青藏高原植被降水利用效率的空间格局及其对降水和气温的响应

Spatial pattern of vegetation precipitation use efficiency and its response to precipitation and temperature on the Qinghai-Xizang Plateau of China

植被降水利用效率(precipitation use efficiency,PUE)是反映生态系统水、碳循环相互关系的重要指标。该文利用GLOPEM-CEVSA模型模拟了青藏高原2000-2008年植被净初级生产力(net primary production,NPP),以97个野外草地样点实测地上净初级生产力(above-ground net primary productivity,ANPP)对模拟NPP进行验证,模拟NPP与ANPP线性显著相关(R2=0.49,p<0.001)。利用降水量空间插值数据,分析了近9年青藏高原植被PUE的空间分布、主要植被类型的PUE及其与降水量之间的变化关系。结果表明:2000-2008年青藏高原地区植被年平均PUE沿东南向西北递减,降水量和气温对植被PUE有着重要的影响;PUE在不同植被类型间差异较大,其中农田PUE最高,高寒草甸PUE高于高寒草原。在不同降水区域植被PUE与降水量的关系不同,降水量低于90mm的区域,植被PUE值最低((0.026±0.190)gC·m-2·mm-1,平均值±标准偏差)、波动最大(变异系数CV=721%),与降水量和气温不相关(p=0.38)。降水量为90-300mm的地区,植被PUE较低((0.029±0.074)gC·m-2·mm-1,平均值±标准偏差)、波动较大(CV=252%),与降水量和气温显著相关(p<0.001),降水量和气温能够解释PUE空间变化的43.4%,其中降水量的影响是气温的1.7倍。降水量为300-650mm的区域占整个研究区的45%,主要植被类型为高寒草原,植被PUE较高((0.123±0.191)gC·m-2·mm-1,平均值±标准偏差),CV为155%;植被PUE的空间变化与降水量和气温极显著相关(p<0.001),降水量和气温能够解释植被PUE空间变化的97.8%,但以气温影响为主导,其影响是降水量的1.5倍。降水量为650mm的区域,植被PUE达到最高(0.26gC·m-2·mm-1)。降水量为650-845mm的区域主要是西藏林芝地区,植被以常绿针叶林为主,PUE最高((0.210±0.246)gC·m-2·mm-1,平均值±标准偏差)、波动最小(CV=117%);降水量和气温可解释植被PUE空间变化的93.1%(p<0.001),降水量的影响是气温的3.5倍,但其影响为负。

Aims Precipitation use efficiency （PUE） is a key to understanding the coupling between ecosystem carbon and water cycles. Our objective was to probe the spatial PUE pattern and its response to climate on the Qinghai- Xizang Plateau to better understand mechanisms of vegetation productivity and improve ecosystem process models. Methods GLOPEN-CEVSA model was applied to estimate net primary production （NPP） by using the Fraction of Photosynthetically Active Radiation Absorbed by Vegetation （MOD15A2）, and spatially interpolated meteorological data in 2000-2008. The modeled NPP was significantly correlated with the observed above-ground net primary productivity （R^2 = 0.49, p 〈 0.001, n = 97）. The PUE was calculated as the ratio of NPP to the annual sum of precipitation. Important findings The spatial pattern of PUE showed large differences among vegetation types. Crops had the highest PUE, and alpine meadow had higher PUE than alpine steppe. These differences were related to the precipitation and temperature distribution on the plateau. The PUE was relatively stable and the lowest value of （0.026 ± 0.190） g C·m^-2·mm^-1 （mean ± standard deviation） with the highest coefficient of variance （CV） of 721% was where precipitation was 〈 90 mm. Where precipitation was 90-300 mm, PUE was relatively stable and also low （（0.029±0.074） g C·m^-2·mm^-1） with relatively high CV （252%）. Together precipitation and air temperature in this precipitation range explained 43.4% of the spatial variance of PUE, and the effect of precipitation was 1.7 times that of temperature （p 〈 0.001）. The area with precipitation from 300-650 mm, mainly covered by alpine steppe （45%）, had relatively high PUE （（0.123 4±0.191） g C·m^-2·mm^-1） with a CV of 155%. The significant correlation of PUE with climate factors explained 97.8% spatial variance of PUE. Temperature had the dominant role, having 1.5 times the effect of precipitation. With increasing precipitation, PUE reached a peak of 0.26 g C·m^-2·mm^-1 at 650 mm of precipitation and then showed a decreasing trend. The precipitation of the mountainous Nyingchi region, Xizang, is 〉845 mm, and the region is mainly covered with evergreen needleleaf forest. It has relatively high PUE （（0.210 ± 0.246） g C·m^-2·mm^-1） with a minimum CV of 117%. Temperature and precipitation together explained 93.1% of the spatial variation of PUE for Nyingchi. Precipitation was negatively correlated with PUE and its effect was 3.5 times that of temperature.