TY - JOUR
T1 - Time-varying natural mortality in fisheries stock assessment models
T2 - Identifying a default approach
AU - Johnson, Kelli F.
AU - Monnahan, Cole C.
AU - McGilliard, Carey R.
AU - Vert-Pre, Katyana A.
AU - Anderson, Sean C.
AU - Cunningham, Curry J.
AU - Hurtado-Ferro, Felipe
AU - Licandeo, Roberto R.
AU - Muradian, Melissa L.
AU - Ono, Kotaro
AU - Szuwalski, Cody S.
AU - Valero, Juan L.
AU - Whitten, Athol R.
AU - Punt, A. E.
N1 - Publisher Copyright:
© 2014 © International Council for the Exploration of the Sea 2014. All rights reserved.
PY - 2014/11/4
Y1 - 2014/11/4
N2 - A typical assumption used in most fishery stock assessments is that natural mortality (M) is constant across time and age. However, M is rarely constant in reality as a result of the combined impacts of exploitation history, predation, environmental factors, and physiological trade-offs. Misspecification or poor estimation of M can lead to bias in quantities estimated using stock assessment methods, potentially resulting in biased estimates of fishery reference points and catch limits, with the magnitude of bias being influenced by life history and trends in fishing mortality. Monte Carlo simulations were used to evaluate the ability of statistical age-structured population models to estimate spawning-stock biomass, fishing mortality, and total allowable catch when the true M was age-invariant, but time-varying. Configurations of the stock assessment method, implemented in Stock Synthesis, included a single age- and time-invariant M parameter, specified at one of the three levels (high, medium, and low) or an estimated M. The min-max (i.e. most robust) approach to specifying M when it is thought to vary across time was to estimate M. The least robust approach for most scenarios examined was to fix M at a high value, suggesting that the consequences of misspecifying M are asymmetric.
AB - A typical assumption used in most fishery stock assessments is that natural mortality (M) is constant across time and age. However, M is rarely constant in reality as a result of the combined impacts of exploitation history, predation, environmental factors, and physiological trade-offs. Misspecification or poor estimation of M can lead to bias in quantities estimated using stock assessment methods, potentially resulting in biased estimates of fishery reference points and catch limits, with the magnitude of bias being influenced by life history and trends in fishing mortality. Monte Carlo simulations were used to evaluate the ability of statistical age-structured population models to estimate spawning-stock biomass, fishing mortality, and total allowable catch when the true M was age-invariant, but time-varying. Configurations of the stock assessment method, implemented in Stock Synthesis, included a single age- and time-invariant M parameter, specified at one of the three levels (high, medium, and low) or an estimated M. The min-max (i.e. most robust) approach to specifying M when it is thought to vary across time was to estimate M. The least robust approach for most scenarios examined was to fix M at a high value, suggesting that the consequences of misspecifying M are asymmetric.
KW - Stock Synthesis
KW - model misspecification
KW - natural mortality
KW - population models
KW - reference points
KW - simulation
KW - time-varying
UR - http://www.scopus.com/inward/record.url?scp=84899078077&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84899078077&partnerID=8YFLogxK
U2 - 10.1093/icesjms/fsu055
DO - 10.1093/icesjms/fsu055
M3 - Article
AN - SCOPUS:84899078077
SN - 1054-3139
VL - 72
SP - 137
EP - 150
JO - ICES Journal of Marine Science
JF - ICES Journal of Marine Science
IS - 1
ER -