Nordic hydrology pipeline creates large open snow data set

Hydro-Dataset: A Multi-Modal Open-Source Pipeline for Nordic Hydrological Research and Foundation Model Benchmarking

Machine Learning

Summary

Measuring snow in Nordic countries is important but tricky because existing data is scattered and incomplete. The authors created Hydro-Dataset, which combines weather station data from Sweden and Finland into one big, easy-to-use resource. It includes new snow measurements like how much water is in the snow and how fast it melts. This dataset helps train computer models that predict snow behavior more accurately than current global weather estimates.

What this means in practice

  • For hydrological modelers: Use the Hydro-Dataset to train and validate hydrological models specific to Nordic snow conditions with reliable annotated snow parameters.
  • For environmental monitoring teams: Improve local water resource management by incorporating the harmonized dataset’s detailed snow melt and freeze-thaw metrics.

Authors

Getnet Demil, Md Moinul Islam, Mourad Oussalah

Abstract

Despite world-class snow monitoring by the Swedish Meteorological and Hydrological Institute (SMHI) and Finnish Meteorological Institute (FMI), no station-anchored, multi-parameter snow benchmark dataset exists for the Nordic boreal domain. We present Hydro-Dataset v1.1, a reproducible open-source pipeline that harmonizes 582 SMHI meteorological observation stations across nine hydrological winters (2015/16-2023/24) into a CF-1.9 NetCDF product with 1.8~million annotated station-days. Six derived snow parameters, such as Snow Water Equivalent (SWE), snow phenology, melt rates, degree-day melt factor, rain-on-snow flags, and freeze-thaw cycles, supply benchmark-ready labels for hydrological foundation model training. ERA5-Land cross-validation (Pearson R=0.78, bias=-262mm) confirms observation-anchored SWE provides a more reliable benchmark than reanalysis for machine learning training.