Age-dependent changes in ecosystem carbon fluxes in managed forests in northern wisconsin, USA

The age-dependent variability of ecosystem carbon (C) fluxes was assessed by measuring the net ecosystem exchange of C (NEE) in five managed forest stands in northern Wisconsin, USA. The study sites ranged in age from 3-year-old clearcut to mature stands (65 years). All stands, except the clearcut, accumulated C over the study period from May to October 2002. Seasonal NEE estimates were -655 +/- 17.5 g C m(-2) in the mature hardwood (MHW), -648 +/- 16.8 in the mature red pine (MRP), -195 +/- 15.6 in the pine barrens (PB), +128 +/- 17.1 in the young hardwood clearcut (YHW), and -313 +/- 14.6 in the young red pine (YRP). The age-dependent differences were similar in the hardwood and conifer forests. Even though PB was not part of either the hardwood or conifer chronosequence, and had a different disturbance agent, it still fits the same general age relationship. Higher ecosystem respiration (ER) in the young than in the mature stands was the combined result of earlier soil warming in spring, and higher temperature and greater biological activity in summer, as indicated by temperature-normalized respiration rates. The fire-generated PB had lower ER than the harvest-generated YHW and YRP, where high ER was sustained partly on account of logging residue. During the main growing season, the equivalent of 31 (MHW), 48 (MRP), 68 (PB), 114 (YHW) and 71% (YRP) of daily gross ecosystem production (GEP) was released in ER during the same day. The lower ER:GEP ratio in the mature stands was driven by greater age-dependent changes in ER than GEP. The magnitude of the increase in ER:GEP ratio in spring and fall was interpreted as the extent of the decoupling of ER and GEP. Decoupling (sustained high ER despite decreasing GEP) was observed in YHW, PB and MHW, whereas in coniferous stands (MRP and YRP) the stable ER:GEP ratio suggested preferential use of new photosynthates in ER. The results indicate that a great part of the variation in landscape-level C fluxes can be accounted for by mean stand age and associated parameters, which highlights the need to consider this source of heterogeneity in regional C balance estimates.