EGI OpenIR
全球及中亚地区大气 CO2浓度三维时空分布特征及驱动因子分析
Alternative TitleThe Spatiotemporal Distribution of the Three-Dimensional Atmospheric CO2 Concentration and the Controlling Factors over Global and Central Asia
曹良中
Subtype博士
Thesis Advisor陈曦 ; 张弛 ; Philippe De Maeyer
2020-06-30
Degree Grantor中国科学院大学
Place of Conferral北京
Degree Discipline理学博士
Keyword大气 CO2 时空分布 碳排放 陆地生态系统 风场 Atmospheric CO2 Spatiotemporal Distribution Carbon Emission Terrestrial Ecosystems Atmospheric Circulation
Abstract大气中温室气体浓度的增加所引起的全球气候变暖已成为 21 世纪所面临的最为紧迫、影响最为深远的全球环境问题之一。作为最为重要的人为长生命期温室气体,大气二氧化碳(Carbon Dioxide, CO2) 在全球气候变化和碳循环中扮演重要角色。准确且全面地获取大气 CO2 浓度信息对于研究全球碳循环,减少人类对气候影响估算的不确定性,评价 CO2 减排成效以及制定减排政策具有重要意义。当前大气 CO2 浓度获取方式主要有三种:站点测量、遥感观测以及模型模拟,三种方式各有优缺点。本文首先借助于世界温室气体数据中心(World DataCentre for Greenhouse Gases , WDCGG ) 、 航 空 微 量 气 体 综 合 观 测 网( Comprehensive Observation Network for Trace gases by AirLiner project ,CONTRAIL) 以及总碳柱观测网( Total Carbon Column Observing Network,TCCON) 等站点观测数据验证了温室气体观测卫星(Greenhouse gases ObservingSATellite, GOSAT) 、 大气红外探测仪(Atmospheric Infrared Sounder, AIRS)观测数据以及碳追踪器(Carbon Tracker, CT) 模型模拟数据的数据精度。在精度满足后续分析的前提下,结合气象因子数据、社会经济数据、植被因子数据等辅助数据,利用线性趋势分析、经验正交分解、奇异值分解、相关分析以及延时相关分析等方法对近地面、对流层等不同高度上的全球及中亚地区大气CO2 空间分布特征、年际变化特征、 年变化特征、季节变化特征以及日变化特征进行了定量分析,并分析了碳排放、植被以及大气环流对这种时空分布特征的影响。主要研究内容以及主要结论如下:(1)GOSAT 近地面 CO2 产品、 AIRS 对流层中层 CO2 产品以及 CT2017 模型模拟的 CO2 产品具有高精度和高稳定性,可以用来捕捉不同高度上大气 CO2的时空分布特征。通过比较 GOSAT 近地面 CO2 产品与 18 个符合条件的WDCGG 地面 CO2 测量数据,发现两者之间具有相似的季节波动特征以及增长特征,两者之间具有强相关性,每个站点上两者之间的偏差均小于 2.3 ppm,而相关系数则均大于 0.85。通过比较 AIRS 对流层中层 CO2 产品与航空微量气体综合观测网(Comprehensive Observation Network for Trace gases by AirLiner project,CONTRAIL)在六个区域上基于飞机平台的观测数据,发现两者之间具有相似的季节波动特征以及增长特征,每个区域上上两者之间的偏差均小于 0.4 ppm,而相关系数则均大于 0.94。比较 AIRS 对流层 CO2 产品与 WDCGG 地面 CO2测量数据,发现 AIRS 反演的 CO2 季节波动较小且更晚取得极值。通过对 CT2017CO2 产品与 TCCON 站点观测数据以及 GOSAT、 AIRS 卫星观测数据的比较,发现 CT2017 CO2 与前者之间具有相似的季节波动特征及增长特征,两者之间具有强相关性(R2 = 0.883, RMSE = 2.564 ppm),每个站点上两者之间的偏差均小于 2.3 ppm,而相关系数则均大于 0.7。 CT2017 CO2 与后者空间分布特征基本相似,相关系数均大于 0.65。(2)全球、中亚近地面和对流层中部大气 CO2 浓度具有明显的空间异质性,且这种空间异质性随着高度的增加而逐渐减弱。北半球(陆地)近地面、对流层中部的 CO2 浓度高于南半球(海洋)。在北半球,随着纬度的升高,近地面、对流层中层 CO2 浓度先是显著升高,在 40°N(近地面)或 50°N(对流层中部)以后开始小范围内波动;在南半球,随着纬度的升高,近地面 CO2 浓度逐渐降低,而对流层中部大气 CO2 浓度则没有明显规律。 CO2 浓度高值带随着高度的增加逐步由北半球中高纬度地区转移至低纬度地区,并在 12184 m 高度处形成低纬度 CO2 浓度高,高纬度 CO2 浓度低的格局。中亚地区近地面 CO2浓度自西向东表现出高-低-高的空间分布格局,这一空间格局在 1259 m 高度上仍然存在,而对流层中部 CO2 浓度自北向南表现出低-高-低的空间分布格局。中亚地区年均浓度在对流层的不同高度上始终高于全球平均水平。随着高度的增加,全球及中亚地区大气 CO2 浓度持续减小。(3)全球、中亚近地面和对流层中部 CO2 浓度均具有年增长、季节循环及季节波动特征。在增长率方面,全球所有区域近地面、对流层中层大气 CO2 浓度均呈增长趋势,平均增长率分别为 2.197 ppm/a 和 2.107 ppm/a, 其中北半球增长率高于南半球,陆地增长率高于海洋, 且除了个别区域外,随着纬度的增加,无论是南半球还是北半球,近地面 CO2 浓度年增长率均增加。中亚地区 CO2 浓度的增长率从地面附近开始持续到对流层中上部始终均高于同时期全球平均水平。在空间分布上,中亚北部近地面 CO2 浓度的增长率高于南部,但对流层中层 CO2 浓度的增长率则恰好相反。随着高度的增加,全球 CO2 浓度的增长率先是逐渐减小,然后增加,接着又减小;中亚上空 CO2 的增长率持续减小。在季节变化方面,随着高度的增加,取得最大值及最小值的月份逐渐后移。全球近地面 CO2 浓度 4 月份(春季)时最大, 9 月份(夏季)时最小,中亚近地面 CO2浓度在 3 月份(春季)最大, 8 月份(夏季)最小。全球对流层中层 CO2 浓度在5 月份(春季)最大,在 1 月份(冬季)最小,而中亚对流层中层 CO2 浓度在 4月份(春季)最大, 1 月份(冬季)最小。近地面,随着纬度的增加,北半球季节波动先是明显增大,在 60°N-70°N 达到最大值,接着又逐渐减小;南半球季节波动没有明显规律。在对流层中层,随着纬度的增加,季节波动的幅度逐渐增大。中亚地区的季节波动高于全球平均水平,季节波动随着高度的增加而逐渐减弱。在日变化方面,低层大气的日变化特征较为明显,全球、中亚不同季节的大气 CO2浓度多是在上午 10 点或者 13 点取得最小值,而在夜晚 22 点或者凌晨 1 点取得最大值。中亚地区不同季节的日变化幅度大于全球平均水平。(4)全球、中亚地区的近地面、对流层中层大气 CO2 浓度的持续增长与化石燃料的持续排放正相关,且具有强相关性,其相关系数均大于 0.68。(5)全球、中亚地区的近地面、对流层中层大气 CO2 浓度的空间分布除与碳排放有关外,还与大气环流息息相关。全球近地面 CO2 浓度高值区形成的可能原因是该区域是 CO2 强排放区且气流下沉,全球近地面 CO2 浓度低值区的形成的可能原因是地势高,人类活动少,植被覆盖度高,此外水平运动以纬向运动为主也是可能原因之一。在盛行西风及下沉气流的影响下,欧洲地区排放的CO2 持续输入到中亚地区,这导致在中亚西部地区形成近地面 CO2 浓度高值区。