Abstract
Integrating hydrological models, satellite observations and ground-based measurements remains challenging due to different spatial resolutions. Further, field-based techniques are commonly subject to placement bias in complex mountain topography. Cosmic Ray Neutron Sensing (CRNS) offers an alternative, providing monitoring of soil moisture (SM) and snow water equivalent (SWE) over several hectares. However, the mixed signal requires good knowledge of the research site. Many influencing factors remain unexplored, causing that snow is often perceived as a disturbance in SM observations and partial snow cover remains difficult to interpret. This study evaluates the complementary use of four Sentinel-1/-2 and MODIS based fractional snow cover (FSC) products at the footprint scale of five alpine CRNS sites to identify snow seasonality and corresponding neutron count (NC) ranges. After validating the products against topographic lidar, we (i) establish, (ii) test, and (iii) apply a CRNS footprint-wide classification into snow-free, partial and full snow conditions. We find that all products perform well in comparison to lidar with an overall FSC deviation of 11.8 ± 14.7%. Partial snow cover is best differentiated using product-specific FSC thresholds of approximately 2% and 97.6% (Sentinel) and 0% and 74.8% (MODIS). Identifying seasonal NC ranges allows us to evaluate the station-specific Neutron Level Index (NLI) as an inverse proxy for CRNS uncertainty, ranging from 1.2 to 2.1, and the Hydro-pool Ambiguity Index (HAI), as a proxy for SWE separability after NC rates. Our findings demonstrate that remote FSC observations enhance decision-making a-priori and a-posteriori to the installation and operation of CRNS probes