Back to Search
Start Over
Arctic and subarctic snow microstructure analysis for microwave brightness temperature simulations
- Source :
- Remote Sensing of Environment. 242:111754
- Publication Year :
- 2020
- Publisher :
- Elsevier BV, 2020.
-
Abstract
- Passive microwave (PMW) remote sensing has proven to be a useful approach to characterize the volume of seasonal snowpack in remote northern regions at the synoptic scale. Modeling emitted microwave brightness temperatures (TB) is made possible using a physical radiative transfer model that takes into account microstructural and stratigraphic structure of the snowpack. However, prescribing the microstructure remains a difficult task. This paper aims to find proper microstructure parametrization and the snow emission model formulation that best optimize TB simulations for Arctic and Subarctic snowpacks. Surfaced-based radiometric measurements in conjunction with in-situ snowpack characterization were used for testing different configurations based on the Snow Microwave Radiative Transfer model (SMRT), with two electromagnetic models (Dense Media Radiative Transfer Quasi Crystalline Approximation, DMRT, and Improved Born Approximation, IBA) and two microstructure description theories (Sticky Hard Sphere, SHS, and Exponential, Exp). We compare the performance of three configurations (DMRT-SHS, IBA-SHS and IBA-Exp) with a unique large dataset (119 snowpits with concomitant microwave ground-based radiometer observations) covering a wide range of Arctic and Subarctic snow types in Northern and Eastern Canada. Results show that the input measured microstructure parameters must be scaled up in order to better match simulated and observed TB at 11, 19, 37 and 89 GHz. We show that the IBA-Exp gives the best results, with a Root-Mean-Square Error (RMSE) lower by up to 30% for Subarctic snow and 24% for Arctic snow compare to the other model configurations we used. In addition, we undertake a complementary experiment on isolated homogeneous snow slabs to investigate the sensitivity of the scaling factor to snow microstructure. The retrieved microwave correlation length appears significantly different than the in-situ Debye correlation length. At high frequencies, the observed variability of these scaling factors with frequency and snowpack types means that density, SSA and estimated correlation length seem insufficient to appropriately fully characterize snow microstructure for microwave modeling purposes.
- Subjects :
- Radiometer
010504 meteorology & atmospheric sciences
0208 environmental biotechnology
Types of snow
Soil Science
Geology
02 engineering and technology
Snowpack
Snow
01 natural sciences
020801 environmental engineering
Atmospheric radiative transfer codes
13. Climate action
Brightness temperature
Radiative transfer
Environmental science
Parametrization (atmospheric modeling)
Computers in Earth Sciences
0105 earth and related environmental sciences
Remote sensing
Subjects
Details
- ISSN :
- 00344257
- Volume :
- 242
- Database :
- OpenAIRE
- Journal :
- Remote Sensing of Environment
- Accession number :
- edsair.doi...........ecabab5f5f168bab0ad9d77cbf0aa862