Regional wind monitoring system based on multiple sensor networks: A crowdsourcing preliminary test

• Basic meteorological stations can provide interesting information about short-term wind conditions.
• A regional wind monitoring system based on both low-quality and high-quality stations is the best solution.
• The tested prototype achieves RMSE values of 1.52 m/s and 64.9° for 10-min series in the region of Andalusia.
• The application of crowdsourcing techniques would improve these results.

The availability of updated information of regional wind fields is fundamental for an efficient management and prediction of wind power production. The present paper proposes a regional wind monitoring system based on the integration of different meteorological nets as a direct way to obtain this information. Concretely, we describe a monitoring system for the region of Andalusia supported by 198 stations, able to provide near surface wind field estimations with updating period of 10 min. Each of these stations has different characteristics since they are focused on measuring different environmental parameters, and most of them are below the quality level required by the World Meteorological Organization for wind measurement. Despite this drawback, the proposed monitoring system takes advantage of the high density of measurement points to produce valuable descriptions of the regional wind field. A basic geostatistical model coupled to the system achieves better results than Numerical Weather Prediction models, obtaining RMSE values of 1.52 m/s and 64.9° for speed and directional series respectively. The test confirms the tolerance of this massive monitoring system to data without quality assurance, being an interesting platform to implement crowdsourcing methodologies.