Handout (321.5 kB)
The model is forced with boundary conditions defined from local weather observations, regional and global scale model outputs, or a combination of both. When BLM is forced with observations it becomes an excellent tool for constructing climatological wind fields and analyzing the local effect of large scale atmospheric conditions and their variability.
The model horizontal resolution is 0.05º, which corresponds to an average of 5 km, with 79 points in the x direction (354 km) and 58 points in the y direction (315 km). The vertical domain has 12 levels between the surface and the material top at 2000 m, distributed according to a loglinear spacing. The physical and numerical formulation of the model, along with a high-resolution coast geometry digitization, allows reproducing kilometer scale details of the atmospheric circulation normally unnoticed by other models.
The low-level wind fields are calculated as the ensemble result of a series of daily forecasts obtained by forcing the model with limited local observations. The upper boundary condition is taken from the local radiosonde observation, and the lower boundary condition consists of a surface heating function calculated with the temperature observations of surface weather stations in the region.
Previous studies showed an overall good agreement between the modeled surface winds and the observations at 13 weather stations. The model represents very well the differences in wind speed magnitudes and predominant wind direction sectors across the La Plata River region which is characterized by a robust sealand-breeze daily cycle. The appropriate definition of the landriver surface temperature contrast along with the high-resolution geometry of the river coasts are the fundamental advantages of BLM for resolving small scale details of the low-level atmospheric circulation.
The objective of this study is to evaluate different alternatives for initializing a mesoscale boundary layer model with radiosonde data, with the ultimate purpose of optimizing the ensemble mean method for the simulation of low-level climatological wind fields over regions with limited observations.
The alternatives considered for determining the BLM wind and temperature upper boundary condition from the only radiosonde observation available in the region include 4 groups, namely: i) standard levels; ii) first significant pressure level; iii) level of the temperature inversion base; and iv) interpolated levels at different fixed heights. The preliminary conclusion of the study is that the overall best result, considering the different weather stations and times of the day, is obtained when using the 1000 hPa level data for initializing the BLM model.