E
 03
 D)
 O
 03
OG
 0
 03
 (D
 03
 C/)
 CD
     •c
     O
     Q.
     0
     DC
C
05
D)
_O
O
C
JZ
O

-------
I  *
     Technical Analysis of Response of
   Chesapeake Bay Water Quality Model
             to Loading Scenarios
                   A Report of the
                Modeling Subcommittee
             Chesapeake Bay Program Office
                    Annapolis, MD
                                  )>'->' ^
                     April,1994
•: ';! r-'eci!cn Agency
	L..! Resource

•'A 13107
     Printed by the U.S. Environmental Protection Agency for the Chesapeake Bay Program

-------
Principal Authors
                          Robert V. Thomann
                           Manhattan College
                             Riverdale, NY
                           James R. Collier
                DCRA Water Resources Management Division
                           Washington, D.C.
                              Arthur Butt
                Virginia Department of Environmental Quality
                             Richmond, VA
                           Elizabeth Gasman
                  Maryland Department of the Environment
                             Baltimore, MD
                            Lewis C. Linker
                 U.S. EPA Chesapeake Bay Program Office
                             Annapolis, MD

-------
                        THE MODELING SUBCOMMITTEE
                                    OF THE
                          CHESAPEAKE BAY PROGRAM
James R. Collier, Chairman of the
Subcommittee,
Program Manager
Water Resources Management Division
Washington, DC
Michael Haire
MD Department of the Environment
Chesapeake Bay & Special Projects Program
Baltimore, MD
Ken Bartal
Bureau of Water Quality Management
PA Department of Environmental Resources
Harrisburg, PA
Dr. Kurt Hess
NOAA/National Ocean Service
Rockville, MD
Dr. Arthur Butt
Chesapeake Bay Office
VA Department of Environmental Quality
Richmond, VA
Jerry Hollowell
Susquehanna River Basin Commission
Harrisburg, PA
Dr. Robert Byrne
VA Institute of Marine Science
Gloucester Point, VA
Dr. Steve Jordan
MD Department of Natural Resources
Annapolis, MD
Dr. Elizabeth A. Gasman
Interstate Commission on the Potomac River
Basin
Water Management Administration
MD Department of the Environment
Baltimore, MD
Veronica Kasi
Bureau of Land & Water Conservation
PA Department of Environmental Resources
Harrisburg, PA
James Cox
Division of Soil & Water Conservation
Richmond, VA
Lewis C. Linker, Coordinator of the
Subcommittee
US EPA Chesapeake Bay Program Office
Annapolis, MD
Steve Funderburk
US Fish & Wildlife Service
Chesapeake Bay Estuary Program
Annapolis, MD
Larry Lower
US Army Corps of Engineers
Baltimore, MD

-------
                       THE MODELING SUBCOMMITTEE
                                   OF THE
                         CHESAPEAKE BAY PROGRAM
Alan Pollock
Chesapeake Bay Office
V A Department of Environmental Quality
Richmond, VA
Dr. Gordon Smith
STAC Liaison
Environmental Programs
Applied Physics Laboratory
The Johns Hopkins University
Laurel, MD
Dr. David Schultz
Office of Research
US Geological Survey
Reston, VA
Paul Welle
USDA Soil Conservation Service
Northeast National Technical Center
Chester, PA
Dr. Sybil Seitzinger
Academy of Natural Sciences
Philadelphia, PA

-------
ACKNOWLEDGEMENTS
      The Modeling Subcommittee of the Chesapeake Bay Program would like to acknowledge
the specific contributions of the following individuals in the preparation of this report:

Narendra Panday, Maryland Department of the Environment, Chesapeake Bay and Watershed
      Management Administration, Baltimore, MD,
Veronica Kasi,  Bureau  of Land and  Water Conservation,  Pennsylvania  Department of
      Environmental Resources, Harrisburg, PA,
Robert Byrne, Virginia Institute of Marine Science, Gloucester Point, VA,
Albert Kuo, Virginia Institute of Marine Science, Gloucester Point, VA, and
Carl  Cerco, US Army Corps of Engineers, Waterways Experiment Station, Vicksburg, MS.

The Modeling Subcommittee also  acknowledges the essential guidance and direction provided
by the Model Evaluation Group:

Donald Harleman, Massachusetts Institute of Technology, Cambridge, MA,
Wu-Seng Lung, University of Virginia, Charlottesville, VA, and
Jay L. Taft, Harvard University, Cambridge, MA.

      Special  thanks are also due to those individuals listed below who provided important
guidance for the selection of scenarios and for the subsequent interpretation of results:

Louis W. Bercheni, Bureau of Land and Water Conservation, Pennsylvania Department of
      Environmental Resources, Harrisburg, PA,
Michael Haire, Maryland Department of the Environment, Chesapeake Bay and Special Projects
      Program,
Alan  Pollock,  Virginia  Department of Environmental  Quality,  Chesapeake Bay Office,
      Richmond, VA, and
Ken   Bartal,  Bureau  of  Land  and Water  Conservation,  Pennsylvania   Department of
      Environmental Resources, Harrisburg, PA.

-------
TABLE OF CONTENTS
SECTION
PREFACE
Members of Modeling Subcommittee, Chesapeake Bay Program
Ackno wl edgements
TECHNICAL SUMMARY
I - INTRODUCTION
A. Background
B. Water Quality Goals and Requirements for Interpretation of
Scenarios
C. Rationale For Scenario Choices
D. Summary Listing of Scenarios
II - MODEL LOADS AND SCENARIOS
A. Loading Considerations
1. Introduction
2. Reference loadings
B. Hydrological Sequence
C. Description of Scenarios
D. Scenario Nitrogen Loads
E. Scenario Phosphorus Loads
F. TN/TP Ratio for Input Loads
G. Scenario Carbon Loads
H. Scenario Solids Loads
I. Conclusions
III - CBWQM RESPONSE TO SCENARIO LOADS -
GENERAL CONSIDERATIONS
A. Temporal and Spatial Averaging
B. Tributary Interfaces
C. Fundamentals of Response Surface Analysis
PAGE



S-l

1-1
1-3
1-4
1-5

II- 1
II -3
II -5
II -5
II- 10
11-13
II- 19
II- 19
11-24

III-l
III -5
III -6

-------
IV
- BAY HYDRODYNAMIC TRANSPORT USED IN
SCENARIOS
A. Introduction
B. Main Bay Flows
1. Estimate of hydraulic residence times of bottom and
surface waters
C. Main Bay Salinity - Zonal and Seasonal Averages
D. Tributary Inflows, Outflows, Net Flows
V-
CBWQM RESPONSE TO UNIT LOADS OF
"TRACER" VARIABLES
A. Introduction
B. Response to Conservative Dissolved Tracer
C. Response to Conservative Particle Tracers
VI
- NUTRIENT, PHYTOPLANKTON, CARBON AND
SOD RESPONSE
A. Introduction
B. Nitrogen and phosphorus responses
1 . Nitrogen and Phosphorus Concentrations
2. The DIN/DIP ratio
3. Nitrogen and phosphorus fluxes
4. Nitrogen and phosphorus responses - SAV goals
C. Phytoplankton Response
1 . Base case and comparison to LOT scenarios
2. Biomass reduction comparisons of "40% controllable"
scenarios to LOT N&P
3. Effect on light penetration
4. Primary production response
D. Carbon Response
1 . TOC and net carbon settling to sediment
2. Bay-wide carbon fluxes
E. Sediment Oxygen Demand Response
F. Tidal Tributary Loading to Bay
1 . Potomac and James estuaries
2. All major tidal tributaries
G. Discussion and Conclusions
VII
- DISSOLVED OXYGEN RESPONSE
A. Introduction
B. Seasonal Average Base Case Dissolved Oxygen
1. Analysis of Base case bottom DO

IV -1
IV -1
IV -2
IV -5
IV -5

V-l
V-l
V-3

VI -1
VI -2
VI -2
VI -6
VI -6
VI -10
VI -14
VI- 14
VI -17
VI -17
VI -19
VI -22
VI -22
VI -28
VI -28
VI -31
VI -31
VI -43
VI -43

VII- 1
VII -1
VII- 1

-------
C. Seasonal Average DO Response for LOT and 40% Controllable
Scenarios
D. Anoxic Volume Days Response
E. Anoxic Volume Days Response Surface Analysis
1 . Whole Bay response
2. Response by zone and season
F. Conclusions
VIII - REFERENCES
APPENDIX A
Scenario Descriptions
Summary of Nitrogen and Phosphorus Loadings
Steady State Response Matrices for Tracer Runs
Anoxic Volume Days - Seasonal and Annual Totals
Anoxic Volume Days - Zone Annual Totals
APPENDIX B
Regression Equations from Response Surface Analysis
VII
VII
VII-
VII-
VII-
VII-
VIII
-3
-5
11
11
11
16
-1

A-
A-
A-
A-
A-
1
4
9
10
11

B-
1

-------
PREFACE

In 1983, the Chesapeake Bay Program identified excess nutrients, or eutrophication, as the
primary reason  for water quality decline in the  Chesapeake1.  To quantify the  nutrients
contributing to eutrophication and the nutrient reductions necessary to restore Chesapeake Bay
resources, several water quality models have been developed and applied.

The first model application was the Watershed Model.  The first phase of that model, completed
in 1982, documented the magnitude and source of the point and nonpoint nutrient loads to the
Bay for wet, dry, and average hydrology years1.

To examine the impact of nutrient loads on the mainstem Bay, a steady state water quality model
of the Bay was completed  in  1987.   Using simplified loading estimates and simulation
procedures, this model calculated the average or  steady state summer (June - September)
conditions in the mainstem Bay  .  Results from the steady-state model indicated that a 40%
reduction in nutrient loads would eliminate anoxia (dissolved oxygen concentrations less than 1.0
mg/1) in the mainstem.

To confirm the estimates of anoxia reduction, and to refine estimates of the improvements in Bay
water quality in response to nutrient load reductions, work began on an integrated set of
Chesapeake Bay models in 1987  (shown below).  The linked watershed, hydrodynamic, water
quality, and sediment models were completed in 1992.  This report documents the findings from
the application of these integrated models to evaluating the  technical aspects of various  load
reduction scenarios.
             Integrated Chesapeake Bay  Models
                Watershed
                    Model
                                           Water Quality
                                               Mode!
                                        Sedfeseot
                                          Model
                  Ocear*
                  otjrwfw
                 : Model
Chesapeake Bay Water
Oualtty Model -
                                        Hydrodynamic Model
                                     Linked Water Quality/
                                     Hydrodynamic Model
                                     -i-

-------
The following model refinements were made in the development of the integrated models.

WATERSHED MODEL
The Watershed Model was updated to provide greater detail of atmospheric and agricultural
sources2. This model was used to 1) determine the distribution of the point and nonpoint source
loads and the controllable and uncontrollable portions of the loads; 2) determine the quantity of
loads reduced under different management actions; 3) determine the nutrient loads to the Bay
under different Clean Air Act scenarios; and 4) quantify  the loads under future (year 2000
conditions). These loads were used as input conditions for the Chesapeake Bay Water Quality
Model.

CHESAPEAKE BAY WATER QUALITY MODEL (CBWQM)
The CBWQM is a time variable, three dimensional water quality model coupled  with a model
of sediment processes.  The CBWQM is driven by a hydrodynamic model simulating the hourly
movement of Bay waters over the three  year (1984 - 1986) simulation period.  The sediment
model provides simulation of sediment nutrient sources and sinks.  An ocean boundary submodel
simulates the expected coastal input of loads under different nutrient management conditions.
The details of model development, structure, calibration, and sensitivity are given in separate
reports3'4'5.

This report is divided into seven sections.  Section I briefly describes the Chesapeake Bay Water
Quality  Model and introduces the scenarios analyzed.  Section II describes  the scenarios and
associated nutrient and sediment loads in detail.  The third section provides background for the
more detailed analysis of Bay water quality response that follows.  Sections IV and V summarize
aspects of Bay circulation and its influence on water quality.  Section  VI surveys the nutrient,
phytoplankton, and sediment response to loading reductions.   Section  VI also  examines the
overall nutrient budget and the exchange among  the tributaries  and  mainstem.   Section VII
concludes the report with a detailed examination of the mainstem dissolved oxygen response to
the interactions  among nutrient  loads,  transport,  sediment  processes,  and phytoplankton
processes examined in the previous sections.  A summary of major findings is included in each
section and an overview of these findings is provided in the Technical Summary.

  The report is designed to be a resource for a diverse technical and managerial audience.  With
this eclectic readership in mind, the following suggestions are made.

       1) The harried reader may  want to review the Technical Summary only.
      2) A reader with more time may include the major conclusions  on pages 11-10, VI-11
      through VI-12, and VII-6,  with  reference to areas of interest within  the appropriate
      sections.
      3) The determined reader who persists in  the completion of the entire report will be
      rewarded with a more complete analysis of Bay water quality response to nutrient loading
      reductions,  and will have a greater understanding of how  Bay water quality is expected
      to respond under different loading conditions.
                                        -11-

-------
The work described in this report is by no means complete.  Scenarios have been applied to
develop the tributary loading  allocations of  a 40%  reduction of controllable nitrogen  and
phosphorus. Other scenarios annually track the loads to compare annual reductions with the year
2000 goal.

Further model refinements are now under way.  These refinements will examine the relationship
among  air deposition, water quality,  and key living resource areas including SAV, benthos, and
phytoplankton/zooplankton.    The  refined  model  analysis  of air  deposition,  and  water
quality/living resource interaction will be completed in 1997.
James R. Collier                                           Lewis C. Linker
Modeling Subcommittee Chairman                          Modeling Subcommittee Coordinator
ENDNOTES

1.  M.E. Gillelan et al., 1983.  Chesapeake Bay: A Framework for Action  U.S. EPA Chesapeake Bay Program.
       Annapolis,  MD.

2.  A.S. Donigian et al., 1991. Watershed Model Application to calculate Bay Nutrient Loads: Phase II Findings and
       Recommendations  U.S. EPA Chesapeake Bay Program. Annapolis, MD

3.  C.F. Cerco and  T. Cole, 1993.  Application of the Three-Dimensional Eutrophication Model CE-QUAL-ICM to
       Chesapeake Bay U.S. Corps of Engineers Waterways Experiment Station. Vicksburg, MS.

4.  D.M.  Di Toto and J.J. Fitzpatrick, 1993.   Chesapeake Bay Sediment Flux  Model  U.S. Corps of Engineers
       Waterways Experiment Station.  Vicksburg, MS

5.  B.H. Johnson et al., 1991.  Users Guide  for a Three-Dimensional Numerical Hvdrodvnamic.  Salinity,  and
       Temperature Model of Chesapeake Bay  U.S. Corps of Engineers Waterways Experiment Station.  Vicksburg,
       MS.
                                            -Ill-

-------
                             TECHNICAL SUMMARY
INTRODUCTION

       A modeling framework was constructed for the Chesapeake Bay system  to provide a
credible basis to assist the decision-making process and to further the understanding of Bay water
quality processes and the sensitivity of such processes to external nutrient loading. The modeling
structure consists of a Watershed Model to generate nutrient loads from the Bay sub-basins; a
three-dimensional, time variable hydrodynamic model; and a three-dimensional, time variable
model of water quality coupled to a model of sediment chemistry.
       Extensive calibration analyses of the entire modeling structure was conducted using data
collected primarily during a three year period from 1984-1986. The Chesapeake Bay Program
Modeling Subcommittee completed its initial review of the Chesapeake Bay Water Quality Model
(CBWQM) calibration in May 1991 and concluded the model could provide useful information to
the Bay community, especially with respect to dissolved oxygen problems in the deep water of the
main Bay. Final calibration of the CBWQM was completed in January 1992.
       After completion of the model calibration, the models were used to address management
issues such  as: (1) What impact do reductions or increases in nutrient loads from point and
nonpoint sources delivered by the Bay's major tributaries have on Chesapeake's water quality? and
(2) How much of the nutrient loads to the Bay is natural and how much is related to man-made
sources, and to what extent can loads be controlled?
       The CBWQM provides projections of the expected water quality responses (including
dissolved oxygen concentrations) in the main Bay under a variety of proposed management
scenarios. The purposes of this report are therefore to (a) document the results of a full set of
loading scenario computations, (b) analyze the scenario results across different loading conditions
and (c) interpret the results in the light of the sensitivity of the Bay  model to various loading
conditions.
       A total of 21 scenarios were run for this work and a summary description of the 11
scenarios that formed the basis for the analysis herein is given in the Table below.
Scenario
Number
1
2
3
4
5
6
7
8
10
11
16
Scenario Tag
BASE
40% CONT
40% +CAA
40%CAA+BASIN
LOT
LOT -UPPER
LOT -MID
LOT - LOWER
LOT - N ONLY
LOT - P ONLY
90% RED
Scenario Description
1984-1986 Conditions
40% Reduction of controllable load ("Agreement" states only)
Scenario #2 + Clean Air Act atmospheric reductions
40% + CAA for entire basin
Limit of technology (LOT) for nutrient reductions
Limit of technology for "Upper Bay" only, others at BASE
Limit of technology for "Middle Bay" only, others at BASE
Limit of technology for "Lower Bay" only, others at BASE
LOT for nitrogen only, phosphorus as incidental by N reduction
LOT for phosphorus only, nitrogen as incidental by P reduction
90% reduction in total N & P loading from BASE
                                          S-l

-------
 SCENARIO  NUTRIENT LOADS

       This report is focused on the response of the main Bay to various nutrient loading
scenarios. The "external" loads are comprised of (1) fall line watershed loading, (2) below fall line
watershed load, (3) point source loads, (4) atmospheric loads direct to the water surface of the
Bay, and (5) ocean loading. The "internal" loadings from tidal tributaries to the Bay are calculated
by the CBWQM and are given at the interface of the tributary with the main Bay. Fall line
loadings and below fall line loadings are calculated from the Watershed Model (WSM). Point
source loads were obtained from inventories. Atmospheric loads were estimated based on
available data, and ocean boundary loads were estimated using shelf nutrient data and a simple
exchange model at the mouth of the Bay.
       The three principal reference loadings that established the extent of feasible reductions,
are:
              1. Base Case Load,
             2. Controllable Load, and
             3. Limit of Technology (LOT) Load.
             a. Base Case. The Base Case loading represents the 1984 -1986 loading as a
reference time period. This is the same period as used for the calibration of the CBWQM.
             b. Controllable Load.  Controllable loads are defined as the difference between
Base Case loads and a WSM run using an  all forested basin (excluding NY, WV & DE) with no
point sources. 40% Controllable loads are those agreed to in the Bay agreements.
             c. LOT Load. Limit of Technology loads for Non-Point Source (NPS) inputs
were determined by evaluation of Best Management Practices (BMP) and implemented in the
WSM. Point source (PS) LOT loads were assumed as follows: 3.0 mg/L for TN, 0.075 mg/L for
TP, and 1.0 mg/L for BOD.
      In order to represent year to year variation in hydrology, the scenario runs were conducted
over a ten year period with flows representative of the interval from 1979 to 1988. Wet, dry and
average years during this period were assigned the 1984, 1985 and 1986 hydrologies,
respectively. The total river flows to the Bay for 1984 to 1986 are 487,300 cfs (13,800 m3/s),
459,100 cfs (13,000 m3/s), and 476,700 cfs (13,500 m3/s), respectively. The CBWQM was run for
these ten years in sequence and the final five years were output. In this report, the emphasis is on
year #9, the average hydrologic  year.

TOTAL NITROGEN LOADS

      The Figure below shows the total annual nitrogen (TN) loads used for three of the
principal scenarios: Base Case, 40% Controllable and the Limit of Technology (LOT).  Base and
LOT runs provide an approximate bounding of the feasible range of load reductions. As shown in
this Figure, the major source of nitrogen is from the nonpoint sources making up more than 45%
of the total for all three scenarios. The point sources decline from 19% of total at Base to 5% at
LOT illustrating the greater possible reduction of point source loading relative to nonpoint source
reduction. Atmospheric loading directly to the tidal waters  is about 9% of total. Ocean loading of
TN is a significant component of the total load to the Bay and is estimated to range from 29% to
36% of the TN loading for Base to LOT.

                                         S-2

-------
                   TOTAL NITROGEN LOADS  BY TYPE AND SCENARIO

              t»  250
              2   20°
              O   150

              2   100
              J
              g    50

              H     0
                          (45)
Note: ( ) = % of Total Load

     (50)


              (36)

NFS
PS
ATMO S .

W:
^
OCEAN[^
TOTAL
# 1-Base
207.5
86.8
34.6
131.8
460.7
#2-40% Cont
180.4
53.4
34.6
131.8
400.2
#5 LOT
183.0
17.0
34.6
131.8
366.4
       The  Figure below indicates the percent reduction of external TN loading (excluding
ocean input) from the Base Case for selected scenarios. The increase in % reduction across the
variations of the 40% Controllable (#2) through scenario #4 can be noted. Clean Air Act controls
(CAA) on a basin-wide scale and including the non-basin states increase the percent reduction
from 18% to 25%. It can also be seen that scenario #4 is approaching the LOT scenario #5.  The
LOT-Mid (#7) run removes almost as much TN as the 40% Controllable case.
                REDUCTIONS IN EXTERNAL TOTAL NITROGEN LOAD FOR SELECTED SCENARIOS
                                   EXCLUDING OCEAN LOAD
                    100
                                           SCENARIO
       The relative reduction of TN loading due to point and nonpoint sources indicates that the
point source loading is reduced considerably more than the nonpoint loading in the LOT scenario.
In LOT, the point source loading is reduced about 85% from Base case while the nonpoint
source loading is reduced 14-23%  from Base. This is a reflection of the relative technological
difficulty in reducing nonpoint TN loading as opposed to point TN inputs. It should be noted that
location of such reductions is also important.  The Upper Bay region is dominated by the nonpoint
input of the Susquehanna River so that the contribution from point sources in that region is small
while the middle Bay region is responsible for about 50% of the load reductions (Scenario #5 vs.
#7).
                                          S-3

-------
                                        TOTAL PHOSPHORUS LOADS BY TYPE AND SCENARIO
1 tfl-B.se
NPS
PS
ATMOS.
OCEAN
U 11.1
HI 8.8
il 1.5
d 18.9
TOTAL | 40.3
#2-40% Coat
7.5
5.4
1.5
18.9
33.3
#5 LOT
7.5
0.6
1.5
18.9
28.5
TOTAL PHOSPHORUS  LOADS

       The accompanying
Figure shows an important
difference between TP and TN
sources.  As shown, the ocean
loading of TP dominates the
loading inputs accounting for
as much as 66% of the TP
load at the LOT scenario. It
can also be noted that the
nonpoint and point source
loading are closer in
magnitude than for TN, but
for LOT the nonpoint load is
about the same as for the 40%
Controllable case. This is a
reflection of the fact that most
of the TP is considered controllable.
       The percent reduction of TP from Base for a series of scenarios is given in the Figure to
                                                     the left. The reductions are higher
                                                     than for TN reflecting the greater
                                                     technological control for TP over
                                                     TN. For LOT, a 56% reduction is
                                                     calculated although it should be
                                                     recognized that if the ocean load
                                                     were included, the net percent
                                                     reduction of TP for LOT due to all
                                                     loads drops to 29%.

                                                     The following principal conclusions
                                                     are drawn from this analysis of the
                                                     TN and TP loads.
                                                             1. The upper limit of overall
                                                     total nitrogen reduction from the base
                                                     case is from about 20 - 30% of the
                                                     total input load (excluding input
from the ocean), with PS being more controllable than NPS.
       2. The upper limit of overall total phosphorus reduction from the Base case is from about
30-55% of the total input load (excluding input from the ocean).
       3. The calculated ocean nutrient input load (which is independent of scenario) is estimated
to contribute about 30-35% of the total input nitrogen load and about 45-65% of the total input
phosphorus load.
      too
          REDUCTIONS IN EXTERNAL TOTAL PHOSPHORUS LOAD
                     FOR SELECTED SCENARIOS
        «-<•% CONT.  «4-«%*OB   »*-LOT-Up   «l-LOT-L«w   Kll-LOT-P
             I3-4«V.*CAA   US-LOT    «7-LOT-Mid   ult-LOT-N   «!<-»«% RED
                                          S-4

-------
       4. Deposition of atmospheric nitrogen directly to the Bay waters is about 10% of the Base
case loading.
       5. 40% reduction of controllable nitrogen for nonpoint sources is approximately equal to
the Limit of Technology reduction of nitrogen nonpoint sources.
       6. The application of LOT results in significantly larger percentage reductions in point
source nutrient loadings to the Bay than nonpoint source loadings.
       7. The loading to the Upper Bay is primarily  from the Susquehanna River while the
remainder of the Bay accounts for more than 50% of the load reductions.

CBWQM RESPONSE  - GENERAL CONSIDERATIONS

Since model output for all state variables, time and locations is voluminous, some averaging of
model results over time and space is necessary. Although the CBWQM calculates state variable
concentrations at a time scale  of  hours, such calculations are for computational stability only.
Input information is provided  on a week to week basis and the kinetics that are incorporated in
the model are representative of longer time behavior. The model is considered to represent
processes on a time scale of months, seasons and longer. Therefore, some of the model output
results were averaged over months while other results were averaged over seasons according to
the Table below.
          TEMPORAL PERIODS USED IN AVERAGING MODEL OUTPUT
SEASON
I
II
III
IV
DESCRIPTION
"Winter"
"Spring"
"Summer"
"Fall"
JULIAN DAY
0 -60
61- 150
151 - 270
271 - 365
APPROX. MONTHS
Jan.- Feb.
Mar. - May
June - Sept.
Oct. - Dec.
        NORTH
                                           SCALE
                                          •OSES
                                                50km
The Bay spatial grid scale
horizontally is about 10 km by 5
km by 1.7m and includes from
two to fifteen cells in the vertical
direction for a total of 4029
cells. Two sediment segments
(aerobic and anaerobic) are
incorporated under the water
column segments. Again, in
order to provide tractable
output, water column model
results are averaged spatially
according to the zones indicated
in the Figure below.
                                         S-5

-------
NUTRIENT, PHYTOPLANKTON, CARBON & SOD RESPONSE
       Rain of DIN/DIP
       An important consideration in the behavior of nutrients in the Bay is the degree to which
nitrogen and/or phosphorus limits phytoplankton growth. One measure of which nutrient is
important is the ratio of Dissolved Inorganic Nitrogen (DIN) to Dissolved Inorganic Phosphorus
(DIP). Ratios significantly greater than about 7-10 on a mass basis indicate a tendency toward
                                                phosphorus limitation while ratios
                                                significantly less than that range tend to
                                                indicate nitrogen limitation.
                                                       As shown in the accompanying
                                                figure, for the base case averaged over zone
                                                over the spring, the Bay is calculated to be
                                                phosphorus limited from the head to about  75
                                                km from the mouth. During the summer,
                                                more of the lower Bay becomes nitrogen
                                                limited and during the fall average conditions,
              300      200       100
              Distance (km) Along Mtln Bay (0 = Moalh of Bay)
         Longitudinal variation of surface DIN/DIP ratio for BASE case.
         Averaee over zone over season for average vear.
                                                more than half of the Bay is nitrogen limited.
For the load reduction scenarios, the general tendency is for the LOT N&P and LOT N Only
scenarios to increase the region of the Bay that is nitrogen limited. The LOT P Only however,
decreases the region of nitrogen limitation because of an apparent increased nitrogen transport to
the lower Bay.

              Nitrogen and Phosphorus Fluxes  The Figures below shows the calculated annual
nitrogen and phosphorus fluxes for the main Bay 40% Controllable scenario. Total external load
is the load from fall lines, below fall lines, point sources and direct atmospheric deposition to the
Bay. Net ocean flux is the net exchange at the mouth of the Bay. The settling flux is the gross
settling to the sediment of the Bay. Diffusive flux is the net exchange of dissolved nutrient forms
across  the sediment-water interface. Net sediment flux is the difference between gross settling and
diffusive flux. Denitrification flux is the loss  of nitrogen due to the conversion to nitrogen gas,
primarily in the sediments. Finally, the burial flux is the net loss of nitrogen or phosphorus from
the bottom sediment segment of the model. All fluxes are given in areal units.  It should also be
noted that these fluxes are for the average year (year #9) of the variable hydrology sequence and
              TOTAL NITROGEN FLUX  SCENARIO #2 .40%
          SedtngFlux  DKtimtFlux |]NelS«din»mlHiE(
          54.0 (14%j | 28.8(18%)  24.2(8%) |
                                                          TOTAL PHOSPHORUS FLUX
                                                               MAIN BAY
                                           S-6

-------
               SEASON: MARCH-MAY
as such reflect a flux "snapshot". It can readily be observed that all fluxes do not necessarily add
up to zero because of the dynamic non-steady state nature of the computation.
       The calculated export of nitrogen and import of phosphorus can be noted. For this
scenario, a significant reduction in nitrogen exiting the Bay is estimated while for phosphorus, the
influx of phosphorus from the ocean is calculated to increase over the Base case condition. The
burial  loss of phosphorus is significant and is about equal to the total external load to the Bay.

       Phytoplankton Response Several significant insights were obtained by examining the
model response of phytoplankton biomass and primary production as a function of nutrient
reduction. For the 40% Controllable scenario, phytoplankton biomass is reduced about 10%  in
the spring (with a minimum percent reduction from Base of zero at about 125 km from the
mouth) and about  15% during the summer. As shown in the Figure below, the response of
phytoplankton biomass to LOT, LOT-N Only and LOT-P Only reflected the nitrogen -
phosphorus limitation regions discussed above. For the spring (similar results occur for the
                                                summer), the biomass in the upper  100 km is
                                                reduced almost entirely by reductions in
                                                phosphorus while the biomass in the lower
                                                100 km is reduced by controlling nitrogen.
                                                The increase in percent reduction for LOT- N
                                                Only    over LOT for both N & P in the
                                                lower 100 km is interpreted to result from a
                                                down-Bay transport of nitrogen when
                                                phytoplankton are reduced in the upper Bay
                                               because of control of phosphorus. Because
                                                of biomass reduction in the upper Bay by
                                               phosphorus control, light penetration is
                                                estimated to increase in that region, while for
                                               the lower Bay, light penetration will increase
because of nitrogen removal.
       The response in terms of primary production across scenarios is summarized in the next
Figure. For the 40 % Controllable scenarios (#2 - #4),  it is seen that as the load is increasingly
                                                     reduced for these scenarios, the impact
                                                     on the primary production approaches
                                                     the LOT case. The reductions in the mid
                                                     Bay areas (LOT-Mid) is also shown to
                                                     be  a significant part of the overall
                                                     reduction. The annual averages,
                                                     however tend to mask the actual
                                                     seasonal dynamics of primary production
                                                     and the spatial variability of reductions
                                                     in primary production as a function of
                                                     which nutrient is reduced. Thus, the
                                                     LOT P Only annual average reduction is
                                                     only 10% whereas the spring reduction
                                                     is over 40%. The responses for two
       3««   ISO  200  ISO   100  SO   0
       Distance (km) Along Main Bay (0 = Mouth of Bay)
   Longitudinal variation of percent reduction of surface chlorophyll
  on g/mA2 basis. Average over zone over season for average year.
%REDUCTION FROM BASE CASE
  •2-40% COKT.  *4-«%tC+B   *frLOT-Uo   »«-LOT-L«ir   *ll-LOT-F
       «3-4»%tCAA    »5-LOT    «7-LOT-Mld   H10-LOT-N   »1«-»0%RED
                     SCENARIO
   Reductions In Primary Production for Selected Scenarios. Average Year
                                         S-7

-------
zones over season and for the three LOT scenarios helps explain these results and are shown in
the following two Figures. For Zone 2, the reduction is controlled entirely by phosphorus in the
winter and spring whereas in the summer, the production is controlled equally by nitrogen and
     ZONE *2 X REDUCTION IN PRIMARY PROD.
     60,	
     40

     20

      0

     -20

     -40
                 Q LOT SLOT MB LOT P
         JAN-FEB
                        JUN-8EPT
                                OCT-NOV
     Variation of % reduction from Base of phytoplankton primary
     production. Zone #2. Average over zone over season for average
     year.
                                               ZONE *6 % REDUCTION IN PRIMARY PROD.
                                                60
                                                       JAN-FEB
                                                               MAR-MAY
                                                                      J UN-SEPT
                                                                              OCT-NOV
                                                Variation of % reduction from Base of phytoplankton primary
                                                production. Zone #6. Average over zone over Mason for average
                                                year.
phosphorus. In the fall, nitrogen is more controlling than phosphorus. For the mid-Bay Zone 4
(not shown), phosphorus controls in the winter and spring whereas nitrogen is the controlling
nutrient for the other two seasons. For the summer in Zone 4, LOT P Only results in virtually no
change in production over Base case. For Zone 6, the impact of downstream transport of nitrogen
to the nitrogen poor regions of the Bay is immediately apparent. For the winter and spring
seasons, LOT P Only results in an increase in production over Base case due to this down Bay
transport of nitrogen. In the summer and fall, this effect is less pronounced because of the
relatively lesser impact of phosphorus reductions in the upper Bay regions during these periods.
       These results from the LOT  scenarios provide further evidence of the calculated down
Bay transport of nitrogen by LOT phosphorus load reduction. Such increases in nitrogen increase
primary production in the lower nitrogen limited regions of the Bay and as will be discussed
shortly, have a proportionally smaller impact on the DO of the bottom waters of the Bay. On the
other hand, phosphorus load reductions have a positive impact in the upper Bay zones where the
system is phosphorus limited.

