Numerical simulations of polar lows in the Norwegian Sea
Version of Record online: 18 JAN 2010
1987 Blackwell Munksgaard
Volume 39A, Issue 4, pages 334–353, August 1987
How to Cite
GRøNÅS, S., FOSS, A. and LYSTAD, M. (1987), Numerical simulations of polar lows in the Norwegian Sea. Tellus A, 39A: 334–353. doi: 10.1111/j.1600-0870.1987.tb00312.x
- Issue online: 18 JAN 2010
- Version of Record online: 18 JAN 2010
- Manuscript received 24 June 1986; in final form 13 January 1987
A high resolution version of an operational weather prediction model has been developed. An explicit time integration scheme is used on a limited area polar stereographic grid. The physical parameterization of surface fluxes, fluxes in the free atmosphere, stratiform and convective precipitation are relatively simple and no radiation processes are involved. Simulations are presented for six synoptic situations when polar lows occurred in The Norwegian Sea. Analyses from the European Centre for Medium Range Weather Forecasts(ECMWF) are used as initial analyses. The simulations are verified against published subjective analyses. In one case, analyses from detailed aircraft measurements were available.
In most cases the horizontal grid distance has been 50 km; in one case 25 km resolution was used. The integrations started before any polar low was observed and lasted for 36 or 48 hours. The model creates polar low disturbances at approximately the right position at the right time. The predicted intensity is, however, generally too weak.
All the simulated polar lows show an initial phase where a baroclinic development takes place in a reversed shear flow. Usually the structure is relatively shallow. The low-level trough or low is found to be warm, and conditional vertical instability is connected to it. In this way, a synoptic situation is prepared which is favourable for some CISK-like mechanism which sometimes further develop the polar low. The model does not handle this rapid growth properly, probably due to lack of resolution or incomplete parameterization of convection.
In all cases, cross-sections perpendicular to the trough-like disturbance show a warm occlusion, or a double-front system. The front also exists when precipitation processes are excluded in the model; however, release of latent heat strengthens the vertical circulations generally connected to each branch of the occlusion. A rapid development of a polar low might be connected to locally descending air in between the two circulations; satellite pictures indeed show a cloud-free eye in some cases.