Using the MM5 model for wind prediction in a complex terrain site

摘要

Short term prediction models for wind power need meteorological forecastas as input. The better the quality of the available meteorological forecasts the better the wind power prediction that be obtained for on-line operation models. In Spain, meteorological forecasts are regularly delivered by the National Institute os Meteorology by means of an operational Numerical Weather Prediction System based on the HIRLAM model with a maximum spatial resolution of 0.2° lat x lon. The use of a higher resolution model should produce better meteorological forecasts, and therefore better wind power prediction, particularly on complex terrain sites. Dynamical downscaling technics have been proved as an adequate tool for characterization of regional features linked to complex orography. This technique has already been used by the EPPE (Ente Publico Puertos del Estado, Spain), within the HIPOCAS EU-funded proyect, to produce a high resolution 44-year atmospheric database for the Mediterranean basin through the REMO model. A comparison between this data and MM5 on an offshore site is considered a first step in order to obtain a higher resolution database. The authors have tested the influence of different combinations of parametrizations and nesting strategies for predicting wind on a complex terrain site with MM5 to provide realistic/efficient forecasts adapted to the characteristics of the area and case study.