Simulation of spatio-temporal changes in evapotranspiration in typical mountains
WANGFeiyu1,2,, ZHANChesheng1,, HUShi1, JIAYangwen3, NIUCunwen3, ZOUJing4,5 1. Key Laboratory of Water Cycle and Related Land Surface Processes,Institute of Geographic Sciences and Natural Resources Research,Chinese Academy of Sciences,Beijing 100101,China2. University of Chinese Academy of Sciences,Beijing 100049,China3. State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin,China Institute of Water Resources and Hydropower Research,Beijing 100038,China4. Shandong Provincial Key Laboratory of Ocean Environment Monitoring Technology,Qingdao 266001,China5. Shandong Academy of Sciences Institute of Oceanographic Instrumentation,Qingdao 266001,China 通讯作者:通讯作者:占车生, E-mail:zhancs@igsnrr.ac.cn 收稿日期:2016-01-8 修回日期:2016-08-30 网络出版日期:2017-02-25 版权声明:2017《资源科学》编辑部《资源科学》编辑部 基金资助:国家重点基础研究发展计划(2015CB452701)国家自然科学基金项目(41571019,51209224) 作者简介: -->作者简介:王飞宇, 女,陕西西安人,硕士生,主要从事流域水循环模拟研究。E-mail:wangfy.14s@igsnrr.ac.cn
关键词:蒸散发;时空变化;CLM;CERES;太行山地;横断山地;黔桂喀斯特山地 Abstract Based on a new land surface model CLM_CERES,spatio-temporal changes in evapotranspiration(ET)in three typical mountains of China (Taihang Mountain,Hengduan Mountain,and Qiangui Karst Mountain) in baseline of 1951-2005 and estimation period of 2006-2060 were simulated. The CLM_CERES model was constructed by considering the scheme of water exploitation and utilization,and coupled with the crop growth and development model CERES. This new land surface model was driven by the atmosphere external forcing data of multi-model ensemble data of CMIP5. The simulated results were validated with a data-driven estimate of global land evapotranspiration (MTE data)derived from observations from a global network of micrometeorological tower sites(FLUXNET). The results showed that ET simulated by CLM_CERES was consistent with the MTE data among all three regions at a monthly scale (R2= 0.76~0.88). From 1951 to 2060,the total ET increased significantly with linear trends of 0.981 and 0.757 mm/a (P < 0.001)over Taihang Mountain and Hengduan Mountain,of which vegetation ET dominated. However,the total ET increased insignificantly over Qiangui Karst Mountain,of which soil evaporation dominated. Intra-annual variation of ET in all three regions from 1951 to 2060 showed unimodal curves with a peak region from May to September. The spatial distribution of ET was mainly influenced by climate change and topographic factors over Taihang Mountain and Hengduan Mountain. The spatial distribution of ET over Qiangui Karst Mountain was more complex due to the particularity of dualistic structure between surface and ground water.
