Evapotranspiration interpolation in alpine marshes wetland on the Qingzang Plateau based on machine learning

doi:10.17521/cjpe.2022.0015

Abstract

Abstract:

Aims This study aims to explore a high-precision interpolation method of evapotranspiration based on machine learning to construct high-quality data set of actual evapotranspiration.

Methods Taking the typical alpine marsh wetland on the Qingzang Plateau as the observation station to study evapotranspiration, combined with meteorological factors (net radiation, air temperature, soil heat flux, wind speed, relative humidity, soil volumetric water content), we established a prediction model to construct an actual evapotranspiration data set with a high-precision interpolation method based on combining five methods including multiple linear regression (MLR), decision tree (CART), random forest (RF), support vector regression (SVR) and multi-layer perceptron (MLP).

Important findings 1) The correlation between evapotranspiration and net radiation was the largest in the study area, and soil heat flux was the key factor affecting the evapotranspiration process. 2) The determination coefficients are from 0.58 to 0.83 among five machine learning algorithm models with seven combinations, and the root mean square error ranges from 0.038 to 0.089 mm·30 min^-1. 3) The random forest regression model has the highest determination coefficient, the best model stability and the best interpolation. 4) Interpolated evapotranspiration data had the same diurnal variation trend with net radiation, soil heat flux and ari temperature, but the opposite diurnal variation trend with wind speed and relative humidity. Daily evapotranspiration is mainly concentrated in the growing season, with the daily maximum (8.77 mm) on July 9 and the daily minimum (0.21 mm) on January 30.

Key words: machine learning, alpine marshes, evapotranspiration, cross validation

WANG Xiu-Ying, CHEN Qi, DU Hua-Li, ZHANG Rui, MA Hong-Lu. Evapotranspiration interpolation in alpine marshes wetland on the Qingzang Plateau based on machine learning[J]. Chin J Plant Ecol, 2023, 47(7): 912-921.

