NH026-02
Impact Assessment of Major Flood Events on Agriculture and Infrastructure in Kerala, India from Remote Sensing Data

Friday, 11 December 2020: 16:04
Virtual
Barbara Hofmann1, Eleanor A Ainscoe1, Steven Reece2 and Quillon K Harpham1, (1)HR Wallingford, Wallingford, United Kingdom, (2)University of Oxford, Oxford, United Kingdom
Abstract:
Climate-related hazards in India cause large loss of life and damage to property, as well as critical infrastructure on an annual basis. Between 1995 and 2015 it is estimated that some 805 million people in India were affected by weather-related hazards. A recent report put India among the 10 most disaster-prone countries in the world and ranked floods as the climate-related hazard posing the greatest risk to people. Kerala state in particular endured devastating floods in 2018 and 2019. 2020 is expected to be another year with high monsoonal rainfall in Kerala .

The economic damage and loss assessment for buildings, infrastructure and agriculture can only be completed in detail after the initial response phase. However, a much earlier, initial estimate would be of benefit to discussions on fund mobilization and response strategies. Although several approaches and tools are already being used for post-disaster damage and economic loss assessment, their response time can be lengthy and their levels of detail and accuracy vary significantly.

As part of the Weather and Climate Science for Service Partnership India (WCSSP-India), Satellites for Impact and Vulnerability Assessment in India (SIVA) is researching and developing methods for using Earth Observation data to infer the short-term and long-term impact of severe flooding events in India on buildings, key infrastructure and crops.

SIVA uses optical and SAR data as well as freely available third party products in combination with machine learning algorithms with the aim of developing and testing methodologies to provide fast, reliable information about impacts of flooding to facilitate disaster response and recovery. The same methods can also be applied retrospectively to provide information about the impact of previous flood events for use in future disaster modelling.