Soil respiration (SR) is highly sensitive to future climate change, and particularly to global warming. However, considerable uncertainties remain associated with the temperature sensitivity of SR and its controlling processes. Using 384 field measurement data from 114 published papers and one book, this study quantifies the variation in the seasonal Q10 values of soil respiration, the multiplier by which respiration rates increase for a 10 °C increase in temperature, and its drivers across different sites. No significant correlation between Q10 and mean annual temperature or mean annual precipitation is found when statistically controlling seasonal changes in vegetation activity, deduced from satellite vegetation greenness index observations (normalized difference vegetation index, or NDVI). In contrast, the seasonal amplitude of NDVI is significantly and positively correlated with the apparent Q10 of SR. This result indicates that the variations of seasonal vegetation activity exert dominant control over the variations of the apparent Q10 of SR across different sites, highlighting the ecological linkage between plant physiological processes and soil processes. It further implies that the seasonal variation of vegetation activity may thus dominate the apparent seasonal temperature sensitivity. We conclude that the apparent Q10 value of SR estimated from field measurements is generally larger than the intrinsic temperature sensitivity of soil organic matter decomposition, and thus cautions should be taken when applying apparent Q10 values directly in ecosystem models. Our regression analysis further shows that when the amplitude of NDVI variation approximates 0 (and thus when the seasonality in vegetation activity is marginal), the residual Q10 of SR for soil temperature measured at 5 cm depth is about 1.5.