Fuan Tsai
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Fuan Tsai has a total of 66 co-authors. Here are the top co-authors by number of shared publications:
- Walter Chen
National Taipei University of Technology, Taiwan
5 shared publications - Huan Chang
Department of Civil Engineering, National Central University, Taoyuan City 32001, Taiwan
4 shared publications - Jhe-Syuan Lai
Department of Civil Engineering, Feng Chia University, Taichung 40724, Taiwan
4 shared publications - Tang-Huang Lin
Center for Space and Remote Sensing Research, National Central University, No. 300, Jhongda Rd., Jhongli Dist, Taoyuan City 32001, Taiwan
4 shared publications Chi-Chuan Lo
2 shared publicationsH. Chang
Department of Civil Engineering
2 shared publications
Latest Publications
- Open Access
- 1 August 2026
- Published by MDPI AG in Sustainability
- Vol. 18 (15) , 7779
A Pilot Study of SHAP-Interpreted Machine Learning for Pixel-Level Landslide Classification from High-Resolution DEM and Satellite Imagery
Accurate delineation of current landslide extent is important for hazard assessment, sustainable watershed management, and disaster risk reduction in tectonically active mountainous regions. This study presents a pilot machine learning framework for pixel-level landslide classification in the Laonung (Laonong) Creek Watershed, southern Taiwan, using very high-resolution digital elevation model (DEM) derivatives and SPOT-6 multispectral imagery. Thirteen geomorphometric and spectral features, including slope, curvature, and six spectral indices derived from SPOT-6 bands, were extracted from 96 landslide-containing tiles within a pilot subregion of the watershed; no landslide-free tiles were included in model training or evaluation. Landslide annotations followed a geomorphic-unit delineation protocol in which optical imagery provided the primary evidence of current activity and DEM-derived hillshade supported boundary refinement. Three classifiers were evaluated using column-quartile spatially blocked four-fold cross-validation, with each fold comprising a geographically contiguous range of columns, to reduce spatial leakage: logistic regression (LR), random forest (RF), and XGBoost. All three models substantially outperformed the no-skill baseline for the resampled evaluation dataset (average precision, AP ), achieving mean AP values of , , and for LR, RF, and XGBoost, respectively. The convergence of linear and nonlinear model performance suggests that the dominant discriminatory signal is largely captured by relatively simple spectral and topographic predictors within this pilot dataset, rather than reflecting a general property of landslide classification. SHapley Additive exPlanations (SHAP) analysis across all four spatial folds identified SPOT-6 Band 3 (Red) as the dominant predictor in every fold, with NDVI a robust secondary predictor, consistent with the spectral characteristics of fresh bare-soil landslide surfaces and with the optical cues used in the annotation protocol. The results are interpreted in the context of the pilot dataset’s limited spatial extent, the resampled class distribution used for model evaluation, and unquantified label uncertainty. This study provides a transferable methodological baseline for future, larger-scale landslide classification analysis in the Laonung Creek Watershed and highlights the potential contribution of spatially explicit landslide mapping to sustainability-oriented disaster management.- 1 August 2026
- Published by MDPI AG in Sustainability
- Vol. 18 (15) , 7779
Open AccessA Pilot Study of SHAP-Interpreted Machine Learning for Pixel-Level Landslide Classification from High-Resolution DEM and Satellite Imagery
Accurate delineation of current landslide extent is important for hazard assessment, sustainable watershed management, and disaster risk reduction in tectonically active mountainous regions. This study presents a pilot machine learning framework for pixel-level landslide classification in the Laonung (Laonong) Creek Watershed, southern Taiwan, using very high-resolution digital elevation model (DEM) derivatives and SPOT-6 multispectral imagery. Thirteen geomorphometric and spectral features, including slope, curvature, and six spectral indices derived from SPOT-6 bands, were extracted from 96 landslide-containing tiles within a pilot subregion of the watershed; no landslide-free tiles were