Discussion
Relative to previous studies determining Rm with traditional method that incubation temperature differs from the ambient soil temperature (Jenkinson & Ladd, 1981), we used a novel method to incubate soils at the MAT of their corresponding sites when MAT was higher than 10 °C. As Xu et al. (2017) reported, our measured Rm based on MAT may be more representative for soil microbial activity and also more useful in process modeling because MAT is near to ambient soil temperature. Furthermore, in the previous studies estimating Rm at the same incubation temperature for all soils collected from different sites may have a deviation in Rm values because Rm is generally positively related to incubation temperature (Hamdi et al., 2013; Wang et al., 2019). That is to say, Rm in high latitude biomes with low MAT (e.g., boreal forests) would be overestimated when incubation temperature was higher than their MAT, while in low latitude biomes with high MAT (e.g., tropical forests) Rm would be underestimated. Therefore, we believe that the findings from our novel incubation experiment may somewhat advance our understanding of Rm’s spatial variations and enhance the accuracy of predicting Rm and its feedback to global change under the environment change scenario.