Spatial debiasing is a post-processing step applied to 0.25deg native forecasts and ERA5. It provides a better estimate for the weather for a specific ****lat/lon or a list of grid points in a location_file . Debiasing helps correct inherent biases present in spatial data like ERA5 Reanalysis, a source of  training data for Salient’s models. Salient uses two methodologies depending on the timescale via an API debiasing option(debias=True).

<aside> 📌

Note: Currently, debiased forecasts are not available in Cloudflare.

</aside>

Debiasing Model - Daily Resolution

The debiasing model applies to:

• Salient GEM
• Salient Blend Daily Downscale
• ERA 5 Daily Downscale

The debiasing solution is a hybrid approach depending on the proximity of the location to a weather station. If the location is within 1 km of a good station, then the station-specific model uses a quantile-mapping approach and applies a different bias adjustment depending on the daily temperature (warm vs cold day). Otherwise, the debiasing uses the generalized model that uses training data to learn the complex relationship between ERA5 and local environment to calculate a bias correction factor.

• This is a post-processing model that downscales 0.25deg ERA5-style data to a specific location within a grid cell at a granularity of 100m
• Uses properties of the local environment like the land surface attributes and proximity to water, plus other features of the current weather, to debias the forecast
• This has the biggest impact in coastal regions, mountains, and urban areas, and has a bigger gain for daily high temperatures than for low temperatures
• Should only be used when validating against station data rather than gridded reanalysis data, since the outputs are specific to a precise coordinate

Supported Variables

• Temperature
• Miniumum Temperature
• Maximum Temperature
• HDD
• CDD
• Wind Speed 10m (coming in a future release)
• Precipitation (coming to a future release)