Data Access in Real Time
Predictions
Trends
Tools
Publications
About
Columbia Basin Research hosts this web page for archival purposes only. This project/page/resource is no longer active.

Harvest Modeling Project

From 1996 through 1999, the NMFS Harvest Modeling Project web page coordinated the work of multiple agencies to develop and implement various harvest algorithms including the CRiSP Harvest model, PM Selective Fishery Modeling and Data Retrieval, and the PSC Selective Fishery Simulation Model.

Invitation/Introduction

Creation of a model oversight committee issued October 1996

I wish to invite you to participate in the development of a new model for salmon management. By combining modern programming techniques and statistical-mathematical approaches we seek to develop a flexible model to help meet the future needs of salmon management. Our immediate task is to develop a model for the harvest and migration of salmon on the Pacific coast of North America which should assist managers in evaluating stock specific harvest policies. A long term goal is to help integrate databases, statistical techniques and salmon ecological theory into a flexible modeling system that can be used to evaluate alternative management policies.

The project's success ultimately will be measured in its utility to salmon managers. For this reason we are forming an advisory committee to provide input through all aspects of the project including selecting ecological theory, statistical techniques, model display capabilities, distribution and maintenance of software, protocols for within-season management issues and more. The advisory committee will assist us in developing a solid statistical and biological model in a form usable to salmon managers.

Within a year we hope to have a prototype fish migration/harvest model that can be used to evaluate a stock selective harvest policy. The new model will start from existing work including: a stochastic harvest model (based on the PSC chinook technical committee model), a state-space approach harvest theory, an object oriented software paradigm and a graphical user interface. The new model will use graphical tools to visualize the fishery (see figure) and will run on Window 95, NT and UNIX computer operating systems.

Technical details can be found under the project World Wide Web site, which includes details of the state-space statistical approach and a downloadable stochastic harvest model (CRiSP Harvest) demonstrating the look and feel of the modeling approach.

To describe fish migration and harvest in the fisheries we will apply a state-space model {For a nice discussion see J. Schnute's paper "A general framework for developing sequential fisheries models" Canadian J. of Fisheries and Aquatic Sci., 1994. pp 1676-1688}. In this approach two processes are modeled: a population model describing within-season changes in stock abundance and distribution in the fisheries, and a catch model describing within-season catch from individual stocks in particular fisheries. The model will be modularized. One module describes initial spatial location of fish stocks, a second describes natural and fishing mortalities, and a third describes fish movement patterns between time steps. The model will be calibrated with actual abundance and coded-wire tag recoveries by catch area of particular stocks.

We are inviting you to appoint a member of your organization to our advisory committee. Ideally this person should be currently involved in salmon management, familiar with management needs and have a fair understanding of fisheries science, statistics and computer programming. This person will be ask to attend twice yearly meeting and to review our approaches, recommend additional approaches and identify needed data and model requirements. On our part we will provide quarterly progress reports to the committee and arrange meetings to demonstrate the model. The lifespan of the committee is expected to be two years, from October, 1996 to the end of September, 1998. We will provide to your agency the model with full documentation.

The first advisory meeting is tentatively scheduled for Friday November 22, 1996. To provide the advisors sufficient lead time could you contact me (e-mail or phone) of your appointment by Friday November 15, 1996. The meeting will be held at our downtown Seattle office: Columbia Basin Research Puget Sound Plaza 1325 - 4th Ave., Suite 1820 Seattle, WA 98101-2509. Directions can be found on the project Web page.

The project is funded by the National Marine Fisheries Service and includes a team of biologists, oceanographers, statisticians, mathematicians and computer programmers from the University of Washington and University of Idaho. The senior staff include:

  • Jim Anderson - Principal Investigator, Associate Professor, University of Washington.
  • Ken Newman - Statistician, Assistant Professor, University of Idaho.
  • Jim Norris - Fisheries Biologist, University of Washington.
  • Troy Frever - Programmer, University of Washington.
  • Judy Cress - Database manager, University of Washington,

Meetings & Discussions

Models

  • Coast Model Executable, Data Files, Documentation and Source Code
  • PM Selective Fishery Modeling and Data Retrieval
    • PM Model

Papers & Documents