Urban soils are an important soil organic carbon (SOC) store, but soils are also a natural source of carbon dioxide (CO2) due to soil CO2 efflux derived from decomposition and root related respiration. Little is known about the temporal and spatial variability of SOC stocks and soil CO2 efflux in urban areas, in particular in the southern hemisphere. The objectives of our study were to quantify SOC stocks and soil CO2 efflux in tree (urban forests) and grass dominated (parklands) urban ecosystems and to investigate the underlying processes which determine urban SOC stocks and soil CO2 efflux. Soil organic C stocks were quantified to 30cm depth and soil CO2 efflux was determined using a closed chamber technique at 16 sites across Auckland, New Zealand. Vegetation and soil characteristics were measured to 30cm depth at each study plot. In addition, the diurnal and seasonal variability in soil CO2 efflux was assessed at six sites. Median soil organic C stocks (0–10cm) were significantly higher in parkland soils (4.8kgm−2, IQR: 0.9) compared to urban forest soils (2.7kgm−2, IQR: 0.1), which was largely due to higher bulk densities in parkland soils. In contrast, median soil CO2 efflux was not significantly different between parklands (5.2μmolm−2s−1, IQR: 0.8) and urban forests (4.5μmolm−2s−1, IQR: 1.9) sites. This suggests that increasing soil bulk densities without restricting root growth may enhance SOC stocks without increasing soil CO2 efflux. Soil temperature and soil water content explained 54% (urban parkland) and 71% (urban forests) of the temporal variation in soil CO2 efflux. Urban SOC stocks and soil CO2 efflux were similar to non-urban forests and grasslands in New Zealand, highlighting the importance of considering urban soils when assessing SOC stocks and soil CO2 efflux.