2002年—2018年太湖水体溶解二氧化碳浓度卫星遥感数据集

2002年—2018年太湖水体溶解二氧化碳浓度卫星遥感数据集 Monthly average satellite-estimated dataset of Lake Taihu’s dissolved carbon dioxide concentration from 2002 to 2018 1 2 first-author 齐天赐，1994年生，男，博士研究生， 研究方向为湖泊水环境遥感。E-mail： tcqi@niglas.ac.cn 齐天赐，1994年生，男，博士研究生， 研究方向为湖泊水环境遥感。E-mail： tcqi@niglas.ac.cn tcqi@niglas.ac.cn 1 3 corresp 段洪涛，1979年生，男，研究员， 研究方向为湖泊遥感与智慧流域。E-mail： htduan@niglas.ac.cn 段洪涛，1979年生，男，研究员， 研究方向为湖泊遥感与智慧流域。E-mail： htduan@niglas.ac.cn htduan@niglas.ac.cn 1 1 2 1 1 1 2 中国科学院南京地理与湖泊研究所 中国科学院流域地理学重点实验室, 南京 210008 中国科学院南京地理与湖泊研究所 中国科学院流域地理学重点实验室, 南京 210008 Key Laboratory of Watershed Geographic Sciences, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences,Nanjing 210008, China Key Laboratory of Watershed Geographic Sciences, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences,Nanjing 210008, China 中国科学院大学, 北京 100049 中国科学院大学, 北京 100049 University of Chinese Academy of Sciences, Beijing 100049, China University of Chinese Academy of Sciences, Beijing 100049, China 西北大学 城市与环境学院, 西安 710027 西北大学 城市与环境学院, 西安 710027 College of Urban and Environmental Sciences, Northwest University, Xi'an 710027, China College of Urban and Environmental Sciences, Northwest University, Xi'an 710027, China 湖泊在全球碳循环中发挥着重要作用，而湖泊中溶解CO 2 浓度（ <math id="M1"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653194&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653162&type= 4.40266657 3.72533321 ）控制着其CO 2 通量的方向及大小，是湖泊CO 2 排放估算的关键参数。中国第三大淡水湖——太湖虽然具有长期的野外监测数据，但其样点分布在空间和时间上并不均匀，很可能给其CO 2 排放的估算带来不确定性和偏差。有必要利用更高频率和覆盖范围地遥感手段来弥补野外监测在时空代表性上的不足。本文基于MODIS/Aqua数据反演的叶绿素a浓度、表层水温、漫衰减系数及光合有效辐射产品，通过二次多项式经验模型对太湖藻型湖区表层水体 <math id="M2"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653207&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653206&type= 4.40266657 3.72533321 进行逐像元的估算。并对结果进行数据质量控制和统计平均得到2002-07—2018-12长时序月平均 <math id="M3"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653207&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653206&type= 4.40266657 3.72533321 数据集。数据集由GeoTiff格式储存，使用GCS_WGS_1984地理坐标系，共包含198个文件。产品估算精度验证结果显示，星地同步像元-样点的遥感估算结果与野外数据的偏差在总体上较小（均方根误差RMSE=12.83 μmol·L -1 ，无偏百分比偏差UPD=24.03%）。同时遥感与野外数据估算的年均值在太湖各个湖区表现出很好的一致性（RMSE < 13.24 μmol·L -1 ，UPD < 25.82%），证明数据的可信度。产品的不确定性评估结果显示，在所有输入变量的随机误差影响下，月均 <math id="M4"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653204&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653201&type= 4.40266657 3.72533321 产品最大可能高估约30%。基于本数据集数据统计得到的结果显示，太湖 <math id="M5"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653189&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653186&type= 4.40266657 3.72533321 表现出明显的季节变化，冬春高夏秋低，西部高东部低；且太湖年平均 <math id="M6"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653207&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653206&type= 4.40266657 3.72533321 在数据集覆盖时间段内表现出显著下降趋势（0.80 μmol·L -1 ·a -1 ）。本数据集（下载方式： https：//doi.org/10.5281/zenodo.4729048 https://doi.org/10.5281/zenodo.4729048 ［2021-05-18］）月平均时间尺度同常规生态环境监测对应，便于分析对比，并且提供了空间分异信息，能够辅助研究深入理解太湖 <math id="M7"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653210&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653191&type= 4.40266657 3.72533321 乃至碳循环过程的时空变化规律，值得推广使用。 Lakes play an important role in the global carbon cycle. The dissolved carbon dioxide concentration ( <math id="M8"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653225&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653223&type= 4.48733330 3.72533321 ) controls the direction and amount of the lake CO 2 flux, which makes it one of the keys to the Lake CO 2 emission estimates. Due to the limitations of traditional field surveys on the spatiotemporal representativeness, large efforts of field surveys are still required to fulfil the requirements of monitoring lake <math id="M9"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653268&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653265&type= 4.48733330 3.72533321 dynamics. China’s third largest freshwater lake—Lake Taihu is a hot spot for lake carbon cycle and eutrophication research because of its complex environmental problems. Although Lake Taihu has long-term field limnological observations, including the measurements of physical, chemical, and biological parameters, the spatiotemporal distributions of sampling sites are still limited for the accurate estimation of the CO 2 emissions, which is likely to give uncertainty and deviation to its CO 2 emission estimates. It is necessary to take advantages of high frequency and wide range remote sensing technologies for achieving larger-scale and longer-term estimations of lake <math id="M10"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653219&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653228&type= 4.48733330 3.72533321 dynamics compared to field surveys. In this paper, we used the MODIS-derived chlorophyll- a concentration, lake surface temperature, diffuse attenuation coefficient of photosynthetically active radiation, and photosynthetically active radiation to estimate daily <math id="M11"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653268&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653265&type= 4.48733330 3.72533321 of Lake Taihu (the coefficient of determination R 2 =0.84, root mean square error RMSE=11.81 μmol·L -1 , unbiased percent difference UPD=22.46%). After data quality control, the daily <math id="M12"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653244&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653242&type= 4.48733330 3.72533321 were averaged on a monthly scale to obtain the monthly average <math id="M13"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653239&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653237&type= 4.48733330 3.72533321 of Lake Taihu. The data was stored in GeoTIFF grid format, with the GCS_WGS_1984 geographic coordinate system. The dataset contains 198 files of monthly average <math id="M14"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653268&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653265&type= 4.48733330 3.72533321 of Lake Taihu from July 2002 to December 2018. The uncertainty assessment results of the product show that under the influence of all input variables, the monthly <math id="M15"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653244&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653242&type= 4.48733330 3.72533321 product