       Sediment Oxygen Demand Response The demand of the sediment for oxygen is
calculated by the sediment sub-model of the CBWQM. The water column is coupled to the
sediment model through the settling of particulate nutrients. The sediment oxygen demand (SOD)
                                                 is calculated using the net carbon flux to the
                                                 sediment as the primary input loading.
                                                 Maximum loading of carbon to the sediment
                                                 is during the summer months and is the
                                                 highest in the upper Bay zones. Peak  values
                                                 in this region are about 0.8 gC/m2-d.  The
                                                 Figure shows the variation in the SOD for
                                                 the Base case across the zones and for the
                                                 four seasons. Maximum SOD is calculated
                                                 to occur in the summer and in Zones 3 to 6.
                                                 This is in contrast to the carbon flux to the
                                                 sediment which is maximum in the upper
SOD g/m*2-d
 2 i	
1.5
0.5
                             BASE CASE
    30ft    250    200   150   100    50    0
       Distinct (km) Alouj Mala B»y (0 = Moult of Biy)
  Longitudinal variation of sediment oxygen demand for BASE case.
  Average over zone over season for average year.
                                            S-8

-------
during which there is zero SOD thereby lowering the overall average. On the other hand, the
difference may be related to the labile and refractory components of the carbon used in the model
The fall line paniculate loadings are considered to be all refractory while the point sources are
assumed to be 70% labile and paniculate carbon produced from phytoplankton is assumed to be
55% labile. Thus, while the sediment of the upper zones receive more carbon, the nature of the
carbon is largely refractory in contrast to the middle and lower zones where the carbon results
from primary production and is considerably more labile.  Since the calculated  diagenesis  rate in
Zones 3 and 4 is higher than in Zone 2, one concludes that the variable carbon fractions has an
effect on the SOD in Zone 2 and together with the periods of zero DO is contributing to the
lower calculated SOD in that Zone.
       The percent reductions in SOD for the 40% Controllable scenarios (#2 - #4) in
comparison to the LOT N&P  scenario is displayed in the Figure to the right. The upper Bay
reductions in SOD are higher under this
latter scenario due presumably to the higher
degree of phosphorus removal in the LOT
than in the 40% controllable. The
differences in nitrogen loading are not as
great. For the middle and lower regions of
the Bay, the 40% controllable scenarios
approach the LOT N&P loading in reducing
SOD. In fact, the 40% + CAA + Basin
control is at the LOT level of reduction for
zones 4 through the rest of the Bay.
% REDUCTION OF SOD FROM BASE
40 |	
                         SEASON: JUNE -SEPT
    300   250    200   150    100    50    0
       Distance (km) Along Main Bay (0 = Mouth of Bay)

 Longitudinal variation of percent reduction of SOD from Base case.
 Average over zone over season for average year.
Tidal Tributary Interface Nutrient Loading The net input of the tidal tributaries to the main Bay is
of particular interest since such loadings represent actual contributions to the Bay proper. As part
of the Bay model calculations, mass balances were conducted around the principal tributaries and
the exchange of load across the interfaces of the tributaries was calculated for each of the
scenarios.
                                                 The net flux of nitrogen across the
                                                 interfaces of the major tidal tributaries is
                                                 indicated in the Figure. The three largest
                                                 inputs are the Patapsco/Back, Potomac
                                                 and James estuaries. The Patuxent
                                                 contributes a small input, while the
                                                 remaining two lower Bay tributaries
                                                 receive a net input from the Bay. An
                                                 interesting point of these runs is that the
                                                 Rappahannock and York rivers are
                                                 calculated to receive nitrogen from the
                                                 Bay as opposed to these tributaries
                                                 providing nitrogen to the Bay. Indeed,
NET
40
30
20
10
0
-10
Vi
CO
NITROGEN LOAD mtfctay
- r-|
ro
M
Sl
bul
iil
Kj
js
i
Vtomi
*y f
i ••
I ill 1 J
To Tributary
from Bay y



D PAT/BACK
BPATUX
ED POT
a RAP
a YORK
H JAMES
BASE 40%CONT. LOTNW
illation of annual net nitrogen load from tributaries for Base, 40%
ntrollable and LOT N&P scenarios. Average year.
                                           S-9

-------
     % REDUCTION FROM BASE CASE OF NET LOAD FROM TRIB
     100
under several removal programs (including the 40% Com. and LOT N&P that are shown) the
input net nitrogen load increases from the Bay to the tributary. This is undoubtedly a result of a
complex interaction of transport and nutrient concentration where the gradient from the Bay to
these tributaries is increased under various removal programs. A similar behavior is calculated for
the net phosphorus loadings from the tributaries.
       An important  linearity in the net interfacial load of both TN and TP from the Potomac and
James over the range of loadings from the base to the LOT (not including the geographical runs)
is shown in the Figure to the left. As seen for TN, if the total input load of TN to the Potomac or
                                                James is reduced by, say, 30% from the base
                                                case load, then the net load of TN exiting
                                                from the Potomac or James is reduced about
                                                35% from the base case net load. Therefore,
                                                in spite of the rather complex nonlinear
                                                interactions that exist in the overall model
                                                framework, and the apparent interactions
                                                between the Bay and the tributaries, the
                                                relationship of net load from these two
                                                tributaries to the Bay is directly proportional
                                                to the reduction in external load to the
                                               tributary. However, as noted previously,
                                               loads from the James influence only the
             20      40     60     80      100
        •/• REDUCTION FROM BASE CASE OF INPUT LOAD TO TRIB
    Relationship between % reduction of input load to tributary and net
    load exiting tributary to Bay for TN.
lower Bay and mouth region, while load reductions from the middle Bay including the Potomac
provide significant improvements in the main Bay water quality.

       The ability to examine the behavior of the Bay with the calibrated CBWQM under
different removal levels of nutrients in combination is a particularly important use of the model.
Such behavior is not directly observable in the Bay and can only be predicted by a credible model.
The degree to which phosphorus and nitrogen load reductions have an impact on the water quality
of the Bay is of course an important consideration in the decision making process.
       In general, the Bay can be divided into three broad regions: the upper approximately 100
km of the Bay where control of phytoplankton growth is by phosphorus, the approximately 100
km of the lower Bay where the phytoplankton production is controlled by nitrogen and a middle
Bay region of about 100 km where a transition takes place. The extent of nitrogen control
proceeds up the estuary during the summer and fall and is a function of fresh water hydrology and
resulting circulation. This general conclusion is consistent with interpretations of observations
made on the Bay by a variety of investigators. Modeling shows  that as phosphorus loadings to
the Bay are reduced (with nitrogen loadings remaining at approximately Base levels), excess
nitrogen is transported down the Bay in the surface waters. This transported nitrogen then
stimulates phytoplankton production in this nitrogen limited region of the Bay. This "additional"
relatively labile biomass then settles in the downstream region and contributes to higher SOD in
that area. Phosphorus removal however has a distinctly positive effect in the surface waters of the
upper Bay where spring and summer phytoplankton biomass are reduced considerably more than
if only nitrogen were removed. Such reductions of biomass of 20-30% have an impact on light
penetration, with a 20% increase in light calculated for the 2 m depth at LOT levels.
                                          S-10

-------
       Reductions in nitrogen have of course a direct effect on phytoplankton production in the
nitrogen limited areas and subsequently on the carbon fluxes and the SOD. In addition, the
nitrogen load reductions result in improvement in meeting the DIN habitat requirements for the
SAV.
       It is concluded from the analyses reviewed here, that load reductions of both phosphorus
and nitrogen are necessary to result in reductions in the nutrients, phytoplankton biomass, (with
increases in light penetration) and sediment oxygen demand. Phosphorus load reductions are most
effective in achieving improvement in these measures of water quality in the upper Bay. Nitrogen
removal is required throughout the Bay: in the upper Bay to reduce nitrogen loads that would be
transported down Bay under the phosphorus reduction and in the middle and lower Bay to
directly reduce biomass and hence SOD.
       The Susquehanna River, a non-tidal tributary to the Bay accounts for a majority of the
nutrient input on an average annual basis (about 42% of the TN and 31% of the TP loads). The
net input of nutrients on an annual average basis from  the principal tidal tributaries to the Bay is
exclusively from the Patapsco/Back, Potomac and James estuaries. The Rappahannock and York
estuaries are calculated to receive a  net input of nitrogen and phosphorus from the Bay. For the
Potomac and James estuaries, the net nutrient load exiting the tributary to the Bay is
approximately linear to the external  load of nutrient  to the tributary. Nutrient loads from the
Potomac enter the middle Bay region where water quality impacts persist while nutrient impacts
from the James are limited to the lower Bay and Bay mouth region.
DISSOLVED OXYGEN RESPONSE
                                       DISSOLVED OXYGEN i
                                       12
ig/L
                                                                       BASE CASE
                                                                       SUMMER
                                       10
       The calculated response of the dissolved oxygen (DO) of the Bay assumes particular
importance in analyzing the effects of scenario nutrient load reductions. The dissolved oxygen
focus in this Section is twofold: (1) evaluation of the seasonal (specifically summer) average DO
response, and (2) analysis of the response of the DO concentrations below 1 mg/L, the working
definition of anoxia. The latter quantity is determined by calculating the volumetric extent and
temporal extent of DO below 1 mg/L. These "anoxic volume-days" have units of m3 - days.
       The summer average longitudinal DO profile for the Base case loading condition and for
the average hydrology flow year is
shown here. The summer profile is the
basis for comparison of assessing the
effect of nutrient reduction scenarios.
The rapid drop of the minimum bottom
DO between the spring level of greater
than 5 mg/L to the minimum summer
average level of 0.1 mg/1 can be noted.
The steep increase in the bottom DO
beyond the upper limit of the deep trench
at approximately 260 km is due to a
rapid decrease in depth. The marked
vertical gradient in DO during the
summer can also be seen where average
                                                     SURFACE
                                           300
                                                 250    200    150    100    50
                                               Distance (km) Along Main Bay (0= Mouth of Bay)
                                                I   2  !   3  I  4     5  I  6 I  7
                                                   BOUNDARIES OF BAY ZONES
                      I a
                                          Calculated longitudinal variation of DO for Base case averaged over
                                          SUMMER for average year.
                                         S-ll

-------
surface DO is generally supersaturated due to algal productivity and the bottom DO is responding
to deep water sinks of oxygen. The marked difference in the longitudinal profiles between the
surface and pycnocline levels and the bottom level can also be noted. Examination of the bottom
DO longitudinal summer profile indicates an approximate linear decrease in DO as one progresses
up the Bay. A simple analysis of the behavior of the DO in the bottom waters can be made to help
understand this behavior.
       The principal sinks of oxygen in the bottom water are the sediment oxygen demand
(SOD), the oxidation of the dissolved organic carbon (DOC) and the immediate uptake of oxygen
to satisfy the chemical oxygen demand (COD) of reduced substances released from the sediment.
Phytoplankton respiration is neglected since during the summer the bottom layer phytoplankton
biomass is small. In the CBWQM, the sediment and water column are interactive and not
separated. However, since the output from the sediment model is computed as equivalent SOD
and COD, an analysis can be made considering these processes as external sinks of DO. The rates
of utilization of oxygen in the model are oxygen dependent, but this complication is not
considered here in this  simple analysis. Also, vertical mixing of oxygen is not included which
simplifies the analysis considerably. The DO
concentration is thus given by a linear
equation
       The accompanying Figure illustrates
this behavior. At the head end of the trench
after the approximately 17 days of total
travel time, the total DO decline is then
about 6.5 mg/L or for an initial bottom DO
at the mouth of the Bay of about 7 mg/L, a
DO of about 0. 5 mg/L is calculated. The
analysis represents a general process of
bottom water moving up the Bay,  losing
oxygen during the time of travel of the
parcel (due principally to a constant
f
B '
g.
i.
           COD
   Showing approximate zero order (linear) decrease in bottom DO,
   Base case, Summer, average year.
withdrawal of oxygen to satisfy the SOD)
and arriving at the head end of the trench at anoxic levels.
       The Figures below show the summer average DO for the LOT N&P and the 40%
Controllable scenarios.  For the LOT N&P, the bottom DO is improved, but not to the point of
raising the DO above anoxia (i.e., DO < 1 mg/L) on a summer average basis. Analysis of the
DO I
0.8
0.6
0.4
o.;
-0.1
-0.4
-0.6
>IFF. FROM BASE mg/L

\A
-JoJV
'IhMiBw
^ 	
A AA.
r1
SURFACE BOTTOM
DOOM*-
~]
i 	 tfW

300 250 200 150 100 50 0
DtoUnce (km) Along Main Bay (0° Mouth of Bay)
Longitudinal variation of DO difference between LOT and Base
cue averaged over summer for average year.
BOT
O.I
0.6
0.4
0.2
0
-0.2
-0.4
-0.6
TOM DO DIFF. FROM BASE m(/L
40% CONTROLLABLE -BASE CASE i
SUMMER f
DOGraakr
ItunBw
r\
\
IDOLwa
lunBw

300 250 200 150 100 50 0
Dktanet (km) Along Main Bay (0* Mouth of Bay)
Longitudinal variation of bottom DO difference between 40%
Controllable scenario and Base case averaged over summer for
iverage year.
                                          S-12

-------
                                           ANOXIC VOLUME DAYS (Billion mA3-dayi)
                                           140
range of load reduction indicates that only when the incoming loads are reduced by at least 50%
of the external load is the summer average bottom DO calculated to be greater than 1.0 mg/L.
The 40% controllable scenario improves the bottom DO by about a constant 0.2 mg/L on a
summer average basis which is about half of the LOT N&P scenario.

       Anoxic Volume Days Response As noted above, a useful measure of the degree of anoxia
is the total volume »days where the DO was calculated to be less than 1 mg/L. That is, anoxia is
tracked on a volume basis over time and a sum is tabulated for each scenario. Maximum anoxia
occurs in the summer with about 16%
occurring in the spring and fall and none in
the winter. The longitudinal variation of the
anoxia as shown to the left indicates peak
regions in zones 2 and 3 where about 70%
of the annual total occurs. An additional
24% occurs in Zones 4 and 9 (Eastern
Shore).
       The percent reductions in total
annual anoxic volume days from the Base
case for selected scenarios is shown below.
Complete elimination occurs at 90% N&P
removal (Scenario #16). The upper limit  of
anoxia reduction provided by the best
technology (LOT) is about 30%. Note that
40% controllable +CAA+Basin control (Scenario #4) approaches the improvement from LOT
N&P (#5).
       The relative impact of nitrogen and phosphorus is seen by comparing #10 and #11 with
                                                300   250    100   ISO    100    SO    0
                                                  Distance (km) Along Main Bay (0 - Mouth of Bay)
                                              I   1  I   2   !  » I  4  I   e  I  • I  T I «l
                                                       BOUNDARIES OF BAY ZONES
                                            Longitudinal variation of anoxic volume days for Base case total
                                            over zone for average year.
               % REDUCTION FROM BASE
               100
                &2-40V. CONT. #4-40V.+C+B  #«-LOT-Up  HW-LOT-Low  #11-LOT-P
                     #3-40%+CAA  #S-LOT   #7-LOT-Mld (K10-LOT-N  #16-90% RED
                                      SCENARIO

                 Percent reductions in anoxic - volume - days from Base case for selected
                 scenarios. Average Year
#5. As noted above, LOT P Only has less of an impact on the bottom DO and as shown in Figure
VII - 12, only a 15% improvement in anoxia is calculated which is half that from LOT N Only.
The relationship between nitrogen and phosphorus loadings and the response in terms of anoxic
volume »days is further described by the use of response surface analysis. Combining scenario
runs in a single plot of Bay wide anoxia reduction versus TN and TP load reductions allows the
                                          S-13

-------
visualization of the change in DO improvement as a function of nutrient reductions. Such a
surface is shown below.
                   RESPONSE SURFACE OF REDUCTION OF ANOXIC
                   VOLUME DAYS - WHOLE BAY, AVERAGE YEAR
                   REDUCTION IN ANOXIA
                    A
                  1.00
                  0.75
                  0.50
                                                                       L 0.5
       As seen from this Figure, reductions in TN improve the DO conditions more than
reductions of TP of the same magnitude. It should also be noted that LOT N Only has a greater
reduction in primary production during the summer months in the mid to lower Bay regions. Since
maximum anoxia occurs during the summer, LOT N Only can be expected to also have a
relatively larger impact on summer anoxia than LOT P Only.
       In addition to this hypothesized effect, two other effects may also contribute to the
reduced effect on anoxic volume-day response to the LOT - P Only scenario. As noted
previously, reduced phosphorus loading reduces primary production in the surface waters of the
upper Bay. Such a reduction has the following two consequences:
       (1) the reduced algal growth at the surface decreases surface water DO which in turn
decreases the vertical concentration gradient thereby reducing the exchange of DO and
replenishment of bottom DO in the upper Bay, and
       (2) the reduced algal growth will not assimilate as  much ammonium with the result that
nitrification will increase in the surface waters, decreasing the DO and again decreasing the
vertical transport of oxygen to bottom waters of the upper Bay.
       It can also be noted that the location of where LOT load reductions are applied is also
significant. Thus, comparing Scenarios #6,#7 and #8 indicates that maximum impact on bottom
anoxia  is from load reductions in the mid-Bay region. That is, as Figure VII - 12 indicates, there is
a negligible percent reduction in anoxic volume days under Scenario #8 (LOT -  Lower Bay) as
compared to 8% for Scenario #6 (LOT - Upper Bay) and  21% for Scenario #1 (LOT - Middle
Bay). The minimum impact on anoxia for LOT in the lower Bay only is apparently a consequence
of (a) no net input of nutrients from the Rappahannock and York estuaries (but  an input from the
                                         S-14

-------
Bay into these tributaries), and (b) transport of nutrient input from the James out through the
mouth of the Bay more than transport of nutrients up the Bay proper.

       Conclusions The results presented in this Section indicate the following:
       1. Bottom DO concentrations under Base case conditions reach minimum summer average
levels of less than 1 mg/L. The approximate linear decline in oxygen with distance as one
proceeds up the Bay in the direction of the bottom flows is a result of the distributed sink of
oxygen occasioned principally by the sediment oxygen demand. As such, the minimum bottom
DO at the head end of the trench reflects the accumulated DO depletion of a bottom water parcel
since it entered the Bay. All SOD along the path of bottom water contributes to the DO
depletion.
       2. Feasible reductions in nutrient loadings of about 20 -30% N & P (i.e., LOT and "40%
controllable" scenarios) result in improvement in bottom DO over Base by about 0.2 - 0.4 mg/L
as a summer average.
       3. Load reductions of about 50% or greater result in minimum summer average DO
concentrations above 1 mg/L.
       4. 90% N & P reductions are calculated after the ten year simulation to result in average
summer DO of greater than 5 mg/L.
       5. A measure of anoxia as given by the volumetric and temporal extent of DO less than 1
mg/L (the anoxic volume days) is a maximum in the summer and in Zones 2-4 under Base case.
The feasible load reduction scenarios result in a range of reduction in anoxic volume days of about
20 - 30% from Base. This reduction in anoxia is directly proportional to the load reduction of
nitrogen of about 20-30%.
       6. Response surface analysis of anoxic volume days on a Bay wide basis indicates a
generally linear response in anoxia reduction as a function of nitrogen with little effect due to
phosphorus reductions. The maximum effect of phosphorus is in Zone 4, a region that contributes
a relatively  smaller fraction to the Bay wide total anoxia.
       7. Even though the upper Bay is phosphorus limited, reductions of phosphorus do not
have as significant an effect on anoxic volume days as do nitrogen reductions. The reasons for this
response are complex . Phosphorus controls primary production in the winter and spring while
nitrogen controls primary production in the summer, the period of maximum anoxia. Also, when
only phosphorus is removed there is a calculated increased nitrogen transport to down Bay
nitrogen limited regions which increased downstream SOD.  This effect is apparently coupled
with reduced primary production in the surface waters of the upper Bay resulting in a reduced
vertical DO gradient and less oxygen transferred to the bottom waters of the upper Bay.
       8. The location of where LOT load reductions are applied is also significant. Thus, the
scenarios where LOT reduction were selectively applied by Bay regions (Upper, Mid and Lower)
indicate that maximum impact on bottom anoxia is from load reductions in the mid-Bay region. A
negligible percent reduction in anoxic volume days is calculated for LOT in the  Lower Bay only
as compared to 8% for LOT for the Upper Bay and 21% for LOT in the Middle Bay. The
minimum impact on anoxia for LOT in the lower Bay only is apparently a consequence of (a) no
annual net input of nutrients from the Rappahannock and York estuaries (but rather an input from
the Bay into these tributaries) and (b) possible transport of nutrient input from the James out
through the mouth of the Bay more than transport of nutrients up the Bay proper.
                                         S-15

-------
 TECHNICAL ANALYSIS OF RESPONSE OF CHESAPEAKE BAY WATER QUALITY
                         MODEL TO LOADING SCENARIOS
I. INTRODUCTION

A. BACKGROUND

       A modeling framework was constructed for the Chesapeake Bay system (Figure 1-1) to
provide a credible basis to assist the decision-making process and to further the understanding of
Bay water quality processes and the sensitivity of such processes to external nutrient loading. The
structure includes a Watershed Model, (WSM), (Donigian et al., 1991) to generate nutrient loads
from the Bay sub-basins, a three-dimensional, time variable hydrodynamic model (Johnson et al.,
1991a, 1991b; Dortch, 1990, Blumberg et al., 1991) and a three-dimensional, time variable model
of water quality (Cerco and Cole, 1992) coupled to a model of sediment processes (Di Toro and
Fitzpatrick, 1993). The integrated  latter two models herein designated as the Chesapeake Bay
Water Quality Model (CBWQM) are driven by the hydrodynamic model and loadings generated
by the WSM.
       Extensive calibration analyses of the entire modeling structure was conducted using data
collected primarily during a three year period from 1984-1986. The Chesapeake Bay Program
Modeling Subcommittee completed its review of the CBWQM calibration in May 1991 and
concluded the model could provide useful information to the Bay community.  However,
additional calibration efforts continued during 1991 to improve overall model performance.
Although no significant changes in the model calibration resulted from this continuing effort, the
final calibration did smooth out some spatial variability allowing for a much better match between
observed and predicted values. Final calibration of the CBWQM was completed in January 1992.
       A full documentation of this calibration effort is given in the aforementioned reports. After
completion of the development and calibration of the water quality model of the Bay, a series of
runs were conducted to test and explore model response to loading outside of the calibration
conditions and to provide input into the decision-making process. The preliminary results from
seven initial runs to test overall model behavior were summarized in a  progress report by the
Nutrient Reevaluation Workgroup of the Chesapeake Bay study (Nutrient Reevaluation
Workgroup, 1992). As stated  in that report, the models are to address the following management
issues:
       What impact do nutrient loads from point and nonpoint sources delivered by the Bay's
major tributaries have on Chesapeake's water quality?
       How do these impacts change with reductions or increases in these sources?
       How are these impacts distributed across the Bay's habitats?
       How much of the nutrient loads to the Bay is natural and how much is related to
man-made sources, and to what extent can loads be controlled?
       How long will it take the water quality in the Bay to improve once nutrient controls are
fully implemented?
                                          1-1

-------
Figure I - 1 The Chesapeake Bay



             1-2

-------
One of the key questions answered during the reevaluation was whether meeting the 40 percent
nutrient reduction goal would translate into a significant improvements to the Bay's anoxic
problems.  The CBWQM provides projections of the expected water quality responses including
dissolved oxygen concentrations in the main Bay under a variety of proposed management actions
(e.g., point and nonpoint source controls). A number of scenarios were run simulating water
quality conditions under varying reductions of point (PS) and nonpoint source (NFS) loads.

      The purposes of this report are therefore to address these questions by (a) documenting
the results of a full set of loading scenario computations, (b) analyzing the scenario results across
different loading conditions and (c) interpreting the results in the light of the sensitivity of the Bay
model to various loading conditions.

B. WATER QUALITY GOALS AND REQUIREMENTS  FOR
   INTERPRETATION OF SCENARIOS

      Several water quality goals and requirements can be utilized as guidelines in interpreting
scenario results  and in comparing results between scenarios. These water quality goals and
requirements  are considered from the point of view of the maintenance, protection and
improvement  of the aquatic ecosystem habitat.  The focus is on the Bay fisheries, and supporting
aquatic life communities, including the benthic community  and the submerged aquatic vegetation
(SAV). Details for the assessment and determination of water quality requirements are given in
Chesapeake Bay Program (1993), Dennison et al (1993), Batuik et al (1992), Jordan et al (1992)
and Funderbuck et al (1991). The principle water quality parameters are: dissolved oxygen (DO),
dissolved inorganic nitrogen (DIN), dissolved inorganic phosphorus (DIP), phytoplankton
chlorophyll a, light attenuation coefficient and total suspended solids.

      For further reference in later sections of this report, the DO goals are summarized in Table
1-1 and the SAV requirements are summarized in Table 1-2.

             TABLE 1-1 SUMMARY OF DISSOLVED OXYGEN GOALS1
Dissolved Oxygen Goal
At least 1.0 mg/L at all times
Between 1-3 mg/L for less than 12 hrs and
interval between 1-3 mg/L longer than 48 hrs.
Monthly mean at least 5 mg/L
At least 5 mg/L at all times
Location & Other Specifications
Throughout Bay & tidal tributaries, including
subpycnocline waters
Throughout Bay & tidal tributaries, including
subpycnocline waters
Throughout above-pycnocline waters of Bay &
tidal tributaries
Throughout above-pycnocline waters of
anadromous fish spawning reaches, rivers and
nursery areas
'See CBP (1993) and Jordan et al (1992) for details.
                                         1-3

-------
       TABLE I - 2 SUMMARY OF SAV HABITAT REQUIREMENTS1
SAV Habitat Requirements for One Meter Restoration
Water Quality Parameter
Light Attenuation Coefficient (m"1)
Total Suspended Solids (mg/L)
Chlorophyll a (|^g/L)
Dissolved Inorganic Nitrogen (mg/L)
Dissolved Inorganic Phosphorus (mg/L)
Value
<2
<1.5
<15
<15
<0.15
- <0.02
<0.01
Other Specifications
For TF2 & OL2 regions, April-October
CLP3
For ME2, April-October CLP and for PO2,
March-Nov. CLP
ForTF,OL & ME, April-October CLP and
for PO, March-Nov. CLP
For TF,OL & ME, April-October CLP and
for PO, March-Nov. CLP
For ME , April-Oct. CLP and for PO,
March-Nov. CLP
For TF & OL, April-Oct. CLP and for PO,
March-Nov. CLP
For ME, April-Oct. CLP
SAV Habitat Requirements for Two Meter Restoration
Light Attenuation Coefficient (nV1)
<0.8
For TF,OL & ME, April-October CLP and
for PO, March-Nov. CLP
1 See Batuik et al (1992) and CBP (1993) for details.
2 TF = Tidal Fresh (<0.5 ppt salinity), OL = Oligohaline (<5-5 ppt), ME = Mesohaline (>5-18
ppt) and PO = Polyhaline (>18 ppt).
3 Critical Life Period
C. RATIONALE FOR SCENARIO CHOICES

       The loading scenarios included four groupings of conditions representing the need to
examine model response over a range of conditions, permit assessment of scenario responses
relative to water quality goals and requirements and provide input for management purposes.

              1. Loads within the "feasible" loading reductions representing controllable loads
and limit of technology loading conditions1,
             2. Load reductions to levels beyond the feasible range to low values to test the
sensitivity and stability of the model,
             3. Loading to "pristine" levels to test model response under pre-European
settlement conditions, and
             4. Load scenarios for allocation and management purposes.
1      See  Section II. Model Loads and Scenarios for specifications of "controllable" and "Limit
of Technology" loading conditions.
                                             1-4

-------
D. SUMMARY LISTING OF SCENARIOS
      Table 1-3 below summarizes the scenarios completed for this report. A more complete
description of the scenarios is given in the Appendix.

                  TABLE 1-3 SUMMARY OF SCENARIOS
Scenario
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Scenario Tag
BASE
40% CONT
40% +CAA
40%CAA+BASIN
LOT
LOT -UPPER
LOT -MID
LOT - LOWER
LOT - MID (A)
LOT - N ONLY
LOT - P ONLY
65% LOT
BNR
ALLOCATION 2
50% N & P
90% RED
90% N ONLY
90% P ONLY
31%N-18%P
10%N - 49%P
SUS. REGR.
Scenario Description
1984-1986 Loading
40% Reduction of controllable load ("Agreement" states only)
Scenario #2 + Clean Air Act atmospheric reductions
40% + CAA for entire basin
Limit of technology
Limit of technology for "Upper Bay" only, others at Base
Limit of technology for "Middle Bay" only, others at Base
Limit of technology for "Lower Bay" only, others at Base
Same as #7 except Potomac River basin loads at Base
LOT for nitrogen only, phosphorus at Base
LOT for phosphorus only, nitrogen at Base
Loads at 65% of LOT
Point sources at seasonal BNR
Variable loading by Bay region
50% reduction in total N & P loading from Base
90% reduction in total N & P loading from Base
90% reduction in total N loading only from Base
90% reduction in total P loading only from Base
3 1% & 18% reduction in total N & P, respectively
10% & 49% reduction in total N & P, respectively
Same as Base case (#1), with regression loads for Sus. R. load
                                     1-5

-------
       The 21 scenario runs cover a wide range of loading conditions and hence water quality
response in the Bay. Using the "Base" run #1 as a reference loading condition, other scenarios
include the feasible range (#2-#l 1), responses beyond feasible load reductions (#15-#20)
including an estimate of pristine loading (approximated by run #16) and several management runs
(#12-#14). Run #21 is a comparison run with run #1 but using regression estimates of loading
from the Susquehanna River rather than loadings as generated by the Watershed Model. Details of
the loading into the model are given in Section II.
       In addition to the scenario runs listed above,  a series of "tracer" calculations were also
made.  These runs included release of conservative and non-conservative constituents in various
portions of the Bay model in order to track the behavior of such variables and to provide
additional information for interpreting the more complex water quality runs.
                                           1-6

-------
II. MODEL  LOADS AND SCENARIOS

A. LOADING CONSIDERATIONS
       1. Introduction
       This report is focused on the response of the main Bay to various loading scenarios. As such,
the loadings from tributaries to the Bay (as well as the exchange with the ocean) are evaluated at the
interface of the tributary with the main Bay. Also, loadings external to the water column and loadings
interacting with the sediment are compiled. Figure II -1  shows a schematic of the external loadings
together with the delineation of the Bay into the three large regions of Upper - Bay, Mid-Bay and
Lower Bay as used in Scenarios #6 through #9 and #14.
       As indicated, the external loads are comprised of (1) fall line loading, (2) below fall line load,
(3) point source loads, (4) atmospheric loads direct to the water surface of the Bay, and (5) ocean
loading.
       Fall line loadings and below fall line loadings are calculated from the Watershed Model (WM).
Point source loads were obtained from inventories. Atmospheric loads were estimated based on
available data and ocean boundary loads were estimated using shelf nutrient data and a simple
exchange model at the mouth of the Bay. Cerco and Cole (1992) describe the procedure in detail.
       Tidal tributary interface loadings are the result of fall line loadings, point source inputs and
below fall line loadings together with the within-tributary processes of transport, kinetic and sediment
interactions. The net tributary loads are discussed in some detail in Section VI - F of this report.