Add to citation manager EndNote|Ris|BibTeX

URL: https://www.plant-ecology.com/EN/10.17521/cjpe.2022.0015

https://www.plant-ecology.com/EN/Y2023/V47/I7/912

Figures/Tables 11

References 26

[1]	Ainsworth EA, Rogers A (2007). The response of photosynthesis and stomatal conductance to rising [CO₂]: mechanisms and environmental interactions. Plant, Cell & Environment, 30, 258-270.
[2]	Chen SP, You CH, Hu ZM, Chen Z, Zhang LM, Wang QF (2020). Eddy covariance technique and its applications in flux observations of terrestrial ecosystems. Chinese Journal of Plant Ecology, 44, 291-304. DOI URL
	[陈世苹, 游翠海, 胡中民, 陈智, 张雷明, 王秋凤 (2020). 涡度相关技术及其在陆地生态系统通量研究中的应用. 植物生态学报, 44, 291-304.] DOI
[3]	de Dios VR, Roy J, Ferrio JP, Alday JG, Landais D, Milcu A, Gessler A (2015). Processes driving nocturnal transpiration and implications for estimating land evapotranspiration. Scientific Reports, 5, 10975. DOI: 10.1038/srep10975. DOI PMID
[4]	Deng XY, Liu Y, Liu ZH, Yao JQ (2017). Temporal-spatial dynamic change characteristics of evapotranspiration in arid region of northwest China. Acta Ecologica Sinica, 37, 2994-3008.
	[邓兴耀, 刘洋, 刘志辉, 姚俊强 (2017). 中国西北干旱区蒸散发时空动态特征. 生态学报, 37, 2994-3008.]
[5]	Falge E, Baldocchi D, Olson R, Anthoni P, Aubinet M, Bernhofer C, Burba G, Ceulemans R, Clement R, Dolman H, Granier A, Gross P, Grünwald T, Hollinger D, Jensen NO, et al. (2001). Gap filling strategies for long term energy flux data sets. Agricultural and Forest Meteorology, 107, 71-77. DOI URL
[6]	Li SG, Asanuma J, Kotani A, Davaa G, Oyunbaatar D (2007). Evapotranspiration from a Mongolian steppe under grazing and its environmental constraints. Journal of Hydrology, 333, 133-143. DOI URL
[7]	Li X, Liu TX, Duan LM, Tong X, Wang GL (2020). Crop coefficient simulation and evapotranspiration estimation of dune and meadow in a semiarid area. Arid Zone Research, 37, 1246-1255.
	[李霞, 刘廷玺, 段利民, 童新, 王冠丽 (2020). 半干旱区沙丘、草甸作物系数模拟及蒸散发估算. 干旱区研究, 37, 1246-1255.]
[8]	Li YM (2019). Wheat Yield Forecasting: a Machine Learning Approach Based on Meteorological Factors. Master degree dissertation, Henan Agricultural University, Zhengzhou.
	[李严明 (2019). 基于机器学习的气象因素对小麦产量影响的分析预测. 硕士学位论文, 河南农业大学, 郑州.]
[9]	Liu CF, Zhang ZQ, Sun G, Zha TG, Zhu JZ, Shen LH, Chen J, Fang XR, Chen JQ (2009). Quantifying evapotranspiration and biophysical regulations of a poplar plantation assessed by eddy covariance and sap-flow methods. Chinese Journal of Plant Ecology, 33, 706-718. DOI
	[刘晨峰, 张志强, 孙阁, 查同刚, 朱金兆, 申李华, 陈军, 方显瑞, 陈吉泉 (2009). 基于涡度相关法和树干液流法评价杨树人工林生态系统蒸发散及其环境响应. 植物生态学报, 33, 706-718.] DOI
[10]	Liu K, He QS, Jing CL, Li JY, Chen L (2020). Gap filling method for evapotranspiration based on machine learning. Journal of Hohai University (Natural Sciences), 48(2), 109-115.
	[刘堃, 何祺胜, 荆琛琳, 李金阳, 陈丽 (2020). 基于机器学习的蒸散量插补方法. 河海大学学报(自然科学版), 48(2), 109-115.]
[11]	Liu XH, Wei BQ, Wu LF, Yang P (2020). Applicability of four kinds of artificial intelligent models on prediction of reference crop evapotranspiration in Jiangxi Province. Journal of Drainage and Irrigation Machinery Engineering, 38(1), 102-108.
	[刘小华, 魏炳乾, 吴立峰, 杨坡 (2020). 4种人工智能模型在江西省参考作物蒸散量计算中的适用性. 排灌机械工程学报, 38(1), 102-108.]
[12]	Meng XN, Jiao RL, Liu N, Xia JJ, Yan ZW, Yu S, Lou X, Li HC, Wang LZ, Chen L, Zheng ZY, Zhao N (2020). Extreme summer high-temperature changes in Central Asia based on interpolated data from random forest. Arid Zone Research, 37, 966-973.
	[孟欣宁, 焦瑞莉, 刘念, 夏江江, 严中伟, 于爽, 娄晓, 李昊辰, 王立志, 陈亮, 郑子彦, 赵娜 (2020). 基于随机森林插值的中亚夏季极端高温变化特征. 干旱区研究, 37, 966-973.]
[13]	Niu ZE, Hu KM, He HL, Ren XL, Zhang L, Ge R, Li P, Zheng H, Zhu XB, Zeng N (2019). The spatial-temporal patterns of evapotranspiration and its influencing factors in Chinese terrestrial ecosystem from 2000 to 2015. Acta Ecologica Sinica, 39, 4697-4709.
