The research used the soil of Picea crassifolia forest distribution on altitude of 2 500～3 100 m in the eastern of Qilian mountains as the research object, and using the close-top tube at field sampling methods and laboratory analysis method to study on soil NH4+-N and NO3--N net mineralization rate and net N mineralization and mineralization rate, aimed to explore the relationship between the soil net nitrogen mineralization rate and temperature, precipitation, soil nutrient. With a view to improve productivity of Picea crassifolia forest in Qilian Mountain and guide the Picea crassifolia forest ecosystems of sustainable development and management to provide basal reference materials. The results were summarized as follows: (1) There was different significantly about net mineralization rate of soil nitrate with the increase of altitude. The net ammonia nitrogen mineralization rate, soil net nitrogen mineralization rate and mineralization rate as well. Soil net nitrogen mineralization and the mineralization rate showed the "W" word in the altitude gradient, and changes of the same with nitrate net mineralization rate with increasing of altitude, contrary to the law of nitrogen net mineralization rate. At an altitude of 2 800 m, net nitrate nitrogen mineralization rate and amount of soil net nitrogen mineralization and the mineralization rate achieves had the maximum value, which was 0.372, 160.3, 0.44 mg/(kg. d), respectively, the net ammonia nitrogen mineralization rate had minimum value (0.067 mg/(kg. d)); the lowest value appeared at an altitude of 2 900 m, which was 0.155, 94.7, 0.26 mg/(kg. d), respectively. At an altitude of 3 100 m, the net mineralization rate of ammonia nitrogen achieved the maximum 0.13 mg/(kg. d). (2)Regression analysis showed that soil net nitrogen mineralization were very significantly negative correlation with annual average temperature (P＜0.01), R2=0.717 3; which were very significantly positive correlation with annual rainfall (P＜0.01), R2=0.383 5, concluded that climate change affecting the quantity of soil net nitrogen mineralization degree was: the average annual temperature＞annual precipitation. (3) Regression analysis showed that soil net nitrogen mineralization were extremely significant positive correlation with the soil total nitrogen, organic matter, water content and pH value (P＜0.01), the R2 were 0.910 1, 0.906 0, 0.842 8, 0.797 9, respectively; which were very significant negative correlation with the soil bulk density (P＜0.01), R2=0.222 4. By a R2 value size the soil nutrient affecting the quantity of soil net nitrogen mineralization degree was: soil total nitrogen＞soil organic matter＞soil moisture＞soil pH＞soil bulk density.