United States
Environmental Protection
Agency
EPA Region 10
(OEA-095)
EPA910-R-03-011
September 2003
A PRELIMINARY ANALYSIS OF THE THERMAL
REGIME OF DWORSHAK RESERVOIR
John Yearsley
EPA Region 10
September 2003
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
TABLE OF CONTENTS
1.0 BACKGROUND 1
1.1 INTRODUCTION 1
1.2 REPORT OBJECTIVES 1
1.3 PROJECT CHARACTERISTICS 4
1.3.1 Water Temperature Control 5
1.3.2 Selective Withdrawal Structures 5
1.3.3 Water Temperature Monitoring 10
1.3.4 Reservoir Thermal Structure 10
1.4 CLEARWATER RIVER WATERSHED CHARACTERISTICS 11
1.4.1 Geography and Soils 11
1.4.2 Climate 11
1.4.3 Hydrology 11
1.5 WATER QUALITY AND QUANTITY ISSUES ASSOCIATED WITH DWORSHAK DAM 12
1.5.1 Water Quality Issues 10
1.5.2 Water Quantity Issues 12
1.5.3 Fisheries 13
2.0 DESCRIPTION OF MATHEMATICAL MODEL 14
2.1 CEQUAL-W2 VERSION 3.1 14
2.2 BATHYMETRY 15
2.3 METEOROLOGY 15
2.4 TRIBUTARY FLOW AND TEMPERATURES 19
2.5 OUTFLOWS 19
3.0 MODEL RESULTS 25
3.1 OUTLET TEMPERATURES25
3.2 VERTICAL TEMPERATURE PROFILES 25
4.0 CONCLUSION AND RECOMMENDATIONS 35
4.1 CONCLUSION 35
4.2 RECOMMENDATIONS 35
5.0 REFERENCES 36
LIST OF FIGURES
Figure 1. Location Map 3
Figure 2. Dworshak Dam face, looking from upstream to downstream 5
Figure 3. Dworshak Dam selective withdrawal structure schematic 6
Figure 4. Dworshak pool limitations in September 7
Figure 5. Two-dimensional (x-z) hydrodynamics model such as that used in CE-QUAL-
W2. Source: Scott A. Wells Department of Civil Engineering Portland State
University Portland, Oregon USA 14
Figure 6. Measured and modeled reservoir widths for Dworshak Reservoir 16
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
Figure 7. Typical reservoir cross-section based on exponential shape 17
Figure 8. Model and observed reservoir volumes 18
Figure 9. Water temperature and stream flow stations in the Clearwater River basin .... 20
Figure 10. Observed and simulated surface elevations in Dworshak Reservoir in 2002 21
Figure 11. Tributary inflows to Dworshak Reservoir in 2002 22
Figure 12. Temperature of tributary inflows to Dworshak Reservoir in 2002 23
Figure 13. Inferred outflow schedule at Dworshak Dam during 2002 24
Figure 14. Simulated and observed outlet temperatures at Dworshak Dam during 2002 25
Figure 15. Simulated and observed water temperatures in Dworshak Reservoir near
Dworshak Dam on June 3 0,2002 27
Figure 16. Simulated and observed water temperatures in Dworshak Reservoir near
Dworshak Dam on July 20, 2002 28
Figure 17. Simulated and observed water temperatures in Dworshak Reservoir near
Dworshak Dam on August 8, 2002 29
Figure 18. Simulated and observed water temperatures in Dworshak Reservoir near
Dworshak Dam on August 28, 2002 30
Figure 19. Simulated longitudinal temperature profiles in Dworshak Reservoir on
June 30, 2002 31
Figure 20. Simulated longitudinal temperature profiles in Dworshak Reservoir on
July 20, 2002 32
Figure 21. Simulated longitudinal temperature profiles in Dworshak Reservoir on
Augusts, 2002 33
Figure 22. Simulated longitudinal temperature profiles in Dworshak Reservoir on
August 28, 2002 34
in
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
ACKNOWLEDGEMENTS
This report relies heavily on a project completed by graduate students in Civil Engineering at the
University of Washington. The project was a curriculum element in CEE577, Water Quality
Management. The following students participated in the completion of the report:
Project Leader
Modeling Group
Background Group
Data Group
Scenarios Group
Presentation
Ani Kameenui
Konstantinos Andreadis
Matthew Wiley
Nicoleta Cristea
Marketa McGuire
Tim Brown
Stephanie Kampf
Jaeshin Kim
Darren LeMaster
Kim Braun
Carolyn Fitzgerald
John Koreny
Hamid Abbasi
Deanna Matzen
Kari Moshenberg
Micaela Ellison
Chapin Brackett
Chad Harris
Shannon Nobel
IV
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
1.0 BACKGROUND
1.1 Introduction
EPA Region 10 is developing mathematical models of water temperature as decision support
tools for watershed planning in the main Columbia and Snake rivers. This work has been
primarily for the purpose of developing a Total Maximum Daily Load for thermal energy as
required for water-quality limited river segments under Section 303 (d) of the Clean Water Act.
Products of this effort include a one-dimensional, time-dependent model of thermal energy for the
run-of-the river dams and free-flowing river segments (Yearsley et al, 2001); a two-dimensional
model of Lake Roosevelt (Yearsley, 2002) and a two-dimensional model of Lower Granite
reservoir (EPA 2002). Dworshak Dam and Reservoir on the North Fork of the Clearwater River
in Idaho (Figure 1) is upstream of the four run-of-the-river Snake River dams (Lower Granite,
Little Goose, Lower Monumental, Ice Harbor). Because Dworshak Dam can provide
downstream temperature control, it plays an important role in watershed planning for the
Clearwater, Snake and Columbia rivers. An effective mathematical model of water temperature
may be useful for optimizing use of these temperature control features. This report represents a
preliminary effort to assess the effectiveness of the two-dimensional mathematical model,
CEQUAL-W2 for this purpose.