	[牛忠恩, 胡克梅, 何洪林, 任小丽, 张黎, 葛蓉, 李攀, 郑涵, 朱晓波, 曾纳 (2019). 2000-2015年中国陆地生态系统蒸散时空变化及其影响因素. 生态学报, 39, 4697-4709.]
[14]	Peng HH, Zhao CY, Liang J (2016). Daily variation of evapotranspiration rate of alpine grassland and analysis of its environmental factors in upper reach of Heihe River. Journal of Water Resources & Water Engineering, 27(1), 46-53.
	[彭焕华, 赵传燕, 梁继 (2016). 黑河上游高寒草地蒸散发日变化及其影响因子分析. 水资源与水工程学报, 27(1), 46-53.]
[15]	Qi DL, Li XD, Xiao HB, Zhou WF, Su WJ, Hu AJ, Li F (2015). Study on changing characteristics and impact factor of evaporation over three-river source area in recent 50 years. Resources and Environment in the Yangtze Basin, 24, 1613-1620.
	[祁栋林, 李晓东, 肖宏斌, 周万福, 苏文将, 胡爱军, 李璠 (2015). 近50 a三江源地区蒸发量的变化特征及其影响因子分析. 长江流域资源与环境, 24, 1613-1620.]
[16]	Qiu LS, Zhang LF, He Y, Chen YD, Wang WH (2020). Spatiotemporal variations of evapotranspiration and influence factors in Qilian Mountain from 2000 to 2018. Research of Soil and Water Conservation, 27, 210-217.
	[邱丽莎, 张立峰, 何毅, 陈有东, 王文辉 (2020). 2000-2018年祁连山蒸散发时空变化及影响因素. 水土保持研究, 27, 210-217.]
[17]	Shen ZX, Fu G (2016). Relationships between water use efficiency and environmental temperature and humidity in an alpine meadow in the northern Tibet. Ecology and Environmental Sciences, 25, 1259-1263.
	[沈振西, 付刚 (2016). 藏北高原高寒草甸水分利用效率与环境温湿度的关系. 生态环境学报, 25, 1259-1263.] DOI
[18]	Tian XH, Zhang LF, Zhang X, Chen ZG, Zhao L, Li Q, Tang YH, Gu S (2020). Evapotranspiration characteristics of degraded meadow and effects of freeze-thaw changes in the Three-River Source Region. Acta Ecologica Sinica, 40, 5649-5662.
	[田晓晖, 张立锋, 张翔, 陈之光, 赵亮, 李奇, 唐艳鸿, 古松 (2020). 三江源区退化高寒草甸蒸散特征及冻融变化对其的影响. 生态学报, 40, 5649-5662.]
[19]	Wang FY, Zhan CS, Hu S, Jia YW, Niu CW, Zou J (2017). Simulation of spatio-temporal changes in evapotranspiration in typical mountains. Resources Science, 39, 276-287. DOI
	[王飞宇, 占车生, 胡实, 贾仰文, 牛存稳, 邹靖 (2017). 典型山地蒸散发时空变化模拟研究. 资源科学, 39, 276-287.] DOI
[20]	Wang S, Fu ZY, Chen HS, Ding YL, Wu LP, Wang KL (2017). Simulation of reference evapotranspiration based on random forest method. Transactions of the Chinese Society for Agricultural Machinery, 48, 302-309.
	[王升, 付智勇, 陈洪松, 丁亚丽, 吴丽萍, 王克林 (2017). 基于随机森林算法的参考作物蒸发蒸腾量模拟计算. 农业机械学报, 48, 302-309.]
[21]	Wen XF, Yu GR, Sun XM (2004). Uncertainties in long-term studies of net ecosystem CO₂ exchange with the atmosphere based on eddy covariance technique. Advances in Earth Science, 19, 658-663.
	[温学发, 于贵瑞, 孙晓敏 (2004). 基于涡度相关技术估算植被/大气间净CO₂交换量中的不确定性. 地球科学进展, 19, 658-663.] DOI
[22]	Wever LA, Flanagan LB, Carlson PJ (2002). Seasonal and interannual variation in evapotranspiration, energy balance and surface conductance in a northern temperate grassland. Agricultural and Forest Meteorology, 112, 31-49. DOI URL
[23]	Wu FT, Cao SK, Cao GC, Han GZ, Lin YY, Cheng SY (2018). Water use efficiency of alpine wetland ecosystem. Arid Zone Research, 35, 306-314.
	[吴方涛, 曹生奎, 曹广超, 汉光昭, 林阳阳, 成淑艳 (2018). 高寒湿地生态系统水分利用效率研究. 干旱区研究, 35, 306-314.]
[24]	Yang BP, Chen SB, Yu HY, An Q (2020). Remote sensing estimation of rice yield based on random forest regression method. Journal of China Agricultural University, 25(6), 26-34.
	[杨北萍, 陈圣波, 于海洋, 安秦 (2020). 基于随机森林回归方法的水稻产量遥感估算. 中国农业大学学报, 25(6), 26-34.]
[25]	Zhang L, Wang LL, Zhang XD, Liu SR, Sun PS, Wang TL (2014). The basic principle of random forest and its applications in ecology: a case study of Pinus yunnanensis. Acta Ecologica Sinica, 34, 650-659.
	[张雷, 王琳琳, 张旭东, 刘世荣, 孙鹏森, 王同立 (2014). 随机森林算法基本思想及其在生态学中的应用——以云南松分布模拟为例. 生态学报, 34, 650-659.]
[26]	Zhang MM (2019). Analysis of the Temporal and Spatial Variation of Evapotranspiration and Its Driving Factors in Arid and Semi-arid Region of China from 2000 to 2015. Master degree dissertation, Chang’an University, Xiʼan.
	[张明明 (2019). 2000-2015年中国干旱半干区蒸散发时空变化及其影响因素分析. 硕士学位论文, 长安大学, 西安.]