       2. Reference Loadings
       The three principal reference loadings that established the extent of feasible reductions, are:
             1. Base Case Load,
             2. Controllable Load, and
             3. Limit of Technology (LOT) Load.
       As noted in the Introduction, in addition to these loading "boundaries", a series of other
loading patterns were used in the scenarios to determine Bay response over a wider range of loading
reductions.
             a. Base Case. The Base Case loading represents the calculated response from the
1984 -1986 loading as  a reference time period. This is the same period used for the calibration of the
CBWQM.
             b. Controllable Load. Controllable loads are defined as the difference between Base
Case loads and a WSM run using an all forested (excluding NY, WV & DE) basin with no point
sources.
             c. LOT Load. Limit of Technology loads for NFS inputs were determined by
evaluation of Best Management Practices (BMP) and implemented in the WSM. The basic LOT NPS
reduction components include:
             1.  Conventional tillage was placed into conservation tillage together with simulation of
the retirement of highly credible land by placing additional conservation tillage and  hay land into
pasture,
             2.  Animal waste control where 75% of animal waste acres were converted to pasture
land,

                                           II-1

-------
SUSQUEHANNA
    RIVER
     TO
ATMOSPHERIC
   INPUT
BELOW POTOMAC
     RIVER
      TO
	 te-
SUSQUE-
HANNA
DIX/ED
"UPPER
BAY"
REGION
BACK RIVER TO
BELOW POTOMAC RIVER
"MID - BAY" REGION
"LOWER
BAY" -^
REGION
       POINT SOURCE
       i nan	
                                         LOAD @ TRIBUTARY
                                        INTERFACE WITH RAY
                                BELOW FALL LINE NPS
                                LOAD
     FALL LINE LOAD
     (PS & NPS)
                              "EXTERNAL LOAD'
                              1. FALL LINE
                              LOADS (PS & NPS)
                              2. BELOW FALL
                              LINE NPS LOADS
                              3. POINT SOURCE
                              LOADS
                              4. ATMOSPHERIC
                              LOADS - DIRECT
                              TO BAY
                              5. OCEAN LOAD
              Figure 11-1. Load Categories and Bay Regions for
                         "Geographical" Scenarios
                                 II-2

-------
             3. Nutrient management controls where all cropland and hay land had reduced manure
and fertilized inputs as recommended by state NFS programs,
             4. Pasture controls, such as grazing load stabilization systems, stream protection
systems or spring development, where reductions of 4%, 8% and 8% were taken for TN, TP and
BOD, respectively,
             5. Urban controls, such as wet and dry ponds or infiltration trenches, where a 20%
reduction of TN and TP is assumed,
             6. Structural BMP's on farmland including any physical or constructed practice such as
vegetated filter strips or waterways implemented on cropland, where 4%,8% and 8% reductions in
TN, TP and BOD, respectively, are assumed, and
             7. Forest controls where NH4, PO4 and BOD loads were reduced by State to account
for LOT controls of silviculture, (5%, 7.5% and 10% for PA, MD and VA, respectively).
       Point source (PS) LOT loads were assumed as follows: 3.0 mg/L for TN, 0.075 mg/L for TP,
and 1.0 mg/L for BOD.

B. HYDROLOGICAL SEQUENCE
       In order to represent year to year variation in hydrology, the scenario runs were conducted
over a ten year period with flows representative of the interval from 1979 to 1988. Wet, average and
dry years during this period were assigned the 1984, 1986 and 1985 hydrologies, respectively
together with the associated transport for those years as calculated by the hydrodynamic model. The
total river flows to the Bay for 1984 to 1986 are 487,300 cfs (13,800 mVs), 459,100 cfs (13,000
nVYs), and 476,700 cfs (13,500 mVs), respectively. Figure II - 2 shows the seasonal flow variation for
the average year for three of the major tributaries to the Bay as generated by the WSM and as used
for load generation at the fall lines.
       Table II-1 shows the flows and sequence used in the scenarios.
TABLE H - 1
HYDROLOGICAL SEQUENCE USED IN SCENARIOS
YEAR
#1 - 1979
#2 - 1980
#3 - 1981
#4 - 1982
#5 - 1983
#6 - 1984
#1 - 1985
#8 - 1986
#9 - 1987
#10-1988
HYDROLOGY ASSIGNED
Wet (1984)
Dry (1985)
Dry (1985)
Average (1986)
Average (1986)
Wet (1984)
Dry (1985)
Average (1986)
Average (1986)
Dry (1985)
                                          II-3

-------
  100,000
   50,000
•o
§  20,000

-------
       The CBWQM was run for these ten years in sequence and the final five years were output. In
this report, the emphasis is on year #9, the average year. This choice is dictated by the time to a
dynamic steady state. The water quality response for the average year is the closest to this dynamic
equilibrium.
C. DESCRIPTIONS OF SCENARIOS

       The scenarios discussed in this report are summarized in Table II-2. A complete tabulation of
the input nitrogen and phosphorus loads for each scenario run is presented in Appendix A.

D. SCENARIO NITROGEN LOADS

       Figure II-3 shows the total nitrogen (TN) loads used for three of the principal scenarios: Base
Case, 40% Controllable and the Limit of Technology (LOT). These three runs provide an
approximate bounding of the feasible range of load reductions. As shown in this Figure, the major
source of nitrogen is from the nonpoint sources making up more than 45% of the total for all three
scenarios.  The point sources directly to the Bay or it's tidal tributaries decline from 19% of total at
Base to 5% at LOT indicating a significant reduction of point source loading relative to nonpoint
source reduction. This is discussed further below. Atmospheric loading directly to the tidal waters is
about 9% of total. Ocean loading of TN is a significant component of the total load to the Bay and is
estimated to range from 29% to 36% of the TN loading for Base to LOT.

       Figure II-4 indicates the % reduction of external TN loading (excluding ocean input) from the
Base Case for selected scenarios.  The increase in % reduction across the variations of the 40%
Controllable (#2) through scenario #4 can be noted. Clean Air Act controls (CAA) on a basin-wide
scale and including the non-basin states increase the % reduction from 18% to 25%. It can also be
seen that scenario #4 is approaching the LOT  scenario #5. The LOT-Mid (#7) run removes almost as
much TN as the 40% Controllable case.

       The distribution of TN loads by tributary and Bay shore area is plotted in Figure II-5 for three
scenarios.  The Susquehanna River input is the largest and makes up about 47-60% of the  total load
over the three scenarios. The reduction in the  Susquehanna load from the Base Case to LOT is about
14% while for the Potomac, the reduction from Base Case to LOT is about 43%. It can also be noted
that the contributions from the smaller tributaries in the Bay shore areas (i.e., West shore MD , West
shore VA and East shore MD/VA) together make up about 15% of the total TN load.

       The variation of the nitrogen components for the Base Case and the Susquehanna  River is
shown in Figure II-6. The N03 -N form is shown as the principal component to the TN load with
some additional contribution from the organic-N form. This contribution of nitrogen species is
approximately representative of the other tributaries.
                                          n-s

-------
I
   o
      w


      I
             I,
           8
11
                                 !.!•§
                                 — « I

                                  • oo;
                                 ^11
                                 «il
                                 Ili
               rii
               in
                                 ,2-S
                                 '« h
                                 II §



                                 III


                                 ^ w 8
                                 * cfl O

                                 5.Si
Q
s

^
                    a

                    1
                        *rj   vo
                                            l-s
                                            *.§.
                                            2 J§



                                            8 § '
                                            •i > '
                          ^1s

                          s>$
                          • p .£*  c ^
            ^ o v;
            .° 3 §
            g u (2
            53 ^ o




            ill


            |l I

            11!
            <« a S

            |£S?
            I g2
            ?2-a
            ;a
-------
I
                       'Eh
                                        8
II
                                                     II-7

-------
  CO
  D
  z


  2
  LL
  O
  CO

  g
  _i
  z
    250
o
O
en
O
o
O
en
O
(50)
                                                  (9)
         .(36)

NPS C
r O ES
ATMOS.l
OCEAN 1
TOTAL
#1-Base
207.5
86.8
34.6
131.8
460.7
#2-40% Cont
180.4
53.4
34.6
131.8
400.2
#5 LOT
183.0
17.0
34.6
131.8
366.4
 Figure II-3.  Total Nitrogen Loads by Type and Scenario,

                    Average Year.

        (Numbers in parentheses = % of total load)
                          SCENARIO
Figure II-4. Reductions in External Total Nitrogen Load for Selected

                          Scenarios.

                         Average Year
                            I-8

-------
                    Base Case
                                  40% Controllable
Limit of Technology
        D Susquehanna Q Potomac       O Patuxent     S Rappa/York
        m James      M West Shore MD m West Shore VA E East Shore MD/VA
     Figure 11-5. Total Nitrogen Loads Into Major Tributaries by Scenario -
                              Average Year
   so
   20
OH

O
V)  1

_O

S 0.5




  0.2



  0.1
                                         Susquehanna River
      Jan.   Feb.  Mar.   Apr.   May  June   July   Aug.  Sep.   Oct.   Nov.  Dec.

                        NH4   N03   Org. N   Tot. N


                   Figure 11-6. Base Case Nitrogen Loads from
                       Susquehanna River. Average Year
                                      -9

-------
       The monthly variation in TN loading for the average year and Base loading is shown in Figure
II-7 for three major tributaries. As indicated, the within-year variation is approximately similar for the
three tributaries. It can also be noted that, for the Susquehanna, the range from the peak loading in
March to the lowest loading (which occurs in September) is considerable and declines about 95%
during this period. Minimum loading for the James is estimated to be during the June-July period
while for the Potomac and Susquehanna rivers, the minimum occurs in September.

       The relative reduction of TN loading due to point and nonpoint  sources indicates that the
point source loading is reduced considerably more than the nonpoint loading between the Base Case
and LOT. This is illustrated in Figure II-8 where the loading across the three Bay regions (see Figure
II-1) is shown. As shown, in progressing from Base to LOT, the point source loading is reduced
about 85% while the nonpoint source loading is reduced from  14-23% in proceeding from Base to
LOT. This is a reflection of the relative technological difficulty in reducing nonpoint TN loading as
opposed to point TN inputs. Also, it should be noted that the Upper Bay region is dominated by the
nonpoint input of the Susquehanna River so that the contribution from point sources in that region is
small.

       Further insight into the relative reduction of the TN loading across scenarios and between
point and nonpoint loading is given in Figure II-9 and 11-10. For the former Figure, the % reduction
of the point source for the 40% Controllable case is 40% by definition. For the nonpoint loading for
that scenarios,  however, the % reduction is only 14% leading to a net 18% reduction as indicated in
Figure  II-4.  Also, as shown the % reduction for the nonpoint inputs for the 40% Cont. scenario is
about equal to  the nonpoint reduction for the LOT case (14% vs.  16%).

       Finally, the relative contribution to the reduction from Base case from three categories of TN
loading and across several scenarios is shown in Figure 11-10.  In this Figure, "Fall Line" includes the
NFS and PS entering the Bay and tidal tributaries, "Below Fall Line"  is the NPS loading entering the
tidal tributaries and Bay and "Point Source" represents the input loading of point sources below the
Fall Line (see also Figure II-1). Again, the significant contribution from  point source reductions is
shown  for all scenarios except the LOT-Upper run which, as noted earlier, is dominated by the
Susquehanna input.

E. SCENARIO PHOSPHORUS LOADS

       Figures II-11 through 11-18 parallel the previous figures for nitrogen but focus on the
phosphorus loading,  reductions in loading and distribution of phosphorus loads.

       Figure II-11 shows an important difference from the TN plot (contrast to Figure II-3). As
shown, the ocean loading of TP dominates the loading inputs accounting for as much as 66% of the
TP load at the LOT scenario. (Reference should be made to Cerco and Cole, 1992 for a detailed
discussion of the ocean boundary condition that gives rise to this significant TP input.) It can also be
noted that the nonpoint and point source loading are closer in magnitude than for TN, but for LOT
the nonpoint load is about the same as for the 40% Controllable case.

                                          11-10

-------
30


10
   g>    3
   U
   +*
   Z
   •S    1

   o
   OH
   •g  0.3
   a
   o
   s  0.1
      0.03
      0.01
            B	s'
                        QSusquehenna

                        	V-"-"-"Q.
                                              James
           Jan.  Feb.  Mar.   Apr.   May   June  July  Aug.  Sep.   Oct.   Nov.  Dec.

           Figure II-7. Base Case Nitrogen Loads from Three Principal
                             Rivers. Average Year
     200,000
  150.000
     100.000
z
K


1
      50,000
              POINTSpURCg .
                 LOADING

                    83 - 85%
                    REDUCTION
                             DBASE
           UPPER BAY   LOWER BAY
                 MID - BAY
                                  200,000
                                   150.000
                                 i
                                 b 100,000
                                        o
                                        o
                                   50,000
NONPOINT
SOURCE




-
-
LOADING

w
4
i
i










i
-^
	 .1.4
Rl





21
RE
( — i
\%,
§
R
DBASE
^LOT

/.
DUCTION


DUCTION
M
feuCTION
                                         UPPER BAY  LOWER BAY
                                               MID - BAY
Figure 11-8. Longitudinal Distribution of Total Nitrogen Point & Nonpoint Loads
for Base Case and Limit of Technology (Nonpoint = Fall Line + Below Fall Line
                                  Loads)
                                    11-11

-------
       Figure 11-9. Reductions in Point and Nonpoint Nitrogen
                             Loading.
                           Average Year
   100
 1°°
 I60
   40
   20
     40« CONT.  LOT-Uppw  LOT_LoMT  LOT P ONLY
           LOT    LOT-MM  LOT N ONLY
                                         100
I  80
                                       O
   60
                                          40
   20
NOTE: 40%
CONT. & LOT
ARE APPROX.
EQUAL.
     4MCONT.  LOT4jpi»r  LOT-LOW  LOT p ONLY
          LOT     LOT-Md   LOT N ONLY
    Figure 11-10. % OF Total Nitrogen Reduction Due to Fall Line, Below Fall
                 Line and Below Fall Line Point Source Loads.
                                 Average Year
pa
O 100
  60 -
  40 -
   20 -
yi





«
_


A '

" » IF

-


j
J




»
|
:'::






























M

«









U



















: , »
; :














41
M

;



3}


i
T













	











^^
H

;!!i!!:i
;:;;;,:


FALL
LINE
BELOW
FALL LINE
POINT
SOURCE


Note "Fall Line" =

PS
+ NPS entering
from upper basins
"Below Fall Line" =
M
^


3
"1
NFS entering tidal
tnb
or Bay
"Pant Source" =
PS entering below
Fall Line
      40%(fONT.         ;LOT-Upper j        LOT-Lower        LOT P ONLY
              '.  LOT   ,        i LOT-Mid         LOT N ONLY
                             SCENARIO
              Ex. 47% Of Total Nitrogen Load Reduction Was Due To Reductions
              In Below Fall Line Point Sources
                                    1-12

-------
       The % reduction of TP from Base for a series of scenarios is given in Figure 11-12. For TP,
the reductions are higher than for TN reflecting the relative increase in technological control for TP
over TN. For LOT, a 56% reduction is calculated although it should be recognized that if the ocean
load were included, the net % reduction of TP from Base due to all loads drops to 29%.

       Figure 11-13 (for TP) is similar to Figure II-5 (for TN) except that the contribution from the
James river is significantly greater in TP than TN. The relative input of TP from the Susquehanna
river is less than TN ranging from 33% at Base to 41% at LOT (as opposed to 47-60% for TN).
Figure 11-14 shows that the principal phosphorus species calculated for the Susquehanna is the
organic-P form, considered a less available form for phytoplankton uptake in the CBWQM. The
seasonal variation in TP load for the average flow year is shown in Figure 11-15 and shows a pattern
similar to that for TN. (A discussion of the ratio of the TN/TP for the input loads is given below.)

       Figures 11-16 through 11-18 differ from comparable figures for TN (Figures II-8 to 11-10) in
the increased reduction of TP from nonpoint sources. Figures 11-16 and  17 indicate that although PS
reductions from Base to LOT are about 96%, NFS reductions are about 45% or twice the reductions
for NPS TN. Figure 18 illustrates this further by showing that the relative contribution to the total
reduction from point sources is 36% for the 40% Cont. case. (Figure 11-18 is similar to Figure 11-10
where "Fall Line" represents both NPS and PS loadings from above the Fall line, "Below Fall Line"
are the NPS loads entering below the Fall line and "Point Source" are the direct inputs of PS below
the Fall Line.)
       In contrasting these TN and TP load estimates, relative to PS reductions, decreases in the
NPS TN loading are calculated to be more difficult technologically than  NPS TP loading.  This is
presumably a reflection of the assumptions made throughout the general process of load generation.
F. TN/TP RATIO FOR INPUT LOADS

       The ratio of the TN/TP loading provides a guideline for determining which nutrient may be
limiting in the control of phytoplankton biomass. A TN/TP ratio of 7 (by mass) represents the
"Redfield Ratio", i.e., the elemental stoichiometry of oceanic algal cells. In general, TN/TP values
significantly less than about 7-10 indicates potential nitrogen limitation while TN/TP ratios
significantly greater than 7-10 indicates potential phosphorus limitation.  Figure 11-19 shows the
TN/TP ratio for the input loads as described earlier. Several points can be noted. For the average year
shown in this figure, the overall TN/TP ratio is close to the range where nitrogen or phosphorus may
be limiting. Ocean TN/TP is at the Redfield ratio (an assumed shelf water TN of 0.37 mg/L and TP of
0.053 mg/L was used). The NPS loading TN/TP is significantly above 7-10 indicating potential
phosphorus limitation from that source, but because the ocean load is a significant portion of the total
load to the Bay, the overall TN/TP is decreased. It should also be recalled that these values are for the
entire year and monthly variations are to be expected as shown in the next Figure.

       Figure 20 shows these monthly variations in TN/TP for three significant tributaries. The
Susquehanna river TN/TP is always significantly above 7-10 while  the Potomac and James tributary
loading is in the region of 7-10 during several important months of the year.

                                           II- 13

-------
r«^u
OF POUND!
Ol
• MILLIONS
Ol O
Q.
- n

-



(28)
(5

JA

2)
(3)
IP
ZL





(22)
(1

&

6)
15)
M
•L,





(26)

$


jjA
IL




NPS C
PS 1
ATMOS.l
OCEAN E
TOTAL
#1-Base
11.1
8.8
1.5
18.9
40.3
#2-40% Cont
7.5
5.4
1.5
18.9
33.3
#5 LOT
7.5
0.6
1.5
18.9
28.5
     Figure 11-11.  Total Phosphorus Loads by Type and Scenario,
                         Average Year.
             (Numbers in parentheses = % of total load)
100
                              SCENARIO
     Figure 11-12. Reductions in External Total Phosphorus Load for Selected
                                Scenarios.
                              Average Year
                                1-14

-------
          8
       CO
       0


       §6
       Q.
       UL
       O
       CO 4

       O
       Q.
          0
3.3
                                                                1.2
                 Base Case     40% Controllable Limit of Technology
        O Susquehanna ED Potomac       d Patuxent      @ Rappa/York
        1 James       m West Shore MD B West Shore VA S East Shore MD/VA
       Figure 11-13. Total Phosphorus Loads Into Major Tributaries by
                         Scenario - Average Year
3


0. °*

   0.03
   0.01
  0.003
  0.001
            _l	l_
                                         Susauehanna River
      Jan.   Feb.   Mar.   Apr.   May   June   July   Aug.  Sep.   Oct.   Nov.  Dec.

                            P04   Org. P   Tot. P




                Figure 11-14. Base Case Phosphorus Loads from

                       Susquehanna River. Average Year
                                     -15

-------
    0.3
   0.03
   0.01
  0.003
  0.001 L-L
                             i	I	i
       Jan.   Feb.   Mar.   Apr.   May   June   July   Aug.  Sep.   Oct.   Nov.  Dec.
        Figure 11-15. Base Case Phosphorus Loads from Three Principal
                            Rivers. Average Year

POINT SOURCE PHOSPHORUS LOADING kg/day
hO *• O> 00 p
§o o o o
o o o o
o o o o



POINT SOURCE
LOADING





oe
REDL




%
C1

ION

DBASE 10'000
^LOT
3 8,000
o
2 6,000
o
| 4,000
Q.
| 2,000
o
z
n



-
r— |


NONPOINT
SOURCE
LOADING
REDUCTION
1





I
1



|
DBASE
^LOT
                 MID - BAY
                                                      MID - BAY
  Figure 11-16. Longitudinal Distribution of Total Phosphorus Point & Nonpoint
Loads for Base Case and Limit of Technology (Nonpoint = Fall Line + Below Fall
                                Line Loads)
                                    11-16

-------
         Figure 11-17. Reductions in Point and Nonpoint Phosphorus
                                 Loading.
                               Average Year
  100
   80
   60
o
g
   40
   20
                            1
     WUCONT.   LOT-UppK   LOT4jMn(  LOT P ONLY
          LOT    LOT-MW   LOTNOMY
100
 80
 60
 40
 20
   40* C0t*r  LOT-Uppr   LOT-UXMT  LOT P ONLY
        LOT     LOTMU   LOT N ONLY
    Figure 11-18. % OF Total Phosphorus Reduction Due to Fall Line, Below
                      Fall Line and  Point Source Loads.
                                Average Year

 100
  80 -
  60 -
  40
  20 -
M


_


-

«' : pit,

-



-








'. 43
M



U
1





















9














1















M





4* '

37




m1-.








n
I













n














.• 71


























HP
^n?



FALL
LINE
BELOW
FALL LINE
POINT
SOURCE
Note "Fall Line" =

41
43


2S
w
i
1


\











t







PS
+ NPS entering
from upper basins
"Below Fall Line" =
NPS entering tidal

trib
or Bay
"Pant Source" =
PS entering below
Fall Line



      40%CONT.         LOT-Upper         LOT-Lower        LOT P ONLY
            ,    LOT           LOT-Mid         LOT N ONLY
            4                 SCENARIO

            I	Ex. 36% of Total Phosphorus Load Reduction Was Due To
                Reductions In Below Fall Line Point Sources
                                  11-17

-------


NPS 1
PS m
ATMOS.l
OCEAN 1
TOTAL [I
#1-Base
18.7
9.9
23.0
7.0
11.4
#2-40% Cont
24.0
9.9
23.0
7.0
12.0
#5 LOT
24.4
28.0
23.0
7.0
12.8
 Figure 11-19. TN/TP Ratio by Load Type and Scenario, Average Year.
  30
to 9K
O Z5

a.


-------
G. SCENARIO CARBON LOADS

       Total incoming carbon loads were estimated from point source inventories and from
Watershed model runs. Table II-3 is a summary of the allochthonous (external) carbon loads for the
Base, 40% controllable and LOT scenarios. Figure 11-21 shows the organic carbon loading by basin
and scenario.
TABLE H - 3
ORGANIC CARBON LOADINGS
(Average Year)
SCENARIO
#1 - Base
#2 - 40% Controllable
#5 - LOT
CARBON LOAD
(10A6 Ibs/yr)
484
327
300
% REDUCTION
FROM BASE

32
38
       In contrast to nitrogen and phosphorus loads, a significant source of organic carbon is also
generated internally by phytoplankton primary production. As will be seen later in this report, the
principal carbon loading to the Bay is from this autochthonous (internal) source. Therefore the
external carbon sources assume a reduced role in terms of the impact of controlling these sources.
Table II -3 indicates an overall reduction from the Base case of about 32% for Scenario #2 and a
relatively small improvement to 38% for LOT. Figure 11-21 shows the principal external carbon
source to be the Susquehanna River followed by the Potomac River. These results also indicate that
the 40% controllable scenario is close to the Limit of Technology for organic carbon loading.
H. SCENARIO SOLIDS LOADS

       The suspended solids in the Bay are of importance in determining the extent of light extinction
and penetration and hence influence the production of plant biomass. The CBWQM does not model
the fate and transport of suspended solids. As currently configured, the CBWQM calculates changes
in light penetration or extinction by tracking changes in phytoplankton chlorophyll. The concentration
of suspended inorganic solids (as empirically linked to incoming river flows) is therefore not changed
across scenario runs. However, an examination of input solids loading as generated by the Watershed
Model is of interest to qualitatively determine the potential impact of nonpoint nutrient controls on
light penetration.
                                          11-19

-------
  400
  300
•5200
W

i
  100
                   Base
                                     223
                                         40% Controllable
100
                                                                    Limit of Tech.
   LJ Susquehanna  Ej Potomac I1T11 Patuxent g Rappa/York ^ James |H West Shore MD H West Shore VA ^ East Shore MD/VA


    Figure 11-21.  Total Organic Carbon Loading by Scenario and Basin.

                                    Average Year
                                           II-20

-------
       Briefly, the Watershed Model calculates sediment delivery based on the following scheme:
Land Processes
Erosion
Rate

Storage

Factors
River Processes

Scour Sediment
       The simulated erosion rate is dependent on rainfall, energy, antecedent soil moisture, and
percentage of exposed soil.  These values are largely input as data sets or are simulated internally.
       Sediment storage is considered to be sediment dislodged by rainfall or plowing and available
for transport. Sediment in storage is calculated at each time step as a balance of sediment detachment
and re-attachment. Sediment storage for each land use is consistent throughout the model.
       Transport factors  move the field storage sediment to the river.  These parameters are selected
 to match calculations of annual erosion for crop, pasture, and forest land based on National
Resource Inventory (NRI) data. (The NRI is a national data base of land use and characteristics such
as cover, slope and estimated erosion rates.) Gross erosion estimates are reduced by a delivery ratio
to represent deposition within smaller sub-watersheds.  The NRI data are at a county level.
       Deposition velocities are user supplied parameters. Critical shear stresses are adjusted to
simulate observed sediment concentrations at USGS river monitoring stations.
       Suspended sediment is a Watershed Model state variable which is directly comparable with
observed suspended sediment at monitoring stations.
       Figure 11-22 shows the calculated suspended solids loading for two principal inputs for the
base case and over several years.  The Potomac River sediment load is about an order of magnitude
higher than that of the Susquehanna River. This is due  to the reservoirs at the terminus of the
Susquehanna River which act as sedimentation systems. Also, for the Potomac, the sediment load did
not drop during the 1985 dry year, as would be expected and as calculated for the Susquehanna. The
reason for the similar solids loading for the Potomac during the dry year is a large storm in November
1985 which crossed the upper Potomac. The WSM calculated significant bed load scour from this
storm and subsequently transported this sediment load  downstream. It can also  be seen that the load
for the Susquehanna can vary by about one order of magnitude between differing hydrologic years.
       Currently, only two controls/scenarios result in reduced sediment loads:
       1. conversions of conventional to conservation  tillage; and
       2. other LOT land use changes (crop land use into  pasture land use).
        These controls therefore reduce sediment loading through land use changes. It is important to
note that no changes were made to account for sediment reduction from farm practices in the LOT
scenario. The delivered LOT sediment reductions should therefore be considered extremely
conservative.
       Figures 11-23 and 11-24 show the WSM calculated  sediment loads for differing scenario and
controls. The "3-state forest" loading is considered to be the level of uncontrollable sediment loads
delivered from Bay program states.  As in the case of nutrients, New York in the Susquehanna basin,
and West Virginia in the Potomac basin are left at base case land use (and base  case sediment loads).
       It is clear from these figures that the reduction in sediment load from base to LOT is small,
 but the previously mentioned important caveat should  be noted. The year to year variation in solids

                                           11-21

-------
SUSPENDED SOLIDS LOAD (tons/yr)

5,000.000
2,000,000


1,000,000


 500,000


 200,000


 100.000


  50,000


  20,000


  10,000
Susquehanna River
       1984
           ret)
         1985
          (Dry)
YEAR
1986
(Average)
  1987
(Average)
     Figure II-22. Variation of Suspended Sediment Loads from
       Susquehanna and Potomac Rivers for Different Years
                                  I-22

-------
              SUSQUEHANNA RIVER
     250.000
    . 200.000
   9


   8

   S
   Q

   5
   a.
   V)

   
-------
 loading is considerable and will tend to reduce the effectiveness of sediment control measures. For
the Potomac, compared to other years, 1985 has less spread in the differences of load among the
scenarios, particularly between base and forest. This is attributed to the previously mentioned
November 1985 storm.
       Generally, sediment loads of basins can be characterized as transport dominated processes in
large basins and source dominated sediment loads predominate in small basins (Walling, 1983).  This
may explain to some extent the little difference among the few sediment reduction scenarios, i.e., the
basin is not yet in steady state between sediment input from the land and transport of stored loads in
the river. This is, a least, entirely consistent with the literature, which generally finds little immediate
effect of Best Management Practice (BMP) on discharged sediment load (Illinois EPA, 1983;
Walling, 1983). Additional runs would be desirable to further characterize the extent of sediment
control, but these large basin are "in-stream process" dominated with respect to sediment loads. As
indicated, it is known that the response of such system in the short run to sediment BMP's is not as
great as it would be if the sediment loading were dominated by edge of stream loads.
I. SECTION II - CONCLUSIONS

       The following principal conclusions are drawn from the loads given in this section.

