United States
                    Environmental Protection
                    Agency
Water Engineering Research
Laboratory
Cincinnati OH 45268
                    Research and Development
EPA/600/S2-85/043  June 1985
&EPA         Project  Summary
                    Characterization  of the  Water
                    Quality of  the  Lower
                    Mississippi  River
                    W. M. Grayman
•
-------
is fully documented in a separate report
of the same title (see Project Report
ordering information at back).


Introduction
  The water quality characteristics  of
surface water used as a source of drinking
water is important information for water
utilities. This  characterization may be
used to  make decisions about modifying
the water treatment  process, to design
water quality monitoring programs, or to
identify desirable reduction in  pollutant
discharges to the surface water.
  Characterization of water quality takes
several forms. When available, in-stream
water quality samples may be analyzed to
identify  spatial and temporal trends and
to define the movement and transforma-
tion of pollutants in surf ace water. Mathe-
matical models operating at a wide range
of detail may be applied  to address a
range of scenarios or what-if questions.
For example, the impact of alternative
discharge limitation policies or varying
hydrologic conditions may be studied with
such models
  A recent  study (Conceptual  Develop-
ment  of a Toxic Screening Model, EPA/
600/9-84/018,  September  1984,  U.S.
Environmental Protection Agency, Cincin-
nati, Ohio) demonstrated the application
of a water  quality model  linked to the
Reach File System, a nationwide com-
puterized  data base of stream related
information and application programs
maintained  by the U.S. Environmental
Protection Agency (EPA). This screening
model was applied to the lower Missis-
sippi River to characterize the concentra-
tions  of organics in  the river  resulting
from point source discharges. This study
concluded that the joint use of the Reach
File System and a water quality model
provided an economic and feasible means
of performing such characterization.
Another study demonstrated the applica-
bility  of water quality models to predict
toxic  pollutant concentrations  resulting
from upstream industrial discharges.
  The present study was conducted as a
natural  extension of the toxic screening
application to the lower Mississippi River.
Specific areas investigated in this study
include further statistical analysis of the
water quality data on the lower Missis-
sippi  River, identification  of additional
sources of  data for  characterizing the
water quality of the lower Mississippi
River, and a review of models and methods
that may be used in this further character-
ization. This information is intended as a
source document for future water quality
management studies with specific appli-
cation to the lower Mississippi River


