CO₂ sources and sinks in urban and suburban areas of a northern mid-latitude city

Abstract

Urban environments can modify the local climate and are net CO₂ emitters, both of which can affect the life of an increasing proportion of the world’s population that lives in and around cities. Observational information on the seasonal and diurnal variations of CO₂ fluxes $\left({\text{F}}_{{\text{CO}}_2}\right)$ as well as on the contribution from the different CO₂ sources and sinks is needed at fine spatial resolution to help develop and validate weather prediction and atmospheric dispersion models. This study reports on ${\text{F}}_{{\text{CO}}_2}$ measured using the eddy covariance technique for a two-year period over a suburban and an urban residential area as well as a reference agricultural site located in the Montreal region, which experiences cold winters and warm summers. The seasonal and diurnal variability of ${\text{F}}_{{\text{CO}}_2}$ and its response to incoming light and to air temperature was analysed. Typical weekday and weekend vehicular traffic CO₂ emissions (E_VT) were estimated from inventory data for both residential sites. The urban site was a net source of CO₂ throughout the entire measurement period on the order of 200 t CO₂ ha⁻¹ year⁻¹. The suburban site was a winter source and a summer daytime sink of CO₂ resulting in a net source of 50 t CO₂ ha⁻¹ y⁻¹. Lower emissions at the suburban site are attributed to the large biological uptake in summer and to its low anthropogenic emissions. Higher emissions at the urban site are partly associated with its greater population and building density, promoting higher emissions from vehicular traffic and heating fuel combustion. The cold climate induced increased heating fuel combustion as compared to cities under a warmer climate.

Introduction

Increasing attention is being focused on understanding the exchanges of heat, mass and momentum over cities; the roughness of urban areas can influence local climate and mesoscale phenomena, which in turn will affect the majority of the world’s population that now live in and around cities. At the same time, increasing computational power has allowed weather prediction and atmospheric dispersion models to act at finer spatial resolution and to incorporate detailed surface characteristics and turbulent exchange properties for different urban surfaces (industrial, downtown, urban, suburban, exurban). However, the development, calibration and validation of such models are limited by the small number of observational studies available in the literature. The Environmental Prediction in Canadian Cities (EPiCC) research network (www.epicc.uwo.ca) was established to provide observations of turbulent exchanges and surface characteristic using flux towers and remote sensing to test and couple the urban surface parameterisation scheme Town Energy Budget (Masson, 2000) to a Soil–Vegetation–Atmosphere Transfer model (Noilhan and Planton, 1989).

It is well stablished that urban form modifies the local climate such that heat islands are created (Oke, 1982) which in turn decrease human comfort and lead to increased energy costs for cooling for example. Vegetated areas have been shown to mitigate urban heat through evapotranspiration (e.g. Sailor, 1998), but less well studied is the relationship between urban vegetation cover and the reduction of urban CO₂ emissions. Urban areas are major sources of CO₂ emissions which in turn can affect the global C cycle and public health (Svirejeva-Hopkins et al., 2004, Tong and Soskolne, 2007). The net CO₂ exchange from urban ecosystems results from the combination of emissions from fossil fuel burning and from sources and sinks of biological origin (Pataki et al., 2006). Net CO₂ exchange depends on climate conditions, characteristics of the built environment and human behaviour; each regulating the temporal variability of biological (vegetation and animal metabolism) sources and sinks as well as fossil fuel emissions (Cervero and Murakami, 2010, Groffman et al., 2006, Holtzclaw et al., 2002, Matese et al., 2009, Sailor et al., 1998, Zegras, 2010). CO₂ fluxes have been quantified for a limited number of cities around the world (Coutts et al., 2007, Grimmond et al., 2002, Grimmond et al., 2004, Moriwaki and Kanda, 2004, Velasco et al., 2005, Velasco et al., 2009, Velasco and Roth, 2010, Vogt et al., 2006), but less so in cities experiencing cold winters where emissions from heating can represent a major component of the CO₂ budget (Jarvi et al., 2009, Nemitz et al., 2002). Such studies have typically focused on anthropogenic CO₂ emissions from fossil fuel burning, and have neglected the role of vegetation. Hence, long-term observational studies reporting on the annual and seasonal variation of net CO₂ exchanges as well as on the environmental drivers that can affect CO₂ fluxes are still lacking, especially regarding the role of the vegetation in urban environments.

Micrometeorological techniques can be used to directly measure the integrated CO₂ fluxes over local urban areas. The eddy covariance (EC) technique (Baldocchi, 2003) permits high temporal resolution measurements of CO₂ exchange to be made continuously over multiple years to determine the diurnal, seasonal and annual variability over specific neighbourhoods of an urban landscape. The EC technique thus represents an ideal method to quantify the net CO₂ exchange of specific urban ecosystems at different time scales and can be used to gain information on the response of CO₂ fluxes to urban characteristics and environmental factors. These measurements can be used to evaluate inventories of CO₂ emissions, estimates of vehicular traffic at short time scales, and remote sensing information to determine the contribution of different sources to the net CO₂ exchange (Matese et al., 2009, Soegaard and Moller-Jensen, 2003, Velasco et al., 2005).

Montreal is the second largest city in Canada and is characterised as having a humid continental climate with cold, snowy winters and warm, humid summers. Montreal features diverse neighbourhood types from detached family suburban homes on grass and tree-covered lots to dense attached row housing with narrow alleyways; coupled with the wide range of environmental conditions experienced throughout the annual cycle, this city is an ideal location to study the relationship between CO₂ fluxes, urban characteristics and environmental factors.

From fall 2007 to fall 2009, three EC flux towers monitored surface CO₂ fluxes over agricultural, suburban and urban areas of the Montreal region as part of the EPiCC research network. This unique two-year dataset of CO₂ fluxes over Montreal represents an exceptional opportunity to quantify the net CO₂ exchange of residential areas subject to cold winters and to study the influence of environmental factors and human behaviour on CO₂ fluxes at different time scales. The objective of the present study is to quantify the net CO₂ exchange of residential areas of varying density in Montreal and to determine the role of vehicular traffic and biological sources and sinks on the net CO₂ fluxes. Specific objectives of the study were: 1) to quantify the net CO₂ exchanges of the three sites on daily to annual time scales; 2) to estimate the vehicular traffic CO₂ emissions; 3) to determine the response of CO₂ fluxes to temperature and light levels within each site; and 4) to determine the response of CO₂ fluxes to directional surface cover fractions.