Machine intelligence applied to animal bioclimatology: Predictive models between the thermal environment and adaptive and productive responses for farm animals

• Grant number: 22/14250-8
• Support Opportunities: Scholarships in Brazil - Doctorate
• Effective date (Start): April 01, 2023
• Effective date (End): September 01, 2025
• Field of knowledge: Agronomical Sciences - Agricultural Engineering - Rural Buildings and Ambience
• Principal Investigator: Iran José Oliveira da Silva
• Grantee: Robson Mateus Freitas Silveira
• Host Institution: Escola Superior de Agricultura Luiz de Queiroz (ESALQ). Universidade de São Paulo (USP). Piracicaba , SP, Brazil
• Associated scholarship(s): 23/16733-9 - COMBINING MACHINE LEARNING AND PATH ANALYSIS TO IDENTIFY ADAPTIVE MECHANISMS IN FARM ANIMALS, BE.EP.DR

Abstract

Future projections on global warming reveal an increase of up to 2°C in the temperature of the globe by the year 2050, combined with the increase in the food production demand due to the growing human population on the planet, which installs a paradox. In this case, it reinforces the need to carry out studies in order to understand the impact of climate on the thermoregulation physiology of farm animals, as well as the impact of heat stress on productive performance. However, the general field of heat stress biology lacks an integrated theory and associated predictive models to allow quantitative research connecting the multiple mechanisms and outcomes involved in animals' adaptation to the environment. In this context, the application of machine learning is one of the statistical tools available for assertive decision making. Here we propose to use morphological, thermoregulatory, hormonal, hematological and productive responses of five goat breeds (animal model) from different climatic conditions to investigate the potential causal effects underlying the complex relationships of climate × animal adaptation × animal production, besides simulating and revealing the adaptive profile of these populations in different environmental scenarios and developing an adaptability index. The choice to evaluate the goat species is justified by the fact that goats are considered the ideal animal model for climate change, due to their phenotypic plasticity in scenarios of rapid changes in tolerance limits for temperature, food and water resources. This study is part of a regular project supported by FAPESP in the area of animal ambience and machine learning associated with the development of products, services and innovation, enabled by technologies aimed at agriculture and livestock 4.0. This study is expected to contribute to the expansion of scientific frontiers regarding the revelation of complex processes that will help to understand adaptive responses in livestock to climate change, besides identifying phenotypic markers that can be used to quantify responses to thermal stress.