(more details: Drazen Glasnovic & Branka Ivancan-Picek, MHS of Croatia)
Recently, a great deal of attention in DRU Zagreb has been paid to the examination of the forecast-to-forecast differences that can be used as an indication of the forecast quality and its skill. This refers to the successive forecasts consistency and takes into account the fact that their agreement is highly related to the possible uncertainties in the initial data.
The figures (1, 2) illustrate two examples of the forecast consistency check based on the HRID vertical time cross-sections in which two successive forecasts are compared. On the left cross-sections the family of corresponding isolines (temperature, potential temperature, specific humidity and equivalent potential temperature) are superimposed, while shaded areas on the right cross-sections indicate the forecast error calculated at every grid-point, as a deviation of the new forecast in respect to the old one.
|
Home |