Next Article in Journal
LACC2.0: Improving the LACC Algorithm for Reconstructing Satellite-Derived Time Series of Vegetation Biochemical Parameters
Next Article in Special Issue
Fracture Electromagnetic Radiation Induced by a Seismic Active Zone (in the Vicinity of Eilat City, Southern Israel)
Previous Article in Journal
Self-Supervised Monocular Depth Estimation Using Global and Local Mixed Multi-Scale Feature Enhancement Network for Low-Altitude UAV Remote Sensing
Previous Article in Special Issue
Statistical Assessments of InSAR Tropospheric Corrections: Applicability and Limitations of Weather Model Products and Spatiotemporal Filtering
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Communication

Identifying Potential Landslides in Steep Mountainous Areas Based on Improved Seasonal Interferometry Stacking-InSAR

1
College of Earth Science, Chengdu University of Technology, Chengdu 610059, China
2
State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology, Chengdu 610059, China
3
College of Geological Engineering and Geomatics, Chang’an University, Xi’an 710064, China
4
Kay Laboratory of Earth Exploration and Information Techniques (Chengdu University of Technology), Ministry of Education, Chengdu 610059, China
*
Author to whom correspondence should be addressed.
Remote Sens. 2023, 15(13), 3278; https://doi.org/10.3390/rs15133278
Submission received: 23 April 2023 / Revised: 20 June 2023 / Accepted: 23 June 2023 / Published: 26 June 2023
(This article belongs to the Special Issue Geological Applications of Remote Sensing and Photogrammetry)

Abstract

:
Landslides are a major concern in the mountainous regions of southwest China, leading to significant loss of life and property damage. Therefore, it is crucial to identify potential landslides for early warning and mitigation. stacking-InSAR, a technique used for landslide identification in a wide area, has been found to be faster than conventional time-series InSAR. However, the dense vegetation in southwest China mountains has an adverse impact on the coherence of stacking-InSAR, resulting in more noise and inaccuracies in landslide identification. To address this problem, this paper proposes an improved seasonal interferometry stacking-InSAR method. It uses Sentinel-1 satellite data from 2017 to 2022. The study area is the river valley section of the G213 road from Wenchuan County to Mao County. The study reveals the characteristics of seasonal decoherence in the steep mountainous region, and identifies a total of 21 potential landslides using the improved method. Additionally, optical satellite imagery and LiDAR data were used to assist in the identification of potential landslides. The results of the conventional stacking-InSAR method and the improved seasonal interferometry stacking-InSAR method are compared, showing that the latter is more effective in noise suppression caused by low coherence. Their standard deviations were reduced by 46%, 22%, 10%, and 14%, respectively, using the quantitative statistics for the four tested areas. The proposed method provides an efficient and effective approach for detecting potential landslides in the mountainous regions of southwest China. It can serve as a valuable technical reference for future landslide identification studies in this area.

Graphical Abstract

1. Introduction

In southwest China’s mountainous regions, geological hazards are frequent, with landslides being one of the most hazardous types, especially in river valleys. They are concealed, sudden, and unpredictable, posing a serious threat to local lives and property [1,2,3,4,5]. For instance, in 2017, a landslide in Mao County’s Xinmo Village buried the entire village and resulted in 83 deaths and disappearances [6,7]. In 2018, two collapses in quick succession near Jinsha River’s Baige blocked the river and destroyed over 100 km of downstream roads and bridges due to post-failure floods [8,9,10]. In June 2020, a sudden rainstorm in Danba County triggered mudslides and activated the Aniangzhai ancient landslide, causing significant damage to local residents and the economy [11,12,13]. Thus, identifying potential landslides beforehand and implementing disaster prevention and reduction measures are crucial for ensuring local safety and economic development.
Potential landslides refer to slopes with a risk of failure but which have not failed yet, which can be determined and identified according to the displacement signal on the slopes. Recently, spaceborne interferometric synthetic aperture radar (InSAR) technology has become an important tool for identifying potential landslides [14,15]. InSAR technology uses SAR images to extract deformation information with high precision, with the advantages of all-day, all-weather, and wide coverage. Several scholars have conducted research on identifying potential landslides using InSAR [16,17,18,19,20,21,22]. Most of these studies have employed time-series InSAR methods such as SBAS-InSAR [23], PS-InSAR [24,25], etc. These conventional time-series InSAR methods are effective in filtering out errors such as atmospheric delays and terrain residuals, and detecting deformation with high accuracy; their computational efficiency is relatively low and they require good coherence for a long time [26,27]. Moreover, the mountains of southwest China are covered in lush vegetation, which leads to poor coherence, particularly in summer. As a result, SAR interferometry produces low coherence or decoherence signals in most areas, limiting the usefulness of conventional time-series InSAR methods. The stacking-InSAR method involves stacking deformation phases acquired at different times, making it suitable for identifying long-term deformation in areas with low coherence. Zhang et al. discovered that stacking-InSAR can detect potential landslides in low-coherence areas that cannot be detected via SBAS-InSAR in Li County and Mao County of southwest China. Additionally, the processing speed of stacking-InSAR is faster, but the quantitative accuracy is lower than that of the SBAS-InSAR method [28]. Liang et al. also demonstrated that stacking-InSAR can detect more potential landslides in the Ya’an area of southwest China compared to other time-series methods [29]. The above research reveals that stacking-InSAR is fast in processing and better at detecting low-coherence regions, while its quantitative deformation results are less accurate than those of conventional time-series InSAR. Therefore, further research is needed to suppress noise in stacking-InSAR in mountainous areas and improve its accuracy.
This study aims to reduce the noise and improve the accuracy of the stacking-InSAR method. The Sentinel-1 data from 2017 to 2022 in this area were utilized to investigate the temporal coherence of each interferogram. An improved seasonal interferometry stacking-InSAR method was proposed to effectively identify potential landslides. Performance evaluations of a comparison with the conventional stacking-InSAR method were performed. This method plays an important role for the wide-area automated identification of potential landslides and contributes to effective disaster prevention.