Dworshak Dam was completed in 1971. The reservoir reached full pool and began operating in
1973. The project is operated for flood reduction, hydroelectric power generation, recreation,
water quality, and fish and wildlife uses. The US Army Corps of Engineers (USAGE) is one of
the Action Agencies1 that operates or markets power from the Federal Columbia River Power
System (FCRPS). Management of Dworshak Dam is the responsibility of the USAGE as part of
the coordinated regulation of the 10-dam FCRPS. This responsibility includes support of the
Basinwide Recovery Strategy for endangered salmonids and is described in the 1995 Biological
Opinion and the 2000 Biological Opinion. The Biological Opinions require that Dworshak Dam
provide flow augmentation for the Snake River for the benefit of migrating juvenile salmon and
steelhead from April through August. Since the project has temperature control structures that
provide the capability for controlling the temperature of water releases, Dworshak Dams is also
used to provide cold water during the summer for purposes of reducing water temperatures in the
Snake River.
The multiple uses that Dworshak Dam provides are in conflict at times, particularly in the
summer. High flows in July and August are desirable to downstream migrants, while high flows
and low temperature are best for adult fall Chinook in September. Though cold water releases
may be beneficial for migrating Snake River salmon and steelhead and for adult fall Chinook,
they have a negative impact on growth rates of fish at the hatchery just downstream from the
dam. Additional conflicts arise because recreation uses in the reservoir behind Dworshak Dam
rely on pool elevations sufficient to launch pleasure boats.
1.2 Report Objectives
Finding an operational strategy for Dworshak Dam that addresses the needs of flood control,
hydroelectric power generation, water quality and fish and wildlife uses in the Clearwater and
1 The Bonneville Power Administration (BPA), the US Army Corps of Engineers (USAGE) and the Bureau of
Reclamation (BOR) comprise the Action Agencies; the agencies that operate or market power from the Federal
Columbia River Power System.
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
Snake rivers is not, as one might expect, straightforward. A report developed in response to
Reasonable and Prudent Alternative (RPA) 143 of the 2000 Biological Opinion outlines a
program of data collection and mathematical model development to provide decision support pre-
season planning of an operational strategy for the project (RPA Workgroup, 2003). One-
dimensional mathematical models of water temperature have been used in this role with
considerable success for river segments downstream from Dworshak Dam since 1999 (EPA,
2000). However, that given the water depth and long residence time of water in the reservoir
behind the dam, a two-dimensional mathematical can provide valuable information regarding
water quality conditions within the Dworshak pool and future options for flow augmentation.
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
Prajact Study £Jt«
''Dwonhak Ftessvolr
NEVADA
NOT TO SCALE
UTAH
Rgum 1
Dwaneheh RB
LocaHor Map
Figure 1. Location Map.
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
A two-dimensional water quality model that has been widely applied to projects like Dworshak
Dam and Reservoir is CEQUAL-W2 (Cole and Buchak, 1995; Cole, 1997). Data input
requirements for CEQUAL-W2 are considerable. Recent data collection programs initiated by
the Action Agencies, in response to the requirements of RPA 143 of the 2000 Biological Opinion,
as well as other data collected in earlier programs provide sufficient input data for the
development of a preliminary model of water temperature in the reservoir. The objective of this
report is to evaluate the model's utility as a decision support tool for this project and to provide an
assessment of data needs for further modeling.
1.3 Project Characteristics1
Dworshak Dam is a multiple purpose concrete gravity structure completed in 1973 that controls
water from a 2440-mile drainage area on the North Fork Clearwater River, Idaho.
The reservoir provides 3.5 million acre-feet of total storage and 2.0 million acre-feet of flood
control storage (USAGE 1986). Dworshak Dam has a maximum height of 717 feet and a crest
length of 3,287 feet. The Dworshak Reservoir lies within a narrow, steep canyon and extends
53.6 miles upstream on the North Fork Clearwater River with a width range of 0.5 to 2 miles.
Dworshak Dam is equipped with multilevel gates that are adjustable for selective withdrawal
between full pool elevation of 1600 ft MSL (mean sea level) to the minimum pool elevation of
1445 ft MSL. Power facilities at the dam include two 90 megawatt generating units and one 220
megawatt generating unit, with room for future installations of three additional 220 megawatt
generating units (USAGE 1986).
The primary role of Dworshak Dam is to meet the overall flood regulation plan for the Columbia
River, as described in the Columbia River Treaty Flood Control Plan by the Corps. The flood
control plan involves three reservoir regulation periods: October through March (reservoir
evacuation period), May through July (reservoir refill period), and July through August (summer
recreation). Accordingly, rules for the evacuation of storage space at each reservoir have been
developed.
The overall storage plan calls for the following:
• September - December: Draw the reservoir down to elevation 1558 feet,
National Geodetic Vertical Datum (NGVD) by December 15 to provide 700,000
acre-feet (AF) of flood control storage. This is accomplished by using normal
power load discharges.
• January - March: Maintain 700,000 AF of storage space, plus any additional that
is required based on volume forecasts of available runoff.
• April - July: Refill at a rate that will provide safe flood control based on runoff
volume forecasts, yet will allow a 95 percent probability of refill.
• July - August: Attempt to hold full pool or maximum elevation achieved during
filing.
