Sources of uncertainty
Let's explore the question of why model forecasts may go wrong! The short video looks at three different aspects which result in our imperfect knowledge of the atmosphere and their bearing on the NWP process.
Quantifying uncertainty
In the previous YouTube video, the sources of uncertainty have been discussed. Now we will be looking at how one can quantify the uncertainty and use it in our NWP models.
Ensemble forecast performance
Verification of model performance is an integral part of the forecasting process. Verification allow us to monitor the forecast quality, to improve the forecast quality and to compare the quality of different forecast systems. In the video you will be shown verification scores used to assess the performance of an ensemble system
Ensemble forecasts: products highlights
In this video you will be shown a range of typical products derived from the ensemble prediction system. These products, by summarising and tailoring the information contained in the 51 realisation of a forecast, help forecasters in their daily duties to provide the most appropriate service to their customers
Communicating confidence
Communicating the uncertainty of the forecast is vital to forecast users, but it has proven to be very difficult to provide such information in a way that can be easily understood. Strategies to communicate this uncertainty have been or are being developed in many NMHS around the world. In recent times there has been a greater emphasis on using the concept of confidence in a forecast. This confidence can be estimated using the uncertainty information provided by NWP models. In the YouTube video, you will see examples of such uncertainty information to be used to determine weather scenarios then presented to the general public.
Your task
After having listened to the five Youtube videos on different aspects of "Ensemble Forecasting", please answer the following questions:
- Why do you think forecasts can go wrong?
- How does ensemble forecasting address (or does not address) the reasons for 'providing wrong forecasts' you have outlined above?
- What is the EDA?
- What constitutes a 'good' ensemble forecast?
- What is the set up of the ECMWF Ensemble forecasting system?
Please upload your answers here (include your name on the document)