Home | Our members | Contact us | Français | |  

Joëlle Voglimacci


Étudiante 2è cycle

Department of Applied Geomatics, Université de Sherbrooke

2500, boul. de l'Université
Université de Sherbrooke
Quebec, Canada
J1K 2R1

819.821.8000 extension 62506






Research project

Development of a snow depth mapping approach using radar satellite data and multisource in-situ data

Significant warming has occurred in the Arctic over the past four decades at a much faster rate then the rest of our planet. Negative anomalies of spatial and temporal trends snow and permafrost are now relatively well documented. Those trends lead to a series of strong climate-related feedbacks, which in turn contribute to the arctic amplification that is currently observed.

More specifically, changes in snow depth and distribution lead to changes in the spatial patterns of landscape freeze/thaw cycles, which in turn have a significant impact on how the cryosphere responds to climate change. While recent research efforts attempted to monitor soil thermal state from space, lingering uncertainties remain with regards to the various processes governing changes in snow and permafrost spatial and temporal distribution.

In this context, this project proposes to develop a snow depth mapping approach combining satellite radar data and in-situ snow geophysical measurements. More specifically, we present a methodology to map snow at high spatial resolution using a recently developed spatialized snow simulation method using the SNOWPACK model. This simulation platform will be forced with a precise digital elevation model (DEM) and continuous meteorological observations available on Qikiqtaruk (Herschel Island) in the Beaufort Sea, near the Yukon Coast. Snow height, microstructure and density will be simulated at a 250, 500 and 1000m and outputs will be compared and validated using field measurements acquired during annual field campaigns. Since the model is not suited to arctic snow conditions, we will use the Groot Zwaaftink et al. Antarctic version of SNOWPACK that showed promising improvements in preliminary work within our group. The simulations will allow the production of spatially-distributed snow information that will be used to develop a snow depth mapping algorithm using TERRA-SAR data. The available frequency at 9.6 GHz will allow a proper penetration depth in dense arctic snowpacks, while decreasing the sensitivity of the backscatter to large depth hoar layers. A statistical retrieval approach of depth using field measurements, model outputs and satellite observations will be developed and applied regionally on Qikiqtaruk, where soil thermal data are available. This will improve our empirical understanding of the effect of snow variability on freeze/thaw timing.

© 2020 All rights reserved | Adapted from an original design by BinaryTheme.com