Agriculture is exposed to a wide spectrum of risks and is influenced by Land-atmosphere interactions. In spite of their importance, the representation or modelling of these interactions in the existing weather and climate models are constrained because of the involvement of a complex set of process that is difficult to be monitored. Agro-hydrological parameters are the variables that vary with land-atmosphere interactions yet will retain the memory of the anomalies happened in past. Agro- hydrological (A-H) parameters (such as soil moisture, evapotranspiration, and NDVI) due to its inherent memory can hence be used in modelling L-AI to predict the risk associated in agriculture production.

Dynamic forecasting of agriculture yield is possible by ensembling A-H variables with a machine learning algorithm. With the emergence of satellite sensors, A-H parameters can be monitored with much ease. Agriculture risk and related issues should be tackled from a policy-making and governance angle. Agriculture risk estimation may help decision-makers and stakeholders to take appropriate steps or formulate policies to eradicate or reduce the impacts of agricultural risk and thus safeguarding the life and interests of the farmers.

Satyukt Analytics Pvt Ltd has developed an innovative algorithm that merges data from different satellites to provide agro-hydrological parameters with high spatial and temporal resolution under all weather conditions. These variables will help in monitoring the farm without going individually to all the farms. The analysis will provide insight towards farming practices underlining the actions required immediately.