温度系数（Q10，温度每变化10 ℃，呼吸速率的相对变化）不仅可以用来描述不同森林非同化器官（根系和树干）和土壤对温度升高的敏感性，并由此断定它们在全球变暖进程中的不同表现，而且是其呼吸总量定量估计中必不可少的参数。虽然目前已经进行了大量的研究，但不同研究者结论并不一致，影响我们对问题的整体把握。因此，有必要综合以往文献进行统计分析。该文综合大量文献，评述了林木非同化器官和土壤的Q10值频率分布、不同研究方法对Q10值的可能影响并探讨了它们对温度升高的敏感性。结果表明，不同非同化器官和土壤的Q10值差异较大，但具有相对稳定的分布中心范围。其中，土壤呼吸Q10值中，频率分布最集中的区域是2.0～2.5，占23%，其中超过80%的测定结果在1.0～4.0之间，中位数为2.74。 根系呼吸的Q10值，频率分布最集中的区域2.5～3.0，占33%，而大部分（>80%）的研究结果在1.5～3.0之间，中位数为2.40。树干呼吸的Q10值中，频率分布最集中的区域是1.5～2.0，占42%，而90%以上的测定结果在1.0～3.0之间，中位数为1.91。通过对比，发现不同非同化器官Q10值不同（树干<根系<根系与土壤共同体<去除根系土壤）。其中树干和根系的Q10值显著低于去除根系土壤的Q10值（p<0.05），表明土壤微生物活动对于未来全球变暖的反应要比木质化器官更敏感。此外，常绿植物的根系和树干呼吸的Q10值与落叶树木对应值差异不显著，说明同化器官叶片的着生时间长短对非同化器官Q10的影响不大。不同的CO2分析方法（碱吸收法，红外线测定技术和气相色谱方法）对土壤呼吸Q10值测定结果的影响不显著（p>0.10），根系分离方法（断根测定和壕沟隔断测定）也对根系呼吸的Q10值影响也不显著（p>0.10）。但是，与活体测定相比，离体测定树干呼吸显著提高了其Q10值。总体来看，不同林分相同非同化器官以及不同非同化器官呼吸的Q10值相对稳定但仍具有较大的差异性，研究方法也对结果产生一定影响，在进行呼吸总量的定量估计中应该注意这一点。今后研究的重点是进一步把影响森林非同化器官呼吸的外在因素和内在因素综合考虑于Q10值相关模型中，以便准确定量估计其呼吸总量，而研究难点是深入研究Q10值具有较大变异性的原因（如温度适应性）和内在机理以便更好的表征不同器官和生态系统组分对全球变暖的敏感性。
The temperature coefficient, Q10 (Fractional change in rate with a 10 ℃ increase in temperature), can describe the response of organisms to temperature increases as a result of global warming. It is also a necessary parameter for estimating CO2 efflux. Although many studies have focused on Q10 values, reported values are highly variable. To better understand the sensitivity of forests to global warming, we reviewed and summarized reported Q10 values in the literature. Our specific objectives were the following: 1) to calculate the frequency distribution of Q10 values for soil, tree root and tree stem respiration and compare the temperature sensitivity of these different forest ecosystem compartments; 2) to determine the Q10 values of evergreen and deciduous tree species and examine the methodological influences on their calculation; and 3) to discuss future Q10-related studies. We found that most Q10 values reported for soil, root and stem respiration fell within a relatively narrow range although there were some outliers. For soil respiration, the median Q10 value was 2.74 with 23% of the values falling between 2.0 - 2.5 and 80% falling between 1.0 to 4.0. The median Q10 value for root respiration was 2.40 with 33% of the values falling between 2.5 - 3.0 and 80% between 1.0 - 3.0. The median Q10 value for stem respiration was 1.91 with 90% of the values falling between 1.0 - 3.0. The stem respiration Q10 value was significantly less than both the root and soil respiration Q10 values. There were no significant differences between the Q10 values for root and stem respiration of evergreen and deciduous trees (p>0.10). Methods for CO2 analysis (Soda lime absorption, IRGA and chromatograph analysis) and root separation methods (Excised root and trenched box) did not have a significant effect on Q10 values of soil and root respiration (p>0.10), but in vitro measurements of stem respiration yielded a significantly higher Q10 value than in vivo methods (p<0.05). In general, although the Q10 values of stem and root respiration fell within a relatively narrow range, there still was considerable variation between and within reported values for stems and roots. More attention should be paid to the quantitative estimation of total CO2 efflux by Q10 related models. Future research should focus on the biochemical, environmental and biological factors that control respiration for more precise estimation of total CO2 efflux. The greatest challenge is to better understand the underlying mechanisms that result in the variation in Q10 values between habitats and tree components to make Q10 values more universal for representation of temperature sensitivity to global warming.