Modeling of maximum and minimum temperature of warm-water fish breeding pond using machine learning methods

Articles in Press

Document Type : Original Article

Authors

1 ​Research Institute of Meteorological and Atmospheric Science (RIMAS), Climate Research Institute (CRI), Mashhad, Iran

2 Expert of Gilan Agricultural Meteorology Research Center

3 Agricultural meteorological Research center of Guilan

4 South of Iran Aquaculture Research Institute, Iranian Fisheries Science Research Institute, Agricultural Research, Education, and Extension Organization (AREEO), Ahvaz, Iran

10.22125/agmj.2024.410815.1157

Abstract

Fish’s body temperature is almost the same as the temperature of the water in which they live. Water temperature and fish metabolism are closely related. Temperature is a vital environmental factor that strongly affects the nutrition and growth of fish. Temperature changes cause stress in fish, and disease outbreaks occur after sudden temperature changes or when temperatures are chronically close to their maximum tolerance. Therefore, according to the importance of the subject, in this article, with the help of daily observation data of Rasht agricultural meteorological station in Gilan province, related to the period of June 2016 to November 2018, the maximum and minimum temperature variables of fish breeding pond using several Machine learning methods is modeled. For this purpose, artificial neural network, gradient boosting and random forest methods were used to model the maximum and minimum temperature of the ponds of a fish breeding complex. The results obtained from the evaluation of the performance of these methods on the test data showed that for modeling the minimum temperature, the neural network (with a root mean square of 1.93 and a correlation of 0.92) and for modeling the maximum temperature, the random forest model (with a root mean square of 1.61 and a correlation of 0.95 ) are more accurate. One of the applications of the proposed modeling is that by using meteorological forecast data as an input to the proposed model, it is possible to predict the maximum and minimum temperature for the fish breeding pond on a daily basis.

Keywords

Main Subjects

Journal of Agricultural Meteorology

Articles in Press, Accepted Manuscript
Available Online from 29 January 2024

Files

Share

How to cite

Statistics