资源环境科技发展态势分析平台

This study investigates snowmelt and streamflow responses to cloudiness variability across the mountainous parts of the western United States. Twenty years (1996-2015) of Geostationary Operational Environmental Satellite-derived cloud cover indices (CC) with 4-km spatial and daily temporal resolutions are used as a proxy for cloudiness. The primary driver of nonseasonal fluctuations in daily mean solar insolation is the fluctuating cloudiness. We find that CC fluctuations are related to snowmelt and snow-fed streamflow fluctuations, to some extent (correlations of <0.5). Multivariate linear regression models of daily snowmelt (MELT) and streamflow (AQ) variations are constructed for each month from February to July, when snowmelt is most active. Predictors include CC from five antecedent days up to the current day. The CC-MELT and CC-AQ associations vary with time and location. The results show the dominance of negative correlations between CC and MELT, exemplifying the cloud-shading (or clear-sky) effect on snowmelt. The magnitude of the CC-MELT association (R-2) amounts to 5-61%, typically peaking in May. These associations fade earlier in summer during dry years than wet years, indicating the differing responses of thicker versus thinner snowpack. The CC-AQ association displays a less consistent pattern, with R-2 amounting to 2-47%. Nevertheless, MELT and AQ fluctuations exhibit spatially extensive patterns of correlations with daily cloudiness anomalies, indicating that the effects of cloudiness often operate over regional spatial scales.

Plain Language Summary Much of the water supply in the western United States originates as mountain streams, which derive much of their water from snowmelt. The primary driver of mountain snowmelt is solar energy, and cloud cover regulates how much solar energy can reach the snow surface. Despite this fact, how snowmelt and streamflow respond to cloud cover (or its absence) has not been thoroughly studied. In our study, we describe snowmelt and streamflow responses to cloud cover using satellite images of cloud cover and surface records of snowmelt and streamflow. We find significant snowmelt and daily streamflow rate responses to cloud cover. Importantly, during the peak snowmelt season, snowmelt and streamflow decrease when cloud cover increases, and vice versa, confirming the cloud-shading effect on the snow surface. However, this cause-and-effect process is not so simple. We also find that cloud cover (or its absence) in the previous few days can affect how much snow melts and the streamflow rate is in a day. Snowmelt and streamflow responses to cloud cover are stronger, albeit shorter-lived, in dry years than in wet years, highlighting the relative importance of cloud cover in drier years.