CRiSP1.6 Theory & Calibration Manual: I.1 - General Description

I.1 - General Description

CRiSP.1 models passage and survival of multiple salmon substocks through the Snake and Columbia rivers, their tributaries, and the Columbia River Estuary (Fig. 1). The model recognizes and accounts for several aspects of the life-cycle of migratory fish--fish survival, migration, and passage--and their interaction with the river system in which they live.

Fish survival through reservoirs depends on:

• predator density and activity
• total dissolved gas (TDG) supersaturation levels dependent on spill
• travel time through a reservoir.

Fish migration rate depends on:

• fish behavior and age
• water velocity which in turn depends on flow, cross-sectional area of a reach, and reservoir elevation.

Fish passage through dams (Fig. 2) depends on:

• water spilled over the lip of the dam
• turbine operations
• bypass screens at turbine entrances and fish guidance sluiceways
• fish delay at dams.

CRiSP.1 computes daily fish passage on a release-specific basis for all river segments and dams. In CRiSP.1, passage and survival of fish through a reservoir is expressed in terms of the fish travel time through the reservoir, the predation rate in the reservoir, and a mortality rate resulting from fish exposure to total dissolved gas supersaturation, an effect called gas bubble disease (GBD). Fish enter the forebay of a dam from the reservoir and may experience predation during delays due to diel and flow related processes. They leave the forebay and pass the dam mainly at night through spill, bypass or turbine routes, or the fish are diverted to barges or trucks for transportation. Once they leave the forebay, each route has an associated mortality rate and fish returning to the river are exposed to predators in the dam tailrace before they enter the next reservoir.

Fig. 1 CRiSP.1 map of an abbreviated Columbia Basin river system which includes about thirty fish release points and the major dams

.

Fig. 2 Dam showing fish passage routes. Fish collected in bypass systems are returned to the tailrace or, in some situations, transported downstream.

I.1.1 - CRiSP.1 Submodels

CRiSP.1 integrates a number of submodels that describe interactions of isolated components. Together they represent the complete model. These elements include submodels for: fish travel time, reservoir mortality, dam passage, total dissolved gas supersaturation, and flow/velocity relationship. The structure of CRiSP.1 allows the user to select different formulations of these submodels at run time. In this sense, CRiSP.1 can be configured to simple interactions or it can be set up to consider several ecological interactions. CRiSP.1, as it is presently calibrated, has an intermediate level of complexity: age dependent travel time is implemented (the temporal components of the active migration equation cause the migration rate to increase with time of year), but other age dependent factors are switched off. A brief description of the submodels follows.

Travel Time

The smolt migration submodel, which moves and spreads releases of fish down river, incorporates flow, river geometry, fish age and date of release. The arrival of fish at a given point in the river is expressed through a probability distribution. All travel time factors can be applied or they can be switched off individually, resulting in a simplified migration model.

The underlying fish migration theory was developed from ecological principles. Each fish stock travels at an intrinsic velocity as well as a particular velocity relative to the water velocity. The velocities can be set to vary with fish age. In addition, within a single release, fish spread as they move down the river.

The travel time parameters are calibrated for spring and fall chinook and steelhead from the Snake River Basin and the Upper Columbia River Basin. See also the Juvenile Salmon Travel Time web page at www.cbr.washington.edu/crisp/tt/.

Predation Rate

The predation rate submodel distinguishes mortality in the reservoir, the forebay, and the tailrace of dams. The rate of predation can depend on temperature, smolt age, predator density, and reservoir elevation.

The predation rate parameters are calibrated using laboratory studies of the response of predators to temperature and field studies of smolt migration survival. The model is calibrated for spring and fall chinook and steelhead from the Snake River Basin and the Upper Columbia River Basin.

Gas Bubble Disease

A separate component of the mortality submodel is mortality from gas bubble disease produced by total dissolved gas (TDG) supersaturation. The mortality rate is species specific, and it is adjusted to reflect the relationship of fish length and population depth distribution to TDG supersaturation experienced by the fish.

The gas bubble disease rate is calibrated from laboratory studies.

Dam Passage

Timing of fish passage at dams is developed in terms of a species dependent distribution factor and the distribution of fish in the forebay. Fish guidance efficiency (FGE) can be held constant over a season or it can vary with fish age and reservoir level.

Fish guidance efficiency parameters are calibrated from fish guidance efficiency studies.

Transportation Passage

Transportation of fish at collection dams is in accordance with the methods implemented by the U.S. Army Corps of Engineers. The start and termination of transportation and separation of fish according to species can be determined for any dam under the same rules used to manage the transportation program. Time in transportation and transportation mortality can also be set.

Transportation operations information was used to identify the individual transportation operations from 1975 through 1999.

Total Dissolved Gas Supersaturation

Total dissolved gas (TDG) supersaturation, resulting from spill at dams, can be described by empirical submodels which are an empirical fit of spill data and monitoring data collected by the U.S. Army Corps of Engineers. Alternatively, supersaturation can be described by mechanistic models which include information on geometry of the spill bay and physics of gas entrainment.

The TDG generation equations used for gas production include the newest developments by U.S. Army Corps of Engineers, Waterways Experiment Station (WES) as well as additional work done by Columbia Basin Research. The gas calibration has been verified for 13 dams for the years 1995 through 1999.

Flow

Flow is modeled in two ways: it can be specified at dams using results of system hydroregulation models and historical flows or it can be described in terms of daily flows at system headwaters. When flow is described in headwater streams, the flow submodel generates a random set of seasonal flows that have statistical properties in accordance with the available water over a year. In this fashion, the model statistically reproduces flow for wet, average and dry years. The user controls the mainstem river flows by adjusting the outflow of the storage reservoirs within their volume constraints.

In the historical data files daily flow information, including temperatures and dam operations, are specified for the years 1954 through 1999.

Water Velocity

Water velocity is used in CRiSP.1 as one of the elements defining fish migration. Velocity is determined from flow, reservoir geometry and reservoir elevation.

Reservoir Drawdown

Reservoir elevation is set on a daily basis from elevation information in system hydroregulation models files or from user specified files. As water levels drop, part of the reservoir may become a free-flowing stream.

Stochastic Processes

CRiSP.1 can be run in a Monte Carlo Mode in which flows and model parameters vary within prescribed limits. In this mode, survival to any point in the river can be determined as a probability distribution.

Geographical Extent

CRiSP.1 can describe a river to any desired level of detail by changing a single file--the river description file--containing the latitudes and longitudes of all possible release sites, dams, and river segments as well as many of the physical attributes of these features. All menus and input and output tools automatically configure from the information in this file. In the current distribution, three river description files are available. The default columbia.desc file contains an abbreviated description of the Columbia Basin river system with about thirty fish release points and major dams. Some rivers in the basin are not represented (e.g., Imnaha River or Grande Ronde River).

Additional river description files are available in the CRiSP.1 distribution. They have been modified to reflect changes in the river that could occur under certain proposed management actions. The file columbia_snakedraw.desc does not include any of the Snake River dams-- Lower Granite, Little Goose, Lower Monumental, and Ice Harbor--that are in the default columbia.desc file. This simulates a Snake River drawdown in order to make the Snake River free-flowing. The file columbia_drawdown.desc is similar to columbia_snakedraw.desc. In addition to the removal of the four Snake River dams, the John Day dam has also been removed. This represents the most extensive drawdown scenario that has been considered.

CRiSP1.6 Theory & Calibration Manual: I.1 - General Description