2. Materials and Methods

2.1. Study Area

As depicted in Figure 1, the study area is located in the valley section from Wenchuan County to Mao County, which belongs to the Longmen Mountain Fault area in Sichuan Province, China. The G213 road, which runs alongside the river, is the only crucial link between Wenchuan County and Mao County, and is known as the “lifeline”. However, the region is susceptible to geological hazards, particularly during the rainy season of July–August each year. This vulnerability is exacerbated by the impact of the 2008 Ms 8.0 Wenchuan earthquake, which caused extensive damage to the mountains [1]. As a result, the Minjiang River frequently experiences significant collapses on both sides, leading to road blockages. The study area has an average elevation ranging from 1300 m to 4000 m, with a substantial difference in height, which poses a high risk for large landslides. Moreover, the dense vegetation cover conceals landslides, making them challenging to detect. In recent years, human activities, such as road and water conservancy construction, have accelerated deterioration in the geological environment, leading to a significant increase in natural hazards like landslides.

2.2. Sentinel-1 SAR Data

The Sentinel-1 satellite is an Earth observation satellite of the Copernicus program of the European Space Agency [30]. The Sentinel-1 satellite is equipped with a C-band synthetic aperture radar and precise orbit control, providing a minimum revisit time of 12 days. It offers extensive data coverage for most regions worldwide and provides open source and free data. As a result, Sentinel-1 has become one of the most commonly utilized data sources for research and applications in InSAR technology. The detailed parameters of the dataset used in this study are presented in Table 1. However, its short C-band wavelength and limited penetration capability pose certain limitations, particularly in densely vegetated areas such as the southwestern mountains, where the coherence of the data is poor [29,31].

2.3. Coherence Analysis of Study Area

Interferometric phase noise mainly arises from two sources. The first is due to Doppler decoherence and thermal noise resulting from SAR hardware. The second is caused by excessive temporal or spatial baseline, which is also the most significant error in extracting the deformation phase [32,33]. The second is the primary noise discussed in this study, which always affects the recognition of deformation.
Interferometric coherence is a critical parameter for assessing the quality of interferometric pairs, and the most commonly used measure of interferometric phase coherence is the coherence coefficient [34]. The value of the coherence coefficient is distributed within the interval [0, 1], where 0 means that they are completely incoherent and 1 means that they are completely consistent. Factors such as the thermal noise of the radar system can introduce jumps in the calculation results, which hinders the accurate evaluation of coherence. Therefore, the coherence coefficient calculation method based on SAR image amplitude information is used, and its equation is shown as follows:
γ = n = 1 N m = 1 M μ 1 n , m | μ 2 n , m | n = 1 N m = 1 M | μ 1 ( n , m ) | 2 n = 1 N m = 1 M | μ 2 ( n , m ) | 2
In Equation (1), γ is the coherence coefficient. M and N are the size of the matrix for calculating the coherence; n and m are the number of ranks in the matrix. μ 1 n , m and μ 2 n , m represent the complex values of the primary and secondary image matrices at the image coordinates n , m , respectively. Based on this equation, a window of size n × m was used to calculate any pixel, so that the γ of any pixel could be obtained. The coherence coefficient of each pixel served as an indicator to assess the quality of the interferometric phase. The impact of spatial and temporal baselines on coherence is closely related to the geographic characteristics of the study area. Factors such as topography, vegetation growth, foliage orientation, and changes in soil water content can lead to temporal or spatial decoherence.
To examine the temporal coherence pattern in the region, Sentinel-1 data spanning from 2017 to 2022 were utilized for InSAR analysis. To maintain the highest possible coherence, a temporal baseline of 24 days was set. The spatial baseline was set to 250 m in order to include a larger number of interferometric pairs for the stacking operation. Subsequently, the average coherence coefficient was computed for each interferometric pair in the study area.
As an example, the coherence plots of the best interferometric pairs for each month in 2018 were plotted and ordered chronologically, as shown in Figure 2. The color close to yellow in the plot indicates good coherence, which means that deformation can be detected, while the colors closer to purple-blue indicate poor coherence, meaning that the deformation detection is weak. It is clear from Figure 2 that coherence starts to weaken in April (Figure 2d), and by July (Figure 2g), the overall coherence is the worst, with most of the study area out of coherence and the deformation being undetectable. However, from October (Figure 2j), coherence starts to improve, and from November to the following April, coherence is generally better. Figure 2m displays the filtered differential interferogram of 20180102–20180114 (master image date–slave image date), while Figure 2n shows the filtered differential interferogram of 20180701–20180713. The coherence of Figure 2m is better than that of Figure 2n, with clear coherence fringes visible within the red frame of the study area. In contrast, most of the area within the red frame of Figure 2n is incoherent, and thus, deformation during that time period cannot be detected.
To confirm this pattern, all of the interferometric pairs from 2017 to 2022 on a connectivity diagram were plotted, as illustrated in Figure 3a. The average coherence coefficient values of each pair are indicated by green or red connecting lines, corresponding to good or poor coherence, respectively. The light blue background clearly indicates that coherence is significantly weaker from May to October each year, compared to other months. From Figure 3b, it can also be found that the coherence coefficient changes periodically, with a cycle of about 1 year. The ANOVA-1 (analysis of variance-1) test was used to test whether the study area coherence varied with season. The study area is divided into four seasons a year, with spring from March to May, summer from June to August, autumn from September to November, and winter from November to February. All interferograms were divided into 4 groups of samples according to time for the ANOVA-1 test. Statistical results (Figure 3c) showed that the coherence of the study area varied significantly with season (p value < 0.001), where the coherence was relatively poor in summer and better in winter.
Based on the above study, the seasonal variation pattern in coherence in the study area can be found.

