Estimating commercial fish production in Santa Catarina, Brazil, based on animal transport records and mathematical models

Abstract

This study investigated the correlation between fish transit data, recorded in the Animal Transit Guides (GTA), and fish farming production estimates to develop mathematical models to predict commercial fish production in Santa Catarina. GTA data provided by Cidasc and production estimates made by Epagri between 2015 and 2021 were used. The data was combined and linear regression models were developed based on data from different fish groups. The results showed a significant correlation between GTA data and production estimates, with important correlations for tilapia (R²=0.69) and trout (R²=0.55). When the previous year's production was added to the models as a predictor variable, the models' R² values for tilapia and trout increased to 0.89 and 0.80, respectively. The validation of these two models for 2022 and 2023 indicated good performance, with an excellent fit between the model's predictions and Epagri’s estimates (R² above 0.95). However, the model developed for trout tended to underestimate production. It is concluded that the models produced are effective in predicting the commercial production of tilapia and, to a lesser extent, trout in Santa Catarina.