Abstract: This study aimed to investigate the effects of multiple climatic stresses on the photosynthetic efficiency of tea plants and to devise an efficient, precise stress classification system for real-time monitoring. We focused on the typical tea cultivars grown extensively in Fujian Province and systematically monitored their photosynthetic physiological responses under combined drought and low-temperature stress. Utilizing the collected data, we established a rapid stress classification method based on photosynthetic physiological characteristics and constructed a photosynthesis prediction and early warning model. The results reveal that all tested tea cultivars exhibited a significant decline in leaf photosynthetic efficiency under combined stress, with the decreasing trend displaying a clear linear relationship with increasing stress intensity. Notably, ‘Tieguanyin’ demonstrated a significantly lesser decline in photosynthetic efficiency compared to other cultivars, suggesting its robust stress tolerance. In contrast, ‘Fuding Dabaicha’ showed the least stress tolerance. By selecting photosynthetic physiological parameters highly sensitive to combined stress and employing the K-means clustering algorithm, we developed a rapid stress classification method with an accuracy exceeding 80%. Various models were then used to predict and validate the response of photosynthetic physiological indicators to environmental stress, with the Random Forest (RF) model yielding the highest accuracy. This study provided a reference for the selection and breeding of tea cultivars under diverse climatic conditions. The stress classification method enables swift categorization of combined stress in tea plants, while the RF model facilitates non-destructive monitoring and early warning of photosynthetic physiology, offering significant practical value in tea production.

Key words: tea plant, drought and low-temperature stress, photosynthetic physiology characteristics, clustering algorithm, regression prediction