close

Вход

Забыли?

вход по аккаунту

?

40569%282001%29170

код для вставкиСкачать
+CALIBRATION AND VERIFICATION DATA COLLECTION
FOR THE CHICAGO WATERWAY SYSTEM
Development of a Water Quality Model for Unsteady-State Flow
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
Richard Lanyon (1) and Charles S. Melching (2)
(1) Metropolitan Water Reclamation District of Greater Chicago, 100 East Erie Street, Chicago,
IL 60611; PH (312) 751-5190; FAX (312) 751-5194; email: richard.lanyon@mwrdgc.dst.il.us
(2) Department of Civil and Environmental Engineering, Marquette University, Room 262, P.O.
Box 1881, Milwaukee, WI 53201-1881; PH (414) 288-6080; FAX (414) 288-7521; email:
charles.melching@marquette.edu
Abstract
The flow and water quality processes in the Chicago Waterway System (CWS) are very
complex and critical water quality conditions may result under a wide range of flows. The
Metropolitan Water Reclamation District of Greater Chicago (District) will soon be faced with a
number of difficult water quality management problems including the impact of reduced
discretionary diversions from Lake Michigan for water quality improvement in the summer,
assessment of water quality impairment, and development of total maximum daily load
allocations. A water quality model capable of simulating water quality processes under
unsteady-state flow is being developed to assist water quality management and planning decision
making.
The CWS is a 122 kilometer (76-mile) network of navigable waterways controlled by
hydraulic structures in which the majority of flow is treated sewage effluent. The dominant uses
of the CWS are for commercial and recreational navigation and for urban drainage.
Critical to the development of any water quality model is the adequacy of data for
calibration and verification. Due to the complexity of the CWS, the District will increase the
already extensive amount of quality and quantity data collected. The United States Geological
Survey (USGS) has recently established discharge and stage gages at the three CWS inlet
controls and has operated one near the outlet control for several years. In addition, four tributary
stream inflows are gaged by the USGS and the District measures the treated effluent discharges
at four water reclamation plant (WRP) outfalls.
The District regularly samples constituent concentrations in the discharge from its WRPs,
providing the primary loads that drive the water quality relations in the CWS, and collects
monthly grab samples for analysis, providing periodic instream water quality data at 19
locations. The USGS and District have collected limited quality and quantity data to
characterize wet weather loading from combined sewer overflows. A network of continuous
temperature and dissolved oxygen (DO) in-situ monitors has been installed by the District
throughout the CWS. Monitored data collection began in August 1998 and will continue through
the period required for model development and testing. This rarely available continuous data
will allow rigorous calibration of the water quality model.
Because little information is available on bed sediment, sediment quality, and sediment
oxygen demand in the CWS; the District will begin to collect these data to confirm the default
values used in the models or provide new values. In addition, the District will perform wet
1
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
weather monitoring for peak flow events to properly characterize dynamic transient waterquality characteristics.
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
Introduction
The District is a special purpose local government agency created by the Illinois General
Assembly in 1889 to protect the water quality of Lake Michigan by diverting municipal
wastewater from the Chicago area across the subcontinental divide to the Illinois and Mississippi
Rivers. Today, the District provides wastewater treatment services for an area of 2,260 square
kilometers (872 square miles), which includes the City of Chicago and 124 suburban
communities in Cook County. The District is a regional agency and has a corporate structure
separate from the City of Chicago and Cook County governments. The District’s 885 kilometers
(550 miles) of intercepting sewers collect sewage from local municipalities. The District’s seven
(Calumet, Egan, Hanover Park, Kirie, Lemont, North Side and Stickney) WRPs treat an average
of 61.7 cubic meters per second (1.4 billion gallons) of wastewater per day.
Approximately 122 kilometers (76 miles) of navigable waterways are controlled by the
District to provide for drainage of treated municipal wastewater and urban stormwater runoff.
Over four-fifths of these waterways are part of the Illinois Waterway, a federal inland navigation
system. The CWS is controlled through the use of locks, sluice gate structures and a
hydroelectrical generation station. The CWS conveys runoff from an approximately 2,000
square kilometer (700 square mile) area, treated wastewater effluent from four of the District’s
WRPs (Calumet, Lemont, North Side and Stickney) and diversions directly from Lake Michigan
into the Illinois and Mississippi Rivers via the Des Plaines River.
