We used a life cycle greenhouse gas (GHG) accounting tool to test the sensitivity of Maine’s state-wide forest sector GHG emissions to changes in forest management. Inputs included forest cover data and growth and yield models for the state of Maine. We estimated net GHG emissions over 100- and 300-year time horizons of different management strategies across a range of carbon (C) pools and emission sources. C pools included: 1) storage in above- and below-ground live and dead biomass; 2) storage in forest products in use and in landfills; 3) harvest, transport, and manufacturing emissions; 4) avoided emissions (substitution; bioenergy), and 5) landfill methane fluxes. Continuation of the baseline forest sector was a net GHG sink throughout the 300-year modelling period. Increasing management intensity through greater use of even-aged management increased total emissions compared to the baseline. Scenarios that increase the area of no harvest set asides compared favourably with baseline GHG emissions predicted for reduced harvesting intensity scenarios when product substitution was not considered. The sensitivity of results to inclusion or exclusion of GHG pools, such as avoided emissions through product substitution, illustrates the importance of assumptions when evaluating complex LCA systems.