether or not a problem may truly exist
is strongly  influenced  by a  variety of local, site-specific characteristics.
These include:

        Local climatology and hydrology;

     •  Type  of water body  which  is  the  principal  recipient  of urban
        runoff;

     •  The particular beneficial use desired; and

        Local  attitudes,  resources,  values   —   especially  values  as
        applied  specifically to  individual  water bodies  in  an  area.

The NURP screening  analysis  to date suggests  that there are a lot of places
where urban  runoff  is unlikely to  result  in  problems  (quantifiable  problems
reflected  in standards  violations  or use impairment).  The  same screening
analysis also suggests that there should be a significant number of locations
where urban  runoff could theoretically result in the occurrence of a quanti-
fiable problem.   Some "possibilities  of  problem situations" would have to be
discounted on the basis  of  the  actual beneficial uses.   For  example,  "pro-
blems" based on drinking water standards not being met would be problems only
if the  receiving water  is actually used as a drinking water  source.   Local
attitudes  or alternatives could further  reduce  a theoretical  or potential
problem to the category of no problem from a practical standpoint.

NURP is  unable,  at this time, to estimate the probable relative distribution
of sites in  the nation which fall  into  the above alternate categories.  Its
analysis efforts  are  continuing  to try to  make  this  determination.   At this
time, however,  NURP  is  not highly optimistic about its ability to be able to
make a credible generalization  of this type.  However,  some  rather  sweeping
generalizations that will be true more often than not are offered.  The water
quality  impact  of  urban runoff will be greatest in small urban impoundments,
especially those with no  other  influent  flows.   Aside  from  possible  near-
shore effects,  the water  quality  impact of urban runoff will  diminish with
impoundment size and, consequently, so should the potential for occurrence of
water quality  problems.   Similarly, the  water quality  impacts  of urban run-
off will be greatest  in small streams, especially those that headwater in the
                                    14

-------
urban area.  As  stream  flow and size increase, the potential for water qual-
ity impacts caused  by  urban runoff should decrease  as  should problem poten-
tial.    Furthermore,  in  this  regard,  swiftly flowing streams  should  be less
vulnerable than  more sluggish ones.   Bays and estuaries  are so complex that
even such rather simplistic generalizations as the foregoing are difficult to
justify and the same is true of the near-coastal  oceans.

There are  plateau and  threshold  effects for many pollutants  that influence
their impact on  a given receiving water and,  consequently,  the strategy for
their control.    As  an  example, consider the case where the Level of a toxic
substance is four times the LC-50 value for a species of fish associated with
the desired beneficial  use.   If urban runoff is  the  only source of the toxi-
cant,  and  if  the  implementation  of  a  given control  measure results  in  a
50 percent reduction in loads, the money spent would have been wasted insofar
as  attainment  of  the   designed  beneficial  use  is  concerned.   To be  sure,
pounds  of  pollutants  would  have  been removed at a  (perhaps) cost-effective
rate  of so many dollars  per pound,  but  there  would  still  be none of the
sought-after fish  in the  water and,  consequently, the desired beneficial use
would still  be  denied.    Similarly,  the  removal  of a contaminant  that  is at
such  low levels  that  it is not interfering with  a designated beneficial use,
simply  because  the  substance  is  present  in  urban  runoff, results  in the
incurrence of  costs but  produces  no benefit because the  beneficial  use was
not being impaired in the first place.

While the  tools  are available for conducting cost-effectiveness analyses and
the procedures  are rather  straightforward,  NURP is  not  so  fortunate  in the
area  of benefit-cost analysis.   Damage theory and utility  cost tools  exist,
but their  application  is  far from straightforward and is subjective at best.
The Agency is  currently attempting to remedy this situation,  and successful
approaches, if found,  will be applied by NURP.

Because of the  local  nature of water  quality  problems  in terms of existence
and cause,  and  because  in many cases  much more attractive approaches than
control  of urban runoff will be available for addressing problems where they
are found  to truly exist, it does not  appear  desirable to adopt a universal
requirement for urban runoff control  or treatment.

Urban runoff does appear to contribute significantly to certain water quality
problems in certain areas (there are a number of  well-documented cases in the
NURP  program), and  there  are situations where control of urban runoff is the
only realistic action that is available to address the problem.  However, the
practicality of being able to solve some problems will be constrained in some
cases by technological  limitations, in some cases by costs, and in some cases
by  physical  site  constraints.   Some  examples of such  constraints  include:

     Technological.  Situations which require very high degrees of removal of
     a pollutant  may prove to be beyond the performance capabilities of con-
     trol   measures available.   Example:   Suspended  forms of  a  contaminant
     lend themselves to  removal  whereas,  if the soluble  component is  a sub-
     stantial   fraction  of  the total,  control  measures  may  not be able to
     reduce the total  load sufficiently to achieve desired effects.
                                    15

-------
Cost.   The cost associated with reducing the overall load of a pollutant
from an  entire urban  area  could prove  to be  prohibitive,  even though
relatively cost  effective when  applied  in restricted  areas.   Usually,
the overall urban  runoff  load must be addressed  if the "problem" is to
be successfully attacked.   However,  it may provide real  benefit to apply
efficient, cost  effective controls  serving only  limited  areas  where it
can be done conveniently.

Physical  Site Constraints.   Even where  efficient,  cost-effective  con-
trols may  be  available for reduction of the specific pollutants contri-
buting to  a  particular problem,  the physical  space  to  implement  them
may not be obtainable.   For example, Central  Business  Districts  may be
a  significant pollutant  source,  but  provide  little  opportunity  for
control.
                               16

-------