would overestimate about 30%. The differences between <math id="M16"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653268&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653265&type= 4.48733330 3.72533321 of pixel-sample matchups were small in total (Root mean standard error RMSE=12.83 μmol·L -1 , non-bias percentage deviation UPD=24.03%). The annual average of <math id="M17"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653268&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653265&type= 4.48733330 3.72533321 estimated by field observation and MODIS were consistent with each other in different regions of Lake Taihu (Root mean standard error RMSE < 13.24 μmol·L -1 , non-bias percentage deviation UPD < 25.82%). Based on the monthly average dataset, the <math id="M18"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653268&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653265&type= 4.48733330 3.72533321 of Lake Taihu showed significant seasonal dynamics, which were was low in summer and autumn (June to November) and eastern region, and high in winter and spring (December to May) and western region. Besides, the annual average <math id="M19"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653268&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653265&type= 4.48733330 3.72533321 showed a significant declining trend (0.80 μmol·L -1 ·a -1 , p < 0.01). This monthly average dataset (The download address is https://doi.org/10.5281/zenodo.4729048 https://doi.org/10.5281/zenodo.4729048 ) corresponds to the time scale of traditional limnological and ecological observations, which is suitable for comparison and analysis with traditional field datasets. Besides, the satellite dataset provides more spatial details of <math id="M20"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653268&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653265&type= 4.48733330 3.72533321 . It is very enlightening for better understanding of the biogeochemical process associated with <math id="M21"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653268&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653265&type= 4.48733330 3.72533321 in Lake Taihu. We believed this dataset would be very worth promoting to all researchers focusing on Lake Taihu. 湖泊遥感 MODIS 太湖 二氧化碳 碳排放 湖泊碳循环 数据集 MODIS Lake Taihu carbon dioxide carbon emission lake carbon cycle dataset 关联数据获取： https：//doi.org/10.5281/zenodo.4729048 https://doi.org/10.5281/zenodo.4729048 1　引言 1 近年来，大气中CO 2 浓度持续升高所带来的温室效应给社会和经济可持续发展带来严重威胁，已成为人类社会备受关注的热点问题。尽管全球湖泊总表面积仅占全球非冰川陆地面积的3.7%（ Verpoorter等，2014 Verpoorter等，2014 ），但是湖泊水体往往因为其相对大气过饱和的溶解CO 2 浓度（ <math id="M22"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653269&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653267&type= 4.99533367 4.23333359 ）而成为显著的CO 2 排放源（ Raymond等，2013 Raymond等，2013 ； Borges等，2015 Borges等，2015 ），其排放潜力可能抵消约22%的陆地生态系统CO 2 净吸收（ Ciais等，2013 Ciais等，2013 ），在全球碳收支中起到关键作用。然而，基于有限野外观测或间接计算的 <math id="M23"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 数据升尺度得到的全球CO 2 排放量的研究结果在数值上可以相差数倍（ Cole等，2007 Cole等，2007 ； Marotta等，2009 Marotta等，2009 ； Raymond等，2013 Raymond等，2013 ； Holgerson和Raymond，2016 Holgerson和Raymond，2016 ），准确估算全球湖泊CO 2 排放量仍是一项重大的挑战（ Drake等，2018 Drake等，2018 ），急需修订和改进。 受限于人力物力，目前已有的全球CO 2 排放估算研究中使用的湖泊 <math id="M24"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653304&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653303&type= 4.99533367 4.23333359 数据多是短期的有限样点采样方法，这样的低频观测缺乏时空代表性，难以满足准确估算湖泊CO 2 排放的需求。由于湖泊内部物理和生物地球化学过程的影响（ Balmer和Downing，2011 Balmer和Downing，2011 ； Natchimuthu等， 2017 Natchimuthu等， 2017 ； Xu等，2019 Xu等，2019 ）， <math id="M25"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653297&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653294&type= 4.99533367 4.23333359 表现出显著的时空异质性，需要较高的时空观测频率才能准确刻画 <math id="M26"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653274&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653272&type= 4.99533367 4.23333359 变化的年均水平（ Klaus等，2019 Klaus等，2019 ； Loken等，2019 Loken等，2019 ）。同时，目前在国内外只有极少数湖泊具有长时序、较高频率的连续观测记录（ Seekell和Gudasz，2016 Seekell和Gudasz，2016 ； Xiao等，2020 Xiao等，2020 ； Yan等，2021 Yan等，2021 ），而这些长期监测数据展示出 <math id="M27"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653281&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653278&type= 4.99533367 4.23333359 高动态变化，突出了长期监测对CO 2 排放估算的重要性，也对新的高频率、大范围观测手段提出要求（ 杨平和仝川，2015 杨平和仝川，2015 ； 段魏岩和黄昌，2021 段魏岩和黄昌，2021 ）。 卫星传感器可以提供高频率、连续和大范围的观测数据，弥补野外观测方法的不足（ 张兵 等，2021 张兵 等，2021 ）。利用遥感可反演的环境变量（例如叶绿素a浓度、水温、黄色物质吸收等）来表征水体中控制 <math id="M28"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 的相关生物地球化学过程，从而实现水体 <math id="M29"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653304&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653303&type= 4.99533367 4.23333359 的遥感估算，并重构其大范围、长时序的空间分布。卫星遥感技术在海洋及海岸带水体 <math id="M30"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653302&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653317&type= 4.99533367 4.23333359 估算研究中已经取得了成功的应用（ Bai等，2015 Bai等，2015 ； Chen等，2019 Chen等，2019 ），而在湖泊水体方面的研究也在稳步发展（ Kutser等，2015 Kutser等，2015 ； Qi等，2020 Qi等，2020 ）。对于瑞典湖泊，已有学者作了很多尝试工作，使用黄色物质吸收系数估算 <math id="M31"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653304&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653303&type= 4.99533367 4.23333359 （ Kutser等，2015 Kutser等，2015 ）。对于富营养化湖泊，也有学者建立了以叶绿素a为主要变量的 <math id="M32"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 估算模型，实现了太湖长时序 <math id="M33"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 空间分布的重构（ Qi等，2020 Qi等，2020 ）。 太湖作为中国第三大淡水湖，长期遭受富营养化问题困扰，频繁暴发蓝藻水华（ Duan等，2009 Duan等，2009 ），随之引发的各种环境问题也使得太湖成为生态环境和碳循环研究的热点（ Lee等，2014 Lee等，2014 ； 秦伯强，2020 秦伯强，2020 ），相关学者针对太湖进行了长期的野外观测。然而，虽然太湖的野外测数据时间序列长，但是其在空间和时间上的分布并不均匀（ Xiao等，2020 Xiao等，2020 ），因此基于现有数据得出的估计结果可能存在较大随机性和偏差。为了弥补实测数据的不足，本文以 Qi等（2020） Qi等（2020） 针对太湖构建的 <math id="M34"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 遥感估算模型为基础，先基于MODIS/Aqua数据反演的水色参数等产品，通过经验模型逐像元估算太湖藻型湖区表层水体 <math id="M35"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 ，再经统计平均得到2002-07—2018-12的月平均数据集。月平均 <math id="M36"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653344&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653312&type= 4.99533367 4.23333359 数据集在时间尺度上对应常规生态环境监测频率，在此基础上利用卫星更高的观测频率来增强月均数据的时空代表性，减小采样随机性带来的偏差，为太湖生态环境研究提供支撑。 2　数据研发方法 1 基于覆盖太湖全湖（ 图1 图1 ）的MODIS/Aqua遥感影像数据，本文生产了2002年—2018年太湖月平均 <math id="M37"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 空间分布数据集。