A new approach of fitting biomass dynamics models to data

Area of Science:

• Ecology
• Fisheries Science
• Computational Biology

Background:

• Dynamic resource biomass models are crucial for fisheries management.
• Traditional parameter estimation can be computationally intensive and less accurate.
• Accurate stock assessment requires integrating diverse data sources.

Purpose of the Study:

• To develop and apply a non-traditional, computationally efficient method for fitting dynamic resource biomass models.
• To estimate parameters for bioeconomic models using logistic and Gompertz growth functions.
• To analyze the North-east Arctic cod stock and assess fishing pressure relative to maximum sustainable yield.

Main Methods:

• A variational adjoint technique for dynamic parameter estimation.
• Minimization of a cost function comparing model solutions to observational data.
• Constrained least squares method for parameter estimation in logistic and Gompertz population growth models.

Main Results:

• The variational data assimilation method provides a novel and efficient procedure for analyzing resource systems.
• Parameter estimates for the population dynamics models were found to be reasonable.
• The average fishing mortality for the North-east Arctic cod stock was determined to be significantly above the maximum sustainable yield.

Conclusions:

• The developed data assimilation technique offers an efficient approach to resource system analysis.
• The North-east Arctic cod stock is experiencing fishing pressure exceeding sustainable levels.
• This methodology can improve the accuracy of fisheries stock assessments and inform management decisions.