Cover crops are grown in order to provide agro-ecological services and must be terminated before planting the subsequent cash crop. Winterkill termination (by frost damage) depends on the interaction between crop frost hardiness, temperatures and the development stage reached at the time of sub-zero temperature exposure. Remotely sensing intensity, timing and spatial variation of cover crop frost damage can be useful for modeling and planning purposes. Therefore, in this study Sentinel-2 vegetation indices were employed in order to detect frost damage in four white mustard (Sinapis alba L.) fields located in Northern Italy. We estimated the starting date of frost events by means of vegetation indices (EVI, NDRE, NDVI, MMSR, and CCCI); we quantified and mapped frost damage at the sub-field level, using ground-based frost damage measurements carried out during the 2021/2022 season. As to frost damage quantification, MMSR outperformed the other VIs followed by CCCI and EVI (R2 > 0.55). The adopted procedure to detect starting dates of frost events was successful in most cases, with a one-day and a four-day delay in the two best cases (NDRE). Finally, maps of frost damage were consistent with its observed spatial variation. We demonstrated that it is possible to employ vegetation indices in order to detect cover crop frost damage and thus assessing cover crop winterkill termination efficiency in the field. Further research is needed, involving additional field monitoring of white mustard in more diverse conditions, and extension of the calibration, as well as validation.

No abstract

A scoping review of the relevant literature was carried out to identify the existing N recommendation systems, their temporal and geographical diffusion, and knowledge gaps. In total, 151 studies were identified and categorized. Seventy-six percent of N recommendation systems are empirical and based on spatialized vegetation indices (73% of them); 21% are based on mechanistic crop simulation models with limited use of spatialized data (26% of them); 3% are based on machine learning techniques with integration of spatialized and non-spatialized data. Recommendation systems started to appear worldwide in 2000; often they were applied in the same location where calibration had been carried out. Thirty percent of the studies use advanced recommendation techniques, such as sensor/approach fusion (44%), algorithm add-ons (30%), estimation of environmental benefits (13%), and multi-objective decisions (13%). Some limitations have been identified. Empirical systems need specific calibrations for each site, species and sensor, rarely using soil, vegetation and weather data together, while mechanistic systems need large input data sets, often non-spatialized. We conclude that N recommendation systems can be improved by better data and the integration of algorithms.