涡度相关观测的太湖CH4通量数据插补方法评价

doi:10.13287/j.1001-9332.202210.021

摘要/Abstract

摘要： 涡度相关法是在湖泊开展CH₄通量长期连续观测的重要方法。受到多种因素的影响,CH₄通量观测数据存在大量缺失。为重构完整的CH₄通量时间序列,就需要适宜的数据插补方法。本研究利用太湖涡度通量观测网络东部的避风港站点2014—2017年的常规气象数据及涡度相关观测的CH₄通量数据,在分析半小时尺度以及日尺度CH₄通量影响要素的基础上,测试了非线性回归法以及随机森林算法和误差反向传播算法在半小时尺度及日尺度上插补CH₄通量缺失数据的可行性。结果表明: 在半小时尺度上,避风港站生长季CH₄通量主要受到底泥温度、摩擦风速、气温、相对湿度、潜热通量和20 cm处水温的影响,非生长季主要受到相对湿度、潜热通量、风速、感热通量和底泥温度的影响,而在日尺度上CH₄通量主要受潜热通量和相对湿度的影响。在对CH₄通量缺失数据的插补中,随机森林模型在所有时间尺度上都表现为最佳的插补性能,其中,将日序、太阳高度角、底泥温度、摩擦风速、气温、20 cm处水温、相对湿度、气压和风速作为输入变量的随机森林模型更适用于半小时尺度缺失数据的插补;将日序、底泥温度、摩擦风速、气温、20 cm处水温、相对湿度、气压、风速和向下短波辐射作为输入变量的随机森林模型更适用于日尺度缺失数据的插补;整体上,插补模型对日尺度缺失数据的插补优于半小时尺度。

Abstract: Eddy covariance method has become a key technique to measure CH₄ flux continuously in lakes. A large number of CH₄ flux data was missing due to variable reasons. In order to reconstruct a complete time series of CH₄ flux, it is necessary to find an appropriate gap-filling method to insert the CH₄ flux data gap. Based on the routine meteorological data and CH₄ flux data measured at Bifenggang site in the eastern part of the Taihu eddy flux network during 2014 to 2017, we analyzed the control factors of CH₄ flux at the half-hour scale and daily scale. With those data, we tested that whether nonlinear regression method and two machine learning methods, random forest algorithm and error back propagation algorithm, could fill the CH₄ flux gap at the half-hour scale and daily scale. The results showed that CH₄ flux at the half-hour scale was mainly influenced by sediment temperature, friction velocity, air temperature, relative humidity, latent heat flux and water temperature at 20 cm in the growing season, and was mainly affected by relative humidity, latent heat flux, wind speed, sensible heat flux and sediment temperature in non-growing season. The CH₄ flux at the daily scale was mainly affected by latent heat flux and relative humidity. Random forest model was the best in CH₄ flux data gap filling at both time scales. The random forest model with the input variables of day of year, solar elevation angle, sediment temperature, friction velocity, air temperature, water temperature at 20 cm, relative humidity, air pressure, and wind speed was more suitable for filling the CH₄ flux data gap at the half-hour scale. The random forest model with the input variables of day of year, sediment temperature, friction velocity, air temperature, water temperature at 20 cm, relative humidity, air pressure, wind speed, and downward shortwave radiation was more suitable for filling CH₄ flux data gap at the day scale. The interpolation models could fill the data gap better at daily scale than that at the half-hour scale.

Key words: eddy covariance method, CH₄ flux, gap filling, random forest, back propagation neural network

邱吉丽, 张弥, 蒲旖旎, 张圳, 贾磊, 赵佳玉, 肖薇, 刘寿东. 涡度相关观测的太湖CH₄通量数据插补方法评价[J]. 应用生态学报, 2022, 33(10): 2785-2795.

QIU Ji-li, ZHANG Mi, PU Yi-ni, ZHANG Zhen, JIA Lei, ZHAO Jia-yu, XIAO Wei, LIU Shou-dong. Evaluation of gap-filling methods for CH₄ flux data based on eddy covariance method in the Lake Taihu, China[J]. Chinese Journal of Applied Ecology, 2022, 33(10): 2785-2795.

