Aldiano Rahmadya
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Aldiano Rahmadya has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Fajar Setiawan
National Research and Innovation Agency
2 shared publications - Meti Yulianti
National Research and Innovation Agency
2 shared publications - Ahmad Muhtadi
Universitas Sumatera Utara
1 shared publication Alifia Imananda
Bogor Agricultural University
1 shared publication- Arianto Budi Santoso
Research Center for Limnology and Water Resources National Research and Innovation Agency (BRIN) Bogor Indonesia
1 shared publication - Astried Sunaryani
Doctoral Program of Environmental Engineering, Faculty of Civil and Environmental Engineering, Institut Teknologi Bandung, Kota Bandung 40132, Indonesia
1 shared publication
Latest Publications
- Open Access
- 30 July 2026
- Published by Elsevier in SSRN Electronic Journal
Representation and Hyperparameter Search in Landslide Susceptibility Modeling: Tree Ensembles and One-Dimensional Convolutional Networks under Nested Spatial Validation
Spatial autocorrelation and arbitrary ordering of heterogeneous conditioning factors can distort algorithm comparisons in landslide susceptibility modeling. This study benchmarks four tree ensembles, a tabular multilayer perceptron, a support-vector classifier, penalized logistic regression, a frequency-ratio reference, and three one-dimensional convolutional architectures using 695 landslide locations, 695 pseudo-absence locations, and 13 conditioning factors in Bogor Regency, Indonesia. Models were evaluated by repeated nested five-fold spatial-block cross-validation using 5 km blocks; tuning was confined to inner training blocks, and discrimination, precision–recall performance, Brier score, expected calibration error, and calibration slope were computed from pooled out-of-fold predictions. Tree ensembles achieved ROC-AUC 0.833–0.839 and PR-AUC 0.848–0.852, whereas the convolutional models achieved ROC-AUC 0.794–0.807. CatBoost exceeded ResNet by 0.032 ROC-AUC (95% CI 0.013–0.051), and random forest exceeded the EfficientNet-inspired 1-D MBConv model by 0.036 (95% CI 0.016–0.062); ResNet and the tabular multilayer perceptron were not distinguishable. Across 30 factor orderings, convolutional ROC-AUC varied by approximately 0.05–0.07. Under an equal budget of 24 unique candidates, six search strategies produced best inner-CV AUCs of 0.723–0.729. TreeSHAP analysis of random forest ranked elevation, geology, terrain ruggedness, NDVI, and soil type as the leading contributors. For this small tabular geospatial dataset and evaluation design, model representation and spatial validation were more consequential than added architectural or metaheuristic complexity.
- 30 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessRepresentation and Hyperparameter Search in Landslide Susceptibility Modeling: Tree Ensembles and One-Dimensional Convolutional Networks under Nested Spatial Validation
Spatial autocorrelation and arbitrary ordering of heterogeneous conditioning factors can distort algorithm comparisons in landslide susceptibility modeling. This study benchmarks four tree ensembles, a tabular multilayer perceptron, a support-vector classifier, penalized logistic regression, a frequency-ratio reference, and three one-dimensional convolutional architectures using 695 landslide locations, 695 pseudo-absence locations, and 13 conditioning factors in Bogor Regency, Indonesia. Models were evaluated by repeated nested five-fold spatial-block cross-validation using 5 km blocks; tuning was confined to inner training blocks, and discrimination, precision–recall performance, Brier score, expected calibration error, and calibration slope were computed from pooled out-of-fold predictions. Tree ensembles achieved ROC-AUC 0.833–0.839 and PR-AUC 0.848–0.852, whereas the convolutional models achieved ROC-AUC 0.794–0.807. CatBoost exceeded ResNet by 0.032 ROC-AUC (95% CI 0.013–0.051), and random forest exceeded the EfficientNet-inspired 1-D MBConv model by 0.036 (95% CI 0.016–0.062); ResNet and the tabular multilayer perceptron were not distinguishable. Across 30 factor orderings, convolutional ROC-AUC varied by approximately 0.05–0.07. Under an equal budget of 24 unique candidates, six search strategies produced best inner-CV AUCs of 0.723–0.729. TreeSHAP analysis of random forest ranked elevation, geology, terrain ruggedness, NDVI, and soil type as the leading contributors. For this small tabular geospatial dataset and evaluation design, model representation and spatial validation were more consequential than added architectural or metaheuristic complexity.