Data Base
  Application  of  mathematical models
requires a data base both as a means of
defining parameters for the model and
verifying the  results  of the modeling.
Within the context of the present study,
the required  data  base must  include
information  in three major areas: In-situ
water  quality, stream  flow and flow
dynamics, and point discharge data. Also
required is general, geographically refer-
enced  information on the river and  its
environs.
  The  specific area of  interest  in this
study is the  mamstem of the Mississippi
River between Baton Rouge and New
Orleans, Louisiana.
  In-situ water quality data on the lower
Mississippi River is routinely collected by
several government agencies and water
utilities. Additional data have been col-
lected for special studies or for industries
using the river as a source of water supply
or as a discharge location. The primary
sources of data on the level of organics
concentrations in the  lower Mississippi
River arethe Jefferson Parish Department
of Public  Utilities,  the Sewerage and
Water  Board of New  Orleans,  and  the
Louisiana Department of Environmental
Quality (formerly Louisiana Department
of Natural Resources). Jefferson Parish
and  New  Orleans perform weekly and
daily sampling, respectively, for volatile
and nonvolatile organics. Louisiana col-
lects samples and analyzes  for 29 con-
stituents on  a weekly to bi-weekly basis at
12 stations on the lower M ississippi River.
  The major source of streamflow infor-
mation  is the  elevation and discharge
measurements made  at the  Tarbert
Landing, Mississippi, gage located approx-
imately 75  miles upstream of Baton
Rouge This gage is 8.2 miles downstream
of the  inlet channel  to the Old River
Control Structure, where approximately
30% of the streamflow in the Mississippi
River is diverted to the Atchafalaya River
Basin. The gage is maintained bytheU.S
Army Corps of Engineers,  and elevation
measurements are  taken  every 3 days.
Daily flow is estimated from these meas-
urements and published by  the  U.S.
Geological Survey (USGS) on a yearly
basis. Continuous flow records have been
published by the USGS since 1972, and
they are available in Corps of Engineers
reports dating back to  1936.
  The  USGS  has cooperated  with  the
Louisiana Department of Public Works to
perform several studies aimed at identi-
fying  the time-of-travel  and dispersion
characteristics of solutes  in  the  lower
Mississippi River. Based on dye tracer
studies  performed in  1965, 1974, and
1975, the  USGS has estimated  travel
time and dispersion characteristics under
varying  flow conditions from the Arkan-
sas-Louisiana  state line to below New
Orleans.
  The primary source of discharge infor-
mation  is data collected by EPA in the
process of issuing or monitoring permits
for the National Pollutant Discharge
Elimination  System  (NPDES).  For the
State  of Louisiana,  permits are issued
and monitored by EPA Region VI in Dallas.
Currently,  93 major dischargers are lo-
cated  on the  lower Mississippi  River.
These dischargers are required to submit
quarterly Discharge Monitoring  Reports
(DMR) based on the sampling  require-
ments in their permit. This report will
become a monthly  requirement m the
near future. Information from the DMR is
put into the computerized Permit Compli-
ance System (PCS), and various reports
are generated.
  In the past, few industries have been
required to sample  for specific  organic
compounds, which has resulted in rela-
tively  few discharge data on organics. In
the past few years, many of the permits
written  for the organic chemical indus-
tries have required the industries to report
organic concentrations in terms of aggre-
gate organic indices such as total purge-
able halocarbons and total  purgeable
aromatics. In calculating such aggregate
indices, measurements of the individual
compounds are made and mathematically
summed. The industries are required to
save those individual measurements, and
EPA may request them for a period of up
to 3 years. Thus, though  an extensive
data base on organic chemicals in dis-
charges is  not  currently  available, a
mechanism for obtaining the data is in
place.
  No  information on  minor permits  is
stored in the PCS. These data are stored
in hard  copy files along with monitoring
results.
  The present study assembled a compu-
terized  data base of currently available
information on  organics  in the  lower
Mississippi River. The sources of these
data,  sampling stations, period  of sam-
pling, and pollutants sampled are sum-
marized in Table 1. The sampling data
have been organized in a common format
with the following information stored for
each observation.

-------
Table 1.    Summary of Data Stored in Computerized Project Data Base
Data Source
Jefferson Parish
Jefferson Parish
New Orleans
Louisiana Dept. of
Natural Resources
uses
No. of
Sampling
Stations
1 Feb.
1 Oct.
1 Sept.
12 Dec.
1
Period of
Sampling
1977
1979
1980
1982
1977
- Oct. 1979
-Nov. 1983
- Dec. 1983
-Sept. 1983
-Sept. 1981
No. of
Parameters
153
74f
fit
29
Suspended
Total No. of
Observations*
34.455
1.830
5.190
1 1,339
1.734
U.S. Army Corps of
 Engineers
1977 - 1982
                                                       solids
Streamflow
2.191
*An observation corresponds to a sample analyzed for a single pollutant.
^Additional routine sampling was performed but was not inserted into the project data base.
 • Source code (source of data)
 • Observation number (unique number
   for each sample)
 • Station number (code for each sam-
   pling location)
 • Substation number (lateral location of
   sample within the river)
 • Year
 • Day
 • Constituent number (unique number
   for each constituent)
 • River mile
 • "Less  than" code  (code indicating
   whether  sample  concentration  has
   been given as less than given value)
 • Value (concentration  for  pollutants,
   flow rate for Streamflow)

