Temporal patterns in near-surface CO2 concentrations over contrasting vegetation types in the Phoenix metropolitan area

Abstract Few studies have examined concentrations of near-surface CO 2 in urban areas and none have considered the effect of vegetation type. To investigate this, we monitored near-surface (2 m) CO 2 concentrations and meteorological variables over four contrasting sites in the metropolitan area of Phoenix, AZ, from 15 March to 3 April 2000. These four sites represented a 2×2 factorial design of high productivity turf (golf course fairways dominated by perennial ryegrass, Lolium perenne ) or low productivity desert (dominated by creosote bush, Larrea tridentata ) near the center and at the edge of the Phoenix metropolitan area. Our main objective was to assess the influence of urbanization on near-surface CO 2 concentrations. Additionally, we assessed how well meteorological variables and traffic density patterns appeared to explain hour-to-hour and day-to-day variability in concentrations. Concentrations were higher over sites near the urban center than the metropolitan edge at all hours of the day, with the greatest differences at night. When we attempted to isolate the influence of urbanization by restricting our comparisons to periods when sites were upwind of the urban center, we found that CO 2 concentrations near the urban center averaged 19 ppm higher over the course of the day (396 ppm near the urban center versus 377 ppm at the edge of the metropolitan area), 8 ppm higher during the day-time (383 versus 375 ppm), and 24 ppm higher during the night (409 versus 385 ppm). Concentrations were higher over turf than desert sites during the night, but lower over turf during the middle of the day (09:00–15:00 h), probably due to the greater CO 2 source/sink strength of turf. Patterns of hourly CO 2 concentrations at the sites were similar and involved: (1) high, somewhat variable, night-time concentrations; (2) an abrupt decline in concentrations around sunrise; (3) low, less variable, concentrations through the day-time, and (4) a rise in concentrations around sunset. In general, CO 2 concentrations were low at high photosynthetically active radiation (PAR) and high wind speeds. Regression analyses between logarithmically transformed meteorological variables and hourly CO 2 concentrations revealed that during the day-time, explained the greatest amount of variability in concentrations over turf, whereas wind speed explained the greatest amount of variability over desert. During the night, wind speed explained the greatest amount of variability in hourly CO 2 concentrations at all sites. The variability in day-to-day average CO 2 concentrations was not well correlated with meteorological variables, although night-to-night variability was correlated with wind speed at all sites. Our findings show that near-surface concentrations of CO 2 were enhanced near the center of the Phoenix metropolitan area, even over large patches of vegetation several hectares in size, and vegetation type influenced this enhancement.