Fitted Parameters and Assumptions Used in the StCroix Model
Type
Publication
Authors
Category
SCIWC Library
[ Browse Items ]
Publication Year
1984
Pages
8
Subject
St. Croix Model and Fisheries Management
Tags
Abstract
Introductory letter from W. D. Watt (Fish Habitat, Protection Section, Fisheries Research Branch) to Mr. Ken Beland, Atlantic Sea Run Salmon Commission on June 22, 1984.
Dear Sir:
In the process of preparing a draft Fisheries Management Plan for the St. Croix River (material now being considered by the Steering Committee), Dick Cutting and I became aware of the need to estimate the salmon returns to be expected from stocking efforts and of the desirability of a flexible system for convenient recomputation of those returns. an obvious solution to these requirements was to prepare a computer model of the St. Croix salmon runs arising from past hatchery stocking efforts. This technique has the advantage that 'best guess' estimates based on data from other sites can be incorporated immediately ( for parameters such as upstream fish passage efficiency, and downstream migration survival at turbine sites and headponds) into the model and then varied for sensitivity analysis, thus giving some immediate indication of how successful we may be in our goal of establishing a self-sustaining run of Atlantic Salmon. When our knowledge of local St. Croix conditions becomes more refined, various relevant statisitica parameters (e.g., sea survival, age of smoltification and angling losses) which are presently estimated from donor stock performances at other sites can be fitted to the St. Croix Model, and the return expectations can be recalculated.
Attached is an accounting of the various parameters and their values that are currently incorporated into the model. The model is open-ended and other parameters can be incorporated now if their impact on stock recruitment can be expressed mathematically, and the necessary parameters fitted. Note that certain parameters have pessimistic median and optimistic assumed values.
To illustrate the potential use of the model for salmon restoration and management planning, I have attached some tabular and graphic output showing: (1) a fifteen-year projection of grilse and salmon returns given the stocking efforts completed to date and using median assumptions (table and graph); and (2) 'sensitivity analysis on fish-pass efficacy and downstream mortality (clustered bar chart). The model cannot be expected to give a serous predication of actual returns 15 years into the future, but it should indicate the stock recruitment trend. To achieve a self-sustaining Atlantic salmon run in the St. Croix will require a strong positive growth trend. Under 'median' assumptions (our current 'best guesses'), the model predicts a declining trend, though of course future runs can be sustained by a continuous stocking program. Even when downstream survival or fish-pass efficiency are singularly raised to optimistic assumptions for these two parameters are entered together is a positive growth response indicated. It appears fortunate for our undertaking that the two parameters most amenable to technological improvement are also critical for reproductive success.
I have also tried the model with pure Saint John and pure Narraguagus stock characteristics (instead of allowing them to blend) and similar results were obtained. The model thus suggests that the choice of stock characteristics is not critical in comparison to upstream and downstream fish passage successes.
My conclusion from this analysis is that it is essential to calibrate the river as soon as possible for the upstream fish-pass efficiency and downstream survival parameters.
The model is now available for use as a restoration and management tool. It operates on an H.P 150 micro-computer with 256 k RAM, and uses the spread sheet 'VisiCalc'. I solicite your suggestions for: additional parameters that should be fitted, better estimates for the present parameters, and any other constructive criticisms or suggestions."
Dear Sir:
In the process of preparing a draft Fisheries Management Plan for the St. Croix River (material now being considered by the Steering Committee), Dick Cutting and I became aware of the need to estimate the salmon returns to be expected from stocking efforts and of the desirability of a flexible system for convenient recomputation of those returns. an obvious solution to these requirements was to prepare a computer model of the St. Croix salmon runs arising from past hatchery stocking efforts. This technique has the advantage that 'best guess' estimates based on data from other sites can be incorporated immediately ( for parameters such as upstream fish passage efficiency, and downstream migration survival at turbine sites and headponds) into the model and then varied for sensitivity analysis, thus giving some immediate indication of how successful we may be in our goal of establishing a self-sustaining run of Atlantic Salmon. When our knowledge of local St. Croix conditions becomes more refined, various relevant statisitica parameters (e.g., sea survival, age of smoltification and angling losses) which are presently estimated from donor stock performances at other sites can be fitted to the St. Croix Model, and the return expectations can be recalculated.
Attached is an accounting of the various parameters and their values that are currently incorporated into the model. The model is open-ended and other parameters can be incorporated now if their impact on stock recruitment can be expressed mathematically, and the necessary parameters fitted. Note that certain parameters have pessimistic median and optimistic assumed values.
To illustrate the potential use of the model for salmon restoration and management planning, I have attached some tabular and graphic output showing: (1) a fifteen-year projection of grilse and salmon returns given the stocking efforts completed to date and using median assumptions (table and graph); and (2) 'sensitivity analysis on fish-pass efficacy and downstream mortality (clustered bar chart). The model cannot be expected to give a serous predication of actual returns 15 years into the future, but it should indicate the stock recruitment trend. To achieve a self-sustaining Atlantic salmon run in the St. Croix will require a strong positive growth trend. Under 'median' assumptions (our current 'best guesses'), the model predicts a declining trend, though of course future runs can be sustained by a continuous stocking program. Even when downstream survival or fish-pass efficiency are singularly raised to optimistic assumptions for these two parameters are entered together is a positive growth response indicated. It appears fortunate for our undertaking that the two parameters most amenable to technological improvement are also critical for reproductive success.
I have also tried the model with pure Saint John and pure Narraguagus stock characteristics (instead of allowing them to blend) and similar results were obtained. The model thus suggests that the choice of stock characteristics is not critical in comparison to upstream and downstream fish passage successes.
My conclusion from this analysis is that it is essential to calibrate the river as soon as possible for the upstream fish-pass efficiency and downstream survival parameters.
The model is now available for use as a restoration and management tool. It operates on an H.P 150 micro-computer with 256 k RAM, and uses the spread sheet 'VisiCalc'. I solicite your suggestions for: additional parameters that should be fitted, better estimates for the present parameters, and any other constructive criticisms or suggestions."
Description
This document provides a review of fish management policies, procedures, and methods used in the 1980's with a focus on the St. Croix River and Milltown Dam.
Sections Include:
- For adult transfers from St. John
- Green Lake Hatchery
- Tagged Green Lake smolts stocked for evaluation of downstream mortalities are assumed to return to Milltown with a survival rate of 0.25% (similar to that observed for the 1981 stocking).
- 'Wild' returns to Milltown fishway 1981-83
- 'Wild' returns to Milltown after 1983
- Median assumptions (70% downstream survival and 49% fish-pass efficiency)
- Projected Returns of Grilse and Salmon to St. Croix River from CroINASJ Computer Model (Median Assumptions)
- Projected Returns of Grilse + Salmon to Milltown
Sections Include:
- For adult transfers from St. John
- Green Lake Hatchery
- Tagged Green Lake smolts stocked for evaluation of downstream mortalities are assumed to return to Milltown with a survival rate of 0.25% (similar to that observed for the 1981 stocking).
- 'Wild' returns to Milltown fishway 1981-83
- 'Wild' returns to Milltown after 1983
- Median assumptions (70% downstream survival and 49% fish-pass efficiency)
- Projected Returns of Grilse and Salmon to St. Croix River from CroINASJ Computer Model (Median Assumptions)
- Projected Returns of Grilse + Salmon to Milltown
Number of Copies
1
