Monitoring the Degradation of Island Permafrost Using Time-Series InSAR Technique: A Case Study of Heihe, China

Sai Wang, Bing Xu, Wei Shan, Jiancun Shi, Zhiwei Li, Guangcai Feng2019 · DOI

TL;DR

This study uses time-series InSAR to monitor ground deformation in Heihe, China, revealing significant settlement linked to island permafrost degradation. The findings highlight a strong correlation between rising air temperatures and this degradation, posing risks to infrastructure and underscoring the need for continued monitoring in these sensitive cold regions.

What Problem Does This Paper Address?

Island permafrost, found in regions like Northeast China, is particularly sensitive to climate change. Global warming is causing air temperatures to rise, accelerating permafrost degradation. This degradation leads to localized ground deformation, which can be distinct from broader permafrost subsidence and poses significant risks to infrastructure, such as expressways, and can trigger geological disasters like landslides. While permafrost degradation is a known issue, monitoring its localized impacts, especially in island permafrost zones, has been less studied compared to larger permafrost regions. This paper aims to fill this gap by precisely monitoring and analyzing the ground deformation associated with island permafrost degradation in the Heihe area, China, and understanding its relationship with climate change.

Methods

The researchers employed a time-series Interferometric Synthetic Aperture Radar (InSAR) strategy, specifically a combination of Persistent Scatterer InSAR (PS-InSAR) and Small Baseline Subset InSAR (SBAS-InSAR) techniques. This approach is designed to overcome the limitations of traditional differential InSAR, such as decorrelation and atmospheric noise, enabling the monitoring of small deformations over extended periods.

The core methodology involved:

  • Data Acquisition: Utilizing a stack of 20 L-band ALOS/PALSAR stripmap images acquired between June 2007 and December 2010. The L-band wavelength (approximately 23.6 cm) was chosen for its reduced susceptibility to temporal decorrelation caused by dense vegetation and active layer moisture changes.
  • Pre-processing: Co-registration of all SLC images using GAMMA software, with a reference image selected to minimize decorrelation. Topographic effects were removed using a SRTM Digital Elevation Model (DEM).
  • Interferogram Generation: Creating interferograms by conjugate multiplication of co-registered master and slave SLC images. Topographic phase contribution was simulated and removed using the DEM.
  • Point Target (PT) Selection: A crucial step to ensure reliable deformation measurements. A combination of the Spectral Diversity (SD) and Amplitude Dispersion (DA) methods was used to select highly coherent point targets, mitigating issues with unreliable PTs in low-coherence areas.
  • Phase Smoothing and Unwrapping: Applying a low-pass adaptive phase filter before phase unwrapping using a network programming-based method.
  • Deformation Time-Series Retrieval: Implementing the SBAS-InSAR method to obtain the deformation time-series and mean deformation velocity. A triangular filter was used for atmospheric screen estimation.
  • Analysis: Correlating the derived deformation with topographic factors (altitude, slope) and meteorological data (air temperature).

Datasets

  • ALOS/PALSAR Data: A total of 20 L-band ALOS/PALSAR stripmap images acquired from June 2007 to December 2010 over the Heihe area, China. These images were acquired with Fine Beam Single (FBS) or Fine Beam Double (FBD) polarization modes.
  • SRTM DEM: A 1-arc-second resolution (approximately 30m x 30m) Shuttle Radar Topography Mission Digital Elevation Model was used for topographic correction.
  • Meteorological Data: Air temperature data from the Sunwu County Meteorological Station from 1980 to 2018 were used to analyze the correlation with permafrost degradation.
  • Landsat Images: Used for historical change analysis and identifying island permafrost areas.

Key Insights and Ideas

  • Localized Deformation Hotspots: The study identified distinct areas of significant ground deformation, particularly associated with island permafrost, rather than widespread, uniform subsidence.
  • Altitude Dependence: Permafrost degradation and subsequent deformation are strongly concentrated in lower-altitude valleys and flat areas, aligning with the known distribution of island permafrost.
  • Slope Aspect Matters: Deformation is more prevalent on slopes with less solar radiation (north, northwest, northeast slopes), indicating that shadow and reduced solar input play a role in permafrost stability.
  • Temperature as a Primary Driver: A robust negative correlation was found between rising air temperatures and permafrost deformation, suggesting that warming is the primary factor driving degradation and settlement.
  • Infrastructure Vulnerability: The study highlights the direct threat to infrastructure, such as the Bei'an-Heihe expressway, which has experienced instability and required realignment due to permafrost degradation.
  • Human Activity Amplification: Changes in land use, such as the conversion of forest to agricultural land, are identified as factors that can exacerbate permafrost degradation.

Key Results

  • Deformation Rates: The time-series InSAR analysis revealed deformation rates along the line of sight ranging from -70 to 70 mm/a in the Bei'an-Heihe area between 2007 and 2010.
  • Significant Deformation: Most of the island permafrost areas exhibited significant deformation, substantially exceeding 35 mm/a, indicating active degradation.
  • Correlation with Temperature: A strong negative correlation (average value of -0.80, ranging from -0.78 to -0.85) was found between the linear trend of deformation and the linear trend of temperature increase.
  • Altitude of Maximum Deformation: The highest probability of significant deformation was observed at altitudes between 210-300 m, with a peak around 250 m.
  • Deformation Lag Time: The deformation fitting curves indicated a lag time of approximately 25 days between temperature changes and ground deformation response.

Limitations and Future Work

  • Limitations:
    • The study acknowledges that dense vegetation cover and seasonal changes in active layer moisture can lead to temporal decorrelation, limiting InSAR application. However, the L-band data helped mitigate this.
  • The analysis focused primarily on negative deformation (settlement), with positive deformation (uplift) being briefly discussed but not deeply analyzed for simplicity.
  • While human activities are mentioned as exacerbating factors, their precise quantitative impact on deformation was not the primary focus.
  • Future Work:
    • The authors suggest further improving deformation observation by combining more abundant data (in space and time).
  • Establishing a more comprehensive understanding of the interaction between climate change and permafrost degradation is recommended.
  • Future research should continue to monitor tundra in the Heihe area due to its sensitivity and the ongoing impacts of permafrost degradation.

Summary updated

BibTeX
@misc{wang2019monitoring,
  author = {Sai Wang and Bing Xu and Wei Shan and Jiancun Shi and Zhiwei Li and Guangcai Feng},
  title = {{Monitoring the Degradation of Island Permafrost Using Time-Series InSAR Technique: A Case Study of Heihe, China}},
  year = {2019},
  doi = {10.3390/s19061364},
}
Figure