Satellite measurements of tropospheric trace gas distributions on a regional and urban scale

Abstract

Air pollutant is a serious problem in many urbanised areas. The recent increase in the spatial resolution of satellite instruments has made it feasible to study the regional distribution of air pollutants from space. In this thesis, I developed a novel gridding algorithm and studied the feasibility for evaluating a regional chemistry transport model using a regional OMI NO2 product. The objective is to contribute to the current effort to make satellite measurements more suitable on the local and regional scale. A new gridding algorithm was developed to map measurements from the instrument's frame of reference (level 2) onto a longitude-latitude grid (level 3). It was designed for OMI and can easily be employed for similar instruments. The algorithm reconstructs the trace gas distribution by a parabolic spline surface by accounting for the instrument's spatial sensitivity within each ground pixel. A comparison with an established gridding algorithm shows improved performance for small to moderate measurement errors due to better parametrisation of the distribution. The resulting maps are smoother and extreme values are more accurately reconstructed. To study the feasibility for evaluating a regional chemistry transport model with satellite measurements, a new regional OMI NO2 product was developed for the Pearl River Delta (PRD) region. The NO2 retrieval was improved by using high-resolution ancillary parameters of surface reflectance, NO2 profile shapes and aerosol extinction coefficients to recalculate air mass factors (AMF). The new product reduces the bias between satellite and ground measurements. The CMAQ model was evaluated with the new satellite product and the ground measurements. The mean bias between model and satellite product is consistent with the bias between model and ground network. Therefore, satellite measurements can not only be used to study the trace gas distribution but also be used to estimate the model bias. The novel gridding algorithm and the results of the model evaluation, presented in this thesis, can be very helpful for future studies of air pollutants on local and regional scale.

Date of Award	3 Oct 2014
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Mark Oliver WENIG (Supervisor), Yun Fat Nicky LAM (Supervisor) & Chung Leung Johnny CHAN (Co-supervisor)