Forecasting solar wind high-speed streams (STEREO+CH)


   The STEREO+CH solar wind speed forecast model is based on persistence. STEREO-A in-situ plasma measurements (proton bulk speed) are shifted forward in time (variable; for 2017 ca. ~9 days) according to the angle of the STEREO spacecraft with Earth (see Figure left). As additional input, the information on the evolution of coronal holes (CHs) is applied. By comparing CH areas extracted in EUV data from STEREO and Earth perspective, a speed uncertainty assessment is performed. Derived changes in the CH area (given as ratio between STEREO and SDO) are used to modify the predicted solar wind speed profile arriving at 1AU. The statistical evaluation covers the time range 2008-2012 from which we derive that the STEREO+CH persistence model performs better than an ACE persistence model (increase of 12% for correctly predicted high-speed streams, decrease of misses by 23% and false alarms by 19%). For more details on the method see Temmer, Hinterreiter, Reiss (2018).

SWAMI L5 solar wind forecasting algorithm: STEREO+CH solar wind speed forecast (red curves; upper/lower curves mark the estimated speed uncertainty
range), STEREO-A persistence forecast (gray thick curve for 9 days in advance), and DSCOVR/ACE in-situ measurements (black curve) and ESWF (blue curve).
The uncertainties in the speed forecast are estimated from observed changes in the CH areas (expansion/decay; see bottom panel) causing an
under-/overestimation of the persistence result. The STEREO+CH forecast has a lead time of 4 days.

We greatly acknowledge the support by the FFG/ASAP Programme under grant no. 859729.

This plot is updated every hour.



Example of a successful implementation to correctly forecast the speed variation due to an increase in the size of the coronal hole when rotating from STEREO-A center view to Earth center view (mat@Oct 31, 2019).