太行山地(35.0°N - 41.5°N,111.5°E - 116.5°E)、横断山地(24.5°N - 34.5°N,104.5°E -116.5°E)和黔桂喀斯特山地(21.5°N - 26.5°N,110°E - 116.5°E)分别位于中国华北平原、青藏高原和西南喀斯特地区(见图1),空间分布上从北至南,有着不同的气候和下垫面特征。太行山位于中国东部湿润区半湿润区与西部半干旱区干旱区的分界,面积约12万km2,属暖温带大陆性季风气候,四季分明,降雨分布极为不均,土壤贫瘠,砾石含量高,水分短缺胁迫严重影响农林系统功能,同时也是中国两大地势阶梯的过渡段,海拔50~3000m,峡谷众多,切割深,落差大[16]。横断山地跨四川省西部、西藏东部以及云南省西北部,是中国跨度最大和最典型的南北向山系,且是唯一兼有太平洋和印度洋水系的区域[17],面积约50.6万km2,位于亚热带气候区,水量丰富,干湿季明显[17],区内地形复杂,海拔400~7519m,岭谷高差大,降水、径流、气温等垂直差异显著[18]。黔桂喀斯特山地为西南喀斯特峰丛的重要组成部分,面积约15万km2,属热带-亚热带季风气候,气候温暖,雨量丰沛[19],海拔70~2100m,主要地貌类型为洼地、峡谷[20],由于特殊的岩溶水赋存的二元结构,使得大部分地区岩石裸露,不利于植被演替[21],生态极为脆弱。 显示原图|下载原图ZIP|生成PPT 图 1研究区位置及其高程 -->Figure 1Location and elevation of the research area -->
利用MATLAB随机函数在三个区域内分别选取随机采样点,对CLM_CERES模式模拟结果与MTE数据集的月蒸散量进行格网尺度的对比。在太行山,横断山和喀斯特地区两者的相关系数(R2)分别为0.88,0.76和0.81,均方根误差(RMSE)分别为16.34mm/a,13.39mm/a和19.27mm/a (图 2),两者呈现较好的相关性。此外,对蒸散量区域平均值进行对比,两者的相关系数(R2)分别为0.94,0.87和0.92,均方根误差(RMSE)分别为14.31mm/a,9.22mm/a和13.68mm/a(图 3,见第281页)。1982-2005年太行山地、横断山地和黔桂喀斯特山地多年平均模拟蒸散量分别为602mm/a、556mm/a、796mm/a,MTE蒸散量分别为469mm/a、491mm/a、895mm/a,两者的相对误差分别为28%、11%和-11%,说明在太行山和横断山CLM_CERES模拟值与MTE数据相比存在一定的高估,这与Shi等研究表明在全球大部分地区CLM模式模拟蒸散量高于MTE蒸散量的结果相一致[35]。但在喀斯特山地CLM_CERES模拟值低于MTE数据,这可能是由于CLM_CERES的地表参数输入不能精确描述喀斯特地区复杂的下垫面状况,使模拟值存在一定误差。 显示原图|下载原图ZIP|生成PPT 图 21982-2005年研究区月蒸散量CLM_CERES模拟值与MTE数据在格网尺度的对比 -->Figure 2Validation of simulated monthly evapotranspiration by CLM_CERES with MTE data in study area from 1982 to 2005 based on grid scale -->
显示原图|下载原图ZIP|生成PPT 图 31982-2005年研究区平均月蒸散量CLM_CERES模拟值与MTE观测数据的对比 -->Figure 3Comparison of simulated monthly evapotranspiration with observation based on MTE method in study area from 1982 to 2005 -->
1951-2005年与2006-2060年总蒸散量的空间分布如图 4所示,两个时段内总蒸散量的空间分布规律基本一致,表现为黔桂喀斯特山地>太行山地>横断山地,但三个区域内部蒸散量的变化具有差异性,太行山地和横断山地的总蒸散量大致体现为南高北低的特征,而黔桂喀斯特山地总蒸散量的空间分布差异不明显。 基准期(1951-2005年)太行山平均蒸散量为602mm(图 4a),蒸散量整体上呈现由西北向东南的增加趋势,在阶梯状地貌和气候的共同作用下,土壤水分和植被生长受到影响[44],使得蒸散发与海拔高程由西北向东南的递减趋势相反。横断山平均蒸散量为556mm,等值区主要沿经向山脉向南北延伸,这是由于区域内纵向山体发育,东西受到山体阻隔作用造成[39]。