Comparison of ground-based and airborne snow depth measurements with georadar systems, case study

Abstract

Snow depth measurements with the help of georadar (Ground Penetrating Radar, GPR) were investigated for almost two decades in Scandinavia. For the first tests in the early 1980s, results were of poor quality. Later, data quality improved when different systems were developed. In Norway, emphasis was put on ground-based snow measurements of scientific character; few attempts for operational use were undertaken. In Sweden, airborne, operational snow measurements with georadar were performed since 1986. A helicopter, flying near to the ground, was used as platform. The objective of the presented study was to compare results from ground-based and airborne snow measurements. The radar control units used were comparable, but the antenna configuration and frequencies differed. Also radar data interpretation and the conversion of radar signal travel time into snow depth values varied. Measurements were made at common snow courses. The comparison showed in general high correlation between radar results from both methods. Differing results were found for shallow snow and bare areas. Here, the ground-based method indicated zero or close-to-zero snow depth, whereas the airborne method rarely detected zero snow depth. This phenomenon seemed to be connected to the bigger footprint size of the airborne radar and to the different radar data interpretation methods. On average, the airborne measurements indicated shallower snow depths than ground-based measurements, 4% less in open terrain and 7% less in forest. Comparing snow depth as grid cell values, the best agreement, less than 1% difference, was obtained for the 10 m resolution.