MODELING AND ARTIFICIAL INTELLIGENCE FOR PREDICTING CORN PRODUCTION IN MATO GROSSO

• Grant number: 23/10140-6
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: December 01, 2023
• End date: November 30, 2024
• Field of knowledge: Agronomical Sciences - Agronomy - Agricultural Meteorology
• Principal Investigator: Fabio Ricardo Marin
• Grantee: Ulysses Chaves de Menezes Netto
• Host Institution: Escola Superior de Agricultura Luiz de Queiroz (ESALQ). Universidade de São Paulo (USP). Piracicaba , SP, Brazil

Abstract

This project aims to address the importance of sustainable agricultural production to achieve global food security amidst challenges such as climate change and land use restrictions. Brazil plays a crucial role in this context,being a leader in the agricultural trade balance and meeting the growing demand of developing nations. Maize is an extremely important crop, providing about 60% of global energy intake and being required both for human food and animal feed, as well as for biofuel production.The state of Mato Grosso is a highlight in corn production, mainly in the second crop. Given the scenario of climate variations and the importance of maize for food security, this study proposes the development of an artificial intelligence (AI) algorithm to predict the productivity of maize second crop in the state of Mato Grosso.The project will use space-time data, including climatic and socioeconomic information, to identify climatic indicators in the months prior to the beginning of the harvest, allowing to anticipate possible variations in corn production. The work will seek the best sources of data, including meteorological and climate information,socioeconomic data and information on the El-Niño/Southern Oscillation (ENSO) phenomenon. It is expected that, at the end of the study, the performances of different AI algorithms to estimate corn productivity and identify the main variables that explain the variability of crop production in the state of Mato Grosso will be presented.