Deriving the normalised ion-neutral collision frequency from EISCAT observations, Ann

Abstract. Common programme observations by the EISCAT UHF radar revealed an extended interval, post geomagnetic local noon on 03 April 1992, during which the F-region ion velocity orthogonal to the geomagnetic ®eld was signi®cantly enhanced, to values exceeding 2 km s)1 corresponding to a perpendicu...

Full description

Bibliographic Details
Main Authors: J. A. Davies, M. Lester, T. R. Robinson
Other Authors: The Pennsylvania State University CiteSeerX Archives
Format: Text
Language:English
Published: 1997
Subjects:
Online Access:http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.371.6680
http://hal.archives-ouvertes.fr/docs/00/31/65/17/PDF/angeo-15-1557-1997.pdf
Description
Summary:Abstract. Common programme observations by the EISCAT UHF radar revealed an extended interval, post geomagnetic local noon on 03 April 1992, during which the F-region ion velocity orthogonal to the geomagnetic ®eld was signi®cantly enhanced, to values exceeding 2 km s)1 corresponding to a perpendicular electric ®eld of some 100 mV m)1. Observations from this interval are used to illustrate a method by which estimates of the E-region ion-neutral collision frequency may be derived in the presence of enhanced electric ®eld. From both the rotation of the ion velocity vector and the reduction in the ion velocity magnitude relative to that in the F-region, independent estimates of the normalised ion-neutral collision frequency are made at the UHF E-region tristatic altitudes; the derived values are, in general, lower than model predictions. Although initial calculations assume a stationary neutral atmosphere, ®rst-order estimates of the E-region neutral wind are subsequently employed to calculate revised estimates of the normalised ion-neutral collision frequency; these neutral winds are derived by attributing the di€erence between predicted and observed enhancements in ®eldparallel ion temperature to thermospheric motion. The inclusion of neutral winds, which are themselves not inconsiderable, appears to have only a limited e€ect on the normalised collision frequencies derived. 1