Abstract

Rain-on-snow (ROS) events can alter nival regimes and increase snowmelt, peak river flow, and reduce water storage. However, detection of ROS events is challenging and only the most intense and obvious cases are identified. Rain is known to reduce snow cover and decrease near-infrared reflectance due to increased grain size. This study explored the fingerprints of ROS events on mountain snowpack with a simple typology that classifies changes in snow reflectance using fifteen years of MODIS imagery, reanalysis, and surface hydrometeorological data. The Maipo River Basin, with strong nival regime and a steep topography, in the western Andean mountain range was selected as a case study. Statistical analysis showed two distinct and opposite responses in the near infrared reflectance distribution of snow-covered pixels after precipitation, consistent with the typology for rain or snow events. For the probable ROS events, the daily maximum and minimum temperature increased in the days preceding the event and subsequently decreased, in some cases followed by a less consistent response in river flow. Although much remains to be studied, this approach can be used to expand historical records and improve modelling and detection schemes.