included in model training or evaluation. Landslide annotations followed a geomorphic-unit delineation protocol in which optical imagery provided the primary evidence of current activity and DEM-derived hillshade supported boundary refinement. Three classifiers were evaluated using column-quartile spatially blocked four-fold cross-validation, with each fold comprising a geographically contiguous range of columns, to reduce spatial leakage: logistic regression (LR), random forest (RF), and XGBoost. All three models substantially outperformed the no-skill baseline for the resampled evaluation dataset (average precision, AP ), achieving mean AP values of , , and for LR, RF, and XGBoost, respectively. The convergence of linear and nonlinear model performance suggests that the dominant discriminatory signal is largely captured by relatively simple spectral and topographic predictors within this pilot dataset, rather than reflecting a general property of landslide classification. SHapley Additive exPlanations (SHAP) analysis across all four spatial folds identified SPOT-6 Band 3 (Red) as the dominant predictor in every fold, with NDVI a robust secondary predictor, consistent with the spectral characteristics of fresh bare-soil landslide surfaces and with the optical cues used in the annotation protocol. The results are interpreted in the context of the pilot dataset’s limited spatial extent, the resampled class distribution used for model evaluation, and unquantified label uncertainty. This study provides a transferable methodological baseline for future, larger-scale landslide classification analysis in the Laonung Creek Watershed and highlights the potential contribution of spatially explicit landslide mapping to sustainability-oriented disaster management. - Open Access
- 1 July 2026
- Published by MDPI AG in Sustainability
- Vol. 18 (13) , 6649
Input Modality Ablation for Sustainable Landslide Hazard Management Using U-Net: Fused DEM–Optical vs. Spectral vs. Terrain Representations in a Small-Sample Pilot Study
Rapid and accurate landslide mapping is essential for disaster risk reduction and sustainable land management in landslide-prone mountainous regions. This study presents a U-Net semantic segmentation framework for pixel-wise landslide classification in the Laonung Creek Watershed of southern Taiwan using 96 annotated tiles derived from a very high-resolution DEM and SPOT-6 multispectral imagery. An input modality ablation experiment compares four configurations: a fused DEM–optical composite matching the visual input used by the annotators (annotation-coherent input), SPOT-6 natural color imagery, a DEM-derived terrain stack, and a six-channel multi-source stack combining all SPOT-6 bands with slope and curvature. All configurations use an identical EfficientNet-B0 U-Net architecture under a spatially blocked train/validation/test design with a fixed held-out test set of 29 tiles. The multi-source stack achieves the highest test Average Precision (AP) of 0.556 (95% CI: 0.463–0.643), whereas the annotation-coherent fused composite achieves AP = 0.511 (95% CI: 0.404–0.601); overlapping confidence intervals indicate that neither modality is definitively superior at this test-set size. The terrain-only configuration (AP = 0.152) confirms that optical information is essential for reliable delineation. A key methodological finding is that differential encoder–decoder learning rates caused rapid decoder overfitting; matched rates of substantially stabilized training and are recommended as a conservative default for small-sample segmentation with pretrained encoders. At matched pixel positions, the best DL model achieves AP comparable to a companion Random Forest (DL: 0.847, RF: 0.824), while producing spatially coherent probability maps that support scalable landslide inventory compilation for sustainable hazard management.- 1 July 2026
- Published by MDPI AG in Sustainability
- Vol. 18 (13) , 6649
Open AccessInput Modality Ablation for Sustainable Landslide Hazard Management Using U-Net: Fused DEM–Optical vs. Spectral vs. Terrain Representations in a Small-Sample Pilot Study
