Optimizing carbon sequestration in commercial forests by integrating carbon management objectives in wood supply modeling

This paper provides a methodology for generating forest management plans, which explicitly maximize carbon (C) sequestration at the forest-landscape level. This paper takes advantage of concepts first presented in a paper by Meng et al. (2003; Mitigation Adaptation Strategies Global Change 8:371–403) by integrating C-sequestration objective functions in existing wood supply models. Carbon-stock calculations performed in Woodstock^TM (RemSoft Inc.) are based on C yields generated from volume table data obtained from local Forest Development Survey plots and a series of wood volume-to-C content conversion factors specified in von Mirbach (2000). The approach is used to investigate the impact of three demonstration forest-management scenarios on the C budget in a 110,000 ha forest in south-central New Brunswick, Canada. Explicit demonstration scenarios addressed include (1) maximizing timber extraction either by clearcut or selection harvesting for greatest revenue generation, (2) maximizing total C storage in the forest landscape and in wood products generated from harvesting, and (3) maximizing C storage together with revenue generation. The level of clearcut harvesting was greatest for scenario 1 (≥15 × 10⁴ m³ of wood and ≥943 ha of land per harvesting period), and least for scenario 2 (=0 m³ per harvesting period) where selection harvesting dominated. Because softwood saw logs were worth more than pulpwood ($60 m⁻³ vs. $40 m⁻³) and were strategic to the long-term storage of C, the production of softwood saw logs exceeded the production of pulpwood in all scenarios. Selection harvesting was generally the preferred harvesting method across scenarios. Only in scenario 1 did levels of clearcut harvesting occasionally exceed those of selection harvesting, mainly in the removal of old, dilapidated stands early in the simulation (i.e., during periods 1 through 3). Scenario 2 provided the greatest total C-storage increase over 80 years (i.e., 14 × 10⁶ Mg C, or roughly 264 Mg ha⁻¹) at a cost of $111 per Mg C due to lost revenues. Scenarios 3 and 1 produced reduced storage rates of roughly 9 × 10⁶ Mg C and 3 × 10⁶ Mg C, respectively; about 64% and 22% of the total, 80-year C storage calculated in scenario 2. The bulk of the C in scenario 2 was stored in the forest, amounting to about 76% of the total C sequestered.