[Manual] [Contents] [Prev] [Next]

II.8.3 - Spill

The spill algorithm represents allocations of spill from flow models (HYDROSIM or HYSSR) through Flow Archive Files or the Spill Schedule Tool window.

Flow Archive Spill

When spill is allocated from Flow Archive files, it is identified as a percent of daily averaged flow over multi-day periods. Consequently, for use in CRiSP.1, archive derived spill must be allocated to specific days and hours of the day. Special adjustments to spill allocations in years of low and high water are not implemented at this time. CRiSP.1 considers three types of spill:

Planned Fish Spill is requested by the fisheries agencies. The schedule for this can be obtained from the Flow Archive Files or can be set in the Spill Schedule Window.

Overgeneration Spill occurs when electrical generation demand is less than that available in flow. This is obtained from the Flow Archive File only.

Forced Spill occurs when river flow exceeds powerhouse capacity. This is calculated by CRiSP.1.
CRiSP.1 allocates spill flows in the following order.

First, Planned Fish Spill is allocated. For each period, planned spill is distributed over scheduled spill days and fish spill hours (within those days) using the following steps.

  1. Total modulated flow in the period that occurs in fish spill hours on planned spill days is calculated and designated
    flow_available (in kcfs units).
  2. The requested spill in a period is designated
    spill_request (in kcfs units).
  3. Percent spill during Fish Hours is calculated as
    spill_daily_percent = spill_request/flow_available.
  4. If spill_daily_percent > 100%
    then spill_daily_percent = 100%
    of the flow available in the request periods and the rest is discarded and a warning message is generated.
Second, Overgeneration Spill identified in the flow models for 2 or 4 week periods is evenly distributed over all days in the periods. The following calculations are made on a daily basis.
  1. Overgeneration Spill is added to Planned Fish Spill in Fish Hours every day in a period to yield total spill.
  2. If Total Spill in Fish Hours is now greater than the total flow over the hours then the excess is distributed over the rest of the day.
  3. If Total Spill for the entire day is greater than the total daily modulated flow then the spill is set to the total daily modulated flow.
Third, Forced Spill occurs when river flow exceeds powerhouse capacity. Forced Spill is calculated on the dam time slice periods. This is typically a 2 hour interval. CRiSP.1 uses the following logic:
  1. Calculate the quantity
    flow - powerhouse capacity/flow = possible forced spill
  2. Then, if
    possible forced spill > total fish & overgeneration spill
    assign
    total spill = possible forced spill.
    Otherwise the forced spill is assimilated into fish and overgeneration spills.
Spill from Spill Schedule Tool

Planned Spill can be set by specifying spill information with the Spill Schedule Tool. The following information is entered:

Overgeneration Spill is only applied if a Monte Carlo Mode is used. Forced Spill is calculated as described above and is applied in both Scenario and Monte Carlo Modes.

Spill Caps

The maximum allowable planned spill is set by spill caps at each dam. If planned spill exceeds the cap then spill is limited to spill cap. Forced spill can exceed the spill cap. Spill cap is under the DAM button.

Spill Efficiency

The fraction of fish passed with spilled water is defined by one of seven possible empirical equations that can be selected by the user.

Table 40 Equations used in spill efficiency
Equation number index Equation form
eq (146) a
b
c
d
e
f
g

where

The equations and parameters defining spill efficiency (often called "effectiveness" in the literature) are indicated in Table 41. These values were used beginning with the SOR screening runs of CRiSP.1.

Table 41 Spill efficiency (% fish passed in spillway /% flow passed in spillway).
Dam Spill equation Reference
Wells zeroa Erho et al. 1988, Kudera et al. 1991
Rocky Reach % pass = 0.65 * (% spill) Raemhild et al. 1984
Rock Island % pass = 0.94 * (% spill) + 11.3 Ransom et al. 1988
Wanapum % pass = 15.42 * ln (% spill) Dawson et al. 1983
Priest Rapids % pass = (% spill) ^ 0.82
L. Monumental % pass = 1.2 * (% spill) Johnson et al. 1985
Ransom/Sullivan 1989
The Dalles % pass = 2 * (% spill)
all other dams % pass = (% spill) -
aWells Dam is designed to pass smolts preferentially through the spillway system: about 96% of all smolts pass via the spillway. This is modeled by assigning an FGE value of 96% (range 95-97%) at Wells with a zero spill efficiency for years 1991 on.

[Manual] [Contents] [Prev] [Next]
Columbia River Salmon Passage Model CRiSP.1.5 Theory, Calibration & Validation Manual
Copyright © 1996, Columbia Basin Research. All rights reserved.

web@cbr.washington.edu