MATLAB-Based Standalone Application for Estimation of Growth and Mortality Parameters of Fish
Abstract
The exploited fish populations and the fishery dependent on them require regular monitoring and periodical stock assessment. The resource evaluation is crucial for definition of appropriate management strategies for rational exploitation and long-term sustainable development of the exploited stocks.
The determination of the current state of a given stock is directly related to analysis of historical and present data concerning specific stock parameters which indicates the current status of the stock and suggests relevant definition of healthy exploitation levels. Such key stock parameters are growth and mortality rates, abundance indices, carrying capacity, maturity and reproduction, catch composition and recruitment [1,3,4,6,13,14].
The present article introduces a program system - MATLAB-based standalone application specifically developed to facilitate length-weight relationship of fish analysis outcome, growth and mortality parameters estimates and their confidence limits, normality tests of length-weight-frequency sample data, used in analysis. The proposed system incorporates statistical analysis of the overall significance of the mathematical models and results delivered, as well as methodology to overcome added computational errors in numerical analysis.
Keywords
Download Options
Introduction
The present material introduces a MATLAB-based standalone application specifically developed to support the fish stock assessment procedure. It can be executed in MATLAB environment as well by simply typing the name of the program GR_MORT_PAR_EST in MATLAB workspace. It is focused on length-weight relationship analysis, von Bertalanffy individual growth parameters and mortality parameters estimation, based on length-frequency, length-weight frequency, length-at-age and catch-at-age samples data, by applying variety of methods introduced in fisheries science, which can be selected depending on the input data type and availability. Another incorporated feature is related to validation of the accuracy of the estimates provided by using detailed statistical analysis of the adequacy and the overall significance of the mathematical models obtained, construction of confidence intervals of the parameters estimated and validation of specific distributional assumptions of the sample data under analysis.
Conclusion
The software developed delivers estimates of key growth and mortality parameters to support stock assessment procedure and analysis of historical and present data. It allows analysis of the overall model significance and delivers a statement for the model adequacy.
The program itself does not add computational error to the analysis results and works well with natural, standardized and logtransformed values of the variables under used for the provisions of analysis.
The methods can be repeatedly executed with new or amended data depending on the user’s requirements and data availability.
In case that input data is available to feed few models, which deliver estimates of same stock/individual parameters, comparison of the estimates delivered might results in more realistic values to support the analysis.
The program is set as standalone application and it does not require MATLAB software installation to operate. In case the script needs further modifications as per the users demand it can be easily modified in MATLAB programming environment. The graphic representations dispose the same features as regular MATLAB graphs.
The results are recorded in a designated .txt file and can be further processed with other applications depending on the user needs.