5:15 PM - 7:15 PM
[HGG01-P02] Assessment of Density and Particle Size Distribution of Volcanic Sediments at Merapi and Semeru Volcanoes (Indonesia) Using SfM-MVS Techniques
Keywords:Structure-from-Motion Multi-View Stereo, Bulk density
1. Introduction
Sediment movement in volcanic regions (e.g., debris flows, lahars) increases natural disaster risks and affects watershed environments. Accurate density and grain size data are crucial for disaster prevention and management. Conventional methods (e.g., core cutter, sand replacement) are labor-intensive and may not capture in-situ compression/expansion of particles.
We applied an SfM-MVS-based bulk density estimation method in the Merapi and Semeru Volcano areas in Indonesia. This approach accounts for compression/expansion of inter-particle spaces and enables non-destructive field measurements. We also integrated grain size analysis to assess spatial heterogeneity and depositional characteristics.
2. Methods
2.1. Study Sites and Sampling
Surveys were conducted in:
Merapi Volcano (Gendol and Krasak River basins) Semeru Volcano (around Curah Lengkong)
At each site:
A 50×25 cm frame marked the measurement area. Photographs (overhead/oblique) were taken for SfM. Deposits were excavated, and samples were collected for drying, grain-size analysis, and weight measurement. Post-excavation photographs were taken to build 3D models and calculate volumetric changes.
2.2. SfM-MVS Analysis and Bulk Density Estimation
Images were processed with Agisoft Metashape Pro to generate high-resolution 3D point clouds. CloudCompare was used to calculate volumetric differences before and after excavation. By comparing these volumes with the dry sample weights, we estimated bulk density. This method captures in-situ particle compression/expansion. Accuracy was evaluated using GCP-based RMS and MAE errors.
2.3. Grain Size Distribution Analysis
Samples were dried, sieved, and classified by weight ratio for each grain size. We examined how coarse/fine ratios influence bulk density and depositional characteristics.
3. Results and Discussion
3.1. Volume Measurement and Bulk Density via SfM-MVS
High-resolution 3D models were successfully constructed at both volcano sites. Volumetric differences allowed bulk density estimation, reflecting actual inter-particle void conditions more accurately than traditional methods.
3.2. Grain Size Distribution and Depositional Characteristics
Grain size analysis revealed variations in coarse vs. fine fractions, likely due to selective deposition driven by rainfall and flow velocity. Coarser sites tended to have higher bulk density, whereas finer sites showed lower values. Future work incorporating water content and porosity could enhance accuracy.
4. Conclusion and Future Outlook
We applied an SfM-MVS-based density measurement method in volcanic deposits at Merapi and Semeru, capturing compression/expansion effects. Integrating grain size analysis enabled high-resolution 3D modeling and bulk density estimation, offering insights into depositional characteristics and density variation mechanisms.
Future work should focus on:
Developing refined density models that include water content and porosity Conducting time-series surveys to track sediment dynamics Comparing results across different volcanic regions
Further refinement and wider application of this method can improve sediment disaster risk assessments and watershed management. The ability to capture real void conditions between particles positions SfM-MVS as a valuable tool for disaster prevention and mitigation.
Sediment movement in volcanic regions (e.g., debris flows, lahars) increases natural disaster risks and affects watershed environments. Accurate density and grain size data are crucial for disaster prevention and management. Conventional methods (e.g., core cutter, sand replacement) are labor-intensive and may not capture in-situ compression/expansion of particles.
We applied an SfM-MVS-based bulk density estimation method in the Merapi and Semeru Volcano areas in Indonesia. This approach accounts for compression/expansion of inter-particle spaces and enables non-destructive field measurements. We also integrated grain size analysis to assess spatial heterogeneity and depositional characteristics.
2. Methods
2.1. Study Sites and Sampling
Surveys were conducted in:
Merapi Volcano (Gendol and Krasak River basins) Semeru Volcano (around Curah Lengkong)
At each site:
A 50×25 cm frame marked the measurement area. Photographs (overhead/oblique) were taken for SfM. Deposits were excavated, and samples were collected for drying, grain-size analysis, and weight measurement. Post-excavation photographs were taken to build 3D models and calculate volumetric changes.
2.2. SfM-MVS Analysis and Bulk Density Estimation
Images were processed with Agisoft Metashape Pro to generate high-resolution 3D point clouds. CloudCompare was used to calculate volumetric differences before and after excavation. By comparing these volumes with the dry sample weights, we estimated bulk density. This method captures in-situ particle compression/expansion. Accuracy was evaluated using GCP-based RMS and MAE errors.
2.3. Grain Size Distribution Analysis
Samples were dried, sieved, and classified by weight ratio for each grain size. We examined how coarse/fine ratios influence bulk density and depositional characteristics.
3. Results and Discussion
3.1. Volume Measurement and Bulk Density via SfM-MVS
High-resolution 3D models were successfully constructed at both volcano sites. Volumetric differences allowed bulk density estimation, reflecting actual inter-particle void conditions more accurately than traditional methods.
3.2. Grain Size Distribution and Depositional Characteristics
Grain size analysis revealed variations in coarse vs. fine fractions, likely due to selective deposition driven by rainfall and flow velocity. Coarser sites tended to have higher bulk density, whereas finer sites showed lower values. Future work incorporating water content and porosity could enhance accuracy.
4. Conclusion and Future Outlook
We applied an SfM-MVS-based density measurement method in volcanic deposits at Merapi and Semeru, capturing compression/expansion effects. Integrating grain size analysis enabled high-resolution 3D modeling and bulk density estimation, offering insights into depositional characteristics and density variation mechanisms.
Future work should focus on:
Developing refined density models that include water content and porosity Conducting time-series surveys to track sediment dynamics Comparing results across different volcanic regions
Further refinement and wider application of this method can improve sediment disaster risk assessments and watershed management. The ability to capture real void conditions between particles positions SfM-MVS as a valuable tool for disaster prevention and mitigation.