Estimating individual- and stand-level stem CO2 efflux in a subalpine forest: assessment of different extrapolation methods.

Accurate evaluation of the carbon release by stems is an important objective of many forest carbon cycling studies. However, few studies have assessed the applicability of different extrapolation methods of stem CO₂ efflux. Therefore, a reliable estimation method that considers both temporal and vertical variations is required. To overcome this problem, we conducted a field experiment to measure the stem CO₂ efflux rate of Abies fabri in subalpine forests, and evaluated the effect of three extrapolation methods (area-based estimation, volume-based estimation and integral method) on accurate estimation of stem respiration at the individual and stand levels. First, we calculated the stem carbon loss at the individual level using the three extrapolation methods, and discovered that the diameter at the breast height (DBH) played a major role in determining the magnitudes of stem respiration. Second, we summarized the relationship between the stem CO₂ efflux rate (E_s) and the DBH from previous studies. The E_s for small diameter stems tended to increase with height, but remained constant in trees with large DBH, implying a role of DBH threshold in determination of the E_s effect. Our calculations from three estimation methods supported the hypothesis that a threshold accounted for the different vertical variations in E_s. Specifically, the vertical variation of E_s primarily presented as two categories: E_s decreased with the stem height for small diameters (DBH < 30 or 40 cm), or E_s showed little change for stem height for large diameters (DBH > 60 cm). Based on these trends, we suggest a two-part approach to scale the individual tree stem CO₂ efflux, in which a volume-based scaling method is recommended for trees with small DBH (0–30 or 40 cm) and an area-based estimation is suitable for stems with large DBH (> 30 or 40 cm). Using this approach, the annual amount of stem carbon release for an immature forest and a mature forest were estimated as 1.80 and 4.81 t C ha⁻¹ year⁻¹, respectively. [ABSTRACT FROM AUTHOR]