Mid-term Forecast Of Solar Active Index

The solar active indices indicate the turbulence of magnetic flux on the surface of the Sun. Now one of the most familiar solar indices is the solar 10.7cm radio flux (F10.7) which has been always used as a proxy of solar heating to the earth's upper atmosphere. Because the variations of the density of upper atmosphere cause for changes in satellite orbital drag, the prediction of solar active index has been important in space environment forecasting operation with the development of space technology. Though many researches have been done, the mid-term forecast of solar active index remains complex and difficult.In this thesis, two mid-term forecast model of F10.7 are built on the basis of variations of the index and the theory of solar active region. The major works is summarized as following.1. Considering the quasi-27-day oscillation of the index as chaos, an auto-regressive (AR) model can be effective. The higher error of the prediction presents at the high active phase of the 23rd solar cycle. The model works better when the index varies much regular with 27-day period, which means the AR model can describe the quasi-27-day oscillation of the index. The AR model performs similar prediction accuracy to the United Stated Air Force (USAF) which also does the forecast of F10.7 during September, 2005 to June, 2007. However, the error can become much large when some large active regions appear or disappear.2. Because of the disadvantage of the AR model and the relation of the solar index to the active regions, a new model based on the observation and general evolution of the solar active regions has been built. The area and longitude of an active region have been used as control parameters in the new model. The sunspot decay rate suggests a simple way to forecast the area of active region in next solar rotation. The result indicates that the errors of the new model were mainly 3%~10% smaller than that of AR model when the index is much larger than usual as unexpected solar active regions appear on the disc and the index becomes quite peaceful because the dominant solar activity regions disappear. The prediction accuracy of the new model is similar to the AR model in the decay and low phases of the 23rd solar cycle. That means it is available to improve the operational forecast model of solar index using the information of solar active regions. It is unpredictable to predict the birth of the active regions, which make the new model less precise in the high phase of the solar cycle.