Monitoring and prediction of the LULC change dynamics using time series remote sensing data with Google Earth Engine

Urbanization is a significant global problem affecting many regions facing climate change. Land use land cover (LULC), land surface temperature (LST), urban sprawl, and significant concerns. The research objective was to evaluate the effects of LULC change, pre-urban expansion, and urban growth on LST for 30 years (and 2022) in District Lahore using Landsat (TM, ETM+, and OLI/TIRS) data in Google Earth Engine (GEE). In this study, we concentrate on four major LULC classes identified: urban area, vegetation, barren land, and water bodies through Landsat data and Support Vector Machine (SVM) in GEE. Results showed that 975.6 km 2 (196.5%) of the built-up area increased, while vegetation decreased by 579.15 km 2 (30.4%) from 1992 to 2002. Additionally, the normalized difference built-up index (NDBI) and the normalized difference vegetation index (NDVI) were retrieved to measure the association with LST. A negative and positive correlation was found between NDVI, NDBI, and LST, respectively. The urban heat island ratio index (UHIRI) was also mapped, and an upward trend was displayed during this research. These results are crucial for the division of development and planning to secure the enduring utilization of land resources for future urbanization growth.

This study comprehensively examined the dynamics of LULC changes and their impact on LST in Islamabad, Pakistan, from 1992 to 2022, utilizing the CA and Markov-Chain models. The research successfully mapped and predicted LULC and LST changes, highlighting the significant urban expansion and its correlation with rising temperatures over the past three decades. The study aimed to explore LULC changes and build the CA- Markov-Chain model to predict the distribution of LULC and its relationship with the LST. At the same time, the transition probability matrix was obtained through the Markov-Chain model. The RF algorithm was applied to urban areas classified using Landsat 5, 7, and 8 imageries, which achieved more than 80% accuracy. Moreover, future LST and LULC were predicted using the GEE and the integration of the CA-Markov-Chain al- gorithm. The built-up area in Islamabad increased substantially from 105.63 km2 (11.66%) in 1992 to 447.39 km2 (49.38%) in 2022. This urban growth will continue, reaching 531.82 km2 (55.38%) by 2042. The forest and cropland areas have seen significant reductions, with forest areas decreasing from 152.21 km2 (16.80%) in 1992 to 91.25 km2 (10.07%) in 2022. This trend is expected to continue, with further forest and barren land reductions by 2042. The study observed a rise in LST from 1992 to 2022, with urban areas showing the highest temperature increases. The LST is projected to continue increasing, with the highest temperatures (37–40 °C) expanding in coverage by 2042.

Based on the findings of this study, it is recommended that urban planners and policymakers in Islamabad implement strate- gies to mitigate the adverse impacts of rapid urbanization on LST and LULC. Emphasis should be placed on sustainable urban development practices, including preserving and expand- ing green spaces, effective land use planning, and adopting climate-resilient infrastructure. In addition, enhancing the mon- itoring of urban growth and environmental changes using high- resolution satellite data and integrating socio-economic factors into predictive models will provide more accurate and actionable insights. Collaborative efforts between government agencies, researchers, and local communities are essential to ensure sus- tainable urban expansion and minimize the environmental and health impacts of increased LST and altered LULC.