       1. The "feasible" range of overall total nitrogen reduction from the base case is from about 20
- 30% of the total input load (excluding input from the ocean),
       2. The "feasible" range of overall total phosphorus reduction from the Base case is from about
30-55% of the total input load (excluding input from the ocean),
       3. The calculated ocean nutrient input load (which is independent of scenario) is estimated to
contribute about  30-35% of the total input nitrogen load and about 45-65% of the total input
phosphorus load,
       4. Deposition of atmospheric nitrogen directly to the Bay waters is about 10% of the Base
case loading,
       5. 40% reduction of controllable nitrogen for nonpoint sources is approximately equal to the
Limit of Technology reduction of nitrogen nonpoint sources,
       6. The application of LOT results in significantly larger percentage reductions in point source
nutrient loadings to the Bay than nonpoint source loadings,
       7. The TN/TP ratio for total input load is calculated at about 12 across scenarios, indicating
an input load situation that depending on season and Bay location,  may result in either nitrogen or
phosphorus controlling phytoplankton production,
       8. Reductions in suspended solids between Base and LOT are conservatively estimated at
about 6-14% for the Potomac and Susquehanna Rivers, but would be considerably higher if
reductions would include decreases in sediment due to farm plans.
                                           11-24

-------
III. CBWQM RESPONSE TO SCENARIO LOADS - GENERAL
CONSIDERATIONS
A. TEMPORAL AND SPATIAL AVERAGING

       Since model output is very large for all state variables, time and locations, some averaging
of model results over time and space is necessary. Cerco and Cole, 1992 discuss this averaging in
the context of model calibration and present the details of the averaging used therein. The same
averaging is used for the scenario output. Although the CBWQM calculates state variable
concentrations at a time scale of hours, such calculations are for computational stability only.
Input information is provided on a week to week basis and the kinetics that are incorporated in
the model are representative of longer time behavior. The model is considered to represent
processes on a time scale of months, seasons and longer. Therefore, some of the model output
results were averaged over months while other results were averaged over seasons according to
the Table below.
   TABLE III - 1. TEMPORAL PERIODS USED IN AVERAGING MODEL OUTPUT
SEASON
I
II
III
IV
DESCRIPTION
"Winter"
"Spring"
"Summer"
"Fall"
JULIAN DAY
0 -60
61 - 150
151 - 270
270 - 365
APPROX. MONTHS
Jan.- Feb.
Mar. - May
June - Sept.
Oct. - Dec.
      The Bay spatial grid scale horizontally is about 10 km by 5 km by 1.7m and includes from
two to fifteen cells in the vertical direction. Two sediment segments (aerobic and anaerobic) are
incorporated under the water column segments. Again, in order to provide tractable output, water
column model results are averaged spatially according to the zones indicated in Figure III -1 and
described in Table III - 2 below.

      The water surface areas for each zone and for the entire Bay are also shown in this Table
and are used to compute areal loading and responses  of various constituents as detailed later in
this report.
      The principal Bay tributaries are also divided into zones for calibration analysis (see Cerco
and Cole, 1992). The areas for the tributaries are also indicated in Table III - 2.
                                        III-l

-------
(B
-9
                                                          o
                                                          3
                                                          O
                                                          s
                                                          O
                                                          W)

                                                          S
                                                          OJO
                                                         a
                                                         fl
                                                         o
                                                         N
                         m-2

-------
   TABLE m - 2. ZONES USED FOR SPATIAL AVERAGING OF MODEL OUTPUT
ZONE
1
2
3
4
5
6
7
8
9
DESCRIPTION
Conowingo to Back R.
Back R. to Choptank R.
Choptank R. to Patuxent R.
Patuxent R. to Potomac R.
Pot.R. to Rappahannock R.
Rapp. R. to York R.
York R. to James R.
James R. to Mouth
Eastern Shore
RANGE-km
(0 km = mouth)
330 - 272
272 - 220
220-181
181-138
138-90
90-56
56-27
27-0
159-90
ZONE
LENGTH (km)
58
52
39
43
48
34
29
27
69
ZONE
SURFACE
AREA (109 m2)
0.502
0.977
0.94
0.859
1.04
1.03
0.617
0.462
1.44

TOTAL BAY AREA 7.87
TRIBUTARY SURFACE AREAS
Patapsco R.
Patuxent R.
Potomac R.
Rappahannock R.
YorkR.
James R.
0.1
0.114
0.951
0.365
0.465
0.511
      Averaging over depth is accomplished by dividing the water column into three layers: the
"surface" well-mixed layer (0 - 6.7 m), the "pycnocline" layer (6.7 -12.8 m) and the "bottom"
layer (>12.8 m), as shown in Figure III-2.

      Table III - 3 shows volumes of the Zones by the three levels. Figure III - 3 shows the
cumulative surface and bottom volumes.
                                      ra-3

-------
                                     CONSTITUENT
                                     CONCENTRATION OR MASS
                  "SURFACE"  0-6.7m
                   "PYCNOCLINE"  6.7-
                   12.8m
                  "BOTTOM"  > 12.8 m
  DEPTH OF
  WATER
  COLUMN(m)
Figure III-2. Depth ranges used for vertical averaging of model
variables.
     2E+8
        350     300     250    200    150    100    50      0
            Distance (km) Along Main Bay (0 = Mouth of Bay)

Figure III-3. Cumulative Bay zone volumes of surface and bottom layers.
                             III-4

-------
TABLE III - 3 VOLUMES OF ZONES AND VERTICAL LEVELS
Region or Zone


CONOWINGO TO
BACK RIVER
BACK RIVER TO
CHOPTANK
CHOPTANK TO
PATUXENT
PATUXENT TO
POTOMAC
EASTERN SHORE
POTOMAC TO
RAPPAHANNOCK
RAPPAHANNOCK
TO YORK
YORK TO JAMES
JAMES TO MOUTH
PATAPSCO AND
BACK RIVERS

Mid-point of
Zone- km
301
246
200.5
159.5
124.5
114
73
41.5
13.5

Level One
Volume
mA3
2.26E+09
5.81E+09
6.00E+09
5.72E+09
8.72E+09
7.11E+09
7.05E+09
4.15E+09
3.06E+09
6.10E+08
Level Two
Volume
mA3
4.35E+08
1.77E+09
2.24E+09
2.70E+09
2.54E+09
3.64E+09
3.65E+09
2.75E+09
1.59E+09
1.75E+08
Level Three
Volume
mA3
0
9.75E+08
2.15E+09
2.07E+09
8.59E+08
2.46E+09
1.37E+09
5.35E+08
5.02E+08
87,324,040
TOTAL
VOL
Billion
mA3
2.7
8.6
10.4
10.5
12.1
13.2
12.1
7.4
5.2
0.9
B. TRIBUTARY INTERFACES

       The interfaces between the tributaries and the main Bay are a delineation point for
estimating the flux of nutrients from a Bay tributary to the main stem. (These fluxes are discussed
in Section VI - F.) The interfaces of the tributaries represent a complicated flow pattern where in
general, flow enters the tributary from the main Bay at the bottom and exits the tributary in
surface layers. However, this general pattern may be altered or reversed depending on local
                                         III-5

-------
 geometry. The fluxes of constituents from the tributaries to the main Bay are calculated by
 summing all flows "into" the Bay with associated concentrations and all flows "from" the Bay.
Fluxes are therefore not always necessarily associated with surface flows and bottom flows, but
rather with tributary flows into the Bay and with flows from the Bay to the tributary. The
tributary flows are further discussed below in Section IV.
C. FUNDAMENTALS OF RESPONSE SURFACE ANALYSIS

       Time and cost constraints precluded exhaustive running of the Bay Model to explore the
infinite N and P reduction options. A statistical technique called response surface analysis was
used to enhance and interpret the information contained in the limited number of model runs.
Response surface analysis was used to make and validate generalizations about how the Bay
responds to nutrient reduction, to interpolate between model runs, and to compare the levels of
achievement of different Bay water quality goals under a range of N and P inputs.  In this
technique groups of model runs are considered together so that trends in the model results can be
appreciated. These trends are approximated by a mathematical function, the response surface.  In
the following discussion it may help to think of the following response surface analysis simply  as
an application of interpolation in two dimensions, the dimensions being reductions in nitrogen
and phosphorus load to the Bay.
       The outcome of any Bay model run can be considered to be some function of the nitrogen
and phosphorus loading of that run:
where, for j = 1, 2, ..., mj stands for each one of m different model runs. The residual 6j is the
error of they'th observation. yj is some measurement of Bay response to different levels of nutrient
loadings, such as the reduction in anoxia from scenario j or the level of SAV habitat improvement
associated with that scenario predicted by the model. The function/is called the response
surface. This function represents the workings of the model plus the process of condensing the
model output. It has no precise mathematical form. /summarizes the trends in the model runs and
enables the prediction (within the region covered by the data) of the Bay response for
combinations of N and P loads which were not tested explicitly in a model run.

       When the exact mathematical form of the response surface is unknown, or as in this case,
does not exist, /can be approximated by a quadratic polynomial in N and P
                  Yj =  b,  4- b2N, + b3PJ

       This is a very adaptable form. If the true response surface is flat, the second degree terms
drop out in ordinary least squares regression. If the response surface is curved, the squared terms
are significant. If multiplicative interactions between N and P are important, that shows up in the
cross product term.
                                         III-6

-------
       A prerequisite to response surface analysis is that there must be a unique relationship
between settings of N and P and the response^. That means that the function must return a
unique y} for a unique choice of Nj and P,. Therefore the definitions of N, P and.y must be chosen
carefully, and only model runs that conform to those definitions must be used to fit the
polynomial.

       Model scenarios where N and P reductions are defined as bay-wide same-percent
reductions fit this requirement. Also the 40% controllable N and P reduction scenarios (runs 2, 3,
and 4) approximately fulfill this requirement. Runs 2, 3, and 4 were included even though they
weren't entirely free of geographic loading influence because of the desire to have a few more data
than parameters to be fit.  There are 6 constants in equation 2, and the regressions were more
significant using 10 data than 7. The 10 runs chosen to fit the polynomial were:

Scenario 1           Base Case
Scenario 2           40% controllable N and P reduction from Agreement States
Scenario 3           Scenario 2 with N reductions from Clean Air Act implementation
Scenario 4           40% controllable N and P reduction from all Bay basin states plus Clean
                            Air Act N reductions
Scenario 15          50% N and P reduction from base case
Scenario 16          90% N and P reduction from base case
Scenario 17          90% N reduction from base case
Scenario 18          90% P reduction from base case
Scenario 19          31% N 18% P reduction from base  case
Scenario 20          10% N 49% P reduction from base  case.
       The other runs were not used because of the possibility of unequal geographic loadings
violating the prerequisite.1

       N and P were expressed in one of three sets of units in different regression analyses:  (1)
percent reductions from base case (with base case being 0% reduction of N and P), (2) daily loads
of N and P to the Bay (kg/day), and (3) annual loads (million pounds per year). Changing the
units of the independent variables has no effect on the shape of the surface. Only the scaling of the
axes changes. For simplicity, all results will be reported using the first units only.
1       An attempt to construct a regression which could use all the runs was made; however, there were
insufficient model runs to fit all the variables. Three N and three P variables were defined, one set of N and
P loadings for each of the 3 geographic regions of the Bay. This made a total of 6 independent variable
linear terms, 6 squared terms and 15 cross product terms. Preliminary results were very promising (with
high R2's) and showed that N and P loads from regions 1 and 2 dominate the production of anoxia in the
Bay, but at least 2 more model runs would have been necessary to finalize the analysis.
                                              III-7

-------
       The predicted response^ also assumed various forms, all expressed as percent
improvement over base case for an average year. Using the same 10 scenarios, unique response
surfaces (presented here for the average flow year #9) can be generated for each of these
responses.
       (1) whole Bay anoxia (anoxia defined as DO < 1 mg/L)
       (2) single zone anoxia for each of the 9 Bay zones
       (3) summer, fall, or spring anoxia for the whole Bay
       (4) summer, fall, or spring anoxia for each of the 9 zones
       (5) whole Bay dissolved oxygen habitat goal achievement (see Table 1-1 of Section I)
       (6) whole Bay dissolved inorganic nitrogen goal achievement (see Table 1-2 of Section I).

       Since there was no violation of the dissolved inorganic phosphorus habitat goal in the base
case run, this goal was not analyzed further.
       Presentation of the results of the response surface analysis for DIN is given in Section VI -
B - 4 and for anoxia in Section VII - E.
                                          III-8

-------
IV. BAY HYDRODYNAMIC TRANSPORT USED IN SCENARIOS
A. INTRODUCTION

       The purpose of this section is to briefly review the principal aspects of the hydrodynamic
transport as calculated by the hydrodynamic model of the Bay and which formed the underlying
flow transport used in the scenarios. A general understanding of the flow transport is necessary in
order to more fully interpret the scenario results. A complete description of the hydrodynamic
model is given in Johnson et al, 1991.

B. MAIN BAY FLOWS

       In general, the hydrodynamic flows of the Bay are a complicated function of ocean
boundary condition, winds, river inflows and Bay geometry and bathymetry. Flows vary over the
tide, over days, weeks, and seasons. Surface Bay flows are generally down the Bay toward the
ocean while a return flow along the bottom layers of the Bay occurs from the ocean into the Bay.
The apparent force of the earth's rotation further adds to the circulation by deflecting currents to
the right in the direction of the flow. The simple two layer flow is therefore further impacted and
inflows up the Bay may occur at mid-depth or surface layers as well as bottom layers.
       In order to provide at least a preliminary understanding of the transport of the Bay,
average annual flows have been computed across the interfaces of each of the zonal boundaries.
Inflows are not always in a bottom layer. Similarly, "Outflows" are average flows for each cell
that are leaving an interface. Outflows are not always in the surface layers.
       Table IV - 1 is a compilation of the average annual interfacial flows. (See Figure III -1 for
a map of the zones.) As shown in this Table, for the main Bay, the volume of flows entering and
leaving an interface are large and the net difference between the flows is relatively small. Also,
there is a relatively complicated transport structure in the vicinity of Zone 9, the Eastern Shore
zone. Flows enter and leave this zone from Zones 4 and 5.  Since Zone 5 receives the inflow from
the Potomac estuary, the exchanges between that zone and Zone 9 is large. A flow balance
around zone 5 indicates a significant transport into the Eastern Shore which then exits into Zone
6. Net interfacial flows from the tributaries to the main Bay  are also the differences of two large
inflows and outflows. The tributary flows are discussed more fully below.
       Figure IV - 1 shows the longitudinal profile of the main Bay zonal interfacial flows and  the
general increase in the inflows and outflows as  one proceeds down the Bay can be observed. Also,
shown in this Figure is the net outflow across the interfaces  where the exchange from 5 to 6
incorporates the exchange with zone 9. Except for the Eastern Shore region with its more
complicated transport regime, the increase in the net outflow is relatively small and reflects
primarily the net inflows from the tributaries.
       Figure VI - 2 is a plot of the longitudinal variation in the net outflow from a zone interface
as a percent of the total inflow to the interface. A high of 60% net outflow/inflow occurs at the
head end of the Bay and declines to less than 10% in the lower half of the Bay. The net transport
in the lower Bay is therefore not a significant component of the overall transport which is
governed by the large multi-level Bay circulation.

                                         IV-1

-------
                                    TABLE IV - 1
           AVERAGE ANNUAL INTERFACIAL FLOWS (m'/s) FOR AVERAGE YEAR
INTERFACE
Zone 1 & 2
Zone 2 & 3
Zone 3 & 4
Zone 4 & 5
Zone 4 & 9
Zone 5 & 6
Zone 5 & 9
Zone 9 & 6
Zone 6 & 7
Zone 7 & 8
Zone 8 & Ocean
James & Bay
York & Bay
Rapp. & Bay
Pot. & Bay
Patux. & Bay
Patap.& Bay
FLOW (From Zone # to
Zone #)
2047 (2 to 1)
96 19 (3 to 2)
10042 (4 to 3)
14048 (5 to 4)
104 1(9 to 4)
10457 (6 to 5)
6213 (9 to 5)
6238 (6 to 9)
15949 (7 to 6)
17962 (8 to 7)
26020 (Ocean to 8)
2594 (B to J)
4794 (B to Y)
1615 (B to R)
4258 (B to Pot)
1108(BtoPtx)
1, 866 (B to Ftp)
FLOW (From Zone # to
Zone#)
3252 (1 to 2)
10838 (2 to 3)
11273 (3 to 4)
13875 (4 to 5)
2436 (4 to 9)
10408 (5 to 6)
6309 (5 to 9)
7697 (9 to 6)
17370 (6 to 7)
19414 (7 to 8)
27630 (8 to Ocean)
2750 (J to B)
4825 (Y to B)
1640 (R to B)
4498 (Pot to B)
1115(PtxtoB)
1871(PtptoB)
NET FLOW (From Zone #
to Zone #)
1204 (1 to 2)
1218 (2 to 3)
1230 (3 to 4)
173 (5 to 4)
1395 (4 to 9)
49 (6 to 5)
96 (5 to 9)
1459 (9 to 6)
142 1(6 to 7)
145 1(7 to 8)
1610 (8 to Ocean)
156(JtoB)
31(YtoB)
25 (R to B)
240 (Pot to B)
7 (Ptx to B)
5 (Ftp to B)
       1. Estimate of Hydraulic Residence Time of Bottom and Surface Waters
       An estimate can be made of the hydraulic residence time of the Bay zones by assuming a
two layer Bay where the "Inflows" to a zone are primarily in the bottom layer and a fraction of
the mid-layer volume and the "Outflows" from a zone are primarily in the surface layer and a
fraction of the mid-layer volume. The hydraulic residence time for the bottom layer is estimated
for each zone using the zone volume and the flow rate into the zone (inflow > outflow in most of
the zones). For the surface layer, the residence time is estimated using the flow rate leaving each
zone (inflow < outflow in most of the zones). The volumes of Table III - 3 are distributed into the
two levels by assigning 50% of the mid-layer volume to the surface and to the bottom,
respectively. The complicated transport in Zone 9 is ignored in this analysis. The hydraulic
residence times for the two layers computed with these rules are shown in Table IV - 2.
                                        IV-2

-------




m
<

£


1UU
30

10
1 3
i '
H 0.3
0.1
0.03
n Al

_^

r ,^^^—
y

i
r \ i
i . i . i . i . i . i ,
INFLOW
OUTFLOW

NET OUTFLOW
.......




               300    250    200   ISO    100    50     0
              Distance (km) Along Main Bay (0 = Mouth of Bay)
Figure IV -1. Longitudinal variation of average annual inflows (from
ocean direction) and outflows (to mouth of Bay)across zone
interfaces. Average year.
      70

    O 60
    $

    5"
    * 40
    §30

    to
    a 10
            300      250      200     150      100       50        0
                 Distance (km) Along Main Bay (0 = Mouth of Bay)



Figure IV - 2. Longitudinal variation of net outflow from zone interface
as a percent of total inflow to interface.
                                IV-3

-------





p<5
g
^
.
w
r )
"SURFA<

IMATES
-2
TIME ESTi
^ bd
1-1 CJ
S S
H cn
K
0
\&
i N
ES 5

s
5?
i
z











•gf
BN TJ
il
|H
o

25?
13-
J"l
p

|t

1"?
Ic.
j
o
N
o •£•
l|l
| '%""
O^H
"»
|!
&


if




Is
^




n
|
N








CS
00
oo




CO
-











o
cs
o^
CN









oo

1
o


00
CO








(S
in
(S^


(S







oo
m
o








r— *






00









in
r-




r-
VO
^



00
in
o






r-^~











00


f-
ON
in








00
00
o"


(N







m
(N








ON
1— M






-









T-H
m
r-~'




-
m



oo
r-,'






ON











f-


00
n
00^


in







£








00
cs






m









ON
ON




ON
OO
in



oo
00






oo
s











m
•4



^
ON








00
0
o"


i
in







CS
00








S






*









1—H
ON
in




oo
oo
00
•o



ON
|^






CS
o"











Tf



O
in








o


5







t-.








fN
f)






en









ON
(N
en




m
m'
•>



^
tC






ON
ON"











(N



m
CO








oC


00







H








oo






(S









3
^H'




00
oo



^
r— <






o
ts"











1



ON
ON








O
(N


(U
O
0
OO






t— I








(N
O






-




















ON
00
V— 4














t
o

~o _
G  « (J

—• 
-------
       As shown in this Table, the approximate cumulative residence time of assumed "Bottom"
flows is about 19 days from the mouth to the head of the Bay and conversely, for the assumed
"Surface" flows, the residence time is more than twice as long. It should be stressed that these are
estimates only for the average year and assuming a simple two layer flow pattern. This
approximate estimate of detention times is used later in analysis of scenario results.

C. MAIN BAY SALINITY - ZONAL AND SEASONAL AVERAGES

       An important aspect of the Bay is the degree of vertical stratification resulting from the
salinity and temperature distribution of the Bay waters. The average seasonal and zonal salinity
values provide an additional overview of the Bay circulation. Figures IV - 3 and IV - 4 show the
longitudinal distribution of the spring and summer average salinity for each zone. The marked
longitudinal increase in salinity in the down-bay direction can be seen with an increase in salt from
spring to summer, as of course would be expected. The vertical difference in salinity is also
marked and increases in difference from the mouth of the Bay to the head.

D. TRIBUTARY INFLOWS, OUTFLOWS, NET FLOWS

       As noted in  the preceding Section III, the flows into and out of the tributary interfaces
form an important basis for calculating the net input of nutrients from tributaries to the main Bay.
Minor tributaries, such as the Patapsco, Rappahannock and York contribute less than the major
tributaries and under several scenarios may experience a small but negative flux.  Figure IV - 5
shows the flows at the interfaces of two major tributaries; the Potomac and James estuaries. For
the Potomac, the principal flow into the tributary is in the deeper portion of the interface while the
outflow is generally in the surface layers. For the James, however, there is a calculated inflow in
the surface  layers in the northern portion of the interface. Outflows are generally in the surface
layers of the southern side of the James. The total inflows and outflows together with the net
flows for each of the principal tributaries are shown in Figures IV- 6 and IV - 7. As seen, the total
inflows and outflows are large for each tributary and the net flow represents the difference
between two large estimates. Thus, for the Patapsco and Back Rivers, the net flow of 5 mVs
represents the difference between total inflow and outflow of almost 900 mVs.
       It should be recognized that these flow results are a function of the degree of spatial detail
at the mouth of the  tributary which is relatively coarse in some instances.  A finer spatial detailing
of the tributary and  its interface with the Bay may result in a different interfacial distribution of
flows, but the net flows, due to flow continuity are believed to be approximately correct.
                                         IV-5

-------
  30


  25
§ 20
o
>
I"
J
^ 10
                                                      SURFACE
                                                        MID
                                                      BOTTOM
        300     250    200     150     100      50      0
          Distance (km) Along Main Bay (0 = Mouth of Bay)


 Figure IV -3. Longitudinal variation of SPRING salinity for Base case.
 Average over zone over season for average year.
  30


  25


§ 20



I"
J
M 10
                                                      SURFACE

                                                        MID
                                                      BOTTOM
        300     250     200     150     100      50
          Distance (km) Along Main Bay (0 = Mouth of Bay)
 Figure IV -4. Longitudinal variation of SUMMER salinity for Base
 case. Average over zone over season for average year.
                            IV-6

-------
                                                     8S


                                                     II
                                                     <;
                                                     «£
                                                     if
                                                     il
                                                     I
                                                     r
                                                     £
               55
               (0 LU





               |l
               oc

               2
                   $
§
   f
                              O)
S
                                CO
                                     CO
                                          co
                                          CO
i*
                       u
CO
co
i
^
$
05
CO

^
fN
in
i-.
•
co
Jo
1

*?
^
CO
t
m
on
en
CO

,_
CO
•
o
1
CO
on
•
^.
^>
OJ
1




o
co1
(^
CO
CM








in
§
Tt
CO
CN





^
$

CO
CO
0
^
8








c
I
^
1
I
1
\








o
n
*
n
*
5








o
3
CO
CM
CM








00
<3>
m
m
CM





                                             h
                                                       \£v
                                                        §
                                                      «K

                                                      5i5
                                                               &


                                                               1
                                                                                                             S
                                                                                       i
                                                                                       rC
                                                                                        i
                                                                                       2
                                                                             y
                                                                            XBQ en XjsinqiJj. ujcuj (§rt;vui) Moy ;»N
                                                                  IV-7

-------
V. CBWQM RESPONSE TO UNIT LOADS OF "TRACER" VARIABLES
A. INTRODUCTION

       In order to better understand the behavior of the CBWQM, a series of "tracer" runs were
made. Since the hydrodynamic transport is complicated and recognizing the role of settling
paniculate nutrient forms, computations that assume a conservative variable or a
non-conservative settling variable provide additional insight into Bay dynamics and help interpret
the even more complicated interactions of various nutrient forms in the full CBWQM.
       The CBWQM model was used with individual non-interacting state variables and was run
for seven years using the "average" hydrology of the scenarios. The following procedure was used
for the tracer variables:
       1. Dissolved conservative tracers are input continuously into the model at the fall lines and
ocean boundary of the nine Bay regions at an arbitrary concentration of 100 mg/L,
       2. Conservative particles settling at the same rate as non-living particles in the calibrated
model (1.0 m/d) are released continuously from the fall lines and ocean boundaries,
       3. Conservative particles are released in the surface of the nine Bay regions at a constant
rate (1 gm/m2 - day) and at a settling rate of 1.0 m/d.
       The results can then be examined to see the fraction of the released substance that is
transported throughout the Bay or settled  into the sediment. Normalizing the results by the input
load of the tracer then provides "unit responses" of these conservative substances.  The range of
the runs from dissolved conservative tracer to a tracer that  is settling at a high rate provide
"boundaries" to what would be expected for tracers that have other behavior. For example, algal
settling in the model varies from 0- 0.25 m/d depending on  the algal group and time of year. Also,
for diatoms, in the spring bloom period (January - June), the net settling into the sediment is set
equal to zero.
       Table A - 2 in Appendix A provides a summary of the results in unit response matrix form.
This matrix is termed the "steady state response matrix". In this form, the columns of the response
matrix represent the water quality response in the Bay regions due to an input from a given fall
line or ocean input. The rows represent the impact in a specific region of an input into all other
regions.
B. RESPONSE TO CONSERVATIVE DISSOLVED TRACER

       Figure V - 1 shows the steady state distribution of a conservative dissolved tracer
discharged at various locations. The concentrations are the volume - averaged concentrations for
a zone. The ordinate is the mg/m3 concentration response in a zone per ton/day of input into the
indicated location. Thus, for a dissolved input into the Bay at the Susquehanna River, the plot
shows the volume averaged unit concentration response for each of the nine zones. Similarly, for
the profile labeled "Pot." (Potomac), the plot shows the response in each of the nine zones for a
dissolved input at the fall line of the Potomac River. The plots in Figure V - 1 are therefore the

                                          V-l

-------
1
   100
    10
f    1
    0.1
Ul
g
«  0.01
  0.001
 sus.
 -B-
 BALT.

 POT.

JAMES

OCEAN
          300    250    200    150    100    50      0
           Distance (km) Along Main Bay (0 = Mouth of Bay)
   Figure V -1. Longitudinal variation of steady state response to
   dissolved inputs at indicated locations. Dissolved input for
   tributaries is at fall lines.
      Sus.        Patuxent        Rapp.         James
           Bait. Hbr.       Potomac         York         Ocean
                   SOURCE OF TRACER INPUT

   Figure V - 2. Steady state response in a given zone to dissolved
   inputs at  indicated locations. Dissolved input for tributaries is at fall
   lines.
                             V-2

-------
Bay columns of the first table in Table A - 2.
       The general observation from this Figure is that dissolved inputs are distributed
throughout the Bay in varying degrees, of course, depending on the location of the input.
 Dissolved input at the Susquehanna River results in concentration response that declines almost
logarithmically down the Bay due to dilution and exchange with the ocean. It can also be seen that
dissolved input at the ocean extends up into the Bay as of course is observed from the behavior of
salinity, a conservative dissolved substance. Discharges of a dissolved conservative tracer at the
fall lines of the tributaries has an equal effect on concentration from the tributary interface down
the Bay. Thus, a conservative discharge at the fall line of the Potomac has a similar effect on the
down Bay zone 6 as a dissolved conservative input from the Susquehanna. Of course, within the
Potomac estuary itself, the concentrations are higher than for the main Bay. The concentration in
the Potomac zone immediately below the fall line (see Table A - 2) is about 16.4 mg/m3 per ton/d.
At Zone 5, at the mouth of the Potomac, the concentration is about 17 % of this maximum
concentration. The input from the Potomac is transported up the Bay by up-Bay flow so that at
Zone 2, the concentration response is almost 50% of the response in Zone 5 where the Potomac
enters.
       Inputs of dissolved conservative tracer at the fall line of the James also impacts the entire
Bay, but at a considerably smaller response. Thus, the response in Zone 8 where the James enters
is only 3% of the maximum response immediately below the fall line in the James (23.9 mg/m3 per
ton/d)  The response in Zone 2 is only about 5% of the response in Zone 8 where the James
enters  This probably reflects the considerable dilution and exchange of the James inputs with the
ocean boundary condition.
       Figure V - 2 shows the response from the point of view of a given zone. That is, this
figure shows selected rows of the first table of Table A - 2. The profile labeled "Zone 2" is the
response in that zone due to inputs in the various indicated locations. Thus, for Zone 2,
5.3 mg/m3 per ton/d is calculated to be the response in that zone due to a dissolved conservative
tracer discharged into the Baltimore Harbor. Similarly, 1.4 mg/m3 per ton/d is calculated in Zone
2 as the response from an input at the fall line of the Potomac. It is clear from this plot that inputs
into the upper Bay to as far as at least the Potomac have an impact on Zone 2. Also, for Zone 6,
as an example of a down Bay zone, inputs from all locations north of the Potomac have a similar
effect on that zone. Inputs  in the James have little effect on the main Bay except for Zone 8, the
zone of exchange with the  James. Finally, the response in Zone 8 is virtually identical from inputs
into all locations.

C. RESPONSE TO CONSERVATIVE PARTICLE TRACERS

       1. Inputs at Tributary Fall Lines
       At the other extreme of the preceding tracer, where all material is dissolved and
conservative and therefore tracks flows and reflects dilution, this tracer represents a conservative
variable (in the kinetic, biodegradation sense) that settles with a velocity of 1.0 m/d. This settling
velocity is the same as that used for non-living particles in the CBWQM.
                                          V-3

-------
0.000001
                             Accumulation below tall line in
                             Potomac         James
          300   250   200    150    100    50     0
        Distance (km) Along Main Bay (0 = Mouth of Bay)
 Figure V - 3. Longitudinal variation of steady state response of
 sediment acumulation flux in a zone due to particle inputs at
 indicated locations. Particle in put for tributaries is at fall lines.
 2      0.1
o:
HI
 §
        0.01


       0.001
      0.0001
 8
 < 0.00001
   0.00000
       Sus
Zone
  2

  4

  6

  8
        Ifsquehanna           Potomac              Ocean
                  Baltimore H              James
                     SOURCE OF TRACER INPUT

Figure V - 4. Steady state response in a given zone to particle
inputs at indicated locations. Particle input for tributaries is at fall
lines.
                            V-4

-------
       Figure V - 3 shows the response profiles in a manner similar to Figure V - 1. the difference
is immediately apparent. The ordinate here however is the kg/d accumulated in the sediment of a
given zone per kg/d input at the various indicated locations. For particle input from the
 Susquehanna, almost all the sediment accumulation is in Zones 1 and 2 immediately below the
 input.  Virtually none of the particles on net reach the sediment in zones below these two upper
 zones. Also, in general, particle inputs  at the fall lines do not significantly influence sediments in
the main Bay. Thus, for the Potomac, the accumulation below the fall line in the estuary is about 1
kg/d per kg/d while in the main Bay Zone 5, is less than 0.1% of the peak below the fall line.
       Figure V - 4 is the analog to Figure V - 2 and shows the sediment accumulation in a given
Zone due to input at other locations. With the exception of the input from the Susquehanna, all
other zones receive little accumulation from particle inputs at the fall lines of the indicated
tributaries.

       2. Inputs at Surface of Bay
       The final set of tracer runs released particles at the surface of each of the zones at a fixed
rate of 1 g/m2-d and with a settling velocity of 1.0 m/d. The results are shown in Figures V - 5 and
V - 6 and are analogous to Figures V- 3 and V - 4. In general, the maximum sediment
accumulation is again in the zone into which the particles are released. For example, 45% of the
particle load at the surface of Zone 6, settling at 1 m/d, accumulates in the sediment of zone 6.
For Zone 2, the sediment accumulates only 0.2% of the particle load released in the surface of
Zone 6.

D. CONCLUSIONS

       Caution should of course be exercised in extending the results from these tracer runs too
far. The recycling  processes both in the water column, the non-steady state nature of the inputs
and the dynamic nonlinear behavior that is inherent in the CBWQM preclude any extensive
generalizations of the tracer runs to the variables in the model. However, some useful conclusions
can be  drawn.