Water Quality of the Lower
Mississippi River
  Though no in-depth study of the water
quality of the lower Mississippi River has
been performed previously,  there have
been several studies aimed at specific
aspects of the problem. Studies by Region
VI, the  State  of Louisiana,  and EPA's
Drinking Water Research Division have
been reviewed.  The  Region VI  studies
indicated  the prevalence  of  selected
organics in  the  river and a  substantial
reduction in  total organic  concentrations
from 1977 to 1979. A study by the State
of Louisiana  identified significant fluctua-
tions in  the  levels of  1,2-dichloroethane
in the river and attempted  to correlate the
peaks with occurrences of known spills.
The Drinking Water  Research Division
used a large data base on  organics in the
•iver  collected  at Jefferson  Parish,
      Louisiana. That study noted a significant
      variation  in  concentrations,  with little
      apparent  correlation to Streamflow or
      sediment.
       In the present study,  long-term and
     short-term temporal and spatial variations
     were investigated. Graphical and statis-
     tical  analyses indicated  a  significant
     short-term, high-frequency variation in
     concentrations. Concentration variations
     of 500% to 1000% in samples taken a few
     days or a week apart were not uncommon.
     The significant variation is reinforced by
     statistics derived from information pre-
     sented in Table 2. For the six constituents
     sampled at  New Orleans, the ratio of
     standard deviation to mean (a normalized
     measure of variation known as the coef-
     ficient of variation) ranges from 1.47 to
     8.21, and the ratio of maximum values to
     mean values varies from 16.31 to 181.16.
     A third statistic is  the  percentage  of
     samples in which the constituent is not
     detected (i.e., when the concentration is
     equal to zero or belowthe detection limit).
     This figure varies from 96% to 19%. All
     three statistics indicated significant vari-
     ations in  the observed values of the
     constituents. The results are similar for
     the data collected at Jefferson Parish.
       Analysis of the data at Jefferson Parish
     indicated a moderate, general, long-term
     trend toward decreasing concentrations.
     However,  exceptions to this  trend  do
     occur. For example, the average concen-
     trations for  bromodichloromethane are
     increasing, and those for other constitu-
     ents such as chloroform and trichloro-
     ethylene show significant random vari-
     ation between years rather than a general
     trend.
       Spatial variation in water quality may
     be characterized by differences in any of
three dimensions: the longitudinal (in the
direction of primary advective flow), the
lateral (horizontally perpendicular to the
direction of primary advective flow, and
the vertical. In a river environment,
primary emphasis is generally placed on
longitudinal variation—namely, the
changes in concentration at different
river miles. The sampling data collected
and analyzed by the Louisiana Depart-
ment  of Natural  Resources  provides a
potential resource of analyzing the longi-
tudinal  and lateral variations in  water
quality of the Mississippi River. These
data indicate increased concentrations,
occurrence of detectable  levels of pollu-
tants, and greater lateral variations in
concentration  in  the vicinity of  major
industrial dischargers.  However, defini-
tive statements on longitudinal and lateral
variations  are  limited  by the relatively
short history of water quality data that is
availableflessthan 10 months) and by the
reporting precision, which measures only
to the nearest microgram  per liter (//g/L).
  Several means are available for deter-
mining whether sets of constituents are
correlated. If correlation exists, sampling
requirements may be minimized by  re-
quiring only samples of the independent
constituent. In addition, other inferences
may be drawn concerning the source and
transformations of correlated constitu-
ents. Pollutant  constituents were  com-
pared in pairs using  both graphical and
statistical methods. Little or no correlation
was detected.  Values of  the correlation
coefficient(R)varyfromahighof0.56toa
low of essentially zero, with nearly 80% of
the absolute values of R less 0.1. The data
were  further  investigated  to  identify
whether any relationship existed between
the joint occurrence of detectable levels
of constituents. For each sample, a de-
tected presence in a sample was assigned
a value of 1, and an undetected value was
assigned a value of 0. The correlation
analysis applied to this  transformed data
set indicated no significant improvement
in correlation.


Available Models and Methods
  Over the past 25 years, the develop-
ment and use of computer-based math-
ematical  models of the movement and
transformation  of pollutants in streams,
lakes, and estuaries has been extensive.
In more recent years, models have  been
developed for tracing organics, pesticides,
and other toxics. These models form the
pool of currently available tools that may
serve  as the basis for  modeling of the
lower Mississippi River.

