Soil CO₂ efflux (R_S) is the second largest terrestrial carbon flux. Even a relatively small change in the carbon flow into, or out of, soils could have a large impact on atmospheric CO₂ concentration. Conversion of forest ecosystems can potentially be a major contribution to greenhouse gas emissions. Therefore, reviewing measured rates and spatial and temporal variability of soil respiration from different terrestrial ecosystems and evaluating the effects of environmental factors and human disturbance on them provides essential data for the global carbon budget and its estimated effect on climate change. The representative disturbance gradient (virgin forest < secondary forest < plantation) in the Liangshui Nature Reserve of the Xiaoxing'an Mountains of Northeast China provides an excellent opportunity to examine how soil respiration responds to human disturbances. Throughout the growing season (May-October) in 2010, an infrared gas exchange analyzer was used to measure and compare soil respiration rates and related environmental factors in various forest systems. The forest systems investigated were virgin mixed broadleaved-Korean pine forest, valley spruce-fir forest, selectively cut mixed broadleaved-Korean pine forest, secondary birch forest and Dahurian larch plantation. The results showed that there was significant seasonal variation in soil respiration in the different forest types during the growing season where the peak soil respiration rate was in July and August and the lowest respiration rate was in the early and late growing season. There was a significant exponential correlation between soil respiration and soil temperature at 5 cm (T₅) in the five forest types (P<0.001). Furthermore, soil temperature and soil moisture and the interactions between them could explain 71% of seasonal variation in soil respiration, and incorporating soil moisture into the pure R_S temperature model improved the prediction of R_S in most forest types. The average soil respiration rates in the growing season for the birch, spruce-fir, selectively cut mixed broadleaved-Korean pine, mixed broadleaved-Korean pine, and larch forests were 3.59, 3.52, 3.44, 2.58, and 2.29 μmolCO₂·m^-2·s^-1, respectively and the corresponding Q₁₀ values were 2.32, 2.10, 2.08, 1.95, and 1.84. The average soil respiration rate of the mixed broadleaved-Korean pine forest was 33% and 28% lower than the secondary birch forest and selectively cut mixed broadleaved-Korean pine forest, respectively, but 12% higher than the larch plantation. This indicated that soil respiration responded differently to human disturbance in the mixed broadleaved-Korean pine forest. The standard deviation (SD) of R_S for the six forest types showed a similar seasonal pattern to soil temperature where the mean SD of R_S in the six forest types was 1.03 μmolCO₂·m^-2·s^-1 between late June and early September and 0.51 μmolCO₂·m^-2·s^-1 in the early and late growing season. The average coefficient of variation (CV) among forest types was 26.40%. The CV of soil respiration rate between replicates in each forest type varied from 32.13% to 60.20%, significantly greater than those between plots (14.28%-35.70%), indicating that soil respiration was more variable on the fine scale. The CV of R_S between replicates in the mixed broadleaved-Korean pine forest and selectively cut mixed broadleaved-Korean pine forest (59.54% and 60.20%, respectively) were significantly (P<0.05) greater than those in valley spruce-fir forest and Dahurian larch plantation (44.39% and 32.13%, respectively). Although spatial variations in soil respiration for all the forest types were not significantly related to soil temperature, soil moisture could explain about 15.8%-33.5% of spatial variation of soil respiration within each forest type except the mixed broadleaved-Korean pine forest.