Revealing Grain Size Variability Across the Percolation Zone of the Greenland Ice Sheet through Hyperspectral Imaging of Shallow Firn Cores

Firn grain size directly controls both gas and water transport in the firn column. Grain size is therefore a state variable that needs to be accurately measured or modeled to refine paleoclimate interpretations of ice cores and predict the fate and transport of meltwater generated in ice sheet accumulation zones. Grain growth parameterizations commonly used in firn models have been derived either from examining firn samples from a single site or through observations of snow metamorphism in laboratories. Validating these models remains difficult because of the scarcity of grain size measurements across ice sheets. Manually or digitally measuring grain size from firn samples can be tedious, time consuming, and subjective. Here, we present a new method for quickly and accurately retrieving grain size estimates from 10 m firn cores collected in the percolation zone of western Greenland during the Greenland Traverse for Accumulation and Climate Studies (GreenTrACS). Using a near infrared hyperspectral imager (NIR-HSI; 900-1700 nm) mounted to a linear translation stage, we scanned firn cores and measured the NIR spectral reflectance in a cold laboratory. We leverage the relationship between ice grain size and the area of the first ice absorption feature, centered at 1030 nm, to invert the firn core reflectance for effective grain radius. We compare these estimates of grain size with previously-measured grain diameters from optical firn core imagery and examine spatial variations of grain growth rates within the shallow firn column. We suggest hyperspectral imagery can efficiently provide robust validation data for firn models and increase our understanding of the impact of grain size variability on firn hydrology.