而在纬向上蒸散量从北至南呈递增趋势(图4a),且变化梯度较大,范围从158~834mm,这是因为南部亚热带地区受印度洋西南季风和太平洋东南季风控制,降水量大,光照强,因而总蒸发量大;而北部高山区受海洋暖湿气流影响微弱,海拔大多超过4000m,气温低、湿度大,蒸发量小。在纬度增加和海拔升高两种因素共同作用下,区域内各种自然地理过程呈现过渡性和复杂性特点。黔桂喀斯特山地平均蒸散量为796mm,与其他两个山地相比,其总蒸散量的空间分布特征并不显著。喀斯特地貌的显著特点是剧烈岩溶作用形成的地表地下双重结构,地表渗透强烈,储水能力低,土壤退化严重[40],因此山地蒸散主要受地质背景的制约,并且与复杂的地形地貌和多样的小生境紧密关联[41]。 显示原图|下载原图ZIP|生成PPT 图 41951-2060年研究区多年平均总蒸散量的空间分布 -->Fig.4Spatial distribution of long-term mean annual evapotranspiration in study area from 1951 to 2060 -->
1951-2060年太行山地和横断山地的总蒸散量呈增加趋势,增幅分别为0.981和0.757mm/a(P<0.001),其中基准期(1951-2005年) 蒸散量增加趋势不显著,预估期(2006-2060年)有显著增加趋势(P<0.05) (表1),说明未来气候变化是导致蒸散量变化的主要原因。1951-2060年黔桂喀斯特山地的总蒸散量呈上升趋势(0.190mm/a) (表1),蒸散量在基准期呈现波动下降趋势,在预估期波动上升,但变化趋势均不显著,说明喀斯特地区蒸散量的变化对气候变化的响应并不显著,可能主要受地质条件影响。 Table 1 表1 表11951-2060年研究区多年平均蒸散量的线性变化趋势 Table 1Linear trend of long-term mean annul evapotranspiration of the study area in study area from 1951 to 2060
区域
1951-2005年
2006-2060年
1951-2060年
太行山地
0.188
0.713*
0.981***
横断山地
0.229
0.774***
0.757***
黔桂喀斯特山地
-0.107
0.086
0.190
注:线性趋势单位为mm/a; *,***分别表示通过了0.05和0.001的显著性检验。 新窗口打开 显示原图|下载原图ZIP|生成PPT 图 51951-2060年研究区总蒸散量、植被蒸散量和土壤蒸发量的年际变化 -->Figure 5Inter-annual variation of total evapotranspiration,vegetation evapotranspiration and soil evaporation in study area from 1951 to 2060 注:由于蒸散量变化主要受植被蒸散影响,故图中并未标出土壤蒸发的拟合方程。 -->
1951-2060年CLM_CERES模拟的蒸散量季节变化如图 6所示,太行山地蒸散量季节变化最为显著,与降水量季节变化特征大体一致,夏季(6-8月)蒸散量最大,占全年总蒸散量的49%,而夏季降水量可达到全年降水量的65.7%[46]。横断山地蒸散量季节变化较为平缓(图 6),这可能是由于区域内海拔梯度变化大,不同海拔高度带的气候变化存在响应差异,因此区域平均从一定程度上弱化了研究区的内部差异。每年5月西南季风登陆向北推移,气流受山体阻隔抬升形成大量降水,致使横断山5-8月蒸散量最大,占全年总蒸散量的52%。黔桂喀斯特山地蒸散量季节变化也比较显著,5-9月蒸散量最大,占年总蒸散量的62%,其中春季(3-5月)蒸散量的增加幅度和秋季(9-11月)蒸散量的减小幅度都比较剧烈。 整体上看,三个山地蒸散量季节变化均呈现单峰曲线,其峰值位于5-9月,这与降水变化和植被生长规律密切相关。5月开始太阳辐射增强,近地层气温升高,同时降水增多,土壤水分充足;7-9月植被迅速生长,耗水量增多;10月植被发育成熟并开始衰老,耗水量相应减少[13,45]。但在地形、气候、植被类型、土壤质地等要素的综合影响下,蒸散量的季节变化特征具有区域差异。其中,横断山蒸散量的峰值区域相对平缓,可能由于其冠层截留蒸发量大[4],因此季节变化较为均匀。黔桂喀斯特山地蒸散量的峰值出现时间早于太行山和横断山,这是由于其蒸散量以土壤蒸发为主,5月、6月植株小,覆盖度低,地表大面积裸露,太阳直射使蒸发量显著增大;7-9月降水量大,空气湿度也较大,土壤蒸发响应减小,而且随着植被的生长,覆盖度增大,土壤在植被的遮荫下蒸发量减少[45],同时植被生长大量消耗土壤水,也会影响土壤蒸发,而在此期间由于植被蒸散增大,因此总蒸散量仍然保持在较高水平。 显示原图|下载原图ZIP|生成PPT 图 61951-2060年研究区蒸散量季节变化 -->Fig.6Intra-annual variation of evapotranspiration in study area from 1951 to 2060 -->