- 9 March 2026
- Published by Taylor & Francis in Inland Waters
- Vol. 16 (1)
Stratification and mixing dynamics in tropical polymictic Lake Batur, Indonesia
- 9 March 2026
- Published by Taylor & Francis in Inland Waters
- Vol. 16 (1)
Stratification and mixing dynamics in tropical polymictic Lake Batur, Indonesia
- Open Access
- 1 January 2026
- Published by Wiley in Scientifica
- Vol. 2026 (1)
Spatiotemporal Patterns of Microplastic Distribution in Surface Water and Nearshore Sediments of a Large Tropical Lake: Evidence From Lake Toba, Indonesia
Microplastic (MP) pollution is increasingly documented in freshwater ecosystems, yet lake‐scale evidence from large tropical lakes remains limited. We assessed MPs in surface water and nearshore sediments at 21 stations across Lake Toba, Indonesia, during the dry and rainy seasons. Mean MP abundance was higher in the rainy season than in the dry season in both surface water (1.461 ± 0.836 vs. 0.288 ± 0.110 particles m −3 ) and sediment (3.14 ± 0.59 vs. 1.26 ± 0.31 particles g −1 dry weight). Fragments and particles measuring 1000–2000 μm were dominant, while black and blue particles jointly represented 67%–68% of the observed MPs. Surface‐water abundance was highest in the northern sector, whereas nearshore sediment abundance was higher in the southern sector. These contrasting patterns are consistent with spatial redistribution and retention within the lake, although transport pathways and sediment‐focusing processes were not measured directly. Aquaculture, fishing, tourism, and domestic activities are therefore discussed as potential rather than confirmed sources. This study provides the first seasonally resolved, lake‐wide baseline integrating surface water and nearshore sediments in Lake Toba and identifies priorities for source‐specific monitoring and hydrodynamic investigation.
- 1 January 2026
- Published by Wiley in Scientifica
- Vol. 2026 (1)
Open AccessSpatiotemporal Patterns of Microplastic Distribution in Surface Water and Nearshore Sediments of a Large Tropical Lake: Evidence From Lake Toba, Indonesia
Microplastic (MP) pollution is increasingly documented in freshwater ecosystems, yet lake‐scale evidence from large tropical lakes remains limited. We assessed MPs in surface water and nearshore sediments at 21 stations across Lake Toba, Indonesia, during the dry and rainy seasons. Mean MP abundance was higher in the rainy season than in the dry season in both surface water (1.461 ± 0.836 vs. 0.288 ± 0.110 particles m −3 ) and sediment (3.14 ± 0.59 vs. 1.26 ± 0.31 particles g −1 dry weight). Fragments and particles measuring 1000–2000 μm were dominant, while black and blue particles jointly represented 67%–68% of the observed MPs. Surface‐water abundance was highest in the northern sector, whereas nearshore sediment abundance was higher in the southern sector. These contrasting patterns are consistent with spatial redistribution and retention within the lake, although transport pathways and sediment‐focusing processes were not measured directly. Aquaculture, fishing, tourism, and domestic activities are therefore discussed as potential rather than confirmed sources. This study provides the first seasonally resolved, lake‐wide baseline integrating surface water and nearshore sediments in Lake Toba and identifies priorities for source‐specific monitoring and hydrodynamic investigation.
- 19 March 2025
- Published by Chiang Mai University in Natural and Life Sciences Communications
- Vol. 24 (2)
Utilization of Deep Learning and Unmanned Aerial Vehicle (UAV) for Monitoring Plastic Debris
Plastic debris has become a significant environmental problem due to its persistence in nature and harmful ecological impacts. Recently, Unmanned Aerial Vehicles (UAVs) and deep learning models have been utilized for detection and analysis, as effective plastic debris monitoring is essential for mitigation efforts. While extensive research has focused on coastal areas, riverine environments remain understudied. This study evaluates the capability of UAVs and the You Only Look Once version 5 (YOLOv5) deep learning model in detecting and quantifying macro plastic debris in river systems. The research was conducted in 2021 in the lower reaches of the Citarum River, Muara Gembong District, Bekasi Regency, West Java Province, Indonesia. UAVs were deployed to capture aerial images, which were then processed using the YOLOv5_small and YOLOv5_large models. A total of 416 images were analyzed, and plastic debris was classified based on size and distribution. The results indicate that UAV effectively capture high-resolution images of plastic debris in river environments. The YOLOv5 model successfully identified plastic debris; however, its accuracy requires further improvement, with mean Average Precision (mAP) scores of 16.90% and 19.21% for YOLOv5_small and YOLOv5_large, respectively. Detection performance varied depending on factors such as solar radiation intensity, surface water reflection, and debris characteristics. Future research should explore advanced deep learning architectures, integrate multi-spectral imaging, and assess seasonal variations in plastic debris accumulation. This study provides a foundation for the development of automated, scalable plastic debris monitoring systems to support environmental management and pollution mitigation efforts.
