Alam (2026) Modeling and performance evaluation of machine learning for historical and future land use and land cover dynamics in Chattogram District, Bangladesh

Identification

• Journal: Frontiers in Environmental Science
• Year: 2026
• Date: 2026-09-11
• Authors: M. J. Alam
• DOI: 10.3389/fenvs.2026.1923661

Research Groups

• Department of Sustainability and Social Justice, School of Climate, Environment and Society, Clark University, Worcester, MA, United States
• OpenStreetMap Contributors (2025)
• HydroSHEDS/HydroRIVERS
• NASA/USGS SRTM

Short Summary

This study integrates machine learning-based approaches and Markov Chain simulation to assess historical and future land use and land cover dynamics in Chattogram District, Bangladesh. The results reveal substantial cropland expansion (+32,375 ha) and continued growth of built-up land, accompanied by declines in vegetation, waterbodies, and bare land.

Objective

• Assess spatial and temporal LULC changes between 2006, 2015, and 2025.
• Examine the internal performance characteristics of RF- and SVM-based transition-potential models.
• Simulate the spatial distribution of LULC in 2035.

Study Configuration

• Spatial Scale: Chattogram District, located in southeastern Bangladesh within Chattogram Division (approximately 5,283 km²).
• Temporal Scale: Historical analysis from 2006 to 2025 and future projection up to 2035.

Methodology and Data

• Models used:
  • Random Forest (RF)
  • Support Vector Machine (SVM)
  • Markov Chain simulation
• Data sources:
  • Landsat-based classification
  • Google Earth Engine (GEE)-based LULC classification
  • TerrSet Land Change Modeler (LCM) transition analysis

Main Results

• The RF-classified LULC maps show substantial variation among major land-cover classes in Chattogram District across 2006, 2015, and 2025.
• Cropland expansion (+32,375 ha) and continued growth of built-up land are accompanied by declines in vegetation, waterbodies, and bare land.

Contributions

• This study provides empirical evidence to support sustainable land-use planning and environmental management in coastal-urban systems.
• The results highlight the importance of considering historical land-change patterns when projecting future LULC scenarios.

Funding

• This research was supported by Clark University's Department of Sustainability and Social Justice.

Citation

@article{Alam2026Modeling,
  author = {Alam, M. J.},
  title = {Modeling and performance evaluation of machine learning for historical and future land use and land cover dynamics in Chattogram District, Bangladesh},
  journal = {Frontiers in Environmental Science},
  year = {2026},
  doi = {10.3389/fenvs.2026.1923661},
  url = {https://doi.org/10.3389/fenvs.2026.1923661}
}