TEC over a single station studied with PCA

AZPILICUETA FRANCISCO; NAVA BRUNO; BRUNINI CLAUDIO

Montreal

Simposio; XXXII URSI General Assembly & Scientific Symposium; 2017

URSI

The study long series of the Total Electron Content (TEC) over a single station with the Principal Components Analysis (PCA) technique is relatively new. The usual approach in this studies consists on applying Fourier Transforms, analyzing the dominant frequencies and the subsequent interpretation or association to the physical processes that could be the causes, for example the solar cycle.In this contribution we apply PCA to series of TEC determinations over a set of GNSS stations. The series consist on daily mean values spanning over more than a 11 year solar cycle and covering all of the periodic geometrical configurations of the Sun - Earth system (annual variation of the distance, annual movement of the sub-solar point, etc.). The results show that the first component alone is enough to explain approximately 90% of the total variance of the series. This component can be associated to the local noon segment of the TEC daily pattern. The temporal behavior of the related coefficients is dominated by the variation of the solar EUV flux, mainly due to the 11 year solar cycle and secondly due to annual relative movement of the sub-solar point and annual change of the Sun-Earth distance. Interesting to mention is that overlapped to the previous signals the first coefficient also shows a semi-annual signal related to the so-called semi-annual anomaly which is not present in the flux.Finally, the addition of the second and third components help to reproduce as much as 96% of the total variance. In general the second component could be related to slight corrections (w.r.t. the first component) during the local morning and the third component to slight corrections during local evening.