Quantifying Permafrost Processes and Soil Moisture with Interferometric Phase and Closure Phase PDF Download

Author: Roger John Michaelides
Publisher:
ISBN:
Category :
Languages : en
Pages :

Get Book Here

Book Description
Air temperatures in the Arctic are increasing at twice the global rate, making permafrost regions one of the most vulnerable ecosystems in a changing climate [Jorgenson et al., 2001]. Permafrost, or ground that remains frozen for two or more consecutive years, covers 24% of the Northern Hemisphere and contains 60% of the world's soil carbon [Turetsky et al., 2019a]. Large stores of soil carbon are bound in permafrost, predominantly as carbon dioxide (CO2) and methane (CH4); this bound soil carbon is susceptible to rapid decomposition and release into the atmosphere after thaw [Natali et al., 2019]. As air temperatures rise, permafrost regions experience i) seasonal thawing and freezing, and ii) permanent thaw and loss of frozen ground. These processes modify ecosystems, change land cover and surface hydrologic regimes, and release vast amounts of greenhouse gases into the atmosphere. Due to the amount of permafrost soil carbon susceptible to release into the atmosphere, there is a critical need to monitor permafrost status and vulnerability to change, as well as project future behavior of the permafrost system. The vast spatial extent of permafrost regions and their inaccessibility provides challenges to monitoring efforts. In situ methods of characterizing permafrost processes are spatially sparse, restricting regional studies of permafrost thaw status, and introducing uncertainties into climate models. Remote-sensing techniques are an attractive method for characterizing and monitoring permafrost systems on large scales. Interferometric Synthetic Aperture Radar (InSAR) is a geodetic technique for measuring temporal variations of the surface of the Earth, in which repeated synthetic aperture radar (SAR) images are acquired over a region of interest. These images are then interferometrically combined, and the resulting phase difference between SAR images quantifies surface topography and deformation of the surface of the Earth. InSAR, with its fine spatial resolution and broad coverage, presents an attractive method for regional characterization of permafrost thaw status and active layer thickness at fine resolution. However, in permafrost regions, variations in soil moisture, vegetation, snow cover, and phase changes of pore-bound water and ice all affect the observed deformation and can amplify signal decorrelation. This decorrelation can complicate, and in severe cases preclude, the estimation of surface deformation from InSAR phase observations. In this dissertation, we use the InSAR technique to observe permafrost processes in the discontinuous permafrost zone, with a case study in the Izaviknek Highlands region of the Yukon-Kuskokwim delta in Southwestern Alaska. We measure both centimetric seasonal deformation of permafrost associated with seasonal freeze/thaw processes, as well as long-term, interannual deformation associated with permafrost thaw and degradation. We find significant long-term deformation on the order of centimeters per year associated with a complex of wildfire burns in this region, which we relate to the age of wildfire events, and demonstrate that InSAR successfully captures permafrost dynamics induced by wildfire decades after the original burn. We also introduce a method of quantifying and removing decorrelation phase artifacts from InSAR observations by exploiting closure phase relations within a subset of SAR scenes. We show that decorrelation phase biases on the order of tens of degrees can be successfully characterized and re- moved from the original InSAR signal. Further, we investigate the impact of variable soil moisture on closure phase observations using a new SAR interferometric imaging model that explicitly accounts for signal decorrelation treating scattering surfaces as realizations of stochastic processes. Finally, we construct an algorithm that combines the SAR interferometric imaging model introduced above with direct closure phase observations to estimate changes in surface soil moisture state directly from InSAR phase measurements.