New Estimates of Snow Water Availability in the Northern Regions of North America.

Seasonal snow has a crucial role on freshwater supply in mountainous regions and high latitudes. The advent of remote sensing data and Earth System reanalysis products has opened enormous opportunities for estimating snow water availability at larger scales. Despite these technological advancements, still estimating the water stored in snow and determining its variability in space and time pose major challenges. One major issue is limitations in the benchmarking studies and the fact that while several new datasets are introduced, little is known about their accuracy, reliability and robustness. The second issue is related to the way that water stored in the snowpack is assessed using the concept of Snow Water Equivalent (SWE). Most importantly, maximum annual SWE does not reflect the losses of snow water during winter melts– a phenomenon that has become widespread due to the rising temperature and more frequent winter rain as a result of climate change. In addition, SWE does not take into account snow cover extent, and therefore cannot distinguish whether changes in stored water in the snow correspond to changes in snow depth or snow cover. To address the first challenge, a formal benchmarking is performed to test three key snow fields of a newly released reanalysis product, ERA5-Land, over the area of Canada and Alaska, ~9% of global land in which snow processes have a critical role on water supply. The considered snow variables are snow depth, snow cover and SWE, from which snow density can be also retrieved. The ERA5-Land’s snow depth and SWE fields are intercompared with Canadian Meteorological Centre’s (CMC’s) snow depth and SWE, whereas snow cover field is tested against MODIS satellite observations as the reference. Special care is made to assess how spatial and temporal patterns of change and persistence are reconstructed using ERA5-Land’s snow field over 21 ecological regions that cover the domain. In addition, the spatial patterns discrepancies between ERA5-Land’s snow fields and corresponding reference products are explored to inspect whether they entail there is any significant dependence with latitude, longitude and elevation, which points to a systematic bias in ERA5-Land data. Based on this benchmarking attempt, it is advised against the use of ERA5-Land’s snow depth and SWE estimates in Canada and Alaska, while estimates of snow cover and snow density can be still used although with cautions, particularly for local assessments, which may require bias-correction. To address the second challenge, a new and more physically-appealing metric, Snow Water Availability (SWA), is defined that take into account snow cover extent in conjunction with snow depth and snow density. Based on the findings of the benchmarking attempt, four monthly estimates of SWA are established over Canada and Alaska by integrating CMC snow depth fields with ERA5-Lands’s and CMC’s snow density as well as MODIS’s and ERA5-Land’s snow cover during the water years of 2000 to 2020 at 25×25 km2 spatial resolution. Using these SWA estimates, the implications on water availability over 25 drainage regions in Canada and Alaska are explored and discussed. It is concluded that while Canada and Alaska as a whole has gained substantial amount of SWA during the study period, the strategically important drainage regions in western Canada have lost substantial amount of SWA since the beginning of the century. This can jeopardize regional water resource management in some of the world’s most important food baskets in Canadian Prairies, revealing the urgency for regional adaptation to maintain the water, food and energy security in Canada.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.