Permafrost Dynamics Observatory—Part I: Postprocessing and Calibration Methods of UAVSAR L‐Band InSAR Data for Seasonal Subsidence Estimation

Abstract

We develop and present several calibration and postprocessing methods for seasonal subsidence estimation from InSAR deformation • Novel methods for phase referencing and uncertainty quantification due to nonergodicity within the multilook window are proposed • Residual sources of uncertainty in active layer thickness estimation are discussed and quantified

Validation of Permafrost Active Layer Estimates from Airborne SAR Observations

“…The PDO product uses both L-band InSAR and P-band backscatter to estimate seasonal subsidence, ALT, and soil volumetric water content product at 30 m pixel resolution. By estimating the water content, and assuming a typical subsurface porosity profile, the required ALT to produce the measured subsidence is calculated in a joint inversion framework [12]. Given that the SAR acquisitions are not precisely at the beginning and end of the thaw period, an accumulated degree days of thaw (ADDT) correction is applied to extrapolate the total seasonal subsidence (and therefore ALT) from the measured subsidence, based on the ADDT experienced at each swath.…”

“…Given that the SAR acquisitions are not precisely at the beginning and end of the thaw period, an accumulated degree days of thaw (ADDT) correction is applied to extrapolate the total seasonal subsidence (and therefore ALT) from the measured subsidence, based on the ADDT experienced at each swath. Full details of the postprocessing and joint retrieval can be found in [12]. The ALT product is masked to eliminate pixels with heavy forest cover and InSAR coherence < 0.35 [12].…”

“…Full details of the postprocessing and joint retrieval can be found in [12]. The ALT product is masked to eliminate pixels with heavy forest cover and InSAR coherence < 0.35 [12].…”

“…SAR uncertainty is two to three times larger than GPR due to errors both in the measurement (as a result of having infrequent acquisitions from airborne platforms), as well as assumptions in the conversion from subsidence to ALT. Please refer to [12] for more details on SAR uncertainties.…”

“…Recently, the possibility has arisen of using aircraft-mounted SAR systems to acquire datasets for ALT estimation similar to what had previously been observed using satellites [12]. Although airborne SAR deployments have the disadvantage that they are not placed into Earth orbit, whereas satellites in orbit for long durations can repeat temporal measurements up to every 12 days in the Arctic [13].…”