Abstract: Biochar is a carbon-rich solid product resulting from biomass heated in the absence of oxygen. Biochar application is deemed to have the potential for greenhouse gas mitigation. Dryland farming areas in Northwest China contribute tremendously to greenhouse gas emission. However, little studies have been conducted in the region involving the application of biochar to improve carbon sink and reduce carbon emission, and the optimal biochar application has remained uncertain. The aim of this study was to determine the effects of biochar on methane (CH₄), nitrous oxide (N₂O) and carbon dioxide (CO₂) emissions in the semi-arid regions. Observation was done for the whole growth period of spring wheat and the treatments consisted of six different biochar rates — CK (0 t.hm^-2), T1 (10 t.hm^-2), T2 (20 t.hm^-2), T3 (30 t.hm^-2), T4 (40 t.hm^-2) and T5 (50 t.hm^-2) based on Randomized Complete Block design with three replications. The carbon dioxide analyzer and static chamber-gas chromatographic techniques were used to continuously measure and analyze the greenhouse gases fluxes. Soil moisture and temperature were measured simultaneously with gas measurement. The results showed that dry spring field during whole growth period under different biochar treatments were the sources for CH₄, N₂O and CO₂. The trend of different biochar application rates in the average emission flux of CH₄ was CK (0.005 7 mg·m^-2·h^-1) > T1 (0.004 7 mg·m^-2·h^-1) > T2 (0.003 6 mg·m^-2·h^-1) > T3 (0.003 3 mg·m^-2·h^-1) > T4 (0.002 7 mg·m^-2·h^-1) > T5 (0.000 4 mg·m^-2·h^-1). Similar trend of average emission flux of N₂O were CK (0.230 5 mg·m^-2·h^-1) > T5 (0.151 3 mg·m^-2·h^-1) > T1 (0.144 1 mg·m^-2·h^-1) > T2 (0.135 3 mg·m^-2·h^-1) > T4 (0.125 0 mg·m^-2·h^-1) > T3 (0.098 9 mg·m^-2·h^-1). The average emission fluxes of CO2 were 0.449 2 μmol·m^-2·s^-1 (CK), 0.447 0 μmol·m^-2·s^-1 (T1), 0.430 3 μmol·m^-2·s^-1 (T2), 0.391 4 μmol·m^-2·s^-1 (T3), 0.408 0 μmol·m^-2·s^-1 (T4) and 0.416 4 μmol·m^-2·s^-1 (T5), respectively. The mean emission flux of CH₄ reduced with increasing biochar application level. The results also showed that the mean emissions fluxes of N₂O and CO₂ significantly increased when biochar input exceeded 30 t·hm^-2. Soil temperature and moisture were affected by biochar application. Both soil temperatures at 515 cm depth and soil moistures at 510 cm soil depth in biochar input exceeded 30 t·hm^-2 treatments were significantly different from that of those in the other treatments. The ranges of soil temperature and soil moisture for different soil layers under the CK treatment were highest among all treatments. The application of biochar reduced the variation range of soil temperature and soil moisture at different soil layers. CH₄ flux was significantly negatively correlated with soil temperature, while significantly positively correlated with soil moisture at 510 cm soil depth. The N₂O and CO₂ fluxes were negatively correlated with soil temperature at 1015 cm soil depth, however, they were significantly positively correlated with soil temperature at 2025 cm soil depth. In addition, soil temperature at 1520 cm soil depth also had significant positive correlation with average N2O flux. Moreover, soil moisture at 05 cm soil depth had significant negative correlation with average CH₄, N₂O and CO₂ fluxes. It is concluded that during the crop growth period, reasonable application of biochar could mitigate greenhouse gas emission in dry farmlands in the study area.