Predicting the impact of climate change on myrtle rust

• Ano de defesa: 2025
• Autor(a) principal: Gonçalves, Manoel Penachio
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Austropuccinia psidii; Aquecimento global; Epidemiologia; Epidemiology; Ferrugem da goiabeira; Fotossíntese; Global warming; Guava rust; Modelagem preditiva; Photosynthesis; Predictive modelling
• Link de acesso: https://www.teses.usp.br/teses/disponiveis/11/11135/tde-04062025-080726/

Predicting the occurrence of plant diseases under climate change scenarios is challenging, and forecasting the damage caused by such diseases is even more complex. Myrtle rust (Austropuccinia psidii) is a globally relevant disease that affects hundreds of species within the Myrtaceae family. In Brazil, rust is the most damaging fungal disease of guava trees, impairing photosynthesis, vegetative development, and fruit production. Although current predictive models suggest that the occurrence of this disease will decrease with rising temperatures in the tropics, these projections do not consider the potential damage caused by the pathogen under high-temperature scenarios. The objectives of this study were: (i) to characterize the physio-histopathological damage resulting from the A. psidii-guava interaction under favorable environmental conditions; (ii) to assess the effects of elevated temperatures on monocyclic components and the photosynthetic damage caused by the disease; and (iii) to develop a myrtle rust infection risk model using daily weather variables as predictors, applying it to forecast the future risk of the disease in São Paulo and Pernambuco, the main guava-producing states in Brazil. In Chapter 1, the response of leaf gas exchange to light intensity was analyzed in healthy and diseased guava plants. Colonization of the substomatal chamber by A. psidii altered stomatal dynamics and compromised leaf gas exchange. The pathogen\'s ability to interfere with stomatal regulation may impair the plant\'s response under climate change scenarios. In Chapter 2, experiments under different thermal regimes demonstrated that increased temperature prolonged the rust latent period and reduced lesion density, disease severity, and pathogen sporulation. However, photosynthetic damage was not proportionally reduced, with severe reductions in gas exchange variables observed in infected plants maintained at high temperatures. Furthermore, A. psidii impaired the plants\' thermal dissipation capacity, indicating a compromised ability to acclimatize to environmental changes. In Chapter 3, a predictive model based on machine learning was developed using daily weather variables as predictors. The model demonstrated high performance in estimating infection risk, offering a promising alternative to models that require hourly predictor variables. Projections generated by the model indicated an increased risk of infection in key guava-producing regions in São Paulo, while the risk in Pernambuco is expected to remain low. Together, the risk estimates and quantification of photosynthetic damage caused by the disease under potential future climate scenarios provide a scientific basis for the development of management strategies aimed at mitigating the impacts of myrtle rust on guava cultivation.