Modelling the interannual variability of net ecosystem CO2 exchange at a subarctic sedge fen

This paper presents an empirical model of net ecosystem CO2 exchange (NEE) developed for a subarctic fen near Churchill, Manitoba. The model with obse rved data helps explain the interannual variability in growing season NEE. Five years of tower-flux data are used to test and examine the seasonal beh aviour of the model simulations. Processes controlling the observed interan nual variability of CO2 exchange at the fen are examined by exploring the s ensitivity of the model to changes in air temperature, precipitation and le af area index. Results indicate that the sensitivity of NEE to changing env ironmental controls is complex and varies interannually depending on the in itial conditions of the wetland. Changes in air temperature and the timing of precipitation events have a strong influence on NEE, which is largely ma nifest in gross ecosystem photosynthesis (GEP). Climate change scenarios in dicate that warmer air temperatures will increase carbon acquisition during wet years but may act to reduce wetland carbon storage in years that exper ience a large water deficit early in the growing season. Model simulations for this subarctic sedge fen indicate that carbon acquisition is greatest d uring wet and warm conditions. This suggests therefore that carbon accumula tion was greatest at this subarctic fen during its early developmental stag es when hydroclimatic conditions were relatively wet and warm at approximat ely 2500 years before present.