Since 1972, the District has pursued an aggressive capital improvement program to
expand and increase the capacity of the WRPs to comply with the Clean Water Act and to meet
water quality standards. In addition, deficiencies in natural waterway reaeration have been
supplemented with mechanical aeration at seven locations in the CWS. To protect the CWS and
Lake Michigan from the deleterious effects of combined sewer overflows (CSOs), the Tunnel
and Reservoir Plan (TARP) has been implemented to convey, store and treat excess CSOs before
discharge.
Despite past improvements in water quality in the CWS, there are several challenges to
continued compliance with water quality standards in the future. These include:
•
Diminished allowable diversion of high quality water from Lake Michigan into
the CWS during warm-weather periods.
•
Waterway system flows which are predominantly treated municipal sewage
effluent.
•
Tributary watershed inflows which contain nonpoint source pollutants.
•
Continued pollution from occasional CSOs.
•
Legacy pollutants in sediments which exert an uncertain water quality impact.
•
Commercial navigation which limits the development of quality aquatic habitat.
2
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
•
Increased recreational use, which has resulted from enhanced water quality, has
focused attention on further water quality improvement.
•
Nutrient control criteria may require additional point source wastewater treatment
capacity.
•
Total maximum daily load determinations will require knowledge of the nonpoint
source pollutant contributions and impacts.
Past District studies have relied on a steady-state flow water quality model for capital
improvement planning purposes. The foregoing challenges are significantly impacted by the
hydrologic and hydraulic dynamics of the CWS. Therefore, a model capable of simulating the
dynamics of the CWS is needed
.
Model Development
Water quality decisions to be made in the 21st Century would be greatly aided by the use
of a model that can simulate water quality processes under unsteady-state flow conditions.
Therefore, the objective of this project is to develop, calibrate, and verify a water quality model
for unsteady-state flow conditions. This development, calibration, and verification will rely
heavily on the data collected, including continuous DO data to ensure that the model properly
simulates the water quality processes in the CWS.
Development of the model will proceed sequentially with the first effort devoted to the
routing of unsteady-state flow in the CWS. Once this part is calibrated and verified, attention
will be focused on the water quality processes. A considerable amount of both hydraulic and
water quality data are available for the CWS and are reviewed in this paper.
Hydraulic and Hydrologic Data
The USGS has established discharge and stage gages at the three points where water is
diverted from Lake Michigan into the CWS. The data from these gages provide the primary
upstream boundary conditions for the model. Further, data from the USGS discharge and stage
gage on the Chicago Sanitary and Ship Canal (CSSC) at Romeoville and other data at the
Lockport Powerhouse and Lock collected by the District provide a reliable downstream
boundary condition for the model. Two other secondary boundary conditions are provided by
the USGS discharge and stage gages located on the North Branch of the Chicago River at Albany
Avenue upstream from the confluence with the North Shore Channel, and the Little Calumet
River at Cottage Grove Avenue upstream from the confluence with the Cal-Sag Channel. These
locations are shown in Figure 1. In addition, USGS stream flow data for two gaged tributaries to
the Calumet-Sag Channel, Midlothian and Tinley Creeks will be used to estimate inflows from
ungaged tributary watersheds.
Hourly flow data are available from the District for the treated effluent discharged to the
CWS by each of the four WRPs. In addition, hourly flows discharged to the CWS at three CSO
pumping stations will be estimated from operating logs. There are numerous points at which
CSOs occur, but no flow data are available. However, the times of gate operations are known
from operating logs and these records will be compiled and used to corroborate the times of
3
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
significant flow changes. Although flows in the various branches of the CWS are not measured,
water level is recorded at 12 locations and these records will be used for calibration and
verification. Flow and water level data for calibration and verification of the hydraulic part of
the model are summarized in Tables 1 and 2.