参考文献

[1] Saunois M, Stavert AR, Poulter B, et al. The Global Methane Budget 2000-2017. Earth System Science Data, 2020, 12: 1561-1623
[2] Bastviken D, Cole J, Pace M, et al. Methane emissions from lakes: Dependence of lake characteristics, two regional assessments, and a global estimate. Global Biogeo-chemical Cycles, 2004, 18: GB4009
[3] Natchimuthu S, Sundgren I, Gålfalk M, et al. Spatio-temporal variability of lake CH₄ fluxes and its influence on annual whole lake emission estimates. Limnology and Oceanography, 2016, 61: s13-s26
[4] 贾磊, 蒲旖旎, 杨诗俊, 等. 太湖藻型湖区CH₄、CO₂排放特征及其影响因素分析. 环境科学, 2018, 39(5): 2316-2329
[5] Davidson TA, Audet J, Jeppesen E, et al. Synergy between nutrients and warming enhances methane ebullition from experimental lakes. Nature Climate Change, 2018, 8: 156-160
[6] Sepulveda-Jauregui A, Anthony K, Martinez-Cruz K, et al. Methane and carbon dioxide emissions from 40 lakes along a north-south latitudinal transect in Alaska. Biogeosciences, 2015, 12: 3197-3223
[7] Xiao QT, Zhang M, Hu ZH, et al. Spatial variations of methane emission in a large shallow eutrophic lake in subtropical climate. Journal of Geophysical Research: Biogeosciences, 2017, 122: 1597-1614
[8] Xiao W, Liu SD, Li H, et al. A flux-gradient system for simultaneous measurement of the CH₄, CO₂ and H₂O fluxes at a lake-air interface. Environmental Science and Technology, 2014, 48: 14490-14498
[9] Zhao JY, Zhang M, Xiao W, et al. An evaluation of the flux-gradient and the eddy covariance method to mea-sure CH₄, CO₂, and H₂O fluxes from small ponds. Agricultural and Forest Meteorology, 2019, 275: 255-264
[10] Baldocchi DD, Hicks BB, Meyers TP. Measuring biosphere-atmosphere exchanges of biologically related gases with micrometeorological methods. Ecology, 1988, 69: 1331-1340
[11] Lee XH, Liu SD, Xiao W, et al. 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, 2014, 95: 1583-1594
[12] Erkkilä KM, Ojala A, Bastviken D, et al. Methane and carbon dioxide fluxes over a lake: Comparison between eddy covariance, floating chambers and boundary layer method. Biogeosciences, 2018, 15: 429-445
[13] Zhang M, Xiao QT, Zhang Z, et al. Methane flux dynamics in a submerged aquatic vegetation zone in a subtropical lake. Science of the Total Environment, 2019, 672: 400-409
[14] Iwata H, Nakazawa K, Sato H, et al. Temporal and spatial variations in methane emissions from the littoral zone of a shallow mid-latitude lake with steady methane bubble emission areas. Agricultural and Forest Meteorology, 2020, 295: 108184
[15] Moffat A, Papale D, Reichstein M, et al. Comprehensive comparison of gap-filling techniques for eddy cova-riance net carbon fluxes. Agricultural and Forest Meteoro-logy, 2007, 147: 209-232
[16] Dengel S, Zona D, Sachs T, et al. Testing the applicability of neural networks as a gap-filling method using CH₄ flux data from high latitude wetlands. Biogeoscien-ces, 2013, 10: 8185-8200
[17] Kim Y, Johnson MS, Knox SH, et al. Gap-filling approaches for eddy covariance methane fluxes: A comparison of three machine learning algorithms and a traditional method with principal component analysis. Global Change Biology, 2020, 26: 1499-1518
[18] 于贵瑞, 温学发, 李庆康, 等. 中国亚热带和温带典型森林生态系统呼吸的季节模式及环境响应特征. 中国科学D辑: 地球科学, 2004, 34(增刊2): 84-94
[19] Oikawa PY, Jenerette GD, Knox SH, et al. Evaluation of a hierarchy of models reveals importance of substrate limitation for predicting carbon dioxide and methane exchange in restored wetlands. Journal of Geophysical Research: Biogeosciences, 2017, 122: 145-167