数据研发流程如 图2 图2 所示，总体思路为：首先，获取原始卫星数据及产品，对数据进行预处理得到遥感反射率数据并对数据进行标准化与质量控制；接着，利用已发表的遥感算法反演得到关键水环境参数，进而估算太湖逐像元 <math id="M38"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653339&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653335&type= 4.99533367 4.23333359 ；最后，对 <math id="M39"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653360&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653342&type= 4.99533367 4.23333359 进行月尺度统计得到太湖月平均 <math id="M40"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653377&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653375&type= 4.99533367 4.23333359 空间分布数据集。 2.1　MODIS遥感数据获取、预处理及筛选 2 本数据集使用美国国家航空航天局（NASA）地球观测系统星座中Aqua卫星上搭载的中分辨率成像光谱仪MODIS（Moderate Resolution Imaging Spectrum-radiometer）传感器获取的太湖影像数据及其衍生产品作为模型输入变量估算太湖 <math id="M43"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 。从NASA海洋水色网（ https：//oceandata.sci.gsfc.nasa.gov/ https://oceandata.sci.gsfc.nasa.gov/ ［2021-05-18］）下载太湖区域2002-07—2018-12的MODIS/Aqua的L1A级数据供遥感反射率的计算使用，以及水面日平均光合有效辐射PAR（Einstein m -2 ·d -1 ）产品作为估算 <math id="M44"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 的输入变量。从NASA地球观测系统数据和信息系统网站EOSDIS（ https：//earthdata.nasa.gov https://earthdata.nasa.gov ［2021-05-18］）下载相同时间段的气溶胶光学厚度产品（MYD04_3K）用于大气校正，以及地表温度产品（MYD11A1）作为表层水温（LST）变量来估算 <math id="M45"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653435&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653434&type= 4.99533367 4.23333359 （ Liu等，2015 Liu等，2015 ）。 使用SeaDAS 7.5软件对L1A数据进行替代定标，并结合几何地理文件处理得到L1B级数据。接着使用SeaDAS 7.5中的 l 2 gen 工具，通过气象参数文件以及瑞利散射的查找表对L1B数据进行处理，得到经过水汽、臭氧校正与瑞利散射去除后的反射率（ R rc ）（ Hu等，2004 Hu等，2004 ）。另外，基于已有研究对主流大气校正算法在太湖的性能评估结果（ Shen等，2017 Shen等，2017 ），本文采用6SV模型对太湖L1B数据进行精确大气校正以获取遥感反射率（ R rs ）。具体步骤和参数选择如下： （1）观测几何：从MODIS辅助元数据中获取太阳天顶角、方位角、卫星天顶角和方位角。计算太湖区域平均观测几何输入6S模型。 （2）大气模型：选择中纬度大气模型，并根据影像日期选择冬季或夏季。 （3）气溶胶类型和光学厚度：选择大陆型气溶胶模型。气溶胶光学厚度使用MYD04_3K产品在太湖区域取有效陆地像元平均值作为550 nm处的光学厚度输入。该遥感产品在中国东部湖泊的适用性已得到验证评估（ Shen等，2020 Shen等，2020 ）。 （4）地面和传感器高度：地面高度输入0 km；传感器高度设置-705 km。 （5）光谱响应函数：输入NASA EOSDIS网站提供的MODIS传感器光谱响应函数。 （6）地面反射率特征：选择均一朗伯体和湖泊水体平均光谱。 将具有不同分辨率的MODIS反射率与产品数据重采样采样到250 m的分辨率，以确保所有数据都具有相同的空间分辨率以支持进一步的计算。之后，为便于进行后续一致性检验和水质参数的反演，对数据进行以下处理： （1）将数据进行裁剪，获取太湖区范围（30.8°N—31.6°N，119.8°E—120.8°E）的数据； （2）合成 R rc 数据的真彩色图像（RGB），通过目视解译将存在云覆盖率高、湖体结冰、太阳耀斑的影像排除，保留云覆盖低的影像共1242景，具体时间分布如 表1 表1 所示； 2 年份/年 12 月份/月 2 总计 1 2 3 4 5 6 7 8 9 10 11 12 2002 — — — — — — 4 3 5 9 8 2 31 2003 9 4 5 8 5 4 4 5 8 12 8 7 79 2004 5 10 8 12 6 3 7 7 7 9 4 2 80 2005 2 3 9 9 8 6 3 3 7 11 5 11 77 2006 4 2 6 5 6 4 6 7 6 7 3 6 62 2007 8 7 7 10 8 2 7 7 4 7 6 3 76 2008 2 9 8 3 5 1 4 3 2 4 5 11 57 2009 9 3 5 7 5 3 2 2 5 12 4 4 61 2010 7 5 8 6 5 5 2 16 7 9 9 15 94 2011 3 10 11 11 8 1 5 1 12 7 4 9 82 2012 6 4 10 9 10 4 11 5 8 11 5 6 89 2013 7 3 6 8 10 1 9 8 8 13 10 6 89 2014 7 2 11 7 5 3 3 3 2 15 5 15 78 2015 6 4 3 6 6 4 8 7 6 10 2 6 68 2016 5 11 5 3 2 3 4 5 5 1 7 9 60 2017 9 7 5 9 5 2 5 7 5 8 7 8 77 2018 3 9 10 9 5 8 9 10 4 7 4 4 82 总计 92 93 117 122 99 54 93 99 101 152 96 124 1242 （3）对于存在少量云像元的影像，将1240 nm波段的 R rc 大于0.1的像元视作云像元剔除； （4）使用浮藻指数（FAI）来剔除影像数据中的蓝藻水华像元（FAI > -0.004）（ Hu等，2010 Hu等，2010 ），以避免水华浮渣对水色参数反演的影响； （5）由于水生植被对遥感反射率（ Luo等，2020 Luo等，2020 ）和 <math id="M46"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 动态（ Xiao等，2017 Xiao等，2017 ）都存在显著影响，因此将水生植被区连同陆地像元通过太湖矢量边界剔除。 2.2　关键水色参数反演与 <math id="M47"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653431&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653430&type= 4.40266657 3.72533321 估算 2 叶绿素a浓度（Chla）使用 Shi等（2017） Shi等（2017） 提出的Chla算法反演得到。该算法通过构建645 nm与859 nm波段 R rc 值的归一化差值指数与Chla的经验关系实现对太湖水体Chla的反演。该算法的决定系数 R 2 =0.72，最低平均相对误差为24%，具有较高的反演精度。计算公式如下： <math display="block" id="M48"><mi mathvariant="normal">C</mi><mi mathvariant="normal">h</mi><mi mathvariant="normal">l</mi><mi mathvariant="normal">a</mi><mo>=</mo><mo>-</mo><mn mathvariant="normal">1454.3</mn><mo>×</mo><mfrac><mrow><mi mathvariant="normal">e</mi><mi mathvariant="normal">x</mi><mi mathvariant="normal">p</mi><mfenced separators="|"><mrow><msub><mrow><mi>R</mi></mrow><mrow><mi>r</mi><mi>c</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow></mfenced><mo>-</mo><mi mathvariant="normal">e</mi><mi mathvariant="normal">x</mi><mi mathvariant="normal">p</mi><mfenced separators="|"><mrow><msub><mrow><mi>R</mi></mrow><mrow><mi>r</mi><mi>c</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">859</mn><mo stretchy="false">)</mo></mrow></mfenced></mrow><mrow><mi mathvariant="normal">e</mi><mi mathvariant="normal">x</mi><mi mathvariant="normal">p</mi><mfenced separators="|"><mrow><msub><mrow><mi>R</mi></mrow><mrow><mi>r</mi><mi>c</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow></mfenced><mo>+</mo><mi mathvariant="normal">e</mi><mi mathvariant="normal">x</mi><mi mathvariant="normal">p</mi><mfenced separators="|"><mrow><msub><mrow><mi>R</mi></mrow><mrow><mi>r</mi><mi>c</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">859</mn><mo stretchy="false">)</mo></mrow></mfenced></mrow></mfrac><mo>+</mo><mn mathvariant="normal">69.35</mn></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653448&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653432&type= 89.57733154 11.68400002 式中， R rc （645）、 R rc （859）分别为MODIS数据中645 nm及859 nm波段 R rc 的值。 光合有效辐射漫衰减系数（ K d （PAR））通过490 nm处的漫衰减系数（ K d （490））推算得到，两者的经验关系及 K d （490）的反演根据 Huang等（2017） Huang等（2017） 针对太湖开发的半分析算法反演得到。该算法通过645 nm及531 nm波段的遥感反射率 R rs （645）与 R rs （531）的比值计算645 nm的颗粒后向散射系数 b bp （645）和吸收系数 a （645）（式（ 2 2 ）—（ 5 5 ）），进而计算645 nm的漫衰减系数 K d （645）（ 式（6） 式（6） ），最后由 K d （645）推算得到 K d （490）。由于该算法中各项经验系数是根据模拟与实测数据率定的，因此其遥感数据大气精校正的不确定性不会传递到参数中，故本文应用时不进行算法系数重新率定。