For the analysis we are using the MERRA model. MERRA model is a NASA reanalysis data product which couples numerical modelling with large quantities of empirical data (surface measurements, earth observation data etc.). The model generates a long-term continuous database at a global resolution of 1/2 degree latitude by 2/3 degree longitude. The meteorological data were collected with an hourly frequency. Data is available from 1979 for a range of pressure levels and heights about ground. The analyzed period for specific site is usually more than 20 years.
The MCP-Method (Measurement – Correlation - Prediction) enables the relation of the measured data with one of the several long-term data source(s) located in the vicinity (meteorological station and MERRA data). With the correlation factor and the correlation coefficient, a qualitative and quantitative interpretation of the measured data in respect of the long-term prediction of the wind resource can be made.
Depends on the properties of the data and the site, different methodologies can be applied and specific requirements have to be taken into account: sectorial regression MCP and matrix MCP.
All methods mentioned above we are using to determine the long-term wind assessment and yield over the entire life-cycle of a wind farm.