2.4. The Improved Seasonal Interferometry Stacking-InSAR

The stacking-InSAR was first proposed by the American scholar David T. Sandwell in 1998 [35]. The principle of stacking-InSAR is to calculate the superposition of the interferometric phase using a weighted averaging method, which can obtain the deformation rate of the target region. However, its noise suppression ability is not as good as conventional time-series methods. Nevertheless, stacking-InSAR has fewer processing steps, faster calculation speed, and can include data with lower quality. Therefore, it is more suitable for the rapid identification of potential landslides in the southwest mountainous area [26,27,28,29].
In this study, our focus is on improving the accuracy of stacking-InSAR and addressing the issue of poor coherence in the southwest mountainous region. The correlation between coherence quality and time in the area was analyzed, and a seasonal interferometry stacking-InSAR method was proposed for improvement, as shown in Figure 4. All data processing was performed using GAMMA and MATLAB software. The first step was preprocessing the single look complex (SLC) data from Sentinel-1, which involved cropping and coregistration. The data were then combined into interferometric pairs based on the time-space baseline threshold, with the temporal baseline threshold set at 24 days and the spatial baseline range set between −250 m and 250 m to maintain detectability. The external SRTM-DEM was used for differential interferometry, followed by adaptive filtering. A higher coherence threshold was set to create a phase unwrapping mask for the first unwrapping, where only points with higher coherence were used. Next, ground control points (GCPs) were selected among these points for refinement to remove the orbital error. The refined baseline information was then imported for differential interferometry, filtering, and phase unwrapping. Coherence calculation was performed to eliminate poorly coherent interferometric pairs in the summer of each year. Finally, the stacking process was performed to calculate the displacement velocity. The deforming slopes were identified based on a displacement velocity map. Figure 5 shows that the yellow bars represent the interferometric pairs participating in the stacking calculation, while the purple color represents the eliminated pairs, with the poor-quality eliminated pairs concentrated in the summer of each year.

2.5. Potential Landslide Boundary Delineation

The boundary of potential landslides was primarily established through the integration of displacement information derived from improved stacking-InSAR, satellite optical imagery, and LiDAR data. To begin with, the slopes exhibiting displacement in the study area were delineated based on the available displacement information. Subsequently, satellite optical imagery was employed to refine and define the boundaries of potential landslides. This process followed the criteria utilized by geomorphologists for identifying landslides from aerial photographs, which involved visually analyzing and interpreting various characteristics. These characteristics included shape, size, color, tone, mottling, texture, and pattern of individual or clustered features visible in satellite imagery (as described in references [36,37]).
In addition, airborne LiDAR point cloud data played a crucial role in the analysis. They aided in vegetation removal, ground point classification, and the generation of a high-resolution digital elevation model (HRDEM) [38]. By generating a hillshade from the HRDEM [39], it became easier to identify morphological features associated with landslides, such as scarps, mobilized material, and foot areas [40]. By integrating the methods mentioned above, the boundary for potential landslides was ultimately determined.

3. Results

Improved seasonal interferometry stacking-InSAR identified a total of 21 potential landslides in and around the study area, as depicted in Figure 6. The displacement signal was evident, allowing for the potential landslide boundaries to be clearly outlined on the displacement velocity map. The potential landslides are concentrated primarily on the southeast bank of Minjiang River, in close proximity to Longmen Mountain.
Additionally, satellite remote sensing images and LiDAR data were compared and analyzed with the potential landslides identified via improved seasonal interferometry stacking-InSAR. Figure 7 illustrates some typical identified potential landslides within the study area. By combining remote sensing images and LiDAR data, we could further accurately delineate the boundaries of these potential landslides based on their geomorphic characteristics. L-01 and L-02 are two adjacent potential landslides, and their feet and scarps can be identified in the LiDAR data (Figure 7c). The gully on the right side of L-06 can be detected in both the optical remote sensing image and LiDAR data, and its scarp is clearly visible in the LiDAR data (Figure 7e,f). The scarps of L-08 and L-09 can be identified in the LiDAR data, while the feet of L-10-L-14 can also be clearly recognized in the LiDAR data (Figure 7i). The potential landslide boundaries delineated based on these geomorphic features closely align with the potential landslides identified via improved seasonal interferometry stacking-InSAR. L-08~L-14 are potential landslides that are closely situated and span over 5 km in length. Their displacement areas are enormous, with each exceeding 0.4 km2, and thus they require special attention.
It is important to note that the boundaries of potential landslides depicted using stacking-InSAR, optical satellite imagery, and LiDAR data may exhibit differences. The criteria are described in Section 2.5. The use of InSAR primarily captures the boundaries of the deformation area, which may not accurately represent the precise boundaries of the entire landslide in many cases (L-01, L-08, L-09, etc.). Optical satellite imagery mainly highlights areas experiencing current or historical displacement, but it may not be very accurate in densely vegetated areas or when landslide geomorphic features are not prominent. On the other hand, the LiDAR method directly utilizes hillshade maps to identify landslide geomorphology, making it easier to recognize ancient landslides or those with distinctive geomorphic features.