As noted previously, the July/August operations have been significantly modified in response to
salmonid recovery efforts in the Lower Snake River. Rather than holding at full pool during this
period, the reservoir is drawn down to provide higher flows and lower temperatures downstream.
2 The text and figures associated with Section 1.3 of the report have been excerpted from a USAGE memorandum
entitled "Selective Withdrawal Technology at Dworshak Dam, Idaho" dated August 28, 2002
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
1.3.1 Water Temperature Control
Selective withdrawal technology was incorporated into the design of Dworshak Dam so the
quality of the water released from the dam, especially water temperature, could be managed.
Water temperature management of release flows from the dam is most effective during the
summer. The National Marine Fisheries Service Forum Technical Management Team has used
water from Dworshak Dam to augment and cool Snake River summer flows since the early
1990s. Since 1998, the Dworshak pool has been drafted to elevation 1520 feet, NGVD by
August 31. Typically, target water release temperatures have varied from 48°F (8.9°C) to 51° F
(10.6°C) during July and August. The target water temperatures are selected as a balance between
fish production at the downstream Dworshak National Fish Hatchery and anadromous fishery
needs in the lower Snake River. Cool summer releases from Dworshak Dam typically contribute
from 25 to 45 percent of the Snake River flow.
1.3.2 Selective Withdrawal Structure
Water is released from the dam for water temperature control via three selector gate units and/or
three regulating outlets during normal operating conditions. Each of the multiple level selector
gate units is adjustable so water withdrawal can be made from full pool elevation (1600 feet,
NGVD) to the minimum pool elevation (1445 feet, NGVD). Water passing through the selector
gate units guide water to hydroelectric penstocks located at elevation 1395 feet, NGVD. One of
the selector gate units services a 250-WM generator (hydraulic capacity of 5,800 cubic feet per
second) while the other two selector gate units service two 100-MWgenerators (2,200 cfs each).
The 250-MW selector gate unit is actually composed of one master gate that is operated
simultaneously with two slave gates. The three gates are operated as one selector gate unit. The
two 100-MW selector gate units are each composed of a master and slave gate. The two 100-MW
selector gate units are operated as independent units. The hydraulic capacity of the 250-MW
generator varies from 3515 cfs to 2740 cfs for gross heads of 457 and 630 feet respectively. The
hydraulic capacity of the 100-MW generator varies from 2680 to 1880 cfs for gross heads
between 480 to 630 feet
(Figure 2).
Selector gate units are designed to pass water over the top of the gate units, called an overshot
mode, or under the bottom of the gate units, called an undershot mode. The reservoir elevation
range for operation in the overshot mode is from 1600 to 1500 feet, NGVD. The minimum
design submergence restriction to prevent cavitation is 30 feet, but for safety reasons, 35 feet is
used as the criterion.
The undershot mode operation can be operated in the reservoir pool range between 1600 feet,
NGVD and 1445 feet, NGVD. Between these elevations, there are at least 40-feet of head
provided to flow under the gate to prevent cavitation. The minimum elevation for the undershot
mode is elevation is 1435 feet, NGVD (Figure 3).
It is important to note that there is a zone in the reservoir that cannot be accessed via the selector
gates. The zone that cannot be reached is between elevations 1465 feet (lowest elevation of a
selector gates in the overshot position) and 1430 feet (lowest elevation of the selector gates in the
undershot position) (Figure 4).
There are three regulating outlets at Dworshak Dam available for evacuation of reservoir storage
below the spillway crest when release flows that exceed generation capacity are needed. Their
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
invert elevation is 1350 feet, NGVD. Discharge capacity of the three outlets varies from 23,100
cfs at the minimum pool elevation of 14445 feet to 39,750 cfs at the full pool elevation of 1600
feet, NGVD. Water releases through the regulating outlets can be made while selector gate
releases are
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
Spill-
way
E1.1545
3 Selector gate units:
2 in undershot &
overshot
Pen-
stock
El. 1393
Regulating
Outlets
El. 1350
0
o
o
Figure 2. Dworshak Dam face, looking from upstream to downstream
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
Limits of Selector Gate Positioning.
Limit of Lowest Elevation Overshot Position
water surface at elevation 1500 feet
Overshot limit 1465 feet
Limit of Lowest Elevation
Undershot Position
Undershot limit 1435
1 Penstock elevation 1395
|
Regulating Outlet elevation 1350 feet
Powerhouse
Figure 3. Dworshak Dam selective withdrawal structure schematic
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
Selector
Gate
Positions
30! elevation (a). 1520 feet in September of every year
Minimum
Dvershot
Position
Top of recondite zone @ el 1465 feet-
Water temperature 48° to 50°F
Water temperature 45° to 47°F
bottom of recondite zone (a), el 1435 feet-
Uinimum
Jndershot
osition
Figure 4. Dworshak Reservoir pool limitations in September
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
being made. The temperature of water released through the regulating outlet is typically between
42°F (5.6°C) and 43° F (6.1°C). (Figures 2 and 3).
During high runoff or flood conditions, a spill way (crest elevation 1545 feet) is also used to
evacuate water from the reservoir. Theoretically, the spill way could be used for water
temperature control. The spillway can be used anytime the pool is above elevation 1545 feet,
NGVD. However, the spillway is not used for regular temperature control operations because of
the high Total Dissolved Gas levels that it produces. Typically, the spillway is only used during
high runoff and flood events.
1.3.3 Water Temperature Monitoring
Water temperature is monitored at several locations at Dworshak Dam. The major water
temperature control point is at the Dworshak National Hatchery. Use of the selective withdrawal
flexibility at the project is governed by the resulting water temperatures achieved at the hatchery.
