Measuring mesospheric ionisation

Abstract

The differential absorption method of retrieving D-region electron density profiles from partially reflected MF radar pulses, has existed in many variants since the original Gardner and Pawsey experiment of 1953. Magnetoionic theory predicts that a pulse propagating within the ionosphere undergoes birefringence two magnetoionic modes. The differential absorption method requires that both the differential absorption and the ratio of the reflection coefficients of those modes are evaluated from a model profile. In this thesis, a critical review of the differential absorption method is undertaken. Results obtained using the phase shifted correlation polarimeter at the University of Canterbury's Birdlings Flat radar facility near Christchurch, New Zealand, are synthesised to retrieve electron density measurements. The results indicate that over some height ranges, differential absorption method does not account adequately for ionospheric discontinuities and generates reflection coefficient ratios which are in error. An alternative to the differential absorption method is developed. This alternative uses an optimal estimation inverse method to retrieve both electron density-height and gradient electron density-height profiles. It is asserted that this retrieval, although not resolving small scale discontinuities, does account for them. The retrieval uses a forward model which includes the physical effects of the ionosphere upon a propagating MF pulse. A feature of the forward model is that as well as using a Fresnel reflection model and parameterising pulse width effects, it includes the effects of both absorption and Faraday rotation upon the magnetoionic modes of propagating pulses. Hourly mean electron density-height and gradient .electron density-height profiles are generated for the hours of 10:00 to 14:00 every day, over the time period of January 1994 to December 1999, The results of these retrievals are used for internal and external validation as well as to assess geophysical effects upon the distributions of middle atmospheric nitric oxide. Results of diagnostic tests and the internal validation process indicate that the retrieval consistently converges towards solutions which can be regarded as approximating "true" ionospheric behaviour. The external validation results suggest that the optimal estimation electron densities contain structure which has been observed in other studies and which can be explained in terms of dynamics and chemistry. Analyses of long term and seasonal variability of nitric oxide transport are facilitated by using a simple photochemical model to derive nitric oxide profiles from optimal estimation electron density profiles. Results show that wintertime enhancement of nitric oxide transport to the mesosphere takes place. The extent of this enhancement varies from year to year with the downward extent of vertical transport of thermospheric nitric oxide showing inter-annual variability. Gradient electron densities may indicate the presence of mesospheric gravity wave breaking. They also exhibit a strong seasonal variation with large reductions at equinox possibly indicating a lack of breaking gravity waves at this time of year. The ratio of gradient electrons to electrons shows seasonal variability consistent with the notions that gravity wave breaking occurs over a broader range in winter than in summer and upward propagation of orographic gravity waves is favoured in winter.