4
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Table 1. Water flow data
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
Waterway/Reach
Facility/Location
CSSC
Lockport L&D (b)
Romeoville Road
Lemont WRP
Stickney WRP
SOUTH BRANCH
Racine Avenue PS (c)
CHICAGO RIVER
Columbus Drive
NORTH BRANCH
North Branch PS
Albany Avenue
Touhy Avenue (b) (c)
NORTH SHORE
CHANNEL
North Side WRP
Maple Avenue
CALUMET-SAG
CHANNEL
Tinley Creek (c)
Midlothian Creek (c)
LITTLE CALUMET
RIVER
Cottage Grove Avenue (c)
125th Street PS
Calumet WRP
O’Brien L&D
Distance Upstream
of Lockport
Kilometers (Miles)
0.0 (0.0)
Data Source and
Measurement Type
Data Frequency
Minutes
8.4 (5.2)
15.4 (9.6)
39.4 (24.5)
District, AVM &
Estimated
USGS, AVM
District, Venturi
District, Venuri
60
49.4 (30.7)
District, Estimated
57.1(35.5)
USGS, AVM
67.8 (42.1)
68.6 (42.6)
80.6 (50.1)
District, Estimated
USGS, SDR
USGS, SDR
73.9 (45.9)
81.1 (50.4)
District, Venturi
USGS, AVM
60
5
37.0 (23.0)
41.8 (26.0)
USGS, SDR
USGS, SDR
15
15
56.8 (35.3)
48.9 (30.4)
48.9 (30.4)
57.1 (35.5)
USGS, SDR
District, Estimated
District, Venturi
USGS, AVM
15
60
60
Variable
5
Variable
15
15
15
Variable
60 (a)
5
WRP = Water Reclamation Plant
PS = Pumping Station
L&D = Lock and Dam
(a) = Three Observations Per Day Prior to 2001
(b) = Back-up for Another Location
(c) = Location Is Upstream of CWS
AVM = Acoustic Velocity Meter
SDR = Stage Discharge Relationship
5
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Table 2. Water level data
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
Waterway Reach/Location
CSSC
Lockport L&D
Controlling Works
Romeoville Road
Sag Junction
Western Avenue
CHICAGO RIVER
Columbus Drive
Controlling Works
NORTH BRANCH
Lawrence Avenue
NORTH SHORE
CHANNEL
Maple Avenue
Sheridan Road
CALUMET-SAG
CHANNEL
Southwest Highway
LITTLE CALUMET
RIVER
O’Brien L&D
O’Brien L&D
Distance Upstream
of Lockport
Kilometers (Miles)
Data Source and
Measurement Type
Data Frequency
Minutes
0.0 (0.0)
3.5 (2.2)
8.4 (5.2)
20.3 (12.6)
47.5 (29.5)
District,
District,
USGS,
District,
District,
60
60
15
60
60
57.1 (35.5)
58.3 (36.2)
USGS,
District,
5
60
67.8 (42.1)
District
60
81.1 (50.4)
81.8 (50.8)
USGS,
District,
5
60
31.7 (19.7)
District,
60
57.1 (35.5)
57.1 (35.5)
District,
USGS,
60
5
L&D = Lock and Dam
Water Quality Data
Data for calibration of the water quality part of the model will be provided by the District
and, to a lesser extent, from the Illinois Environmental Protection Agency and Midwest
Generation, Inc. The District regularly samples constituent concentrations in the treated effluent
discharged from its WRPs. These data provide the primary loads that drive the water quality
relations in the CWS during low flows without the contribution of CSOs and other nonpoint
sources. The District also has an ambient water quality monitoring network (AWQMN) in
which grab samples are collected monthly at 19 locations in the CWS and analyzed for a wide
array of constituents.
In 1997, the District installed a network of 20 continuous temperature and DO monitors
along the North Shore Channel, the North and South Branches, the Chicago River and the CSSC.
6
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
Data collection began in August 1998 and will continue through 2002. The District has added to
this DO monitoring network in 2000 to provide 2 locations along the Chicago River and in 2001,
12 locations along the Little Calumet River and Calumet-Sag Channel. These data will allow
rigorous calibration of the water quality model. Water quality data for calibration and
verification of the model are summarized in Table 3.
To estimate the loads from CSOs at the three pumping stations, the District will conduct
sampling during CSO events at each station. This data also will be used to estimate the loading
from CSOs at numerous unmeasured outfalls. Sampling of the two significant tributary inflows
is included in the District’s AWQMN and historic data will be used to estimate these loads. In
addition, the District will sample these two tributaries during periods of significant storm runoff
to provide more detailed data for these short-term, high flow periods.
There are numerous points at which nonpoint sources discharge to the CWS and this
loading will be estimated. However, the District is attempting to identify the location where this
type of inflow is most prominent. Preliminary modeling results may aid in this identification.
Once identified, sampling these locations during periods of significant runoff will be attempted
in an effort to characterize the nonpoint source loading.
