A comparison of EISCAT and SuperDARN F-region measurements with consideration of the refractive index in the scattering volume

[1] Gillies et al. (2009) proposed the use of interferometric measurements of the angle of arrival as a proxy for the scattering region refractive index ns needed to estimate the line-of-sight Doppler velocity of the ionospheric plasma from HF [Super Dual Auroral Radar Network (SuperDARN)] radar observations. This study continues this work by comparing measurements of line-of-sight velocities by SuperDARN with tristatic velocity measurements by the EISCAT incoherent scatter radar from 1995 to 1999. From a statistical viewpoint, velocities measured by SuperDARN were lower than velocities measured by EISCAT. This can, at least partially, be explained by the neglect in the SuperDARN analysis of the lower-than-unity refractive index of the scattering structures. The elevation angle measured by SuperDARN was used as a proxy estimate of ns and this improved the comparison, but the velocities measured by SuperDARN were still lower. Other estimates of ns using electron densities Ne based on both EISCAT measurements and International Reference Ionosphere model values did not increase the SuperDARN velocities enough to attain the EISCAT values. It is proposed that dense structures that were of comparable size to the SuperDARN scattering volume partially help resolve the low-velocity issue. These dense, localized structures would provide the Ne gradients required for generation of the coherent irregularities from which the SuperDARN radar waves scatter, whereas EISCAT incoherent radar measurements provide only the background Ne and not the density of the small-scale structures. The low-velocity SuperDARN results suggest that small-scale dense structures with refractive indices well below unity must exist within the SuperDARN scattering volume and may contribute greatly to the scattering process.