Rapid and accurate landslide mapping is essential for disaster risk reduction and sustainable land management in landslide-prone mountainous regions. This study presents a U-Net semantic segmentation framework for pixel-wise landslide classification in the Laonung Creek Watershed of southern Taiwan using 96 annotated tiles derived from a very high-resolution DEM and SPOT-6 multispectral imagery. An input modality ablation experiment compares four configurations: a fused DEM–optical composite matching the visual input used by the annotators (annotation-coherent input), SPOT-6 natural color imagery, a DEM-derived terrain stack, and a six-channel multi-source stack combining all SPOT-6 bands with slope and curvature. All configurations use an identical EfficientNet-B0 U-Net architecture under a spatially blocked train/validation/test design with a fixed held-out test set of 29 tiles. The multi-source stack achieves the highest test Average Precision (AP) of 0.556 (95% CI: 0.463–0.643), whereas the annotation-coherent fused composite achieves AP = 0.511 (95% CI: 0.404–0.601); overlapping confidence intervals indicate that neither modality is definitively superior at this test-set size. The terrain-only configuration (AP = 0.152) confirms that optical information is essential for reliable delineation. A key methodological finding is that differential encoder–decoder learning rates caused rapid decoder overfitting; matched rates of substantially stabilized training and are recommended as a conservative default for small-sample segmentation with pretrained encoders. At matched pixel positions, the best DL model achieves AP comparable to a companion Random Forest (DL: 0.847, RF: 0.824), while producing spatially coherent probability maps that support scalable landslide inventory compilation for sustainable hazard management. - Open Access
- 18 April 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (8) , 1228
Modeling Long-Term LULC Changes and Future Urban Growth: A Case Study of Ulaanbaatar Using CA-Based Machine Learning
Accelerated urbanization in Ulaanbaatar, Mongolia, has driven substantial changes in Land Use and Land Cover (LULC), threatening sustainable urban ecosystems. This study investigates historical LULC dynamics (2000–2021) and simulates future expansion scenarios through 2050 using a hybrid Machine Learning (ML) and Cellular Automata-Artificial Neural Network (CA-ANN) approach. Multi-temporal classification was performed using Support Vector Machine (SVM) and Random Forest (RF) algorithms. Both classifiers demonstrated high and comparable accuracy; SVM achieved an average Kappa coefficient of 0.8939 while RF achieved 0.8917, a marginal difference that should be interpreted with caution. Change detection analysis revealed a continuous expansion of built-up areas at the expense of dense forest and grassland, a trend driven largely by accessibility factors. Future projections indicate that even as the rate of urbanization may slow, encroachment on green spaces will persist without policy intervention. This research presents a replicable methodological workflow for monitoring urban sprawl and provides evidence to inform sustainable land management and reforestation strategies in rapidly developing urban regions.
- 18 April 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (8) , 1228
Open AccessModeling Long-Term LULC Changes and Future Urban Growth: A Case Study of Ulaanbaatar Using CA-Based Machine Learning
Accelerated urbanization in Ulaanbaatar, Mongolia, has driven substantial changes in Land Use and Land Cover (LULC), threatening sustainable urban ecosystems. This study investigates historical LULC dynamics (2000–2021) and simulates future expansion scenarios through 2050 using a hybrid Machine Learning (ML) and Cellular Automata-Artificial Neural Network (CA-ANN) approach. Multi-temporal classification was performed using Support Vector Machine (SVM) and Random Forest (RF) algorithms. Both classifiers demonstrated high and comparable accuracy; SVM achieved an average Kappa coefficient of 0.8939 while RF achieved 0.8917, a marginal difference that should be interpreted with caution. Change detection analysis revealed a continuous expansion of built-up areas at the expense of dense forest and grassland, a trend driven largely by accessibility factors. Future projections indicate that even as the rate of urbanization may slow, encroachment on green spaces will persist without policy intervention. This research presents a replicable methodological workflow for monitoring urban sprawl and provides evidence to inform sustainable land management and reforestation strategies in rapidly developing urban regions.
