Monitoring surface deformation over permafrost with an improved SBAS-InSAR algorithm: With emphasis on climatic factors modeling

- We propose an InSAR deformation model that considers temperature and precipitation.
- Strong negative correlation holds between temperature/precipitation and deformation.
- Compensating the time lag between temperature and deformation improves the modeling.

Changes of permafrost not only affect regional and global water circulation, carbon deposit and climate warming, but also influence ground ecological, geophysical, and biogeochemical processes in cold region. Monitoring surface deformation over permafrost will help to understand the dynamics of these processes, assess climate changes, and prevent potential damages to engineering structures underlain by permafrost. This article employs an improved small baseline subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) technique to monitor the surface deformation over the Danxiong-Yangbajing area in the southern Qinghai-Tibet Plateau (QTP), with emphasis on climatic factors modeling. A new deformation model considering both the external (mainly climatic) and internal factors (such as tectonic activities and thermal character of frozen soil) is constructed and introduced into the SBAS-InSAR to estimate the surface deformation over permafrost. The results show that the peak-to-peak annual deformation over natural surface is 3.6-5.0 cm, while that along the Qinghai-Tibet Highway (QTH) and the Qinghai-Tibet Railway (QTR) is 2.8-3.7 cm. We also find that the surface deformation over frozen soil is negatively correlated to air temperature and precipitation, with the correlation being − 0.80 to − 0.45 and − 0.95 to − 0.75, respectively. The root mean squares (RMSs) of the residual (un-modeled) deformation got by the proposed method are much smaller than that by the existing methods. We also successfully examine a time lag of about 60 days between the retrieved maximal subsidence and the highest air temperature in warm seasons (from May to September). After compensating for this time lag, the RMSs of the residual deformation reduce significantly and the correlation between the deformation and the temperature rises considerably, varying from − 0.95 to − 0.65. The method presented in this paper is more suitable to model and analyze the surface deformation in permafrost areas.