4. Discussion

It can be observed that the interferometric coherence in the study area is closely related to temporal variations, as shown in Figure 2 and Figure 3. During the summer, the coherence is the poorest, with most areas exhibiting incoherent signals. If the interferograms from summer were included in the stacking calculation, it would introduce a significant amount of noise. In order to validate the effectiveness of the improved seasonal interferometry stacking-InSAR in noise suppression, we compared it with the traditional stacking-InSAR method.
Figure 8 compares the results obtained from conventional stacking and improved seasonal interferometry stacking-InSAR. Both methods are capable of identifying potential landslides with large areas, such as L-01, L-02, and L-06. However, the improved method is significantly better than the conventional method in terms of the noise suppression effect. As shown in the areas circled in white in Figure 8a,d, this discrete speckle is due to the phase change caused by the decoherence noise, which is significantly different from the real landslide deformation in geometry and distribution, and the improved method can effectively suppress this noise (Figure 8b,e). Due to its small area, the deformation information of L-05 is masked by noise in the result processed via conventional stacking (Figure 8a) and cannot be distinguished, while the deformation boundary of L-05 is very obvious in the results processed using the improved method and can be easily identified.
Figure 8g shows the deformation information of L-16 identified using the conventional method, and Figure 8h displays the deformation information of L-16 identified using the improved method. It is evident that the improved method detects a clearer landslide deformation signal with a more accurate boundary, which is consistent with the boundary identified by the optical image in Figure 8i. Moreover, in Figure 8a,b, L-03 and L-04 exhibit clearer and more obvious deformation signals using the improved method. On the other hand, the improved method can detect the deformation signal of L-17 with clear boundaries.
L-17, L-18, and L-19 are similar to L-05 in that they are all masked by noise using the conventional method. However, these potential landslides can be detected using the improved method. In Figure 8j, the deformation signal of L-17 is completely obscured by noise and cannot be identified using the conventional method. In the satellite image (Figure 8l), it can be observed that L-17 is a potential landslide located on an upper slope. The deformation area of this type of landslide is concentrated in the upper part of the slope, and the area of deformation is relatively small and hidden, making it difficult to detect but extremely hazardous. L-18 and L-19 are two neighboring river bank potential landslides. Using the conventional method, the deformation signals of these landslides are confused with noise points (Figure 8m), making it difficult to accurately discern their boundaries. In contrast, the improved method removes the noise points around them and clearly detects their deformation signals.
To further verify the enhancement effect, four typical areas were selected for the statistical analysis of the conventional method and the improvement. The Mann–Whitney U test was used to demonstrate the statistical significance of the observed differences [41]. The p-value for each case was <0.001, indicating significant differences. Figure 9 shows the statistical analysis between the convention and improvement of the four areas. It can be found that the distribution of each area after the improved method is more concentrated than the conventional method (Figure 9a3–c3), which can reflect a certain suppression effect on noise. Also, their standard deviations were calculated and used to quantify the improvement effect. In area I, the STD was reduced by 46% from 1.735 to 0.938 rad/a, and the improvement was also visually significant (Figure 9a1–a3). In area II, the STD was reduced by 22% from 1.297 to 1.013 rad/a. The STD was reduced by 10% from 2.832 to 2.562 rad/a of area III. In area IV, the STD was reduced by 14% from 2.551 to 2.199 rad/a.
The method proposed in this study, which employs improved seasonal interferometry stacking-InSAR, offers a rapid means of identifying widespread potential landslides in the mountainous regions of southwestern China. Nevertheless, it is essential to acknowledge that this method does have certain limitations. Accurately delineating the precise boundaries of landslides remains a rigorous task that necessitates field investigations and analysis by geological experts. Field surveys and detailed examinations are crucial for validating and refining the identified potential landslide boundaries. Furthermore, it should be noted that the method may not detect potential landslides that experience displacement exclusively during the summer season. As the proposed approach relies on seasonal interferometry, any landslides that undergo displacement outside of the specified time period might not be detected using this method.

