We report and compare the results of long-term observations (1998–2006) of monthly mean soil CO2 fluxes and their carbon isotope composition, carried out at 3 sites with contrasting characteristics: 1) a grassland site located in the urban area of Krakow, southern Poland, which was exposed to anthropogenic impact for more than a century; 2) a mixed forest site; and 3) cultivated agricultural field site. A closed-chamber, dynamic sampling system was used to collect monthly cumulative samples of soil-respired CO2. The CO2 collected at the mixed forest site was enriched in 14C with respect to European free-atmosphere continental 14CO2 background (high-altitude station Jungfraujoch in Swiss Alps) by approximately 40%, while the urban site revealed 14C depletion by ∼30% against the same reference. The Δ14C values observed at the agricultural site were lying in between, clustering along the regional reference atmospheric Δ14CO2 trend curve. The Δ14C values of soil-respired CO2 at the urban site turned out to be indistinguishable from the Δ14CO2 values in the local atmosphere. For the estimation of mean turnover time of soil carbon for each of the monitored sites, we used a multicompartment model (MCM) accounting for input of carbon to the soil profile via deposition of fresh organic matter, as well as 3 different sources of CO2 in the soil profile: 1) root respiration; 2) “fast”; and 3) “slow” pools of soil carbon. The estimated mean turnover time of carbon in the “fast” carbon pool was ∼14 yr for both urban grassland and mixed forest sites, and ∼22 yr for the cultivated agricultural field. From the observed differences in Δ14C values of the measured fluxes of soil-respired CO2, we conclude that 14C content of the biogenic component in the local atmospheric CO2 is site-specific and may differ significantly from the regional atmospheric background Δ14CO2 value. Therefore, the assumption widely used in 14C-based assessments of the fossil-fuel contribution local atmospheric CO2 load, stating that 14C concentration in the biogenic CO2 component is equal to that of regional atmospheric reference value, needs to be carefully evaluated on a case-by-case basis.