Assessing Climate Change Impacts on Coastal Dynamics along the Pakistan Coastline Using Remote Sensing and Machine Learning

Sea-level rise (SLR) driven by climate change is accelerating coastal erosion, which necessitates comprehensive assessments of shoreline dynamics to identify areas for targeted risk mitigation and adaptive planning. However, extracting coastlines from medium spatial resolution satellite data remains challenging due to limited large-scale annotated datasets, particularly for coastal risk assessment under future climate change. This study utilizes geospatial and machine learning based multispectral indices approach combining Random Forest (RF) classifier for shoreline extraction from Landsat data to quantify erosion and accretion along Pakistan1,365 km coastline. Shoreline assessment was quantified using linear regression rates (LRR), incorporating SLR and sea surface temperature (SST) trends. Future erosion projections were derived from CMIP6 General Circulation Models (GCMs) under different Shared Socioeconomic Pathways (SSPs) and based on historical erosion for 2020–2050. Results reveal pronounced erosion along Indus Delta, with shoreline retreat reaching − 50.4 ± 1.02 m associated with SLR of 0.015–0.15 m (2000–2020), with rises in SST (297–301 K). Sandspit coast showed accretion of 23.24 km², with shoreline advancement rates of t 48.45 ± 0.23 m, Gwadar Port experienced accretion rates up to 40 m, together with annual SST of 0.02–0.05°C. Under the high emission SSP5-8.5 scenario, persistent erosion of -23 ± 1.35 m is expected in the Indus Delta by 2050. Projections across different Global Warming Levels (GWL) ranging from 1.5 to 5.0°C suggest SLR may reach 0.23 m at 5°C of warming. Our findings underscore the importance of Integrated Coastal Zone Management (ICZM) and reinforce the IPCC’s call to limit global warming to safeguard coastal ecosystems and achieve SDGs 13 and 15 targets by 2030.