5. Conclusions

In this study, an improved seasonal interferometry stacking-InSAR method was proposed for identifying potential landslides in the river valley section from Wenchuan County to Mao County. The method was based on the analysis of the temporal distribution pattern of coherence of Sentinel-1 data. The results of the proposed improved method and the conventional stacking-InSAR method were compared. This study led to the following conclusions: (1) In our analysis of Sentinel-1 data for differential interference in the southwestern mountain vegetation cover region, we found that the quality of coherence was highly time-dependent. Specifically, we observed weaker coherence in May–October each year, compared to other months. The coherence was the worst during July–August, with most of the decoherence signals in the study area, and relatively weaker displacement detection ability in winter. (2) Both the conventional stacking-InSAR and the improved seasonal interferometry stacking-InSAR can detect potential landslides with large displacement areas in the study area. However, potential landslides with small areas located in low-coherence regions cannot be identified using the conventional method. The improved method, on the other hand, effectively reduces noise and detects such potential landslides. From the quantitative statistics, it can be found that the standard deviations of the four tested areas were reduced by 46%, 22%, 10%, and 14%, respectively.
As demonstrated above, the improved seasonal interferometry stacking-InSAR has several advantages in identifying potential landslides in the low-coherence area of the southwest mountains. These advantages include high efficiency and good noise suppression, which provide important technical support for fast and accurate InSAR wide-area identification of potential landslides in high and steep mountainous areas.

Author Contributions

Conceptualization, Z.L. and K.D.; methodology, Z.L., J.D., C.L. and G.T.; visualization, Z.L., J.D. and T.Y.; writing—original draft, Z.L.; writing—review and editing; K.D., Z.L. and X.S. All authors have read and agreed to the published version of the manuscript.

Funding

The work was supported by the fellowship of China Postdoctoral Science Foundation (2020M673322), Sichuan Science Foundation for Outstanding Youth (2023NSFSC1909), the National Natural Science Foundation of China (grant no. 41801391), the State Key Laboratory of Geohazard Prevention and Geoenvironment Protection Independent Research Project (SKLGP2020Z012), and Open Research Fund Program of MNR Key Laboratory for Geo-Environmental Monitoring of Great Bay Area (20220006).

