Forecast "jumpiness"

Closely related to the dynamic activity and accuracy is the problem if "inconsistent" or "jumpy" forecasts: the re-occurring problem that for example today's D+5 is quite different to yesterday's D+6 verifying at the same time. The "jumpiness" is closely monitored by the ECMWF in the same way as the forecast accuracy, measuring the RMS difference between consecutive forecasts verifying at the same time (see 8.4 and Appendix B).