Assimilation of Sentinel-2 Biophysical Variables into a Digital Twin for the Automated Irrigation Scheduling of a Vineyard

Decision support systems (DSS) are needed to carry out precision irrigation. Key issues in this regard include how deal with spatial variability and the adoption of deficit irrigation strategies at field scale. A software application originally designed for water balance-based automated scheduling locally fine-tuned through use sensors has been further developed emerging paradigm both digital twins Internet Things (IoT). The aim research is demonstrate feasibility automatically a commercial vineyard when adopting regulated (RDI) assimilating near real time fraction absorbed photosynthetically active radiation (fAPAR) obtained from Sentinel-2 imagery. In addition, simulations crop evapotranspiration by twin were compared remote sensing estimates using surface energy balance models Copernicus-based inputs. Results showed that regression between instantaneous fAPAR situ measurements intercepted (fIPAR) had coefficient determination (R2) ranging 0.61 0.91, root mean square deviation (RMSD) 0.10. conversion daily step was dependent on row orientation. site-specific successfully adopted an adaptive response allowed spontaneous adjustments order stress vines certain level specific growing stages. Simulations soil components performed well. sensing-estimated actual (two-source Priestley–Taylor modeling approach, TSEB-PTS2+S3) potential (Penman–Monteith approach) RMSD values 0.98 mm/day 1.14 mm/day, respectively.