Prediction of scorpion envenomation risk and the impact of climate change using neurovolutionary network

• Grant number: 24/07110-0
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: March 01, 2025
• End date: February 28, 2026
• Field of knowledge: Biological Sciences - Ecology - Applied Ecology
• Principal Investigator: Rodrigo Hirata Willemart
• Grantee: Welton Dionisio da Silva
• Host Institution: Escola de Artes, Ciências e Humanidades (EACH). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

Scorpion envenomation represents a serious concern for public health worldwide, causing thousands of deaths yearly, in special in tropical and subtropical countries. In Brazil, these animals are the second leading cause of human poisoning, with the northeastern region being the most affected. These accidents are mainly associated with the scorpion Tityus serrulatus, whose population growth is favored by asexual reproduction, high prolificacy, and a lack of basic infrastructure in urban areas, such as poor sewage management and inadequate garbage disposal, promoting profitable environments for these animals. Control measures are implemented by state health departments, although face challenges in mitigating this complex issue. Therefore, this study aims to understand the key factors associated with the increase in scorpion accidents in Brazil, developing an algorithm capable of predicting incidents in the short-term (present time) and long term (future scenarios). In this latter, it will evaluate the potential effects of climate changes under five different socioeconomic projections for the year 2100. It is expected that scenarios with high inequality, consumption and population growth will have a higher risk of scorpion accidents (scenarios SSP3-6.0, SSP4-6.0 e SSP5-8.5). Furthermore, sustainable or moderate socioeconomic development is expected to reduce this risk by 2100 (SSP1-2.6 e SSP2-4.5). To investigate this, data from 32 bioclimatic, socioeconomic, and environmental variables will be used to train an artificial neural network optimized by the genetic algorithm, known as a neuroevolutionary network. These same approaches of neural network will be used to understand how socioeconomic development and climate changes affect the risk of accidents. The results of this study can contribute to decision-making by enabling the efficient allocation of resources and preventive interventions in public health.