An inspection of Analysis increments can suggest to the user the areas where the analysis scheme may not have completely captured the true state of the atmosphere. They show where observations have led to a modification to, and hopefully, an improvement of the field but they do not suggest that a perfect description of the field has been attained. And in any case, new data anywhere, unless fully rejected, will always induce some, usually slight, changes to the structure of the IFS model atmosphere. The influence of the modification of the analysis by latest data, both good and bad, often travels faster downstream than the synoptic systems themselves. For example, a two-day forecast over Europe may be affected by the initial conditions over most of the North Atlantic. Similarly a five-day forecast may be affected by the initial conditions over the North American continent and easternmost North Pacific. There is also an ever-present influence from the subtropical and tropical latitudes. In particular, when a subtropical depression, tropical storm or hurricane enters the mid-latitude westerlies (Fig4.3.1) or large areas of severe convection modify the upper flow (e.g. as in MCS over the United States) the energy of the distorted flow will propagate downstream faster than the ambient winds. The speed of energy transfer may be considered very roughly as 30° longitude per day, although case studies show that in individual cases the value can be very different. It is important to assess the potential influence of such an evolution that may not be adequately resolved by the IFS model and its downstream consequences. Therefore, if the short-range forecast is initially poor or good over the area of interest, this does not mean that the medium-range forecast for the same area is necessarily poor or good also. Any attempt to judge the medium-range performance, as distinct from the short-range performance, should be based upon large upstream areas and also involve the upper-air flow.
Fig4.3.1: Schematic illustration of the typical propagation of forecast errors over the northern hemisphere towards Europe in situations with generally zonal flow. The errors propagate mainly along the storm track, which during the warm season, on average, is displaced polewards. Note however that in individual cases the propagation speed can vary greatly from what is shown here.
Jumpiness or errors in the forecast at Day3 over Europe typically have their origin over the eastern or north-eastern parts of the North American continent. Similarly, jumpiness or errors at Day5 over Europe often have their origin over north-eastern North Pacific and adjacent coastal regions. In rare cases, forecast failures at Day7 have been traced back even further. The Madden-Julian Oscillation can also have a significant downstream effect. During all seasons, but in particular during the summer and autumn, forecast errors associated with disturbances in the tropics or subtropics can move into the zonal westerlies (shown schematically by the green arrows). Extratropical transition of low-latitude weather systems into the mid-latitudes can often cause substantial differences in subsequent evolution and significant forecast variability in forecasts over European areas.
Overview
Community Forums
Content Tools