Orlanski, Isidoro, and Larry J Polinsky, 1984: Predictability of mesoscale phenomena In International Symposium on Nowcasting II: Mesoscale Observations and Very-Short-Range Forecasting, Noordwijk, Netherlands, ESA Scientific & Technical Publications Branch, 271-280. Abstract
A number of simulations with a high-resolution three-dimensional primitive equation model were conducted to assess the impact of initial and boundary data inaccuracies for the simulation of mesoconvective systems and their environments. Attention has been given to the simulations of Pacific comma clouds, frontal squall lines, mesoconvective complexes, and coastal cyclogenesis; all these cases are from FGGE year 1979. The sensitivity of these simulations to surface boundary layer variables and the cloud fraction (latent heat parameterization) has been investigated. Considerable success has been achieved in those simulations. It has been found that, in most mesoconvective systems, the environmental convergence of a preexisting front is responsible for the growth of the storm's vorticity, whereas moisture is essential for its explosive evolution.
Many processes have been proposed as possible forcing mechanisms for mesoscale oceanic variability. The present study shows that atmospheric forcing can be an important source of mesoscale variability in the ocean. We show that the response is linearly proportional to the product of the time scale of the storm and its intensity. We clarify the point that for storms with scales considerably smaller than the barotropic Rossby radius of deformation, the oceanic stratification and the horizontal extent of the storm are the only factors determining the penetration depth of the response, implying that it is not the Rossby radius of deformation but rather the scale of penetration depth (h = (f/N)L) that characterizes the response.
In exploring the effect of differing eddy-viscosity parameterization on oceanic-response, we find no significant qualitative differences, although as one might expect we find quantitative differences in the results. The role of the mixed layer is considered very important in the transfer of surface stresses down into the system. The mixed layer does not seem to be important in determining the characteristic lengths of the problem, however, at least for storms that give a penetration depth considerably larger than the mixed layer (for a mixed layer on the order of 20 m, the storm should be larger than a few kilometers).
The non-linear advection terms seem to affect the adjustment process more by reducing the associated wave energy than by modifying the characteristics of the geostrophic response.
Finally, making the stratification more realistic has no significant impact on the resulting oceanic response.
An analysis is made of the spectral characteristics of the cloud cover observed over Africa for a period of three months. The results indicate the predominance of a 2-2.7 day spectral peak within the vicinity of the Equator (10 degrees N - 10 degrees S) with the intensity of this peak much stronger over land than over the ocean. The peak itself may not be detected if the smallest resolved area used in the data analysis is too large. Coherence was found to be maximum in belt-like configurations along certain latitude bands. The phase difference, although very noisy, indicates a horizontal scale on the order of 2000 kilometers.
A two-dimensional mesoscale atmospheric model is presented and used to study unsteady dynamic processes occurring in the planetary boundary layer (PBL) driven by diurnal heating at the ground. The model reproduces turbulent fluxes of heat and momentum both by explicitly modeling resolvable eddies and by employing a single parameterization at all levels of the model to represent vertical fluxes caused by subgrid-scale eddies. The unsteady behavior of horizontally-averaged profiles of temperature and velocity respond quite realistically to the diurnally-varying heat flux at the ground, particularly with regard to the time variation of lapse rates and the occurrence times of maximum and minimum temperatures at various levels in the lower boundary layer. The spatial variation of predicted atmospheric quantities shows a great deal of resolved eddy activity during the day with a significant remnant persisting through the night at higher levels of the PBL. These eddies account for the predominant means of vertical heat and momentum transfer away from the surfaces with the model realistically reproducing the unsteady behavior of heat fluxes in the PBL. Temporal variation of vertical heat and momentum profiles shows boundary layer activity to be confined to a few hundred meters at night while extending up to a kilometer during the day. A weak heat flux source at the ground with an amplitude of 10% of the maximum daytime heating produced a nocturnal heat island some 60 m high with a maximum city-country temperature contrast of ~1C.