计算公式如下： <math display="block" id="M49"><mi mathvariant="normal">l</mi><mi mathvariant="normal">n</mi><mfenced separators="|"><mrow><msub><mrow><mi>b</mi></mrow><mrow><mi>b</mi><mi>p</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow></mfenced><mo>=</mo><mn mathvariant="normal">3.456</mn><mo>×</mo><mi mathvariant="normal">l</mi><mi mathvariant="normal">n</mi><mfenced separators="|"><mrow><mfrac><mrow><msub><mrow><mi>R</mi></mrow><mrow><mi mathvariant="normal">r</mi><mi mathvariant="normal">s</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow><mrow><msub><mrow><mi>R</mi></mrow><mrow><mi mathvariant="normal">r</mi><mi mathvariant="normal">s</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">531</mn><mo stretchy="false">)</mo></mrow></mfrac></mrow></mfenced><mo>+</mo><mn mathvariant="normal">0.432</mn></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653451&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653449&type= 70.18866730 10.32933331 <math display="block" id="M50"><mi>a</mi><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo><mo>=</mo><mfrac><mrow><mn mathvariant="normal">1</mn><mo>-</mo><mi>U</mi><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow><mrow><mi>U</mi><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow></mfrac><msub><mrow><mi>b</mi></mrow><mrow><mi mathvariant="normal">b</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653471&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653468&type= 45.55066681 8.29733276 <math display="block" id="M51"><mtable columnalign="left"><mtr><mtd><mi>U</mi><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo><mo>=</mo><mfrac><mrow><msub><mrow><mi>b</mi></mrow><mrow><mi mathvariant="normal">b</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow><mrow><mi>a</mi><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo><mo>+</mo><msub><mrow><mi>b</mi></mrow><mrow><mi mathvariant="normal">b</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow></mfrac></mtd></mtr><mtr><mtd><mo>=</mo><mfrac><mrow><mo>-</mo><mn mathvariant="normal">0.084</mn><mo>+</mo><mo stretchy="false">(</mo><mn mathvariant="normal">0.084</mn><mo>+</mo><mn mathvariant="normal">4</mn><mo>×</mo><mn mathvariant="normal">0.17</mn><mo>×</mo><msub><mrow><mi>r</mi></mrow><mrow><mi mathvariant="normal">r</mi><mi mathvariant="normal">s</mi></mrow></msub><msup><mrow><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo><mo stretchy="false">)</mo></mrow><mrow><mn mathvariant="normal">1</mn><mo>/</mo><mn mathvariant="normal">2</mn></mrow></msup></mrow><mrow><mn mathvariant="normal">2</mn><mo>×</mo><mn mathvariant="normal">0.17</mn></mrow></mfrac></mtd></mtr></mtable></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653455&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653474&type= 75.52266693 18.88066673 <math display="block" id="M52"><msub><mrow><mi>R</mi></mrow><mrow><mi mathvariant="normal">r</mi><mi mathvariant="normal">s</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo><mo>=</mo><mfrac><mrow><mn mathvariant="normal">0.52</mn><msub><mrow><mi>r</mi></mrow><mrow><mi>r</mi><mi>s</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow><mrow><mo stretchy="false">(</mo><mn mathvariant="normal">1</mn><mo>-</mo><mn mathvariant="normal">1.7</mn><msub><mrow><mi>r</mi></mrow><mrow><mi>r</mi><mi>s</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo><mo stretchy="false">)</mo></mrow></mfrac></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653479&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653458&type= 43.26466751 9.14400005 式中， b bp （645）为645 nm的颗粒后向散射系数， R rs （645）与 R rs （531）分别为645 nm及531 nm波段的遥感反射率， a （645）为吸收系数， b b （645）为后向散射系数， U （645）为中间变量， r rs （645）为恰在水面以下的遥感反射率。 <math display="block" id="M53"><mtable columnalign="left"><mtr><mtd><msub><mrow><mi>K</mi></mrow><mrow><mi mathvariant="normal">d</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo><mo>=</mo><mn mathvariant="normal">1.25975</mn><mi>a</mi><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo><mo>+</mo></mtd></mtr><mtr><mtd><mn mathvariant="normal">3.27888</mn><mfenced separators="|"><mrow><mn mathvariant="normal">1</mn><mo>-</mo><mn mathvariant="normal">0.61748</mn><mi mathvariant="normal">e</mi><mi mathvariant="normal">x</mi><mi mathvariant="normal">p</mi><mfenced separators="|"><mrow><mo>-</mo><mn mathvariant="normal">3.56323</mn><mi>a</mi><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mrow></mfenced></mrow></mfenced><msub><mrow><mi>b</mi></mrow><mrow><mi mathvariant="normal">b</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo></mtd></mtr></mtable></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653484&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653461&type= 81.61866760 10.15999985 <math display="block" id="M54"><msub><mrow><mi>K</mi></mrow><mrow><mi mathvariant="normal">d</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">490</mn><mo stretchy="false">)</mo><mo>=</mo><mn mathvariant="normal">1.3729</mn><msub><mrow><mi>K</mi></mrow><mrow><mi mathvariant="normal">d</mi></mrow></msub><mo stretchy="false">(</mo><mn mathvariant="normal">645</mn><mo stretchy="false">)</mo><mo>+</mo><mn mathvariant="normal">0.4241</mn></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653498&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653495&type= 54.01733780 4.57200003 <math display="block" id="M55"><msub><mrow><mi>K</mi></mrow><mrow><mi mathvariant="normal">d</mi></mrow></msub><mo stretchy="false">(</mo><mi>P</mi><mi>A</mi><mi>R</mi><mo stretchy="false">)</mo><mo>=</mo><mn mathvariant="normal">0.8392</mn><msub><mrow><mi>K</mi></mrow><mrow><mi mathvariant="normal">d</mi></mrow></msub><msup><mrow><mo stretchy="false">(</mo><mn mathvariant="normal">490</mn><mo stretchy="false">)</mo></mrow><mrow><mn mathvariant="normal">0.9332</mn></mrow></msup></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653503&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653489&type= 46.39733505 4.82600021 太湖水体逐像元的 <math id="M56"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 估算根据 Qi等（2020） Qi等（2020） 针对太湖提出的经验算法进行计算。该模型以Chla、LST、 K d （PAR）及PAR作为输入变量，构建了多元二次多项式经验模型实现对太湖水体 <math id="M57"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 的估算。计算方程如下： <math display="block" id="M58"><mtable columnalign="left"><mtr><mtd><mi mathvariant="normal">l</mi><mi mathvariant="normal">n</mi><mo stretchy="false">(</mo><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub><mo stretchy="false">)</mo><mo>=</mo><mn mathvariant="normal">3.2613</mn><mo>+</mo><mn mathvariant="normal">0.0270</mn><mi>C</mi><mo>-</mo><mn mathvariant="normal">0.9186</mn><mi>L</mi><mo>+</mo></mtd></mtr><mtr><mtd><mn mathvariant="normal">0.7253</mn><mi>K</mi><mo>+</mo><mn mathvariant="normal">0.0335</mn><mi>P</mi><mo>+</mo></mtd></mtr><mtr><mtd><mn mathvariant="normal">0.0011</mn><msup><mrow><mi>C</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msup><mo>+</mo><mn mathvariant="normal">0.2431</mn><msup><mrow><mi>K</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msup></mtd></mtr></mtable></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653521&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653518&type= 62.56866837 13.46199989 式中， C ， L ， K ， P 为经过中心化的Chla，ln（LST），ln（ K d （PAR））和PAR数据。