Broadband Radiometry for Snowpack Travel Time Measurement

Project Overview
Project Abstract: 
This project will demonstrate a remote sensing technology for the direct retrieval of snow water equivalent (SWE) of a snow pack. Broadband radiometry (GHz bandwidth) has promise for the recovery of travel times in layered low-loss and non-scattering media. Broadband microwave noise emitted by the soil under a snow pack travels upwards through the snow pack, and interferes with itself as it also reflects off first the upper and then lower snow pack interfaces before continuing to the point of observation. This selfinterference manifests itself as a spike in the time-domain autocorrelation of the noise, or as a sinusoidal amplitude modulation of narrowband brightness over the spectrum. The delay in the spike, or equivalently the period (in Hz) of the amplitude modulation reveals the RF travel time through the slab of snow pack. This travel time is directly proportional to the sum of the depth of the snow pack and the SWE. Because this is a direct measurement of SWE (as opposed to indirect methods such as those relying upon scatter darkening) we expect that the variability of travel times within a single measurement will reflect the variability of depth and SWE within the scene. Dry snow is sufficiently lossless at microwave frequencies that we expect this method to work under a broad range of circumstances. Scattering in the snow pack must be sufficiently small, so a center frequency below X-band is proposed. Bandwidth is inversely proportional to the minimum measurable travel time, and so we propose a bandwidth of more than 1GHz so that snow packs as thin as 15cm can be measured. Unlike active altimeters, this technique works over a broad range of incidence angles. Field-scale observations have thus far not revealed significant radio frequency interference (RFI), and any RFI that does appear can be attacked with a number of RFI excision techniques recently developed for Aquarius, SMAP, etc. The technique should be robust to the presence of at least some RFI, because it is the temporal/spectral characteristics that constitute the observable, rather than the magnitude of the brightness. We will fabricate a radiometer as described above to complement our suite of truck-mounted microwave radiometers, notably a 19- and 37-GHz dual pol radiometers. These truck-mounted radiometers will then be deployed for at least one winter season in a snowy location, such as the University of Michigan Biological Station in Pellston, MI, for continual monitoring of the seasonal snow pack. The observations will be used to validate the performance of the instrument and the retrieval algorithms, and to explore the synergy of this new observable with more traditional (19/37) remote sensing of snow packs. Areas of particular focus will be on the effects of phenomena unique to snow packs, such as internal layers within the snow pack and effects of metamorphism like the formation of depth hoar. The results of this investigation will inform the validation scheme of products of the coming NASA Snow and Cold Lands Processes (SCLP) mission. It may also provide inputs to the instrument suite configuration for the SCLP spacecraft.
Project Details
Research sites: 
Investigator Info
Funding agency: 
Years research project active: 
2014 to 2016