The state of Assam, primarily in the Brahmaputra basin, is one of the most flood-prone states of India, with devastating floods occurring every year. Rapid urbanization in the floodplains and inadequate water management have further exacerbated the human and infrastructural exposure to floods. Despite efforts in organizing relief camps and setting up safe houses, flood losses in terms of life and property have increased. This study presents a novel dataset and an approach that uses a combination of satellite and ground-based data to assess the population’s vulnerability to these floods. The Sentinel-1 SAR data and India Meteorological Department (IMD) rainfall data are used to determine areas in Assam that suffered from floods during July 2020. Additional datasets employed include the population density maps obtained from Facebook, road density maps from the Global Roads Inventory Project (GRIP), relief camp data from the state government, and infrastructure data from a new national dataset sourced from the Pradhan Mantri Gram Sadak Yojna (PMGSY). These geo-tagged datasets are utilized to develop a vulnerability map that would help the policymakers identify vulnerable regions and accordingly strategize infrastructure and disaster preparedness for floods.

Soil moisture is a vital climatic variable driving environmental and biological processes. Large scale soil moisture can be estimated using Land Surface Models or observed using active and passive microwave remote sensing. Increasing availability of remotely sensed soil moisture retrievals has allowed for constraining Land Surface Model (LSM) estimates. Though the accuracy of remote sensing datasets is constrained by soil roughness, vegetation, and surface temperature, combining them with LSMs allow us to reduce errors in soil moisture estimates. This study assimilates the SMOPs-ASCAT, ESA-CCI and SMAP soil moisture retrievals into a land surface model u sing an ensemble Kalman filter. The open-loop and data assimilated soil moisture outputs are evaluated against the ground-based sensor data. This demonstrates the establishment of an Indian Land Data Assimilation System (ILDAS) with the goal of providing accurate soil moisture products at high spatial and temporal resolution over India.