5 结论
基于考虑了水资源开采利用方案并与CERES农作物生长模型耦合的新型陆面模式CLM_CERES, 利用CMIP5多模式集合数据集驱动该模式,对中国典型山地太行山地、横断山地、黔桂喀斯特山地基准期(1951-2005年)和预估期(2006-2060年)的地表蒸散量进行模拟分析,使用基于全球通量观测网络(FLUXNET)逐点水热通量观测数据的地表蒸散发估算数据集(MTE数据)对基准期模拟结果进行了对比验证,结果显示: (1)太行山地、横断山地、黔桂喀斯特山地CLM_CERES的蒸散量模拟值与MTE数据在月尺度上均具有较好的相关性,两者间的相关性均在0.76~0.88之间。 (2)1951-2060年太行山地和横断山地总蒸散量呈显著增加趋势,增幅分别为0.981mm/a和0.757mm/a(P<0.001),与基准期相比,预估期的总蒸散量整体上均呈现增加趋势。 (3)1951-2060年太行山地和横断山地以植被蒸散为主,其中太行山植被蒸散以植被蒸腾为主,横断山以冠层截留蒸发为主;黔桂喀斯特山地的总蒸散量无显著增加趋势,以土壤蒸发为主,但其总蒸散量的年际波动仍主要受植被蒸散影响。 (4)三个区域蒸散量的季节变化均呈现单峰曲线,峰值位于5-9月,与降水和植被生长的季节变化规律密切相关。太行山地和横断山地蒸散量的空间分布主要受气候和地形影响,黔桂喀斯特山地受其特殊的地表、地下水二元结构影响,对蒸散量的响应机制相对复杂,还需进一步探讨。 The authors have declared that no competing interests exist.
[Zhang RH,Du JP,SunR.Review of estimation and validation of regional evapo-transpiration based on remote sensing [J]. Advances in Earth Science,2012,27(12):1295-1307.] [本文引用: 2]
[Liu SM,Sun ZP,Li XW,et al.A comparative study on models for estimating evapotranspiration [J]. Journal of Natural Resources,2003,18(2):161-167.] [本文引用: 1]
[YangF,Zhang WJ.Transpiration and soil evaporation of different vegetation communities in Taihang Mountain hill-scape [J]. Chinese Journal of Eco-Agriculture,2008,16(1):30-34.] [本文引用: 1]
[4]
LinY,Wang GX,Guo JY,et al.Quantifying evapotranspiration and its components in a coniferous subalpine forest in Southwest China [J]. Hydrological Processes,2012,26(20):3032-3040. [本文引用: 5]
[Wen ZQ,Yang ST,Song WL,et al.The numerical simulation on green water cycle of typical vegetation types in karst area [J]. Geographical Research,2010,29(10):1841-1852.] [本文引用: 1]
[Ma YM,Wang JM.A survey in the study of area evaporation (evapotranspiration)over the hetero-geneous landscape [J]. Plateau Meteorology,1997,16(4):446-452.] [本文引用: 2]
[ChenM,Pan ZD.Estimation of large area evapotranspiration by thermal infrared satellite data [J]. Advances In Water Science,1994,5(2):126-133.]
