Chemistry and transport of layered phenomena in the mesosphere-lower thermosphere

The formation and properties of many naturally occurring layered phenomena in the mesosphere-lower thermosphere (MLT, 50-110 km) are not fully understood. Such layers, through their inclusion in chemistry-climate model studies, offer considerable insight into the chemical and transport properties of the MLT. In part, this thesis reports the first phase of chemistry developments for the Met Office’s new high altitude (~120 km) chemistry-climate model: the Extended Unified Model coupled to the UK Chemistry and Aerosols scheme (Extended UM-UKCA). The initial work reported here has added a realistic representation of neutral chemistry in the MLT, focusing on atomic oxygen (O) and atomic hydrogen (H). This contributes towards the Met Office’s strategic goal of a Sun-to-Earth coupled forecasting system, whilst providing a testbed for associated model performance diagnostics to be developed for the MLT. In particular, an atomic sodium (Na) chemistry diagnostic package is developed and used here to quantitatively attribute the physical and chemical deficiencies of this early version of the Extended UM-UKCA to the magnitude and variation of the resultant Na layer distribution. Motivated by recent Atmospheric Chemistry Experiment (ACE) limb satellite measurements of a layer of enhanced nitrous oxide (N2O) in the MLT, peaking around 94 km, this thesis also reports the development of a novel N2O production parametrisation to explain and allow a first model simulation of the observations.

With suitable extensions to O and H chemistry included, Extended UM-UKCA simulations reproduce the target climatological H profile through the MLT, but show an approximate ×10 O-excess, weighted towards the summer pole. Na chemistry diagnostic simulations reveal this as a cause of a similar magnitude Na-excess and an even greater Na+-excess, although model transport deficiencies also contribute. The expected Na compound partitioning is reproduced, negating some discrepancies in minor compounds, partly attributable to the breakdown of assumptions in the background chemistry. Sub-grid scale physics parametrisations, such as eddy diffusion, should be prioritised in future model versions to improve vertical transport, while the provision of chemical heating will enable a more realistic seasonal temperature gradient to be generated, assisting meridional transport. A further extension to the model lid height (≥ 140 km) is recommended to reduce the impact of atmospheric wave reflection off of the upper boundary. Simulations of the N2O layer, performed in the reference Whole Atmosphere Community Climate Model (WACCM) to overcome noted deficiencies in the Extended UM-UKCA, provide strong quantitative support for the satellite observations. Furthermore, the results show that essentially all of the N2O enhancement occurs through a new mechanism (N2(A) + O2) based on secondary electrons. The contribution of a previously proposed mechanism (N(4S) + NO2) appears to be less important than originally suggested, attributable to no more than 20% of overall N2O simulated at any altitude or latitude band. Therefore, the new mechanism needs to be included in relevant chemistry-climate models for a realistic description of N2O in the MLT.