Spatial-Distribution-Bogor
Major cities in Indonesia, particularly on the island of Java, are facing rapid population growth pressure, with the national population estimated to reach 281.6 million in 2024. Java, which accounts for only 7% of Indonesia's total land area, accommodates approximately 56% of the national population[1]. One of the cities experiencing significant growth is Bogor City, with a projected population of 1.5 million in 2024[2]. Its strategic location has made Bogor City a primary destination for urbanization in West Java, especially during the 2011–2019 period. Consequently, rapid development has driven the expansion of settlements and infrastructure, reducing agricultural land, and decreasing vegetation cover as built-up areas expand[3].
Land use changes, particularly the reduction in vegetation, lead to a decline in Normalized Difference Vegetation Index (NDVI) values. NDVI itself is an indicator used to observe the greenness level or vegetation density in a given area[4]. When vegetation decreases, Relative Humidity (RH) tends to drop, while Land Surface Temperature (LST) increases. This condition causes an increase in the Temperature Humidity Index (THI) value, an indicator that represents the level of temperature and humidity comfort in a region[5].
In the long term, elevated LST and THI can cause serious impacts, such as heat stress, health issues, reduced air quality, and a decline in the quality of life for urban communities[6]. THI is an essential tool for measuring the level of environmental comfort while helping to identify conditions that can affect human physical and mental well-being[7]. If the temperature and humidity remain within an ideal range, the human body can achieve thermal balance, allowing various daily activities to be carried out more comfortably, efficiently, and productively[8].
A remote sensing approach based on Google Earth Engine (GEE) is utilized to analyze satellite imagery rapidly, accurately, and extensively. GEE is widely used in environmental monitoring studies, particularly concerning land use dynamics, LST, and THI. One study noted an increase in built-up land by 87.49 ha over 10 years, accompanied by an LST rise of up to 5°C and a THI increase of 3°C[9]. Built-up land recorded the highest average temperature (27.18°C), whereas vegetated areas had the lowest. Vegetated areas effectively lower temperatures through evaporation and transpiration, while the conversion of green spaces triggers the risk of the Urban Heat Island (UHI) effect[10].
The Random Forest (RF) classification method, based on satellite imagery on the GEE platform, is used to detect land use changes accurately. Random Forest is a machine learning algorithm that operates by constructing multiple decision trees and merging their results to improve classification accuracy. Based on previous research, Random Forest has proven its superiority in accuracy, reaching 89.53%, which is significantly higher than the Support Vector Machine (SVM) method, which only achieved 74.07% when applied to Landsat 8 imagery[11].
Based on the aforementioned issues, an Analysis of the Spatial Distribution Patterns of Land Use, Land Surface Temperature, and Temperature Humidity Index in Bogor City for the years 2004, 2014, and 2024 is required. The year 2024 was selected as the primary focus of the analysis because it reflects the current conditions following a decade of rapid urbanization. By comparing data from 2004, 2014, and 2024, this research is expected to uncover the long-term dynamic changes occurring in Bogor City. The results of this analysis are anticipated to provide in-depth insights into the spatial distribution of land changes, LST elevation, and THI, which can further serve as a foundation for the Bogor City government in developing mitigation strategies and urban planning that are more responsive to the impacts of climate change.
