Integrating Machine Learning and Remote Sensing Using Google Earth Engine to Detect Land Use Land Cover Change in Pakse City, Lao PDR, During 2004-2024

Over the last two decades, rapid urban expansion has reshaped land use and land cover (LULC) patterns in many developing countries, especially in secondary cities where planning control remains limited. The objective of this study is to integrate machine learning and remote sensing using Google Earth Engine (GEE) to detect LULC change in Pakse City, Champasak province, Lao PDR, during 2004-2024. Landsat images and the Random Forest algorithm, a type of machine learning, were employed within the GEE platform for LULC classification. The Markov chain modelling in the Terrset software was used to assess the spatiotemporal dynamics of LULC in Pakse City over the same period. The results include classified LULC maps for the years 2004, 2014, and 2024. LULC was categorized into five types: water bodies, built-up areas, forest areas, agricultural areas, and miscellaneous areas. The classification accuracy assessment showed high reliability, with overall accuracy rates greater than 97% and Kappa coefficients greater than 0.94. The findings also demonstrated a significant change in LULC during the study period, with built-up area increasing significantly from 4.19 km² (2004) to 32.73 km² (2024), or 681.64%, while forest and agricultural areas decreased by 9.46% and 12.75%, respectively. In terms of Markov transition analysis revealed that the main sources of urban expansion were forest and agricultural areas, while miscellaneous areas were the highest transition dynamics. These trends are in line with increasing pressure related to urban expansion and infrastructure development. Thus, the findings can assist urban planners and related parties in Pakse City and similar secondary cities in managing future land expansion and protecting agricultural, forest areas, and the environment.

Authors

Publication Details

Journal
Scientific Journal of Savannakhet University-ວາລະສານວິທະຍາສາດ ມະຫາວິທະຍາໄລສະຫວັນນະເຂດ
Published
2026-11-21
DOI
https://doi.org/10.71026/sjsku.v5i1.14.2026
Primary Topic
Remote Sensing and Land Use
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Integrating Machine Learning and Remote Sensing Using Google Earth Engine to Detect Land Use Land Cover Change in Pakse City, Lao PDR, During 2004-2024

Khandala Khamphila, Sengsouksanh Phanthouvong, Keomany Khamphoumy, Phetamkha Theppanya et al.
Scientific Journal of Savannakhet University-ວາລະສານວິທະຍາສາດ ມະຫາວິທະຍາໄລສະຫວັນນະເຂດ
Remote Sensing and Land Use
article

Integrating Machine Learning and Remote Sensing Using Google Earth Engine to Detect Land Use Land Cover Change in Pakse City, Lao PDR, During 2004-2024

Khandala Khamphila, Sengsouksanh Phanthouvong, Keomany Khamphoumy, Phetamkha Theppanya, Thanouxay Vernavong
article en

Abstract

Over the last two decades, rapid urban expansion has reshaped land use and land cover (LULC) patterns in many developing countries, especially in secondary cities where planning control remains limited. The objective of this study is to integrate machine learning and remote sensing using Google Earth Engine (GEE) to detect LULC change in Pakse City, Champasak province, Lao PDR, during 2004-2024. Landsat images and the Random Forest algorithm, a type of machine learning, were employed within the GEE platform for LULC classification. The Markov chain modelling in the Terrset software was used to assess the spatiotemporal dynamics of LULC in Pakse City over the same period. The results include classified LULC maps for the years 2004, 2014, and 2024. LULC was categorized into five types: water bodies, built-up areas, forest areas, agricultural areas, and miscellaneous areas. The classification accuracy assessment showed high reliability, with overall accuracy rates greater than 97% and Kappa coefficients greater than 0.94. The findings also demonstrated a significant change in LULC during the study period, with built-up area increasing significantly from 4.19 km² (2004) to 32.73 km² (2024), or 681.64%, while forest and agricultural areas decreased by 9.46% and 12.75%, respectively. In terms of Markov transition analysis revealed that the main sources of urban expansion were forest and agricultural areas, while miscellaneous areas were the highest transition dynamics. These trends are in line with increasing pressure related to urban expansion and infrastructure development. Thus, the findings can assist urban planners and related parties in Pakse City and similar secondary cities in managing future land expansion and protecting agricultural, forest areas, and the environment.

Scientific Journal of Savannakhet University-ວາລະສານວິທະຍາສາດ ມະຫາວິທະຍາໄລສະຫວັນນະເຂດVol. 5(1)
Remote Sensing and Land Use
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon
Integrating Machine Learning and Remote Sensing Using Google Earth Engine to Detect Land Use Land Cover Change in Pakse City, Lao PDR, During 2004-2024 — Khandala Khamphila, Sengsouksanh Phanthouvong, et al. · Scientific Journal of Savannakhet University-ວາລະສານວິທະຍາສາດ ມະຫາວິທະຍາໄລສະຫວັນນະເຂດ (2026) | TGRS Research Map | TGRS