Data Availability Statement

The Sentinel-1 data used in this study are downloaded from the European Space Agency (ESA) through the ASF Data Hub website https://vertex.daac.asf.alaska.edu. The DEM data used in the study is available at https://earthexplorer.usgs.gov/. The optical remote sensing imagery used in this study is available at https://earth.google.com/web/.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Yin, Y.; Wang, F.; Sun, P. Landslide Hazards Triggered by the 2008 Wenchuan Earthquake, Sichuan, China. Landslides 2009, 6, 139–152. [Google Scholar] [CrossRef]
  2. Huang, R.; Li, W. Formation, Distribution and Risk Control of Landslides in China. J. Rock Mech. Geotech. Eng. 2011, 3, 97–116. [Google Scholar] [CrossRef] [Green Version]
  3. Lin, Q.; Wang, Y. Spatial and Temporal Analysis of a Fatal Landslide Inventory in China from 1950 to 2016. Landslides 2018, 15, 2357–2372. [Google Scholar] [CrossRef]
  4. Chen, X.L.; Zhou, Q.; Ran, H.; Dong, R. Earthquake-Triggered Landslides in Southwest China. Nat. Hazards Earth Syst. Sci. 2012, 12, 351–363. [Google Scholar] [CrossRef] [Green Version]
  5. Scaringi, G.; Fan, X.; Xu, Q.; Liu, C.; Ouyang, C.; Domènech, G.; Yang, F.; Dai, L. Some Considerations on the Use of Numerical Methods to Simulate Past Landslides and Possible New Failures: The Case of the Recent Xinmo Landslide (Sichuan, China). Landslides 2018, 15, 1359–1375. [Google Scholar] [CrossRef]
  6. Fan, X.; Xu, Q.; Scaringi, G.; Dai, L.; Li, W.; Dong, X.; Zhu, X.; Pei, X.; Dai, K.; Havenith, H.-B. Failure Mechanism and Kinematics of the Deadly June 24th 2017 Xinmo Landslide, Maoxian, Sichuan, China. Landslides 2017, 14, 2129–2146. [Google Scholar] [CrossRef]
  7. Dai, K.; Xu, Q.; Li, Z.; Tomás, R.; Fan, X.; Dong, X.; Li, W.; Zhou, Z.; Gou, J.; Ran, P. Post-Disaster Assessment of 2017 Catastrophic Xinmo Landslide (China) by Spaceborne SAR Interferometry. Landslides 2019, 16, 1189–1199. [Google Scholar] [CrossRef] [Green Version]
  8. Fan, X.; Xu, Q.; Alonso-Rodriguez, A.; Subramanian, S.S.; Li, W.; Zheng, G.; Dong, X.; Huang, R. Successive Landsliding and Damming of the Jinsha River in Eastern Tibet, China: Prime Investigation, Early Warning, and Emergency Response. Landslides 2019, 16, 1003–1020. [Google Scholar] [CrossRef]
  9. Zhang, L.; Xiao, T.; He, J.; Chen, C. Erosion-Based Analysis of Breaching of Baige Landslide Dams on the Jinsha River, China, in 2018. Landslides 2019, 16, 1965–1979. [Google Scholar] [CrossRef]
  10. Feng, W.; Zhang, G.; Bai, H.; Zhou, Y.; Xu, Q.; Zheng, G. A Preliminary Analysis of the Formation Mechanism and Development Tendency of the Huge Baige Landslide in Jinsha River on October 11, 2018. J. Eng. Geol. 2019, 27, 415–425. [Google Scholar]
  11. Zhao, B.; Zhang, H.; Hongjian, L.; Li, W.; Su, L.; He, W.; Zeng, L.; Qin, H.; Dhital, M.R. Emergency Response to the Reactivated Aniangzhai Landslide Resulting from a Rainstorm-Triggered Debris Flow, Sichuan Province, China. Landslides 2021, 18, 1115–1130. [Google Scholar] [CrossRef]
  12. Zhu, L.; He, S.; Qin, H.; He, W.; Zhang, H.; Zhang, Y.; Jian, J.; Li, J.; Su, P. Analyzing the Multi-Hazard Chain Induced by a Debris Flow in Xiaojinchuan River, Sichuan, China. Eng. Geol. 2021, 293, 106280. [Google Scholar] [CrossRef]
  13. Dai, K.; Li, Z.; Xu, Q.; Tomas, R.; Li, T.; Jiang, L.; Zhang, J.; Yin, T.; Wang, H. Identification and Evaluation of the High Mountain Upper Slope Potential Landslide Based on Multi-Source Remote Sensing: The Aniangzhai Landslide Case Study. Landslides 2023, 20, 1405–1417. [Google Scholar] [CrossRef]
  14. Hooper, A.; Bekaert, D.; Spaans, K.; Arıkan, M. Recent Advances in SAR Interferometry Time Series Analysis for Measuring Crustal Deformation. Tectonophysics 2012, 514–517, 1–13. [Google Scholar] [CrossRef]
  15. Dai, K.; Li, Z.; Xu, Q.; Burgmann, R.; Milledge, D.G.; Tomas, R.; Fan, X.; Zhao, C.; Liu, X.; Peng, J.; et al. Entering the Era of Earth Observation-Based Landslide Warning Systems: A Novel and Exciting Framework. IEEE Geosci. Remote Sens. Mag. 2020, 8, 136–153. [Google Scholar] [CrossRef] [Green Version]
  16. Sun, Q.; Zhang, L.; Ding, X.L.; Hu, J.; Li, Z.W.; Zhu, J.J. Slope Deformation Prior to Zhouqu, China Landslide from InSAR Time Series Analysis. Remote Sens. Environ. 2015, 156, 45–57. [Google Scholar] [CrossRef]
  17. Yin, Y.; Zheng, W.; Liu, Y.; Zhang, J.; Li, X. Integration of GPS with InSAR to Monitoring of the Jiaju Landslide in Sichuan, China. Landslides 2010, 7, 359–365. [Google Scholar] [CrossRef]
  18. Liu, X.; Zhao, C.; Zhang, Q.; Lu, Z.; Li, Z.; Yang, C.; Zhu, W.; Liu-Zeng, J.; Chen, L.; Liu, C. Integration of Sentinel-1 and ALOS/PALSAR-2 SAR Datasets for Mapping Active Landslides along the Jinsha River Corridor, China. Eng. Geol. 2021, 284, 106033. [Google Scholar] [CrossRef]
  19. Dai, K.; Li, Z.; Tomás, R.; Liu, G.; Yu, B.; Wang, X.; Cheng, H.; Chen, J.; Stockamp, J. Monitoring Activity at the Daguangbao Mega-Landslide (China) Using Sentinel-1 TOPS Time Series Interferometry. Remote Sens. Environ. 2016, 186, 501–513. [Google Scholar] [CrossRef] [Green Version]
  20. Dai, K.; Feng, Y.; Zhuo, G.; Tie, Y.; Deng, J.; Balz, T.; Li, Z. Applicability Analysis of Potential Landslide Identification by InSAR in Alpine-Canyon Terrain—Case Study on Yalong River. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2022, 15, 2110–2118. [Google Scholar] [CrossRef]
