Avaliação de Produtividade e Saúde Vegetal em Mangueira a partir de Dados de Luminosidade de Abertura de Copa

Abstract

The São Francisco Valley is a central hub for mango production and export in Brazil, playing a crucial role in the local economy. Mango cultivation and viticulture lead to economic activity in the region. Pruning is essential for promoting plant ventilation and reducing pests and diseases, but the lack of defined parameters makes its evaluation subjective. The research aims to correlate mango tree light data with fruit quality to determine an ideal light level. Plants were divided into plots, and 100 fruits from each were assessed. Data analysis in Python involved correlation graphs, decision trees, and confusion matrices. Despite some observed damage, promising results show a strong correlation between higher light levels and better-quality fruits for export. The decision tree effectively classified fruits, identifying higher-quality samples with lower incidences of pests and diseases. The confusion matrix performed satisfactorily predicting batches with more exported fruits. It is concluded that increased light and canopy openness are associated with better fruit quality. The study suggests establishing an ideal light level to optimize the production of high-quality mangoes with further research.