       1. Conservative Dissolved Substance
       The tracer runs using a conservative dissolved substance indicate that the Susquehanna
River and Baltimore Harbor have equivalent unit influences on Zones 2 and 3. Dissolved input
into the Potomac at the fall line influences the Bay both up and down from the entrance of the
Potomac but with major unit influence in Zone 4 and below. The unit influence of dissolved input
at the fall line of the James is limited to Zone 8 where it is believed that most of the James
dissolved loads to the Bay leave the system at the mouth. Although the unit influence of the ocean
is small in the upper Bay, the actual effect may be significant due to the large oceanic load.
Overall, as shown in Figure V-l, the Bay dissolved substance response is indifferent as to where
load  inputs occur upstream in the system.
                                          V-5

-------
                                       Particle input into surface of
         1


|      0.1


*     0.01
oc
£
1    0.001

i
I   0.0001


§  0.00001
X

  0.000001
                      Zone
                       2


                       4


                       6


                       8
              300    250    200    150     100     50      0
              Distance (km) Along Main Bay (0 = Mouth of Bay)
   Figure V - 5. Longitudinal variation of steady state response of
   sediment acumulation flux in a zone due to particle inputs at the
   surface of the indicated zone.
          1


        0.3

        0.1


       0.03


       0.01


      0.003


      0.001


     0.0003


     0.0001
Accumulation in given zone due to —\
particle inputs at surface of all
other zones                  w

                        Zone
                          2


                          4


                          6


                          8
               300     250     200     150     100     50      0

             Distance (km) Along Main Bay (0 = Mouth of Bay)
   Figure V -6. Longitudinal variation of steady state response of
   sediment acumulation flux in a given zone due to particle inputs at
   the surface of all other zones.
                               V-6

-------
       2. Particle Inputs at Fail Lines
       Overall, the  results indicate that for the particle settling rates of 1.0 m/d used for the
tracer, sediment accumulation occurs almost entirely in the immediate vicinity of the input. Thus
sediment loads from the Susquehanna are mostly deposited immediately downstream. Some
 Baltimore Harbor sediments escape to the main Bay but virtually no sediment response is
calculated outside of the tributaries into which the load was input (see Figure V-3). Particles
imported from the ocean are distributed more widely than particles released at the fall lines.
Roughly one third of the oceanic particles are exported from the Bay.

       3. Particle Inputs at Surface of Bay
       For the runs where particles were introduced into the surface of the zone and allowed to
settle at 1 m/d, almost all  of the particles in Zone 5 and above (including the Eastern Shore) are
retained while some fraction of the particles formed in Zone 6 and below are lost. Particles input
at the surface are transported upstream and downstream of the zone in which they are formed.
The particle distribution is roughly bell shaped (see Figure V-5). In Zones 1 and 2, over 70% of
the material formed in the zone is deposited within the zone. The fraction declines with distance
downstream to only about 10% in Zone 8.
       Particles with  settling rates less than 1.0 m/d, such as the phytoplankton groups will of
course be distributed more widely and impact sediments over a wider area, the upper bound of
which is given by the dissolved tracer results. Since the algal group settling rates are considerably
less than the 1.0 m/d (i.e., 0.1 - 0.25 m/d) and are set at zero during certain stages of the
calculation, one can conclude from these tracer runs that algal  particles are probably distributed
widely throughout the Bay system and production in surface layers impact sediments in zones
outside the immediate  production zones.
                                          V-7

-------
 VI. NUTRIENT, PHYTOPLANKTON, CARBON AND SOD RESPONSES

 A. INTRODUCTION
       The purpose of this chapter is to summarize the nitrogen, phosphorus, phytoplankton,
 carbon and sediment oxygen demand responses for the scenarios. The focus is two-fold: (a) How
 the nutrients and resulting carbon fluxes impact the DO under various scenarios, and, (b) How the
 reduction in nutrients impacts living resources habitat, as indicated by DIN, DIP, chlorophyll
 biomass and light penetration.
       In order to set the stage for the subsequent analyses, Figure VI-1 shows the calculated
 longitudinal summer average DO for the Base case. As shown, the minimum bottom DO occurs at
 the head end of the deep trench in Zone #2. The principal reasons for this response under the Base
 case are explored in this chapter and the next  as a means for assessing the impact of the nutrient
 reduction scenarios on the resulting DO.
                  10
                in
                I6
                o
o

8 2
O Z
                'PYCNOCUNE
                                       BOTTOM
                      300    250    200    150    100    60     0
                         Distance (km) Along Main Bay (0- Mouth of Bay)

                    I   1    I  2  I   3 I   4  I   5  I  6  I 7 I  6|
                              BOUNDARIES OF BAY ZONES
                                 (Sw Figure III -1)

                   Hgure VI - 1. Calculated longitudinal variation of DO for BOM ca
                   over summer for average year.
       Comparisons are drawn in this section between the Base case nutrient and carbon
conditions and a choice of several scenarios including the LOT scenarios (LOT N&P, LOT N
only and LOT P Only) and the 40% Controllable scenarios.  These comparisons provide a
bounding of feasible responses and assist in examining the trade-off between nitrogen and
phosphorus removals.
       Throughout this section, Figures are presented as concentration profiles or other variables
as a function of distance along the main axis of the Bay. In most cases, the points plotted
represent the zonal averages for the particular case or season and are plotted at the mid-point of
the zone. Zone 9, the Eastern Shore zone is plotted midway between Zones 4 and 5.
                                        VI-1

-------
B. NITROGEN AND PHOSPHORUS RESPONSES

       1. Nitrogen and Phosphorus Concentrations
       Figures VI-2 and VI-3 show the longitudinal profiles for TN and TP averaged over zone
and season for the Base case. The marked differences in the profiles can be noted, due principally
to a significant input of phosphorus from the ocean boundary (see Cerco and Cole, 1992 for
discussion of this boundary condition). For TN, spring concentrations exceed summer
concentrations throughout the Bay with minimum values generally in the fall.
       For TP, the variation longitudinally and seasonally is relatively small compared to TN
since TP varies by about a factor of two while TN varies by about a factor of four. This is a
consequence of the loading to the Bay from the Susquehanna and other rivers as well as the ocean
(see Section II).
       The spring nutrient conditions are important reference points for subsequent
phytoplankton growth. The March-May profiles (which are similar to the winter) and
June-September profiles for TN and DIN under Base and LOT cases are shown in Figures
VI-4a,b. Figures VI-5a,b focus in on the DIN for a range of scenarios: Base, LOT (N&P), LOT
for TN only and LOT for TP only. The differences between the spring and summer periods can be
noted where the latter season generally has higher nutrient levels than in the spring. The reduction
in the spring DIN from Base to LOT is most pronounced in the lower reaches of the  Bay. It can
also be noted that the DIN becomes a significant smaller proportion of the TN as one progresses
down the Bay. This is partly a result of the ocean boundary condition for TN and DIN. The half
saturation  constant ( a measure of the degree to which the  phytoplankton growth rate is
controlled  by the nutrient) as used in the CBWQM is shown on the plot and indicates that in the
upper Bay the DIN is significantly above this constant and  approaches this constant as one
proceeds down the Bay, showing potential nitrogen limitation in the mid to lower portion of the
Bay.
       Inspecting Figures VI-5a,b,  it can be seen that the LOT P Only scenario increases the DIN
in the upper and mid Bay over Base DIN levels and significantly over the LOT N Only DIN
levels. As will be discussed again later, this is a result of reduced phytoplankton under LOT P
only in the upper Bay allowing down-Bay transport of nitrogen which would otherwise have been
taken up by the phytoplankton. The difference between LOT N&P and LOT N Only  where the
latter  results in lower DIN is therefore another reflection of the Increased transport of nitrogen
down the Bay under phosphorus reductions in the upper Bay.
       Figures VI-6 a,b and VI-7a,b show similar plots for the TP and DIP. Again, summer DIP
is higher than spring levels. For the  spring, the upper Bay DIP is seen to be close to or below the
half saturation constant for phosphorus indicating that the upper Bay is more phosphorus limited
than nitrogen limited. The input of phosphorus from the ocean results in concentrations
significantly above the half saturation constant for the lower Bay. For the summer, DIP is
considerably above the half saturation constant.  Figure VI-7a shows that for LOT N Only, the
phosphorus is increased in the lower Bay over the Base case, apparently because of an analogous
mechanism as noted  above for DIN. That is, with reduced biomass in the mid Bay because of
nitrogen reduction, more phosphorus is transported to the lower Bay, but because the lower Bay
tends  to be nitrogen  limited, the impact of such phosphorus transport is negligible.
                                         VI-2

-------
i  1
E,
3 0.5

  0.3

  0.2


  0.1
                                                        BASE TN
                                                         Winter

                                                         Spring

                                                        Summer

                                                          Fall
         300    250     200    150    100     50      0
          Distance (km) Along Main Bay (0 = Mouth of Bay)
   Figure VI - 2. Longitudinal variation of surface nitrogen. Average
   over zone over season for average year.
   0.1

  0.07
B.0.05
£

o
«0.03
c

  0.02
  0.01
                                                       BASE TP
                                                        Winter
                                                        Spring

                                                        Summer

                                                          Fall
         300     250    200    150     100     50      0
          Distance (km) Along Main Bay (0 = Mouth of Bay)
   Figure VI - 3 . Longitudinal variation of surface phosphorus.
   Average over zone over season for average year.
                              VI-3

-------
 6  o   P  P
        P  p
                          I
                          I
S?    8

  §    |

?!    ;

  °    o
si    §
                          if
                          15
                                                          B.
                                                        «?   T
                                                        o   o
                                            8  P
                                            P  O
                                                        (I/Bui) uo|]BJ)unuo3
                                                                                    *
                                                                                  S
  SB

  -5




8%
*- n

  S


a!
        s  §
(n/flui) uonBjyj*9uoo

      II

      I?
      SP~
      9&
                                                 1!
                                   z|° i O i
                                   i-Tz i °- ,

                                   2  5 V
                                     °  °"   I  §
       !

j    !
                                                                           s
                                                                            "
                                                                           8*
                                                              P
                                                                                   .f


                                                                                    *

-------
8    £!  5.
O    O  O
8    §  8
o    o  o
                            •      I

                            •!    !«

                              i    »S
                            si    is
                                    To §
                                    C M
       (T/ftju) uoflBjjueouoo
                                                                   Q_
                                                                   «J
  Q   q
                                                             8   5
                                                             o   o
 3    8   8   j
 o    o   o   '

(I/Bui) UOIIBJ»U»OUOO
I|§|3'§
t 'P=Ta !5

                            8
                                    1?

                                    s&
                                    € t:
                                                                       -
                                                                   -1  O   Q
                                                             o   o   o    99
                                                             b   b   P    P   P
                                                                      o    oo
                       I
                     81
                                                                                          §1
                                                                                            3.

-------
       2. The DIN/DIP Ratio
       The ratio of DTN4o DIP is an important measure of whether nitrogen or phosphorus is
important in controlling phytoplankton growth. Ratios significantly greater than about 7-10 on a
mass basis indicate a tendency toward phosphorus limitation while ratios significantly less than
that range tend to indicate nitrogen limitation. Figure VI-8 shows the DIN/DIP ratio for the base
case averaged over zone over season. The "Redfield Ratio" of 7.2:1 on a mass basis is shown.
During the average spring conditions for this average flow year, the Bay is phosphorus limited
from the head to about 75 km from the mouth. During the summer, more of the lower Bay
becomes nitrogen limited and during the fall average conditions, more than half of the Bay is
nitrogen limited.
       Figure VI-9 shows that the general tendency is for the LOT N&P and LOT N Only
scenarios to increase the region of the Bay that is nitrogen limited. The LOT P Only however,
decreases the region of nitrogen limitation because of increased nitrogen transport to the lower
Bay.
       Figure VT-10 displays the DIN/DIP ratio in a similar manner as the preceding figure but
for the summer condition. Again, reducing the nutrient input increases the down-Bay region of
nitrogen limitation and the effect of the LOT P Only is again to decrease nitrogen limitation while
the LOT N Only increases N-limitation significantly over the LOT N&P case.
       These results are in general agreement with the work of Fisher et al., 1992 who present
data and analyses to support the hypothesis that during the spring, phosphorus (and silica) limit
growth while N limits growth during the summer. Their data for August 1987 show a
considerably larger region of the estuary being controlled  by nitrogen than shown here for summer
conditions. This is probably due to a combination of effects: e.g., the relatively long averaging
period used here for "summer" (June - September), and the differing fresh water inflow
hydrographs and associated nutrient loading.

       3. Nitrogen and Phosphorus Fluxes
       A second mode of interpreting the scenario results is to examine the changes in the fluxes
of the principal processes of nutrient inputs, exchanges and sediment interactions. Figures VI-11
through VI- 16 show the nitrogen fluxes for the main Bay. Total external load is the load from fall
lines, below fall lines, point sources and direct atmospheric deposition to the Bay (See also Figure
II-1). Net ocean flux is the net exchange at the mouth of the Bay. The settling flux is the gross
settling to the sediment of the Bay. Diffusive flux is the net exchange of dissolved nutrient forms
across the sediment-water interface. Net sediment flux is the difference between gross settling and
diffusive flux. Denitrification flux is the loss of nitrogen due to the conversion to nitrogen gas,
primarily in the sediments. Finally, the burial flux is the net loss of nitrogen from the bottom
sediment segment of the model. All fluxes are given in areal units. It should also be noted that
these fluxes are for the average year (year #9) of the variable hydrology sequence and as such
reflect a flux "snapshot". It can readily be observed that all fluxes do not necessarily add up to
zero because of the dynamic non-steady state nature of the computation.
                                          VI-6

-------
882"
  OIIVH dio/Nia
                1
               s,   »i
               "a   *S
               8*   I!
               ''   II
i*0
 I
n
               81
                    Ii

 ouva dio/Nia
                                     ilisisist
                                     oa's^'b1
                                     I 1  I
                                                         §
                                                        §1
                                                        Si
                                           OIIVM dia/Nia
                                          o
                                          i
                                          5 ^
                                          U
                                          € «
                                          3?
                                          *S
                                      3   5 i
                                 	      §2
                                  »  5      f|

-------
3
o
ii
   la
   : CM
R COLUMN
            ii.
             i a
             ;d
             15?
             i?-
             iS

             5 3?
             "iS
o
ii
3

I
m
                       eo
ll
ii
|1

1
N
                        I
i
Sfc
8!
                                   8
                                   rfS
                                   i ii
                 1?
                 I-
                 £ CO
                                                            J
                                                     \li
ive Flux

24.4(33%)
                                                It
                                                f «

                                                            1
                                                            3
s
ct
 i
          |
                 i to
                rv
             i «
             IS
                         o
                         X
                         c
                         S
                         i
                         &
                                             I
                                                    K
                                                    :3 — .

                        5*
                         5.
                                                            5
                                                            s
                                                            5

                                                            I
                                                                  oo

-------
Units: mo N
    m*2-day
(10%) = % DECREASE
FROM BASE CASE
TOTAL NITROGEN FLUX
       MAIN BAY
SCENARIO #2 - 40%
CONTROLLABLE
                         Total External Load
                          +42.0(18%)
   Figure VI -15. Total Nitrogen Flux and % Decrease from Base Case for
                      Main Bay - Average Year
Units: moN
    m*2-day
(10%) = % DECREASE
FROM BASE CASE
TOTAL NITROGEN FLUX
      MAIN BAY
SCENARIO #16 -90%
N & P REMOVAL
                         Total External Load
                           +4.6(91%)
                                                   (% Change > 100% = Input
                                                   into Bay from Ocean)
   Figure VI -16. Total Nitrogen Flux and % Decrease from Base Case for
                      Main Bay - Average Year
                                VI-9

-------
       The net nitrogen flux to the ocean occurs in all scenarios except the 90% N& P removal
(Figure VI-16). For the feasible loading reduction scenarios, the net burial of nitrogen is
 approximately constant over the scenarios. An increase in denitrification flux over Base Case is
calculated for both LOT N&P and LOT P Only.
       For phosphorus, a net input from the ocean occurs, and for all the scenarios in Figures VI
- 17 through VI-22, the ocean input flux is substantial and exceeds the Base case flux. Burial flux
is again approximately constant across the scenarios. The influx of phosphorus from the ocean has
a significant impact on the fluxes, which can be seen by comparing, for example, the net flux of
nitrogen to the sediment with that of phosphorus. For nitrogen, the net sediment flux is about
50% of the external load  while for phosphorus, the net flux to the sediment is about 100-150% of
the external phosphorus loading. The extra phosphorus is calculated by the CBWQM to be fluxed
into the Bay from the ocean.
       4. Nitrogen and Phosphorus Responses Interpreted in Terms of SAV Goals
       The submerged aquatic vegetation (SAV) habitat requirements include a goal attaining of
less than 0.15 mg/1 dissolved inorganic nitrogen (DIN) for mesohaline and polyhaline present and
potential locations of SAV colonization during the annual growth period of the vegetation (see
Table 1-2 of Section I). The extent of any model run in meeting the DIN goal can be tabulated,
and then normalized to the base case scenario, so that all runs can be designated a percent
improvement over base case (with base case as zero improvement, and complete compliance with
the DIN goal as 1.00 = 100% improvement). For uniform baywide nutrient reductions the
achievement of the DIN goal can be visualized as the response surface shown in Figure VI - 23
(See Section III - C for details on generation of response surfaces). This Figure shows DIN goal
achievement (percent improvement over base case) on the vertical axis as a function of nitrogen
and phosphorus load reductions (also expressed as percent improvement over base case) on the
horizontal plane. As expected, the model predicts that the DIN goal responds strongly to
nitrogen removals. Previous discussion of scenario results have indicated the downstream
transport of nitrogen from phosphorus removal  scenarios. The response surface DIN figure
illustrates this result by indicating that phosphorus reductions caused the model to predict that
DIN concentrations would increase somewhat in SAV habitat areas. DIN goal in this figure is for
the cumulative effect during an average year for the entire Bay. Though seasonal and local effects
in many cases are quite different from the overall total response, the total response indicates the
dominant trend.
       The dissolved inorganic phosphorus (DIP) goal for SAV is that for critical growing
periods DIP concentrations must be less than 0.02 mg/1 in actual and potential tidal fresh,
oligohaline, and polyhaline SAV zones, and less than 0.01 mg/1 in mesohaline SAV zones. (Table
1-2). This goal was met 100% by the base case  scenario.
                                         VI-10

-------

<
>,
U
I
o.
i
r*
^
i
5

D)

                  Is
Flux
-42%
N«(
10
                in w
                ?!
ain % Cha
as« m Fhjx
                           J
                               1?
                                              s
                                              ?.
                                              *«&
                                              S.E

-------
Units: ma P
    m*2-day
(10%) = % DECREASE
FROM BASE CASE
TOTAL PHOSPHORUS FLUX
        MAIN BAY
SCENARIO #2 - 40%
CONTROLLABLE
                          Total External Load
                           +2.3(31%)
                                        •\7
                          WATER COLUMN
Settling Flux
3.3 (15%)
i

Diffusive Flux
0.8(11%)
J
i
*
Net Sediment Flux
2.5 (16%)
i
i
>
^
                                                    (Negative % Change =
                                                    Increase in Flux from Base)
                            SEDIMENT
                    t
    Burial Flux
    -3.0 (5%)
   Figure VI - 21. Total Phosphorus Flux and % Decrease from Base Case
                     for Main Bay - Average Year
Units: mfl P
    m*2-day
(10%) = % DECREASE
FROM BASE CASE
TOTAL PHOSPHORUS FLUX
        MAIN BAY
SCENARIO #16-90%
N & P REMOVAL
                          Total External Load
                           +0.3(91%)
                                                       Net Ocean Flux
                                                       +1 .2 (-20%)
                                                    (Negative % Change =
                                                    Increase in Flux from Base)
   Figure VI - 22.  Total Phosphorus Flux and % Decrease from Base Case
                     for Main Bay - Average Year
                                 VI-12

-------
     REDUCTION IN
     POOR DIN HABITAT

       A
    0.25
    0.00
     1.00
1.00
Figure VI - 23. Response surface analysis of dissolved inorganic nitrogen (DIN)

                      over whole Bay, average year.
                                 VI-13

-------
C. PHYTOPLANKTON RESPONSE

       1. Base Case and Comparison to LOT Scenarios
       Following the analysis of N and P behavior under the Base case and different scenarios,
the response of the phytoplankton biomass can now be examined. Since it was concluded from the
preceding analysis that the Bay is phosphorus limited in the upper reaches and nitrogen limited in
the lower reaches, it is appropriate to check whether the phytoplankton response is calculated to
be consistent with this nutrient behavior. First, Figure VI-24 shows the longitudinal and seasonal
average phytoplankton for the Base case on a ug chlorophyll/L basis. Peak concentrations are
calculated for the spring followed by a general decline in the summer and fall and an increase in
the winter in preparation for the spring bloom. Figure VI-25 shows the chlorophyll biomass on a g
chlorophyll/m2 basis and indicates that peak areal biomass occurs in Zone #4  region (Patuxent to
Potomac zone). Summer biomass is relatively constant with distance down the Bay. This is in
contrast to the concentration plot which indicates peak concentrations of phytoplankton in Zone
#2.
       Figure VI-26 shows the distribution of spring surface chlorophyll on an areal basis for
several scenarios consistent with the nutrient analyses. The impact of phosphorus removal or
nitrogen removal is clear. LOT P Only results in biomass  comparable to LOT N&P for the upper
Bay but increases the biomass in the lower Bay above LOT N&P.  Indeed, the LOT P  Only
results in chlorophyll biomass in the lower Bay at approximately the same levels as the Base case.
On the other hand, LOT N Only has little effect in the upper Bay (recall that this scenario does
include some P reduction) while in the  lower Bay, LOT N Only is more effective in reducing
biomass than LOT N&P. These results reflect the nutrient transport issues discussed above.
       Figure VI-27 through 30 show  the chlorophyll response as a percent reduction from the
Base case for the four seasons. Beginning with the winter, the effect of P only and N only is seen
clearly where the upper Bay reductions for P only are equivalent to LOT N&P while the lower
Bay reductions are comparable to LOT N&P only when N is removed. Note LOT N&P is
approximately constant along the Bay because of the effect of P in the upper Bay and N in the
lower Bay. The next figure for spring shows a similar picture although here the LOT P only
actually results in no change in biomass or a slight increase in biomass in the lower Bay.
The summer profile (Figure VI-29) shows a higher removal in the upper Bay for LOT  N&P while
the N only case is approximately constant throughout the Bay. The percent reductions for the fall
season (Figure VI-30) show a dominance of N only behavior indicating that the LOT N&P
response is due principally to N reduction except for the vicinity of Zone #2.
       It is concluded from this analysis that chlorophyll biomass is controlled in the upper Bay
by phosphorus and therefore reductions in phosphorus are necessary to reduce biomass in that
region.  The mid to lower portion of the Bay is controlled by nitrogen and nitrogen reductions are
necessary to control biomass in that region. Control of either nutrient by itself at LOT  levels
would not be as effective as controlling both.  Indeed, it is calculated that controlling only P at
LOT levels would not result in any improvement in the lower Bay (and may degrade the lower
Bay) because of increased nitrogen transport down to that area. Similarly, controlling only N at
LOT levels has substantially less impact on the upper Bay biomass than P removal. Controlling
both nutrients at LOT levels results in consistent overall biomass reductions.
                                         VI-14

-------
                       I
                      « ID

                      P
                      I J
                 •     fi
                  f   II

                 at



                 •I

                 81
(Z»«upow9 B) uon«|U*ouoo
                      53
                      « M


                      III
                                                                           in
                 J
                 S-6
      Is
      L8
      1
       m

       I
it   |i
 a   |S
So   5 o
 1   ]l
§!   1!
J   !l
                §i
         O  10   O
     *?i
     5&
     |S
     II
                                               6 5

                                                            §
                                                                  •c •
                          S W
                          il

                          $1
                          532
N    «>    T-    u
     5    °    1
       ZvUl/MOIHO B

-------
  -

O O
 ?   *£

11f
§!   **
 a   IN
     Is

     iS
               s
  esegiucujuofpnpeH*
                    *H
                    J5&
                    ^ «J ^!
                                             85°

                                          •reguioyuqpnpey %
                                                        82

                                                         a

                                                        sl
                                                     §
                                                          i*

                                                           i
                                                          1 1
                                                            8*
Ml*
 8   8
    ujojjuoipnp«a%
                    °
                s   5 a

               81   II


                    II
                      8,
                                      B'B
                                   988
                                           I!


                                           ^
                                           % i

                                           I*
                                            $
                                           tsl
                                           S •>
                                           i "*

                                           ^
                                           ii
                                           Is
                                           ??.
                                           alii
                                           »f&
                                                          sJI
                                                               I
                                                             3 ii ra


                                                             ^ll
                                                                  CO

-------
       2. Biomass Reduction Comparisons of "40% Controllable" Scenarios to LOT N&P
       The preceding results can also be compared to the 40% Controllable scenarios (#2,#3 and
#4). Figures VI-31  and 32 show the percent reduction of phytoplankton biomass by comparing
these three scenarios to LOT N&P. For the spring period, 40% Controllable reductions in
biomass are about two-thirds of those for LOT N&P and for Scenario #4 (40% Cont. +CAA +
Basin) the mid to lower Bay response is comparable to LOT N&P. The upper Bay reductions are
however about the  same as 40% Controllable. For the summer reductions (Figure VI-32), similar
responses are calculated.
       The 40% Controllable scenario without CAA and all basin controls results in biomass
reductions from Base from about 0-20% for the spring and from about 10-20% in the summer.
Additional controls beyond the basic 40% Controllable have maximum impact in the mid to lower
Bay with some further improvement (over Scenario #2) in the upper Bay.

       3. Effect on Light Penetration
       The effect of the preceding reductions in phytoplankton biomass on light penetration can
also be evaluated.  Such an effect is important for protection of the Submerged Aquatic
Vegetation where the emphasis is on plants in the more shallow regions of the Bay. In the analysis
that follows however, it should be stressed that responses in light intensity and light extinction are
averaged over a zone and over a season. The results therefore do not necessarily refer directly to
the shallows of the  Bay but are only a zone wide indication of changes in light penetration.
       The general equation for light penetration is given by
                                I = I0e~keZ

for I and I0 as the solar radiation at depth z and at the surface, respectively and for ke as the
extinction coefficient (1/m). The CBWQM calculates the light extinction coefficient as a function
of incoming river flow (as an assumed relationship to incoming suspended solids), added to a
background level and a linear relationship of extinction to phytoplankton chlorophyll
concentration (see Cerco and Cole, 1992 for complete discussion). Briefly, the light extinction is
composed of three components:
                               ke =
where keb is a minimum extinction coefficient ( a function of Bay location), k^ is the extinction
coefficient due to suspended solids as related to incoming river flows and k^ is the extinction
coefficient due to the phytoplankton. The latter coefficient is linearly related to the phytoplankton
chlorophyll (ug/L), Ch, as

                                  = 0.017Ch
                                        VI-17

-------
                    SEASON: MARCH-MAY
ID

CO
30


20


10
  -10
   LOT N&P

  40% CONT

  40%+CAA

40%CAA+BASIN
        300   250   200    150    100   50     0
       Distance (km) Along Main Bay (0 = Mouth of Bay)
   Figure VI - 31. Longitudinal variation of percent reduction of
   surface chlorophyll on g/mA2 basis.  Average over zone over season
   for average year.
   40
   30
CD
   20
o
?
tx.
#
   10
  -10 u
                    SEASON: JUNE -SEPT.
                                                LOT N&P

                                               40% CONT

                                               40%+CAA
                                                 --e—
                                             40%CAA+BASIN
        300   250   200    150    100    50     0
       Distance (km) Along Main Bay (0 = Mouth of Bay)
   Figure VI - 32. Longitudinal variation of  percent reduction of
   surface chlorophyll on g/m*2 basis. Average over zone over season
   for average year.
                            VI-18

-------
       As incorporated in the CBWQM therefore, the part of the extinction coefficient that is
controllable by nutrient reduction of phytoplankton is directly proportional to that biomass. The
phytoplankton biomass reductions shown in Figures VI - 27 through 30 can therefore be
interpreted as reductions in light extinction of that portion that is related to the phytoplankton.
       Figure VI-33 shows the reduction in total light extinction coefficient for three reference
scenarios. As seen, the LOT N&P case results in light extinction reductions of up to about 12%.
This reduction as shown is entirely due to the reduction of phosphorus in the upper Bay as
indicated by the result that the LOT P-only run is identical to the LOT N&P case. The small %
reduction in the lower Bay is seen to be entirely a function of the nitrogen removal as shown by
the LOT N Only. Note that the removal of phosphorus alone is calculated to result in no change
in the light extinction or even a slight worsening of conditions in the lower Bay.
       If attention is directed to the change in light intensity at a fixed depth, z, then the following
equation is relevant:
                                 I',

where I, and I2 are, respectively, the ratio of light at depth z to surface light for a reference
scenario (i.e. Base case) and a nutrient reduction scenario and similarly, kel and ke2 are the
associated extinction coefficients for the respective scenarios. Figure VI - 34 shows the %
increase in light intensity from Base case for a depth of 2m. Again, the strong influence of
phosphorus removal in the upper Bay is evident while the importance of nitrogen removal for the
lower Bay is also indicated.