以上处理计算均在MATLAB 2016a软件中进行。 2.3　月平均 <math id="M59"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653543&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653540&type= 5.58799982 4.23333359 空间分布数据生成及产品展示 2 由于经数据质量控制后仍然可能存在少量异常像元，从而产生异常估算结果，故将日尺度的 <math id="M60"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 数据中数值小于0及超过176 μmol·L -1 （太湖野外采集数据中 <math id="M61"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 平均值+3×标准差（ Qi等，2020） Qi等，2020） ）的数据视为异常值进行剔除。之后，将日尺度数据按月份分别进行平均，得到2002-07—2018-12的MODIS月平均 <math id="M62"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 空间分布数据集， 图3 图3 展示的是由该数据集绘制的月平均 <math id="M63"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 空间分布。 基于数据集进行统计分析，发现太湖的月平均 <math id="M66"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 表现出较大的季节变化（ 图3 图3 ）。在夏季和秋季（6—11月）， <math id="M67"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653594&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653582&type= 4.99533367 4.23333359 较低，而在冬季和春季（12—5月）， <math id="M68"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653615&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653613&type= 4.99533367 4.23333359 较高。最低值发生在10月，最高值发生在1月。结合各湖区虚拟站点的 <math id="M69"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 时间序列变化，太湖中 <math id="M70"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653949&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653946&type= 4.99533367 4.23333359 的空间分布在冬春季差异很大，而在夏季和秋季则有所减弱（ 图3 图3 、 图4（a） 图4（a） ）。此外，1—5月，湖中 <math id="M71"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653624&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653611&type= 4.99533367 4.23333359 的最大值在空间上有所变化，从西南地区向西北湖区移动。在年际尺度上（ 图4（b） 图4（b） ），太湖年平均 <math id="M72"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653627&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653629&type= 4.99533367 4.23333359 在2005年前较为平稳，在2005年达到峰值，此后逐年降低。2003年—2018年整个时段太湖年平均 <math id="M73"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653647&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653644&type= 4.99533367 4.23333359 表现出显著下降趋势（0.80 μmol·L -1 ·a - 1 ）。 3　 <math id="M74"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653635&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653649&type= 6.60400009 5.67266655 遥感精度评估 1 3.1　验证数据获取及统计分析方法 2 本文中使用的野外观测验证数据来自太湖湖泊生态系统研究站（以下简称太湖站）2003年—2015年月度及季度湖泊水质调查的数据，通过水温、pH和总碱度计算得到表层水体 <math id="M75"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 。根据水文和营养状态将去除水生植被区的太湖分为西北湖区、梅梁湾、贡湖湾、湖心区及西南湖区5个子区域（ Xiao等，2017 Xiao等，2017 ），分别位于5个子区域的共21个采样点分布如 图1 图1 所示。太湖站野外观测数据同MODIS日尺度数据匹配的星地同步数据经过匹配准则（采样点采集时间与卫星过境时间差不超过±5 h，匹配像元周围3×3像元窗口内数据值的变异系数小于10%）筛选，共得到的80个匹配点对（ Qi等，2020 Qi等，2020 ）。匹配点对用于 <math id="M76"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653655&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653653&type= 4.99533367 4.23333359 估算模型的精度评估与模型不确定性分析。精度评估所使用的均方根误差（RMSE）、平均偏差（MB）、平均比（MR）、无偏百分比偏差（UPD）和平均相对偏差（MRD）等统计指标的定义与计算公式，模型不确定性分析方法的细节，请参考 Qi等（2020） Qi等（2020） 。 3.2　模型性能评估 2 Qi等（2020） Qi等（2020） 提出的 <math id="M77"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653949&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653946&type= 4.99533367 4.23333359 估算模型具有较高的精度，决定系数 R 2 =0.84，均方根误差RMSE=11.81 μmol·L -1 ，无偏相对误差UPD=22.46%，模型训练 <math id="M78"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653949&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653946&type= 4.99533367 4.23333359 变化范围为7.73—119.34 μmol·L -1 。星地同步数据的对比结果显示（ 图5（a） 图5（a） ， 表2 表2 ），模型在整个藻型湖区的总体验证精度较好（RMSE=12.83 μmol·L -1 ，UPD=24.03%），其中西北湖区、梅梁湾、湖心区和西南湖区的验证精度优于总体精度（RMSE < 12.22 μmol·L -1 ，UPD < 24.98%），而贡湖湾的验证精度较总体精度要差（RMSE=16.48 μmol·L -1 ，UPD=32.97%）。此外，贡湖湾的MR接近1，说明模型估算结果在总体均值上无偏，通过平均得到的月均值能在一定程度上抵消像元尺度的误差。值得注意的是，西南湖区的匹配样点只有2个，其统计指标可能缺乏统计意义。此外，太湖观测数据显示，MODIS/Aqua传感器过境时刻的实测 <math id="M79"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 与24 h连续观测得到的日均 <math id="M80"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653949&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653946&type= 4.99533367 4.23333359 高度一致（ R 2 =0.97， p < 0.01）（ Qi等，2020 Qi等，2020 ），说明卫星估算的 <math id="M81"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653840&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653839&type= 4.99533367 4.23333359 具有全天均值的代表性，适用于长时间尺度 <math id="M82"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653682&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653667&type= 4.99533367 4.23333359 动态变化的监测评估。 湖区 RMSE/μmol·L -1 UPD/% MRD/% MB/μmol·L -1 MR N 西北湖区 9.89 21.47 17.10 3.66 1.17 9 梅梁湾 12.22 20.38 0.70 -2.25 1.01 35 贡湖湾 16.48 32.97 -4.35 -7.17 0.96 16 湖心区 12.19 24.37 15.94 5.39 1.16 18 西南湖区 4.44 24.98 -1.22 -0.46 0.99 2 全藻型湖区 12.83 24.03 4.92 -0.81 1.05 80 1;2; 不确定分析结果显示， <math id="M91"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 估算模型的估算结果受输入变量的随机误差影响较小（RMSE=10.24 μmol·L -1 和UPD=26.14％），仅占模型构建数据集范围（7.73—119.34 μmol·L -1 ）的约9%。同时，在随机误差的影响下模型的估算结果存在高估（MR > 1），相比真实值最大可能高估约30%，这一高估情况可能传递至月平均 <math id="M92"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 数据。更多关于模型性能评估的细节请参考 Qi等（2020） Qi等（2020） 。 3.3　数据集精度评估 2 由于野外观测的采样频率为月度或季度，其时空代表性不足以直接评估月平均 <math id="M93"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 遥感数据集的精度，故将野外观测数据与MODIS月均 <math id="M94"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653754&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653752&type= 4.99533367 4.23333359 数据集统计至年尺度，即计算各个子区域中所有采样点或像元的年平均 <math id="M95"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653747&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653756&type= 4.99533367 4.23333359 值，对两者进行比较以评估数据集的估算精度。结果显示两者保持着较好的一致性（ 图5（b） 图5（b） ）。其中，西北湖区和西南湖区的两个数据集年均值差异要大于其他3个湖区，西北湖区MODIS估算结果偏低，西南湖区MODIS估算结果偏高，而全湖年平均值的差异要小于各子区域年均值的差异（ 表3 表3 ）。这主要是由于野外采样点空间分布不均匀及部分季度样点时间频率低导致的代表性不足对年平均估算结果的影响。西北湖区的月度采样点主要靠近河口区域，而开阔水域的样点区域均为季度采样点；西南湖区的采样点则全为季度采样点；而其他3个湖区中的采样点基本是月度采样点，进而导致了年均值估算的较大差异。 湖区 RMSE/μmol·L -1 UPD/% MRD/% MB/μmol·L -1 MR 西北湖区 13.24 20.40 -15.74 -9.34 0.84 梅梁湾 6.99 15.77 6.59 1.33 1.07 贡湖湾 7.29 18.21 6.32 0.98 1.06 湖心区 6.54 19.15 9.51 1.87 1.10 西南湖区 10.15 25.82 29.33 7.08 1.29 全藻型湖区 5.32 13.22 3.79 0.71 1.04 综上所述，虽然MODIS估算结果与野外观测数据仍有一定偏差，但是主要是由于数据本身时空代表性差异导致，本数据集对于深入理解太湖 <math id="M98"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653840&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653839&type= 4.99533367 4.23333359 乃至碳循环过程的时空变化规律仍具有极高的应用潜力，值得推广使用。 