“The Elephant in the Room”: Exploring Natural Mortality Uncertainty in Statistical Catch at Age Models

Abstract

The natural mortality rate (M) of a fish stock is typically highly influential on the outcome of age-structured stock assessment models, but at the same time extremely difficult to estimate. In data-limited stock assessments, M usually relies on a range of empirically or theoretically derived M estimates, which can vary vastly. This article aims at evaluating the impact of this variability in M using seven Mediterranean stocks as case studies of statistical catch-at-age assessments for information-limited fisheries. The two main bodies carrying out stock assessments in the Mediterranean and Black Seas are European Union’s Scientific Technical Economic Committee for Fisheries (STECF) and Food and Agriculture Organization’s General Fisheries Commission for the Mediterranean (GFCM). Current advice in terms of fishing mortality levels is based on a single “best” M assumption which is agreed by stock assessment expert working groups, but uncertainty about M is not taken into consideration. Our results demonstrate that not accounting for the uncertainty surrounding M during the assessment process can lead to strong underestimation or overestimation of fishing mortality, potentially biasing the management process. We recommend carrying out relevant sensitivity analyses to improve stock assessment and fisheries management in data-limited areas such as the Mediterranean basin.

Natural mortality: Theory, estimation and application in fishery stock assessment models

“…However, this algorithm does not admit a unique solution and we thus proposed a variant (Martiradonna 2012) by imposing an adult mortality constraint that allows us to obtain explicit formulas for the parameter estimation. This variant has been applied recently in literature (Mannini et al 2020, Pantazi et al 2020) to some Mediterranean stocks as case studies of statistical catch-at-age assessments. This work is aimed at showing the gain in stability of our finding, applying the two methods to different stocks in the Mediterranean Sea.…”

Editorial: Data-limited research in stock assessment to increase the understanding of fisheries resources and inform and improve management efforts

“…Source: www.scopus.com. Mannini et al, 2020;Omori et al, 2021;Rudd al., 2021;Sańchez-Maroño et al, 2021).…”

Informing Management of Henrys Lake, Idaho, using an IntegratedCatch‐at‐AgeModel

“…Consequently, assumptions are often made about the rate of natural mortality—a commonly recognized shortcoming of fisheries stock assessment models (Maunder and Piner 2015; Mannini et al. 2020). This is problematic when evaluating the proposed alternative management actions at Henrys Lake because opening the fishery for a longer time period could potentially affect fishing mortality, whereas changing stocking rates will likely affect natural mortality.…”

“…However, separately estimating fishing and natural mortality is not possible by combining just angler catch data with fishery-independent survey data using the traditional framework, as the parameters are confounded (Quinn and Deriso 1999). Consequently, assumptions are often made about the rate of natural mortality-a commonly recognized shortcoming of fisheries stock assessment models (Maunder and Piner 2015;Mannini et al 2020). This is problematic when evaluating the proposed alternative management actions at Henrys Lake because opening the fishery for a longer time period could potentially affect fishing mortality, whereas changing stocking rates will likely affect natural mortality.…”