Improving the blend of multiple weather forecast sources by Reliability Calibration

Abstract

Creating a forecast that is seamless across time yet is optimal at each forecast validity time is often achieved by blending forecasts from multiple Numerical Weather Prediction models (or using other forecast sources, such as an extrapolation nowcast). With the increasing usage of convection‐permitting ensemble models at shorter lead times, the blending of these forecasts with longer‐range ensemble models with parameterized convection can lead to a clear transition from one forecast source to another. This is particularly noticeable when visualizing the evolution of the gridded forecast. Calibrating the forecast sources with a common truth prior to blending provides a method of improving forecast skill whilst also unifying the characteristics of the forecasts to create a smoother blend throughout the evolution of the forecast. In this work, a non‐parametric method for calibrating the reliability of the forecast without degrading the forecast resolution is assessed for its usability for gridded precipitation rate and total cloud amount forecasts. Reliability is markedly improved resulting in a similar skill between forecast sources during the blending period. Further refinements to the technique removed artefacts in the gridded forecasts. Caveats, including a reduction in sharpness following calibration, are also presented.

A seamless blended multi‐model ensemble approach to probabilistic medium‐range weather pattern forecasts over the UK

“…Probability forecast calibration, of which many potential methods exist as discussed in Vannitsem et al (2021). Example methods applicable to this study may include bias correction of the underlying pressure fields before they are assigned to weather pattern definitions (as in Büeler et al, 2021) or by applying more generalized forecast probability calibration techniques, such as reliability calibration (Rust et al, 2023).…”

“…However, the practicalities of such an approach increases when multiple models are used, such as in this study. Therefore, more generalized forecast probability calibration techniques may be appropriate, such as reliability calibration (Rust et al, 2023).…”