1,2;3; 4　数据集存储格式与命名 1 数据集中月平均 <math id="M105"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653822&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653820&type= 4.99533367 4.23333359 以Geotiff格式储存，数值单位为μmol·L -1 ，被掩膜的像元及被剔除的异常像元数值为NaN。数据文件的地理坐标系为GCS_WGS1984，数据命名方式为“Aqua_TH_yyyymm_CO2monthly.tif”，其中yyyy为年份，mm为月份。整个数据集共198个文件（下载方式： https：//doi.org/10.5281/zenodo.4729048 https://doi.org/10.5281/zenodo.4729048 ［2021-05-18］），从2002-07—2018-12，时间段内数据无缺失月份。 5　数据集优势与应用价值 1 本文提供的太湖月平均 <math id="M106"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653840&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653839&type= 4.99533367 4.23333359 数据集相比于前人发表文献中的数据，在时空代表性上均有提升与改进，更加适合相关但非对应专业学者分析使用。首先，在统计月平均数据时 <math id="M107"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653845&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653843&type= 4.99533367 4.23333359 表现出剧烈的时空变异（ 图6 图6 ），即使是在 <math id="M108"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653850&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653848&type= 4.99533367 4.23333359 较为均一的湖心区和西南湖区，以及夏秋季节（6—9月），其平均 <math id="M109"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 变异系数都大于15%，证实了传统野外观测方法在监测太湖 <math id="M110"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653829&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653856&type= 4.99533367 4.23333359 动态变化方面的局限性与不足。本数据集生产过程中每个月可利用影像数量为6.27±3.04景（平均值±标准差， 表1 表1 ），基本满足 <math id="M111"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 动态监测的频率需要（ Klaus等，2019 Klaus等，2019 ）。 1;2;3; 其次，本数据集经过质量控制与统计处理，除了少数月份因有效影像数据不足导致的数据缺失外，每个月份都能展示出完整的太湖藻型湖区 <math id="M114"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653897&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653893&type= 4.99533367 4.23333359 空间分布（ 图3 图3 ）。这相比于以往基于野外采样数据插值得到的空间分布结果（ Xiao等，2020 Xiao等，2020 ）具有更加合理且丰富的空间细节，有利于深度挖掘 <math id="M115"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 的时空动态变化。相比于日尺度的空间分布结果（ Qi等，2020 Qi等，2020 ），虽然时间分辨率有所下降，但是月平均数据极大程度的削减了由于云层、蓝藻水华覆盖所导致的数据缺失，降低了非遥感专业学者使用数据的门槛，而且月尺度的时间尺度可以更好的匹配传统生态环境野外调查的时间频率，支持其研究分析。 本数据集展示的太湖月尺度 <math id="M116"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653923&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653920&type= 4.99533367 4.23333359 空间分布，对太湖CO 2 通量估算以及区域碳收支研究有着重要意义。相比于以往以发表或共享的基于野外调查采样点数据估算的多年平均CO 2 通量空间分布（ Xiao等，2020 Xiao等，2020 ）或是基于涡度相关方法测量的梅梁湾观测站点的2012年—2014年CO 2 通量数据（ Xiao等，2014 Xiao等，2014 ），本数据集提供了太湖CO 2 通量更加细致的空间分异信息。以2018年的月平均 <math id="M117"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 数据为例，其与大气CO 2 浓度的饱和度比显示出剧烈的空间变化（ 图7 图7 ）。例如，秋季（9—11月）饱和度比在西北湖区沿岸表现出类似于河口羽状流的CO 2 排放热点（饱和度比 > 1），在开阔水域表现出大小不一的CO 2 吸收区域（饱和度比 < 1）。这些空间信息不仅有助于减小太湖CO 2 通量估算的不确定性，还有助于深入理解湖泊碳循环的时空过程。 6　结论 1 本文使用基于MODIS数据反演得到的水色参数以及NASA发布的卫星产品，通过经验模型对太湖藻型湖区表层水体 <math id="M120"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653949&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653946&type= 4.99533367 4.23333359 进行逐像元的估算，并对结果进行统计平均得到2002-07—2018-12的长时序月平均 <math id="M121"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 数据集，以期与历史湖泊调查数据的时间频率一致，并通过统计平均以增强数据集的时空代表性，进而提升其对深入理解湖泊中 <math id="M122"><msub><mrow><mi>c</mi></mrow><mrow><mi mathvariant="normal">C</mi><msub><mrow><mi mathvariant="normal">O</mi></mrow><mrow><mn mathvariant="normal">2</mn></mrow></msub></mrow></msub></math> http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653953&type= http://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=28653909&type= 4.99533367 4.23333359 时空变化规律与湖泊碳循环过程研究的应用潜力。本文给出了估算模型本身的性能评估以及不确定性分析结果，并进一步评估了遥感数据集与野外数据年均值的一致性，研究人员可以以此为依据判评估本产品对自身研究的适用性，并进一步分析其研究结果的不确定性大小及来源。此外，本文还展示了数据集的优势特点与应用价值，希望能给相关学者以启发。 参考文献（References） Bai Y , Cai W J , He X Q , Zhai W D , Pan D L , Dai M H and Yu P S . 2015 . A mechanistic semi-analytical method for remotely sensing sea surface pCO 2 in river-dominated coastal oceans: a case study from the East China Sea . Journal of Geophysical Research: Oceans , 120 ( 3 ): 2331 - 2349 [ DOI: 10.1002/2014JC010632 http://dx.doi.org/10.1002/2014JC010632 ] A mechanistic semi-analytical method for remotely sensing sea surface pCO2 in river-dominated coastal oceans: a case study from the East China Sea Balmer M B and Downing J A . 2011 . Carbon dioxide concentrations in eutrophic lakes: undersaturation implies atmospheric uptake . Inland Waters , 1 ( 2 ): 125 - 132 [ DOI: 10.5268/IW-1.2.366 http://dx.doi.org/10.5268/IW-1.2.366 ] Carbon dioxide concentrations in eutrophic lakes: undersaturation implies atmospheric uptake Borges A V , Darchambeau F , Teodoru C R , Marwick T R , Tamooh F , Geeraert N , Omengo F O , Guérin F , Lambert T , Morana C , Okuku E and Bouillon S . 2015 . Globally significant greenhouse-gas emissions from African inland waters . Nature Geoscience , 8 ( 8 ): 637 - 642 [ DOI: 10.1038/ngeo2486 http://dx.doi.org/10.1038/ngeo2486 ] Globally significant greenhouse-gas emissions from African inland waters Chen S L , Hu C M , Barnes B B , Wanninkhof R , Cai W J , Barbero L and Pierrot D . 2019 . A machine learning approach to estimate surface ocean pCO 2 from satellite measurements . Remote Sensing of Environment , 228 : 203 - 226 [ DOI: 10.1016/j.rse.2019.04.019 http://dx.doi.org/10.1016/j.rse.2019.04.019 ] A machine learning approach to estimate surface ocean pCO2 from satellite measurements Ciais P , Sabine C , Bala G , Bopp L , Brovkin V , Canadell J , Chhabra A , DeFries R , Galloway J , Heimann MJones C , Le Quéré C , Myneni R B , Piao S and Thornton P . 2013 . Carbon and other biogeochemical cycles // Climate Change 2013: The Physical Science Basis . Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY, USA : Cambridge University Press: 465 - 570 Carbon and other biogeochemical cycles Cole J J , Prairie Y T , Caraco N F , McDowell W H , Tranvik L J , Striegl R G , Duarte C M , Kortelainen P , Downing J A , Middelburg J J and Melack J . 2007 . Plumbing the global carbon cycle: integrating inland waters into the terrestrial carbon budget . Ecosystems , 10 ( 1 ): 172 - 185 [ DOI: 10.1007/s10021-006-9013-8 http://dx.doi.org/10.1007/s10021-006-9013-8 ] Plumbing the global carbon cycle: integrating inland waters into the terrestrial carbon budget Drake T W , Raymond P A and Spencer R G M . 2018 . Terrestrial carbon inputs to inland waters: a current synthesis of estimates and uncertainty . Limnology and Oceanography Letters , 3 ( 3 ): 132 - 142 [ DOI: 10.1002/lol2.10055 http://dx.doi.org/10.1002/lol2.10055 ] Terrestrial carbon inputs to inland waters: a current synthesis of estimates and uncertainty Duan H T , Ma R H , Xu X F , Kong F X , Zhang S X , Kong W J , Hao J Y and Shang L L . 