New snow metrics for a warming world

Abstract

Snow is Earth's most climatically sensitive land cover type. Air temperature and moisture availability are first-order controls on snowfall. Maximum snowfall occurs at temperatures very near 0°C, so even a slight increase in temperature will shift a snowy winter to one with midseason rainfall and melt events. Traditional snow metrics are not able to adequately capture the changing nature of snow cover. For example, April 1 snow water equivalent (SWE, the amount of water represented by the snowpack) is used as a streamflow predictor. Still, it cannot express the effects of midwinter melt events, now expected in warming snow climates. The multiple impacts of a changing snowpack require a suite of climate indicators derived from readily measured or modelled data that serve as proxies for relevant snow-related and climate-driven processes. Such indicators need to be simple enough to "tell the story" of snowpack changes over space and time, but not overly simplistic as to be conflated with other variables or, conversely, overly complicated in their interpretation. This paper describes a targeted set of spatially explicit, multi-temporal snow metrics for multiple sectors, stakeholders, and scientists. They include metrics based on satellite data from NASA's Moderate Resolution Imaging Spectroradiometer, meteorological observations and snow data from ground-based stations, and climate model output. We describe and provide examples for Snow Cover Frequency (SCF), Snow Disappearance Date (SDD), snowstorm temperature (ST), At-Risk Snow (ARS), and Frequency of a Warm Winter (FWW).

Tracking the Evolution of Snow Drought in the U.S. Pacific Northwest at Variable Scales

“…To examine the degree to which commonly used satellite data can complement snow-drought interpretation, we also incorporated SCA observations from the Moderate Resolution Imaging Spectroradiometer (MODIS). Although snow metrics from satellite data are increasingly common (Crumley et al, 2020;Nolin et al, 2017Nolin et al, , 2021, it is not known whether these can be used as proxies for snow droughts directly. The SNOTEL, the combined University of Arizona SWE (Broxton et al, 2019;Zeng et al, 2018) and PRISM (Daly, 2013) data sets and MODIS snow cover (Hall & Riggs, 2016) data sets complement each other in both spatial and temporal domains.…”

Observed and predicted trends in Icelandic snow conditions for the period 1930–2100

“…The annual SCF was also calculated for each ensemble member as the number of times a model grid cell contained snow per year divided by the number of days in that year. The 1 April SWE was used as it has a long history of use as a snow metric for streamflow forecasting and the SCF was used as it has been suggested as a more appropriate snow metric for a changing climate (Nolin et al, 2021).…”

“…The MODIS instruments on the Terra and Aqua satellites (Dietz et al, 2012) provide a good balance of spatial and temporal resolution, with two daily observations and 500 m × 500 m pixels (Aalstad et al, 2020). An important variable for snow remote sensing is the snow cover frequency (SCF), the number of days with snow cover divided by the number of valid observations per year (Nolin et al, 2021), which is related to e.g., growing season length and habitability (Callaghan et al, 2011). The SCF is a key variable in the Earth's energy balance (Cohen, 1994) and can be used to analyze the impacts of climate change on the cryosphere (Brown and Mote, 2009).…”

A western United States snow reanalysis dataset over the Landsat era from water years 1985 to 2021

“…Moreover, snow exhibits significant spatial heterogeneity due to variability in snowfall, redistribution and ablation controlled by local meteorological conditions, landcover, forest cover, and other physiographic characteristics 6 , especially in mountainous regions with high terrain complexity. The in situ snow stations that do exist are typically located in forest clearings, mid-elevations and flat terrain that do not necessarily sample the underlying heterogeneity of SWE 7,8 . Hence, in situ networks tend to provide an incomplete picture of the spatial patterns of SWE and how point-scale SWE integrates to basin-scale water volumes.…”