Power generation via the three turbines is accomplished using a combination of the releases via
the selective withdrawal structures (in overshot and/or undershot modes). When needed release
flows exceed the generation capacity of the three generators, the non-power producing regulating
outlets are also used.
Water temperature data is available from resistance thermal devices (RTDs) embedded in the face
of the dam during the time of construction. They are located at elevations 1574, 1549, 1524,
1499, 147^ 1449^ 1394, 1349, 1324, 1299, 1249, 1199, 1149, 1099, and 1049 feet, NGVD. In
addition, four floating sensors follow reservoir fluctuations at depths of 1, 5, and 10 and 20 feet.
One sensor is located on the downstream side of the dam at the nose pier of powerhouse bay
number 2.
1.3.4 Reservoir Thermal Structure
The thermal structure of the reservoir can be monitored throughout the summer season using the
RTDs. The reservoir displays typical thermal stratification patterns seen in reservoirs with a warm
epilimnion layer where the surface waters can exceed 68°F (20°C), a rapidly cooling middle zone
metalimnion layer, and a deep, cold hypolimnion layer. The thermocline, the limnologically
important depth in the metalimnion that exhibits the greatest water temperature change per depth,
typically occurs about forty to fifty feet from the surface and is usually about 55°F.
The project operators set the selector gates using the water temperature data from the RTDs to
obtain the desired target at the hatchery. Typically, the hatchery does not want water less than
48°F (8.9°C), and would prefer water that is about 50°F (10.0°C). Consequently, the selector
gates are continually changed throughout the summer to withdraw water from near to but below
the thermocline. By September, the zone of water below the thermocline that is between
45°(7.2°C) and 50°F falls into the zone that is unavailable for operational use. Consequently,
release waters in September are either about 42°F(5.6°C) or about 55°F(12.8°C), depending on
whether the selector gates are in the overshot or undershot position (Fig. 3 and 4).
10
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
1.4 Clearwater River Watershed Characteristics
The Clearwater River watershed has a total drainage area of approximately 9,570 square miles. It
is located in north central Idaho and is a major tributary of the Snake River, which is a major
branch of the Columbia River. Basin elevations range from 750 to 9,000 ft MSL.
The general course of the North Fork Clearwater is westerly from the Bitterroot Range in the
eastern and northeastern areas of the basin. The North Fork joins the Clearwater River
approximately 40 miles east of Lewiston and then drains a densely forested, sparsely populated,
undeveloped area of 2,440 square miles above the Dworshak Dam. Topography and runoff
characteristics naturally divide the Clearwater River basin into two major drainage areas, referred
to as the Upper and Lower Clearwater River watersheds (USAGE 1986). The lower watershed
consists primarily of steep barren hills and plateaus. The plateaus are generally areas used for
dry-farming methods and growing crops of wheat, peas, and lentils (USAGE 1986).
1.4.1 Geography and Soils
The headwaters of the North Fork Clearwater River begin in a mountainous area underlain by
metamorphic rocks of the Belt Series (MT) and igneous granite rocks of the Idaho Batholith.
There are various rock types that underlay the Clearwater River drainage area. Valley walls in
the lower river course are metamorphic rocks of the Orofino Series and lavas of the Columbia
Plateau occur at various locations in the western reaches of the basin. The lava flows are thought
to have changed the original drainage pattern of the Clearwater basin resulting in the current
course of the North Fork Clearwater River which cuts across old valleys and ridges.
Soils in the basin are composed of the major rock types, including decomposed granite and
sedimentary materials. The soil layer over the basin is considered to be thin and is underlain by
an impervious rock. This impervious rock layer contributes to the basin's high runoff (USAGE
1986).
1.4.2 Climate
Mild summers, cold winters, and abundant snowfall between the months of November through
April characterize the climate of the Clearwater River watershed. Snow accumulation generally
begins in late September or October and continues to increase until April or May. The
accumulation of snow over the Clearwater River basin influences the manner in which Dworshak
Reservoir is operated (USAGE 1986). Temperatures within the watershed fluctuate month-to-
month and year-to-year. The mean annual precipitation for the basin varies from 24 to 70 inches
depending on elevation. On the average 60 to 70 percent of the total annual precipitation falls
during the months of October to March, 20 to 30 percent during April through July and 5 to 15
percent during August and September.
1.4.3 Hydrology
A major part of the annual runoff for the Clearwater River basin originates in the upper
watershed, as a combination of winter rain and spring snowmelt. Heavy precipitation, dense
timber, sparse population, and very limited development characterize principal runoff areas
within the upper watershed. The stream flow pattern in the upper watershed heavily influences
the operation of Dworshak Reservoir. Stream flows are generally low from late July through
February, with increasing flows in March, and high flows during April through June. The
11
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
average annual runoff for the North Fork Clearwater above Dworshak Dam based on flow data
for the period 1974-2002 reported on the Columbia River Data Access in Real Time (DART)
Web site (http://www.cqs.washington.edu/dart') is 5540 cfs. The reservoir residence time, based
on the average annual flow and reservoir volume at full pool, is 316 days.
The lower watershed consists of approximately 1,530 square miles or about 16 percent of the total
basin. Natural March to July stream flow in the lower watershed represents about 5 percent or
less of the total March to July runoff from the entire Clearwater River Basin. Peak discharges
and runoff from the lower watershed are considered to have little or no impact on the operation of
Dworshak Reservoir (USAGE 1986).