[8]
RangwalaI,Miller JR.Climate change in mountains:a review of elevation-dependent warming and its possible causes [J]. Climatic Change,2012,114(3-4):527-547. [本文引用: 1]
[Gao YC,LongD.Progress in models for evapo-transpiration estimation using remotely sensed data [J]. Journal of Remote Sensing,2008,12(3):515-528.] [本文引用: 1]
[Liu SF,Lin ZH.Validation of common land model using field experiment data over typical land cover types in East Asia [J]. Climate and Environment Research,2005,10(3):684-699.] [本文引用: 1]
[12]
KumarS,MerwadeV.Evaluation of NARR and CLM3. 5 outputs for surface water and energy budgets in the Mississippi River Basin [J]. Journal of Geophysical Research,2011,116(D8):185-212. [本文引用: 1]
[Luo LH,Zhang YN,ZhouJ,et al.Simulation and application of the land surface model CLM driven by WRF in the Tibetan Plateau [J]. Journal of Glaciology and Geocryology,2013,35(3):553-564.] [本文引用: 3]
[Zhu SG,Chen HS,ZhouJ.Simulations of global land surface conditions in recent 50 years with three versions of NCAR Community Land Models and their comparative analysis [J]. Trans Atmos Sci,2013,36(4):434-446.] [本文引用: 1]
[YongB,Zhang WC,Liu CS.Advances in the coupling study of hydrological models and land-surface models [J]. Journal of Glaciology and Geocryology,2006,28(6):961-970.] [本文引用: 1]
[Fan KF,Yang DC.Geo-morphologic system in Taihang Mountain area [J]. Journal of Changchun Institute of Technology (Natural Sciences Edition),2006,7(1):51-53.] [本文引用: 1]
[WangJ,Meng JJ.Characteristics and tendencies of climate change in the South western karst region of China in the recent 45 years [J]. Acta Scientiarum Naturalium Universitatis Pekinensis,2007,43(2):223-229.] [本文引用: 1]
[Wang LC,Shi YL.Formation process and rational use of water resources and transform of rainfall,surface water and underground water in karst mountains area in Southwest China [J]. Scientia Geographica Sinica,2006,26(2):173-178.] [本文引用: 1]
[Yao CH,Jiang ZC, Yuan DX.Vegetation karst effects on the karst area of southwest China [J]. Acta Geoscientia Sinica,2001,22(2):59-164.] [本文引用: 1]
[22]
ManabeS.Climate and the ocean circulation 1:I. The atmos-pheric circulation and the hydrology of the earth&#x02019;s surface [J]. Monthly Weather Review,1969,97(11):739-774. [本文引用: 1]
[TuG,LiuB,Wang SY.Spatial-temporal characteristics of surface dry/wet status in northeast China by NCAR/CLM3.5 [J]. Scientia Geographica Sinica,2012,32(6):746-751.] [本文引用: 1]
[25]
Oleson KW,DaiY,BonanG,et al. Technical Description of the Community Land Model (CLM)[EB/OL].(2014-05)[2016-01-08]. URL [本文引用: 4]
[ZouJ.Effects of Groundwater Exploitation and Crop Growth on Regional Climate[D]. Beijing:University of Chinese Academy of Sciences,2013.] [本文引用: 3]
[27]
Taylor KE,Stouffer RJ,Meehl GA.An overview of CMIP5 and the experiment design [J]. Bulletin of the American Meteorological Society,2012,93(4):485. [本文引用: 1]
[28]
Baldocchi D. TURNER REVIEW No.15.'Breathing'of the terrestrial biosphere:lessons learned from a global network of carbon dioxide flux measurement systems [J]. Australian Journal of Botany,2008,56(1):1-26. [本文引用: 1]
[29]
JungM,ReichsteinM,CiaisP,et al.Recent decline in the global land evapotranspiration trend due to limited moisture supply [J]. Nature,2010,467(7318):951-954. [本文引用: 3]
[30]
JungM,ReichsteinM,BondeauA.Towards global empirical upscaling of FLUXNET eddy covariance observations:Validation of a model tree ensemble approach using a biosphere model [J]. Biogeosciences,2009,6(10):2001-2013. [本文引用: 1]
[31]
JungM,ReichsteinM,Margolis HA,et al.Global patterns of land-atmosphere fluxes of carbon dioxide,latent heat,and sensible heat derived from eddy covariance,satellite,and meteorological observations [J]. Journal of Geophysical Research:Biogeosciences (2005-2012),2011,116(G4):245-255. [本文引用: 1]
[32]
Xiao JF,Zhuang QL,Baldocchi DD,et al.Estimation of net ecosystem carbon exchange for the conterminous United States by combining MODIS and AmeriFlux data [J]. Agricultural and Forest Meteorology,2008,148(11):1827-1847.
