Day 4

Detailed paper information

Back to list

Paper title Assessing risk of forcibly displaced persons: Creation of a digital elevation model of the island of Bhasan Char from Sentinel-1 Stripmap products
  1. Andreas Braun University of Tuebingen Speaker
Form of presentation Poster
  • D1. Managing Risks
    • D1.01 Satellite EO for Geohazard Risks
Abstract text In 2018 the government of Bangladesh started planning the relocation of refugees of the Rohingya minority which were fleeing violent persecution in Myanmar. An island in the Bay of Bengal (Bhasan Char) was selected which is located around 60 kilometers from the mainland and was not inhabited before. On this island, the construction of 1.500 buildings is planned to host 100.000 persons.
Currently, the island has a size of around 40 square kilometers is considered a comparably recent landform which developed from silts washed from the Himalayans since around 2010. Human rights organizations strongly criticize the plans because they do not consider the island a safe place in case of tidal waves, monsoonal rains and sea-level rise. Currently, the island hosts around 13.000 people.
To assess the risk of the relocated persons, information on the topography is required. However, globally available digital elevation models, such as SRTM, AW3D30, or the Copernicus DEM do not contain usable data in this area because it was masked as sea surface.
In this study, the potential of synthetic aperture radar interferometry (InSAR) base on the Sentinel-1 mission to create a digital elevation model the island is evaluated. While the standard acquisition mode of Sentinel-1 is the interferometric wide swath (IW) mode, collecting images in with the Terrain Observation with Progressive Scans SAR (TOPSAR) technique at a spatial resolution of 5 x 20 meters, Stripmap (SM) products were available in this area at a spatial resolution of 2.2 x 3.5 meters. These allowed to calculate interferograms for the precise delineation of topographic variations. Different image pairs were used and analyzed according to their temporal and perpendicular baselines.
Because of the largely natural surfaces and wet conditions on the island, phase decorrelation led to partially unusable results. However, these could be mitigated by phase filtering and systematic masking to generate a DEM with sufficient quality. An independent accuracy assessment was undertaken based on height measurements from the ICESat-2 mission which covered the island with several tracks. A height accuracy of 75% was achieved before post-processing. Several post processing techniques are still under development and expected to increase the DEM quality to 90 %.
The digital elevation model can serve as an input to risk assessments related to tidal waves and sea level rise to test if the current adaptation measures (embankments, height of the buildings above ground) are substantially protecting the people living on Bhasan Char.