1.5 Water Quality and Quantity Issues Associated with Dworshak Dam
Although water quality within the Clearwater River basin is generally high, water quality and
quantity problems have occurred in the Snake River downstream from the dam as a result of
activities within the Clearwater basin. Some of these issues are directly influenced by the
operation of the Dworshak Dam and Reservoir. The Dworshak Dam has also affected resident
and anadromous fish communities in the Clearwater River basin as well as the Snake and
Columbia Rivers. Water quality, quantity and fisheries issues associated with the Dworshak Dam
and Reservoir are discussed below.
1.5.1 Water Quality Issues
Post-dam conditions within the Dworshak Reservoir, the Clearwater River and in the Snake River
differ greatly from the water quality conditions associated with the free-flowing river. Water
temperature, dissolved oxygen concentrations, turbidity, and total dissolved gas concentrations in
the Clearwater River, both above and below the dam, have been affected by the impoundment.
Water temperatures in the Lower Clearwater River have been impacted by the thermal structure
of the reservoir during periods of stratification. During the summer-fall reservoir stratification
period, June-November, river temperatures in the Clearwater River below the dam vary from 7 to
16°C. Water temperatures in the Clearwater River are cooler in June-September and warmer
from October to November as a result of the discharges from Dworshak Dam (USAGE 1986).
Operations of the Dworshak Dam have also resulted in slower warming spring temperatures and
slower cooling fall temperatures.
Dissolved oxygen concentrations during summer months are lower in the reservoir than pre-dam
conditions. Dissolved oxygen levels during the summer months can fall to 35 to 45 percent of
saturation at the deepest parts of the reservoir.
Dissolved gas concentrations have also increased due to the construction of the Dworshak Dam.
During periods of spill from the dam, the total concentration of dissolved gas downstream of the
dam has been measured at levels above the State of Idaho's 110 percent standard. This standard
was designed for waters designated for salmonid spawning. Total dissolved gas measurements
have ranged from 97 percent to a maximum of 128 percent (USAGE 1986). Typically, Dworshak
is operated to avoid releasing water over the spillway in amounts that exceed the 110 percent
standard.
1.5.2 Water Quantity Issues
The construction of the Dworshak Dam significantly altered the flow regime of the North Fork
Clearwater River, Clearwater River and the Snake River. The reservoir stores large volumes of
12
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
water that would have flowed freely to the Snake and Columbia Rivers. The Dworshak Reservoir
has reduced flood events and peak flows on the Snake and Columbia Rivers. The Dworshak Dam
operations now focus on flow augmentation to provide water for the Dworshak Steelhead
Migration Hatchery located below the dam, and to augment summer flows to cool the Snake
River for salmonids.
1.5.3 Fisheries Issues
Because the Dworshak Dam does not have a fish passage facility, the entire anadromous wild
steelhead and chinook fishery was eliminated from the Clearwater River with the exception of the
lower 1.9 miles of the river. Kokanee salmon, smallmouth bass, bull trout and west slope
cutthroat are present within the system above the dam and are influenced by reservoir operations.
While bull trout are listed as an endangered species and affected by dam operations, their
subpopulations are stable and secure. Above the dam, reservoir fluctuations due to flood control
and power peaking have interfered with riparian vegetation and reservoir limnology, causing
variable primary productivity and loss of resident fish stocks within the reservoir (Idaho
Department of Water Resources, 2000)
13
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
2.0
DESCRIPTION OF MATHEMATICAL MODEL
2.1 CEQUAL-W2 Version 3.1
The version 3.1 of the mathematical water quality model CE-QUAL-W2 (Cole and Buchak,
1995) was selected for this project to simulate water temperatures in the Dworshak Reservoir.
CE-QUAL-W2 is a two-dimensional, longitudinal/vertical, hydrodynamic and water quality
model (Figure 5). Because the model assumes lateral homogeneity, it is best suited for relatively
long and narrow water bodies exhibiting longitudinal and vertical water quality gradients (Cole
and Buchak 1995).
The water quality algorithms of CE-QUAL-W2 include modeling of 21 constituents in addition to
temperature including nutrient, phytoplankton, and dissolved oxygen (DO) interactions during
anoxic conditions (Cole 1997). Each of which is modeled based on laterally averaged equations
of momentum, continuity, and transport. Complete details on model theory and structure, and an
extensive bibliography for theoretical development and application are given in Cole and Buchak
(1995).
X
f
w
u
Q
out
Figure 5. Two-dimensional (x-z) hydrodynamics model schematic. Source: Scott A. Wells
Department of Civil Engineering Portland State University Portland, Oregon USA.
http://cv-nt.technion.ac.il/courses/19225/Lecturel_intro_files/frame.htm
The time-varying solution technique of the model is based on an implicit, finite difference
scheme that results from the simultaneous solution of the horizontal momentum equation and the
free-water surface equation of vertically integrated continuity (Bales and Giorgino 1998). The
computational time step is variable throughout the simulation to ensure numerical stability.
14
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
Application of CEQUAL-W2 to the prototype requires the following input data:
Bathymetery
Meteorology
Tributary Inflows and Temperatures
Outlet Flows
The most complete set of data for purposes of this assessment are from Calendar Year 2002. The
data used to implement CEQUAL-W2 for Dworshak Dam and Reservoir are described below.
