Collie-Sissenwine Analysis
(CSA)

The Collie-Sissenwine Analysis (CSA) model is a stage-based model that estimates the abundance of two classes, defined as recruits and post-recruits. Recruits are assumed to become fully available for the fishery over the model's unit time step. The model requires indices of abundance for these two stages and estimates of the total number of animals removed via the fishery.  

The model provides estimates of both abundance and mortality rates. The user can flexibly specify alternative hypotheses regarding process error and observation error, including models that use process error only or observation error only. The underlying error structure for observation and process error can be specified as either normal and lognormal. If lognormal error is specified the user can choose whether or not to use bias correction methods. Bootstrapping can be based on either randomly resampling from the set of original error terms or by generating random errors from a parametric distribution with parameters derived from the original set of errors.  

The graphical user interface has full features for model specification, review, and visualization.

The calculation engine is based on an adaptation of previous implementations of the model including DELPOP8 by Dr. Jeremy S. Collie at the University of Rhode Island, an APL version by Dr. Ramon Conser at NMFS Southwest Science Center, La Jolla, CA, and by various specific versions developed by the Population Dynamics branch at NMFS Northeast Science Center, Woods Hole, MA.

The input format has been revised and enhanced.  All former input files are converted automatically to the new format.

Back to Top 

Collie, J. S. and M. P. Sissenwine, 1983. Estimating population size from relative abundance data measured with error. Can. J. Fish. Aquat. Sci., 40: 1871-1879.

Back to Top 

For recent changes please see the CSA documentation, or contact User Support.

Back to Top 

Older Version Changes