View skill metrics and skill scores for selected models. Go to Validation for additional details.
Choose the system of measurement you want to use.
SI - International System of units is the metric system.
US - United States measurement system commonly used in the United States and most U.S. territories.
Display skill metrics | scores for the selected region.
Limit your view to a specific area by selecting from a list of shapefiles that may be associated with your account. See Upload Shapefiles and Locations for additional detail.
| Name | Variable | Unit | Description |
|---|---|---|---|
| Cooling Degree Days | cdd | °C | °F |
| Heating Degree Days | hdd | °C | °F |
| Precipitation | precip | mm/day | Precipitation per day. To convert to an hourly rate (mm/hr), divide by 24. |
| Temperature (2m) | temp | °C | °F |
| Solar Insolation | tsi | (kWh/m²) per day | The amount of solar radiation received on a given surface area in a given time period. |
| Wind Speed (10 m) | wspd | m/s or mph | The rate at which air is moving horizontally past a given point at 10m or 100m height above ground level |
Salient offers a suite of skill metrics to assess the forecast error. We offer metrics that assess the error of full probabilistic distributions (CRPS), categorical forecasts (RPS), and even simple deterministic forecast values (MAE).
Continuous ranked probability score (CRPS): CRPS is a scoring rule used to assess the accuracy of probabilistic forecasts of continuous variables. CRPS captures both the accuracy and level of certainty (narrower predicted range) as shown in the figure below. It measures the difference between the cumulative distribution function of the predicted probabilities and the cumulative distribution function of the observed outcomes, rewarding forecasts that are accurate (how close the prediction is to the observed historical value) and precise or sharp (how close the predicted values are to each other). A lower CRPS indicates better calibration and accuracy of the probabilistic forecasts.
Ranked Probability Score (RPS): RPS is a scoring rule used to assess the quality of categorical probabilistic forecasts. It measures the accuracy of predicted probabilities for different categories in comparison to observed outcomes. RPS takes into account the ranking of predicted probabilities and assigns a score based on how well the forecasted probabilities match the observed outcomes. A lower RPS indicates better calibration and accuracy of the probabilistic forecasts.
Mean Absolute Error (MAE): MAE answers the question, “What is the average magnitude of the forecast errors?”
Range: 0 to ∞. Perfect score: 0.
Characteristics: Simple, familiar. Does not indicate the direction of the deviations.
See Forecasts and Hindcasts for additional details.
| Model | Description |
|---|---|
| Salient v8 or v9 | The Salient Blend model is a multi-model blend of our proprietary AI models (natively probabilistic and calibrated by design) with properly calibrated dynamical (GEFS and ECMWF) models |
| NOAA GEFS v12 | Global Ensemble Forecast System, version 12 |
| ECMWF ENS | European Centre for Medium-Range Weather Forecasts, Ensemble Extended range |
| ECMWF SEAS5 | European Centre for Medium-Range Weather Forecasts, fifth generation seasonal forecast system |
| 30 year Climatology | 30-year stationary baseline with smoothing and no trend. It is probabilistic and properly cross-validated. This is equivalent to NOAA’s normals, e.g. 1991-2020, 1981-2010, 1971-2000. |
| Salient Climatology | Salient’s proprietary 40-year trailing average based on the global ERA5 dataset. It includes a linear estimate of trend and is smoothed/tuned for optimal performance. We apply a series of low-pass filtering operations to remove day to day noise, but maintain the sharpness of seasonal transitions to improve the usability and accuracy of our proprietary climatology. It is probabilistic and properly cross-validated. Reliability diagrams and the tercile (above-below-near normal) view are based on the Salient climatology. |
| 10 year Normal | Simple 10 year trailing average |
| 5 year Normal | Simple 5 year trailing average |
Select a benchmark model to display a skill score.