2009 . Two-decade reconstruction of algal blooms in China’s Lake Taihu . Environmental Science and Technology , 43 ( 10 ): 3522 - 3528 [ DOI: 10.1021/es8031852 http://dx.doi.org/10.1021/es8031852 ] Two-decade reconstruction of algal blooms in China’s Lake Taihu Duan W Y and Huang C . 2021 . Research progress on the carbon cycle of rivers and lakes . China Environmental Science , 41 ( 8 ): 3792 - 3807 Research progress on the carbon cycle of rivers and lakes 段魏岩 , 黄昌 . 2021 . 河流湖泊碳循环研究进展 . 中国环境科学 , 41 ( 8 ): 3792- 3807 [ DOI: 10.19674/j.cnki.issn1000-6923.20210414.003 http://dx.doi.org/10.19674/j.cnki.issn1000-6923.20210414.003 ] 河流湖泊碳循环研究进展 Holgerson M A and Raymond P A . 2016 . Large contribution to inland water CO 2 and CH 4 emissions from very small ponds . Nature Geoscience , 9 ( 3 ): 222 - 226 [ DOI: 10.1038/ngeo2654 http://dx.doi.org/10.1038/ngeo2654 ] Large contribution to inland water CO2 and CH4 emissions from very small ponds Hu C M , Chen Z Q , Clayton T D , Swarzenski P , Brock J C and Muller-Karger F E . 2004 . Assessment of estuarine water-quality indicators using MODIS medium-resolution bands: initial results from Tampa Bay, FL . Remote Sensing of Environment , 93 ( 3 ): 423 - 441 [ DOI: 10.1016/j.rse.2004.08.007 http://dx.doi.org/10.1016/j.rse.2004.08.007 ] Assessment of estuarine water-quality indicators using MODIS medium-resolution bands: initial results from Tampa Bay, FL Hu C M , Lee Z , Ma R H , Yu K , Li D Q and Shang S L . 2010 . Moderate Resolution Imaging Spectroradiometer (MODIS) observations of cyanobacteria blooms in Taihu Lake, China . Journal of Geophysical Research : Oceans , 115 ( C4 ): C 04002 [ DOI: 10.1029/2009jc005511 http://dx.doi.org/10.1029/2009jc005511 ] Moderate Resolution Imaging Spectroradiometer (MODIS) observations of cyanobacteria blooms in Taihu Lake, China Huang C C , Yao L , Huang T , Zhang M L , Zhu A X and Yang H . 2017 . Wind and rainfall regulation of the diffuse attenuation coefficient in large, shallow lakes from long-term MODIS observations using a semianalytical model . Journal of Geophysical Research: Atmospheres , 122 ( 13 ): 6748 - 6763 [ DOI: 10.1002/2017jd026955 http://dx.doi.org/10.1002/2017jd026955 ] Wind and rainfall regulation of the diffuse attenuation coefficient in large, shallow lakes from long-term MODIS observations using a semianalytical model Klaus M , Seekell D A , Lidberg W and Karlsson J . 2019 . Evaluations of climate and land management effects on lake carbon cycling need to account for temporal variability in CO 2 concentrations . Global Biogeochemical Cycles , 33 ( 3 ): 243 - 265 [ DOI: 10.1029/2018GB005979 http://dx.doi.org/10.1029/2018GB005979 ] Evaluations of climate and land management effects on lake carbon cycling need to account for temporal variability in CO2 concentrations Kutser T , Verpoorter C , Paavel B and Tranvik L J . 2015 . Estimating lake carbon fractions from remote sensing data . Remote Sensing of Environment , 157 : 138 - 146 [ DOI: 10.1016/j.rse.2014.05.020 http://dx.doi.org/10.1016/j.rse.2014.05.020 ] Estimating lake carbon fractions from remote sensing data Lee X , Liu S D , Xiao W , Wang W , Gao Z Q , Cao C , Hu C , Hu Z H , Shen S H , Wang Y W , Wen X F , Xiao Q T , Xu J P , Yang J B and Zhang M . 2014 . The Taihu Eddy Flux Network: an observational program on energy, water, and greenhouse gas fluxes of a large freshwater lake . Bulletin of the American Meteorological Society , 95 ( 10 ): 1583 - 1594 [ DOI: 10.1175/bams-d-13-00136.1 http://dx.doi.org/10.1175/bams-d-13-00136.1 ] The Taihu Eddy Flux Network: an observational program on energy, water, and greenhouse gas fluxes of a large freshwater lake Liu G , Ou W X , Zhang Y L , Wu T F , Zhu G W , Shi K and Qin B Q . 2015 . Validating and mapping surface water temperatures in lake taihu: results from modis land surface temperature products . IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 8 ( 3 ): 1230 - 1244 [ DOI: 10.1109/JSTARS.2014.2386333 http://dx.doi.org/10.1109/JSTARS.2014.2386333 ] Validating and mapping surface water temperatures in lake taihu: results from modis land surface temperature products Loken L C , Crawford J T , Schramm P J , Stadler P , Desai A R and Stanley E H . 2019 . Large spatial and temporal variability of carbon dioxide and methane in a eutrophic lake . Journal of Geophysical Research: Biogeosciences , 124 ( 7 ): 2248 - 2266 [ DOI: 10.1029/2019JG005186 http://dx.doi.org/10.1029/2019JG005186 ] Large spatial and temporal variability of carbon dioxide and methane in a eutrophic lake Luo J H , Pu R L , Duan H T , Ma R H , Mao Z G , Zeng Y , Huang L S and Xiao Q T . 2020 . Evaluating the influences of harvesting activity and eutrophication on loss of aquatic vegetations in Taihu Lake, China . International Journal of Applied Earth Observation and Geoinformation , 87 : 102038 [ DOI: 10.1016/j.jag.2019.102038 http://dx.doi.org/10.1016/j.jag.2019.102038 ] Evaluating the influences of harvesting activity and eutrophication on loss of aquatic vegetations in Taihu Lake, China Marotta H , Duarte C M , Sobek S and Enrich-Prast A . 2009 . Large CO 2 disequilibria in tropical lakes . Global Biogeochemical Cycles , 23 ( 4 ): Large CO2 disequilibria in tropical lakes GB 4022 [ DOI: 10.1029/2008gb003434 http://dx.doi.org/10.1029/2008gb003434 ] Natchimuthu S , Sundgren I , Gålfalk M , Klemedtsson L and Bastviken D . 2017 . Spatiotemporal variability of lake pCO 2 and CO 2 fluxes in a hemiboreal catchment . Journal of Geophysical Research: Biogeosciences , 122 ( 1 ): 30 - 49 [ DOI: 10.1002/2016jg003449 http://dx.doi.org/10.1002/2016jg003449 ] Spatiotemporal variability of lake pCO2 and CO2 fluxes in a hemiboreal catchment Qi T C , Xiao Q T , Cao Z G , Shen M , Ma J G , Liu D and Duan H T . 2020 . Satellite estimation of dissolved carbon dioxide concentrations in China’s Lake Taihu . Environmental Science and Technology , 54 ( 21 ): 13709 - 13718 [ DOI: 10.1021/acs.est.0c04044 http://dx.doi.org/10.1021/acs.est.0c04044 ] Satellite estimation of dissolved carbon dioxide concentrations in China’s Lake Taihu Qin B Q . 2020 . Shallow lake limnology and control of eutrophication in Lake Taihu . Journal of Lake Sciences , 32 ( 5 ): 1229 - 1243 Shallow lake limnology and control of eutrophication in Lake Taihu 秦伯强 . 