- Open Access
- 15 April 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (8) , 1181
Application of PSInSAR Monitoring for Large-Scale Landslide with Persistent Scatterers from Deep Learning Classification
The Persistent Scatterers InSAR (PSInSAR) technology, which utilizes pixels with stable phases to extract ground deformation, is an effective tool for large-scale, long-period surface monitoring applications. It has been widely applied to land subsidence monitoring, earthquake research, and infrastructure risk management. Furthermore, some studies have successfully employed this method to monitor the progressive motion of creeping in landslide areas. However, these regions containing active landslides are usually covered by canopy layers, which cause low coherence in InSAR processing and reduce the number of stable pixels, thereby preventing long-term period monitoring in those areas. In this study, the supervised deep learning model, U-Net, based on a convolutional neural network, is applied to the differential InSAR dataset acquired from Sentinel-1 to improve persistent scatterer selection. A well-processed PSInSAR result, utilizing 55 Sentinel-1 images acquired from 5 November 2014 to 19 December 2017, is introduced as a dataset for model training. The pixel-based Persistent Scatterer (PS) labels used for model training are identified using the StaMPS software. The model is designed to identify the distributed scatterer (iDS) index using a single pair of SAR images. As a result, more iDS pixels can be obtained from a single interferogram, indicating a significant improvement over the StaMPS algorithm. The line-of-sight velocity and time series of PS pixels from the model prediction show a long-term uplift on the upper slope, which represents downslope sliding in the target area. Furthermore, some iDS pixels exhibit a seasonal deformation on the lower part of the slope. The capability for these additional deformation analyses underscores the potential of this new deep-learning-based approach.
- 15 April 2026
- Published by MDPI AG in Remote Sensing
- Vol. 18 (8) , 1181
Open AccessApplication of PSInSAR Monitoring for Large-Scale Landslide with Persistent Scatterers from Deep Learning Classification
The Persistent Scatterers InSAR (PSInSAR) technology, which utilizes pixels with stable phases to extract ground deformation, is an effective tool for large-scale, long-period surface monitoring applications. It has been widely applied to land subsidence monitoring, earthquake research, and infrastructure risk management. Furthermore, some studies have successfully employed this method to monitor the progressive motion of creeping in landslide areas. However, these regions containing active landslides are usually covered by canopy layers, which cause low coherence in InSAR processing and reduce the number of stable pixels, thereby preventing long-term period monitoring in those areas. In this study, the supervised deep learning model, U-Net, based on a convolutional neural network, is applied to the differential InSAR dataset acquired from Sentinel-1 to improve persistent scatterer selection. A well-processed PSInSAR result, utilizing 55 Sentinel-1 images acquired from 5 November 2014 to 19 December 2017, is introduced as a dataset for model training. The pixel-based Persistent Scatterer (PS) labels used for model training are identified using the StaMPS software. The model is designed to identify the distributed scatterer (iDS) index using a single pair of SAR images. As a result, more iDS pixels can be obtained from a single interferogram, indicating a significant improvement over the StaMPS algorithm. The line-of-sight velocity and time series of PS pixels from the model prediction show a long-term uplift on the upper slope, which represents downslope sliding in the target area. Furthermore, some iDS pixels exhibit a seasonal deformation on the lower part of the slope. The capability for these additional deformation analyses underscores the potential of this new deep-learning-based approach.
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
Integrating Topographic Features and Satellite Images with Deep Learning for Landslide Mapping and Monitoring
Landslide is one of the most commonly happened and threatening natural hazards in Taiwan. Because of the complicated terrain, geological, geotechnical and weather conditions landslides are frequently triggered by earthquakes, typhoons or heavy rainfalls in Taiwan, and sometimes result in serious damages. Satellite imagery is one of the commonly used sources to determine the extent of landslides for mapping, inventorying, assessment and hazard mitigation decision support. However, conventional image-based landslide detection approaches rely only on spectral and two-dimensional spatial characteristics, which may not be able to achieve high accuracy and difficult to differentiate different landslide-related terrain interpretations. This research integrates topographic features with high resolution satellite images to improve landslide detection and monitoring effectiveness and efficiency. In addition to just including height information from three-dimensional (3D) point clouds or digital elevation/terrain models (DEM/DTM), multi-scale landslide-related