The aim of this research is to explore the analysis of methods allowing a synergetic use of information exchange between Earth Observation (EO) data and growth models in order to provide high spatial and temporal resolution actual evapotranspiration predictions. An assimilation method based on the Ensemble Kalman Filter algorithm allows for combining Sentinel-2 data with a new version of Simple Algorithm For Yield (SAFY_swb) that considers the effect of the water balance on yield and estimates the daily trend of evapotranspiration (ET). Our study is relevant in the context of demonstrating the effectiveness and necessity of satellite missions such as Land Surface Temperature Monitoring (LSTM), to provide high spatial and temporal resolution data for agriculture. The proposed method addresses the problem both from a spatial point of view, providing maps of the areas of interest of the main biophysical quantities of vegetation (LAI, biomass, yield and actual Evapotranspiration), and from a temporal point of view, providing a simulation on a daily basis of the aforementioned variables. The assimilation efficiency was initially evaluated with a synthetic, large and heterogeneous dataset, reaching values of 70% even for high measurement errors of the assimilated variable. Subsequently, the method was tested in a case study in central Italy, allowing estimates of the daily Actual Evapotranspiration with a relative RMSE of 18%. The novelty of this research is in proposing a solution that partially solves the main problems related to the synergistic use of EO data with crop growth models, such as the difficult calibration of initial parameters, the lack of frequent high-resolution data or the high computational cost of data assimilation methods. It opens the way to future developments, such as the use of simultaneous assimilation of multiple variables, to deeper investigations using more specific datasets and exploiting the advanced tools.

Novel approaches and algorithms to estimate crop physiological processes from Earth Observation (EO) data are essential to develop more sustainable management practices in agricultural systems. Within this context, this paper presents the results of different research activities carried out within the ESA-MOST Dragon 4 programme. The paper encompasses two research avenues: (a) the retrieval of biophysical variables of crops and yield prediction; and (b) food security related to different crop management strategies. Concerning the retrieval of variables, results show that LAI, derived by radiative transfer model (RTM) inversion, when assimilated into a crop growth model (i.e., SAFY) provides a way to assess yields with a higher accuracy with respect to open loop model runs: 1.14 t·ha⁻¹ vs 4.42 t·ha⁻¹ RMSE for assimilation and open loop, respectively. Concerning food security, results show that different pathogens could be detected by remote sensing satellite data. A k coefficient higher than 0.84 was achieved for yellow rust, thus assuring a monitoring accuracy, and for the diseased samples k was higher than 0.87. Concerning permanent crops, neural network (NN) algorithms allow classification of the Pseudomonas syringae pathogen on kiwi orchards with an overall accuracy higher than 91%.