Joko Widodo
Research and Innovation Agency (BRIN), Jakarta 10340, Indonesia

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

Land Subsidence Prediction and Geo Environmental Predictor Importance Assessment in Jakarta, Semarang, and Pekalongan Using InSAR and Ensemble Machine Learning Luthfy Azh-Zhaviery Puary; Panca Dewi Pamungkasari; Endah Tri Esti Handayani; Rendi Handika; Babag Purbantoro; Joko Widodo; Damy Matheus Heezen
Green Intelligent Systems and Applications Volume 6 - Issue 2 - 2026
Publisher : Tecno Scientifica Publishing

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53623/gisa.v6i2.1233

Abstract

Land subsidence along the northern coast of Java (Pantura) poses a significant threat to coastal infrastructure and flood resilience in densely populated areas. This study integrates 12 geo-environmental variables with Persistent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) data to predict land subsidence in DKI Jakarta, Semarang City, and Pekalongan Regencyusing Random Forest (RF) and XGBoost regression models. Model performance was evaluated using mean absolute error (MAE), root mean square error (RMSE), and the coefficient of determination (R²) on an independent 20% holdout dataset. Across the three study areas, the optimized ensemble models demonstrated strong predictive performance. RF outperformed XGBoost in DKI Jakarta (image, image) and Semarang City (image, image), whereas both models showed comparable performance in Pekalongan Regency(image). Overall, RF consistently achieved equal or superior predictive accuracy, although the best performance in Pekalongan Regencywas obtained using RF without hyperparameter tuning. Feature importance analysis identified GEOLOGY_CODE as the dominant predictor in DKI Jakarta, while Land Surface Temperature (LST) was consistently important in Semarang City and Pekalongan Regency . Spatial analysis further identified Penjaringan, Genuk, and Kedungwuni as areas experiencing high levels of subsidence. The proposed framework provides a reproducible approach for district-scale land-subsidence assessment and supports risk-informed spatial planning across the Pantura coastal corridor.
Land Subsidence Analysis Using Machine Learning Algorithm Random Forest Method in DKI Jakarta Camelia Nur Hidayah; Panca Dewi Pamungkasari; Sari Ningsih; Muhammad Fauzan Azhiman; Joko Widodo; Elfady Satya Widayaka
Green Intelligent Systems and Applications Volume 5 - Issue 1 - 2025
Publisher : Tecno Scientifica Publishing

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53623/gisa.v5i1.606

Abstract

Land subsidence is an environmental phenomenon that causes the earth's surface to decline gradually or suddenly. Land subsidence occurred in DKI Jakarta due to various factors such as excessive groundwater exploitation, infrastructure loads, and geological conditions. The purpose of this study was to analyze land subsidence in DKI Jakarta and the distribution of existing land subsidence. The results were compared with previous findings using PS-InSAR. Land subsidence was predicted using the Random Forest algorithm. Random Forest, as a type of machine learning, was able to reduce noise and minimize the impact of overfitting through ensemble techniques. Researchers used four metrics, Mean Absolute Error (MAE), Root Mean Square Error (RMSE), R², and Kling-Gupta Efficiency (KGE), to assess the accuracy of the algorithm. The analysis results of land subsidence in DKI Jakarta using Random Forest aligned with the PS-InSAR method. It was observed that areas experiencing land subsidence were predominantly in North and West Jakarta compared to other regions. Furthermore, the prediction of land subsidence using the 2017–2021 dataset indicated a decrease of up to -60 mm/year.
Temporal Analysis of Land Subsidence in DKI Jakarta Using the Long Short-Term Memory (LSTM) Model Heni Fitriany; Panca Dewi Pamungkasari; Yunan Fauzi Wijaya Wijaya; Muhammad Fauzan Azhiman; Yasuhito Nagase Nagase; Joko Widodo
Green Intelligent Systems and Applications Volume 5 - Issue 2 - 2025
Publisher : Tecno Scientifica Publishing

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53623/gisa.v5i2.880

Abstract

This research investigated temporal patterns of land subsidence in DKI Jakarta by applying a Long Short-Term Memory (LSTM) model to deformation measurements derived from Persistent Scatterer Interferometric Synthetic Aperture Radar (PS-InSAR) observations acquired between 2017 and 2021. Because the original PS-InSAR time series was characterized by uneven acquisition intervals, the deformation records were first resampled into a uniform 11-day sequence to obtain a consistent temporal structure for modeling. Preprocessing steps, comprising outlier handling, temporal resampling, and feature normalization, were performed to ensure that the model could capture deformation behavior reliably. The LSTM architecture employed three stacked recurrent layers and was trained using the Adam optimizer with Smooth L1 Loss and an early-stopping strategy. Model evaluation demonstrated excellent agreement between predicted and observed deformation, yielding R² = 1.000, MSE = 0.104, RMSE = 0.322 mm, and KGE = 0.998. Compared with a previously developed Random Forest model (R² = 0.9995, RMSE = 0.3314 mm), the LSTM approach exhibited more stable temporal learning and was better suited for long-horizon deformation forecasting. Spatial analysis revealed that northern Jakarta, particularly Cengkareng, Tanjung Priok, and Pantai Indah Kapuk, continued to experience the greatest cumulative subsidence (> −30 mm), whereas areas in the south, such as Jagakarsa and Kebayoran Baru, showed minimal deformation (< −5 mm), aligning with known geological and anthropogenic conditions. Overall, integrating PS-InSAR time series with an LSTM framework provided a more coherent and temporally consistent method for characterizing subsidence behavior in Jakarta. The outcomes of this study offered a scientific basis for developing intelligent monitoring tools to support mitigation efforts and sustainable urban planning in regions affected by land subsidence.