Abstract The present study focuses on the Mahanadi River Delta (MRD) coastline, a highly populated coastal region in India where shoreline changes cause significant land loss and community relocations. A comprehensive approach to shoreline dynamics is undertaken by employing Synthetic Aperture Radar (SAR) and optical remote sensing (RS) imagery through the coupled modified normalized difference water index–tasseled cap transformation method for land cover and shoreline extraction. Additionally, future composite risk maps are generated by integrating projected population and household data with erosion risk, along with an evaluation of current adaptation measures. The 230.5 km MRD shoreline was segmented into five littoral cells (LC-A to LC-E), with 23,239 transacts using the Digital Shoreline Analysis System (DSAS). Landsat satellite images from 1990 to 2010, Sentinel-2 optical data (2015), and Sentinel-1 SAR data (2020) are used, together with linear regression rate (LRR) and end point rate (EPR) techniques, to estimate the rate of shoreline shifting. It is observed that Bhitarakanika in LC-E and Saralikud in LC-D are suffering maximum erosion and accretion, respectively. Validation using root-mean square error mapping demonstrates that EPR is a more reliable forecast method. Short-term shoreline change analysis revealed distinct patterns of erosion and accretion across the delta. In contrast, LULC changes between 1990 and 2020 – characterized by urban expansion, vegetation and sand loss and increased water spread – were found to significantly influence shoreline stability. The study predicts that around 65% of the shoreline will be threatened with erosion by 2050 and 37% of coastal mouzas will be very vulnerable to erosion. A composite risk map for the future is also prepared, using erosion risk levels, predicted household numbers, and population data sets. Field visits were undertaken to assess significant erosion and accretion zones and ascertain their causes. The existing adaptation measures along the MRD coastlines were evaluated based on the Net Shoreline Movement values and prescriptive measures for crucial MRD areas were suggested. Research highlights <UnorderedList Mark="Bullet"> <ItemContent> This study uses combined SAR and optical imagery for cloud-resilient, high-resolution shoreline monitoring. </ItemContent> <ItemContent> Comprehensive temporal assessment of both short-term and long-term shoreline dynamics capturing both episodic and decadal variability. </ItemContent> <ItemContent> Developed a composite erosion risk map by integrating shoreline change data with projected population density, female population density and household distribution. </ItemContent> <ItemContent> Carried out the erosion hazard mapping at the Mouza (village) level, enabling location-specific adaptation strategies. </ItemContent> </UnorderedList>

Spatio-temporal shoreline change and future composite erosion risk mapping using SAR and optical imagery: a case study of the Mahanadi River Delta coastline

期刊论文
发表日期：

2026年3月26日

查看全文

Debabrata Mishra,
Ratnakar Swain,
Pratyush Kumar Dora

摘要

Abstract

The present study focuses on the Mahanadi River Delta (MRD) coastline, a highly populated coastal region in India where shoreline changes cause significant land loss and community relocations. A comprehensive approach to shoreline dynamics is undertaken by employing Synthetic Aperture Radar (SAR) and optical remote sensing (RS) imagery through the coupled modified normalized difference water index–tasseled cap transformation method for land cover and shoreline extraction. Additionally, future composite risk maps are generated by integrating projected population and household data with erosion risk, along with an evaluation of current adaptation measures. The 230.5 km MRD shoreline was segmented into five littoral cells (LC-A to LC-E), with 23,239 transacts using the Digital Shoreline Analysis System (DSAS). Landsat satellite images from 1990 to 2010, Sentinel-2 optical data (2015), and Sentinel-1 SAR data (2020) are used, together with linear regression rate (LRR) and end point rate (EPR) techniques, to estimate the rate of shoreline shifting. It is observed that Bhitarakanika in LC-E and Saralikud in LC-D are suffering maximum erosion and accretion, respectively. Validation using root-mean square error mapping demonstrates that EPR is a more reliable forecast method. Short-term shoreline change analysis revealed distinct patterns of erosion and accretion across the delta. In contrast, LULC changes between 1990 and 2020 – characterized by urban expansion, vegetation and sand loss and increased water spread – were found to significantly influence shoreline stability. The study predicts that around 65% of the shoreline will be threatened with erosion by 2050 and 37% of coastal mouzas will be very vulnerable to erosion. A composite risk map for the future is also prepared, using erosion risk levels, predicted household numbers, and population data sets. Field visits were undertaken to assess significant erosion and accretion zones and ascertain their causes. The existing adaptation measures along the MRD coastlines were evaluated based on the Net Shoreline Movement values and prescriptive measures for crucial MRD areas were suggested.

Research highlights

This study uses combined SAR and optical imagery for cloud-resilient, high-resolution shoreline monitoring.

Comprehensive temporal assessment of both short-term and long-term shoreline dynamics capturing both episodic and decadal variability.

Developed a composite erosion risk map by integrating shoreline change data with projected population density, female population density and household distribution.

Carried out the erosion hazard mapping at the Mouza (village) level, enabling location-specific adaptation strategies.