Abstract. The increasing availability of remotely sensed data have offered unprecedented possibilities for monitoring and analysis of environmental variables, including boosting recent studies in the field of ecosystem resilience relying on indicators derived from timeseries analysis, such as the temporal autocorrelation of vegetation indices. A forest ecosystem with decreased resilience will be more susceptible to external drivers and their change and could shift into an alternative system configuration by crossing a tipping point. Nevertheless, remote sensing data quantifying vegetation and forests properties inherently carry information related to the climate as well, which has to be accounted for before performing any modelling exercise. In this paper, we aim to present the general workflow and the challenges encountered in processing and analysing the historical, high-frequency and high-resolution timeseries of vegetation and climatic data. The final aim is training a machine learning model (Random Forest) in order to model and explore the performance and importance of a set of climatic and environmental metrics in predicting an indicator of the resilience of forests. In this case, the resilience of forests is quantified through the temporal autocorrelation (TAC) of the kernel NDVI (kNDVI). Climatic and environmental predictors include 2-meter air temperature, total precipitation, vapour pressure deficit, surface solar radiation, forest cover and soil organic carbon content. Results show a good performance of the Random Forest model and the ranking in the importance of the predicting variables captured in terms of background climate and climate variability. This application allows to separate and identify the main drivers of the temporal autocorrelation of kNDVI.