Disentangling critical drivers of stem CO2 efflux from Pinus elliottii trees in Subtropical China

Stem CO2 efflux (E-s) plays a critical role in forest carbon budgets and net ecosystem CO2 exchanges, but there is still a significant knowledge gap on Es and its controlling factors in subtropical forests where forest productivity and transpiration are both very high. In this study, E-s and the possible controlling factors such as stem temperature (T-s), sap velocity (v(s)), and other climatic variables were simultaneously measured in a Pinus elliottii plantation of Subtropical China from January 2014 to July 2015. Temporal dynamics of E-s followed similar trends as T-s at a 1-cm depth with bell-shaped curves. The monthly E-s values were significantly higher during the fast-growing season (April to October) than in the slow-growing season (November to next March). However, temperature sensitivity (Q(10), the relative increase of E-s with a 10 degrees C rise in temperature) fluctuated throughout the entire year without a clear pattern. Significant and exponential relationships were found between E-s and T-s, with correlation factors higher during the slow growing season than in the fast-growing season. Additionally, the coefficients of determination of Es to stem temperature were highly divergent with respect to tree size during the fast-growing season but not in the slow-growing season. The residuals (Delta E-s), calculated as the difference between the modeled fluxes (E-p) based on night-time data at zero sap flow and the measured fluxes (E-m) during the daytime when sap flow occurred, became more prominent during the fast-growing season. Thus, significant and positive correlations were observed between the ratio of Delta E-s, to E-p and v(s) during the fast-growing season (r(2) = 0.59, p < 0.01) but not in the slow-growing season. Combined with the maximum value of v(s), sap flows could potentially reduce the measured CO2 efflux to up to 25% of those predicted values on temperature alone during the daytime. Our results clearly demonstrated that temperature was not sufficient to quantify E-s, and thus, the effect of sap flow on E-s must be integrated into any models simulating stem respiration and carbon budgets in forest ecosystems. (C) 2017 Elsevier B.V. All rights reserved.