topographic features are derived from 3D DTM generated from airborne LiDAR surveys or stereo satellite images. These features include: slope, curvature, surface roughness, topographic position index, geomorphons and geo-hydrological features etc. These features are essential for identifying important landslide terrains, such as hummocky, crown scarps, toe bulges, gullies and the like. Based on the calculated topographic features, landslide candidate areas can be identified according to a developed scoring/classification equation. The derived topographic features and resultant landslide candidate (scores) are integrated with high resolution satellite images for landslide detection. A deep learning model based on ResUet is utilized to identify landslide areas. The developed framework was applied to analyze multi-temporal satellite images and digital terrain data of a mountainous watershed region in southern Taiwan. Preliminary results indicate that integrating topographic features with satellite images can improve the performance of landslide detection and is an effective approach for long-term monitoring of large- areas vulnerable to landslide hazards.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessIntegrating Topographic Features and Satellite Images with Deep Learning for Landslide Mapping and Monitoring
Landslide is one of the most commonly happened and threatening natural hazards in Taiwan. Because of the complicated terrain, geological, geotechnical and weather conditions landslides are frequently triggered by earthquakes, typhoons or heavy rainfalls in Taiwan, and sometimes result in serious damages. Satellite imagery is one of the commonly used sources to determine the extent of landslides for mapping, inventorying, assessment and hazard mitigation decision support. However, conventional image-based landslide detection approaches rely only on spectral and two-dimensional spatial characteristics, which may not be able to achieve high accuracy and difficult to differentiate different landslide-related terrain interpretations. This research integrates topographic features with high resolution satellite images to improve landslide detection and monitoring effectiveness and efficiency. In addition to just including height information from three-dimensional (3D) point clouds or digital elevation/terrain models (DEM/DTM), multi-scale landslide-related topographic features are derived from 3D DTM generated from airborne LiDAR surveys or stereo satellite images. These features include: slope, curvature, surface roughness, topographic position index, geomorphons and geo-hydrological features etc. These features are essential for identifying important landslide terrains, such as hummocky, crown scarps, toe bulges, gullies and the like. Based on the calculated topographic features, landslide candidate areas can be identified according to a developed scoring/classification equation. The derived topographic features and resultant landslide candidate (scores) are integrated with high resolution satellite images for landslide detection. A deep learning model based on ResUet is utilized to identify landslide areas. The developed framework was applied to analyze multi-temporal satellite images and digital terrain data of a mountainous watershed region in southern Taiwan. Preliminary results indicate that integrating topographic features with satellite images can improve the performance of landslide detection and is an effective approach for long-term monitoring of large- areas vulnerable to landslide hazards.
- Open Access
Building Footprint Extraction for Large-Scale Basemaps Using Very-High-Resolution Satellite Imagery
Accurate building footprint is a fundamental element of large-scale base maps, which serve as critical inputs for urban planning, infrastructure development, environmental monitoring, and disaster management. While building footprint extraction and geometric regularization have been widely studied, their combined application for automated, large-scale basemap generation using very-high-resolution satellite imagery has received limited attention. To address this gap, this study proposes an integrated framework that leverages deep learning and geometric regularization to efficiently extract and refine building footprints for large-scale base maps. The framework first enhances spectral, spatial, and textural features of very-high-resolution satellite imagery through pan-sharpening, NDVI computation, GLCM-based texture analysis, and PCA. A Mask R-CNN model is then trained on multi-band imagery to segment building footprints, followed by geometric regularization to simplify and align polygons along dominant structural orientations. Object-based evaluation on ground-truth buildings demonstrates high performance, with 97.6% precision, 91.6% recall, and a 94.5% F1-score. The proposed systematic framework substantially reduces production time compared to manual stereo-plotting, requiring less than an hour per 5.29 km2 map sheet in operational production, representing a more than 35-fold efficiency gain. While minor geometric inaccuracies and merged adjacent buildings persist, the methodology offers a robust, scalable, and efficient approach to support large-scale base map production.
