GEMS: Generalizable empirical model of snow accumulation and melt (version 1.0)
R script and corresponding files for running the Generalizable Empirical Model of Snow accumulation and melt (Umirbekov, Essery and Müller, 2023). The model leverages the capabilities of machine learning methods and incorporates empirical relationships between daily changes in snow water equivalent (SWE) and precipitation, temperature, and topographic factors to produce estimates of daily SWE.
The corresponding files include four variations of the pretrained Support Vector Regression, based on a number of the required inputs. The accompanying two datasets include observations from independent Snowpack Telemetry Network (SNOTEL) and the Snow Model Intercomparing Project (ESM-SnowMIP) stations, which were used to evaluate GEMS performance.
- "GEMS_v1_script.Rmd" - the main script for running the GEMS model;
- "SVR_GEMS_7P.rds", "SVR_GEMS_5P.rds", "SVR_GEMS_6P.rds" and "SVR_GEMS_4P.rds" - four variations of the pretrained Support Vector Regression, which is an integral part of GEMS. The four versions differ in terms of number of required input variables;
- "SnowMIP.csv" - in-situ daily meteorological and snow observations from ten reference sites of the ESM-SnowMIP, used to evaluate GEMS performance. The dataset is based on Ménard et al 2019, and contains aggregated daily estimates of precipitation, temperature (mean, max, min) and is supplemented with site specific heat-load index;
- "SNOTEL_ext.csv" - in-situ daily meteorological and snow observations from 520 SNOTEL stations from 2013 to 2022, used to evaluate GEMS performance. Original SNOTEL temperature records were corrected for bias in the SNOTEL temperature sensor following equation suggested by Brown et al 2019. The dataset is supplemented with station specific heat-load index.
Brown, C. R., Domonkos, B., Brosten, T., DeMarco, T., & Rebentisch, A. (2019). Transformation of the SNOTEL Temperature Record – Methodology and Implications. 87th Annual Western Snow Conference. Reno, NV.
Ménard, C. B., Essery, R., Barr, A., Bartlett, P., Derry, J., Dumont, M., Fierz, C., Kim, H., Kontu, A., Lejeune, Y., Marks, D., Niwano, M., Raleigh, M., Wang, L., and Wever, N. (2019). Meteorological and evaluation datasets for snow modelling at 10 reference sites: description of in situ and bias-corrected reanalysis data, Earth Syst. Sci. Data, 11, 865–880, https://doi.org/10.5194/essd-11-865-2019.
Umirbekov, A., Essery, R., & Müller, D. (2023). GEMS v1.0: Generalizable empirical model of snow accumulation and melt based on daily snow mass changes in response to climate and topographic drivers. Geosci. Model Dev. Discuss. [preprint], https://doi.org/10.5194/gmd-2023-103.
||233.0 MB||Preview Download|
||5.4 MB||Preview Download|