Skobeltsyn Institute of Nuclear Physics
Lomonosov Moscow State University

The article presents a hierarchical approach for the predicting of high-speed quasistationary solar wind streams. This approach integrates various types of data in a single forecasting
system by means of an ensemble of experts. Daily values of coronal hole areas obtained from solar
images taken in the ultraviolet spectral range and information about solar wind velocity from previous
solar rotation were used. The images for coronal hole area calculation were obtained from satellite
PROBA2/SWAP in the spectral range centered at a wavelength of 17.4 nm and the AIA instrument of
SDO spacecraft centered at wavelengths of 19.3 nm and 17.1 nm. The data from May to December
2010 which corresponds to the rising phase of the 24-th solar activity cycle was used. At the first level
of hierarchy several simple model estimates of the solar wind velocity were obtained using the input
data of each type. At the second level of hierarchy the final forecast of solar wind velocity was formed
based on these estimates. Described method allows improving the forecast accuracy and shows robust
results on examination data set with periods of rapid change of SW flow recurrence. In addition, the
data gaps from one spacecraft do not have a critical impact on the final forecasting result of the whole
system.