5:15 PM - 6:45 PM
[ACG42-P01] Seasonal surface displacement related to ground freeze-thaw cycles in Fairbanks, interior Alaska, revealed by ALOS-2 InSAR
Keywords:Permafrost, Ground surface displacement, Alaska, Interferometric Synthetic Aperture Radar, ALOS-2
Thawing permafrost due to recent climate changes have impacted on local to global scales (e.g., Biskaborn et al., 2019). However, direct observations of permafrost dynamics are very difficult because of its existence underground. Since ground surface is displaced (uplifted or subsided) by freezing and thawing of frozen soil, the spatial distribution of the displacement has been increasingly studied using Interferometric Synthetic Aperture Radar (InSAR), which is nowadays an essential tool to monitor permafrost dynamics.
The subsurface structure in the permafrost regions can be divided into two major layers: the active layer (seasonally frozen and thawed layer), which is about 1-3 meters below the ground surface, and the permafrost that exists below the active layer. Since the magnitude of surface displacement associated with the freezing and thawing of the active layer is related to its thickness and soil moisture content, a phase change model (e.g., Liu et al., 2012) based on the Stefan's equation for thawing depth is often used in calculations. This model allows the active layer thickness to be calculated from the surface displacement, and the thickness is often compared and discussed with field observation data. However, this model results in zero active layer thickness at locations where the ground surface displacement is zero. Furthermore, since the uplift associated with frost heave is due to formation of ice lenses, the amount of ground surface displacement can be underestimated even if all existing soil water content changes in volume to ice as in the phase change model (Abe et al., 2022). Therefore, it is necessary to construct a physical model based on ground temperature and soil moisture content to explain ground surface displacement associated with freezing and thawing of frozen soil. This study aims to 1) clarify the spatial and temporal variation of seasonal displacement of the ground surface using InSAR, and 2) model the ground surface displacement based on temperature and soil moisture data in frozen soil obtained from field surveys.
Our study area is Poker Flat Research Range (PFRR) in Fairbanks, interior Alaska, managed by University of Alaska Fairbanks. We used the Advanced Land Observing Satellite-2 (ALOS-2) L-band SM1 SAR data (3 m spatial resolution) obtained from 2021 to 2023 to perform InSAR time-series analysis (Berardino et al., 2002; Abe et al., 2022; Abe and Iijima, 2024) for calculating seasonal surface displacement at each year. We also conducted field survey at the beginning of October 2023 to measure surface elevation and thaw depth at the time, and installed equipment for measuring ground temperature, soil water content, and ground surface displacement.
Our analysis resulted in ~10 cm seasonal line-of-sight (LOS) changes in PFRR. The temporal changes showed that seasonal thaw settlement had occurred in June-September/October and subsequent frost upheaval until December. The LOS signals indicating surface uplift during mid-winter (January to March) were also detected in the two consecutive years (2021 and 2022). One possible reason for this is the presence of talik (a layer of unfrozen ground during winter). A recent study has revealed that talik has been widely distributed in Alaska (Farquharson et al., 2022), and a continuous ground temperature in PFRR has recorded that unfrozen layers in winter have been emerging since 2018 (Kobayashi et al., 2023). This suggest that talik may provide soil water for ice lens formation and cause ground surface uplift during mid-winter.
The subsurface structure in the permafrost regions can be divided into two major layers: the active layer (seasonally frozen and thawed layer), which is about 1-3 meters below the ground surface, and the permafrost that exists below the active layer. Since the magnitude of surface displacement associated with the freezing and thawing of the active layer is related to its thickness and soil moisture content, a phase change model (e.g., Liu et al., 2012) based on the Stefan's equation for thawing depth is often used in calculations. This model allows the active layer thickness to be calculated from the surface displacement, and the thickness is often compared and discussed with field observation data. However, this model results in zero active layer thickness at locations where the ground surface displacement is zero. Furthermore, since the uplift associated with frost heave is due to formation of ice lenses, the amount of ground surface displacement can be underestimated even if all existing soil water content changes in volume to ice as in the phase change model (Abe et al., 2022). Therefore, it is necessary to construct a physical model based on ground temperature and soil moisture content to explain ground surface displacement associated with freezing and thawing of frozen soil. This study aims to 1) clarify the spatial and temporal variation of seasonal displacement of the ground surface using InSAR, and 2) model the ground surface displacement based on temperature and soil moisture data in frozen soil obtained from field surveys.
Our study area is Poker Flat Research Range (PFRR) in Fairbanks, interior Alaska, managed by University of Alaska Fairbanks. We used the Advanced Land Observing Satellite-2 (ALOS-2) L-band SM1 SAR data (3 m spatial resolution) obtained from 2021 to 2023 to perform InSAR time-series analysis (Berardino et al., 2002; Abe et al., 2022; Abe and Iijima, 2024) for calculating seasonal surface displacement at each year. We also conducted field survey at the beginning of October 2023 to measure surface elevation and thaw depth at the time, and installed equipment for measuring ground temperature, soil water content, and ground surface displacement.
Our analysis resulted in ~10 cm seasonal line-of-sight (LOS) changes in PFRR. The temporal changes showed that seasonal thaw settlement had occurred in June-September/October and subsequent frost upheaval until December. The LOS signals indicating surface uplift during mid-winter (January to March) were also detected in the two consecutive years (2021 and 2022). One possible reason for this is the presence of talik (a layer of unfrozen ground during winter). A recent study has revealed that talik has been widely distributed in Alaska (Farquharson et al., 2022), and a continuous ground temperature in PFRR has recorded that unfrozen layers in winter have been emerging since 2018 (Kobayashi et al., 2023). This suggest that talik may provide soil water for ice lens formation and cause ground surface uplift during mid-winter.