This paper describes a model developed to estimate potential fish production in the Yakima River Basin. During the past 30 years, planners have addressed the costs and benefits of enhancing fish habitat in the Yakima River Basin. Recently, the Fish and Wildlife Service conducted a flow study on the river system using the Incremental Flow Incremental Methodology (IFIM). Concurrently, the Bureau of Reclamation developed a computer program that simulates flow in each of 31 river reaches for a 52-year period (assuming a repeat of historical flows) for the Yakima Irrigation Project. The challenge facing project planners was to combine the IFIM output (Weighted Usable Area (WUA) vs. flow) with the operations program to estimate potential fish production for various operational scenarios.

Many investigators have reported fish densities in various macrohabitat types such as pools or riffles, although few have assessed fish densities at the microhabitat level. Everest and Chapman (1972) evaluated chinook salmon densities in various microhabitats. They reported fish density in preferred habitat and the mean density in utilized habitat as 6.5 age-0 chinook/m² (0.6/ft²) and 1.8 age-0 chinook/m² (0.17/ft²), respectively. In IFIM terminology, these data suggest that in preferred habitat where the Joint Preference Factor (JPF) is 1.0 (i.e., ideal depth, velocity, cover, etc.) a square foot of habitat supports an average of 0.6 fish. I assumed a linear relation between JPF and fish density, and concluded that the mean JPF for utilized habitat was 1.8/6.5 = 0.28. A more descriptive way of defining WUA is that a unit of WUA is a unit of optimum equivalent habitat or, in the above case, the area required to support 0.6 fish. It takes 1 square foot of preferred habitat to support 0.6 fish and 3.5 square feet (i.e., 6.5/1.8) of marginal habitat with a JPF of 0.28 to support 0.6 fish. However, in each case a single unit of WUA is being described. Thus, for the calculations made in establishing parameter values for our computer model, we used 0.6 fish per unit of WUA as the conversion factor.

The above-derived ratio of fish per unit of WUA is for age-0 juveniles during July and August. Approximately 30% of these August age-0 juveniles survive to the smolt stage (McIntyre 1983, Lindsay et al. 1986). Fast at al. (1985) estimated the egg-to-migrant survival for this species in the Yakima River as 5.6%, and Lindsay at al. (1986) reported survival for the same period as 6.3%. Information verified through spawning surveys in the Yakima Basin indicates that 2.9 spring chinook salmon escape into the upper watershed for every redd constructed. However, with the ongoing improvements to fish passage facilities it is expected that the spawners per redd ratio will likely decrease to 2.5. The fecundities of the Naches stock and the upper Yakima stock were evaluated by Major and Mighell (1969) and reevaluated by Wassernum1, who reported that the weighted mean fecundity for spring chinook in the Yakima Basin is 4,870 eggs per female.

We then developed habitat ratios (Bovee 1982) to determine if rearing or spawning habitat limited production. Using the above information, the rearing WUA for spring chinook was converted to spawner equivalents as follows: 1 Unit WUA_rear = 0.6 fry x 0.30 sW9W x 100 score x 1 r fry5.6 smolts 4,870 eggs I Unit WUA. = 0.00066 equivalent redds or I Unit WUA = 0.00066 redds x 2.5 spawners redd = 0.00165 equivalent spawners (1)

We also calculated potential production from the WUA_spawn by considering that the average size of a spring chinook redd, including disturbed area and protected territory is 176 square feet (Burmer 1951). Thus: 176 Units WUA,.. = 1 redd 1 Unit WUA, = 0 00568 eguivalent redds 2L 1 Unit WUA,. = 0.00568 redds x 2.5 spawners redd = 0.0142 equivalent spawners(2)

A fully seeded unit of WUA_spawn would produce enough fry to fully seed nine units of WUA_rear (i.e., 0.0142/0.00165). This is the "Habitat Ratio" of Bovee (1982).

The Bureau of Reclamation operations model yielded mean monthly flow in cubic feet per second. The first step in converting mean. monthly flow to potential fish production was to create a subroutine containing the WUA vs. flow (Q) functions for the various reaches and then to allow the model to convert the monthly flows to WUA for spawning and for rearing. The model, with a 52-year period of analysis, permitted us to analyze time-series to track limiting life stages for each water year and year class.

We estimated fish production based on rearing for any given reach by averaging the WUA available during July, August, and September. Production in the Yakima Basin can be limited by severely low flows or by severely high irrigation flows during this period. In most instances, the simulation suggests that production (based on rearing) in the Yakima Basin is limited by high flows during the irrigation season. Fish production based on spawning was calculated as the average of the WUA available during the appropriate spawning months.

We then calculated 1) summed WUA for rearing for a given species for the entire basin; 2) summed WUA for spawning the previous fall for the entire basin; 3) WUA totals converted to common units; and 4) production for a given water year, based on the life stage having the lowest equivalent value. The common unit used in the model was "equivalent redds." In the computer simulations, simulated production was primarily governed by rearing habitat availability.

For the WUA_spawn, we applied the conversion factor (from Equation (2)) to calculate equivalent redds; then we considered incubation flows to get net equivalent redds, and then we made the comparison with rearing equivalent redds. The shape of the incubation conversion curve was determined by comparing stage-discharge relationships from IFIM data with the spawning depth preference curves developed for spring chinook in the Yakima Basin. We calculated a ratio of minimum incubation Q to maximum spawning Q, converted that to "% redds viable" using the incubation conversion curve, and multiplied that number by the original WUA_spawn equivalent redds to obtain net spawning equivalent redds.

