Changes in carbon storage and net carbon exchange one year after an initial shelterwood harvest at Howland Forest, ME

Although many forests are actively sequestering carbon, little research has examined the direct effects of forest management practices on carbon sequestration. At the Howland Forest in Maine, USA, we are using eddy covariance and biomet- ric techniques to evaluate changes in carbon storage following a shelterwood cut that removed just under 30% of aboveground biomass. Prior to harvest, the stand contained about 76 Mg C/ha (30 m 2 /ha basal area) in aboveground and belowground live bio- mass. Harvesting removed about 15 Mg C/ha (SEM 2.1) and created about 5.3 Mg C/ha (SEM 1.1) of aboveground and 5.2 Mg C/ha (SEM 0.7) of root/stump detritus. Leaf-area index (LAI) and litterfall declined by about 40% with harvest. Approxi- mately half of the harvested wood was used for paper products and half for longer-lived wood products. Eddy covariance mea- surements in a nearby unharvested stand indicated that net eco- system exchange (NEE) averages about 1.8 Mg C/ha/year of C sequestration. A comparison of NEE at unharvested and har- vested stands, both preharvest and postharvest, indicated that NEE declined following the harvest by about 18%, which is less than expected based on basal area and LAI changes. Soil respi- ration declined slightly (but nonsignificantly, P 0.23) with har- vest, suggesting no major soil C loss after harvest. When decay of paper and wood products is included in a preliminary carbon budget, we calculate a postharvest net source of C to the atmo- sphere for at least 5 years, assuming preharvest growth rates of trees. How quickly the carbon balance becomes positive will de- pend largely on whether postharvest growth rates increase.