       4. Primary Production Response
       Primary production of the phytoplankton is an important variable reflecting the net
increase of the phytoplankton areal biomass per unit time. Figures VI-35 through VI-37 show the
Base case seasonal variation of the primary production and chlorophyll for three zones of the Bay:
Zone 2 in which the minimum bottom DO occurs, Zone 4, a transition region and Zone 6,
representing a down-Bay area. Several interesting points emerge. For the upper Bay zone 2, peak
biomass is in the spring whereas peak primary production occurs in the summer. Zones 4 and 6
indicate a similar pattern although less evident than for the upper Zone.  The spatial gradient in
production can also be noted where during the summer, Zone 2 production is at about 1 gC/m2-d
while for Zone 6, the production during the same season is about two-thirds less. Since the
biomass is virtually constant over the three Zones,  this would tend to indicate that the net growth
rate of the phytoplankton is impacted in the lower Bay, presumably by limitations of nitrogen.
       Figures VI-38 through VI- 40 show the percent reduction in primary production for the
LOT, LOT N Only and LOT P Only scenarios. For Zone 2, the reduction is controlled entirely by
phosphorus in the winter and spring whereas in the summer, the production is controlled equally
by nitrogen and phosphorus. In the fall, nitrogen is more controlling than phosphorus. For the
mid-Bay Zone 4, phosphorus controls in the winter and spring whereas nitrogen is the controlling
nutrient for the other two seasons. Indeed, it can be noted that for the summer in Zone 4, LOT P
                                         VI-19

-------
                  SEASON: MARCH-MAY
                                                   LOT N&P

                                                  LOT N ONLY
                                                     •••*-••
                                                  LOT P ONLY
     300    250    200    150    100     50     0
     Distance (km)  Along Main Bay (0 = Mouth of Bay)
Figure VI - 33. Longitudinal variation of percent reduction of light
extinction coefficient. Average over zone over season for average
year.
SEASON: MARCH - MAY
£40
S
0
E
CM
§>30
«
1
§ 20
£
3,0
C
1
I 0
0 0
.zz
*


INCREASE IN LIGHT IN
UPPER BAY DUE _
ENTIRELY TO 	 7 M
PHOSPHORUS / / \\
REDUCTION / / \\

\ I •••••*••
\ \ 5K-.
Mn liJDDnVrT"UCMT HI IP \ 4 A A "*'-V
TO NITROGEN / \ ^*
REDUCTION / \/^v

i . i , i . i . i . i
LOT N&P
— + —
LOT N ONLY
•••*• •
LOT P ONLY
-B-







     300    250    200    150    100     50     0
     Distance (km) Along Main Bay (0 - Mouth of Bay)
Figure VI - 34. Longitudinal variation of percent increase in light @
2 meters.  Average over zone over season for average year.
                           VI-20

-------
 ZONE
            0.01

PRPRODjCAn*2-dC
CHLORuCNorAn'20
JAN-FEB
0.21
0.03
MAR-MAY
0.56
0.10
JON-SEPT
O.M
0.05
OCT-NOV
O.M
0.03
Figure VI - 35. Variation of primary production and average
phytoplankton blomass, Base Case. Average over zone over
season for average year.
ZONE *2 % REDUCTION IN PRIMARY PROD.
 801	
 40

 20

  0


-20

-40
                                                                                                                                         DLOT
                                                                                                                                         BLOTN
                                                                                                                                         • LOTP
                                                                                             JAN-FEB
                                                                                                        MAR-MAY   JUN-SEPT   OCTMOV
 Figure VI - 38. Variation of % reduction from Base of phyto
 plankton primary production. Zone #2. Average over zone over
 season for average year.
 ZONE #4
             0.5
             0.3
             0.2

             0.1

            0.05
            003
            0.02

            0.01

PPRODgC/m»2-dQ
gChlor/m«2 ^
JAN-FEB
0.4»
0.11
MAR-MAY
o.ei
o.ia
JUN-SEPT
0.64
0.06
OCT-NOV
0.32
0.05
Figure VI - 36. Variation of primary production and average
phytoplankton blomass, Base Case. Average over zone over
season for average year.
ZONE *4 % REDUCTION IN PRIMARY PROD.
 BO
 40


 20

  0


-20

-40
                                                 DLOT
                                                 HLOTN
                                                 • LOTP
                                                                                             JAN-FEB
                                                                                                        MAR-MAY   JUNSEPT
                                                                                                                             OCTHOV
 Figure VI - 39 . Variation of % reduction from Base of phyto
 plankton primary production. Zone *4. Average over zone over
 season for average year.
 7.ONE #6
             0.5
             0.3
             0.2

             0.1

            0.05
            0.03
            002

            0.01

PRPROOgC*n»2-dn
CHLOR(jChlO(*l«2ES
JAN-FEB
0.06
0.11
MAR-MAY
0.13
0.14
JUN-SEPT
0.34
0.05
OCT-NOV
0.10
0.04
Figure VI - 37. Variation of primary production and average
phytoplankton blomass, Base Case. Average over zone over
season for average year.
ZONE #6 % REDUCTION IN PRIMARY PROD.
 60
                                                                                              JAN-FEB    MAR-MAY   JUN-SEPT    OCT-NOV
  Figure VI - 40 . Variation of S reduction from Base of phyto
  plankton primary production. Zone 1M. Average over zone over
  season for average year.
                                                                   VI-21

-------
 Only results in virtually no change in production over Base case. Finally, for Zone 6, the impact
of downstream transport of nitrogen to the nitrogen poor regions of the Bay is immediately
apparent. For the winter and spring seasons, LOT P Only results in an increase in production over
Base case due to this down Bay transport of nitrogen. In the summer and fall, this effect is less
pronounced because of the relatively lesser impact of phosphorus reductions in the upper Bay
regions during these periods.
       These results from the LOT scenarios provide further evidence of the calculated down
Bay transport of nitrogen by LOT phosphorus load reduction. Such increases in nitrogen increase
primary production in the lower nitrogen limited regions of the Bay and as will be seen in the next
Section, have a proportional less impact on the DO of the bottom waters of the Bay. On the other
hand, phosphorus load reductions have a positive  impact in the upper Bay zones where the system
is phosphorus limited.
       Figure VI - 41 displays the reduction in primary production for a range of selected
scenarios. For the 40 % Controllable scenarios (#2 - #4), it is seen that as the load is increasingly
reduced for these scenarios, the impact on the primary production approaches the LOT case. The
annual averages however tend to mask the actual  seasonal dynamics of primary production as
discussed in the preceding paragraphs. Thus, the LOT P only annual average reduction is only
10% whereas the spring reduction is over 40%. The impact of reductions in the mid Bay areas
(LOT-mid) is also shown to be a significant part of the overall reduction although again seasonal
variations may mask the impact of reductions in the upper Bay regions.

D. CARBON RESPONSE

       1. TOC and Net Carbon Settling to Sediment
       Analysis of the organic carbon concentrations and fluxes aids in the interpretation of the
scenarios since this variable has a direct impact on the DO.  The Total Organic Carbon (TOC),
that is,  algal biomass, dissolved plus particulate labile and particulate refractory carbon,  for
bottom waters and the Base case is shown in Figure VI-42  for the four seasons. Interest is
centered  on the bottom TOC because of the direct relationship to the DO in those waters. A
general down Bay longitudinal gradient is calculated except for winter when carbon is essentially
constant with distance.
       Figure VI-43 shows the percent reduction in bottom TOC during the spring for the LOT
scenarios. It is seen that for the LOT- P only case, the bottom TOC in down Bay waters is
increased over LOT of N&P perhaps indicating the effect of increased production in the surface
waters  during that period. It can also be noted that the bottom TOC for the LOT - P Only case is
higher than LOT N&P over a greater distance up  the Bay.  This may be due to up-Bay advective
transport of TOC from the lower Bay. Figures VI- 44 through VI-46 show this behavior more
clearly. In these figures, the percent reduction in the TOC  net settling to the sediment is shown
for the  four seasons and three zones. For Zone 2,  the reduction in carbon settling is largely a
function of the reduction in phosphorus loading with the exception of the fall period. For Zone 6,
however, the carbon settling is reduced primarily as a result of nitrogen reductions. Indeed, the
LOT P Only scenario is calculated to result in a small increase in carbon net settling during the
spring over Base case. For the other seasons, the reduction in phosphorus has a minimal effect on
the carbon settling in Zone 6.

                                         VI-22

-------
100
                              SCENARIO

     Figure VI-41. Reductions in Primary Production for Selected Scenarios.
                              Average Year
                               VI-23

-------
o
S3
a
                                                    BASE TOC
                                                      Winter
   Spring

  Summer

    Fall
       300    250    200    150    100     50      0
        Distance (km) Along Main Bay (0 = Mouth of Bay)
   Figure VI - 42  . Longitudinal variation of  bottom TOC. Average over
   zone over season for average year.
    5
  4.5
    4

i3-5
£
§   3
5=

I"
  1.5
                    SEASON: MARCH - MAY
   TOC
   BASE
 LOT N&P

LOT N ONLY

LOT P ONLY
        300    250   200    150    100    50     0
        Distance (km) Along Main Bay (0 = Mouth of Bay)
   Figure VI-43 . Longitudinal variation of bottom TOC. Average over
   zone over season for average year.
                            VI-24

-------
CM
5

-------
        55?
        si
        §8
        19
o
I
I
                        ll
                           5
                           5
LLl
        las
    i 0
    I +

I
                           I
                           i
                                        i

-------
Units: a C
    mA2-day	
(10%) = % DECREASE
FROM BASE CASE
ORGANIC CARBON FLUX
       MAIN BAY
                          Total External Load
                             + 0.02
SCENARIO #2 - 40%
CONTROLLABLE
              Primary Production
               0.33(14%)
                          WATER COLUMN
                            Settling Flux
                           0.32 (14%)
      Figure VI - 51. Organic Carbon Flux for Main Bay- Average Year
Units: g C
    mA2-day	
(10%) = % DECREASE
FROM BASE CASE
ORGANIC CARBON FLUX
       MAIN BAY
                          Total External Load
                             + 0.005
SCENARIO #16-90%
N & P REMOVAL
              Primary Production
               0.03 (92%)
                          WATER COLUMN
                            Settling Flux
                           0.07(81%)
                                                    (% Change > 100% = Input
                                                    into Bay from Ocean)
     Figure VI - 52. Organic Carbon Flux for Main Bay - Average Year
                                VI-27

-------
       2. Bay-Wide Carbon Fluxes
       Figures VI-47 through VI-52 show the Bay-wide carbon fluxes for selected scenarios and
include the Base case fluxes as well as the percent reductions from the Base Case. It is
immediately clear that the internal primary carbon production dominates the loading of carbon and
that virtually all of the carbon so produced is retained in the Bay and is input to the sediment.
Also of the carbon flux to the sediment, about 80% is calculated to be diagenetically degraded.
Overall for the Bay, the percent  reductions in fluxes are not sensitive to the LOT scenarios. For
the 40% controllable scenario, carbon fluxes are generally reduced by about 14% across the key
flux elements including the diagenesis flux. Burial fluxes of carbon are approximately constant
across the scenarios. Scenario #16, the 90% N&P removal case is the only scenario calculated to
result in an input of carbon from the ocean.

E. SEDIMENT OXYGEN DEMAND RESPONSE

       The demand of the sediment for oxygen is calculated by the sediment sub-model of the
CBWQM and is described in detail in Di Toro et al., (1992). The water column is coupled to the
sediment model through the settling of paniculate nutrients. The sediment oxygen demand (SOD)
is calculated using the net carbon flux to the sediment as the primary input loading.
       Figure VI-53 shows the monthly variation in the net carbon flux to the sediment for the
three zones discussed earlier in the carbon flux analysis. Maximum loading to the sediment is
during the summer months and is the highest in the upper Bay zones. Peak values in this region is
about 0.8 gC/m2-d.  The percent reductions for the three zones shown as a function of the time of
year are shown in the succeeding three figures (VI-54 through VI-56). (Reference should also be
made to Figures VI -44 through VI-46 which show similar plots but by individual zone.) The
most notable feature of Figures VI-54 through VI-56 is the dramatic effect of LOT P Only versus
LOT N Only. For the former case (VI-56), phosphorus removal has a maximum impact in Zone 2
throughout the year whereas the impact on Zone 6 is to increase (over Base) the carbon settling
to the sediment in the spring and during the summer decrease the settling over Base by only about
5%. On the other hand, LOT N only has its maximum effect in Zone 6 up through July and then
equally affects the three zones after that. These results are yet another reflection of the preceding
discussion indicating the effect of down Bay transport of nitrogen which now is  seen to
significantly affect the net carbon flux to the sediment. Given this carbon sediment flux behavior,
it is now important to examine the resultant behavior of the SOD.
       Figure VI-57 shows the variation in the SOD for the Base case across the zones and for
the four seasons. Maximum SOD is calculated to occur in the summer and in Zones 3 to 6. This
is in contrast to the carbon flux to the sediment which is maximum in the upper Bay Zone 2
region. The lower SOD in Zone 2 may be a result of the periods of zero DO in Zone 2 during
which there is zero SOD thereby lowering the overall average. On the other hand, the difference
may be related to the labile and refractory components of the carbon used in the model.  The fall
line particulate loadings are considered to be all refractory while the point sources are assumed to
be 70% labile and particulate carbon produced from phytoplankton is assumed to be 55% labile
(Cerco and Cole, 1992). Thus, while the sediment of the upper zones receive more carbon, the
nature of the carbon is largely refractory in contrast to the middle and lower zones where the

                                        VI-28

-------
  
-------
 • «
  s
i|
8888!??*°
 asva noud oos do Noaonasax
     8*
     !»
                           *
     8
§i   *i
                                             liMi
                                             j-fz ' °- 7
                                             §  §'§
                                 y
                                                                  i   !i
                                                                 1   P
                                                                 »i   j|
                                                                 si   i"
                                                                 s|   fi
        If
                        I  g  £  g  »  O  «

                         3SVB WOUd OOS dO NOIiOnQ3W%
                                                           O
                                                           CO
                     1
                     §1
                           il
                           fi
                           B'B
                                          f
     /
     i
     i>
>  /
                                            8
                                               .
                                            si
                                            'i
                                            if
                             8    8     S
                         3SVB W0«d COS dO NOIiOD03«%

-------
 carbon results from primary production and is considerably more labile.  Since the calculated
diagenesis rate in Zones 3 and 4 is higher than in Zone 2, one concludes that the variable carbon
fractions has an effect on the SOD in Zone 2 and together with the periods of zero DO is
contributing to the lower calculated SOD in that Zone.
       Therefore, it appears that maximum SOD levels are calculated for the middle Bay region
as a result of the variable labile and refractory nature of the external and internal carbon loadings
to the Bay. The effect of load reductions therefore would be expected to have varying influences
on the SOD (and hence DO) as a result of the variable carbon fluxes and degree of carbon lability.
Figures VI-58 through 60 show the percent reduction in SOD from the Base case for the LOT
scenarios for three seasons. Directing attention to the summer season, it is seen the LOT P Only
has a significant effect on the SOD  in the first three zones, but has relatively little effect on the
SOD in the middle and lower zones of the Bay.  In contrast, the maximum reduction in SOD in
those regions is due to nitrogen reduction. Indeed, one can see again that for LOT N Only, the
reduction in SOD is higher than for LOT N&P,  due presumably to the impact of nitrogen
transport increases for the latter scenario. Although, the percent reduction in SOD in the upper
three zones is controlled by phosphorus, the SOD is relatively small in the first two zones, so that
the maximum impact of the reduction is considerably more in Zones 3 through 6.
       Figure VI-61 shows the percent reductions in SOD for the 40% Controllable scenarios (#2
- #4) in comparison to the LOT N&P scenario. The upper Bay reductions in SOD are higher
under this latter scenario due presumably to the higher degree of phosphorus removal in the LOT
than in the 40% controllable. The differences in nitrogen loading are not as great (see Section II).
For the middle and lower regions of the Bay, the 40% controllable scenarios approach the LOT
N&P loading in reducing SOD. In fact, the 40% + CAA + Basin control is at the LOT level of
reduction for zones 4 through the rest of the Bay.

F. TIDAL TRIBUTARY LOADING TO BAY

       The net input of the tidal tributaries to the main Bay is of particular interest since such
loadings represent actual contributions to the Bay proper. As part of the Bay model calculations,
mass balances were conducted around the principal tributaries and the exchange of load across
the interfaces of the tributaries was  calculated for each of the scenarios. For each tributary
interface, the  cells that had flows entering the estuary were separated from the cells that had
flows leaving the estuary. All cells with inflows to the tributary were not necessarily at the bottom
of the interface. This review first focuses on the Potomac and the James estuaries as illustrations
of the dynamic and interactive behavior between tributaries and the Bay. The section closes with
a review of the net nutrient input from all of the tributaries.

       1. Potomac and James Estuaries
       Figure VI - 62 shows the dynamic behavior of the loads (for the average hydrologic year)
at the interface between the Potomac estuary and the main Bay. "To Bay" includes gross output
to the Bay while "From Bay" is gross input from the Bay to the Potomac. "Net" is the difference
between inflow mass and outflow mass.  As seen, peak loadings to/from the Bay occur in the
spring. Net loads to the Bay from the Potomac represent the difference between two large
loadings entering the tributary and leaving the tributary. It can also be noted that the net transport
of nitrogen and phosphorus is small outside of the spring period.
                                        VI-31

-------
                 SEASON: JUNE -SEPT
  40
UJ

< 35


030
oc.

0 25


S 20
O
D
UJ  ,
OC  5
                                        s—.
   LOT N&P


   40% CONT
   40%+CAA
     --e—

40%+CAA+BASIN
       300   250    200    150   100   50     0

      Distance (km)  Along Main Bay (0 = Mouth of Bay)
   Figure VI - 61. Longitudinal variation of percent reduction of SOD
   from Base case. Average over zone over season for average year.
                            VI-32

-------
              POTOMAC INTERFACE
     TOTAL NITROGEN - BASE CASE - YEAR 9 (AVERAGE)
   400

f
£ 300
o
| 200


O 100
1-100
z

•*! -200
O

*~ -300
               FROM BAY
                      5678
                     MONTH OF YEAR 9
                                   9   10  11   12
              POTOMAC INTERFACE
    TOTAL PHOSPHORUS - BASE CASE - YEAR 9 (AVERAGE)
  15
10
1
E

2
3
to
£
O
a
w
O
I
  -10
                 FROM BAY
     1
                                      10   11   12
      2345678!
                  MONTH OF YEAR 9
Figure VI -62. (Top) Monthly variation of TN at Potomac interface
with Bay. (Bottom) Monthly variation of TP at Potomac interface with
Bay.
                        VI-33

-------
       The Potomac results can be contrasted to those from the James estuary shown in Figure
VI - 63. Here there is a less pronounced spring peak of loading in either nutrient and the net
loading from the James to the Bay extends throughout the year. Net flows show similar patterns.
Note also the change in scale. In addition, while the net nitrogen loadings from the Potomac are
substantially larger (by a factor of two) than that from the James, net phosphorus loadings are
very similar. Finally, it can be recalled from the tracer studies that inputs from the James tend to
remain highly localized to the lower Bay.
       Further results for the range from the Base case to the LOT case are shown in Figures VI
- 64 to VI-69 for the Potomac and James estuaries. Figure VI - 64  shows that for the average
hydrology year, the net output load of total nitrogen for the base and LOT cases is calculated at
40% and 33%, respectively of input load. These results are similar to those for the James total
nitrogen flux (Figure VI - 65). The net phosphorus loading is about half of total nitrogen for the
Potomac (Figure VI - 66). The James is significantly different as seen in Figure VI - 67 where for
the  base case, the net phosphorus from the James to the Bay is calculated at 38% of the input load
whereas for LOT,  the net load is only about 2% of the input total phosphorus load. This is due
primarily to the large influx of phosphorus from the ocean boundary to the James via inflows into
the  bottom of that estuary. (Note that the influx of phosphorus  remains approximately the same
over the range of  loading from the base case to the LOT.)
       Figures VI - 68 and VI - 69 display the relative percent  change of the interfacial loads
from the base case loads over the range from the 40% controllable case (S02) to the LOT case
including the geographical runs (S06-S08) where LOT was  selectively applied to the three
geographical regions of the Bay, discussed in Section II. Scenario Loads.  For the Potomac, the
maximum reduction in TN and TP loads is about 50%. The  effect of LOT in the Upper Bay only
or LOT in the lower Bay only on the Potomac loading to the Bay is negligible, indicating little
influence from up-Bay or down-Bay loading on the net Potomac load to the Bay. Figure VI - 69
shows that for the James up-Bay and mid-Bay load LOT reductions have  no influence on the
James. The rather  substantial reductions from the base case  for  the James TN and TP are to be
noted.
       Figure VI - 70 shows a rather remarkable linearity in the net load from these two
tributaries over the range of loadings from the base to the LOT (not including the geographical
runs).  As seen for  TN, if the total input load of TN to the Potomac or James  is reduced by, say,
30% from the base case load, then the net load of TN exiting from the Potomac or James is
reduced about 35% from the base case net load. For TP an approximate linearity is also observed.
Therefore, in spite of the rather complex nonlinear interactions that exist in the overall model
framework, and the apparent interactions between the Bay and  the tributaries, the relationship of
net  load from these two tributaries to the Bay is directly proportional to the reduction in external
load to the tributary.
       In summary, the net loading of TN from the Potomac and James estuaries is about 40%
of the input TN load for the base case and average hydrology and for TP is about 33% of input
TP  loads to those two tributaries. The relationship between  percent reduction from base case of
input load to the tributaries and the resulting percent reduction  from base case of the net load
exiting from the tributaries is approximately linear. Thus,  a 20% reduction from base case in TN
and TP input load  to the Potomac and James estuaries results in about a 20% reduction in net
loading of TN and TP from the two tributaries to the Bay.

                                        VI-34

-------
                JAMES INTERFACE
     TOTAL NITROGEN - BASE CASE - YEAR 9 (AVERAGE)
  100
£


O
E  50
UJ

1   '
2 -50
                          TO BAY
                     FROM BAY
                                 I	[_
         2   3   4   5   6   7   8    9   10  11   12
                     MONTH OF YEAR 9
                JAMES INTERFACE
    TOTAL PHOSPHORUS - BASE CASE - YEAR 9 (AVERAGE)
£10

g  4
to
CO
O
X
Q.
                          TO BAY
       FROM BAY
5   6   7   8
MONTH OF YEAR 9
                                        10   11   12
  Figure VI -63. (Top) Monthly variation of TN at James interface with
  Bay. (Bottom) Monthly variation of TP at James interface with Bay.
                        VI-35

-------
 POTOMAC ESTUARY - INPUT AND OUTPUT TN LOADS
                      BASE CASE
  TOTAL NITROGEN million Ibs/yr

  140

  120 I-            	1169

  100

  80

  60

  40

  20

   0
        INPUT   OUT-SURFACE  IN-BOTTOM

AVERAGE HYDROLOGY YEAR (YR.9)
                      NET = 40% OF
                      INPUT LOAD
                       NET OUT
POTOMAC ESTUARY - INPUT AND OUTPUT TN LOADS
                LIMIT OF TECHNOLOGY
  TOTAL NITROGEN million Ibs/yr
  140

  120

  100

  BO

  60

  40

  20

   0
                      NET = 33% OF
                      INPUT LOAD
        INPUT
OUT-SURFACE  IN-BOTTOM
NET OUT
AVERAGE HYDROLOGY YEAR (YR.9)
   Figure VI -64. (Top) Input and output TN loads at Potomac interface
   with Bay, Base case.(Bottom) Input and output TN loads at
   Potomac interface, LOT.
                         VI-36

-------
  JAMES ESTUARY - INPUT AND OUTPUT TN LOADS
                      BASE CASE
  TOTAL NITROGEN million Ibs/yr
  50
                    456
                                      NET=45% OF
                                      INPUT LOAD;
  40
  30
20
  10
        INPUT    OUT-SURFACE  IN-BOTTOM    NET OUT

AVERAGE HYDROLOGY YEAR (YR.9)
  JAMES ESTUARY - INPUT AND OUTPUT TN LOADS
                LIMIT OF TECHNOLOGY
  TOTAL NITROGEN million Ibs/yr
  50
  40
  30
  20
  10
                    310
                                       NET = 33% OF
                                       INPUTLOAD
        INPUT
              OUT-SURFACE  IN-BOTTOM
NET OUT
AVERAGE HYDROLOGY YEAR (YR.9)
   Figure VI -65. (Top) Input and output TN loads at James interface
   with Bay, Base case.(Bottom) Input and output TN loads at James
   interface, LOT.
                         VI-37

-------
POTOMAC ESTUARY - INPUT AND OUTPUT TP LOADS
                     BASE CASE
 TOTAL PHOSPHORUS million Ibs/yr
 5
                   424_
        375
                              3.40
       INPUT    OUT-SURFACE  IN-BOTTOM

AVERAGE HYDROLOGY YEAR (YR 9)
                                      NET = 22% OF
                                      INPUT LOAD
                                         0.84
                        NET OUT
POTOMAC ESTUARY - INPUT AND OUTPUT TP LOADS
               LIMIT OF TECHNOLOGY
 TOTAL PHOSPHORUS million Ibs/yr
 5
                                     NET = 23% OF
                                     INPUT LOAD
       INPUT
OUT-SURFACE  IN-BOTTOM
NET OUT
AVERAGE HYDROLOGY YEAR (YR.9)
   Figure VI -66. (Top) Input and output TP loads at Potomac interface
   with Bay, Base case.(Bottom) Input and output TP loads at Potomac
   interface, LOT.
                         VI-38

-------
  JAMES ESTUARY - INPUT AND OUTPUT TP LOADS
                     BASE CASE
 TOTAL PHOSPHORUS million Ibs/yr
 5
                                     NET = 38% OF
                                     INPUT LOAD
       INPUT    OUT-SURFACE  IN-BOTTOM    NET OUT

AVERAGE HYDROLOGY YEAR (YR.9)
  JAMES ESTUARY - INPUT AND OUTPUT TP LOADS
                LIMIT OF TECHNOLOGY
 TOTAL PHOSPHORUS million Ibs/yr
 5
                   238
                              236
       INPUT
OUT-SURFACE  IN-BOTTOM
NET OUT
AVERAGE HYDROLOGY YEAR (YR.9)
  Figure VI -67. (Top) Input and output TP loads at James interface
  with Bay, Base case.(Bottom) Input and output TP loads at James
  interface, LOT.
                         VI-39

-------
POTOMAC INTERFACE TOTAL NITROGEN NET LOADS
               % CHANGE FROM BASE
 % CHANGE FROM BASE
 100
  40
  20 -
   0
   802-40% CONT      S06-LOT-UP      S08-LOT-LOW
             SOS-LOT        S07-LOT-MID      S13-SEAS BNR


NOTE: NEGATIVE PERCENT = NET LOAD GREATER THAN BASE CASE LOAD
       POTOMAC INTERFACE TOTAL PHOSPHORUS
          NET LOADS - % CHANGE FROM BASE
  % CHANGE FROM BASE
  100
  20 -
   0
    802-10% CONT      S06-LOT-UP      S08-LOT-LOW
             SOS-LOT       S07-LOT-MID     S13-SEAS BNR
NOTE: NEGATIVE PERCENT = NET LOAD GREATER THAN BASE CASE LOAD
 Figure VI -68. (Top) Percent change from Base case of net TN loads
 at Potomac interface with Bay. (Bottom) Percent change from Base
 of net TP loads at Potomac interface with Bay.
                        VI-40

-------
 JAMES INTERFACE TOTAL NITROGEN NET LOADS - %
                 CHANGE FROM BASE
  % CHANGE FROM BASE
  100
  80
  60
  40
  20
   0
               72.7
                                     73.9
                      (0.2)
                              (0.3)
    802-40% CONT      S06-LOT-UP       S08-LOT-LOW
             SOS-LOT       S07-LOT-MID      S13-SEAS BNR


NOTE: NEGATIVE PERCENT = NET LOAD GREATER THAN BASE CASE LOAD
     JAMES INTERFACE TOTAL PHOSPHORUS NET
            LOADS - % CHANGE FROM BASE
  % CHANGE FROM BASE
  100
100.0
   0
    S02-40% CONT      S06-LOT-UP       S08-LOT-LOW
              SOS-LOT        S07-LOT-MID      S13-SEAS BNR
NOTE: NEGATIVE PERCENT = NET LOAD GREATER THAN BASE CASE LOAD
  Figure VI -69. (Top) Percent change from Base case of net TN loads
  at James interface with Bay. (Bottom) Percent change from Base of
  net TP loads at James interface with  Bay.
                         VI-41

-------
                    TOTAL NITROGEN

  % REDUCTION FROM BASE CASE OF NET LOAD FROM TRIB
  100
   80


   60


   40


   20
       POTOMAC
                                    JAMES
     0        20        40        60        80       100
        % REDUCTION FROM BASE CASE OF INPUT LOAD TO TRIB
                   TOTAL PHOSPHORUS

  % REDUCTION FROM BASE CASE OF NET LOAD FROM TRIB
  100
   80
   60
   40
   20
                                        JAMES
POTOMAC
     0         20        40        60        80        100
        % REDUCTION FROM BASE CASE OF INPUT LOAD TO TRIB

Figure VI -70. (Top) Relationship between % reduction of input load
to tributary and net load exiting tributary to Bay for TN.  (Bottom)
Relationship between % reduction of input load to tributary and net
load exiting tributary to Bay for TP.
                          VI-42

-------
       2. All Major Tidal Tributaries
       Figures VI-71 through VI-73 show the net flux of nitrogen for all of the major tidal
tributaries.  The most interesting point of these runs is that the Rappahannock and York rivers are
calculated to receive input nitrogen load from the Bay as opposed to these tributaries providing a
net input to the Bay. Indeed, under several removal programs (e.g., 40% Cont. and LOT N&P)
the input net nitrogen load increases from the Bay to the tributary (see Figure VI-73). This is
undoubtedly a result of a complex interaction of transport and nutrient concentration where the
gradient from the Bay to these tributaries is increased under various removal programs. The
relative magnitude of the net nitrogen loads by tributary can also be seen in Figure VI-71. For the
Base case, the three largest inputs are the Patapsco/Back, Potomac and James estuaries. The
Patuxent contributes a small input and as noted previously, the remaining two lower Bay
tributaries receive a net input from the Bay.
       Figures VI-74 and VI-75 show a similar behavior for the net phosphorus loadings from the
tributaries. Again, the Rappahannock and York rivers are calculated to receive a net input of
phosphorus from the Bay. The change in loading under different control scenarios is
approximately similar to that of nitrogen.

G. DISCUSSION  AND CONCLUSIONS

       This  section has reviewed the results of the scenarios from the point of view of the
behavior of nitrogen, phosphorus, phytoplankton (and associated effect on light penetration),
carbon and sediment oxygen demand. Particular attention was paid to the behavior of the
phytoplankton production and resulting carbon fluxes as a result of reductions in nitrogen and
phosphorus,  either together or separately. The ability to examine the behavior of the Bay with the
calibrated CBWQM under different removal levels of nutrients in combination is a particularly
important use of the model. Such behavior is not directly observable in the Bay and can only be
predicted by a credible model. The degree to which phosphorus  and nitrogen load reductions have
an impact on the water quality of the Bay is of course an important consideration in the decision
making process.
       In general, the Bay can be divided into three broad regions: the upper approximately 100
km of the Bay where control of phytoplankton growth and production is by phosphorus, the
approximately 100 km of the lower Bay where the phytoplankton production is controlled by
nitrogen and a middle Bay region of about 100 km where a transition takes place. The extent of
nitrogen control proceeds up the estuary during the summer and fall and is a function of fresh
water hydrology and resulting circulation. This general conclusion drawn for the Base case is
consistent with interpretations of observations made on the Bay  by a variety of investigators (see,
e.g., Fisher et al., 1992).
       What is not obvious from the existing data is that as phosphorus loadings to the Bay are
reduced (with nitrogen loadings remaining at approximately Base levels), excess nitrogen is
transported down the Bay in the surface waters. This transport of nitrogen then is calculated to
stimulate phytoplankton production in this nitrogen limited region of the Bay. This "additional"
relatively labile biomass then settles in the downstream region and contributes to higher SOD in
that area.
                                         VI-43

-------
OBBBOQ
                          o     o


      A*p/)Ui QV01 N39OM1W13N
                                         1
                                         ii
                                            i
                                         e o.
                                         *• a

                                         5?
      X«p/«ui QV01 N3OOU1IN13N
                                                                       DSBE3OE3
                                                                    3
                                                                    S
                                                                       S    g    8    °    S?   8    S

-------
                                                  D PAT/BACK
                                                    PATUX
                                                    POT
                                                    RAP
                                                  BYORK
                                                  ED JAMES
 -0.5
           BASE
40% CONT.
LOT N&P
Figure VI - 74 . Variation of annual net phosphorus load from
tributaries for Base, 40% controllable and LOT N&P scenarios.
Average year.
 -0.5
                                                  D PAT/BACK
                                                    PATUX
                                                    POT
                                                  m YORK
                                                  a JAMES
         LOT - UPPER
                        LOT - MID
                                     LOT - LOWER
Figure VI - 75 . Variation of annual net phosphorus load from
tributaries for "Geographical" scenarios, #6 -#8. Base case. Average
year.
                          VI-45

-------
       Phosphorus removal however has a distinctly positive effect in the surface waters of the
upper Bay where spring and summer phytoplankton biomass are reduced considerably more than
if only nitrogen were removed. Such reductions of biomass of 20-30% have an impact on light
penetration with a 20% increase in light calculated for the 2 m depth at LOT levels.
       Reductions in nitrogen have  of course a direct effect on phytoplankton production in the
nitrogen limited areas and subsequently on the carbon fluxes and the SOD. In addition, the
nitrogen load reductions result in improvement in meeting the DIN habitat requirements for the
SAV.
       It is concluded  from the analyses in this Section, that load reductions of both phosphorus
and nitrogen are necessary to result in reductions in the nutrients, phytoplankton biomass, (with
increases in light penetration) and sediment oxygen demand. Phosphorus load reductions are most
effective in achieving improvement in these measures of water quality  in the upper Bay. Nitrogen
removal is required throughout the Bay: in the upper Bay to reduce nitrogen loads that would be
transported down Bay  under the phosphorus reduction and in the middle and lower Bay to
directly reduce biomass and hence SOD.
       The impact of these reductions in phytoplankton carbon and SOD on the DO, especially of
the deep bottom waters of the Bay trench is explored in the next Section.
       The net input of nutrients from the principal tidal tributaries to the Bay is exclusively from
the Patapsco/Back, Potomac and James estuaries. The Rappahannock and York estuaries are
calculated to receive a  net input of nitrogen and phosphorus from the Bay. For the Potomac and
James estuaries, the net nutrient load exiting the tributary to the Bay is approximately linear to the
external load of nutrient to the tributary.
                                         VI-46

-------
VII. DISSOLVED OXYGEN RESPONSE

A. INTRODUCTION

       The calculated response of the dissolved oxygen of the Bay assumes particular importance
in analyzing the effects of scenario nutrient load reductions. The dissolved oxygen focus in this
Section is twofold: (1) evaluation of the seasonal (specifically summer) average DO response, and
(2) analysis of the response of the DO concentrations below 1 mg/L, the assigned level of anoxia.
The latter quantity is determined by calculating the volumetric extent and temporal extent of DO
below 1 mg/L. These "anoxic volume-days" have units of m3 - days.