2.2 Bathymetry
Bathymetry data for the project was obtained from the limnological study of Dworshak Reservoir
funded by the USAGE (Falter et al, 1975). Table 4 in Falter et al (1975) describes the reservoir
volume as a function of surface elevation and various figures in the document provide an estimate
of the bottom slope. Reservoir widths at full pool were estimated from topographical maps at
intervals of approximately one kilometer. Reservoir widths estimated in this way as a function of
distance are shown in Figure 6. Accuracy and computational efficiency is much higher in
applications of CEQUAL-W2 when the geometry varies smoothly. In light of this, a fourth-order
curve was fit to the measured widths as shown in Figure 6. This curve, modified slightly as
described below, was used to estimate reservoir widths at full pool for input to CEQUAL-W2.
The finite difference grid characterizing the reservoir comprised 26 longitudinal segments, each
of which was 3200 meters long for a total modeled reservoir length of 83.2 kilometers. Each
segment was further subdivided into vertical layers 3.0 meters in thickness. The deepest segment,
the segment adjacent to the dam, had a total of 66 active vertical segments.
Since actual reservoir soundings were not available for this study, reservoir widths as a function
of depth were estimated based on the assumption that the glacially-carved valley of the North
Fork has the shape of an exponential curve. It was assumed that the exponential curve had the
same shape along the entire length of the modeled portion of the reservoir. In addition, a
multiplier was incorporated into the width estimate to account for the embayments associated
with Elk Creek and the Little North Fork. For input to CEQUAL-W2, top widths (labeled
"Smoothed" in Figure 6) were estimated at 3200 meter intervals using the polynomial fit. The
multiplier and power to which the exponent was raised were adjusted so as to provide a good fit
to the observed reservoir volume as reported in the limnologic study by Falter et al (1975). A
typical cross section based on the exponential assumption is shown in Figure 7.
Figure 8 compares the observed and modeled volumes between full pool (1600 feet/487.69
meters NGVD) and the elevation to which the reservoir has been drafted for fish passage since
1998 (1520 feet/463.30 meters NGVD).
2.3 Meteorology
Hourly observations of air temperature, dew point, and wind speed and direction are available at
the Dent Acres site adjacent to the reservoir (Figure 9). The Dent Acres site is part of the US
Bureau of Reclamation's (USER) AgriMet system. Observations are available at this site
beginning in
15
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
20 40 60 80
Distance from Dam - kilometers
100
Figure 6. Measured and modeled reservoir widths of Dworshak Reservoir
16
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
-600 -400 -200 0 200 400 600
Reservoir Width - meters
Figure 7. Typical reservoir cross-section based on an exponential shape
17
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
2500
2000
1500
1000
500
5 10 15 20
Pool Elevation (meters below full pool)
Measured
Model
25
Figure 8. Modeled and observed reservoir volumes for Dworshak Reservoir
18
-------
Preliminary Analysis of the Thermal Regime ofDworshakDam and Reservoir September 2003
April 2002 and data are available at the AgriMet Web site (http://mac 1 .pn.usbr.gov/agrimet).
Cloud cover is not observed at this site and it was, therefore, necessary to assume that the average
hourly cloud cover for each day of the year based on the 54-year period of record at Lewiston,
Idaho was adequate.
2.4 Tributary Inflow and Temperature
Upstream boundary conditions included measured daily stream flow and water temperature from
the USGS gauging station 13340600, located at the inflow to Dworshak Reservoir on the North
Fork of the Clearwater River (see Figure 9). The watershed represented by this gage has an area
of 1440 square miles, approximately 60% of the total watershed above Dworshak Dam. The
remaining 40% is not gauged. It was therefore assumed that the ungauged flow could be
estimated by performing a water balance of the reservoir. Outflow at the dam and reservoir
elevation are available on the DART site. The daily-average flows from the ungauged portions,
Qungaged, of the watershed were estimated from:
Qungaged = AV/At - Qout ~ Ql3340600 (1)
Where,
AV/At = the daily change in reservoir volume in time, At,
Qout = the total daily outflow from the reservoir,
Qi33406oo= the gauged daily flow at USGS gage 13340600.
Application of Eq. (1) resulted in negative flows on a few days. The estimated flow on days for
which this occurred was calculated as the average of the preceding day and the following day.
The observed reservoir elevation and the elevation simulated by the water budget are shown in
Figure 10.
Input temperatures for the ungauged portion of the watershed were assumed to be the same as
those observed at USGS gage 13340600. Tributary inflow and temperature for the calendar year
2002 are shown in Figures 11 and 12.
2.5 Outlet Flows
Hourly observations of spill and total outflow from Dworshak Dam are available from the DART
Web site. These data are not specific as to the configuration and withdrawal schedule associated
with the temperature control structure. These data are apparently recorded by the USAGE (C.
Knaak, personal communication), but were not available at the time of model development.
Since this study is being treated as a preliminary analysis of the thermal structure of Dworshak
Reservoir, the outlet flow data were reverse engineered to provide a reasonable comparison with
observed outlet temperatures. The withdrawal schedule for outflows as a function of time and
location is shown in Figure 13. It was assumed that there were only three outflow elevations
(spillway elevation and two selector gate elevations). Elevations of selector outlets were chosen
such that they were within the range of the selective withdrawal structure or penstocks as
described above in Section 1.3. Outflows from each selector outlet were calculated such that the
total flow-weighted average temperature (from all three outflow locations) was approximately
equal to the observed downstream temperature.
19
-------
Preliminary Analysis of the Thermal Regime ofDvtorshakDam and Reservoir September 2003
September 2003
060308
Little Nortti Fc>rt< Gfearwate*' Rivet
"'3340T60
133-4
Worth Fork ijesrwalef Riv«
u
1,1 Ml,.,
9000 0 8000 16000
Idaho: Dworshak Reservoir
Figure 9. Water temperature and stream flow stations in the Clearwater River basin
20
-------
Preliminary Analysis of the Thermal Regime ofDvtorshakDam and Reservoir September 2003
September 2003
• Observed
• Modeled
455
0 50 100 150 200 250 300 350 400
Time - days
Figure 10. Measured and modeled pool elevation in Dworshak Reservoirs in 2002.