-------
Table 2.   Statistical Characteristics of Selected Pollutants at New Orleans and Jefferson Parish. September 1980 to November 1983
                                                                                   Concentrations
  Sample
   No.
Constituent Name
   No. of
Observations
   No. of
 Non-Zero
Observations
Average
 (fJ9/LJ
Standard
Deviation
 (W/L)
                                                                                   Minimum
                                                                                     Value
Maximum
  Value
  tfjg/Lj
Jefferson Parish samp/ing data:

     7   Chloroform
    11   Bromodichloromethane
    15   Chlorodibromomethane
    19   Bromoform
    23   1.2-Dichloroethane
    35   Trichloroethylene
    39   1,1,2-Tnchloroethane
    43   Carbon Tetrachloride
    47   Dichloromethane
    50   Bromochloromethane
    54   Dichloroethylene
    58   Chlorobenzene
   126   Tetrachloroethylene
   134   1,1,2.2-Tetrachloroethane

New Orleans sampling data.
                               118
                               118
                               118
                                58
                               118
                               118
                                58
                               118
                                58
                                58
                                58
                                58
                                58
                                58
                 117
                  76
                  14
                   1
                 111
                  67
                  17
                  39
                  29
                   0
                   0
                   0
                   0
                   0
                 4946
                 192.0
                  16.7
                   0.3
                1144.6
                  65.6
                  56.0
                  18.8
                  66.9
                   00
                   00
                   0.0
                   0.0
                   0.0
               554.5
               3085
                66.2
                 2.6
              1801.6
               179.0
               218.2
                61.2
               130.7
                 00
                 00
                 0.0
                 0.0
                 0.0
                0.0
                0.0
                00
                00
                0.0
                00
                00
                00
                00
                0.0
                0.0
                0.0
                00
                00
  3850.0
  1850.0
   470.0
    20.0
 101200
  1491 0
  1570.0
   6000
   740.0
     0.0
     0.0
     0.0
     0.0
     0.0
7
11
15
23
35
43
Chloroform
Bromodichloromethane
Chlorodibromethane
1 ,2-Dichloroethane
Trichloroethylene
Carbon Tetrachloride
847
847
847
847
847
847
600
241
37
687
32
93
300.4
91 3
138
1363.0
138
27 1
4409
3940
959
2788.8
113.3
1307
0.0
0.0
0.0
0.0
0.0
0.0
49000
7100.0
2000.0
34800 0
2500.0
17000
  Models used in studying the movement
 and transformation  of  pollutants  in  a
 water body are generally composed of
 two components: a hydraulic component
 and a  water quality component. The
 hydraulic component is used to calculate
 the  spatial and temporal  variation  of
 discharge  and  depth within the water
 body. The  water quality component is
 used to determine the spatial  and  tem-
 poral variation of pollutant concentrations
 in the water body. The following models
 were reviewed in terms of their potential
 application to water quality modeling of
 the lower Mississippi River.

 The Routing and Graphical
 Display System
  The Routing and  Graphical Display
 System  (RGDS) is  a modular  set  of
 computer-based files and  programs de-
 veloped for EPA. The RGDS uses files of
 the EPA Reach File System and performs
 routing of  pollutants through  a stream
 network using simplified first-order decay
 functions.  This  system can produce  a
 variety of graphical displays of informa-
 tion generated by the routing process or
 other information  stored in EPA-main-
 tained data files.
                               Water Quality Analysis
                               Simulation Program
                                 The Water Quality Analysis Simulation
                               Program  (WASP)  is a general-purpose
                               water quality  model developed for EPA.
                               Water bodies  are  represented in WASP
                               by interconnecting compartments. These
                               compartments may be linked  to permit
                               one-, two-, or three-dimensional models.
                               WASP is a time-varying model in terms of
                               both loading  conditions, flow, and ex-
                               change coefficients. The kinetic/reactive
                               equations governing pollutant  transfor-
                               mations and interactions may be repre-
                               sented by simple linear  relationships,
                               interactive linear relationships, or a wide
                               range of  nonlinear interactive  relation-
                               ships.