B. SEASONAL AVERAGE BASE CASE DISSOLVED OXYGEN

       Figures VII -1 and VII - 2 show the model calculated spring and summer average
longitudinal DO profile for the Base case loading condition and for the average hydrology flow
year. The summer profile is the basis for comparison of assessing the effect of nutrient reduction
scenarios. The rapid drop of the minimum bottom DO between the spring level of greater than 5
mg/L to the minimum summer average level of 0.1 mg/1 can be noted. The steep increase in the
bottom DO beyond the upper limit of the deep trench at  approximately 260 km is due to a rapid
decrease in depth. The marked vertical gradient in DO during the summer can also be seen where
average surface DO is generally supersaturated due to algal productivity and the bottom DO is
responding to deep water sinks of oxygen.  The marked difference in the longitudinal profiles
between the surface and pycnocline levels and the bottom level can also be noted. Some
interpretation of the bottom DO is given in the next sub-section.

       1. Analysis of Base Case Bottom DO
       Examination of the bottom DO longitudinal summer profile indicates an approximate
linear decrease in DO as one progresses up the Bay. A simple analysis of the behavior of the DO
in the bottom waters can be made to help understand this behavior.
       The principal sinks of oxygen in the bottom water are the sediment oxygen demand
(SOD), the oxidation of the dissolved organic carbon (DOC) and the immediate uptake of oxygen
to satisfy the chemical oxygen demand (COD) of reduced substances released from the sediment.
Phytoplankton respiration is neglected since during the summer the bottom layer phytoplankton
biomass is small. In the CBWQM, the  sediment and water column are interactive and not
separated. However, since the output from the sediment model is computed as equivalent SOD
and COD, an analysis can be made considering these processes as external sinks of DO. The rates
of utilization of oxygen in the model are oxygen dependent, but this complication is not
considered here in this simple analysis. Also, vertical mixing of oxygen is not included which
simplifies the analysis considerably. (See Kuo et al, 1991, for a more detailed analysis of bottom
DO in the Rapphannock River which includes vertical exchange processes.)
       For these assumptions, consider then the following equation for DO in bottom waters:

                       ,,dc    dc     S   v                 COD
                         dx = dt7 = ~H "^DOCaDOCCDOC	~

                                        VII-1

-------
   12
2
    6
Q
o

-------
where c is the DO (g/m3), u is the average velocity of bottom waters (m/d), t* is the average time
of travel (days) for x measured in the up Bay direction beginning from the Bay mouth, S is the
SOD (g/m2-d), H is the depth of the bottom mixed layer (m), KDOC is the respiration rate of the
dissolved organic carbon (1/d), a,^ is the stoichiometric oxygen equivalent of the DOC (2.67 mg
DO/mg carbon), c^ is the dissolved organic carbon concentration (g/m3) and COD is the areal
uptake of oxygen due to immediate chemical oxygen demand of reduced substances (g/m2-d).
       Notice that in this simplified form, the DO is a conservative variable (i.e., there are no
kinetic terms for the state variable, only sources and sinks). The DO concentration is thus given
by a linear equation as

                       c = c0 - (S/H + K^a^c^  + COD/H}t*
for c0 as the initial bottom DO at t* = 0, i.e., at the approximate mouth of the Bay. Figure VII - 3
illustrates this behavior. Calculated values for the various sink terms can now be assigned to
determine the relative magnitude of the impact of each process.
       The average summer SOD over the distance from the mouth of the Bay to about 260 km
is approximately 1.2 g/m2-d (see Figure VI - 57). K^ is given as 0.01/d and c^ is about 2 mg/L
(see Cerco and Cole, 1992). COD is calculated to occur only in the region from about 200-260
km at about 0.5 g/m2-d. Using Table IV - 2, the travel time is about 13.6 days to 200 km and
about 17.2 days to reach 260 km. For an assumed bottom depth of about 4m, the bottom DO is

             c = c0  - {(0.3 + 0.05) t* }                     for 0<=t*<=13.6
             = c0 -  {(0.3 +0.05)t* +0.12(t*-13.6)}       for 13.6 
-------
                           si I
                            M

                           .Xfl
                            I la
 O    CO    <0    •«    CM    O

      1/OuJ N30AOO 03ATOSSIQ
o   eo    ID   ^   CN    o

 I/Bui N30AXO Q3AX)SSia
                               II
                               I
O    «    «'    •»    CM

     1/&U N30XXO QBATOSSia

-------
 summer average basis. Figure VII - 5 shows the summer average bottom DO profile for several
scenarios. Only when the incoming loads are reduced by 50% is the summer average bottom DO
calculated to be greater than 1.0 mg/L. It should of course be recognized that one of the DO
objectives as summarized in Table I -1 is for DO concentrations to be greater than 1.0 mg/L at all
times. It is of interest to note that a significant change in the slope of the bottom DO profile does
not occur until beyond the 50% removal level and at 90% removal the change in slope is
significant, reflecting the substantial reduction in oxygen sinks.
       The differences in summer average DO between a given scenario and the Base case are
shown in Figures VII -6 through VII - 9. For the LOT - Base case (Figure VII - 6),  surface DO is
decreased as would be expected from the reduction of phytoplankton biomass in that layer. The
difference in bottom DO of about 0.4 mg/L is approximately constant over the distance from 75
km to 270 km indicating that the LOT has  a linear slope approximately the same as the Base case.
Comparing the LOT N Only and LOT P Only in the vicinity of km 270, there appears to be a
beneficial additive effect of removing both  nutrients. Removal of N and P alone results in a DO
improvement in that vicinity of about 0.3 mg/L, but LOT N&P results in an improvement of
another 0.1 mg/L. The 40% controllable scenario (Figure VII - 9) improves the bottom DO by
about a constant 0.2 mg/L on a summer  average basis which is about half of the LOT N&P
scenario.
       The LOT P Only case shows significantly less DO improvement than LOT N Only,
indicating that phosphorus reductions do not have as significant an effect on the DO in bottom
waters as do nitrogen reductions. This is a  result of all of the previously discussed processes that
are a consequence of phosphorus reductions, namely, the increased transport of nitrogen to down
Bay regions resulting in a stimulation of biomass,  resulting in an increase in carbon deposition to
the sediment and subsequent increase in  SOD. As shown by the simple analysis  above, it is the
SOD in the down Bay region that has more of an effect on trench  DO that up Bay SOD. The net
result of all these processes is that while  phosphorus load reductions have a significant impact on
up Bay biomass, the bottom DO is not affected as significantly as  with nitrogen removal.

D. ANOXIC VOLUME DAYS RESPONSE

       As noted above, a useful measure of the degree of anoxia is the total volume - days where
the DO was calculated to be less than 1 mg/L. That is, model output is tracked on a cell by cell
basis over time and a product sum is accumulated over zone, season and annually for each
scenario. A complete listing of the anoxic volume days for all scenarios across season is given in
the Appendix as Table A-3 and for anoxic volume days by zone across the year is given in the
Appendix as Table A-4.
       Figures VII -10 and VII -11 show the seasonal and zone totals, respectively for the Base
case. As expected, maximum anoxia, as defined by the anoxic volume days occurs in the summer
with about 16% occurring in the spring and fall. The longitudinal variation of the anoxia shown in
Figure VII -  11 indicates peak regions in zones 2 and 3 where about 70% of the annual total
occurs. An additional 24% occurs in Zones 4 and 9 (Eastern Shore).
       The percent  reductions in total annual anoxic volume days from the Base case for selected
scenarios is shown in Figure VII -  12. Complete elimination occurs at 90% N&P removal

                                        VII-5

-------
 S'S
       ii
o   o   o   o       999
     I/*" 3SV9 WOMd 'ddlQ Od
                                                                     o   o       99
     i/ouj 3SV8 noud -ddia oa
                                                              S'B
                                                          5!
                                                                                    II
                                                             o
  D   •»   CV|   O   0(   •»   <0
  ooo       999
   I/Bui 3SV9 WOMd 'ddlO OQ
                                                                                                            CO
                                                                                                            >

-------
    350

    300

    250
    150
  g 100
  o
  g  50
  z
           WINTER    SPRING    SUMMER
FALL
   Figure VII -10. Seasonal toals of anoxic volume days for Base case
   for average year.
1200

3 100
E
| 50

S-
eo 20
5
-   2
§
I   1
         300     250    200     150     100     50      0
            Distance (km) Along Main Bay (0 = Mouth of Bay)
      |     1    I    2   I   3  I   4   I    5   I  8  I  7  |  B|
                    BOUNDARIES OF BAY ZONES
                       (See Figure III -1)

  Figure VII -11. Longitudinal variation of anoxic volume days for
  Base case total over zone for average year.
                            VII-7

-------
UJ
                                 SCENARIO
    Figure VII-12. Percent reductions in anoxic - volume - days from Base
    case for selected scenarios. Average Year
       100
        80
     m
     o
     s
     a.
     #
        60
        40
        20
                                          LOT N&P

                                        LOT N ONLY

                                        LOT P ONLY
             300    250   200    150    100     50     0
             Distance (km) Along Main Bay (0 = Mouth of Bay)
1   I   2   |  3  |   4  |    5   |  6
        BOUNDARIES OF BAY ZONES
            (See Figure III -1)
                                                      8|
       Figure VII - 13. Longitudinal variation of percent reduction from
       Base case of total  anoxic volume days for average year, by zone
                                  VII-8

-------
 (Scenario #16). The feasible range of anoxia reduction from 40% controllable to LOT is about
20-30%. Note that 40% controllable +CAA+Basin control (Scenario #4) approaches the
improvement from LOT N&P (#5).
       The relative impact of nitrogen and phosphorus is also seen by comparing #10 and #11
with #5. As noted above, LOT P Only has less of an impact on the bottom DO and as shown in
Figure VII -12, only a 15% improvement in anoxia is calculated which is half that from LOT N
Only. The reasons for this difference have been hypothesized previously as being due primarily to
increased nitrogen transport to down Bay nitrogen limited regions. The subsequent simulation of
phytoplankton results in increased deposition of relatively labile carbon to the sediment in the
down Bay areas. The resulting SOD is therefore elevated with a concomitant lessening of
improvement in anoxia. However, it should also be noted that LOT N Only does have a greater
reduction in primary production during the summer months in the mid to lower Bay regions (see
Figures VI - 38 to VI - 40). Since maximum anoxia occurs during the summer, LOT N Only can
be expected to also have a relatively larger impact on summer anoxia than LOT P Only.
      In addition to this hypothesized effect, two other effects may also contribute to the
reduced effect on anoxic volume-day response to the LOT - P Only scenario. As noted in Section
VI, reduced phosphorus loading reduces primary production in the surface waters of the upper
Bay. Such a reduction has the following two consequences:
      (1) the reduced algal growth at the surface decreases surface water DO (see Figures VII -
6 through VII - 8) which in turn decreases the vertical concentration gradient thereby reducing the
exchange of DO and replenishment of bottom DO in the upper Bay, and
      (2) the reduced algal growth will not assimilate as much ammonium with the result that
nitrification will increase in the surface waters, decreasing the DO and again decreasing the
vertical transport of oxygen to bottom waters of the upper Bay.
      It can also be noted that the location of where LOT load reductions are applied is also
significant. Thus, comparing Scenarios #6,#7 and #8 indicates that maximum impact on bottom
anoxia is from load reductions in the mid-Bay region. That is, as Figure VII - 12 indicates, there is
a negligible percent reduction in anoxic volume days under Scenario #8 (LOT - Lower Bay) as
compared to 8% for Scenario #6 (LOT - Upper Bay) and 21% for Scenario #7 (LOT - Middle
Bay). The minimum impact on anoxia for LOT in the lower Bay only is apparently a consequence
of (a) no net input of nutrients from the Rappahannock and York estuaries (but an input from the
Bay into these tributaries, see Section VI, Figures VI - 71 to VI - 75), and (b) possible transport
of nutrient input from the James out through the mouth of the Bay more than transport of
nutrients up the Bay proper.
      Figures VII -13 and VII -  14 show the percent reduction of anoxia from Base as a
function of zone annual totals. Caution should be exercised in interpreting these plots since a high
percent reduction as in the lower Bay zones generally means a reduction from an already low
value. Figure VII -13 indicates the relatively low percent reduction of the LOT P only case and as
a result the improvement given for LOT N&P is almost all due to nitrogen reduction.
                                        VII-9

-------
  100
o>
CB
m
   80
I  60
   40
   20
     LOT N&P

     40% CONT

     40%+CAA

  40%+CAA+BASIN
        300   250   200    150   100   50     0
      Distance (km) Along Main Bay (0 = Mouth of Bay)
          1   I   2  |   3  |  4  |   6  |  6
                 BOUNDARIES OF BAY ZONES
                    (See Figure III -1)
81
   Figure VII -14. Longitudinal variation of percent reduction from
   Base case of total anoxic volume days for average year, by zone ,
                             VII-10

-------
E. ANOXIC VOLUME DAYS RESPONSE SURFACE ANALYSES

       1. Whole Bay Response
       The relationship between nitrogen and phosphorus loadings and the response in terms of
anoxic volume days can be explored using the response surface analysis discussed in Section III.
It has become apparent that the response of the DO to reductions in phosphorus, for example, is
less than that from reductions in nitrogen. Response surface analysis of the anoxic volume days
provides a quantitative means for relating load reduction to this overall measure of anoxia.
       The response surface analysis for the anoxia goal indicates that there is a strong
relationship between nitrogen reduction and improvement in anoxia and a lesser overall
relationship between improvement in anoxia and phosphorus reduction. Figure VII -15 shows the
relationship between nutrient removal and the improvement in anoxia from the base case scenario
for the whole Bay in an average year. Nutrient reduction is scaled from 0 to 1 with 0 being Base
case nutrient removal and 1  indicating 100% removal, with analogous scaling for anoxia. The
.analysis indicates that  100% removal of phosphorus (with 0% removal of nitrogen) would fall
well short of eliminating  anoxia. A 90% reduction in nitrogen totally eliminates anoxia.
       Broken down by  season (Figure VII -  16), and recalling that the major contribution to the
annual anoxia is during the summer (see Figure VII -10), it is clear that the whole Bay full year
response is dominated by the whole Bay  summer response. The two surfaces are quite similar,
with a strong linear response to nitrogen removal and a much weaker response to phosphorus
removal. Spring reduction approaches 100 % but the amount of anoxia during this season is small.
Nitrogen is the controlling nutrient in these graphs, as evidenced by the slight improvement in
anoxia achieved by relatively large reductions in  phosphorus. However, in the spring, phosphorus
reduction plays a greater role in the control of anoxia compared to other seasons.

       2. Response by Zone and Season
       Similar analyses can be made by regressing the responses in the different Bay zones on
bay-wide nutrient reductions. Not all  surfaces were represented equally well by the statistical
analyses as discussed below. Representative responses are shown in Figures VII - 17 through VII
- 18 which indicate  surfaces for Zones 2 and 4.  For Zone  2, summer, the insignificant role of
phosphorus is clear.  However, for Zone 4 (Figure VII -18), phosphorus plays an increasing role
in reducing anoxia. This was not apparent in earlier analyses since load reductions that were
presented did not extend over the range of reductions shown in these surfaces (i.e., 0 - 50%). It
should be remembered that Zone 4 accounts for only about 10% of the total anoxic volume days
in the Bay.
       The individual zones showed a somewhat greater response to phosphorus reduction than
seen in the whole Bay surfaces. These details were obscured in the whole Bay analysis because
they were overwhelmed by the profound  summer anoxia in Zones 2 and 3.
       As noted, some response surfaces are fit better than others. For example for Zone 1,  the
surface does not predict zero improvement in anoxia for zero reduction of N and P (i.e. there is a
large residual for the base case datum). Generally the best fits are calculated for the zones with
                                        VII-11

-------
 REDUCTION IN ANOXIA

   A


I. DO
0.25
0.00
  1.00
J.J^- i.oo
                      ,cr,0,   o.oo  -^  *******
    Figure VII - 15. Response surface of total  anoxic volume days for average

    year.
                              VII-12

-------
o>
 J-,
 (0
 0)
 00
 c
.1—1
 t-
 o,
CO
 O5

 IM
 CO
 0)
 (0
                                                             c
                                                             o
                                                                                                                         en
                                                                                                                        ffl
                                                                                                                        
 t_
 03
 0>
Ex

 flj
Oi

 U
 (0
 
-------
 a
oo
o>
 u,
 (0
 
-------
Oi

 L.

 (6
 0)
 00
 G
CO

x>
  
 >«

 •—>
  (0
 fe,

 T3
                         m    CM
o»
 u,
 co
 0>
>"


3
c^,

 03
                                                       Oi

                                                        u
                                                        (0
 e
 3
CO

 6
                                                       C



                                                       
                                                       o

                                                      'I
                                                                                                             oo


                                                                                                              I
                                                  VIM5

-------
strong anoxic events. Despite occasional weakness, the overall success of this procedure can be
seen in Table VII -1, which lists the coefficients of determination (r2) for the surfaces shown in
the preceding Figures. (A listing of the regression equations is given in Appendix B.) In general,
most of the fits would be considered excellent, with small residuals and significant parameters
resulting in the high r2 values. Spring was the most difficult season to fit, consistent with the
general pattern of strong regressions corresponding to strong anoxia. However, spring anoxic
volume days is not a significant fraction of the annual total. Likewise all the regressions for Zone
4 were relatively less strong. However, even the weak regressions contain valid "general trend"
information on the influence of N and P reductions on anoxia.
                                     Table VII -1
             Coefficients of Determination (r2)  for Seasonal Response Surface
       Region
Full Year
Spring
Summer
Fall
Whole Bay
Zone 1
Zone 2
ZoneS
Zone 4
Zone 5
Zone 6
Zones 7 & 8
Zone 9
0.9804
0.9990
0.9846
0.9666
0.9405
0.9905
0.9769
no anoxia
0.9852
0.9895
0.9486
0.9895
0.9269
0.8776
no anoxia
no anoxia
no anoxia
0.9169
0.9839
0.9993
0.9800
0.9730
0.9483
0.9905
0.9769
no anoxia
0.9884
0.9816
0.9991
0.9676
0.9415
0.8686
no anoxia
no anoxia
no anoxia
0.9983
F. CONCLUSIONS

       The results presented in this Section indicate the following:

       1. Bottom DO concentrations under Base case conditions reach minimum summer average
levels of less than 1 mg/L. The approximate linear decline in oxygen with distance as one
proceeds up the Bay in the direction of the bottom flows is a result of the distributed sink of
oxygen occasioned principally by the sediment oxygen demand. As such, the minimum bottom
DO at the head end of the trench reflects the accumulated DO depletion of a bottom water parcel
since it entered the Bay. All  SOD along the path of bottom water contributes to the DO
depletion.
       2. Feasible reductions in nutrient loadings of about 20 -30% N & P (i.e., LOT and "40%
controllable" scenarios) result in improvement in bottom DO over Base by about 0.2 - 0.4 mg/L

                                         VII-16

-------
as a summer average. Load reductions of about 50% or greater result in minimum summer
 average DO concentrations above 1 mg/L. 90% N & P reductions are calculated after the ten
year simulation to result in average summer DO of greater than 5 mg/L.
       3. A measure of anoxia as given by the volumetric and temporal extent of DO less than 1
mg/L (the anoxic volume days) is a maximum in the summer and in Zones 2-4 under Base case.
The feasible load reduction scenarios result in a range of reduction in anoxic volume days of about
20 - 30% from Base. This reduction  in anoxia is directly proportional to the load reduction of
nitrogen of about 20-30%.
       4. Response surface analysis of anoxic volume days on a Bay wide basis indicates a
generally linear response in anoxia reduction as a function of nitrogen with little effect due to
phosphorus reductions. The maximum effect of phosphorus is in Zone 4, a region that contributes
a relatively smaller fraction to the Bay wide total anoxia.
       5. Even though the upper Bay is phosphorus limited, reductions of phosphorus do not
have as significant an effect on anoxic volume days as do nitrogen reductions. The reasons for this
response are complex . Phosphorus controls primary production in the winter and spring while
nitrogen controls primary production in the summer, the period of maximum anoxia. Also, when
only phosphorus is removed there is a calculated increased nitrogen transport to down Bay
nitrogen limited  regions which increased downstream SOD. This effect is apparently coupled
with reduced primary production in the surface waters of the upper Bay resulting in a reduced
vertical DO gradient and less oxygen transferred to the bottom waters of the upper Bay.
       6. The location of where LOT load reductions are applied is also significant. Thus, the
scenarios where LOT reduction were selectively applied by Bay regions (Upper, Mid and Lower)
indicate that maximum impact on bottom anoxia is from load reductions in the mid-Bay region. A
negligible percent reduction in anoxic volume days is calculated for LOT in the Lower Bay only
as compared to 8% for LOT for the  Upper Bay and 21% for LOT in the Middle Bay. The
minimum impact on anoxia for LOT in the lower Bay only is apparently a consequence of (a) no
annual  net input of nutrients from the Rappahannock and York estuaries (but rather an input from
the Bay into these tributaries) and (b) possible transport of nutrient input from the James out
through the mouth of the Bay more than transport of nutrients up the Bay proper.
                                        VII-17

-------
                               VIII. REFERENCES

Batuik, R., R. Orth, K. Moore, W. Dennison, C. Stevenson. L. Straver, V. Carter, N. Rybicki, W.
Hickman, S.Kollar, and S. Bieber. 1992. Chesapeake Bay submerged aquatic vegetation habitat
requirements and restoration targets: A technical synthesis. CBP/TRS 83/92, Annapolis, MD.

Blumberg, Af, B.H.  Johnson, R.H. Heath, B.B. Hsieh, V.R. Pankow, K.W.  Kim and H.L. Butler,
1991. Data employed in the development of a three-dimensional, time-varying numerical
hydrodynamic model of Chesapeake Bay. Tec. Rept. HL-91-1, Waterways Exp. Sta., Corps of
Eng., Visksburg, Miss., prepared for US army Eng. District, Baltimore, MD, 180 pp.

Cerco, C.F. and T. Cole,  1992. Application of the three-dimensional eutrophication model
CE-QUAL-ICM to Chesapeake Bay. Draft Report, XIV Sections.

Chesapeake Bay Program, 1993.  Chesapeake Bay living resources habitat-based water quality
restoration priorities. Draft report prepared  for CBP, 29 pp + Figs.

Dennison, W. C., Orth, R. I, Moore, K. A., Stevenson, J. C., Carter, V., Kollar, S., Bergstrom,
P. W., & Batuik, R. A. (1993). Assessing water quality with submersed aquatic vegetation.
BioScience, 43(2), 86-94.

Di Toro, D.M. and J.F. Fitzpatrick, 1993. Chesapeake Bay sediment flux model. Prepared by
HydroQual, Inc. for U.S.  Env. Pro. Agency, Ches.  Bay Prog. Office and U.S. Army Engineer
District, Baltimore, MD. Monitored by Environ. Lab. U.S. Army Eng. Waterways Exper. Sta.,
Vicksburg, MS. Contract Report EL-93-2, 200 pp.

Donigian, A.S., Jr., B.R. Bicknell, A.S. Patwardhan, L.C. Linker, D.Y. Alegre, C-H. Chang, R.
Reynolds,  1991. Watershed model application to calculate Bay nutrient loadings. Final findings
and recommendations.  Aqua Terra Conslutants, Comp. Sci. Corp.Chesapeake Bay Program
Office, USEPA, Annapolis Md. 283 pp.

Dortch, M.S., 1990.  Three-dimensional \, Lagrangian residual transport computed from an
intratidal hydrodynamic model. Tech. Report EL-90-11, Waterways Exp. Sta., Corps of Eng.,
Vicksburg, Miss., prepared for US Army Eng. Dist. Baltimore, MD, 264 pp.

Fisher, T.R., E.R. Peele, J.W. Ammerman, and L.W. Harding, Jr., 1992. Nutrient limiation of
phytoplankton in Chesapeake Bay. Mar. ECol. Prog. Ser. 82:51-63.

Funderbuck, S.L., S.J. Jordan, J.A. Mihursky and D. Riley. 1991. Habitat requirements for
Chesapeake Bay living  resources. Prepared for Living Resources Subcommittee, Chespaeake Bay
Program, Annapolis MD.  2nd Edition, 23 Chapters + App.

Illinois Env. Pro. Agency, 1983.  Soil erosion and sediment transport. Dynamics on the Blue
Creek watershed. IEPAAVPC/83-004, Springfiled, IL.

                                        VIII-1

-------
Johnson, B.H., R.E. Heath, B.B. Hsieh, K.W. Kim and H.L. Butler, 1991. User's guide for a
three-dimensional numerical hydrodynamic, salinity, and temperature model of Chesapeake Bay.
Tech. Kept. HL-91-20. Waterways Exp. Sta. Corps of Eng., Vicksburg, Miss., prepared for US
Army Eng. District, Baltimore, MD, 41 pp + Append.

Johnson, B.H., K.W. Kim, R.E. Heath and H.L. Butler, 1991. Verification of a three-dimensional
numerical hydrodynamic model of Chesapeake Bay. Tech. Kept. HL-91-7. Waterways Exp. Sta.
Corps of Eng., Vicksburg, Miss., prepared for US Army Eng. District, Baltimore, MD.

Jordan, S., M. Olson, C. Stenger, K. Mountford, and R. Batuik. 1992. Chesapeake Bay dissovled
oxygen goral for restpration of living resources habitats. CBP/TRS/88/92. Annapolis, MD.

Kuo, A.Y., Park, K. and Moustafa, M.Z., 1991. Spatial and temporal  variabilities of hypoxia in
the Rappahannock River, Virginia. Estuaries,  14, 113-121.

Nutrient Reevaluation Workgroup, 1992. Progress report of the Baywide nutrient reduction
reevaluation. Printed by USEPA for the Chesapeake Bay Program, Annapolis, MD, 68 pp.

Walling, D.E.,  1983. The sediment delivery problem. J. of Hydrology, 65:209-237.
                                        VIII-2

-------
                   APPENDIX A
                SCENARIO DESCRIPTIONS

TABLE A -1  SUMMARY OF SCENARIO NITROGEN AND PHOSPHORUS
                     LOADINGS

TABLE A - 2 STEADY STATE RESPONSE MATRICES FOR TRACER RUNS

  TABLE A - 3 ANOXIC VOLUME DAYS - SEASONAL AND ANNUAL
                      TOTALS

   TABLE A - 4 ANOXIC VOLUME DAYS - ZONE ANNUAL TOTALS

-------
                                           APPENDIX A.
                                     SCENARIO DESCRIPTIONS
 SCENARIO 1
 Base Case scenarios where 1985 loads from Bay Agreement States with nitrogen and phosphorus ocean boundary
             conditions computed based on mass balance outside the Bay mouth. The scenario is run with Water
             Quality Model (WQM) calibration optimized to the Phase II Watershed Model (WSM) loads.
             Atmospheric loads are to the water surfaces only throughout the Bay, its tributaries and the river
             reaches in the Bay Agreement States.

 SCENARIO 2
 "40%" Reduction of controllable carbon, nitrogen and phosphorus loads to Bay from Bay Agreement States only
             (i.e., does not include NY, WV, & DE). The controllable portion of the Base Case was determined
            by subtracting the load generated by a 3 State all forested watershed with no point sources from the
             Base case. Ocean boundary conditions were computed by mass balance.

 SCENARIO 3
40% + CAA for Bay Agreement States Only simulates a forty-percent reduction of controllable  carbon, nitrogen,
            and phosphorus loads to the Bay from the Bay Agreement states, combined with implementation of
            the 1990 Clean Air Act (CAA). Nitrate (CAA) load reductions were performed individually for each
            watershed and for each year according to guidelines provided by the Chesapeake Bay Program Office
            on May 8, 1992.  Atmospheric nitrate loads to the water surface were reduced fourteen percent. The
            net result was roughly an additional three percent (3%) nitrogen load reduction.

 SCENARIO 4
 40% •+• CAA for Bay Basin run simulates a forty-percent reduction of controllable carbon, nitrogen, and
             phosphorus loads to the Bay from the entire Bay watershed including Delaware, New York, and
             West Virginia, combined with implementation of the 1990 Clean Air Act (CAA). This differs from
             Scenario 3 in which controllable loads were reduced in the Bay Agreement states only. Nitrate
             (CAA) load reductions were performed individually for each watershed and for each year according
            to guidelines provided by the Chesapeake Bay Program Office on May 8, 1992. Atmospheric nitrate
            loads to the water surface were reduced fourteen percent. The net result was roughly an additional
            three percent (3%) nitrogen load reduction.

 SCENARIO 5
 Limit of Technology (LOT) run for Bay Agreement States only and atmospheric loads  consistent with Base Case
             Scenario. Ocean  boundary conditions were computed by mass balance.
 SCENARIO 6
 LOT Upper Bay run is where Limit of Technology nutrient controls were implemented in the oligohaline region
            of the Bay.  LOT  point source nitrogen and phosphorus controls were implemented,  along with the
            most comprehensive best management practices for NFS controls in the entire Susquehanna River
            basin and in the below fall line basins denoted "Coastal 11" to just above Back River.  Atmospheric
            loads were consistent with Base Case Scenario and ocean boundary conditions were  computed by
            mass balance.

 SCENARIO 7
 LOT Middle Bay run is where Limit of Technology nutrient controls were implemented in the mesohaline region
            of the Bay.  LOT point source nitrogen and phosphorus controls were implemented,  along with the
            most comprehensive best management practices for NPS controls in the middle Bay region (Patapsco
            and Back, Patuxent, and Potomac River basins). Atmospheric loads were consistent with Base Case
            Scenario and ocean boundary conditions were computed by mass balance.
                                               A-l

-------
 SCENARIO 8
LOT Lower Bay run is where Limit of Technology nutrient controls were implemented in the polyhaline region
            of the Bay. LOT point source nitrogen and phosphorus controls were implemented, along with the
            most comprehensive best management practices for NPS controls in the lower Bay region
            (Rappahannock, York, and James River basins).  Atmospheric loads were consistent with Base Case
            Scenario and ocean boundary conditions were computed by mass balance:.

 SCENARIO 9
LOT - MID (A) This run investigates the Bays response to Limit of Technology N and P controls from Back
           River to just above Potomac River; however, unlike Scenario 7, fall line and below fall line PS and
           NPS loads within the Potomac River and basin were left at Base Case levels.