[20] Papale D, Reichstein M, Aubinet M, et al. Towards a standardized processing of net ecosystem exchange measured with eddy covariance technique: Algorithms and uncertainty estimation. Biogeosciences, 2006, 3: 571-583
[21] Morin TH, Bohrer G, Frasson RPDM, et al. Environmental drivers of methane fluxes from an urban tempe-rate wetland park. Journal of Geophysical Research: Biogeosciences, 2014, 119: 2188-2208
[22] 蒲旖旎, 贾磊, 杨诗俊, 等. 太湖藻型湖区CH₄冒泡通量. 中国环境科学, 2018, 38(10): 3914-3924
[23] Aben RCH, Barros N, Donk E, et al. Cross continental increase in methane ebullition under climate change. Nature Communications, 2017, 8: 1682
[24] 秦伯强. 浅水湖泊湖沼学与太湖富营养化控制研究. 湖泊科学, 2020, 32(5): 1229-1243
[25] 肖启涛. 太湖CH₄通量的空间格局及影响因子分析. 博士论文, 南京: 南京信息工程大学, 2017
[26] Xiao W, Liu SD, Wang W, et al. Transfer coefficients of momentum, heat and water vapour in the atmospheric surface layer of a large freshwater lake. Boundary-Layer Meteorology, 2013, 148: 479-494
[27] Zhang Z, Zhang M, Cao C, et al. A dataset of microclimate and radiation and energy fluxes from the Lake Taihu eddy flux network. Earth System Science Data, 2020, 12: 2635-2645
[28] Baldocchi D, Finnigan J, Wilson K, et al. On measu-ring net ecosystem carbon exchange over tall vegetation on complex terrain. Boundary-Layer Meteorology, 2000, 96: 257-291
[29] Webb EK, Pearman GI, Leuning RG. Correction of flux measurements for density effects due to heat and water-vapor transfer. Quarterly Journal of the Royal Meteorological Society, 1980, 106: 100
[30] Moncrieff JB, Massheder JM, Bruin HD, et al. A system to measure surface fluxes of momentum, sensible heat, water vapour and carbon dioxide. Journal of Hydrology, 1997, 88: 589-611
[31] Mauder M, Foken T. Documentation and Instruction Manual of the Eddy Covariance Software Package TK2. Bavaria: Department of Micrometeorology, University of Bayreuth, 2004
[32] Falge E, Baldocchi D, Olson R, et al. Gap filling stra-tegies for defensible annual sums of net ecosystem exchange. Agricultural and Forest Meteorology, 2001, 107: 43-69
[33] Knox SH, Matthes JH, Sturtevant C, et al. Biophysical controls on interannual variability in ecosystem-scale CO₂ and CH₄ exchange in a California rice paddy. Journal of Geophysical Research: Biogeosciences, 2016, 121: 978-1001
[34] Deventer MJ, Griffis TJ, Roman DT, et al. Error cha-racterization of methane fluxes and budgets derived from a long-term comparison of open- and closed-path eddy covariance systems. Agricultural and Forest Meteorology, 2019, 278: 107638
[35] Breiman L. Random forests. Machine Learning, 2001, 45: 5-32
[36] Tyralis H, Papacharalampous G, Langousis A. A brief review of random forests for water scientists and practitioners and their recent history in water resources. Water, 2019, 11: 910
[37] Sturtevant C, Ruddell BL, Knox SH, et al. Identifying scale-emergent, nonlinear, asynchronous processes of wetland methane exchange. Journal of Geophysical Research: Biogeosciences, 2016, 121: 188-204
[38] Ziegler A, König IR. Mining data with random forests: Current options for real-world applications. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 2014, 4: 55-63
[39] Díaz-Uriarte R, Andrés SAD. Gene selection and classification of microarray data using random forest. BMC Bioinformatics, 2006, 7: 3
[40] Mahabbati A, Beringer J, Leopold M, et al. A comparison of gap-filling algorithms for eddy covariance fluxes and their drivers. Geoscientific Instrumentation Methods and Data Systems, 2021, 10: 123-140

