Premium
The effects of changing resolution on mesocale modelling of line convection and slantwise circulations in FASTEX IOP16
Author(s) -
Lean Humphrey W.,
Clark Peter A.
Publication year - 2003
Publication title -
quarterly journal of the royal meteorological society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.744
H-Index - 143
eISSN - 1477-870X
pISSN - 0035-9009
DOI - 10.1256/qj.02.57
Subject(s) - meteorology , convection , cyclone (programming language) , horizontal resolution , dropsonde , climatology , hydrostatic equilibrium , cyclogenesis , atmospheric convection , grid , geology , numerical weather prediction , environmental science , tropical cyclone , atmospheric sciences , geography , geodesy , physics , computer science , field programmable gate array , quantum mechanics , computer hardware
Data are presented from modelling a rapidly developing cyclone, in intensive observation period 16 of the Fronts and Atlantic Storm Tracks Experiment, using the new non‐hydrostatic version of the Met Office Unified Model at a number of different horizontal and vertical resolutions. The model produces a good representation of the system, including multiple fronts with line convection, cloud heads and slantwise cross‐frontal circulations similar to those seen in the observations. Results from the model with various grid lengths, from 60 km nested down to 2 km with 45 vertical levels, show that more realistic detail is produced as the grid length is reduced but that the large‐scale structures remain consistent. By spectral analysis it is found that structures on scales shorter than approximately five grid lengths are attenuated. The line convection at the fronts is aliased onto the grid‐length scale except in the 2 km model where evidence of a three‐dimensional structure starts to appear. The 2 and 4 km models have also been run with enhanced vertical resolutions. It is concluded that whilst a 24 km 45‐level model is sufficient to represent the overall frontal structure, a model with resolution at least 2 km in the horizontal and 90 vertical levels is required to represent small‐scale multiple slantwise circulations similar to those observed in the dropsonde data. © Crown copyright 2003.