 SCENARIO 10
LOT Nitrogen Only run is where LOT nitrogen controls were implemented throughout Watershed Model and
           LOT nitrogen limits were specified at all point sources (3.0 mg/1). Point source phosphorus was left
           at Base Case. Although scenario 10 has  the same N & P overall removal rate, this run differs from
           Scenario #19 where there was a 31% N and 18% P removal uniformly applied to all tributaries.
           Atmospheric loads were consistent with Base Case  Scenario and ocean boundary conditions were
           computed by mass balance.

SCENARIO 11
LOT Phosphorus Only run is where LOT phosphorus controls were implemented throughout the Watershed
           Model and LOT phosphorus limits were  specified at all point sources (0.075 mg/1). Point source
           nitrogen was left at Base Case. Although Scenario  11 has the same N & P overall removal rate, this
           run differs from Scenario #20 where there was a 10% N and 49% P removal uniformly applied to all
           tributaries. Atmospheric loads were consistent with Base Case Scenario and  ocean boundary
           conditions were computed by mass balance.

 SCENARIO 12
65% Limit of Technology 65% Limit of Technology with Clean Air Act run was made using a load reduction
           from the Base Case equivalent to 65% of the difference between limit of technology loads and Base
           Case loads.  Additionally, segment dependent reductions in atmospheric loads of nitrate over the
           land surface and non-tidal portion of the  water surface  in the watershed were applied.
           SCENARIO 13
Allocation 2 - Seasonal BNR This scenario consists of NPS loads at limit of technology and point sources loads
           at three-stage biological nutrient removal (BNR) for the months of May through November. The
           average effluent value for TN is 8.0 mg/1 from May to November, and at base case effluent loads for
           the remaining five months. The average effluent value for TP is 1.5 mg/1 for the entire year.

SCENARIO 14
Allocation 3 This run investigates regional control strategies similar to previously run Scenarios 6 to 8.  Loads to
           Geo-region 1, from Conowingo to Back River, were reduced 73% of the of the difference between
           Base Case and LOT loads. Loads to Geo-region 3, Potomac to mouth, were set at the 40% reduction
           level with clean air act. Atmospheric nitrogen loads were reduced 10%.

SCENARIO 15
50% Nitrogen and Phosphorus Reduction reduces above and below fall line loads of carbon, nitrogen and
           phosphorus to the Bay by 50% each.
                                              A-2

-------
 SCENARIO 16
90% Load Reductions of 1985 carbon, nitrogen and phosphorus loads to the Bay.  Atmospheric loads to all water
            surface are eliminated. Ocean boundary conditions were computed by mass balance.

 SCENARIO 17
90% Nitrogen Reduction reduces existing nitrogen loads to the Bay including atmospheric deposition to the
            water surface. Phosphorus and carbon loads are left at Base Case as were the nitrogen boundary
            conditions at the ocean mouth.

 SCENARIO 18
90% Phosphorus Reduction reduces existing phosphorus loads to the Bay including atmospheric deposition to
            the water surface.  Nitrogen and carbon loads are left  at Base Case as were the phosphorus boundary
            conditions at the ocean mouth.

 SCENARIO 19

31% N -18% P Load Reduction Run where Base Case nitrogen loads to the Bay are reduced 31% while
            phosphorus loads are reduced 18% for Bay Agreement States only. Atmospheric loads were
            consistent with Base Case Scenario and ocean boundary conditions were computed by mass balance.

 SCENARIO 20
10% N _ 49% P Load Reduction Run where Base Case nitrogen  loads to the Bay are reduced 10% while
            phosphorus loads are reduced 49% for Bay Agreement States only. Atmospheric loads were
            consistent with Base Case Scenario and ocean boundary conditions were computed by mass balance.

Tracer Runs These runs trace the transport of dissolved and paniculate substances in the Bay.  It includes
            transport in the nine Bay zones and four of the Bay's major tributaries (Susquehanna, Patapsco-Back,
            Potomac, and James) in addition to the ocean.
                                              A-3

-------
TABLE A - 1
SUMMARY OF SCENARIO NITROGEN AND PHOSPHORUS
LOADINGS
(ALL LOADINGS IN KG/DAY)
Note: Scenarios are listed in the order in which each was calculated. See Table II - 2 for description of scenarios.
NITROGEN PHOSPHORUS
#1-BASE BASE

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#16
90% RED

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#2
40%
Controllable

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#5
LOT

Fall Line
Below Fall Line
Point Source
Wet
360,463
109,824
86,357.
43,025.
599,669


Wet
36,046.
10,982.
8,636.
0
55,664.


Wet
303,387
81,477.
51,814.
43,025.
479,703


Wet
301,733
87,772.
13,153.
Dry
166,182
53,971.
86,357.
43,025.
349,535


Dry
16,618.
5,397.
8,636.
0
30,651.


Dry
143,701
41,704.
51,814.
43,025.
280,244


Dry
140,227
43,629.
13,153.
Average
220,823
53,208.
86,357.
43,025.
403,413


Average
22,082.
5,321.
8,636.
0
36,039.


Average
194,429
40,983.
51,814.
43,025.
330,251


Average
186,722
42,371.
13,153.
Mean
226,895
64,836.
86,357.
43,025.
421,113


Mean
22,689.
6,484.
8,636.
0
37,809.


Mean
195,929
49,370.
51,814.
43,025.
340,139


Mean
191,126
51,954.
13,153.





























Fall Line
Below Fall Line
Point Source
Atmosphere
Total

90% RED

Fall Line
Below Fall Line
Point Source
Atmosphere
Total

40%
Controllable

Fall Line
Below Fall Line
Point Source
Atmosphere
Total

LOT

Fall Line
Below Fall Line
Point Source
Wet
25,792.
6,386.
7,359.
1,823.
41,360.


Wet
2,579.
639
736
0
3,954.


Wet
18,483.
4,131.
4,416.
1,823.
28,853.


Wet
16,165.
4,359.
274
Dry
11,584.
5,142.
7,359.
1,823.
25,908.


Dry
1,158.
514
736
0
2,408.


Dry
8,288.
3,190.
4,416.
1,823.
17,717.


Dry
5,971.
3,245.
274
Average
13,050.
3,779.
7,359.
1,823.
26,011.


Average
1,305.
378
736
0
2,419.


Average
9,215.
2,452.
4,416.
1,823.
17,906.


Average
6,712.
2,524.
274
Mean
15,012.
4,846.
7,359.
1,823.
29,040.


Mean
1,501.
485
736
0
2,722.


Mean
10,698.
3,083.
4,416.
1,823.
20,020.


Mean
8,306.
3,179.
274
A-4

-------
NITROGEN PHOSPHORUS
#5 (Cont.)
Atmosphere
Total
#19
31%N-18%P

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#20
10%N-49%P

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#6
LOT - Upper

Fall Line, Upper
Fall Line,
Middle
Fall Line, Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
it src, Lower
Atmosphere
Total
#7
LOT - Mid


43,025.
445,683


Wet
248,719
75,779.
59,586.
43,025.
427,109


Wet
324,417
98,842.
11,121.
43,025.
544,005


Wet
205,359
90,019.
30,006.
3,901.
71,106.
33,092.
63
56,263.
29,738.
43,025.
562,572


Wet

43,025.
240,034


Dry
114,666
37,240.
59,586.
43,025.
254,517


Dry
149,564
48,574.
77,721.
43,025.
318,884


Dry
95,385
43,475.
12,525.
2,349.
32,624.
17,755.
63
56,263.
29,738.
43,025.
333,202


Dry

43,025.
285,271


Average
152,368
36,714.
59,586.
43,025.
291,693


Average
198,741
47,887.
77,721.
43,025.
367,374


Average
148,782
38,527.
9,910.
2,938.
34,398.
14,590.
63
56,263.
29,738.
43,025.
378,234


Average

43,025.
299,259


Mean
156,557
44,737.
59,586.
43,025.
303,906


Mean
204,205
58,353.
77,721.
43,025.
383,304


Mean
138,739
50,805.
14,975.
2,895.
41,030.
19,556.
63
56,263.
29,738.
43,025.
397,089


Mean





































Atmosphere
Total

31%N-18%P

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#20
10%N-49%P

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#6
LOT - Upper

Fall Line, Upper
Fall Line,
Middle
Fall Line,
Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
Pt src, Lower
Atmosphere
Total
#7
LOT - Mid


1,823.
22,621.


Wet
21,149.
5,237.
6,034.
1,823.
34,243.


Wet
13,154.
3,257.
3,753.
1,823.
21,987.


Wet
6,748.
10,108.
3,932.
249
3,818.
2,055.
2
2,431.
4,858.
1,823.
36,024.


Wet

1,823.
11,313.


Dry
9,499.
4,216.
6,034.
1,823.
21,573.


Dry
5,908.
2,622.
3,753.
1,823.
14,106.


Dry
2,540.
4,366.

1,699.
162
2,960.
1,605.
2
2,431.
4,858.
1,823.
22,446.


Dry

1,823.
11,333.


Average
10,701.
3,099.
6,034.
1,823.
21,657.


Average
6,656.
1,927.
3,753.
1,823.
14,159.


Average
4,106.
3,669.

1,152.
231
2,555.
780
2
2,431.
4,858.
1,823.
21,607.


Average

1,823.
13,583.


Mean
12,310.
3,973.
6,034.
1,823.
24,141.


Mean
7,656.
2,471.
3,753.
1,823.
15,703.


Mean
4,008.
5,236.

1,927.
207
2,970.
1,365.
2
2,431.
4,858.
1,823.
24,826,


Mean
A-5

-------
NITROGEN PHOSPHORUS
#7 (Cont.)
Fall Line, Upper
Fall Line,
Middle
Fall Line, Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
Pt src, Lower
Atmosphere
Total
#8
LOT-Lower

Fall Line, Upper
Fall Line,
Middle
Fall Line, Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
Pt src, Lower
Atmosphere
Total
#10
LOT N ONLY

Fall Line
Below Fall Line
Point Source
Atmosphere
Total


240,516
71,831.
30,006.
5,627.
56,431.
33,092.
356
9,348.
29,738.
43,025.
519,970


Wet
240,516
90,019.
24,543.
5,627.
71,106.
27,444.
356
56,263.
3,744.
43,025.
562,643


Wet
294,285
87, 773.
13,153.
43,025.
438,236


110,180
34,293.
12,525.
3,592.
26,487.
17,755.
356
9,348.
29,738.
43,025.
287,299


Dry
110,180
43,475.
10,550.
3,592.
32,624.
14,794.
356
56,263.
3,744.
43,025.
318,603


Dry
128,868
43,629.
13,153.
43,025.
228,675


172,380
30,181.
9,910.
4,219.
27,738.
14,590.
356
9,348.
29,738.
43,025.
341,485


Average
172,380
38,527.
7,759.
4,219.
34,398.
11,718.
356
56,263.
3,744.
43,025.
372,389


Average
179,081
42,371.
13,153.
43,025.
277,630


161,127

14,975.
4,250.
32,976.
19,556.
356
9,348.
29,738.
43,025.
355,508


Mean
161,127
50,805.
12,232.
4,250.
41,030.
16,094.
356
56,263.
3,744.
43,025.
388,925


Mean
182,037
51,955.
13,153.
43,025.
290,169





































Fall Line, Upper
Fall Line,
Middle
Fall Line,Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
Pt src, Lower
Atmosphere
Total
#8
LOT- Lower

Fall Line, Upper
Fall Line,
Middle
Fall Line,
Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
Pt src, Lower
Atmosphere
Total
#10
LOT N ONLY

Fall Line
Below Fall Line
Point Source
Atmosphere
Total


11,760.
6,656.
3,932.
483
2,751.
2,055.
71
181
4,858.
1,823.
34,570.


Wet
11,760.
10,108.
2,761.
483
3,818.
1,361.
71
2,431.
92
1,823.
34,708.


Wet
19,618.
4,359.
7,324.
1,823.
33,124.


5,516.
2,338.
1,699.
353
2,062.
1,605.
71
181
4,858.
1,823.
20,506.


Dry
5,516.
4,366.
1,093.
353
2,960.
1,021.
71
2,431.
92
1,823.
19,726.


Dry
8,775.
3,245.
7,324.
1,823.
21,167.


8,229.
2,017.
1,152.
436
1,821.
780
71
181
4,858.
1,823.
(2.)


Average
8,229.
3,669.
588
436
2,555.
472
71
2,431.
92
1,823.
20,366.


Average
9,813.
2,524.
7,324.
1,823.
21,484.


7,850.
3,073.
1,927.
412
2,103.
1,365.
71
181
4,858.
1,823.
23,664.


Mean
7,850.
5,236.
1,225.
412
2,970.
869
71
2,431.
92
1,823.
22,978.


Mean
11,359.
3,179.
7,324.
1,823.
23,685.

A-6

-------
NITROGEN PHOSPHORUS
#11
LOT P ONLY

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#18
90% P ONLY

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#3
40% + CAA

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#4
40% CAA+
BASIN

Fall Line
Below Fall Line
3oint Source
Atmosphere
Total
#9
LOG-MID (A)

Fall Line, Upper


Wet
318,761
87,772.
85,418.
43,025.
534,976


Wet
360,463
109,824
86,357.
43,025.
599,669


Wet
293,485
78,426.
51,814.
40,428.
464,153


Wet
269,338
75,508.
51,814.
40,428.
437,088


Wet
240,516


Dry
155,346
43,629.
85,418.
43,025.
327,418


Dry
166,182
53,971.
86,357.
43,025.
349,535


Dry
138,238
39,946.
51,814.
40,428.
270,426


Dry
127,956
39,384.
51,814.
40,428.
259,582


Dry
110,180


Average
202,772
42,371.
85,418.
43,025.
373,586


Average
220,823
53,208.
86,357.
43,025.
403,413


Average
187,334
39,214.
51,814.
40,428.
318,790


Average
171,883
38,474.
51,814.
40,428.
302,599


Average
172,380


Mean
206,999
51,954.
85,418.
43,025.
387,397


Mean
226,895
64,836.
86,357.
43,025.
421,113


Mean
188,926
47,349.
51.814.
40,428.
328,517


Mean
173,803
46,245.
51,814.
40,428.
312,290


Mean
161,127





































LOT P ONLY

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#18
90% P ONLY

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#3
40% + CAA

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#4
40% CAA+
BASIN

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#9
LOT-MID (A)

Fall Line, Upper


Wet
16,164.
4,359.
274
1,823.
22,620.


Wet
2,579.
639
736
182
4,136.


Wet
18,483.
4,131.
4,416.
1,823.
28,853.


Wet
16,459.
3,986.
4,416.
1,823.
26,684.


Wet
11,760.


Dry
5,969.
3,245.
274
1,823.
11,311.


Dry
1,158.
514
736
182
2,591.


Dry
8,287.
3,190.
4,416.
1,823.
17,716.


Dry
7,265.
3,103.
4,416.
1,823.
16,607.


Dry
5,516.


Average
6,711.
2,524.
274
1,823.
11,332.


Average
1,305.
378
736
182
2,601.


Average
9,215.
2,452.
4,416.
1,823.
17,906.


Average
8,088.
2,376.
4,416.
1,823.
16,703.


Average
8,229.


Mean
8,305.
3,179.
274
1,823.
13,581.


Mean
1,501.
485
736
182
2,904.


Mean
10,697.
3,083.
4,416.
1,823.
20,019.


Mean
9,433.
2,989.
4,416.
1,823.
18,661.


Mean
7,850.

A-7

-------
NITROGEN PHOSPHORUS
#9 (Cont.)
Fall Line, Mid.
Fall Line, Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
Pt src, Lower
Atmosphere
Total
#13
BNR

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#14
ALLOC. 2

Fall Line, Upper
Fall Line,Mid.
Fall Line, Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
Pt src, Lower
Atmosphere
Total

88,520.
30,006.
5,627.
59,803.
33,092.
356
33,330.
29,738.
43,025.
564,013


Wet
309,179
87,772.
62,692.
40,428.
500,071


Wet
209,291
73,657.
20,813.
4,361.
81,276.
40,501.



38,723.
468,622

42,208.
12,525.
3,592.
28,111.
17,755.
356
33,330.
29,738.
43,025.
320,820


Dry
144,258
43,629.
62,692.
40,428.
291,007


Dry
96,494.
34,832.
8,968.
2,740.
49,810.
30,930.



38,723.
262,497

37,263.
9,910.
4,219.
29,595.
14,590.
356
33,330.
29,738.
43,025.
374,406


Average
193,508
42,392.
62,692.
40,428.
339,020


Average
150,352
30,705.
7,130.
3,312.
51,027.
28,198.



38,723.
309,447

49,492.
14,975.
4,250.
35,043.
19,556.
356
33,330.
29,738.
43,025.
390,893


Mean
196,942
0
62,692.
40,428.
352,025


Mean
140,597
40,946.
10,602.
3,293.
56,590.
31,751.



38,723.
322,502


































Fall Line, Mid.
Fall Line,Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
Pt src, Lower
Atmosphere
Total
#13
BNR

Fall Line
Below Fall Line
Point Source
Atmosphere
Total
#14
ALLOC. 2

Fall Line, Upper
Fall Line.Mid.
Fall Line,Lower
BFL, Upper
BFL, Middle
BFL, Lower
Pt Src, Upper
Pt Src, Middle
Pt src, Lower
Atmosphere
Total

9,864.
3,932.
483
3,020.
2,055.
71
486
4,858.
1,823.
38,352.


Wet
18,212.
4,359.
3,695.
1,823.
28,089.


Wet
8,120.
7,630.
2,589.
331
3,811.
4,200.



1,823.
28,504.

4,142.
1,699.
353
2,324.
1,605.
71
486
4,858.
1,823.
22,877.


Dry
7,652.
3,245.
3,695.
1,823.
16,415.


Dry
3,354.
2,910.
1,128.
231
3,087.
3,910.



1,823.
16,443.

3,540.
1,152.
436
2,007.
780
71
486
4,858.
1,823.
23,382.


Average
8,645.
2,525.
3,695.
1,823.
16,688.


Average
5,234.
2,483.
772
306
2,791.
3,383.



1,823.
16,792.

5,046.
1,927.
412
2,336.
1,365.
71
486
4,858.
1,823.
26,174.


Mean
10,161.
3,180.
3,695.
1,823.
18,859.


Mean
5,059.
3,683.
1,278.
281
3,113.
3,757.



1,823.
18,995.
A-8

-------
TABLE A-2 STEADY STATE RESPONSE MATRICES FOR TRACER RUNS







tesult
n
Concentrations
n
"hese
Segments

















droduce
^cumulations
n
These
Zones

















3roduce
^cumulations
n
'hese
tones















Concentration
Tig-day/m**3-t
>ay
>ay
>ay
jay
>ay
>ay
>ay
say
>ay
am
irk
ap
»t
>at
>aht h







Accumulation
bayl
bay 2
>av3
bay 4
ay 9
bay 5
lay 6
>ay7
bays
>alt h
am
yrk
ap
pot
pat






Accumulation
Day
toy. 	
Day
bay
bay
bay
bay
bay
ay
bait h
am
irk
•at
sot
Mrt





^formalized by
>n
1
2
3
4
g
5
6
7
8
1
1
1
1
1
10







ig-day/kg-day
1
2
3
4
9
5
6
7
8
10
1
1
1
1
1






kg-day/kg-day
1
2
3
4
9
5
6
7
8
10
1
1
1
1
1



Dissolved Loa
Susquehanna
_oad

8 9251 3751
560207816
4.53860617
3 66278356
2 74439993
2.84070476
202991947
1.0937541
0 64273346
0 37053778
1.39847951
1 79699641
1 82372437
3.17993074
53979314
















































's From These
Baltimore H


1 5067631
529100529
4 54429083
3 62956563
2.70417307
2.78951186
1 98146015
1 05873443
061870626
0 3573562
1 34141918
1 74144479
1.76811316
312553337
16 9637523





'article Loads
Susquehanna

0 85498306
011179623
001996046
0 00438633
000104524
000111215
00001658
0 00003448
000001035
001059217
1 801 £-06
00000142
000002329
000039142
000021422




^article Loads
Zone 1

0 77994334
01644414
001335794
000219687
000030923
000036434
00000363
6414E-06
1 864E-06
001866077
3 528E-07
2 S65E-06
4 955E-06
000011666
000009606



Sources
3atuxent


054246811
1 67429746
215601767
344238115
2 55887218
289778153
1 96844519
1 06148967
0 62666257
0 36342166
1 36309767
1.77767547
1 87465896
222 952263
1 68175772





rrom These a
Saltimore H

0 02272921
029413365
004611994
00095708
0 00208722
000234342
00003207
000006282
000001794
061571457
3089E-06
0 00002489
000004277
0 00084373
0 00044433




o the Surface
Zone 2

0 04680254
071892351
015170566
0 02245628
000292525
0 00353481
000034118
0 00005793
00000167
004116202
3 228E-06
0 00002354
0 00004633
000115306
000097135




3otomac


044838388
1 36504297
1 64352168
211890094
201522129
2 82599059
1 8408169
1 04315082
0 62042026
0.35543924
1 36425363
1 85987943
16 4423208
1 56332486
1 39503784





jurces
Dotomac

8616E-07
000003886
000014137
0 00047553
000023149
000156492
0 00032459
0 00007628
000002298
1 180E-06
3 777E-08
00000325
000006091
099140292
4151E-06




of These Zone
Zone 3

000173556
011189774
068147309
0 13766988
001697778
002154469
00020754
000035002
000010131
000241336
00000198
000014305
000028611
000717983
000604674




Rappahannoc


014252631
0 42497332
050496523
058495714
0 63524522
070820201
1 18007613
0 95729735
062247778
0 35540381
1 52688145
77 4342555
041429036
044164916
0 4348746






James

1 086E-09
7105E-08
3411E-07
9612E-07
5505E-08
4 053E-08
00000264
000008365
0 00072773
3105E-09
0 98626712
000001145
6 794E-08
1 799E-07
6 397E-09




,
Zone 4

000039286
0 02318557
011386255
0 51450425
011965372
0 13943402
001247507
000201443
000057635
00005109
000011281
00008146
0 00163634
004042153
001867762




Cork


0 02693308
0 07960751
009438495
010820901
012393984
011917292
013967066
03639541
0 69787662
050100296
31 9340526
0 10844736
0 07484059
0 08246766
008127593






Dcean

000001184
000046145
000164298
000435045
0 02420156
0 01977849
011604252
0 20507373
021392213
000001854
0 02973328
001581669
000051407
000091116
000003092





Zone 9

000005187
0 00220263
00076817
003391252
0 73433428
009169562
0 08777856
0 00955991
0 00237108
000008465
000046363
000308957
000332314
0 00553793
0 00025325




James


0011171
0 03345541
0 04007739
004618113
005268795
005015432
0 05504883
008119127
067400267
238572185
008769808
0 03737102
003155519
0 03547079
00341464




























Zone 5

000008845
000454063
0 02385334
013341736
014636914
0 46475921
009572222
00166754
000481808
000009022
000097017
000713105
00154602
0 06056276
00014873




3cean


012613553
038563223
045843306
052649438
0 60824754
056085334
06707877
0 76706586
0 80452281
045343152
061641912
043092235
03583187
040007106
039851225




























Zone 6

000001296
000092291
000492919
001985488
005966105
0.11531846
0.44764873
0.12594884
003271446
0.00001755
000645092
0.04463407
003041148
0 00540282
00002344


















































Zone 7

3146E-06
000022456
00011803
000449297
00203966
0 02567996
015689221
0 25937677
0 10449043
4220E-06
001805619
006984789
000180119
000112377
00000517


















































tone 8

2011E-06
000008096
000029983
000081293
0004459
0 00365147
002090618
004418114
012878187
3181E-06
0 03244236
0 OO376309
000008901
000016736
5 833E-06
                        A-9

-------
1
i
w
I
UJ
I
o
5?
co
LU

I

















ANOXIC


ANOXIC





O
|


O
s
f


z
1
g
—
V)
I




3
|
co
o
-j

O
V)
VOL-DAYS


oi
LL.
LU
LL
Q
ANNUAL SUM
Of
LU
It
Q
£
SEASONAL


SCENARIO

oL
LU
LU
LU
O
ANNUAL SUM

LL
LU
u!
i

o
UJ
LU
i
i

i
Billion m*34
z
V)

NUMBER

01
i
(0

LU
1
V)
NUMBER


0

o

T-




o

o




8
S
CM
CM
T-




8
o

CM



$
a>
8
CO
••-




8
o
o
CO
CO

5
T-
Z
co
%


i-


8
o
1
8
o
8




o

o

CM




0

O

CM


8
8
O
CM
CM




is.

CO
CM
CM


%
O)
8
CO
CM




CO
<0
8
CO
CM
CM
a'
§
8
s
s

CM


8
R'
§
So



CM


O

O

CO




O

O

CO


R
,
in
CM
CO




5


CM
CO


8
CM
CM
CO
CO




8
O)
8
CO

s
co
1
S
R
R

CO


s
CM
8
8
CM

CO


O

O

•*•




o

o





8
o>
CM
•*




^

00
CM



CO*
o>
T-
8
CO
•




0

o

in




o

0

in


Q
.
8
o
CM
in




5

in
CM
in


R
£
R
CO
in




00

8
CO
in
In
S
8
CM'
r-
£
S


in


8
co
8
s
9


in


o

o

CO




o

o

(O


8
8
o
CM
(O




5
8
a
CM
CD


S
9
d
CO
CO




$



*—
s
co
CM
o>




s
CO
00
CM
o>


K
8
o
CO
O)




s
o>
8
CO
o>
8
8
8
5
ts.

o>


CM

CD
CO


o>


o

o

8




o

o

o


s
in

CM
Pi




8
s
CO
CM
o


s
0)
8
CO
Pi




CO

s
CO
o
CO
00
CO
8
8
a

Pi


00
8
CO
8

00

o
•t-

-------

SCE
^
                      8
                      8':
                      88'
                             in
                             oci
                        88:
                                       o
                                       +"
                                                      8!
                                                8;
                                                Hi
                          88
                                                         8!
                                                       + i
                                                       %'
                                                       g:
                                     a
                                                      s
                                                ai
            o
            +
                     £
              g
              g
              in
                                 o
                                 T
                                            in
                                            CM
             *
             CO
             •»
                                               o
                                               •r-
                                               +
                                                                 S
          s
                                                         s
                                           8
S
8
8
8
                   $
          K
8
8
8
6
S
                        8
         8
8
8
8
8
3
3
S
                                                                             3
                                                                     8'
g
8
S
8
8
                                     i
                               g
                               CM
                                            O

                                            +"
                                              a
                                                  8
                                          8
                                                         "
                                                ai
                                                CM
                                           8
                                           S
                            8
        !«o
        f<
               8
                 8
        I
o


I
in
                       ™,
m
                         «N
                                        IN
 s
i in
                                              3

                                              7

                                                        O

                                                        7
      8
SCE

-------
                               APPENDIX B

        REGRESSION EQUATIONS FROM RESPONSE SURFACE ANALYSES



ANALYSIS CONDITION AND EQUATION                       R-SQUARE

WHOLE BAY FULL YEAR 9
a=-.027648 +1.159833*n+ .267241*p - .308414*n*p;                       0.9804

 WHOLE BAY SPRING YEAR 9
a = .020025 + 3.331184 * n - 2.488719 * N*n + .996558 * p*p - .986442 * n*p ;   0.9895

WHOLE BAY SUMMER YEAR 9
a = -.014463 + 1.146319 * n + .205676 * p*p -  .216003 * n*p ;                0.9839

WHOLE BAY FALL YEAR 9
a = -.053316 + 2.028102 * n - .948373 * n*n +  .54995 * p*p - .541366 * n*p ;     0.9816

ZONE 1 FULL YEAR 9
a = .013408 + .964316 * n + . 144493 * n*n + .448454 * p*p -.460693 * n*p ;      0.9990

ZONE 1 SPRING YEAR 9
a = .093315 + 3.571793 * n - 2.854708 * n*n+  1.147789 * p*p - 1.138452 * n*p ;  0.9486

ZONE 1 SUMMER YEAR 9
a = .001148 + . 184689 * n + 1.026181 * n*n + .285626 * p*p - .30611 * n*p ;     0.9993

ZONE 1 FALL YEAR 9
a = -.004358 + .853158 * n + .29092 * n*n + .360827 * p*p - .371816 * n*p ;      0.9991

ZONE 2 FULL YEAR 9
a = -.048278+ 1.16711 *n+.408651 * p*p - .409146 * n*p ;                 0.9846

 ZONE 2 SPRING YEAR 9
a = -.018261 + 3.267295 * n - 2.36681 * n*n +  1.209613 * p*p - 1.198076 * n*p ;  0.9895

ZONE 2 SUMMER YEAR 9
a = -.104308 + 1.205255 * n + .36289 * p*p - .355202 * n*p ;                 0.9800

ZONE 2 FALL YEAR 9
a = -.065619 + 2.2201  * n - 1.148448 * n*n + .668454 * p*p - .649709 * n*p ;     0.9676

ZONE 3 FULL YEAR 9
a-.017157+ 1.130903 *n + .301765 * p*p-.315226 * n*p ;                 0.9666
ZONE 3 SPRING YEAR 9
a = .11723 + 3.601399 * n - 2.917184 * n*n+ 1.112482 * p*p - 1.107638 * n*p ;   0.9269
                                   B-l

-------
ZONE 3 SUMMER YEAR 9
a = -.075862 + 1.462213 * n - .289957* N*N + .325289 * p*p - .322662 * n*p ;     0.9730

ZONE 3 FALL YEAR 9
a = -.078155+ 2.891246 *n- 1.872233 * n*n +.709474 * p*p - .692031 * n*p;     0.9415

ZONE 4 FULL YEAR 9
a = .072407 + 2.13494 * n + 1.477261 * p - 1.221717 * n*n - .798939 * p*p -.860225 * n*p ;
                                                                  0.9405
ZONE 4 SPRING YEAR 9
 a =.112264+ 1.720432 *n + 2.226041 * p - .808812 * n*n- 1.406036 * p*p - 1.100668 * n*p
                                                                  0.8776

ZONE 4 SUMMER YEAR 9
a = .066993 + 2.195415 * n +  1.369103 * p - 1.28241 * n*n - .710283 * p*p - .826618 * n*p ;
                                                                  0.9483
ZONE 4 FALL YEAR 9
a=.11823 + 1.608643 *n + 2.38506 *p-. 691964 * n*n - 1.557117* p*p - 1.128823 * n*p ;
                                                                   0.8686
ZONE 5 FULL YEAR 9 AND SUMMER (SAME EQUATIONS)
a = .028591 + 3.308827 * n - 2.475876 * n*n + .313846 * p*p - .310162 * n*p ;      0.9905

ZONE 6 FULL YEAR 9 AND SUMMER (SAME EQUATIONS)
a = -.023671 + 3.705698 * n -  2.84665 * n*n ;                               0.9769

ZONE 9 FULL YEAR 9
a = -.026057 + 1.767562 * n -  .69466 * n*n ;                                0.9852

ZONE 9 SPRING YEAR 9
a = -.040017 + 2.561638 * n -  1.538052 * N*n ;                              0.9169

ZONE 9 SUMMER YEAR 9
a = -.02601 + 1.640766 * n - .555784 * n*n ;                                0.9884

ZONE 9 FALL YEAR 9
a = -.006908 + 2.521603 * n -  1.55424 * n*n ;                               0.9983

DO HABITAT IMPROVEMENT
a = -.005567 + 0.648366 * n -  0.226809* n*n + 0.162846*p*p-0.156327*n*p       0.9977

DIN HABITAT IMPROVEMENT
a = -.018216 + 3.36393 * n - 1.520132* n*n-0.849344*p                      0.9983
                                    B-2

-------
U.S. Environmental Protection Agency
  Chesapeake Bay Program Office
        410 Severn Avenue
       Annapolis, MD21403
        1-800-YOURBAY

-------