| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 4676553 | Cold Regions Science and Technology | 2009 | 9 Pages |
A controlled experiment was conducted at the Cold Region Research and Engineering Laboratory (CRREL) facilities for retrieving freshwater ice characteristics using a Frequency Modulated Continuous Wave (FM-CW) radar system. The purpose of this study is to quantify the interactions between a backscattered radar signal and ice cover characteristics such as thickness and air bubble content; and to provide a comprehensive data set for testing a theoretical radar backscattering model currently under development. The experiment was held in an indoor test basin (36.5 m × 9.4 m × 2.5 m), where an ice sheet was grown from 0 to 24 cm thickness and monitored with a series of radar and geophysical measurements. Backscattering of the ice cover using an in-house FM-CW C-band radar was repeatedly measured at incidence angles varying from 0 to 50° for four polarisation combinations, HH, VV, HV, and VH. Ancillary data, such as ice thickness, crystalline structure and air bubble content, air temperature and hoarfrost water equivalent were collected to support the analysis and interpretation of the radar data.Analysis of the radar backscattered signal indicates that the received power is notably affected by ice thickness. For example, the HH/HV ratio of power signals correlated with ice thickness between 8 and 24 cm with R2 value ranging from 0.75 to 0.90. It was shown through modeling of the radar reflectivity that the lack of correlation at less than 8 cm was caused by the coherent addition of the multiple reflections at the air/ice and the ice/water boundaries. These results are encouraging as for the potential of retrieving lake and river ice thickness from satellite SAR sensors operating at C-band frequencies and at 20 to 50° incidence angles, such as Radarsat 1 and 2 and Envisat, although their current radiometric resolution may limit ice thickness retrieval to thicker ice covers. Also, a number of unresolved issues must be addressed before taking full advantage of these sensors for operational retrieving of freshwater ice thickness. For one, additional controlled experiments must be carried out with ice covers embracing a range of surface roughness values. Ice covers with more diversified air inclusions and snow cover characteristics also need to be investigated, which probably can be best done by deploying radar systems next to natural ice covers.