21
-------
Preliminary Analysis of the Thermal Regime ofDvtorshakDam and Reservoir September 2003
September 2003
1200 T
o
o
o
S
o
o
?5
o
o
55
o
o o
5 S
o o
o
So
o
o> o
O 1-
Figure 11. Tributary inflows to Dworshak Reservoir during the calendar year 2002 based
on data from USGS gage station 13340600.
22
-------
Preliminary Analysis of the Thermal Regime ofDvtorshakDam and Reservoir September 2003
September 2003
O
O
O
Si
O
O
?5
O
O
55
O
O O
<3 N:
o o
o
£3
o
o o
35 5
O 1-
o
Si
Figure 12. Temperature of tributary inflows during the calendar year 2002 to Dworshak
Reservoir based on data from USGS gage station 13340600.
23
-------
Preliminary Analysis of the Thermal Regime ofDvtorshakDam and Reservoir September 2003
September 2003
350
CM CM CM CMCMCMCMCMCMCMCMCM
CM CM CM CMCMCMCMCMCMCMCMCM
Spill
Elevation 460
Elevation 431
T- CM
IO tO
OO CO
T- CM
Figure 13. Inferred outflow schedule at Dworshak Dam during 2002
24
-------
Impacts of the Dworshak Dam/Reservoir on the Clearwater and Snake Rivers September 2003
3.0 MODEL RESULTS
3.1 Outlet Temperatures
Observed and simulated outflow temperatures are shown in Figure 14. The simulation results
demonstrate that it is possible to configure the modeled outflow in a simplified manner so as to
match the observed fairly well as demonstrated in Figure 14. However, this is quite different
from saying that the model correctly simulates outlet temperatures, since the outflow
configuration was reverse-engineered to do so. A better assessment of model performance can be
made at such time as operations data are available.
16
O
cs cs cs
O O O
O O O
es es es
cs cs cs
o o o
o o o
es es es
cs
o
o
es
cs
o
o
es
cs
o
o
es
es es es
o o o
o o o
es es es
Si ?5
CO
o
^ S3
Figure 14. Simulated and observed outlet temperatures at Dworshak Dam during 2002.
3.2 Vertical Temperature Profiles
On June 9, 2002, the USAGE began observing water temperature profiles near Dworshak Dam as
part of RPA 143. These observations have been used to evaluate temperature simulations
obtained with CEQUAL-W2. Simulated and observed temperature profiles for June 30, 2002;
July 20, 2002; August 8, 2002; and August 28, 2002 are shown in Figures 15-18. In general, the
25
-------
Impacts of the Dworshak Dam/Reservoir on the Clearwater and Snake Rivers September 2003
simulated and observed are similar. Again, more detailed information regarding operations will
be needed to assess model performance.
3.3 Longitudinal Temperature Profiles
Isopleths of width-averaged simulated water temperature on June 30, 2002; July 20, 2002; August
8, 2002; and August 28, 2002 are shown in Figures 19-22, respectively. There are no
observational data, other than at the single site near the dam, to compare with the simulated
longitudinal profiles. Based on the simulated results, one would conclude that dominant
mechanism for distribution of thermal energy during the summer low flow period is vertical
mixing rather than longitudinal advection. This is consistent with observations of reservoir
temperature by Falter et al (1975) in 1972-1974 and also with the reservoir residence time of
316 days.
26
-------
Impacts of the Dworshak Dam/Reservoir on the Clearwater and Snake Rivers September 2003
500
490
480
Q 470
£
"3
460
450
c
.2 440
1
EJ 430
420
410
400
<> m
•1
*.
10 15
Temperature - deg C
20
• Simulated
HourO
HourS
Hour 11
:Hour 17
25
Figure 15. Simulated and observed water temperatures in Dworshak Reservoir at
Dworshak Dam on June 30, 2002
27
-------
Impacts of the Dworshak Dam/Reservoir on the Clearwater and Snake Rivers September 2003
Q
O
0)
•55
c
o
1
^)
LLI
490
480
470
460
450
440
430
420
410
400
10 15 20
Temperature - deg C
25
• Simulated
HourO
HourS
Hour 11
:Hour 17
30
Figure 16. Simulated and observed water temperatures in Dworshak Reservoir at
Dworshak Dam on July 20, 2002
28
-------
Impacts of the Dworshak Dam/Reservoir on the Clearwater and Snake Rivers September 2003
490 -i
480
470
Q
§ 460
2
o> 450
"5
c 440
o
o) 430
LLI
420
410
400
(
i
*
X*
»**
K
K
rx
s
rj"1
(• Simulated
• Hour 0
HourS
Hour 1 1
X Hour 17
) 5 10 15 20 25
Temperature - deg C
Figure 17. Simulated and observed water temperatures in Dworshak Reservoir at
Dworshak Dam on August 8, 2002.
29
-------
Impacts of the Dworshak Dam/Reservoir on the Clearwater and Snake Rivers September 2003
470
460
Q
O
0)
0>
c 440
c
o
1 43°
0)
LLI
(
<
•
<
K «
f*
m
m
4
^
4
•
•7
(• Simulated
• Hour 0
HourS
Hour 1 1
X Hour 17
) 5 10 15 20 25 30
Temperature - deg C
Figure 18. Simulated and observed water temperatures in Dworshak Reservoir at
Dworshak Dam on August 28, 2003
30
-------
Preliminary Analysis of the Thermal Regime ofDvtorshakDam and Reservoir September 2003
48O
Distance from Dam - Km
3OO
ao
24
22
20
18
16
14
12
10
8
"
r
2
Jo
Figure 19. Simulated longitudinal temperature profiles in Dworshak Reservoir on
June 30, 2002.