Open AccessBuilding Footprint Extraction for Large-Scale Basemaps Using Very-High-Resolution Satellite Imagery
Accurate building footprint is a fundamental element of large-scale base maps, which serve as critical inputs for urban planning, infrastructure development, environmental monitoring, and disaster management. While building footprint extraction and geometric regularization have been widely studied, their combined application for automated, large-scale basemap generation using very-high-resolution satellite imagery has received limited attention. To address this gap, this study proposes an integrated framework that leverages deep learning and geometric regularization to efficiently extract and refine building footprints for large-scale base maps. The framework first enhances spectral, spatial, and textural features of very-high-resolution satellite imagery through pan-sharpening, NDVI computation, GLCM-based texture analysis, and PCA. A Mask R-CNN model is then trained on multi-band imagery to segment building footprints, followed by geometric regularization to simplify and align polygons along dominant structural orientations. Object-based evaluation on ground-truth buildings demonstrates high performance, with 97.6% precision, 91.6% recall, and a 94.5% F1-score. The proposed systematic framework substantially reduces production time compared to manual stereo-plotting, requiring less than an hour per 5.29 km2 map sheet in operational production, representing a more than 35-fold efficiency gain. While minor geometric inaccuracies and merged adjacent buildings persist, the methodology offers a robust, scalable, and efficient approach to support large-scale base map production.
- Open Access
- 18 October 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (20) , 3477
Deep Learning-Based Multitemporal Spatial Analytics for Assessing Reclamation Compliance of Coal Mining Permits in Kalimantan with Satellite Images
Monitoring reclamation compliance is important to ensure mining activities follow environmental regulations and reduce land degradation. Yet, few studies directly assess compliance by linking multitemporal satellite data with mining permits. This study presents a multitemporal spatial analytics approach to evaluate reclamation compliance in coal mining permit areas in South Kalimantan, Indonesia. Using satellite imagery from 2016 to 2021, a U-Net-based deep learning classification model classified five land surface types (topsoil, subsoil, vegetation, coal bodies and water bodies) with 0.94 accuracy and a Kappa statistic of 0.91. However, this relatively high accuracy was influenced by the dominance of vegetation compared to more challenging classes such as topsoil and subsoil, which remain subject to misclassification. Analysis of temporal transitions revealed patterns of surface disturbance and delayed reclamation, particularly shown by increased subsoil and reduced vegetation. These changes were integrated with coal mining permit boundaries to derived compliance ratios (CR) ranging from 0.32 to 1.44 across nine permit holders, most of which showed moderate to excellent compliance levels. This indicates that reclamation efforts have been generally being implemented, with several permit holders exceeding expectations, while a few others still need to improve. Reclamation Activity Index (RAI) was developed to classify annual performance and showed strong alignment with the U-Net-based deep learning classification model for surface change trends. The proposed approach provides a scalable and practical tool to support evidence-based monitoring and enforcement of mining reclamation policies.
- 18 October 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (20) , 3477
Open AccessDeep Learning-Based Multitemporal Spatial Analytics for Assessing Reclamation Compliance of Coal Mining Permits in Kalimantan with Satellite Images
Monitoring reclamation compliance is important to ensure mining activities follow environmental regulations and reduce land degradation. Yet, few studies directly assess compliance by linking multitemporal satellite data with mining permits. This study presents a multitemporal spatial analytics approach to evaluate reclamation compliance in coal mining permit areas in South Kalimantan, Indonesia. Using satellite imagery from 2016 to 2021, a U-Net-based deep learning classification model classified five land surface types (topsoil, subsoil, vegetation, coal bodies and water bodies) with 0.94 accuracy and a Kappa statistic of 0.91. However, this relatively high accuracy was influenced by the dominance of vegetation compared to more challenging classes such as topsoil and subsoil, which remain subject to misclassification. Analysis of temporal transitions revealed patterns of surface disturbance and delayed reclamation, particularly shown by increased subsoil and reduced vegetation. These changes were integrated with coal mining permit boundaries to derived compliance ratios (CR) ranging from 0.32 to 1.44 across nine permit holders, most of which showed moderate to excellent compliance levels. This indicates that reclamation efforts have been generally being implemented, with several permit holders exceeding expectations, while a few others still need to improve. Reclamation Activity Index (RAI) was developed to classify annual performance and showed strong alignment with the U-Net-based deep learning classification model for surface change trends. The proposed approach provides a scalable and practical tool to support evidence-based monitoring and enforcement of mining reclamation policies.
