Introduction
| |
 |
| |
Integrated watershed (HSPF) and receiving
water (CH3D-FC) model developed to simulate FC release,
transport, and fate in Sinclair and Dyes Inlets, Puget
Sound, WA. (larger view) |
 |
|
An integrated watershed and receiving water model was developed
to simulate fecal coliform (FC) fate and transport in Sinclair
and Dyes Inlets, Puget Sound, WA. The integrated model consisted
of a watershed model, an empirical FC loading model, and an
estuarine fate and transport model.
The watershed model, consisting of 15 Hydrologic Simulation
Program Fortran (HSPF) submodels, was deployed to simulate
watershed hydrology for streams (open channel flows), stormwater
catchments areas (piped flows), and shoreline drainage areas
(overland flows) for each drainage basin (DSN) within the watershed.
The empirical FC loading model developed from sampling data
gathered from the watershed was used to estimate FC concentrations
in surface streams and stormwater outfalls as a function of
upstream land use and land cover (LULC). Flow and FC concentrations
for discharges from waste water treatment plants (WWTP) were
estimated by interpolating data reported on monthly discharge
monitoring reports (DMRs) submitted by each facility.
The estuarine model, Curvilinear Hydrodynamics in Three Dimensions
(CH3D) previously calibrated to match the hydrodynamics of
the Inlets and modified to include FC kinetics (CH3D-FC), was
used to simulate the release, transport, and fate of FC loading
from watershed pour points corresponding to 39 stream mouths,
44 stormwater outfalls, four WWTP discharges, and 44 shoreline
drainage areas.
The output from HSPF was used as input
to CH3D-FC. The time-varying flows produced by HSPF for each
of the stream, stormwater, and shoreline pour points were used
to calculate the loads based on the loading concentration assigned
to each basin. The loads were read into CH3D-FC along with
the
loads from the WWTPs to simulate the total FC loading for each
simulation scenario. The estuarine CH3D-FC model was run to
simulate the tides, circulation conditions, fresh water, and
FC inputs occurring during individual storm events (10 days)
that
occurred in 2004, and over the course of Water Year 2003 (WY2003)
from October 1, 2002, to September 30, 2003 (364 days).
Note about grids:
CH3D uses a curvilinear grid that is
represented by Cartesian rows and columns. The grid developed
for Sinclair and Dyes Inlets contains 91 rows and 96 columns
(91x96 grid) with a resolution
of about 100-150 m (300-450 ft). A higher resolution
grid was developed to reduce “initial” dilution
in areas of low flushing such as the mouths of Clear, Chico,
and Karcher Creeks, and other areas including Oyster Bay, Ostrich
Bay, Phinney Bay, and near the Shipyard. This higher resolution
grid has 94 rows and 105 columns (94x105
grid) with a resolution of about 30-50 m (100-150 ft)
in those areas.
A total of 20 simulation scenarios were
run to verify model performance, assess sensitivity and uncertainty,
and provide results needed to establish waste load and load
allocations for the Total Maximum Daily Load of FC loading in
Sinclair and Dyes Inlets. Model verification consisted of comparing
model predictions to observed data collected during three storm
events sampled in 2004 and observed data collected during WY2003.
|