  21. Rosi, A.; Tofani, V.; Tanteri, L.; Tacconi Stefanelli, C.; Agostini, A.; Catani, F.; Casagli, N. The New Landslide Inventory of Tuscany (Italy) Updated with PS-InSAR: Geomorphological Features and Landslide Distribution. Landslides 2018, 15, 5–19. [Google Scholar] [CrossRef] [Green Version]
  22. Xu, Q.; Guo, C.; Dong, X.; Li, W.; Lu, H.; Fu, H.; Liu, X. Mapping and Characterizing Displacements of Landslides with InSAR and Airborne LiDAR Technologies: A Case Study of Danba County, Southwest China. Remote Sens. 2021, 13, 4234. [Google Scholar] [CrossRef]
  23. Berardino, P.; Fornaro, G.; Lanari, R.; Sansosti, E. A New Algorithm for Surface Deformation Monitoring Based on Small Baseline Differential SAR Interferograms. IEEE Trans. Geosci. Remote Sens. 2002, 40, 2375–2383. [Google Scholar] [CrossRef] [Green Version]
  24. Ferretti, A.; Prati, C.; Rocca, F. Nonlinear Subsidence Rate Estimation Using Permanent Scatterers in Differential SAR Interferometry. IEEE Trans. Geosci. Remote Sens. 2000, 38, 2202–2212. [Google Scholar] [CrossRef] [Green Version]
  25. Ferretti, A.; Prati, C.; Rocca, F. Permanent Scatterers in SAR Interferometry. IEEE Trans. Geosci. Remote Sens. 2001, 39, 8–20. [Google Scholar] [CrossRef]
  26. Osmanoğlu, B.; Sunar, F.; Wdowinski, S.; Cabral-Cano, E. Time series analysis of InSAR data: Methods and trends. ISPRS J. Photogramm. Remote Sens. 2016, 115, 90–102. [Google Scholar] [CrossRef]
  27. Xu, Y.; Li, T.; Tang, X.; Fan, H.; Wang, Y. Research on the Applicability of DInSAR, Stacking-InSAR and SBAS-InSAR for Mining Region Subsidence Detection in the Datong Coalfield. Remote Sens. 2022, 14, 3314. [Google Scholar] [CrossRef]
  28. Zhang, L.; Dai, K.; Deng, J.; Ge, D.; Liang, R.; Li, W.; Xu, Q. Identifying Potential Landslides by Stacking-InSAR in Southwestern China and Its Performance Comparison with SBAS-InSAR. Remote Sens. 2021, 13, 3662. [Google Scholar] [CrossRef]
  29. Liang, J.; Dong, J.; Zhang, S.; Zhao, C.; Liu, B.; Yang, L.; Yan, S.; Ma, X. Discussion on InSAR Identification Effectivity of Potential Landslides and Factors That Influence the Effectivity. Remote Sens. 2022, 14, 1952. [Google Scholar] [CrossRef]
  30. Torres, R.; Snoeij, P.; Geudtner, D.; Bibby, D.; Davidson, M.; Attema, E.; Potin, P.; Rommen, B.; Floury, N.; Brown, M. GMES Sentinel-1 Mission. Remote Sens. Environ. 2012, 120, 9–24. [Google Scholar] [CrossRef]
  31. Cao, C.; Zhu, K.; Song, T.; Bai, J.; Zhang, W.; Chen, J.; Song, S. Comparative study on potential landslide identification with ALOS-2 and sentinel-1A data in heavy forest reach, upstream of the Jinsha River. Remote Sens. 2022, 14, 1962. [Google Scholar] [CrossRef]
  32. Zebker, H.A.; Villasenor, J. Decorrelation in Interferometric Radar Echoes. IEEE Trans. Geosci. Remote Sens. 1992, 30, 950–959. [Google Scholar] [CrossRef] [Green Version]
  33. Simons, M.; Rosen, P.A. Interferometric Synthetic Aperture Radar Geodesy. Geodesy 2007, 3, 391–446. [Google Scholar]
  34. Touzi, R.; Lopes, A.; Bruniquel, J.; Vachon, P.W. Coherence estimation for SAR imagery. IEEE Trans. Geosci. Remote Sens. 1999, 37, 135–149. [Google Scholar] [CrossRef] [Green Version]
  35. Sandwell, D.T.; Price, E.J. Phase gradient approach to stacking interferograms. J. Geophys. Res. Solid Earth 1998, 103, 30183–30204. [Google Scholar] [CrossRef] [Green Version]
  36. Weirich, F.; Blesius, L. Comparison of satellite and air photo based landslide susceptibility maps. Geomorphology 2007, 87, 352–364. [Google Scholar] [CrossRef]
  37. Fiorucci, F.; Cardinali, M.; Carlà, R.; Rossi, M.; Mondini, A.; Santurri, L.; Ardizzone, F.; Guzzetti, F. Seasonal landslide mapping and estimation of landslide mobilization rates using aerial and satellite images. Geomorphology 2011, 129, 59–70. [Google Scholar] [CrossRef]
  38. Eeckhaut, M.V.D.; Poesen, J.; Verstraeten, G.; Vanacker, V.; Nyssen, J.; Moeyersons, J.; Beek, L.V.; Vandekerckhove, L. Use of LIDAR-derived images for mapping old landslides under forest. Earth Surf. Process. Landf. 2007, 32, 754–769. [Google Scholar] [CrossRef]
  39. Van Den Eeckhaut, M.; Poesen, J.; Verstraeten, G.; Vanacker, V.; Moeyersons, J.; Nyssen, J.; Van Beek, L.P.H. The effectiveness of hillshade maps and expert knowledge in mapping old deep-seated landslides. Geomorphology 2005, 67, 351–363. [Google Scholar] [CrossRef]
  40. Jaboyedoff, M.; Oppikofer, T.; Abellán, A.; Derron, M.H.; Loye, A.; Metzger, R.; Pedrazzini, A. Use of LIDAR in landslide investigations: A review. Nat. Hazards 2012, 61, 5–28. [Google Scholar] [CrossRef] [Green Version]
  41. Mann, H.B.; Whitney, D.R. On a test of whether one of two random variables is stochastically larger than the other. Ann. Math. Stat. 1947, 18, 50–60. [Google Scholar] [CrossRef]
Figure 1. (a) Geographic information of the study area. (b) Location of study area. (c) Three-dimensional scenes of the study area based on Google Earth. (df) Photos of slopes in study area.
Figure 1. (a) Geographic information of the study area. (b) Location of study area. (c) Three-dimensional scenes of the study area based on Google Earth. (df) Photos of slopes in study area.
Remotesensing 15 03278 g001
Figure 2. (al) Interferometric coherence diagrams of the study area by month. (m) Interferogram of summer. (n) Interferogram of winter. The digits at the bottom right of each picture represent the dates of master image and slave image.