- 19 March 2025
- Published by Chiang Mai University in Natural and Life Sciences Communications
- Vol. 24 (2)
Utilization of Deep Learning and Unmanned Aerial Vehicle (UAV) for Monitoring Plastic Debris
Plastic debris has become a significant environmental problem due to its persistence in nature and harmful ecological impacts. Recently, Unmanned Aerial Vehicles (UAVs) and deep learning models have been utilized for detection and analysis, as effective plastic debris monitoring is essential for mitigation efforts. While extensive research has focused on coastal areas, riverine environments remain understudied. This study evaluates the capability of UAVs and the You Only Look Once version 5 (YOLOv5) deep learning model in detecting and quantifying macro plastic debris in river systems. The research was conducted in 2021 in the lower reaches of the Citarum River, Muara Gembong District, Bekasi Regency, West Java Province, Indonesia. UAVs were deployed to capture aerial images, which were then processed using the YOLOv5_small and YOLOv5_large models. A total of 416 images were analyzed, and plastic debris was classified based on size and distribution. The results indicate that UAV effectively capture high-resolution images of plastic debris in river environments. The YOLOv5 model successfully identified plastic debris; however, its accuracy requires further improvement, with mean Average Precision (mAP) scores of 16.90% and 19.21% for YOLOv5_small and YOLOv5_large, respectively. Detection performance varied depending on factors such as solar radiation intensity, surface water reflection, and debris characteristics. Future research should explore advanced deep learning architectures, integrate multi-spectral imaging, and assess seasonal variations in plastic debris accumulation. This study provides a foundation for the development of automated, scalable plastic debris monitoring systems to support environmental management and pollution mitigation efforts.
- Open Access
- 1 January 2025
- Published by Faculty of Agriculture, Brawijaya University in Journal of Degraded and Mining Lands Management
- Vol. 12 (2) , 7379-7388
Developing algorithms for estimating total suspended solids (TSS) using unmanned aerial vehicle: A case study in the Upper Citarum River, Indonesia
Monitoring total suspended solids (TSS) is essential as suspended sediments impact the environment and human health in various ways. However, TSS data are limited in many regions because the methods currently applied through in situ measurements are time-consuming and labor-intensive. The study aimed to develop algorithms to estimate TSS using data derived from UAVs and field measurements. Remote sensing technology, such as unmanaged aerial vehicle (UAV), was applied to obtain imagery data to estimate TSS content. These results were then compared with laboratory analysis of in-situ water samples, determined by gravimetric methods following standard protocols. The results showed that the algorithm developed using three-band ratios, the blue/green + red/green + NIR (near infra red)/green, produces a high R2 (0.70), indicating that this combination is reliable for use in estimating TSS content in a river section. The high accuracy of the red band for suspended sediment prediction is attributed to its spectral signature in turbid water, which shows higher reflectance compared to clean water. The results of this study have the potential to help river managers obtain TSS data quickly at a relatively low cost.
- 1 January 2025
- Published by Faculty of Agriculture, Brawijaya University in Journal of Degraded and Mining Lands Management
- Vol. 12 (2) , 7379-7388
Open AccessDeveloping algorithms for estimating total suspended solids (TSS) using unmanned aerial vehicle: A case study in the Upper Citarum River, Indonesia
Monitoring total suspended solids (TSS) is essential as suspended sediments impact the environment and human health in various ways. However, TSS data are limited in many regions because the methods currently applied through in situ measurements are time-consuming and labor-intensive. The study aimed to develop algorithms to estimate TSS using data derived from UAVs and field measurements. Remote sensing technology, such as unmanaged aerial vehicle (UAV), was applied to obtain imagery data to estimate TSS content. These results were then compared with laboratory analysis of in-situ water samples, determined by gravimetric methods following standard protocols. The results showed that the algorithm developed using three-band ratios, the blue/green + red/green + NIR (near infra red)/green, produces a high R2 (0.70), indicating that this combination is reliable for use in estimating TSS content in a river section. The high accuracy of the red band for suspended sediment prediction is attributed to its spectral signature in turbid water, which shows higher reflectance compared to clean water. The results of this study have the potential to help river managers obtain TSS data quickly at a relatively low cost.