Unlocking Opportunities: How Calibrated Long-range Forecasts Benefit Energy Meteoorlogists.

Energy meteorologists are increasingly asked to forecast beyond the traditional deterministic forecast lead time. The subseasonal numerical model forecasts produced by the ECMWF, NOAA, and other agencies are critical sources of information for meeting this need. Forecasts from the original model ensemble tend to be overconfident, suggesting a higher frequency of occurrence than actually happens. As a result, the stakeholders of Energy meteorologists using the basic model output to prepare long-range forecasts end up disappointed with the results.

The concept of forecast calibration was introduced many years ago. Dynamical model subseasonal and seasonal forecasts are not widely available despite their advantages. Calibration is a relatively complex process requiring access to reforecast histories and attention to detail to ensure correct results. However, forecast calibration improves the reliability of long-range forecasts, meaning the frequency of the observational event they forecast, such as above-normal wind speeds or temperatures, occurs with the probability the calibrated forecasts suggest they will occur. This presentation will discuss the calibration process and the benefits of long-range forecast calibration relative to the role of the energy meteorologist. Those benefits include:

• Forecast confidence is directly proportional to the probability shown on a map,
• Probabilistic forecasts allow the energy meteorologist to communicate forecast risks better,
• “Forecasting the forecast,” which is the capability of revealing guidance about the likely medium-range model forecast, when they are released at a later date.