ERA5 and ReKIS Data for High-Resolution Climate Downscaling

Predictors from the ERA5 reanalysis and predictand from the ReKIS database for statistical downscaling. The dataset is provided by Dánnell Quesada‐Chacón of Technische Universität Dresden. Data is saved in the ".rda" format for use with the R programming language.

Use Cases

• Training deep learning models for statistical downscaling based on the described predictor and predictand variables.
• Comparing reanalysis data (ERA5) with regional observational data (ReKIS) for climate model validation.
• Developing reproducible workflows for high-resolution climate projections using the provided data format.

Strengths

• Data originates from established sources: the ERA5 global reanalysis and the regional ReKIS database.
• The dataset is associated with a specific, peer-reviewed methodological paper on deep learning for downscaling.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset size are unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

ERA5 reanalysis and ReKIS (https://rekis.hydro.tu-dresden.de)

Collection Method

Likely compiled for the paper "Repeatable high-resolution statistical downscaling through deep learning".