European Association for the Development of Renewable Energies, Environment and Power Quality (EA4EPQ) International Conference on Renewable Energies and Power Quality (ICREPQ’10) Granada (Spain), 23th to 25th March, 2010 Mesoscale modelling for wind resource evaluation purposes. A test case in complex terrain A. Soares 1 , P. Pinto 1 and R. Pilão 2 1 MEGAJOULE, SA. Rua Eng. Frederico Ulrich 2650 4470-605 Moreira da Maia (Portugal) Phone: +351 22 0915480, Fax: +351 22 9488166, e-mail: ana.soares@megajoule.pt , paulo.pinto@megajoule.pt 2 Instituto Superior de Engenharia do Porto, ISEP Rua Dr. António Bernardino de Almeida 341, 4200-072 Porto (Portugal) Phone: +351 22 8340500, Fax: +351 22 9537352, e-mail: rmp@isep.ipp.pt Abstract. The aim of this work was the preliminary wind resource assessment of a complex terrain site, located in Bulgaria, using a virtual wind data series, taken from mesoscale modelling Weather Research and Forecasting system (WRF). The meso-micro scale coupling was carried out applying the virtual wind data series to the point where once a local wind measurement was conducted. This way, the comparison between the two data sets, virtual and local measurement, could be done. Using the mesoscale virtual information the characterization of the wind was done for the place of study. The microscale model Wind Atlas Analysis and Application Program (WASP) was used to make the wind resource mapping and the calculation of the estimated wind resource and annual production of energy, taking into account the defined configuration for the wind farm and the turbine model chosen. The obtained differences, in terms of the magnitude of the estimated resource and its own dominant orientation indicates that the application of mesoscale models in very complex topography terrains should be done with extreme care. Key words Mesoescale, WRF, WASP, Wind Farm 1. Introduction The classical methods that use measurement data obtained locally and the use of microscale simulation models, represent the state of the art regarding the assessment of wind resource potential. However, its use is only possible if the site was already monitored by a local station to measure the characteristics of the wind for a representative period - minimum one year. With the great evolution of the wind energy sector verified in recent times, the need in obtaining preliminary estimates of production before the site has been representatively characterized by local measurements, is gaining increased importance. The use of “virtual” data sets obtained by mesoscale modelling, to get a first approach to the production of a particular wind farm project is a tool that although not replacing the classical methods, can make a significant contribution in this subject. This work presents an overall mesoscale methodology approach applied on a Bulgarian complex site, the specific details of the modelling, the final results and remarks about validation, accuracy and application. 2. Methodology Regional climate simulations need an extensive quantity of data, covering at least 30 years, and a computational system capable of describing in great detail the effects of local features. In order to obtain a comprehensive large-scale climate data the NCEP/NCAR Reanalysis database is used. This project, conceived by the National Oceanic and Atmospheric Administration (NOAA) and National Centre for Environmental Prediction (NCEP, part of the National Centre for Atmospheric Research), consists of a database of global observations from various sources (standard meteorological observations, buoys, satellites, and several others) reconstructed using advanced quality control and modelling techniques [1]. This data seamlessly describes the atmosphere in several vertical levels, with a spatial resolution of 2.5 degrees. An initial large-scale analysis is conducted to determine the year (or range of years) that best describes the average conditions in several locations across the desired area for which long term climate data is available [2]. The simulation of the regional wind climate is performed by the Weather Research and Forecasting system (WRF) [3]. This state of the art mesoscale model is currently used by numerous institutions around the world, This numerical modelling system consists of several modules especially created to ingest observational data and