USING THE PEAT CORE RECORD TO HELP PREDICT FUTURE PEATLAND C DYNAMICS - PROBLEMS AND UNCERTAINTIES

According to current estimates, peatlands in the Boreal and Subarctic regions of the northern hemisphere store 270 – 450 Pg of carbon, sequestered at a long-term average rate of 66 - 96 Tg per year. In order to predict whether these ecosystems will continue to act as a carbon sink under future climatic change scenarios, we need to improve our understanding of climate effects on carbon cycling in peatlands. Studies that have directly examined contemporary peatland C dynamics have mostly been short-term, and the paleoecological record contained in peat deposits can potentially contribute valuable information about past peatland – climate relationships over decadal to millennial timescales. However, reconstruction of past C accumulation rates from peat core data is problematic, because apparent accumulation rates reflect the balance between rates of organic matter production and its subsequent decay within the peat column. Models that can be fitted to dated profiles in order to infer net rates of organic matter accumulation in the past assume that organic matter input and decay rates have remained constant over time. Irrespective of whether these assumptions are realistic, they make such models ill-suited to the prediction of future changes in C dynamics resulting from climate change. Models that allow for a dynamic response of peatlands to environmental change in the past are necessarily simulation-based, i.e. not designed to be fitted to core data. The number of driving variables involved and possibility of non-unique solutions makes it impossible to validate such models using data from individual cores. Furthermore, there are unresolved questions relating to both the spatial and temporal scaling of information derived from peat cores, which complicates comparisons between paleoecological data and information on contemporary C cycling in peatlands. I here highlight some of these problems, and examine whether and how data on past carbon dynamics can make a useful contribution to our understanding of the present and future C balance of peatlands.