Figure 2. (al) Interferometric coherence diagrams of the study area by month. (m) Interferogram of summer. (n) Interferogram of winter. The digits at the bottom right of each picture represent the dates of master image and slave image.
Remotesensing 15 03278 g002
Figure 3. (a) Interferometric pairs connection diagram and time distribution of coherence. (b) Coherence coefficient time-series plot. (c) Coherence coefficient of different seasons box plot.
Figure 3. (a) Interferometric pairs connection diagram and time distribution of coherence. (b) Coherence coefficient time-series plot. (c) Coherence coefficient of different seasons box plot.
Remotesensing 15 03278 g003
Figure 4. Flow of improved seasonal interferometry stacking-InSAR.
Figure 4. Flow of improved seasonal interferometry stacking-InSAR.
Remotesensing 15 03278 g004
Figure 5. Interferometric pairs preference statistics chart.
Figure 5. Interferometric pairs preference statistics chart.
Remotesensing 15 03278 g005
Figure 6. Identification of potential landslides based on improved seasonal interferometry stacking-InSAR.
Figure 6. Identification of potential landslides based on improved seasonal interferometry stacking-InSAR.
Remotesensing 15 03278 g006
Figure 7. Improved seasonal interferometry stacking-InSAR results (the left), optical remote sensing image results (the mid), and the identification results of LiDAR data (the right). (ac) Stacking identification results, optical remote sensing image, and LiDAR data of L-01 and L-02. (df) Stacking identification results, optical remote sensing image, and LiDAR data of L-06. (gi) Stacking identification results, optical remote sensing image, and LiDAR data of L-08–L-14.
Figure 7. Improved seasonal interferometry stacking-InSAR results (the left), optical remote sensing image results (the mid), and the identification results of LiDAR data (the right). (ac) Stacking identification results, optical remote sensing image, and LiDAR data of L-01 and L-02. (df) Stacking identification results, optical remote sensing image, and LiDAR data of L-06. (gi) Stacking identification results, optical remote sensing image, and LiDAR data of L-08–L-14.
Remotesensing 15 03278 g007
Figure 8. Comparison of improved seasonal interferometry stacking-InSAR using the conventional stacking method. The rows from left to right are as follows: conventional method, improved seasonal interferometry stacking-InSAR method, and satellite remote sensing images. (af) The improved method is significantly better than the conventional method in terms of noise suppression effect. (gi) The improved method can detect a clearer landslide deformation signal with a more accurate boundary. (jo) The improved method can detect potential landslides which were not detected using the conventional method in low-coherence areas.
Figure 8. Comparison of improved seasonal interferometry stacking-InSAR using the conventional stacking method. The rows from left to right are as follows: conventional method, improved seasonal interferometry stacking-InSAR method, and satellite remote sensing images. (af) The improved method is significantly better than the conventional method in terms of noise suppression effect. (gi) The improved method can detect a clearer landslide deformation signal with a more accurate boundary. (jo) The improved method can detect potential landslides which were not detected using the conventional method in low-coherence areas.
Remotesensing 15 03278 g008
Figure 9. Statistical analysis between convention and improvement of 4 test areas. (a1d1) Displacement velocity map based on conventional method of the 4 areas, respectively. (a2d2) Displacement velocity map based on improved method of the 4 areas, respectively. (a3d3) Histogram comparison of displacement velocity between convention and improvement in 4 areas, respectively, and standard deviation of them.
Figure 9. Statistical analysis between convention and improvement of 4 test areas. (a1d1) Displacement velocity map based on conventional method of the 4 areas, respectively. (a2d2) Displacement velocity map based on improved method of the 4 areas, respectively. (a3d3) Histogram comparison of displacement velocity between convention and improvement in 4 areas, respectively, and standard deviation of them.
Remotesensing 15 03278 g009
Table 1. Sentinel-1 dataset parameter information.
Table 1. Sentinel-1 dataset parameter information.
ParameterValue
Temporal coverage2017.11–2023.02
Number of images155
Orbital directionAscending
Wavelength5.6 cm
Azimuth/Range pixel spacing13.95 m/2.33 m
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Li, Z.; Dai, K.; Deng, J.; Liu, C.; Shi, X.; Tang, G.; Yin, T. Identifying Potential Landslides in Steep Mountainous Areas Based on Improved Seasonal Interferometry Stacking-InSAR. Remote Sens. 2023, 15, 3278. https://doi.org/10.3390/rs15133278

AMA Style

Li Z, Dai K, Deng J, Liu C, Shi X, Tang G, Yin T. Identifying Potential Landslides in Steep Mountainous Areas Based on Improved Seasonal Interferometry Stacking-InSAR. Remote Sensing. 2023; 15(13):3278. https://doi.org/10.3390/rs15133278

Chicago/Turabian Style

Li, Zhiyu, Keren Dai, Jin Deng, Chen Liu, Xianlin Shi, Guangmin Tang, and Tao Yin. 2023. "Identifying Potential Landslides in Steep Mountainous Areas Based on Improved Seasonal Interferometry Stacking-InSAR" Remote Sensing 15, no. 13: 3278. https://doi.org/10.3390/rs15133278

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop