Emissions of Greenhouse Gases from Urban Xi'an, China - Direct Measurements by Eddy Covariance

Throughout the world and especially in Asia, rapid urbanization is resulting in an increasing number of very large cities. In these areas, the rate of development can outpace the perceived need for environmental regulation, and frequently there are inadequate resources available to monitor pollution or enforce compliance with those environmental regulations that do exist. These limitations obviously impact air quality on a local scale, but cities also have significant environmental impacts on regional and even global scales. In order to understand and mitigate these impacts on the surrounding environment, it is first necessary to robustly characterize the pollutant emissions themselves. This can be a significant challenge. Major discrepancies arise when comparing emissions inventories based on bottom-up compilations of source types, number, and activity levels to estimates inferred from satellite observations and other large-scale techniques. Direct measurements of neighborhood-scale emission fluxes via micrometeorological approaches provide a means to resolve these differences. Such measurements can be used to quantify the integrated vertical exchange for a wide variety of greenhouse gases and other pollutants, typically with spatial footprints of tens of square kilometers and with temporal resolutions of ~30 minutes. Here we present the results of an urban flux study conducted in Xi'an, China in August 2011. For the study a 23 m tower was erected atop the ~100 m tall administration building at Xi'an Jiaotong University. From the tower, we employed an eddy covariance approach to measure concentrations and fluxes of carbon dioxide (CO₂), methane (CH₄), nitrous oxide (N₂O), and carbon monoxide (CO). Here we present an analysis of the air-surface exchange of these gases. Results indicate that while our study site in Xi'an was a net source of these species, the greenhouse gas fluxes were significantly smaller than at other sites around the world and exhibited a different diurnal pattern. We attribute these results to two factors: 1) the relatively low traffic density at the Xi'an study site relative to other urban flux sites; and 2) the presence of a large urban park in the northerly sector of the study footprint, where the vegetative sink for CO₂ was often greater than anthropogenic sources. Overall the analysis suggests that even in heavily urbanized regions land use and activity profiles can have significant impacts on air pollutant emissions.