31
-------
Preliminary Analysis of the Thermal Regime ofDvtorshakDam and Reservoir September 2003
48O
46O
Distance from Da
3OO
24
22
20
18
16
14
12
10
8
6
4
2
0
ao
Figure 20. Simulated longitudinal temperature profiles in Dworshak Reservoir on July 20,
2002.
32
-------
Preliminary Analysis of the Thermal Regime ofDvtorshakDam and Reservoir September 2003
46O
3OO
24
22
20
18
16
14
12
10
8
6
4
2
0
2O 4O
Distance from
6O
- km
ao
Figure 21. Simulated longitudinal temperature profiles in Dworshak Reservoir on August
8, 2002.
33
-------
Preliminary Analysis of the Thermal Regime ofDvtorshakDam and Reservoir September 2003
*±=
CO
46O
44 O
42O
4OO
38O
360
34O
32O
3OO
24
22
20
18
16
14
12
10
8
6
4
2
2.O 4O
Distance from
6O
im - km
8O
Figure 22. Simulated longitudinal temperature profiles in Dworshak Reservoir on
August 28, 2002.
34
-------
Preliminary Analysis of the Thermal Regime of Dworshak Dam and Reservoir September 2003
4.0 CONCLUSION and RECOMMENDATIONS
4.1 Conclusion
Based on these preliminary results, CEQUAL-W2 appears to be well-suited for simulating water
temperatures in Dworshak Reservoir. Further testing with more detailed operations data will be
required to confirm this, however.
4.2 Recommendations
Although these results are promising, additional work is necessary before accepting of CEQUAL-
W2 as a decision support tool for managing water temperature releases from Dworshak Dam and
Reservoir. Most of the this work is associated with expanding the monitoring program and
includes:
• Recording and reporting water releases from the penstocks, regulating tubes and
spillway and the daily configuration of the temperature control structures
• Observing cloud cover at the Dent Acres AgriMet weather station
• Improved monitoring of flow and temperature in the ungauged portion of the North
Fork of the Clearwater River
In addition, some modifications to the CEQUAL-W2 software may be of value. For example, it
does not appear to be the case that CEQUAL-W2 version 3.1 has the capability to simulate the
operation of Dworshak Dam's temperature control structures. Although this study suggests that
approximating the temperature control structures with three to five fixed outlets can lead to
reasonable result, it may be worthwhile to consider adding algorithms that accommodate the
movable shutters of the temperature control structure.
35
-------
Preliminary Analysis of the Thermal Regime of Dworshak Dam and Reservoir September 2003
6.0 REFERENCES
Cole, T.M. and E.M. Buchak. 1995. CE-QUAL-W2: A Two-Dimensional, Laterally Averaged,
Hydrodynamic and Water-quality Model. Version 2.0. User's Manual. Vicksburg,
Mississippi. Instruction Report EL-95-1. US Army Engineer Waterways Experiment
Station. 57 pp+app.
Cole, T.M., 1997. CE-QUAL-W2: A hydrodynamic and water quality model for rivers,
estuaries, lakes, and reservoirs, http://www.wes.army.mil/el/elmodels/w2info.html
EPA Region 10. 2000. A Retrospective Analysis of Water Temperature Management In the
Lower Snake River Using Coldwater Release from Dworshak Dam Summer 2000
http://vosemite.epa.gov/rlO/water.nsf/59f3b8c4fc8c923988256b580060f5d9/Ob791dl5aa
01034988256a7300605adf?QpenDocument
EPA, 2002, Temperature Simulation of the Snake River above Lower Granite Dam
Using Transect Measurements and the CE-QUAL-W2 Model: U.S. Environmental
Protection Agency, Region X, Seattle, WA. EPA 910-R-02-008
Falter, C.M., J.M. Leonard and J.M. Skille. 1975. Early limnology of Dworshak Reservoir,
Part 1 - LIMNOLOGY. Final Report Submitted to US Army Corps of Engineers,
Walla Walla, Washington for Contract No. DACW68-72-C-0142.
Idaho Department of Water Resources. 2000. Dworshak operation plan. Prepared for
Idaho Water Resources Board. Adopted December 21, 2000.
National Marine Fisheries Service (NMFS). 2000. Biological Opinion: Operations of the
Federal Columbia River Power System.
RPA Workgroup, 2003. Water Temperature Modeling and Data Collection Plan for Lower
Snake River Basin. Bi-Op Measure 143. Final Draft Report.
United States Army Corps of Engineers (USAGE). 1986. Water Control Manual for Dworshak
Dam and Reservoir, North Fork Clearwater River, Idaho. Walla Walla District.
United States Fish and Wildlife Service (USFWS). 2000. Biological Opinion: Effects to Listed
Species from Operations of the Federal Columbia River Power System.
Yearsley, J., D. Kama, S. Peene and B. Watson. 2001. Application of a 1-D heat budget model
to the Columbia River system. Final report 901-R-01-001 by the U.S. Environmental
Protection Agency, Region 10, Seattle, Washington
Yearsley,!. 2002. Columbia River temperature assessment: Simulation of the thermal regime of
Lake Roosevelt. EPA910-R-03-003. EPA Region 10, Seattle, Washington.
36
------- |