P. T. Jayachandran

Ground‐based amplitude measurements of Global Navigation Satellite System signal during ionospheric scintillation are analyzed using prevalent data analysis tools developed in the fields of fluid and plasma turbulence. One such tool is the structure function of order q , with q = 1 to q = 6, which reduces to the computation of the second‐order difference in the GPS signal amplitude at various time lags, and allows for the exploration of dominant length scales in the propagation medium. We report the existence of a range where the structure function is linear with respect to time lag. This linear time segment could be considered as an analog to the inertial range in the context of neutral and plasma turbulence theory. Quantitatively, the slope of the structure function in the linear range is in good agreement with the scaling exponent determined from in situ measurements of the electrostatic potential at low altitude (E‐region) and the electron density at the topside ionosphere (F‐region). This in turn suggests the conjecture that scintillation could be considered a proxy for ionospheric turbulence. Furthermore, we have found that the probability distribution function of the second‐order difference in the signal amplitude has non‐Gaussian features at large time lags; a result that seems inconsistent with equilibrium statistical physics which suggests a Gaussian distribution for the conventional random‐walk processes. This could indicate that intermittency may occur at large time lags.