在天山山脉附近时,由于西部输入 CO2 浓度逐渐稀释,地势阻隔作用以及上升气流的影响,该区域附近形成近地面 CO2 浓度低值区。全球对流层中部大气CO2 浓度低值区形成的可能原因是该地区存在较强的大气下沉运动,另外该地区强碳汇亚马逊森林的存在以及 CO2 强排放区对此区域影响较小也是对流层中部大气 CO2 浓度低值区形成的原因。全球对流层中部大气 CO2 浓度高值区形成的可能原因是该区域是 CO2 强排放区,在西风的作用下,形成 CO2 浓度高值带。从欧洲输送到中亚干旱区的 CO2 在上升气流的作用下,向上运动至对流层中部,使得中亚干旱区对流层 CO2 浓度在本地地面 CO2 排放较低的情况下比欧洲高。由于存在天山的阻隔,中亚干旱区对流层中部 CO2 浓度的高值区被分成了两部分。(6)全球、中亚地区的近地面、对流层中层大气 CO2 浓度的季节变化与植被固碳能力的季节变化有关,特别是北半球。大气 CO2 浓度季节波动幅度与温度以及光照强度的季节波动幅度有关。全球、中亚近地面、对流层中层 CO2 浓度与总初级生产力(Gross Primary Productivity, GPP) 均呈负相关关系,只是对流层中部 CO2 浓度相对 GPP 的波动有滞后(全球、中亚: 4 个月)且纬度越低,CO2 的滞后时间越短。近地面 CO2 浓度与 GPP 的相关性大小与该区域植被覆被类型有关。 哈萨克斯坦北部、新疆北部和天山周边地区近地面 CO2 浓度与 GPP具有高度相关性,该区域内土地类型为森林、草地和农作物区。哈萨克斯坦北部(50°N 以北)、天山以及阿尔泰山区域内 GPP 升高时,中亚 40°N 以北对流层 CO2 浓度显著减少。
Other AbstractGlobal warming caused by the increase of greenhouse gas concentration in theatmosphere has become one of the most urgent and widespread global environmentalproblems in the 21st century. As the most important man-made long-life greenhousegas, atmospheric CO2 plays an important role in global climate change and carboncycles. Obtaining precise and comprehensive information on atmospheric CarbonDioxide (CO2) concentrations is of great significance for the study of global carboncycles and reducing the uncertainty in the estimation of human impacts on climatechange. Currently, there are three main ways to obtain the atmospheric CO2concentration: station measurement, remote sensing observation and model simulation.Each method has its advantages and disadvantages. In this dissertation, the dataaccuracy of atmospheric CO2 products from Greenhouse gases Observing SATellite(GOSAT), Atmospheric Infrared Sounder (AIRS) and the CarbonTracker model wereverified by station observations from World Data Centre for Greenhouse Gases(WDCGG), Comprehensive Observation Network for Trace gases by AirLiner project(CONTRAIL) and Total Carbon Column Observing Network (TCCON). On thepremise that the accuracy of CO2 products meets the needs for subsequent analysis, weuse CO2 products from GOSAT, AIRS and CT2017 and other auxiliary data (e.g.,meteorological data, socioeconomic data, vegetation factor data) to survey thespatiotemporal distribution characteristics of atmospheric CO2 and the controllingfactors using linear regression, empirical orthogonal functions (EOFs), singular valuedecomposition (SVD), correlation analysis, and delay correlation analysis. The mainresearch contents and conclusions are as follows.(1) AIRS CO2 product, GOSAT CO2 product, CT2017 CO2 product are highlyaccurate and stable and can be used to capture the spatiotemporal variation ofatmospheric CO2 concentrations at different altitudes.(2) The distribution of near-surface and mid-tropospheric CO2 concentrations overthe globe and Central Asia displays notable spatial heterogeneity, which decreases withincreases of altitude. heterogeneity. Overall, the CO2 concentrations of the near-surfaceand middle troposphere are significantly higher in the Northern Hemisphere (NH) thanthe Southern Hemisphere (SH) and significantly higher over land than over the ocean.In the NH, with the increase of latitude, the CO2 concentrations of the near-surface and middle troposphere first increased significantly and then began to fluctuate in a smallrange after 40°N (near surface) or 50°N (middle troposphere). In the SH, with theincrease of latitude, the CO2 concentrations gradually decreased at near surface;however, there is no obvious regularity