PubAg

Main content area

The performance of RegCM4 over the Central America and Caribbean region using different cumulus parameterizations

Author:
Martínez-Castro, Daniel, Vichot-Llano, Alejandro, Bezanilla-Morlot, Arnoldo, Centella-Artola, Abel, Campbell, Jayaka, Giorgi, Filippo, Viloria-Holguin, CeciliaC.
Source:
Climate dynamics 2018 v.50 no.11-12 pp. 4103-4126
ISSN:
0930-7575
Subject:
air, climate, climate models, data collection, islands, temperature, wet season, Caribbean, Central America
Abstract:
A sensitivity study of the performance of the RegCM4 regional climate model driven by the ERA Interim reanalysis is conducted for the Central America and Caribbean region. A set of numerical experiments are completed using four configurations of the model, with a horizontal grid spacing of 25 km for a period of 6 years (1998–2003), using three of the convective parameterization schemes implemented in the model, the Emanuel scheme, the Grell over land-Emanuel over ocean scheme and two configurations of the Tiedtke scheme. The objective of the study is to investigate the ability of each configuration to reproduce different characteristics of the temperature, circulation and precipitation fields for the dry and rainy seasons. All schemes simulate the general temperature and precipitation patterns over land reasonably well, with relatively high correlations compared to observation datasets, though in specific regions there are positive or negative biases, greater in the rainy season. We also focus on some circulation features relevant for the region, such as the Caribbean low level jet and sea breeze circulations over islands, which are simulated by the model with varied performance across the different configurations. We find that no model configuration assessed is best performing for all the analysis criteria selected, but the Tiedtke configurations, which include the capability of tuning in particular the exchanges between cloud and environment air, provide the most balanced range of biases across variables, with no outstanding systematic bias emerging.
Agid:
5964550