Spatial variation and controlling factors of temperature sensitivity of soil respiration in forest ecosystems across China

Aims Our objective was to determine the spatial variation of the temperature sensitivity of soil respiration (Q₁₀) and it’s controlling factors in forest ecosystems across China. Methods Based on published papers, the field measurement data of soil respiration were collected to build the dataset of annual Q₁₀ in forest ecosystems across China. Further, the spatial variation and the drivers of Q₁₀ in different forest types were analyzed. Important findings The results showed that 1) Q₁₀ ranges from 1.09 to 6.24, with a mean value (± standard error) of 2.37 (± 0.04) and no significant difference among different forest types; 2) When all forest types were considered, Q₁₀ increased with increasing latitude, altitude, soil organic carbon content (SOC) and soil total nitrogen content (TN), but decreased with increasing longitude, mean annual temperature (MAT) and mean annual precipitation (MAP). Climate (MAT, MAP) and soil (SOC, TN) factors together explained 32.8% variations in Q₁₀. MAT and SOC were considered as the primary factors driving the spatial variation of Q₁₀. 3) Q₁₀ of different forest types responded differently to climate and soil factors. Q₁₀ decreased with the increase of MAP in the deciduous needleleaf forest (DNF), while Q₁₀ showed no significant correlation with MAP in other forest types. Q₁₀ increased with the increase of TN in evergreen broadleaved forest (EBF), deciduous broadleaved forest (DBF), evergreen needleleaf forest (ENF), and the sensitivity of Q₁₀ to TN was the highest in EBF and the lowest in ENF. Although Q₁₀ showed concentrated distribution trend, more attention should be paid to the large range of variation in future C budget studies. The primary driving factors and the response to environmental factors of Q₁₀ varied among forest types. Under the scenario of future climate change, Q₁₀ may vary divergently among different forest types. Therefore, the divergent responses of key parameters of carbon cycle in different forest types to climate change should also be considered in future carbon-climate models.