Carbon dioxide fluxes in a suburban area of Beijing

Abstract

Over the last two decades, the eddy covariance (EC) technique has been applied to measure the flux of CO2 in various ecosystems. Among terrestrial ecosystems, urban areas play an important role because most of the anthropogenic emissions of CO2 originate in these areas. A better understanding of CO2 exchange between urban ecosystems and the atmosphere is important for quantifying urban contributions to the global carbon cycle as well as for evaluating the urban climate impact on and response to global climate changes. In the past two decades, studies of CO2 levels in urban environment mainly focused on the concentration in the ambient and emission inventory based on energy consumption data in China. To date, CO2 monitoring in Chinese cities using EC techniques is rare because it is difficult to find a representative urban surface and install EC instruments on the available tower. We investigated CO2 fluxes over a suburban, low-density residential area (Shunyi) in the north east Beijing. Suburban areas are growing rapidly in China and are potentially an important land-use category for anthropogenic CO2 emissions. We collected the EC measurements from the meteorological tower of Shunyi Meteorology Center from November 1, 2008 to October 31, 2009. The half-hourly CO2 fluxes were calculated by computing the mean covariance of CO2 fluctuations with the fluctuating vertical velocity observed by the EC system. Turbulence Knight2 (TK2) software package was employed for quality assurance and quality control of the EC data. TK2 provides a quality flag for each half-hourly flux data, and the highest quality data when the quality flag was less than 3 were used to analyze the daily variation and spatial distribution characteristics of CO2 flux during different seasons, the summer daytime CO2 flux as a function of photosynthetically active radiation (PAR), and the nighttime CO2 flux as a function of soil temperature at 10 cm depth during the measurement period. The data gaps were filled using the back propagation neural network to calculate the annual total CO2 emissions. Our results indicate that CO2 emissions were less affected by road vehicles and household activities at the Shunyi measurement area. Fuel consumption for heating during winter significantly increased CO2 emissions; and because of the carbon uptake by plants for photosynthesis during the growing season, the average daily flux measured in winter was significantly higher than that in other seasons. Patterns of spatial variation of CO2 flux showed that areas with a high proportion of man-made surfaces had higher CO2 values, while vegetation covered areas had lower CO2 values. The consistently positive CO2 flux throughout the year indicates that the analyzed suburban surface is a net source of CO2 to the atmosphere. The total annual CO2 emission from our study area was 13.6 kg m-2 a-1; this value is higher than the emissions from suburban residential areas in foreign cities. We infer that this high value is because of the low vegetation coverage and higher number of commercial and industrial buildings in our study area. Our data will be added to the global database of CO2 fluxes and it can be used for future planning of urban development with regard to reducing the CO2 emission.