Estimation of Frost Hazard for Tea Tree in Zhejiang Province Based on Machine Learning

Tea trees are the main economic crop in Zhejiang Province. However, spring cold is a frequent occurrence there, causing frost damage to valuable tea buds. To address this, regional frost-hazard early-warning system needed. In this study, area was estimated based on topography and meteorology, as well longitude latitude. Based support vector machine (SVM) artificial neural networks (ANNs), multi-class classification model proposed estimate of disasters using cases from 2017. Results two models were compared, optimal parameters adjusted through multiple iterations. The highest accuracies 83.8% 75%, average 79.3% 71.3%, Kappa coefficients 79.1% 67.37%. SVM selected establish spatial distribution Province 2016. Pearson’s correlation coefficient between prediction results meteorological yield 0.79 (p < 0.01), indicating consistency. Finally, importance factors assessed sensitivity analysis. show that relative humidity wind speed key influencing accuracy predictions. This study supports decision-making for hazard defense facing frost.