- Open Access
- 11 September 2025
- Published by MDPI AG in Sustainability
- Vol. 17 (18) , 8188
Integrating CMIP6 and Remote Sensing Datasets for Current and Future Flood Susceptibility Projections Using Machine Learning Under Climate Change Scenarios in Demak District for Future Sustainable Planning
Among various natural hazards, floods stand out due to their frequency and severe impact on society and the environment. This study aimed to develop a flood susceptibility model for Demak District, Indonesia, by integrating remote sensing data, machine learning techniques, and CMIP6 Global Climate Model (GCM) data. The approach involved mapping current flood susceptibility using Sentinel-1 SAR data as the flood inventory and applying machine learning algorithms such as MLP-NN, Random Forest, Support Vector Machine (SVM), and XGBoost to predict flood-prone areas. Additionally, future flood susceptibility was projected using CMIP6 GCM precipitation data under three Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2.6, SSP2-4.5, and SSP5-8.5) covering the 2021–2100 period. To enhance the reliability of future projections, a multi-model ensemble approach was employed by combining the outputs of multiple GCMs to reduce model uncertainties. The results showed a significant increase in flood susceptibility, especially under higher emission scenarios (SSP5-8.5), with very high susceptibility areas growing from 16.67% in the current period to 27.43% by 2081–2100. The XGBoost model demonstrated the best performance in both current and future projections, providing valuable sustainable planning insights for flood risk management and adaptation to climate change.
- 11 September 2025
- Published by MDPI AG in Sustainability
- Vol. 17 (18) , 8188
Open AccessIntegrating CMIP6 and Remote Sensing Datasets for Current and Future Flood Susceptibility Projections Using Machine Learning Under Climate Change Scenarios in Demak District for Future Sustainable Planning
Among various natural hazards, floods stand out due to their frequency and severe impact on society and the environment. This study aimed to develop a flood susceptibility model for Demak District, Indonesia, by integrating remote sensing data, machine learning techniques, and CMIP6 Global Climate Model (GCM) data. The approach involved mapping current flood susceptibility using Sentinel-1 SAR data as the flood inventory and applying machine learning algorithms such as MLP-NN, Random Forest, Support Vector Machine (SVM), and XGBoost to predict flood-prone areas. Additionally, future flood susceptibility was projected using CMIP6 GCM precipitation data under three Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2.6, SSP2-4.5, and SSP5-8.5) covering the 2021–2100 period. To enhance the reliability of future projections, a multi-model ensemble approach was employed by combining the outputs of multiple GCMs to reduce model uncertainties. The results showed a significant increase in flood susceptibility, especially under higher emission scenarios (SSP5-8.5), with very high susceptibility areas growing from 16.67% in the current period to 27.43% by 2081–2100. The XGBoost model demonstrated the best performance in both current and future projections, providing valuable sustainable planning insights for flood risk management and adaptation to climate change.
- Open Access
- 16 July 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (14) , 2460
Meteorological Drivers and Agricultural Drought Diagnosis Based on Surface Information and Precipitation from Satellite Observations in Nusa Tenggara Islands, Indonesia
Agriculture accounts for 29% of the Gross Domestic Product of the Nusa Tenggara Islands (NTIs). However, recurring agricultural droughts pose a major threat to the sustainability of agriculture in this region. The interplay between precipitation, solar radiation, and surface temperature as meteorological factors plays a key role in affecting vegetation (Soil-Adjusted Vegetation Index) and agricultural drought (Temperature Vegetation Dryness Index) in the NTIs. Based on the analyses of interplay with temporal lag, this study investigates the effect of each factor on agricultural drought and attempts to provide early warnings regarding drought in the NTIs. We collected surface information data from Moderate-Resolution Imaging Spectroradiometer (MODIS). Meanwhile, rainfall was estimated from Himawari-8 based on the INSAT Multi-Spectral Rainfall Algorithm (IMSRA). The results showed reliable performance for 8-day and monthly scales against gauges. The drought analysis results reveal that the NTIs suffer from mild-to-moderate droughts, where cropland is the most vulnerable, causing shifts in the rice cropping season. The driving factors could also explain >60% of the vegetation and surface-dryness conditions. Furthermore, our monthly and 8-day TVDI estimation models could capture spatial drought patterns consistent with MODIS, with coefficient of determination (R2) values of more than 0.64. The low error rates and the ability to capture the spatial distribution of droughts, especially in open-land vegetation, highlight the potential of these models to provide an estimation of agricultural drought.
