青海湖流域植被的净初级生产力估算

[目的]定量估算青海湖流域2001—2011年草地净初级生产力(net primary productivity,NPP),查明其时空演化特征,为流域草地可持续利用与生态建设及相关的政策制订提供科学依据。[方法]选取陆地生态系统碳循环模型(CASA模型),逐像元模拟2001—2011年青海湖流域草地生态系统NPP的时空变化。[结果](1)2001—2011年青海湖流域草地年均NPP为1.12×1013g/a,单位面积平均值为168.03g/(m2·a);(2)NPP在空间分布上呈现东南高,西北低的格局,这与流域水热因子在空间上的分布一致。近11a流域草地年均NPP总体呈上升趋势,年增加率约为1.74g/(m2·a),湖区北部、东部为主要增加区域;(3)青海湖流域草地NPP具有明显的季节变化特征,7月草地NPP达到最高,1月NPP最低。其中5—9月生长季草地的NPP占到了全年的90.40%。[结论]所选模型模拟精度较高,能够较好地反映流域NPP的空间分布规律和时间变化特征。可以认为改进后的CASA模型在气候资料稀缺的该地区进行模拟是可行的。

Estimating Net Primary Productivity of Alpine Grassland in Qinghai Lake Basin

[Objective] The net primary productivity(NPP) of vegetation from 2001 to 2011 in Qinghai Lake basin(QLB) was estimated quantitatively, and its space-time evolution characteristics were analyzed to provide a scientific basis for sustainable utilization of vegetation, ecological conservation and related policy making.[Methods] This paper estimated the NPP of Qinghai Lake basin grassland using carnegie-ames-stanford approach(CASA) model.[Results] (1) The average annual total NPP grassland in Qinghai Lake basin was 1.12×10¹³ g/a between 2001 and 2011, and the average per unit value was 168.03 g/(m²·a). (2) NPP spatial distribution presented a pattern that it was higher in the southeast region, and lower in the northwest, similar pattern was observed for the hydro-thermal condition of the whole basin. Annual Qinghai Lake basin NPP showed an upward trend from 2001 to 2011. Annual increase rate of per unit value of grassland NPP was 1.74 g/(m²·a). The northeast and east part of the basin were the main increase area of NPP. (3) Also, there existed obvious seasonal variations of NPP in Qinghai Lake basin that the highest value appeared in July and the lowest was in January. The NPP from May to September accounted for 90.40% of whole year.[Conclusion] CASA showed high accuracy for the simulation in our case. It well reflected the spatial distribution and time variation of the basin NPP. These implied that the improved CASA model is feasible to be used for NPP simulation in the region without enough climate data, like that in our case.