ENSEMBLE MODELING OF FALL-RUN CHINOOK PRODUCTION IN THE KLAMATH BASIN, OR AND CA

As part of the upcoming Secretarial Determination to either retain or remove 4 hydropower dams from the Klamath River, a population dynamics model capable of predicting annual abundances of fall-run Chinook with uncertainty was developed. The initial approach was to construct a conceptual model that: 1) captured the biological processes of interest; 2) captured existing hypotheses about environmental or anthropogenic factors thought to affect the population dynamics; 3) identified sources of data for those factors; 4) identified pathways for management actions; and 5) if using statistical methods, identified sources of data that could be used as indices of abundance. Several existing models were embedded in a new full life cycle model for fall Chinook, an adult migration model, an upper basin production model (EDT), an outmigrant mortality model (SALMOD), a retrospective ocean survival model, and a harvest model. Using the conceptual model as a blueprint, a life-stage model (e.g., Leslie-matrix type model) with transition among stages described by stage-specific Beverton-Holt functions was constructed.

The modeling development effort involved four Federal agencies, USGS, FWS, USBR, and NOAA Fisheries, and three consultants, as well as university, State, Tribal and other Klamath Basin stakeholders that all provided models, data, and expert opinion.

The model was used to compare existing conditions (hydropower dams in place) and several alternatives without dams. Preliminary results of model application for analysis of the Secretarial Determination alternatives (with dams, historical and future, and without dams, future predictions) will be presented.