Scaling in sub- substorm scale fluctuations in epsilon, AU and AL- solar cycle dependence.

A key property of magnetospheric energy release is intermittent, non-Gaussian fluctuations in geomagnetic indices and in single station magnetometer data occurring on shorter than substorm timescales. However, the turbulent solar wind is also intermittent, so that the Poynting flux into the magnetosphere shares these qualitative features. We study the properties of fluctuations of the AU and AL geomagnetic indices and of the epsilon parameter which is a measure of the solar wind driver. We perform analyses that provide quantitative measures within the framework of models for turbulence and for critical phenomena; that is, we find the exponent that captures the self-similarity in the time series, and the functional form of the non- Gaussian Probability Density Function (PDF) that expresses its intermittency. Generalized structure function analysis is accompanied by PDF rescaling. Fluctuations in all quantities are found to exhibit self-similar statistics for up to 1-2 hours. We divide the data into intervals of solar maximum and minimum and find that whereas fluctuations in epsilon and AU change their properties with the solar cycle, fluctuations in AL do not. This places strong statistical constraints on the propagation of information from these below- substorm scale fluctuations from the solar wind to the magnetosphere as seen by the indices.