Remotely Sensed Freeze-Thaw from the Soil Moisture Active Passive Instrument to Inform the Timing of Seasonal Load Restrictions in Alaska

Abstract

To balance the competing interests of minimizing road damage while maximizing economic opportunities on low-volume roadways located in seasonal frost areas, road management agencies typically apply spring load restrictions (SLR). The Alaska Department of Transportation and Public Facilities (ADOT&PF) uses an observational approach to determine SLR, which utilizes temperature data probes (TDP). Although their TDP network is fairly extensive, comprehensive coverage of all roads within the ADOT&PF domain is not feasible. Passive microwave spaceborne freeze and thaw (FT) observations may be useful for bridging these observational gaps. This study compared satellite FT retrievals with TDP data at 15 sites in Alaska to determine whether satellite-retrieved FT can inform on road FT; whether there are substantial morning (AM) and afternoon (PM) satellite-retrieved FT differences; and assess how Soil Moisture Active Passive (SMAP)-based SLR application dates compare with ADOT&PF SLR-applied dates. Our evaluation showed that SMAP AM (PM) FT retrievals are highly representative of FT in roads with an overall accuracy of 79% (83%). SMAP-retrieved FT states corresponded to individual TDP locations 52–95% of the time, and FT accuracy exceeded 80% at 9 (11) out of 15 TDP sites for AM (PM) data. SMAP thaw detection was very consistent with respect to road data: (1) interannual variation of FT accuracy at individual TDP sites was low (16%) and (2) strong SMAP thaw signals were detected about 10-days ahead of all roads having thawed according to TDP data.