                               Exposure Analysis
                               Modeling System
                                 The Exposure Analysis Modeling Sys-
                               tem (EXAMS) was developed by EPA to
                               evaluate  the  ultimate (steady-state) be-
                               havior of synthetic organic chemicals in
                               aquatic ecosystems. The  system was
                               designed to  provide information  on
                               chemical exposure  (ultimate  expected
                               environmental concentrations), fate (dis-
                                            tribution of the chemical in the system),
                                            and persistence (time required for effec-
                                            tive purification of the system). A system
                                            is represented by a series of compart-
                                            ments that correspond to levels within
                                            the water column (littoral and benthic ma
                                            river, and epilimnion and hypohmnion in
                                            a  lake) or  to different two-dimensional
                                            locations. A wide range of transforma-
                                            tions is represented, including ionization,
                                            sorption, photolysis, hydrolysis, biolysis,
                                            oxidation,  and volatilization   Hydraulic
                                            transport  between compartments is
                                            through advection and  dispersion

                                            Chemical Transport and
                                            Fate Model
                                             The Chemical Transport and Fate Model
                                            (TOXIWASP)  was developed by  EPA's
                                            Environmental  Research Laboratory in
                                            Athens, Georgia,  as a means of simula-
                                            ting the transport  and fate of toxic
                                            chemicals  in water bodies. The  model
                                            combines the chemical  transformation
                                            processes adapted for the EXAMS model,
                                            the  transport and program structure of
                                            the WASP model, and a simple sediment
                                            balance algorithm.  Like  WASP, it is  a
                                            dynamic model designed to  operate in
                                            conjunction with a hydraulic model or

-------
other source of advective/diffusive flow
information.

Sediment- Contaminant
Transport Model
  The Sediment-Cohtaminant Transport
Model (SERATRA) is an unsteady-state,
two-dimensional (longitudinal and verti-
cal),  finite  element, sediment-contami-
nant transport model. The sediment-
contaminant interaction  is represented
bythree sub-models: a sediment transport
sub-model; a dissolved contaminant
transport sub-model; and a particulate
contaminant transport sub-model.

Adjusting Parameters for
Specific Situations
  To apply a  water quality model to a
specific situation, the parameters of the
model must be adjusted so the model will
represent,  to the degree possible, the
results of the actual processes occurring
within the water body. The fine tuning of
parameters generally encompasses both
the selection of some parameters based
on past experience or commonly accepted
literature values and  their further ad-
justment to match field data for the
specific situation being modeled.

Recommendations
  The lower Mississippi River provides a
 good framework for demonstrating water
 quality  management,  with particular
 emphasis on the use of surface water as a
 source of drinking water. The availability
 of a significant data base and applicable
 models, and the general interest in identi-
 fying water quality problems and  pro-
 posing mitigating solutions contribute to
 this area's applicability for further study
 in this field.
  To support future efforts in the area of
 water quality management, the following
 actions are recommended to augment the
 present data base:

 • The data base on  discharge data for
   the lower Mississippi  River should be
   expanded by requesting EPA Region VI
   to secure supplemental specific organic
   discharge data from industries on the
   river.
 • Scheduling and quality control should
   be coordinated among the monitoring
   programs of Louisiana, New Orleans,
   Jefferson Parish, and others to assure
   a comparable and compatible water
   quality data base for modeling efforts.
   The State should consider modifying
   its monitoring program to provide
  greater precision in reporting organic
  concentrations and to stagger the time
  of sampling to correspond with  the
  approximate  travel  time  between
  stations.
• Industries along the river conducting
  water quality monitoring studies should
  be requested to provide this informa-
  tion as part of an overall water quality
  data base for the lower Mississippi
  River.

  The full report was submitted in fulfill-
ment  of Contract No. 2211NAST  by
Walter M. Grayman  under the sponsor-
ship of the U.S. Environmental Protection
Agency.
  Walter Grayman is with Walter M. Grayman Consulting Engineers. Cincinnati,
    OH 45229.
  'Richard G. Eilers is the EPA Project Officer fsee below).
  The complete report, entitled "Characterization of the Water Quality of the Lower
    Mississippi River," (Order No. PB 85-190 981 /AS; Cost: $13.00, subject to
    change) will be available only from:
          National Technical Information Service
          5285 Port Royal Road
          Springfield. VA 22161
          Telephone: 703-487-4650
  The EPA Project Officer can be contacted at:
          Water Engineering Research Laboratory
          U.S. Environmental Protection Agency
          Cincinnati, OH 45268
                                                                      *U.S.Government Printing Office: 1985 - 559-111/10860

-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
        OCDC329   PS

        U  S  ENVIR  PROTECTION AGENCY
        REGION 5  LIBRARY
        220  S  DEARBORN  STFSFT
        CUCAGO               IL

-------