in the middle troposphere. With the increase oflatitude, the high value zone of CO2 concentrations gradually transferred from themiddle and high latitude areas of the NH to the low latitude areas and formed a patternof high CO2 concentrations in low latitudes and low CO2 concentrations over 12,184 maltitude. The near-surface CO2 concentrations in Central Asia show a high-low-highspatial distribution pattern from west to East, which still exists at a height of 1,259 m,while the CO2 concentrations in the middle troposphere show a low-high-low spatialdistribution pattern from north to south. The average annual concentration in CentralAsia is always higher than the global average in different troposphere heights. Withincreases of altitude, the atmospheric CO2 concentrations over the world and CentralAsia continue to decrease.(3) Near-surface and mid-tropospheric CO2 concentrations over the globe andCentral Asia display the characteristics of continuous growth and seasonal cycles. Interms of growth rate, CO2 concentrations over all regions of the world show anincreasing trend, with an average annual growth rate of 2.197 ppm/a at near surface and2.107 ppm/a at middle troposphere. The growth rate over the NH is significantly higherthan over the SH and significantly higher over land than over the ocean. Except forsome regions, the annual growth rate of near-surface and mid-tropospheric CO2concentrations increases in both the NH and SH with increases of latitude. The growthrate of CO2 concentrations over Central Asia is higher than that over the global averagein the same period from near the ground to the middle-upper troposphere. The growthrate of near-surface CO2 concentrations in the north of Central Asia is higher than thatin the south, although the growth rate of middle troposphere CO2 concentrations is justthe opposite. With increases of altitude, the growth of global CO2 concentrations firstdecreases, then increases, then decreases; the growth rate of CO2 over Central Asiacontinues to decrease. In the aspect of seasonal variation, with the increase of height,the months with maximum and minimum values gradually move backward. The nearsurface CO2 concentrations over the world is the highest in April (Spring) and the lowestin September (Summer). The mid-tropospheric CO2 concentration over Central Asia isthe highest in March (Spring), and the lowest in August (Summer). With the increaseof latitude, the amplitudes of seasonal fluctuations of near-surface CO2 concentration over the NH first increase gradually, reaches the maximum value at 60°N-70°N, andthen decreases gradually, and no obvious regularity is observed over the SH. Theamplitudes of seasonal fluctuations of mid-tropospheric CO2 concentration increasesgradually with the increase of latitude. The amplitudes of seasonal fluctuations overCentral Asia are larger than that over the globe at different altitudes. With increases ofaltitude, the amplitudes of seasonal fluctuations of atmospheric CO2 gradually weaken.In the aspect of diurnal variation, the characteristics of daily variation in the loweratmosphere are more obvious. The atmospheric CO2 concentration in the differentreasons over the world and Central Asia mostly reach the minimum value at 10:00 or13:00 and maximum value at 22:00 or 1:00. The amplitudes of diurnal variation overCentral Asia in different seasons is larger than that over the globe.