Spatial variability of daytime CO₂ concentration with landscape structure across urbanization gradients, Shanghai, China

Cities play an important role in the global carbon cycle. However, direct measurements of CO₂ concentration in urban environments are still very limited. Using Shanghai as a case study, this paper investigated the spatial pattern of atmospheric CO₂ concentration and its relationship with landscape structure across urbanization gradients. From March to April 2014, CO₂ concentrations were measured at 2 m above ground level with a near-infrared gas analyzer along 6 transects with a total length of 335 km. The results showed that the mean near-surface CO₂ concentration among the 6 transects was 445.8 ± 40.5 ppm. The average CO₂ concentration in the inner city was higher (55.1 ppm) than that in the suburban area. Also, CO₂ concentration showed a significant spatial heterogeneity, with the highest CO₂ concentration in the northwest and the lowest in the southeast, in accordance with the urbanization gradients. Further analysis indicated that the spatial variability of CO₂ concentration was mainly influenced by the urban landscape structure and depended largely on the percent of impervious surface cover (ISA) with a positive correlation and on the lower explanatory power for the percent of vegetation cover (Veg) with a negative correlation. This indicated that the trend in atmospheric CO₂ in urban areas was likely to depend more on fossil fuel emissions than on vegetation change. The study also found that the Pearson's correlation (R) between CO₂ concentration and ISA or Veg achieved its highest value when the buffer distance was 5 km, which could be described by the stepwise regression equation CO₂ = 0.99ISA − 0.18Veg + 378.18 (R² = 0.44, p < 0.01).