Low Cost and Compact FMCW 24 GHz Radar Applications for Snowpack and Ice Thickness Measurements

• Published in: Sensors
• Main Authors: Patrick Pomerleau, Alain Royer, Alexandre Langlois, Patrick Cliche, Bruno Courtemanche, Jean-Benoît Madore, Ghislain Picard, Éric Lefebvre
• Format: Text
• Language: English
• Published: Multidisciplinary Digital Publishing Institute 2020
• Subjects: Frequency-Modulated Continuous-Wave (FMCW) radar; lake ice thickness; snow water equivalent; snow density; snow wetness; snow monitoring; snow boreal forest; subarctic snow taiga; Arctic snow tundra; Antarctica; Antarctic; Arctic; Antarc*; Subarctic; taiga; Tundra
• Online Access: https://doi.org/10.3390/s20143909

Monitoring the evolution of snow on the ground and lake ice—two of the most important components of the changing northern environment—is essential. In this paper, we describe a lightweight, compact and autonomous 24 GHz frequency-modulated continuous-wave (FMCW) radar system for freshwater ice thickness and snow mass (snow water equivalent, SWE) measurements. Although FMCW radars have a long-established history, the novelty of this research lies in that we take advantage the availability of a new generation of low cost and low power requirement units that facilitates the monitoring of snow and ice at remote locations. Test performance (accuracy and limitations) is presented for five different applications, all using an automatic operating mode with improved signal processing: (1) In situ lake ice thickness measurements giving 2 cm accuracy up to ≈1 m ice thickness and a radar resolution of 4 cm; (2) remotely piloted aircraft-based lake ice thickness from low-altitude flight at 5 m; (3) in situ dry SWE measurements based on known snow depth, giving 13% accuracy (RMSE 20%) over boreal forest, subarctic taiga and Arctic tundra, with a measurement capability of up to 3 m in snowpack thickness; (4) continuous monitoring of surface snow density under particular Antarctic conditions; (5) continuous SWE monitoring through the winter with a synchronized and collocated snow depth sensor (ultrasonic or LiDAR sensor), giving 13.5% bias and 25 mm root mean square difference (RMSD) (10%) for dry snow. The need for detection processing for wet snow, which strongly absorbs radar signals, is discussed. An appendix provides 24 GHz simulated effective refractive index and penetration depth as a function of a wide range of density, temperature and wetness for ice and snow.