Paddy rice is a staple food in South Asia. Climate change-induced heavy rainfall has increased the risk of loss and damage to rice crops, increasing production uncertainty. However, due to the unavailability of timely and accurate loss and damage data, governments, development partners, and insurance companies cannot respond properly and farmers may not receive appropriate compensation for their losses. To fulfill the long-standing problem of the data gap, this study aims to develop a semi-automated processing chain (toolbox) for the estimation of rice-crop loss and damage using time-series satellite-based imageries coupled with bioclimatic datasets using machine learning models. The flood intensity, rice phenology, bio-climatic data, along with historical productivity will be used to estimate the loss and damage at plot level. The study will be based in the ex-post case of the Ganga-river basin of Bangladesh, India, and Nepal in the recent past.
Accurate data on losses and damages can assist government officials, insurance companies, and banks in making informed decisions and taking appropriate action in response to adverse events. Therefore, we will sensitize stakeholders by disseminating scientifically accepted methods, results, case studies, research publications, policy briefs, and blog posts etc. to encourage adoption. This can help to build better preparedness through financial inclusion and enhance the resilience of farmers, promoting sustainable agriculture.