Abstract:The sand-fixation plantations played an important role in cutting down desertification process in northern China. Soil nitrogen mineralization, which is a key ecological process in poor sandy ecosystems, can be used as a basic index to assess ecosystem function of tree plantations. In this study, soil nitrogen mineralization process and its availability were examined by the methods of in situ closed-top PVC tube incubation and ion exchange resin bags under plantations of pine and poplar, and their neighboring grasslands in the southeastern Keerqin region of China. Results indicated that the accumulative net mineralization in growing season (from April 15 to October 15, 2004) in the grassland (8.06 μg•g-1) and Pinus densiflora (9.06 μg•g-1) plantation was lower than that in both Pinus sylvestris var. mongolica (18.36 μg•g-1) and Populus simonii (17.88 μg•g-1) plantations (p<0.05). The ratios of nitrification rate to net mineralization rate in grassland, P. sylvestris var. mongolica, P. densiflora, and Populus simonii plantations were 0.88, 0.91, 0.82, 0.87, respectively, which indicated that the soil nitrification rate in the study area is more intensive than soil net ammonification processes, and the NO-3-N was the major N form absorbed by plants in the study area, which might due to the higher soil pH values (>6.6) and strong human disturbances in the study area. The percentages of net mineralization amount to soil total N contents in grassland, P. sylvestris var. mongolica, P. densiflora, and Populus simonii plantations were 2.49%, 4.42%, 2.77%, 4.26%, respectively, implying that N turnover rates in grassland and P. densiflora plantation were very low, while those in P. sylvestris var. mongolica and Populus simonii were relatively higher. The accumulative mineral N absorbed by resin from May 15 to October 1 in both the grassland (283.50 μg•g-1 dry resin) and Populus simonii plantations (297.00 μg•g-1 dry resin) was the lowest, while thatin P. sylvestris var. mongolica (440.10 μg•g-1 dry resin) was intermediate, and P. densiflora (835.65 μg•g-1 dry resin) was the highest (p<0.05). Soil relative nitrification rate (which is expressed by the ratios of the soil NO-3-N to mineral N) in the grassland, P. sylvestris var. mongolica, P. densiflora and Populus simonii plantations were 0.88, 0.87, 0.92, and 0.90, respectively, which also implied that soil NO-3-N accounted for most of the soil mineral N. From this study, we conclude that the in situ closed-top PVC tube incubation method is more accurate than the ion exchange resin method in measuring seasonal soil N mineralization rate, while it is more convincing to use both methods in exploring both soil N availability and N absorption by plants. Based on the corresponding results, we concluded that the three tree species mentioned above were selective sand-fixation tree species in the study area, and the rank of suitability was P. densiflora, P. sylvestris var. mongolica and Populus simonii. However, to comprehensively assess the ecosystem functions in the study area, the knowledge of N balance in sand-fixation plantations is necessary.