OCF Meteorograms

Operational Consensus Forecasts (OCF) employs multiple linear regression to develop forecasts using spatiotemporal interpolation of model values of surface and upper air variables (as potential predictors) and site observations (as predictands) in the dependent data set. Basic features of OCF are as follows:

• 1 Use forecasts interpolated from > 5 numerical models.
• 2 Bias-correct each of the forecast series over a certain previous number of days.
• 3 Calculates errors of each bias-corrected forecast series.
• 4 Select the smallest set of n ‘medians’ and use their inverse as weights.
• 5 To make new OCF, normalise weights and multiply by corresponding bias-corrected forecasts.
• 6 Point Based – Observatory level forecast only