In an article published in Remote Sensing, researchers suggested a technique to indirectly observe the deformation features of landslides by extracting anomalous vegetation coverage information, particularly for the high-mountain inaccessible landslides in southwest China.
Here, the team used optical remote sensing data from Gaofen-1 (GF-1) to extract vegetation coverage anomaly details at the Jizong Shed-Tunnel landslide, situated on the main road leading to Tibet.
The authors then employed a time series analysis to determine the spatial and temporal features of the vegetation coverage anomaly data. Sentinel-1 data was analyzed using the small baseline subsets interferometry synthetic aperture radar (SBAS-InSAR) technique to collect time-series surface deformation data.
Finally, the findings of the two approaches were examined and validated. The observations demonstrated a clear drop in vegetative coverage (VC), with the highest percentage rise of 8.77% for medium and low vegetation cover. An evident deformation of the landslide surrounding terrain, with the maximum settlement rates between zero and 30 mm/year, was observed.
The time-series analysis revealed that the shift trends for the two approaches were almost identical. This study demonstrated the degree to which the strategy of employing abnormal vegetation coverage information to track the Jizong Shed-Tunnel landslide was reliable and practical. It offered fresh insight and a useful supplementary landslide monitoring tool.
The Need for Better Landslide Monitoring Tools
The destruction and instability of the soil, rock or other artificial materials due to gravitational action is called a landslide. Due to its unexpectedness, destructive strength, and propensity for subsequent disasters, a landslide has become one of the most catastrophic natural disasters, resulting in significant human casualties and economic losses worldwide.
Landslide tragedies are widespread in China, specifically in the southwest regions, owing to the vast undulated landscape, loose soil, and heavy rain.
The region is mountainous adjacent to the G214 National Highway, which is regarded as a “lifeline” between the Sichuan and Tibet regions.
A landslide obstructs the G214 National Highway’s usual flow, resulting in significant human casualties and financial damages. Therefore, landslide monitoring and identification are of utmost necessity to prevent and manage landslide dangers.
There has been much research on landslide monitoring. Interferometry synthetic aperture radar (InSAR) and the global navigation satellite system (GNSS) techniques are frequently used for landslide monitoring.
The global positioning system (GPS) is an extensively utilized GNSS component. High precision, excellent automation, no requirement for visibility between landslide monitoring stations, and real-time all-weather monitoring are significant benefits of GPS landslide monitoring.
With its ability to operate in all weather conditions and throughout the day, InSAR technology can gather data on surface deformation over a vast region and an extended period.
InSAR can also create a regional digital elevation model (DEM) with paired radar images, which is particularly important for regions lacking access to topography data, and supply essential information needed for landslide monitoring.
In this paper, the authors assessed the vegetation coverage to obtain the abnormal information of the Jizong Shed-Tunnel landslide from the GF-1 optical remote sensing data to validate the probability of using vegetation shifts for landslide monitoring.
The landslide area’s time-series surface deformation was obtained using the SBAS-InSAR approach, which more readily gained comprehensive landslide monitoring than the GPS method to assist the indirect optical remote sensing monitoring.
The two features were integrated to analyze the spatial and temporal properties of landslide creep and confirm the viability and efficacy of the method.
This study provided a new idea and monitoring technique for high-mountain landslides in the southwestern region that could successfully supplement the landslide monitoring methods. It also used abnormal vegetation coverage information to track the Jizong Shed-Tunnel landslides efficiently and sequentially.
The Jizong Shed-Tunnel Landslide was situated in Yunnan Province’s Diqing Tibetan Autonomous Prefecture, on the eastern bank of the Jinsha River, in Ladong Mountain.
Here, the normal fault, primarily formed of slate, volcanic rock, and limestone, was close to the Jizong Shed-Tunnel landslide. The entire slope ranged from 30 to 45 degrees. The research area received considerable precipitation.
The Jizong Shed-Tunnel landslide underwent a significant slide in 2015 due to road building and precipitation followed by a slow creep.
Sentinel-1 A satellite radar optical remote sensing images were selected for surface deformation monitoring. At the same time, GF-1 optical remote sensing images were chosen for vegetation coverage anomaly data extraction.
The GF-1 image was superimposed with the average surface deformation rate map produced by the SBAS-InSAR technology. Only the coherent subsidence locations along the line of sight (LOS) direction assessed the surface subsidence correctly.
The Jizong Shed-Tunnel landslide typically settled at a rate of zero to 30 mm/year in the LOS direction. Therefore, the center point of the vegetation coverage anomaly area was chosen to approximately represent the cumulative surface deformation of the two regions since the cumulative surface deformation acquired in the SBAS-InSAR was displayed by points.
The correlation coefficient between the vegetation coverage and the cumulative surface deformation was 0.988. In contrast, the R2 of the linear regression method was 0.977 between the two parameters. These findings indicated that the two methods had a considerable linear correlation.
The pixels with medium and low vegetation coverage continuously increased with the surface deformation subsidence, which supported tracking the Jizong Shed-Tunnel landslide using atypical vegetation data. The findings demonstrated the accuracy analysis reliability in the selected study area.
Significance of the Study
This paper proposed a technique to indirectly monitor landslides’ deformation features by retrieving the abnormal vegetation coverage data from optical remote sensing images. This method addressed the shortcomings of traditional InSAR and GPS technologies in landslide monitoring, including manpower inaccessibility, large ups and downs, and lush vegetation coverage.
The vegetation anomaly data of the Jizong Shed-Tunnel landslide was monitored using GF-1 optical remote sensing images from 2013 to 2020. Also, the SBAS-InSAR technology was employed to retrieve surface deformation data for the study region from 2017 to 2020.
In this study, the authors used GF-1 time series optical remote sensing data to calculate the green normalized difference vegetation index (GNDVI) and then determined the amount of vegetation coverage in each scene.
The retrieved information on vegetation anomalies was subjected to multitemporal quantitative and qualitative analysis, and the results showed that the vegetation coverage declined from 2013 to 2015. These findings demonstrated the strategy's viability for tracking the Jizong Shed-Tunnel landslide using the information on vegetation anomalies.
According to the SBAS-InSAR technique based on Sentinel-1 optical remote sensing data, the Jizong Shed-Tunnel landslide was at a slow creep stage as the main surface deformation area was found at the back edge of the landslide. The surface subsidence rate ranged from zero to 30 mm/year.
Guo, Q., Tong, L., Wang, H. (2022) A Monitoring Method Based on Vegetation Abnormal Information Applied to the Case of Jizong Shed-Tunnel Landslide. Remote Sensing, 14(22), 5640. https://www.mdpi.com/2072-4292/14/22/5640
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.