Deriving Snow Depth From ICESat-2 Lidar Multiple Scattering Measurements: Uncertainty Analyses

The application of diffusion theory and Monte Carlo lidar radiative transfer simulations presented in Part I of this series of study suggests that snow depth can be derived from the first-, second- and third-order moments of the lidar backscattering pathlength distribution. These methods are now app...

Full description

Bibliographic Details
Published in:Frontiers in Remote Sensing
Main Authors: Xiaomei Lu, Yongxiang Hu, Xubin Zeng, Snorre A. Stamnes, Thomas A. Neuman, Nathan T. Kurtz, Yuekui Yang, Peng-Wang Zhai, Meng Gao, Wenbo Sun, Kuanman Xu, Zhaoyan Liu, Ali H. Omar, Rosemary R. Baize, Laura J. Rogers, Brandon O. Mitchell, Knut Stamnes, Yuping Huang, Nan Chen, Carl Weimer, Jennifer Lee, Zachary Fair
Format: Article in Journal/Newspaper
Language:English
Published: Frontiers Media S.A. 2022
Subjects:
snow depth
snow water equivalent
ICESat-2 lidar
pathlength distribution
multiple scattering
Geophysics. Cosmic physics
QC801-809
Meteorology. Climatology
QC851-999
Arctic
Sea ice
Online Access:https://doi.org/10.3389/frsen.2022.891481
https://doaj.org/article/5430fed14ed84b4bbabb36438d341313
Description
Summary:The application of diffusion theory and Monte Carlo lidar radiative transfer simulations presented in Part I of this series of study suggests that snow depth can be derived from the first-, second- and third-order moments of the lidar backscattering pathlength distribution. These methods are now applied to the satellite ICESat-2 lidar measurements over the Arctic sea ice and land surfaces of Northern Hemisphere. Over the Arctic sea ice, the ICESat-2 retrieved snow depths agree well with co-located IceBridge snow radar measured values with a root-mean-square (RMS) difference of 7.8 cm or 29.2% of the mean snow depth. The terrestrial snow depths derived from ICESat-2 show drastic spatial variation of the snowpack along ICESat-2 ground tracks over the Northern Hemisphere, which are consistent with the University of Arizona (UA) and Canadian Meteorological Centre (CMC) gridded daily snow products. The RMS difference in snow depths between ICESat-2 and UA gridded daily snow products is 14 cm, or 28% of the mean UA snow depth. To better understand these results, we also discuss the possible sources of errors in ICESat-2 derived snow depths, including surface roughness within the laser footprint, atmospheric forward scattering, solar background noise, and detector dark current. Simulation results indicate that the snow depth errors would be less than 5 cm if the standard deviation of pulse spreading due to surface roughness is within 50 cm. Our results demonstrate that the ICESat-2 lidar measurements can be used to reliably derive snow depth, which is a critical geophysical parameter for cryosphere studies including sea ice thickness estimation and also provides important constraints in the modeling of terrestrial hydrological processes.

Similar Items