SUMMARYThe FRONTS 87 project was a European experiment to make mesoscale observations to deduce the dynamics of active cold fronts approaching north-west Europe, with a minimum of orographic influences. In this paper data from dropsoundings made by the Meteorological Office's C-130 research aircraft are analysed. The aircraft flew a pattern with 4 or 5 runs oriented approximately at right angles to the front, but with each run displaced in the along-front direction by about 100 km. The dropsondes gave soundings to a height of approximately 7 km with a cross-frontal spacing of 20 km at best and 60 km on average. Between 30 and 50 soundings were made in each event, over an area of 500 X 500 km2. These data are unique in their mesoscale resolution over a synoptic-scale region; this was made possible by the ability to track up to 5 sondes in the air at any time.Several new aspects of cold-front dynamics have been identified from these data; the approach here is to treat each run as an instantaneous cross-frontal section. The discussion is centred on the hypotheses postulated before the experiment; these were intended to allow verification or refutation of aspects of models of frontal dynamics such as those embodied in semi-geostrophic theory. A minimal and purely objective analysis has been performed on the data presented here, but the high resolution of the dropsonde observations permits evaluation of differentiated quantities, such as potential vorticity (PV), with some confidence. Issues raised here include the degree of thermal-wind balance, the structure of conserved variables such as equivalent potential temperature and absolute momentum, the cross-frontal circulation and the role of diabatic processes such as the evaporation of snow, and the potential vorticity structure on the mesoscale and its interpretation. The intention is to present an integrated view of the dynamical structure of fronts in the light of theoretical rather than conceptual or airflow models.

SUMMARYA three-dimensional primitive equation model of the stratosphere and mesosphere is described. The model was forced at its lower boundary by observed lOOmb height fields for the wavenumber 2 stratospheric warming period of February 1979, and correctly simulated the reversal of the high latitude circulation. This behaviour contrasts with earlier model simulations of wavenumber 2 warmings in which forcing of a climatological zonally symmetric initial circulation by a stationary wave perturbation led to an initial reversal of the circulation in low latitudes. In a series of idealized experiments we show that wave-wave interaction played no essential role in this simulation and that the ingredients leading to its success were firstly the non-climatological initial wind structure and secondly an imposed longitudinal phase speed for the upward propagating planetary wave at the lower boundary. These studies also demonstrate that in the prewarming period the direction of propagation of planetary wave activity, as represented by integral curves of the so-called Eliassen-Palm flux, may be qualitatively described by the WKBJ limit of the quasi-geostrophic potential vorticity equation. In this limit, the Eliassen-Palm flux is simply related to the zonal mean refractive index, leading to two complementary diagnostics for studying wave, mean-flow interaction.

SIJMMARYA case study investigating the relationship of ice phase microphysical structure to differential reflectivity is presented. The distribution of ice crystal types in light precipitation is discussed and shown to agree with a theoretical model of an individual precipitation fallstreak. The particle characteristics were found to reflect sorting by size and type associated with fall through weak wind shear as well as growth in regimes of varying temperature and humidity within the fallstreak.The observed differential reflectivities were compared with the in situ particle measurements and the association of high values with planar crystal structures demonstrated. The particle size distribution was found to have a marked influence on radar returns, particularly through the obscuring effects of large particles of indefinite shape. This effect gave rise to relative variations of differential reflectivity which were found to differ from the predictions of monodisperse models. In particular large dendritic crystals of low density were found in this case to be associated with stronger signatures than denser plate-like crystals which from theoretical studies might be expected to produce the greatest differential reflectivity. Because of these complicating factors under stratiform conditions it was concluded that differential reflectivity data must be interpreted with considerable caution in the absence of in situ measurements or other supporting data.

Abstract. Midlatitude cyclonic cloud systems are common occurrences that significantly impact our climate. In this review, attention is paid to those physical characteristics of these cloud systems with large-scale impacts that must be accounted for in climate simulations. Such attributes include atmospheric forcing, internal structure, surface influences, cloud layering, microphysics, precipitation, water cycling, and radiation. Because of their present limitations associated with, for example, grid sizes and simplified parameterizations, climate models cannot account for all the crucial impacts of these cloud systems. Future advances in the representation of these systems within climate models will need to rely in part on rigorous assessments of model capabilities in a variety of conditions. system. They contribute substantially to global and regional water and energy cycles because of organized vertical motions and latent heat exchanges occurring within the clouds, thereby redistributing and altering the vertical profiles of momentum and moisture. These clouds account for a major fraction of the precipitation (and therefore atmospheric heating) in midlatitudes, and they are associated with some of the main avenues for the latitudinal transport of moisture through the atmosphere [e.g., Lambert, 1988]. These cloud systems also account for a substantial portion of the large-scale cloud radiative forcing found in the midlatitudes. This is due in part to the production of high-level clouds [Wylie and Menzel, 1989;Wild et al., 1995] and to the enhanced cloud water path that occurs in association with these systems. In addition, the interaction of these layer cloud systems with orography often leads to enhanced clouds and precipitation. Because mountains are the source regions for many of the world's rivers, the role of orography in affecting the small-scale distributions (<20 km) of precipitation from these clouds is a critical problem IReinking, 1995]. Air-sea interaction is also important; deep water formation in the North Atlantic (and possibly other areas) may be enhanced through the associated cold air outbreaks linked with these cloud systems. The generation of the downward convection in these bodies of water represents a major aspect of the global oceanographic circulation [Webster, 1994].Global climate prediction models simulate some of the coarse characteristics of these cloud systems [e.g., Lambert, 1995], but they do not account for finer details [Bengtsson, 1995]. For example, McFarlane et al. [1992] showed that current models display a midlatitude peak in precipitation associated with these systems, though The purpose of this article is to summarize some of the current understanding of clouds associated with midlatitude cyclonic storms, with a focus on the processes and phenomena associated with them that can affect climate. The article brings together many recent developments in the study of these systems and relates these to their large-scale impact. It is intended that this article will draw the ...

Observations from 58 dropwindsondes released in a mesoscale array during the FRONTS 92 experiment are interpreted in the context of satellite imagery to derive the mesoscale structure and evolution of parts of a frontal cyclone developing over the eastern North Atlantic. A conceptual model involving the intertwining of 'dry intrusion' and 'cloud head' flows is corroborated and is used to provide the framework for interpreting the detailed mesoscale behaviour. In the cold air, two distinct dry-intrusions were responsible for two cold fronts, trailing south-westwards from the tip of the cloud head. Both were surface features at the beginning of our study but the leading one evolved into an upper front with mid-level convection as the dry intrusion responsible for it overran the warm sector. Areas of both high and low potential vorticity were indicated within the dry intrusions. Upon encountering a critical level of zero system-relative velocity at the top of the moist boundary layer, the dry-intrusions' arrival was associated with the development of multiple dry and moist laminae near the top of the boundary layer. The vertical wavelength of the laminae was about 1 km and they extended over 200 km in the front-normal direction, with a slope of typically 1 in 60. Although most parts of the laminae were subsaturated, their circulations combined with the double structure of the dry intrusions to produce multiple shallow cloud-lines within the boundary layer. These formed as an extension of the south-western tip of the cloud head associated with the developing cyclone. Possible mechanisms for generating the observed structures are discussed.