Vandeskog et al. (2026) Simulation and evaluation of local daily temperature and precipitation series derived by stochastic downscaling of ERA5 reanalysis

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-06-24
• Authors: Silius Mortensønn Vandeskog, Thordis Linda Thorarinsdottir, Alex Lenkoski
• DOI: 10.5194/hess-30-3875-2026

Research Groups

• Norwegian Computing Center
• European Centre for Medium-Range Weather Forecasts (ECMWF)

Short Summary

This paper proposes a computationally efficient stochastic downscaling method to obtain simulations of local daily temperature and precipitation series at any given location based on ERA5 reanalysis data and available local observation data. The method combines elements of model output statistics and weather generators, improving the representation of local temperature and precipitation compared to ERA5.

Objective

• Develop a computationally efficient approach to simulate local daily temperature and precipitation series using ERA5 reanalysis data and local observation data.
• Evaluate the performance of the proposed downscaling method against state-of-the-art dynamical downscaling methods.

Study Configuration

• Spatial Scale: Europe-wide, with a focus on central Europe (ALP-3 domain)
• Temporal Scale: Daily resolution, with a time period of over 70 years

Methodology and Data

• Models used:
  • Generalised additive models (GAMs) for downscaling temperature and precipitation
  • Auto-regressive moving average (ARMA) models for modeling temporal dependence
• Data sources:
  • ERA5 reanalysis data set
  • Global Surface Summary of the Day (GSOD) data set for local weather station observations

Main Results

• The proposed downscaling method improves the representation of local temperature and precipitation compared to ERA5 at over 4000 locations in Europe.
• The method is shown to perform well in capturing distributional properties of the weather variables, including tail properties and precipitation occurrence.

Contributions

• The paper contributes a new computationally efficient approach for simulating local daily temperature and precipitation series using ERA5 reanalysis data and local observation data.
• The study provides a comprehensive evaluation of the proposed downscaling method against state-of-the-art dynamical downscaling methods.

Funding

• This research was funded by [insert funding agencies or projects, e.g., Norwegian Research Council, European Union's Horizon 2020 program].

Citation

@article{Vandeskog2026Simulation,
  author = {Vandeskog, Silius Mortensønn and Thorarinsdottir, Thordis Linda and Lenkoski, Alex},
  title = {Simulation and evaluation of local daily temperature and precipitation series derived by stochastic downscaling of ERA5 reanalysis},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-3875-2026},
  url = {https://doi.org/10.5194/hess-30-3875-2026}
}