[33]
BeerC,ReichsteinM,TomelleriE,et al.Terrestrial gross carbon dioxide uptake:Global distribution and covariation with climate [J]. Science,2010,329(5993):834-838. [本文引用: 2]
[34]
Bonan GB,Lawrence PJ,Oleson KW,et al.Improving canopy processes in the Community Land Model version 4 (CLM4)using global flux fields empirically inferred from FLUXNET data [J]. Journal of Geophysical Research:Biogeosciences (2005-2012),2011,116(G2):96-101. [本文引用: 1]
[35]
ShiX,MaoJ,Thornton PE,et al.Spatiotemporal patterns of evapotranspiration in response to multiple environmental factors simulated by the community land model [J]. Environmental Research Letters,2013,8(2):199-201. [本文引用: 2]
[36]
YangF,Zhang WJ.Estimation of evapotranspiration with drainage lysimeters in the Taihang Mountain area,China [J]. Forestry Studies in China,2009,11(4):219-224. [本文引用: 2]
[Chen BL,Pan JY,Liao SS,et al.Evaporation calculation of in Guangxi and its temporal and spatial distribution characteristics [J]. Journal of Meteorological Research and Application,2008,29(1):29-33.] [本文引用: 1]
[38]
ZouJ,XieZ,YuY,et al.Climatic responses to anthropogenic groundwater exploitation:A case study of the Haihe River Basin,Northern China [J]. Climate Dynamics,2014,42(7-8):2125-2145. [本文引用: 1]
[ Zhu GF,He YQ,PuT,et al.Spatial distribution and temporal trends in potential evapotranspiration over Hengduan Mountains region from 1960 to 2009 [J]. Acta Geographica Sinica,2011,66(7):905-916.] [本文引用: 1]
[Zhao ZQ,Hou LS,Cai YL.The process and mechanism of soil degradation in karst area in southwest China [J]. Earth Science Frontiers,2006,13(3):185-189.] [本文引用: 1]
[Chen HS,Wang KL.Soil water research in karst mountain areas of southwest China [J]. Research of Agricultural Modernization,2008,29(6):734-738.] [本文引用: 1]
[Zhang XB,Wang KL.Ponderation on the shortage of mineral nutrients in the soil-vegetation ecosystem in carbonate rock-distributed mountain regions in southeast China [J]. Earth and Environment,2009,37(4):337-341.] [本文引用: 1]
[ZhangX,Xue JH,Xu XT,et al.Nutrient dynamics and hydrological process of karst forests in mountainous area of central Guizhou Province [J]. Journal of Plant Ecology (Chinese Version),2007,31(5):757-768.] [本文引用: 1]
[Yang YH,WatanabeM,Wang ZP,et al.Impacts of temperature and precipitation changes on soil moisture of Taihang Mountains [J]. Acta Geographica Sinica,2004,59(1):56-63.] [本文引用: 2]
[YangF,Zhang WJ.Transpiration and soil evaporation of different vegetation communities in Taihang Mountain hill-scape [J]. Chinese Journal of Eco-Agriculture,2008,16(1):30-34.] [本文引用: 3]
[Li FD,Song XF,Zhang QY,et al.Wavelet analysis of precipitation and its distribution in the piedmont region of Taihang mountains-a case study at Luancheng station [J]. Journal of Applied Meteoro-logical Science,2005,16(3):367-373.] [本文引用: 1]