The Catch-Survey Analysis (CSA) method of fish stock assessment: an evaluation using simulated data

摘要

The Catch-Survey Analysis (CSA) method of assessment aims at estimating stock abundance from relative indices by filtering measurement error in the latter through a simple two-stage population dynamics model. The method is not widely used and the associated literature is still limited. The objective of this work is to improve current understanding of the properties of the method, using data generated from a known fully age-structured population. A sensitivity analysis confirms that CSA is capable of providing reliable information about general stock trends. However, absolute estimates are sensitive to the parameter relating survey catchabilities of the two stages, which needs to be estimated externally with methods that warrant further validation. Biased results are also obtained when changes in the catchability of the fleet providing indices are not corrected for during data preparation. Conventional diagnostics fail to detect violations of the constant-q assumption and improved diagnostics are thus needed. CSA is proved to be a very valuable method to support management advice in data-limited contexts in which age data are lacking or uncertain. Subject to some refinements, it can even challenge VPA-based assessments, notably when the latter use uncertain age data.