Prediction of the daily spatial variation of stem water potential in cherry orchards using weather and Sentinel-2 data

The common practice for irrigation management is to apply the water lost by evapotranspiration. However, we could manage the irrigation by monitoring the plant's water status by measuring the stem water potential (Ψs), which is currently costly and time-consuming. The primary goal of this work is to predict the daily spatial variation of Ψs using machine learning models. We measured Ψs in two orchards planted with sweet cherry tree variety Regina, and we monitored 30 trees weekly and biweekly in the central part of Chile, during two seasons, 2022–2023 and 2023–2024, and between October and April. To predict the Ψs, we used the random forest (RF), extreme gradient boosting (XGBoost), and support vector machine (SVM) models. We selected vapor pressure deficit (VPD), reference evapotranspiration (ET o ), relative humidity, and temperature as weather predictors. Also, we used as predictors spectral vegetation indices (VIs) and biophysical parameters derived from Sentinel-2. We compared two schemes, one for estimation and another for prediction. We discovered that XGboost and RF worked best for both. The estimation had an R 2 of 0.76 and an RMSE of 0.24 MPa. The prediction, on the other hand, had an R 2 of 0.59 and an RMSE of 0.36 MPa. The analysis of importance variables reveals that weather predictors, such as VPD, ET o , and temperature, have a higher weight in the model. These are followed by VIs that use short-wave infrared regions, which highlight the moisture stress index (MSI) and the disease and water stress index (DWSI). Future research should tackle the challenge of generalizing the approach, despite its methodological strengths, given that it was based on only two orchards. • Sentinel-2 and weather data allow us to model the daily spatial Ψs. • XGBoost and RF achieve higher performance for the estimation and prediction of Ψ s . • ET0, VPD, and T° are the best predictors, followed by SWIR spectral data. • For estimation, the model has R 2 of 0.76; for prediction, 0.59.