Variations of Water and Carbon Fluxes in the Dryland of North China
To avoid encountering 504 error when downloading big files,please take CAJ download or PDF download by chapter as your first priority.
【Author in Chinese】 龚婷婷；
【Author's Information】 清华大学， 水利工程， 2017， 博士
【Abstract in Chinese】 我国北方荒漠区生态环境脆弱,其陆面与大气间水碳循环过程对气候变化和人类活动十分敏感。近年来,在气候变化背景下,该地区的气温不断升高;同时,人类活动范围的不断扩大和强度的不断增高改变了该地区的土地利用和植被覆盖,从而影响了地表水碳通量。因此,深入研究我国北方荒漠区水碳通量的时空变化规律及其对环境变化的响应机制,可为该地区的生态环境保护措施提供科学依据。本论文以我国北方典型荒漠区（主要包括内蒙及黄土高原北部荒漠区）为研究对象,首先基于榆林站和通辽站涡度相关系统的观测数据,分析了典型荒漠灌丛生态系统水碳通量的主要特征及其环境控制因素。结果表明,榆林站相较于通辽站,蒸散发（ET）强度较大、碳汇功能较明显,水分条件是造成该差异的主要因素。两站点ET的季节变化均主要受辐射控制,总初级生产力（GPP）和生态系统呼吸量(Reco)的季节变化主要受温度控制,净碳交换量（NEE）的季节变化主要受植被物候变化控制,NEE对土壤水极端干旱的敏感程度大于其对极端高温的敏感程度。其次,本研究结合榆林站2011-2016年间由人类活动导致的土地利用变化情况,定量分析了人类活动对水碳通量的影响。结果表明,从植被破坏到植被恢复过程中,ET和归一化ET均增加,且在植被恢复阶段,归一化ET增加更快、更明显。GPP,Reco和NEE的年际变化分别有84.5%,59.2%和81.6%是由人类活动导致。不合理的人类活动会加速该地区蒸散发速率、降低净碳吸收量,进而导致该地区植被生长环境恶化,加剧生态脆弱性。最后,本文进一步结合位于我国北方典型荒漠区的其他9个通量站的观测数据,采用支持向量回归（SVR）方法实现由站点到区域的水碳通量升尺度计算,并分析该地区水碳通量的时空变化规律。结果表明,SVR方法对站点的ET和NEE的模拟精度（R2）分别为0.76和0.64;研究区域的ET和NEE在2001-2015年间的平均年总量分别为204±6.3 mm和-63±7.2 gC/m2/yr,且ET和NEE在空间上均呈现阶梯状分布,由东南向西北逐渐减小。
【Abstract】 The dryland in northern China has a vulnerable eco-environment,where the water and carbon cycles are sensitive to climate change and human-induced land use/cover change(LUCC).During the past several decades,the significant increase of air temperature and the expansion of human activities have been changing the land use/cover types in this region.Therefore,studying on the spatial and temporal variations of water and carbon fluxes in the dryland ecosystem is vital and necessary,as well as its responses to environmental changes.This study will provide the scientific understandings for the eco-environmental protection in the typical dryland of north China,especially in Inner Mongolia and northern Loess Plateau.Two typical Sandy Land(Mu Us Sandy Land and Horqin Sandy Land)located in northern Loess Planteau and Inner Mongolia have been selected for field observations,and eddy covariance(EC)techniques were used to observe the water and carbon fluxes.Based on the observations,firstly,we analyzed the diurnal and seasonal patterns of water and carbon fluxes,as well as the main environmental controlling factors.The results showed that the evapotranspiration(ET)rate and carbon sequestration ability at the Yulin station that located in the Mu Us Sandy Land is higher and more significant than those at the Tongliao station that located in the Horqin Sandy Land.Regarding the seasonal variation,the dynamics of ET at Yulin station and Tongliao station are both controlled by radiation.Gross primary productivity(GPP)and ecosystem respiration(Reco)are mainly influenced by temperature,while net ecosystem exchange(NEE)is controlled by vegetation phenological change.The extreme water and temperature conditions have obvious effects on carbon fluxes,especially the extreme dry water condition.Secondly,with the LUCCs that resulted from human activities during 2011-2016 in Yulin station,this study quantitatively analyzed the impact of human-induced LUCC on water and carbon fluxes,respectively.The results showed that both ET and normalized ET were increased in the land degradation and vegetation rehabilitation processes,and the increase of normalized ET was faster and more obvious in the vegetation rehabilitation process.Human activities explained 84.5%,59.2%,and 81.6% of the inter-annual variability of GPP,Reco,and NEE.This study indicates that human-induced LUCC will accelerate the ET rate in this region and reduce the carbon sequestration capacity,which may lead to deterioration of the vegetation growth environment and more vulnerable ecosystem.Thirdly,this study collected data of water and carbon fluxes from other nine flux stations that are located in the typical dryland of north China.Together with the large scale remote sensing data(including surface temperature(LST)and vegetation index(NDVI))and meteorological data(including short solar radiation(Rs)and wind speed(Ws)),the support vector regression(SVR)method was used to develop an upscaling model.The gridded water and carbon fluxes were estimated by SVR method,and the special and temporal patterns of water and carbon fluxes were analyzed in the dryland of north China.The results showed that the SVR model could simulate ET and NEE with reasonable accuracy at all flux stations,with the determination coefficients of 0.76 and 0.64,respectively.The annual mean values of ET and NEE during 2001-2015 were 204±6.3 mm and-63±7.2 g C/m2/yr,respectively.In the typical dryland of north China,ET and NEE are gradually decreasing from the southeast to the northwest.