GIS-Based Evaluation of Land Subsidence and Land Use Change Using Spatial Interpolation Techniques in North Semarang

Abstract

Land subsidence has become one of the most critical environmental hazards affecting coastal cities in Indonesia, particularly North Semarang, where rapid urban expansion, industrial development, and intensive groundwater extraction have accelerated ground deformation and increased the vulnerability of urban infrastructure. Understanding the spatial relationship between land use change and land subsidence is therefore essential for supporting sustainable urban planning and disaster risk mitigation. This study aims to evaluate land subsidence patterns using spatial interpolation techniques and to investigate the influence of land use change on subsidence dynamics in North Semarang. The study integrates Geographic Information System (GIS)-based spatial analysis with multi-temporal geospatial datasets, including land subsidence monitoring data, SPOT-6 satellite imagery, and official land use maps covering the 2013–2023 period. Land use changes were identified through image interpretation, field verification, digitization, and overlay analysis using ArcGIS. Four interpolation methods—Inverse Distance Weighted (IDW), Natural Neighbor, Kriging, and Spline—were evaluated using the Root Mean Square Error (RMSE) statistic to determine the most reliable approach for modeling the spatial distribution of land subsidence. The results indicate substantial conversion of undeveloped land into industrial, warehousing, commercial, and transportation areas, particularly around Tanjung Mas Port, demonstrating rapid urbanization during the study period. Evaluation of the interpolation methods shows that the Kriging model achieved the lowest RMSE value (0.3843), outperforming IDW (0.4101), Natural Neighbor (0.4890), and Spline (0.5428), indicating the highest predictive accuracy for representing land subsidence patterns. The spatial analysis further reveals that areas experiencing the most intensive urban development coincide with zones of higher subsidence rates, which increased from 5.47 cm/year in 2013 to 6.03 cm/year in 2023, while land use conversion reached approximately 88.45 ha/year. These findings demonstrate a strong spatial association between urban expansion and land subsidence in North Semarang. This study demonstrates that GIS-based spatial interpolation combined with multi-temporal land use analysis provides an effective framework for identifying land subsidence hotspots and evaluating their relationship with urban development. The findings offer valuable scientific evidence for spatial planning, groundwater management, and coastal risk mitigation, while supporting the development of sustainable land-use policies in rapidly urbanizing coastal environments.Received: 2024-02-16 Revised: 2024-02-23  Accepted: 2026-08-25 Published: 2026-08-31