(4) The continuous increase of near-surface and mid-tropospheric CO2concentrations over the world and Central Asia is positively related to the continuousemissions from fossil fuels, with the correlation coefficients all greater than 0.68.(5) The spatial distribution of near-surface and mid-tropospheric CO2concentrations over the world and Central Asia is not only related to carbon emissionsbut also to the atmospheric circulation. The possible reason for the formation of thehigh-value areas of near-surface CO2 concentrations over the world is that there arestrong CO2 emissions and sinks of the airflow in the area. The possible reason for theformation of the low-value areas of near-surface CO2 concentration over world is thathigh terrain, few human activities and high vegetation coverage all exist in the areas. Inaddition, zonal motion is the chief form of horizontal motion. Under the influence ofprevailing westerlies and sinking of the airflow, CO2 emissions from Europe continueto be imported into Central Asia, which leads to the formation of a high-value area ofnear-surface CO2 concentrations over the western region of Central Asia. Under theinfluence of the input CO2 emission gradually diluting, the terrain blocking and therising of the airflow, the low-value area of near-surface CO2 concentration over CentralAsia is formed near the Tianshan Mountains. The low-mid-tropospheric CO2concentrations areas are formed as a result of relatively intense subsidence in theatmosphere, the presence of the Amazon rainforest and the lack of impact of high-CO2emissions. The high-mid-tropospheric CO2 concentrations areas are formed due to highCO2 emissions and the horizontal and vertical motions of winds. The formation of lowCO2 concentrations centres is principally related to vegetation. The formation of lowCO2 concentrations centres over Central Asia is principally related to vegetation. In addition, the low-CO2 concentrations centres are located over subregions with eitheropen or relatively high-elevation terrain. The formation of high-CO2 concentrationsareas is mainly affected by the land surface type, terrain, and atmospheric circulation.(6) The seasonal variation of near-surface and mid-tropospheric CO2concentrations over the world and Central Asia is related to the seasonal variation ofcarbon sequestration capacity of vegetation, especially in the NH. The seasonalfluctuation range of atmospheric CO2 concentrations is related to the seasonalfluctuation range of temperature and light intensity. There is a negative correlationbetween the near-surface and mid-tropospheric CO2 concentration over the world andCentral Asia and Gross Primary Productivity (GPP); however, carbon exchanges of landecosystems have no immediate impact on mid-tropospheric CO2 concentrations. GPPhas a four-month-delayed impact on mid-tropospheric CO2 concentrations over CentralAsia and globe. A lower latitude corresponds to a shorter time delay in the fluctuationsin CO2 concentrations.
Subject Area地图学与地理信息系统
Language中文
Document Type学位论文
Identifierhttp://ir.xjlas.org/handle/365004/15408
Collection中国科学院新疆生态与地理研究所
研究系统
Affiliation中国科学院新疆生态与地理研究所
First Author Affilication中国科学院新疆生态与地理研究所
Recommended Citation
GB/T 7714
曹良中. 全球及中亚地区大气 CO2浓度三维时空分布特征及驱动因子分析[D]. 北京. 中国科学院大学,2020.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[曹良中]'s Articles
Baidu academic
Similar articles in Baidu academic
[曹良中]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[曹良中]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.