In the first water year a returning year class was calculated in the first few water years in the model. In Water Year 5 we can calculate spawning and rearing habitat for spring chinook and determine which of these governs production, but if the production in Water Year 1 was extremely low due to a drought, full production for Water Year 5 would probably not be realized due to underseeding (unless continual hatchery supplements are anticipated). Thus we have developed a term called Recovery Equivalent Redds (RFR) for each water year. This term is the number of equivalent redds in Water Year n-4 plus a constant recovery potential that recognizes that density dependent mortality in Water Year n will be somewhat decreased if production is below carrying capacity. The RER term was determined by assuming that production can expand from the lowest level predicted by our model to optimum production in three generations (12 years). This assumption is based on the premise that compensation will occur when fish densities are below carrying capacity (Beverton and Holt 1957).

We calculated the production for a particular year; when that year class returned, we added the recovery potential constant and if this RER was lower than spawning equivalent redds and rearing equivalent redds, then the calculated potential production for the year under consideration could be no higher than the RER value. The RER value rarely came into play in the model, the only occasions being the severe drought years in the historical record used as simulation flows. For example, the poor production during the 1941 draught not only affected predicted production in 1941 due to poor rearing flows and in 1942 due to low spawning flows, but it also decreased predicted production in subsequent years when those year classes returned-that is, 1945-46 and 1949-50. A good example is the low return of adults experienced in the Naches River in 1980 due to the drought of 1977 (Naches spawners are predominantly 5-year-olds) and the subsequent low number of smolts emigrating from this subdrainage when habitat conditions were good.

Thus for each water year, production was determined by the lowest of: 1) net spawning equivalent redds, 2) rearing equivalent redds, and 3) recovery equivalent redds. Average annual production was then calculated by simply computing the average of the production for the 52 years analyzed.

This methodology can be used to estimate potential production based on microhabitat availability. In the Yakima Basin, factors other than microhabitat, such as inadequate screens and ladders and excessive harvest, can also act to depress fish production. Using this method, we estimated that the Yakima Basin can, on the average, support an. escapement of roughly 30,000 spring chinook spawners provided that state-of-the-art fish screens and ladders are maintained, the Bureau of Reclamation operates the Yakima Project to meet fishery flow targets while fulfilling irrigation demands, and harvest of spring chinook salmon is properly managed.

I recommend that this methodology for estimating potential production be used with caution and with a full understanding of the assumptions involved. Efforts should be taken to minimize nonhabitat-related effects such as inadequate passage facilities and overharvest. Further investigations are needed in order to:

1. Better define microhabitat preferences for fry, fingerling, overwinter, and spawning life stages;

2. Validate the assumption that WUA is an indicator of relative as well as absolute fish abundance;

3. Develop multivariate preference functions that consider the interrelationships of microhabitat variables;

4. Continue long-term monitoring to develop survivorship curves and to describe stock recruitment relationships; and

5. Refine the Yakima IFIM to incorporate state-of-the-art techniques.

¹Wasserman, biologist, Yakima Indian Nation, Toppenish, WA 98948. Pers. commun., 1983.

Beverton, R. J. H., and S. J. Holt. 1957. On the dynamics of exploited fish populations. Fish. Invest., London, Series 2, 19:1-633.

Bovee, K D. 1982. Instream flow information paper No. 12, 249 p. U.S. Fish and Wildlife Service, WELUT, Ft. Collins, CO 80521.

Burner, C. J. 1951. Characteristics of spawning nests of Columbia River salmon. U.S. Fish and Wildlife Service, Fish. Bull. 61:1-14.

Everest, F. H., and D. W. Chapman. 1972. Habitat selection and spatial interaction by juvenile chinook salmon and steelhead trout in two Idaho streams. J. Fish. Res. Board Can. 29:91-100.

Fast, D., L. Wasserman, and J. Hubble. 1985. Yakima River spring chinook enhancement study. Yakima Indian Nation report to Bonneville Power Administration, Project No. 82-16, 86 p. (Available from Yakima Indian Nation, P.O. Box 151, Toppenish, WA 98948.)

Lindsay, IL B., W. J. Knox, M. W. Fletcher, B. J. Smith, L. S. Lutz, and E. A. Clean. 1986. Study of wild spring chinook salmon in the John Day River System, 87 p. Bonneville Power Administration and Oregon Department of Fish and Wildlife. Portland, Oregon (Available from Bonneville Power Administration, P.O. Box 3621, Portland, OR 97208.)

McIntyre, J. D. 1983. Program in development of guidelines for outplanting, 74 p. National Fisheries Research Center, U.S. Fish and Wildlife Service. (Available from National Fisheries Research Center., Bldg. 204, Naval Station Puget Sound, Seattle, WA 98115.)

Major, R L., and J. L. Mighell. 1969, Egg to migrant survival of spring chinook salmon in the Yakima River, Washington. Fish, Bull., U.S. 67(2):347-349.

Q:

When you looked at the effects of flows in the Yakima River, what was the effect of flows on temperature effects?

A:

Temperature is only a problem below Sunnyside, where the gradient becomes steep. In general, this is not a problem for spring chinook salmon.

Q:

In rivers that haw been channeled, such as the Trinity River, how can you apply PHABSIM with only one flow?

A:

What I meant is that one applies three different flows independently to get a range of hydraulic conditions.

Q:

Is 30,000 spring chinook salmon a fairly firm estimate? What about other species? Can you base the potential production on what one used to see in those systems?

A:

This is what we came up with, so we used it. With the IFIM analysis, you use the available data or you will go broke fast. The best habitat has been cut off by the dams, so we know that the streams won't produce what they used to. We tried to look at historical records, but this is probably good only in a gross way.

A:

There are many different combinations; for example, sometimes cover is weighted heavily, other times velocity. One needs an answer that is practical.

Table of Contents