SciELO - Scientific Electronic Library Online

 
vol.30 número1  suppl.SympEl Niño and Multidecadal Climate Change: A Global PerspectiveHydrographic Variability along the North Eastern Mexican Pacific índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Revista

Articulo

Indicadores

Links relacionados

Compartir


Investigaciones marinas

versión On-line ISSN 0717-7178

Investig. mar. v.30 n.1 supl.Symp Valparaíso ago. 2002

http://dx.doi.org/10.4067/S0717-71782002030100006 

Response of Air Surface Temperatures
over Central America to Oceanic
Climate Variability Indices

Eric J. Alfaro

School of Physics (LIAP-DFAOP), Center for Research
in Geophysical Sciences (CIGEFI) and Center for
Research in Marine Sciences and Limnology (CIMAR),
University of Costa Rica, 2060-Ciudad Universitaria
Rodrigo Facio, San José, Costa Rica.,
E-mail: ejalfaro@cariari.ucr.ac.cr.

In this study air surface temperature data at 337 grid points (0.5° latitude x 0.5° longitude), over Central America, from a monthly data set were analysed. The Central American air surface temperature showed homogeneous variations in two important aspects: the annual cycle and the seasonal variations. These variations were explained mainly by the first principal components. The region also shows a clear positive trend which is in agreement with the global temperature pattern that supports the idea of global warming.

Multiple regression models were fitted between the temperature first principal components (dependent variables) and several sea surface temperature indices (independent variables) for all seasons. The resulting models showed that the ENSO indices, Niño 3 and 4, have the largest influence over the region when compared with the influence of the other indices, being positively correlated with all the first surface temperature principal components. This explains more than 53% of the variance in the region, and could be indicative of latent and sensible heat transfer from the ocean to the overlying atmosphere. All these models had percentage of detection greater than 50% and false positive rates lower than 19%. Also, despite the Skills of the models fixed being high (> 0.6), the methodology is recommended for its use in a categorical way for climate predictions. Analysis of the categorical scores showed predictability potential, because they had high probability of detection and low false positive rate scores.

These results agree with the Transfer Function Model fitted to Sea Surface Temperature Anomalies (SSTA) of the Gulf of Tehuantepec, Papagayo, Panama and Quepos time series. These models show that Niño 3.4 has the strongest influence over the region when compared with the influence of the other indices, being positively correlated with all the SSTA series. The models show the influence of this index on the relative depth of the thermocline along the Central American Pacific Coast.

On the decadal scale, the first mode of the Air Surface Temperature Anomalies (ASTAs) seems to be related to the eastern tropical Pacific, the second mode to the tropical north Atlantic and the central tropical Pacific and the third mode to the tropical south Atlantic.

The results of these regression models and the correlation analysis suggest that, in general, eastern Pacific (mainly) and Atlantic variability in both interannual and decadal scales has a clear positive influence over the air surface temperature in Central America, in which warm (cold) SSTA events are correlated with warm (cold) ASTA events. This remote or local connection could be active through several processes, such as the modulation of the size and SST magnitude of the warm pool near Central America, or by inducing some anomalous troposphere patterns (e.g. PNA or wind shear). The first process affects the sensible and latent heat transfer from the ocean to the atmosphere and the second could alter the seasonal behaviour, directly or through feedback processes, of some atmospheric variables, including SLP, surface wind and precipitation, which are related to the air surface temperature in the region.

On the other hand, precipitation decadal variability was positively correlated with decadal SSTA variability in the Tropical North Atlantic (TNA) and in the eastern tropical Pacific. The modulation of the TNA on this time scale is very similar to that proposed for the interannual scale, but is of opposite effect for the Niño 3 index, showing that warm (cold) events are associated with convergence (divergence) at low levels over Central America and divergence (convergence) at high levels. This flow pattern could be associated with processes that work in favour (against) the deep convection over the region.

Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons