FAO Home>Fisheries & Aquaculture
FAO of the UN

Various equations allowing for the introduction of an environmental variable into surplus production models were proposed by Freon (1988). CLIMPROD helps the user select the most appropriate model for a particular case, according to objective criteria. It resembles a simple expert system and uses artificial intelligence language (PROLOG) to converse with the user.

The basic input and output is summarized as follows. CLIMPROD requires annual dataseries on catch and effort of a fishery on a single stock, and annual (or seasonal) data series on an environmental variable known to influence the abundance or the catchability of this stock. At least 12 years' observation are required. The software provides a statistical and graphical description of the dataset and then allows an appropriate model to be chosen, fitting it using a nonlinear regression routine and trying to assess the fit with parametric and nonparametric tests before presenting the tables and graphs of the results. These results may explain how environment and fishing effort governed the yields of the fishery during the period under study. Based on an e stimate of the next two years' data on effort and environmental regime, an exploratory prediction is proposed for catches per unit of effort (CPUE), and a graph gives variations in "maximum sustainable yields mSY)" and "optimal fishing efforts (fmax)" according to different environmental regimes.

CLIMPROD can also be used to fit the three conventional surplus production models (linear, exponential, generalized) without environmental influence or to fit the relationship between environment and production without fishing influence. In both cases nonparametric tests and graphical facilities are available. Because of the poor predictive ability of surplus production models, limitations in forecasting capacity have been included, with the possibility of using CLIMPROD as a training tool in mind.

Key Features:

  1. Data input
  2. Monovariate statistics and graphs of raw data
  3. Examination of timeseries
  4. Bivariate graphics of raw data
  5. Computer assisted choice of the model
  6. Useri mposed choice of the model
  7. Model fitting
  8. Statistical tests of robustness (validation)
  9. Output
    1. Description of the dataset
    2. Mathematical results of the model fitting
    3. Graphical results of the model fitting
    4. Limited predictions
    5. Summary of expert decisions

System Requirements:

  • CLIMPROD requires the following hardware and operating system:
  • an IBM or 100% compatible Personal Computer (PC, XT, AT)
  • at least 512 KB RAM (random access memory)
  • one disk drive for loading the program and a hard disk with at least 1.5 MB available
  • MSDOS, Version 3.0 or higher
  • a standard DOS keyboard
  • a printer if you want an output of the results and graphs (at the moment the software is not interfaced with plotters)
  • a mathematical coprocessor is not required, but will be very useful for reducing computer time in model fitting and validating

Statistics and Information Branch (FIAS)

Fisheries and Aquaculture Department
Food and Agriculture Organization of the United Nations
Viale delle Terme di Caracalla

Email: Fish-Statistics-Inquiries

Powered by FIGIS