| Boreal forests play a critical role in the global carbon cycle. Although the amount of carbon stored in the aboveground biomass of these forests is low, their peat-rich soils contain nearly half of global soil carbon stocks. Boreal regions are predicted to warm at a rate faster than the global mean over the next century, bringing the long-term stability of these soil carbon stocks into question. Warming-induced losses of carbon from these forests and soils represent a potential positive feedback to global climate.; We quantified the long- and short-term controls on the carbon balance of forests in Manitoba, Canada using several techniques. We measured carbon fluxes at 30 m in a mature black spruce forest using the eddy covariance technique. We developed and deployed towers to measure carbon fluxes at 2 m in two land cover types using a modified Bowen ratio method. Lastly, we analyzed data from a six-site chronosequence established by collaborators to investigate the effect of stand age on carbon exchange, and used the data to parameterize a satellite-driven model for net carbon exchange.; The mature black spruce ecosystem shifted from a source (+61 g C m -2, 1995) to a sink (-21 gCm-2, 2004) of CO2 during the study period. Temperature, water table depth, and radiation were critical factors regulating half-hourly rates of photosynthesis, respiration, and net carbon exchange. Soil thaw and drainage were coupled, leading to high water tables and inhibited respiration in spring and early summer. Longer growing seasons did not enhance carbon sequestration due to increases in both photosynthesis and respiration. Ecosystem water balance influenced annual carbon exchange, with water surpluses inhibiting respiration and enhancing sequestration. A shift from negative to positive water balances during the study period drove the transition of the ecosystem from a carbon source to a sink.; Along the chronosequence, stand age and successional stage exerted a strong influence on annual photosynthesis, respiration, and net carbon balance. The model we developed accurately predicted diurnal carbon exchange at all six sites. This model shows promise as a tool for scaling-up site observations to produce regional estimates of carbon balance. |
