Developing Tools for Ecological Forestry and Carbon Management in Longleaf Pine
Dr. Lisa Samuelson | Auburn University
The objective of this research was to develop the research knowledge necessary to create carbon management models that can be used to improve integrated natural resource management of longleaf pine. Specific objectives were to quantify carbon in ecosystem pools in longleaf pine forests located across the species’ range, and develop longleaf pine carbon models that can be used to evaluate life-cycle carbon balance, forest structure and biodiversity, and yield of forest products in longleaf pine forests.
Carbon density in above and belowground live and dead plant mass and in soils was measured in stands ranging in age from 5 to 118 years located across the range of longleaf pine at US Army Fort Benning, US Army Fort Polk/Kisatchie National Forest, Marine Corps Base Camp Lejeune and Eglin Air Force Base. Carbon in decaying longleaf pine tap roots was also measured. Field data were used to directly quantify ecosystem carbon and calibrate and validate the longleaf pine carbon models. The longleaf pine carbon model is driven by two linked forest carbon cycle models: an even-aged longleaf pine model (LLM-EA), which enables simulation of scenarios for young (0-70 years) planted stands being managed for transition toward uneven-aged structures with silvicultural tools such as thinning and prescribed fire, and a single-tree-based longleaf pine model (LLM-ST) which enables simulation of older (50 to > 200 years) stands which are managed with silvicultural tools such as single tree or group selection harvests and prescribed fire.
Variation in soil C concentration was largely dependent on depth. Pyrogenic C was a minor component of soil organic carbon (SOC) of bulk soil, comprising 5-7% of the total SOC stock and thus did not contribute substantially to long-term C accumulation in soil within the forest stand where it was produced. There was no enhanced accumulation or deposits of pyrogenic C with depth (to 1 m) and pyrogenic C declined in proportion with SOC, conversely the proportion of SOC resistant to oxidation in SOC increased with depth. The residence time of SOC in surface soils was hundreds of years and at depths > 50 cm residence time increased to thousands of years. The project team's results indicate that soils are not strong sinks for atmospheric CO2 compared to C accumulation in biomass in longleaf pine forests. Measured ecosystem C stock (all pools except soil C) increased with stand age with a predicted asymptotic maximum of 119 Mg C ha-1. At 100 years, ecosystem C was 69% of the predicted maximum. Ecosystem C was driven by increases in aboveground live tree C related to stand age and basal area. Live root C, the sum of below-stump C, ground penetrating radar measurement of lateral root C and live fine root C, was on average 32% of ecosystem C. Live understory C, forest floor C, down dead wood C and standing dead wood C were a small fraction of ecosystem C. Long-term accumulation of live tree C combined with the larger role of belowground accumulation of live lateral root C than in other forest types indicates a role of longleaf pine forests in balancing the more rapid C accumulation and C removal associated with more intensively managed forests. In addition, longleaf pine tap root necromass was found to contribute to long-term forest C sequestration.
Using the LLM-EA developed in this project to evaluate the effects of silvicultural management systems on C sequestration over a 300 year simulation period, project team concludes that: i) site productivity was the major factor driving C sequestration in longleaf pine stands; ii) increasing rotation length increased C storage; iii) prescribed burning had a small effect on C sequestration; and iv) for medium quality sites, C sequestration of thinned 75-year rotation longleaf pine stands was similar than unthinned 25-year rotation slash pine stands. The results support the use of unthinned, long rotation longleaf pine stands for C offset projects. In model simulations with the LLM-ST, the project team found no significant sensitivity to initial placement of trees in the stand lattice after a few decades of simulation and coefficients of variation for tree biomass were typically less than 10%, and this variation also includes stochastic mortality and recruitment processes. Sensitivity to fire return interval increased with longer simulation periods. Large increases in mean longleaf pine dbh, height, and basal area after 250 years with a ten year fire return time reflect a population with poorer recruitment of smaller individuals. Although the mean longleaf tree size was larger, the smaller population of trees resulted in decreased longleaf biomass and C accumulation. Hardwood biomass increased with the ten year fire return time as a consequence of reduced fire mortality and competition from longleaf pine trees. The team examined a range of harvest intensities for both single tree and group selection harvests. The simulations indicate that a harvest return time of 20 years is not likely sustainable, particularly at the highest harvest intensities. With a target of 35 trees ha1 removed every 20 years, longleaf pine aboveground biomass was reduced by 75% of the non-harvested stand. Hardwood dominance increases dramatically in this scenario, even with frequent fire management. Increasing the harvest return time to 80 years appears to be sustainable, with most stand characteristics similar to the range found in non-harvested stands.
Allometric equation derived from comprehensive field measurements provide predictive equations for determining carbon in above and belowground biomass of longleaf pine trees across the range. Tap root decay functions allow prediction of residual carbon from previous harvests or known tree mortality. The LLM-ST model was extended to include carbon dynamics, coarse woody debris, and both single tree and group selection harvesting. The project's integrated model framework allows simulation of the transition from longleaf pine plantations to old-growth savanna management regimes. The LLM-EA provides an important new tool for estimating C stocks for regional assessments and for guiding future longleaf pine management.