Carbon Storage and Sequestration Potential Assessment in Southeastern Bangladesh: An Integrated PLUS-InVEST Modeling Approach

Abstract

Southeastern Bangladesh (SEB) is undergoing rapid land use and land cover (LULC) change that threatens its carbon storage and sequestration (CSS) potential, yet the region's LULC-carbon dynamics remain poorly understood. This study integrates the Patch-generating Land Use Simulation (PLUS) model, the InVEST carbon model, and the Geodetector method to assess CSS trajectories in SEB from 1990 to 2041. Six-class decadal LULC maps (1990–2020), derived from Landsat imagery, were used to calibrate the PLUS model and simulate three 2041 scenarios: Natural Evolution, Ecological Protection, and Urban Development. The model achieved a Kappa coefficient of 0.82 during validation. Built-up area expanded by over 80% between 1990 and 2020, driven mainly by cropland and vegetation loss, while wetland was the only natural land cover that expanded. The InVEST model estimated a net carbon loss of 0.48 Tg over three decades, driven mainly by cropland and vegetation loss, with wetland serving as the natural carbon sink. The Ecological Protection scenario can reverse this trend, projecting net carbon gain by 2041. Geodetector analysis identified elevation and slope as the strongest individual drivers of spatial carbon storage distribution. All factor pairs showed synergistic interaction effects, most notably among slope, elevation, and socioeconomic drivers. These findings offer an evidence-based foundation for effective land-use planning that can safeguard future carbon storage in one of the most ecologically sensitive regions of Bangladesh.