Table 3. District instream water quality data
Waterway Reach/Location
Distance Upstream
of Lockport
Kilometers (Miles)
Data Type
0.2 (0.1)
8.4 (5.2)
15.3 (9.5)
18.2 (11.3)
21.1 (13.1)
34.1 (21.2)
37.0 (23.0)
42.3 (26.3)
48.4 (30.1)
AWQ, DO
DO
AWQ
DO
AWQ, DO
DO
AWQ
AWQ, DO
AWQ
49.7 (30.9)
54.7 (34.0)
55.2 (34.3)
DO
DO
AWQ
56.3 (35.0)
56.8 (35.3)
57.6 (35.8)
58.3 (36.2)
DO
DO
AWQ
DO
CSSC
Lockport Forebay
Romeoville Road
Stephens Street
Argonne Intake
Route 83
Baltimore & Ohio Railroad
Harlem Avenue
Cicero Avenue
Damen Avenue
SOUTH BRANCH
Loomis Street
Jackson Boulevard
Madison Street
CHICAGO RIVER
Clark Street
Michigan Avenue
Lake Shore Drive
Controlling Works
7
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Table 3. District instream water quality data (continued)
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
Waterway Reach/Location
Distance Upstream
of Lockport
Kilometers (Miles)
Data Type
56.0 (34.8)
56.3 (35.0)
58.4 (36.3)
61.8 (38.4)
62.9 (39.1)
65.0 (40.4)
67.3 (41.8)
67.8 (42.1)
DO
AWQ
DO
DO
AWQ
DO
AWQ
DO
71.1 (44.2)
72.7 (45.2)
75.2 (46.7)
78.1 (48.5)
80.3 (49.9)
81.4 (50.6)
DO
AWQ
DO
DO
AWQ
DO
20.8 (12.9)
26.6 (16.5)
31.7 (19.7)
33.3 (20.7)
38.5 (23.9)
41.8 (26.0)
45.1 (28.0)
44.1 (27.4)
AWQ, DO
DO
DO
DO
AWQ, DO
DO
AWQ
DO
46.8 (29.1)
50.5 (31.4)
51.0 (31.7)
55.2 (34.3)
57.9 (36.0)
AWQ, DO
AWQ
DO
DO
AWQ, DO
NORTH BRANCH
Kinzie Street
Grand Avenue
Division Street
Fullerton Avenue
Diversey Boulevard
Addison Street
Wilson Avenue
Lawrence Avenue
NORTH SHORE CHANNEL
Devon Avenue
Touhy Avenue
Main Street
Simpson Street
Central Avenue
Linden Street
CALUMET SAG CHANNEL
Route 83
104th Street
Southwest Highway
Harlem Avenue
Cicero Avenue
Kedzie Avenue
Ashland Avenue
Division Street
LITTLE CALUMET RIVER
Halsted Street
Indiana Avenue
C&WI Railroad
Conrail Railroad
130th Street
AWQ = Monthly Grab Sample Analysis for Several Constituents
DO = Hourly Dissolved Oxygen Monitor Observations
8
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
Water Quality Simulations
At a minimum the water quality model will simulate temperature and concentrations of
the following constituents: DO, carbonaceous biochemical oxygen demand (CBOD), suspended
solids, ammonia nitrogen, nitrate nitrogen, phosphorus, and chlorophyll.
The model will include simulation of the resuspension of sediment by medium and high
flow events and the transport of suspended sediments. Steady-state flow water quality models
typically include the effect of oxygen demanding substances in the sediment on the DO in the
overlying water column as a sink term in the oxygen-balance equation. However, for unsteadystate flow it is necessary to consider the possible changes in sediment oxygen demand (SOD)
with flow rate and time.
Because little information is available on bed sediment, sediment quality, and SOD in the
CWS, default values will be applied in the model. These default values will be checked and
modified using available data. Normally, SOD is regarded as a substantial factor in the oxygen
balance in waterways similar to the CWS. As a result the District will collect samples for
sediment quality analysis and perform in-situ SOD measurements at critical locations.
Preliminary modeling results may aid in selecting the SOD measurement locations.
The impact of thermal loads from cooling water discharges (two thermal electric
generating plants and other thermal dischargers) on DO concentrations will be assessed. Further,
the extent to which thermal loads adversely impact the ability to meet the DO standard shall be
determined. Water quality models typically do not include the equations and algorithms for
thermal transport in rivers such as those developed for Midwest Generation, Inc. thermal power
plants along the Upper Illinois Waterway. Water quality models simulate temperature to
properly adjust reaction coefficients and the saturation concentration of DO. The model
developed will reasonably simulate temperature in order to estimate water quality effects of
thermal discharges, but will not include a detailed thermal model of the CWS since thermal
sources do not appear to be significantly impacting DO concentrations in the CWS.