- 16 July 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (14) , 2460
Open AccessMeteorological Drivers and Agricultural Drought Diagnosis Based on Surface Information and Precipitation from Satellite Observations in Nusa Tenggara Islands, Indonesia
Agriculture accounts for 29% of the Gross Domestic Product of the Nusa Tenggara Islands (NTIs). However, recurring agricultural droughts pose a major threat to the sustainability of agriculture in this region. The interplay between precipitation, solar radiation, and surface temperature as meteorological factors plays a key role in affecting vegetation (Soil-Adjusted Vegetation Index) and agricultural drought (Temperature Vegetation Dryness Index) in the NTIs. Based on the analyses of interplay with temporal lag, this study investigates the effect of each factor on agricultural drought and attempts to provide early warnings regarding drought in the NTIs. We collected surface information data from Moderate-Resolution Imaging Spectroradiometer (MODIS). Meanwhile, rainfall was estimated from Himawari-8 based on the INSAT Multi-Spectral Rainfall Algorithm (IMSRA). The results showed reliable performance for 8-day and monthly scales against gauges. The drought analysis results reveal that the NTIs suffer from mild-to-moderate droughts, where cropland is the most vulnerable, causing shifts in the rice cropping season. The driving factors could also explain >60% of the vegetation and surface-dryness conditions. Furthermore, our monthly and 8-day TVDI estimation models could capture spatial drought patterns consistent with MODIS, with coefficient of determination (R2) values of more than 0.64. The low error rates and the ability to capture the spatial distribution of droughts, especially in open-land vegetation, highlight the potential of these models to provide an estimation of agricultural drought.
- Open Access
- 28 April 2025
- Published by MDPI AG in Engineering Proceedings
- Vol. 92 (1) , 32
Pansharpening Remote Sensing Images Using Generative Adversarial Networks
Pansharpening is a remote sensing image fusion technique that combines a high-resolution (HR) panchromatic (PAN) image with a low-resolution (LR) multispectral (MS) image to produce an HR MS image. The primary challenge in pansharpening lies in preserving the spatial details of the PAN image while maintaining the spectral integrity of the MS image. To address this, this article presents a generative adversarial network (GAN)-based approach to pansharpening. The GAN discriminator facilitated matching the generated image’s intensity to the HR PAN image and preserving the spectral characteristics of the LR MS image. The performance in generating images was evaluated using the peak signal-to-noise ratio (PSNR). For the experiment, original LR MS and HR PAN satellite images were partitioned into smaller patches, and the GAN model was validated using an 80:20 training-to-testing data ratio. The results illustrated that the super-resolution images generated by the SRGAN model achieved a PSNR of 31 dB. These results demonstrated the developed model’s ability to reconstruct the geometric, textural, and spectral information from the images.
- 28 April 2025
- Published by MDPI AG in Engineering Proceedings
- Vol. 92 (1) , 32
Open AccessPansharpening Remote Sensing Images Using Generative Adversarial Networks
Pansharpening is a remote sensing image fusion technique that combines a high-resolution (HR) panchromatic (PAN) image with a low-resolution (LR) multispectral (MS) image to produce an HR MS image. The primary challenge in pansharpening lies in preserving the spatial details of the PAN image while maintaining the spectral integrity of the MS image. To address this, this article presents a generative adversarial network (GAN)-based approach to pansharpening. The GAN discriminator facilitated matching the generated image’s intensity to the HR PAN image and preserving the spectral characteristics of the LR MS image. The performance in generating images was evaluated using the peak signal-to-noise ratio (PSNR). For the experiment, original LR MS and HR PAN satellite images were partitioned into smaller patches, and the GAN model was validated using an 80:20 training-to-testing data ratio. The results illustrated that the super-resolution images generated by the SRGAN model achieved a PSNR of 31 dB. These results demonstrated the developed model’s ability to reconstruct the geometric, textural, and spectral information from the images.