One of the major limitations of tropical hydrostatic mesoscale models initialized through synoptic and subsynoptic scale data and operating on a grid resolution of 20-50km is that they show limited skill in simulating the spatial and temporal distribution of precipitation despite well-predicted synoptic and subsynoptic-scale flow patterns. In general the models show a precipitation spin-up phase, however once activated, the model tends to become convectively overactive through the synoptic scale-mesoscale feedback mechanism. Our experience suggests that for the tropics where the mesoscale circulations are convectively active, the precipitation spin-up arises from the inability of the initial synoptic-scale analysis to produce dynamic, thermodynamic and moisture fields capable of supporting convection on the mesoscale. In particular, the problem lies in the wrong initial analysis of the velocity divergence and moisture fields. The overactive convection in the model arises when the model fails to initiate convection over some of the areas that are observed to precipitate, hence feeding the available moisture into the model activated regions. / It is shown in this study that this spin-up and overactive convection problem can be overcome by carrying out a dynamic Newtonian assimilation coupled with physical initialization during a preforecast integration phase of the model. In particular it is shown that the Newtonian assimilation of the rotational component of the wind and surface pressure coupled with physical initialization of surface fluxes of moisture, cumulus parameterization and outgoing longwave radiation (OLR) enables the model to build the divergence and moisture fields at the required location. The cumulus initialization is done through a humidity reanalysis via a "Reverse Kuo" algorithm. The surface latent heat flux initialization is done by reanalysis of the model's lowest level humidity fields through a "Reverse similarity algorithm", and the OLR initialization is done through a humidity reanalysis at the upper levels by matching the model OLR with satellite observations. / Source: Dissertation Abstracts International, Volume: 53-07, Section: B, page: 3543. / Major Professor: T. N. Krishnamurti. / Thesis (Ph.D.)--The Florida State University, 1992.
| Identifer | oai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_76712 |
| Contributors | Yap, Kok-Seng., Florida State University |
| Source Sets | Florida State University |
| Language | English |
| Detected Language | English |
| Type | Text |
| Format | 254 p. |
| Rights | On campus use only. |
| Relation | Dissertation Abstracts International |
Page generated in 0.0021 seconds