2020 . 浅水湖泊湖沼学与太湖富营养化控制研究 . 湖泊科学 , 32 ( 5 ): 1229- 1243 [ DOI: 10.18307/2020.0501 http://dx.doi.org/10.18307/2020.0501 ] 浅水湖泊湖沼学与太湖富营养化控制研究 Raymond P A , Hartmann J , Lauerwald R , Sobek S , McDonald C , Hoover M , Butman D , Striegl R , Mayorga E , Humborg C , Kortelainen P , Dürr H , Meybeck M , Ciais P and Guth P . 2013 . Global carbon dioxide emissions from inland waters . Nature , 503 ( 7476 ): 355 - 359 [ DOI: 10.1038/nature12760 http://dx.doi.org/10.1038/nature12760 ] Global carbon dioxide emissions from inland waters Seekell D A and Gudasz C . 2016 . Long-term pCO 2 trends in Adirondack Lakes . Geophysical Research Letters , 43 ( 10 ): 5109 - 5115 [ DOI: 10.1002/2016gl068939 http://dx.doi.org/10.1002/2016gl068939 ] Long-term pCO2 trends in Adirondack Lakes Shen M , Duan H T , Cao Z G , Xue K , Loiselle S and Yesou H . 2017 . Determination of the downwelling diffuse attenuation coefficient of lake water with the Sentinel-3A OLCI . Remote Sensing , 9 ( 12 ): 1246 [ DOI: 10.3390/rs9121246 http://dx.doi.org/10.3390/rs9121246 ] Determination of the downwelling diffuse attenuation coefficient of lake water with the Sentinel-3A OLCI Shen M , Duan H T , Cao Z G , Xue K , Qi T C , Ma J G , Liu D , Song K S , Huang C L and Song X Y . 2020 . Sentinel-3 OLCI observations of water clarity in large lakes in eastern China: implications for SDG 6.3.2 evaluation . Remote Sensing of Environment , 247 : 111950 [ DOI: 10.1016/j.rse.2020.111950 http://dx.doi.org/10.1016/j.rse.2020.111950 ] Sentinel-3 OLCI observations of water clarity in large lakes in eastern China: implications for SDG 6.3.2 evaluation Shi K , Zhang Y L , Zhou Y Q , Liu X H , Zhu G W , Qin B Q and Gao G . 2017 . Long-term MODIS observations of cyanobacterial dynamics in Lake Taihu: Responses to nutrient enrichment and meteorological factors . Scientific Reports , 7 : 40326 [ DOI: 10.1038/srep40326 http://dx.doi.org/10.1038/srep40326 ] Long-term MODIS observations of cyanobacterial dynamics in Lake Taihu: Responses to nutrient enrichment and meteorological factors Verpoorter C , Kutser T , Seekell D A and Tranvik L J . 2014 . A global inventory of lakes based on high-resolution satellite imagery . Geophysical Research Letters , 41 ( 18 ): 6396 - 6402 [ DOI: 10.1002/2014GL060641 http://dx.doi.org/10.1002/2014GL060641 ] A global inventory of lakes based on high-resolution satellite imagery Xiao Q T , Xu X F , Duan H T , Qi T C , Qin B Q , Lee X , Hu Z H , Wang W , Xiao W and Zhang M . 2020 . Eutrophic Lake Taihu as a significant CO 2 source during 2000-2015 . Water Research , 170 : 115331 [ DOI: 10.1016/j.watres.2019.115331 http://dx.doi.org/10.1016/j.watres.2019.115331 ] Eutrophic Lake Taihu as a significant CO2 source during 2000-2015 Xiao Q T , Zhang M , Hu Z H , Gao Y Q , Hu C , Liu C , Liu S D , Zhang Z , Zhao J Y , Xiao W and Lee X . 2017 . Spatial variations of methane emission in a large shallow eutrophic lake in subtropical climate . Journal of Geophysical Research: Biogeosciences , 122 ( 7 ): 1597 - 1614 [ DOI: 10.1002/2017JG003805 http://dx.doi.org/10.1002/2017JG003805 ] Spatial variations of methane emission in a large shallow eutrophic lake in subtropical climate Xiao W , Liu S D , Li H C , Xiao Q T , Wang W , Hu Z H , Hu C , Gao Y Q , Shen J , Zhao X Y , Zhang M and Lee X . 2014 . A flux-gradient system for simultaneous measurement of the CH 4 , CO 2 , and H 2 O fluxes at a lake-air interface . Environmental Science and Technology , 48 ( 24 ): 14490 - 14498 [ DOI: 10.1021/es5033713 http://dx.doi.org/10.1021/es5033713 ] A flux-gradient system for simultaneous measurement of the CH4, CO2, and H2O fluxes at a lake-air interface Xu Y J , Xu Z and Yang R J . 2019 . Rapid daily change in surface water pCO 2 and CO 2 evasion: a case study in a subtropical eutrophic lake in Southern USA . Journal of Hydrology , 570 : 486 - 494 [ DOI: 10.1016/j.jhydrol.2019.01.016 http://dx.doi.org/10.1016/j.jhydrol.2019.01.016 ] Rapid daily change in surface water pCO2 and CO2 evasion: a case study in a subtropical eutrophic lake in Southern USA Yan X C , Ma J , Li Z C , Ji M , Xu J , Xu X G , Wang G X and Li Y M . 2021 . CO 2 dynamic of Lake Donghu highlights the need for long-term monitoring . Environmental Science and Pollution Research , 28 ( 9 ): 10967 - 10976 [ DOI: 10.1007/s11356-020-11374-y http://dx.doi.org/10.1007/s11356-020-11374-y ] CO2 dynamic of Lake Donghu highlights the need for long-term monitoring Yang P and Tong C . 2015 . Emission paths and measurement methods for greenhouse gas fluxes from freshwater ecosystems: a review . Acta Ecologica Sinica , 35 ( 20 ): 6868 - 6880 Emission paths and measurement methods for greenhouse gas fluxes from freshwater ecosystems: a review 杨平 , 仝川 . 2015 . 淡水水生生态系统温室气体排放的主要途径及影响因素研究进展 . 生态学报 , 35 ( 20 ): 6868- 6880 [ DOI: 10.5846/stxb201406231298 http://dx.doi.org/10.5846/stxb201406231298 ] 淡水水生生态系统温室气体排放的主要途径及影响因素研究进展 Zhang B , Li J S , Shen Q , Wu Y H , Zhang F F , Wang S L , Yao Y , Guo L N and Yin Z Y . 2021 . Recent research progress on long time series and large scale optical remote sensing of inland water . Journal of Remote Sensing , 25 ( 1 ): 37 - 52 Recent research progress on long time series and large scale optical remote sensing of inland water 张兵 , 李俊生 , 申茜 , 吴艳红 , 张方方 , 王胜蕾 , 姚月 , 郭立男 , 殷子瑶 . 2021 . 长时序大范围内陆水体光学遥感研究进展 . 遥感学报 , 25 ( 1 ): 37- 52 [ DOI: 10.11834/jrs.20210570 http://dx.doi.org/10.11834/jrs.20210570 ] 长时序大范围内陆水体光学遥感研究进展 10.11834/jrs.20221279 1007-4619（2022）01-0231-12 齐天赐，段洪涛，曹志刚，沈明，肖启涛，刘东，马金戈.2022.2002年—2018年太湖水体溶解二氧化碳浓度卫星遥感数据集.遥感学报，26（1）： 231-242 齐天赐，段洪涛，曹志刚，沈明，肖启涛，刘东，马金戈.2022.2002年—2018年太湖水体溶解二氧化碳浓度卫星遥感数据集.遥感学报，26（1）： 231-242 Qi　T C， Duan　H T， Cao　Z G， Shen　M， Xiao　Q T， Liu　D and Ma　J G. 2022. Monthly average satellite-estimated dataset of Lake Taihu’s dissolved carbon dioxide concentration from 2002 to 2018. National Remote Sensing Bulletin， 26（1）：231-242 Qi　T C， Duan　H T， Cao　Z G， Shen　M， Xiao　Q T， Liu　D and Ma　J G. 2022. Monthly average satellite-estimated dataset of Lake Taihu’s dissolved carbon dioxide concentration from 2002 to 2018. National Remote Sensing Bulletin， 26（1）：231-242 国家自然科学基金（编号：41971309， 41801093， 41901299） Supported by National Natural Science Foundation of China (No. 41971309, 41801093, 41901299) 2021-05-18 2022-01-07 2022-05-31 2021-07-02 关联数据获取 1 https：//doi.org/10.5281/zenodo.4729048 https：//doi.org/10.5281/zenodo.4729048 遥感学报 1 26