Abstract:

Low-level climatological wind fields over the La Plata River region of South America are synthesized with a dry, hydrostatic mesoscale boundary-layer numerical model. The model is forced at the upper boundary with the 1200 UTC local radiosonde observations and at the lower boundary with a land-river differential heating function defined from the daily meteorological observations of the region. The climatological wind field is defined as the mean value of a series of individual daily forecasts, employing two methods. The simplified method considers a 192-member ensemble (16 wind directions and 12 wind-speed classes at the upper boundary). Each member has a probability of occurrence that is determined from the 1959-1984 observations; the daily method uses a total of 3,248 days with available data during the same period. In both methods each realization is a daily forecast from which the mean wind distributions at 0300, 0900, 1500 and 2100 local standard time are calculated and compared to the observations of five meteorological stations in the region. The validation of the climatological wind fields for both methods is evaluated by means of the root-mean-square error of the wind-direction frequency distribution and mean wind speed by wind sector. The results obtained with the two methods are similar, and the errors in wind speed are always smaller than those in wind direction. The combined errors of wind direction and wind speed show that the ensemble method is outperformed by the daily method, on average by meteorological station in only one out of five of them, and on average by the time of the day in only one out of 4 h. The conclusion of the study is that the ensemble method is an appropriate methodology for determining high resolution, low-level climatological wind fields, with the boundary-layer model applied to a region with a strong diurnal cycle of surface thermal contrast. The proposed methodology is of particular utility for synthesizing wind fields over regions with limited meteorological observations, since the 192-member matrix can be easily defined with few observing points, as well as in the case of relatively incomplete records. © 2011 Springer Science+Business Media B.V.