Despite considerable improvements in numerical models in recent years, weather events associated with extratropical cyclones continue to present forecasters with a challenge. For most cyclogenesis events model forecasts display, at the least, subtle errors in track and intensity of the system. Occasionally, errors can be significant - a depression can deepen suddenly and unexpectedly, with major implications for wind strength as well as other weather elements. Many of the weather events within extratropical cyclones are essentially mesoscale and, on occasions, a single cyclone can be associated with several simultaneous ‘severe’ events such as gales, heavy rain, snow and thunderstorms. This paper describes how the forecaster can add value to numerical model products by using information from a wide variety of sources such as satellite and radar imagery, as well as by monitoring hourly surface observations. Examples are shown in which the forecaster has to make adjustments to model predictions, reconcile conflicting model guidance, and add the all-important mesoscale detail.

A brief description of the ECWMF Ensemble Prediction System is given, including the method for producing initial perturbations, products and verification, and planned developments.

A comparison between potential vorticity (PV) fields and water vapour imagery shows a close relationship in the region of the polar front. Examples are given showing the nature of this relationship and how a mis-match between the PV and the imagery can be used to detect errors in model analyses and forecasts. The nature of the relationship varies according to the synoptic situation and some of the possible pitfalls encountered by not applying additional checks using model humidity or ‘pseudo imagery’ fields are illustrated.

This brief paper further develops the objective front-plotting methodology described in Hewson (1996), to enable the tips of frontal wave cyclones to also be objectively identified. The method embraces a new definition of frontal waves, but is analogous to operational practice in that these waves are located where cold and warm fronts join. It is suggested that for the early stages of cyclonic development the new methodology will perform better than those previously published.

The dry intrusion is a coherent region of air descending from near tropopause-level. It often has a clear signature in satellite imagery, especially in the water vapour channel, where it is seen as a ‘dark zone’. Parts of dry intrusions are characterised by high potential vorticity and, upon approaching a low-level baroclinic zone, rapid cyclogenesis may be expected to ensue. The leading edges of dry intrusions are defined by cold θw-fronts (moisture fronts). In some places the dry intrusion undercuts rearward-ascending warm air to give an ana-cold front. In other places it overruns the warm air to produce an upper cold θw-front in advance of the surface cold front. Here the dry intrusion is associated with the generation of potential instability and its eventual release as showers or thunderstorms. Identification of dry intrusions provides the forecaster with additional nowcasting evidence that is especially helpful when issuing severe weather warnings. The identification of water vapour dark zones associated with dry intrusions can also form the basis of methods for validating NWP models. Through their relationship to high potential vorticity, they can provide guidance for bogussing NWP models in situations of potentially severe weather. This article provides an introduction to the structure and behaviour of dry intrusions and their relationship to other aspects of extra-tropical cyclones.

A physical discussion of vorticity and its change by stretching and tilting is presented. The discussion naturally leads on to the conservation of quasi-geostrophic potential vorticity, q, and the flow associated with anomalies in q. The extension to full potential vorticity (PV) and the associated conservation and inversion properties is made. The use for illustrating and understanding synoptic developments of maps of PV on isentropic surfaces and potential temperature (θ) on PV surfaces is shown by considering two recent examples in the North Atlantic-Europe regions. The first is a summer situation and the second is the major blocking in January 1996. Concluding comments about the role of diabetic processes and other possible uses of these diagnostics are made.

Quasi-geostrophic (Q-G) theory can be used to infer the vertical circulation associated with a synoptic distribution of pressure and temperature in much the same sense as geostrophic theory is used to infer a three-dimensional distribution of the horizontal circulation. For the purposes of subjective analysis, a diagnosis of the vertical circulation and implied surface development can be based on identifying the sign and magnitude of a geostrophic forcing function associated with the particular distributions of geopotential and thickness shown on a conventional 500 hPa analysis chart. Examples of idealised circulation features are used to demonstrate both the strengths and the weaknesses of this approach, with reference to an analysis based on three different methods of evaluating the geostrophic forcing.

A divergence-based procedure for diagnosing development, based on a two-layer model of the atmosphere, is discussed. It is pointed out that thinking in terms of vorticity advection and thermal advection is directly equivalent to considering divergence due to advective and isallobaric components of the ageostrophic wind. It is shown that such an approach is similar in basis to the quasi-geostrophic omega equation, but it is argued that it is more suitable for subjective application. However, simply estimating vertical velocity or instantaneous pressure tendency from synoptic charts is of questionable accuracy and little operational use, especially since such diagnostics are much more reliably obtained from NWP models. It is necessary to think in terms of the development and movement of features in the flow of the upper as well as the lower troposphere, and their vertical interactions.

With the aid of a generalised frontogenesis vector and an expanded Q-vector diagnosis, the frontogenetic and frontolytic effects in numerical model analysed winds are determined for the case of a typical cyclogenesis over the North Atlantic. These effects, when occurring in the geostrophic part of the flow, give rise to circulations transverse to the frontal zone and the resulting vertical motions are of importance for the dynamics of the cyclone. The forcing of the transverse circulatory motions, which is given by Q-vector divergences across the isotherms, is computed and compared with the forcing resulting from divergences along the frontal zone. The analyses for 850 and 500 hPa show that frontogenesis dominates at the beginning of the cyclogenesis as well as at its final stage, whereas in the meantime different patterns result for both levels. A significant feature in this case was the development of a new frontal zone north of the deepening low through interaction with the parent low further north. Comparison of the different parts of the omega forcing reveals that the forcing through Q-vector divergences along the isotherms is the stronger one during the mature stage, but that the transverse circulations contribute significantly to the total forcing. This is especially true for the cold front in the lower troposphere and for the newly developing frontal zone in both levels. At the final stage the influence of the transverse circulatory forcing is even more important and determines to a great extent the shape of the total forcing. The outcome of this study corresponds very well with the findings from other investigations and with the fields derived from a vortex model.

There are several extant techniques for diagnosing the dynamics and structure of synoptic-scale flow. The essence of three of these approaches (the Q-field diagnosis, Lagrangian airstreams and the PV-perspective) is outlined, and their relative characteristics illustrated by examining two synoptic cases. It is shown that the three approaches can independently reveal key features of the events, and that the views they offer are complementary.

A brief account is given of the case work undertaken as part of the Summer Study Week on ‘Extra-tropical Cyclones’ held in Reading during July 1996. It is shown that different approximate methods of assessing synoptic-scale vertical velocity can give widely differing results. Computation of the full quasi-geostrophically forced vertical velocity is recommended as a more informative procedure. Forecasts for the track of an explosively deepening north Atlantic cyclone are compared at different lead times. The comparison encompasses raw model forecasts and model forecasts adjusted by groups of case work participants. The groups achieved a large improvement over a model forecast at the 2-3 day lead time, but only marginal improvements thereafter. It is suggested that improved understanding of development mechanisms would enable the forecaster to appreciate better the range of possible outcomes in cyclogenetic situations.