Joint Centre for Mesoscale Meteorology (JCMM)
JCMM Internal Report No.97
Forecast skill of the ECMWF model using targeted observations during FASTEX
by A Montani, A J Thorpe, R Buizza and P Unden
The impact of targeted observations on forecast accuracy is investigated for five case studies of cyclogenesis during the Fronts and Atlantic Storm Track Experiment (FASTEX). Calculations of localised singular vectors (SVs) have been made from the ECMWF/IFS forecasting model to find the sensitive regions for each case study. Two sets of analyses have been prepared: a "control" set consisting of analyses in which all the FASTEX data have been removed from the operational analysis and a "perturbed" set obtained by adding to the control the data from the dropsondes deployed in regions highlighted as sensitive by SVs.
Analysis and forecast differences have been compared with the structure of the SVs as they evolve during the forecast.
Re-runs of the model, starting from the two different analyses, have been performed. The impact of the dropsonde data on the forecast skill has been assessed on the basis of both subjective (e.g. mean-sea-level pressure forecast errors) and objective (e.g. root-mean-square errors) scores. Experiments of sensitivity to the numbers of dropsondes used have also been performed. Re-runs of the model from perturbed analyses almost always give better objective scores than those starting from the control ones. The impact is more evident if the verification area coincides with the regions over which the SVs have been optimised. In this case, short-range prediction errors (up to day 2) are reduced on average by 15%, while, in the best case, the gain in predictability amounts to about 37%.
Changing the geographical location of a perturbation from a sensitive to a non-sensitive region has a dramatic impact on the growth of the disturbance. In both cases, the perturbation travels from the target to the verification region, but it does not grow when it is placed in a region that SVs indicate as non-sensitive.