Partitioning Urban CO₂ Fluxes Within The EC Flux-tower Footprint

CO₂ flux observations from eddy-covariance?EC? towers for homogenous land cover are not very likely in a patchy urban surface. Using the EC method to directly measure the land-atmosphere CO₂ fluxes for one land cover type is a big challenge due to the heterogeneity of land use in urban areas. Partitioning CO₂ fluxes from the heterogonous urban underlying surface is central to up-scaling the EC observations into an urban scale.In this thesis, we present an approach for urban carbon sink partitioning based on measured CO₂ fluxes from an open-path EC system in Tianjin university?39o6?26 "N, 117o9?50" E? from 1st Dec., 2012 to 15 th May, 2013. We first established a multi-scale time series data set. We then applied Kormann-Meixner footprint model?KM? to delineate a study area of 600 m radius, which contributes more than 80% of the total CO₂ fluxes. The sources and sinks of CO₂ fluxes in the study area are composed of four parts, i.e. transportation, heating, human respiration and water-vegetation, which vary remarkably in accordance with seasons and academic calendars. Traffic emissions in the winter vocation are significantly lower than that in the semester, and can be ignored in the Spring Festival. Heating emissions are confined to the heating period that ends around 15 th Mar. Water-vegetation contribution shifts from the winter?up to 15 th Mar., 2013? to early spring?up to 26 th Apr., 2013? and, the early spring to spring days. CO₂ emissions from heating and human respiration were estimated by telephone interviews and archival data, respectively. Therefore, we derived the emission factor for each component of the CO₂ fluxes by solving a set of simultaneous equations that defines the above representative periods.The analysis shows that the heating?0.015 mg.m-2.s-1? is the least contributor for the CO₂ flux footprint and human respiration the largest?0.200 mg.m-2.s-1? among the four components. The traffic emission?0.147 mg.m-2.s-1? is the second largest contributor and a major cause of the fluctuation of the daily, weekly and monthly CO₂ fluxes. The traffic emission partition, however, was inconsistence with the simulated result by the International Vehicle Emission Model?IVE? in this study, in which a few parameters may be underestimated during the survey. We find that water-vegetation?-0.121 mg?m-2?s-1? is a stable carbon sink, yet needs to be verified in a longer term observation.