Data-driven Discovery of Snow Cover Parameterization

All data were derived from SNOTEL, Version 1, and were preprocessed for training symbolic regression models in convenience. Train and test data are range from water years of 2001-2009, and of 2010-2018. All NaN value were removed from data and shape as [Sample, feat]. The dataset contains preprocessed dimensonless features: The name and unit of each feature were listed in sequence as follows: 1. Snow depth (mm) 2. Snow water equivalent (mm) 3. Standard deviation of sub-grid topography (m) 4. Air temperature (K) 5. Precipitation (mm/day) 6. 1/snow density (mm/mm) 7. 1/Standard deviation of sub-grid topography (m^-1) The name and unit of target were listed as follows: 1. Snow cover fraction [%] Some own defined constant: surface roughness (0.1 m) 0 degree of temperature (273.16 K) own defined std threshold (200 m) mean SWE (122.3 mm)