No abstract

An unusually deep (961 hPa) hurricane-like polar low over the Barents Sea during 18-21 December 2002 is studied by a series of fine-mesh (3 km) experiments using the Weather Research and Forecasting (WRF) model. The simulated polar low was similar to hurricanes and similar previous case-studies in that it had a clear, calm and warm eye structure surrounded by moist convection organized in spiral cloud bands, and the highest surface wind speeds were found in the eye wall. The proximity to the sea ice and the high surface wind speeds (about 25 m s −1 ) during the deepening stage triggered extremely high surface sensible and latent heat fluxes at the eye wall of about 1200 and 400 W m −2 , respectively. As the polar low moved eastward and weakened, maximum surface sensible and latent heat fluxes dropped to about 600 and 300 W m −2 , respectively. Two types of sensitivity experiments were designed to analyse the physical properties of the polar low. Firstly, physical processes such as condensational heating and sensible and/or latent heat fluxes were switched off-on throughout the simulation. In the second type, these processes were turned off-on after the polar low had reached its peak intensity, which minimized the deformation of the polar-low environment, making it suitable to study the direct effect of physical processes on the mature vortex. The experiments suggest that the deepening stage of the polar low was dominated by baroclinic growth and that upper-level potential vorticity forcing contributed throughout its life cycle. After the deepening stage, the baroclinicity vanished and the polar low was fuelled by surface sensible heat fluxes while latent heat fluxes played a minor role.

ABSTRACT:We present HIRLAM simulations of a deep extratropical cyclone that developed off the southeast coast of Greenland on 2-3 March 2007. The purpose of the simulations is to understand the role of orographic forcing for the cyclone evolution, relating the results to previous model studies. The cyclone evolution was preceded by a powerful cold air outbreak over Greenland, starting on 27 February, manifested by a southward movement of an upper-level potential vorticity (PV) anomaly from 80• N to 60• N. In addition to a CONTROL run, starting at 0000 UTC 2 March, which captures the main features of the cyclone evolution quite well, we have carried out simulations in which Greenland's orography was removed (NOGREEN), as well as simulations with different starting times. In the NOGREEN simulation starting at 0000 UTC 2 March, the cyclone deepens more rapidly than in CONTROL, due to a stronger cold advection on the rear side, leading to a more rapid baroclinic energy conversion. Furthermore, the cyclone position is shifted northward by 500 km, compared to the CONTROL run. A very different result is found in the NOGREEN simulations that were started 24-36 hours earlier, as the cyclone off Greenland's southeast coast is now displaced eastwards by hundreds of km, and more so as the run starts earlier. The results indicate a phase-locking by Greenland of a transient PV anomaly, indicating a mechanism for understanding cyclogenesis in this area. Without Greenland's orography, the PV anomaly is unconstrained, and the curvature of its southward trajectory is larger.

SUMMARYIn this paper simulations of three selected severe winter storms over the North Atlantic are presented. Modifications of potential vorticity (PV) fields according to features in water vapour (WV) images are combined with information from singular vectors (SVs) in an attempt to improve the initial state over data-sparse regions. The apparent mismatch between features in the WV images and the upper-level PV anomalies in the numerical analyses is corrected, mainly at levels indicated as sensitive by the fastest-growing SVs. The procedure is based on the fact that modification of the analysis in regions where SVs show a pronounced signal may have a large impact on the simulation. Model reruns, based on the inverted corrected PV fields, were then performed. Three cases are presented: the two French Christmas storms of 1999, and the storm affecting UK, the North Sea and southern Norway on 30 October 2000. In these cases a substantial improvement of the simulations of the storms was achieved.The PV modifications were carried out by a digital analysis system, implemented at the Norwegian Meteorological Institute. This system allows the PV modifications to be done interactively within an operational time limit.

SUMMARYModi cations of numerical potential vorticity (PV) elds according to features in water vapour (WV) images are combined with singular-vector dynamics in an attempt to improve the numerical analysis in the study of a severe winter storm. The apparent mismatch between features in the WV images and upper-level PV anomalies in the numerical analysis is corrected at levels indicated as sensitive by the fastest growing singular vectors. Model reruns, based on the inverted PV elds, are then carried out. Though numerical weather prediction (NWP) re-runs are highly successful in improving the simulation of the storm, they produce a degraded simulation of a second wave which followed rapidly in the wake of the rst one. Some success is achieved by enhancing its associated upperlevel PV anomaly at sensitive levels. The impact on the large-scale ow arising from localized PV modi cations is discussed and may point to a reason why the NWP runs based on PV modi cation occasionally fail to improve the simulation, e.g. the second wave in this case.