All water quality data available for a period that reflects current conditions (i.e.,
supplemental aeration stations in operation, TARP online, etc.) will be compiled and reviewed.
This review will be used in the establishment of the time series of water quality at the boundaries
of the model. It will also establish the data available for model calibration and verification.
Once the data are organized, the time periods for calibration and verification will be determined.
Because of the availability of the continuous DO concentration data beginning in August 1998
and of continuous stage and discharge records at the boundaries beginning at most locations in
1996, the period beginning August 1998 most likely will be the best for calibration and
verification. However, earlier data may be useful for proper characterization of the operation of
the supplemental aeration stations, establishment of boundary conditions, and general checking
of model simulation results.
Time series of daily load data from the Calumet, Lemont, North Side and Stickney WRPs
will be compiled for the calibration and verification periods. This load data will be
disaggregated to the time interval necessary for routing on the basis of available data on flow and
load variations within a day. If possible, daily load data for other point sources will be estimated
and compiled from NPDES permit records and the Permit Compliance System.
The parameters affecting water quality simulation in the selected model will be calibrated
such that available instream constituent concentration data are matched and the parameter values
remain within reasonable ranges determined from previous experience with the model or
9
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
principles of environmental chemistry. The calibration will proceed step-wise; constituent-byconstituent such that the parameters affecting a given constituent will be determined by its
concentration and that of its breakdown products, but not other related constituents. That is,
nitrogen cycle and phosphorus cycle relations and CBOD reactions will be determined without
regard to DO concentrations.
Continuous DO concentration data have rarely been used in model development because
most models in the U.S. have focused on steady-state, low flows for regulatory purposes;
whereas in Europe, where water quality models for unsteady-state flows are more commonly
used, continuous DO data typically are not available. Therefore, calibration to continuous data
will provide a new challenge in the development of an unsteady-state flow water quality model.
Most likely a new approach to calibration will be developed and new insights on water quality
simulation and processes will be gained.
Initial calibration will primarily focus on periods where nonpoint source flows are not
reaching the CWS. The reason for this is that during periods without nonpoint source flows, the
flows and load to the system are known more accurately and, thus, the model can be more
reliably calibrated. Periods with nonpoint sources including CSOs will then be partially
calibrated. That is, simulations will be made using the best available information on nonpoint
source flows and loads. Small discrepancies between simulated and observed values will be
compensated for by appropriate, small changes in model parameters or assumed loads.
However, larger discrepancies may (1) require significant changes in assumed flows and loads,
(2) indicate that additional data are necessary for proper model calibration during periods with
nonpoint source flows to the CWS, or (3) identify previously undetected flows or loads to the
system.
Loads from all point sources will be inventoried and made a part of the modeling process.
This will include point sources from available NPDES permit records and data for nonpoint
sources of flow and pollutant loads including the measured tributary inflows, CSOs, storm
sewers, highway drains, etc.
As a result of the development, calibration, and verification of the water quality model, a
review will be made for any deficiencies or gaps in the location or frequency of ambient waterquality data in the CWS. Where deficiencies or gaps are found, a recommendation shall be made
for the collection of additional monitoring data to support ongoing use of the model. These
deficiencies and gaps may be identified at any point in the process of model development,
calibration, and verification.
Conclusions
An extensive set of flow, hydraulic, geometry, and water quality data are available for the
development, calibration and verification of a model of water quality under unsteady-state flow
conditions in the CWS. Because the CWS is an artificial system and the primary inflows from
the WRPs are well know, this study should serve as an interesting test for water quality models
for unsteady-state flow conditions. The study also will provide insight on the relative
importance of point sources (for a system with high treatment standards), nonpoint sources, and
legacy pollutants, primarily SOD. The results of this modeling study will be presented at future
specialty conferences and in the refereed literature as they become available. Model
development began in September 2000. Unsteady-state flow routing is expected to be complete
10
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Downloaded from ascelibrary.org by University Of Florida on 10/25/17. Copyright ASCE. For personal use only; all rights reserved.
by June 2001. Water quality modeling for low and medium flows will be completed and
approaches for high flows will be evaluated by August 2002.
11
Copyright ASCE 2004
Bridging the Gap
World Water Congress 2001
Документ
Категория
Без категории
Просмотров
3
Размер файла
95 Кб
Теги
29170, 40569, 282001
1/--страниц
Пожаловаться на содержимое документа