涡度相关观测的太湖CH₄通量数据插补方法评价

Evaluation of gap-filling methods for CH₄ flux data based on eddy covariance method in the Lake Taihu, China

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	赵允格, 吉静怡, 张万涛, 明姣, 黄琬雲, 高丽倩. 黄土高原生物土壤结皮分布时空分异特征 [J]. 应用生态学报, 2024, 35(3): 739-748.
[2]	袁佳玉, 熊立, 吴志伟, 朱诗豪, 康平, 李顺. 江西省赣州市南康区松材线虫病发生特征 [J]. 应用生态学报, 2024, 35(2): 507-515.
[3]	任孟林, 郭妍, 陈伯轩, 范佳乐, 胡同欣, 孙龙. 红松人工林地表可燃物火蔓延速率预测模型 [J]. 应用生态学报, 2023, 34(8): 2091-2100.
[4]	赵月琴, 马秀静, 赵琬婧, 张治军, 孙晓新. 三江平原垦殖湿地恢复对温室气体排放的影响 [J]. 应用生态学报, 2023, 34(8): 2142-2152.
[5]	杨玉丽, 肖辉杰, 辛智鸣, 范光鹏, 李俊然, 贾肖肖, 汪立韬. 基于激光雷达与高光谱的荒漠绿洲农田防护林衰退程度评估 [J]. 应用生态学报, 2023, 34(4): 1043-1050.
[6]	王怡婧, 丁启东, 张俊华, 陈睿华, 贾科利, 李小林. 基于无人机高光谱遥感和机器学习的土壤水盐信息反演 [J]. 应用生态学报, 2023, 34(11): 3045-3052.
[7]	董灵波, 梁凯富, 张一帆, 刘兆刚. 基于Landsat 8时间序列数据的翠岗林场森林类型划分 [J]. 应用生态学报, 2022, 33(9): 2339-2346.
[8]	贾磊, 张弥, 蒲旖旎, 赵佳玉, 谢燕红, 肖薇, 刘寿东, 石婕. 箱体特征对箱式法观测水-气界面CO₂和CH₄通量的影响 [J]. 应用生态学报, 2022, 33(6): 1563-1571.
[9]	崔晏华, 刘淑德, 张云雷, 徐宾铎, 纪毓鹏, 张崇良, 任一平, 薛莹. 海州湾春季短蛸的栖息分布特征及其与环境因子的关系 [J]. 应用生态学报, 2022, 33(6): 1686-1692.
[10]	魏宇宸, 赵美芳, 朱昌达, 张秀秀, 潘剑君. 基于景观及微地形特征的丘陵区土壤属性预测 [J]. 应用生态学报, 2022, 33(2): 467-476.
[11]	王深华, 江军, 刘丰彩, 俞梦笑, 陈洋, 许萍萍, 常中兵, 王应平. 中国成熟天然林土壤有机碳垂直分异特征 [J]. 应用生态学报, 2021, 32(7): 2371-2377.
[12]	秦雯怡, 陈果, 李小臻, 王翔, 王鹏. 基于机器语言的岷江上游流域表层土壤氢氧稳定同位素空间分布模拟 [J]. 应用生态学报, 2021, 32(12): 4327-4338.
[13]	王世航, 卢宏亮, 赵明松, 周玲美. 基于不同特征挖掘方法结合广义提升回归模型估测安徽省土壤pH [J]. 应用生态学报, 2020, 31(10): 3509-3517.
[14]	李哲, 张沁雨, 邱新彩, 彭道黎. 基于高分二号遥感影像树种分类的时相及方法选择 [J]. 应用生态学报, 2019, 30(12): 4059-4070.
[15]	朱迪恩, 徐小军, 杜华强, 周国模, 毛方杰, 李雪建, 李阳光. 基于MODIS时间序列反射率数据的雷竹林LAI反演 [J]. 应用生态学报, 2018, 29(7): 2391-2400.