Spatial and temporal variability in the ice-nucleating ability of alpine snowmelt and extension to cloud frozen fraction

Abstract

Ice nucleating particles (INPs) produce ice from supercooled water droplets through heterogeneous freezing in the atmosphere. INPs have often been collected at the Jungfraujoch research station (at 3500 m a.s.l.) in central Switzerland; yet spatially diverse data on INP occurrence in the Swiss Alps are scarce and remain uncharacterized. We address this scarcity through our Swiss alpine snow sample study which took place during the winter of 2018. We collected a total of 88 fallen snow samples across the Alps at 17 different locations and investigated the impact of altitude, terrain, time since last snowfall and depth on freezing temperatures. The INP concentrations were measured using the homebuilt DRoplet Ice Nuclei Counter Zurich (DRINCZ) and were then compared to spatial, temporal and physiochemical parameters. Boxplots of the freezing temperatures showed large variability in INP occurrence, even for samples collected 10 m apart on a plain and 1 m apart in depth. Furthermore, undiluted samples had cumulative INP concentrations ranging between 1 and 200 INP mL-1 of snowmelt over a temperature range of -5 to -19 oC. From this field-collected dataset, we parameterized the cumulative INP concentrations per m-3 of air as a function of temperature with the following equation c_air^* (T)=e^(-0.7T-7.05), comparing well with previously reported precipitation data presented in Petters and Wright, 2015. When assuming (1) a snow precipitation origin of the INPs, (2) a cloud water content of 0.4 g m-3 and (3) a critical INP concentration for glaciation of 10 m-3, the majority of the snow precipitated from clouds with glaciation temperatures between -5 and -20 °C. Based on the observed variability in INP concentrations, we conclude that studies conducted at the high-altitude research station Jungfraujoch are representative for INP measurements in the Swiss Alps. Furthermore, the INP concentration estimates in precipitation allow us to extrapolate the concentrations to a frozen cloud fraction. Indeed, this approach for estimating the liquid water to ice ratio in mixed phase clouds compares well with aircraft measurements, ground-based lidar and satellite retrievals of frozen cloud fractions. In all, the generated parameterization for INP concentrations in snowmelt could help estimate cloud glaciation temperatures.