Globally, wetlands provide the largest terrestrial carbon (C) store, and restoration of degraded wetlands provides a potentially important mechanism for climate change mitigation. We examined the potential for restored saltmarshes to sequester carbon, and found that they can provide a modest, but sustained, sink for atmospheric CO2. Rates of C and nutrient cycling were measured and compared between a natural saltmarsh (high- and low-shore locations), claimed arable land on former high-shore saltmarsh and a managed realignment restoration site (high- and low-shore) in transition from agricultural land to saltmarsh 15 years after realignment, at Tollesbury, Essex, UK. We measured pools and turnover of C and nitrogen (N) in soil and vegetation at each site using a range of methods, including gas flux measurement and isotopic labelling. The natural high-shore site had the highest soil organic matter concentrations, topsoil C stock and below-ground biomass, whereas the agricultural site had the highest total extractable N concentration and lowest soil C/N ratio. Ecosystem respiration rates were similar across all three high-shore sites, but much higher in both low-shore sites, which receive regular inputs of organic matter and nutrients from the estuary. Total evolution of 14C-isotopically labelled substrate as CO2 was highest at the agricultural site, suggesting that low observed respiration rates here were due to low substrate supply (following a recent harvest) rather than to inherently low microbial activity. The results suggest that, after 15 years, the managed realignment site is not fully equivalent to the natural saltmarsh in terms of biological and chemical function. While above ground biomass, extractable N and substrate mineralisation rates in the high-shore site were all quite similar to the natural site, less dynamic ecosystem properties including soil C stock, C/N ratio and below-ground biomass all remained more similar to the agricultural site. These results suggest that reversion to natural biogeochemical functioning will occur following restoration, but is likely to be slow; we estimate that it will take approximately 100 years for the restored site to accumulate the amount of C currently stored in the natural site, at a rate of 0.92 t C ha−1 yr−1.