Abstract
A primitive equation, dry, hydrostatic, and incompressible mesoscale boundary layer model is used to simulate the high-horizontal-resolution low-level wind field “climatology” over the La Plata River region in South America. The horizontal model domain has 79 × 58 points (350 km × 316 km), with a horizontal resolution of 0.05°. The model climatological field is the ensemble result of a series of daily forecasts obtained by forcing the model with limited local observations. Each ensemble member produces a daily forecast that participates in the definition of the wind climatology with a probability calculated with the local observations. The upper boundary condition is taken from the only local radiosonde observation, and the lower boundary condition consists of a surface heating function calculated with the temperature observations of the surface weather stations in the region. The study, conducted during the period of 1959–84, reveals an overall good agreement between the observed and the modeled surface wind climatological fields at five weather stations in the region. The model represents very well the differences in the wind speed magnitudes and predominant wind direction sectors throughout a region that displays a strong sea–land-breeze daily cycle. The average root-mean-square value of the model relative error is 31% for wind direction and 23% for wind speed. Model errors vary throughout the day with the minimum in the morning and afternoon and the maximum at night. The seasonal climatology shows the minimum wind direction error in winter and the maximum error in summer, whereas the wind speed errors reveal no seasonality. The annual wind direction error is very similar to the winter minimum error. The conclusion of the study is that the proposed ensemble mean method is useful for synthesizing high-resolution climatological low-level wind fields over regions with a strong diurnal cycle of surface thermal contrasts and a limited number of available weather stations.
* Current affiliation: National Meteorological Service, Buenos Aires, Argentina
+ Member of the National Research Council of Argentina (CONICET)
Corresponding author address: Guillermo J. Berri, Servicio Meteorológico Nacional, 25 de Mayo 658, C1002ABN Buenos Aires, Argentina. Email: gberri@smn.gov.ar
