Abstract:

Aims Stem CO₂ efflux (E_s) is an important component of annual carbon budget in forest ecosystems, but how biotic and environmental factors regulate seasonal and inter-specific variations in E_s is poorly understood. The objectives of this study were: (1) to compare seasonal dynamics in E_s for four temperate coniferous tree species in northeastern China, including Korean pine (Pinus koraiensis), Korean spruce (Picea koraiensis), Mongolian pine (Pinus sylvestris var. mongolica), and Dahurian larch (Larix gmelinii); and (2) to explore factors driving the inter-specific variability in E_s during the growing and non-growing seasons.
Methods Ten to twelve trees for each tree species were sampled for E_s and stem temperature at 1 cm depth beneath the bark (T_s) measurements in situ with an infrared gas analyzer (LI-6400 IRGA) and a digital thermometer, respectively, from July to October 2013 and March to July 2014. The daily stem circumference increment (S_i), sapwood nitrogen concentration ([N]), and related environmental factors were monitored simultaneously.
Important findings The temporal variation in E_s for the four tree species overall followed the changes in T_s throughout the study period, with the maxima occurring in the summer months (late May to early July) characterized by higher temperature and more rapid stem growth and the minima in spring (late March to April) or autumn (October) having lower temperature. T_s accounted for 42%-91% and 56%-89% of variations in E_s during the growing (May to September) and non-growing (other months) seasons, respectively. Furthermore, apart from T_s, we also found significant regression relationships between E_s and S_i, relative air humidity and [N] during the growing season, but their forms and correlation coefficients were species-dependent. These results indicated that T_s was the dominant environmental factor affecting seasonal variations in E_s, but the magnitude of the effect varied with tree species and growth rhythm. Mean E_s for each of the four tree species was significantly higher in the growing season than in the non-growing season, whereas within the season there were also significant differences in mean E_s among the tree species (all p < 0.05). The temperature sensitivity of E_s (Q₁₀ value) did not differ significantly among the tree species during the growing season, ranging from 1.64 for Dahurian larch to 2.09 for Mongolian pine, but did differ during the non-growing season which varied from 1.80 for Korean pine to 3.14 for Dahurian larch. Moreover, Korean spruce, Mongolian pine and Dahurian larch had significantly greater Q₁₀ values in the non-growing season than in the growing season (p < 0.05). These findings suggested that the differences of the response of E_s to temperature change for different tree species were mainly from the non-growing season. Because the seasonality and inter-specific variability in E_s for these temperate coniferous tree species were primarily controlled by multiple factors such as temperature, we conclude that using a single annual temperature response curve to estimate the annual E_s may lead to more uncertainty.

Key words: stem CO₂ efflux, seasonal dynamics, inter-specific variations, driving factors, temperate forest, temperature