	月份 Month
	1	2	3	4	5	6	7	8	9	10
有效记录 Active record	1 206	1 206	1 382	1 309	1 110	1 057	1 152	1 282	1 075	936
缺失数据 Missing data	282	138	106	131	378	383	336	206	365	240
缺失率 Absence rate (%)	19	10	7	9	25	27	23	14	25	20

	月份 Month
	1	2	3	4	5	6	7	8	9	10
有效记录 Active record	1 206	1 206	1 382	1 309	1 110	1 057	1 152	1 282	1 075	936
缺失数据 Missing data	282	138	106	131	378	383	336	206	365	240
缺失率 Absence rate (%)	19	10	7	9	25	27	23	14	25	20

气象因子 Meteorological factor	月份 Month										平均 Mean
气象因子 Meteorological factor	1	2	3	4	5	6	7	8	9	10	平均 Mean
净辐射 Net radiation	0.80	0.83	0.69	0.72	0.77	0.76	0.79	0.83	0.81	0.69	0.769
土壤热通量 Soil heat flux	0.74	0.43	0.54	0.86	0.84	0.87	0.84	0.84	0.84	0.85	0.765
相对湿度 Relative humidity	-0.62	-0.68	-0.69	-0.74	-0.71	-0.75	-0.78	-0.80	-0.78	-0.72	-0.727
气温 Air temperature	0.63	0.64	0.56	0.75	0.73	0.69	0.77	0.78	0.66	0.69	0.690
风速 Wind velocity	0.74	0.74	0.71	0.61	0.59	0.43	0.39	0.29	0.46	0.63	0.559
土壤温度 Soil temperature (5 cm)	0.29	0.07	0.35	0.60	0.52	0.53	0.55	0.46	0.41	0.48	0.426
土壤温度 Soil temperature (10 cm)	-0.12	0.12	0.13	0.04	-0.07	-0.01	0	0.21	0.13	0.15	0.058
土壤温度 Soil temperature (20 cm)	-0.37	0	0.11	-0.21	-0.22	-0.13	-0.34	-0.20	0	0.03	-0.133
土壤温度 Soil temperature (30 cm)	-0.32	0.10	0.12	0.13	-0.08	-0.03	-0.10	-0.04	0.09	0.12	-0.001
土壤温度 Soil temperature (40 cm)	0	-0.06	0.14	0.10	0.08	0.10	0.15	0.08	0.12	0.18	0.089
土壤含水率 Soil moisture content (5 cm)	0.26	0.08	0.11	-0.08	0.30	0.22	0.04	0.36	0.40	0.33	0.202
土壤含水率 Soil moisture content (10 cm)	-0.24	0	0.04	0.07	0.04	-0.08	-0.23	-0.03	-0.10	0.11	-0.042
土壤含水率 Soil moisture content (20 cm)	-0.24	0.10	0.07	0.12	0.05	-0.01	-0.14	-0.08	-0.11	0.10	-0.014
土壤含水率 Soil moisture content (30 cm)	-0.02	-0.10	0.09	0.15	-0.02	0.03	-0.15	-0.02	-0.09	0.11	-0.002
土壤含水率 Soil moisture content (40 cm)	-0.03	0.04	0.10	0.15	-0.02	0	0	0	-0.05	0.15	0.034

气象因子 Meteorological factor	月份 Month										平均 Mean
气象因子 Meteorological factor	1	2	3	4	5	6	7	8	9	10	平均 Mean
净辐射 Net radiation	0.80	0.83	0.69	0.72	0.77	0.76	0.79	0.83	0.81	0.69	0.769
土壤热通量 Soil heat flux	0.74	0.43	0.54	0.86	0.84	0.87	0.84	0.84	0.84	0.85	0.765
相对湿度 Relative humidity	-0.62	-0.68	-0.69	-0.74	-0.71	-0.75	-0.78	-0.80	-0.78	-0.72	-0.727
气温 Air temperature	0.63	0.64	0.56	0.75	0.73	0.69	0.77	0.78	0.66	0.69	0.690
风速 Wind velocity	0.74	0.74	0.71	0.61	0.59	0.43	0.39	0.29	0.46	0.63	0.559
土壤温度 Soil temperature (5 cm)	0.29	0.07	0.35	0.60	0.52	0.53	0.55	0.46	0.41	0.48	0.426
土壤温度 Soil temperature (10 cm)	-0.12	0.12	0.13	0.04	-0.07	-0.01	0	0.21	0.13	0.15	0.058
土壤温度 Soil temperature (20 cm)	-0.37	0	0.11	-0.21	-0.22	-0.13	-0.34	-0.20	0	0.03	-0.133
土壤温度 Soil temperature (30 cm)	-0.32	0.10	0.12	0.13	-0.08	-0.03	-0.10	-0.04	0.09	0.12	-0.001
土壤温度 Soil temperature (40 cm)	0	-0.06	0.14	0.10	0.08	0.10	0.15	0.08	0.12	0.18	0.089
土壤含水率 Soil moisture content (5 cm)	0.26	0.08	0.11	-0.08	0.30	0.22	0.04	0.36	0.40	0.33	0.202
土壤含水率 Soil moisture content (10 cm)	-0.24	0	0.04	0.07	0.04	-0.08	-0.23	-0.03	-0.10	0.11	-0.042
土壤含水率 Soil moisture content (20 cm)	-0.24	0.10	0.07	0.12	0.05	-0.01	-0.14	-0.08	-0.11	0.10	-0.014
土壤含水率 Soil moisture content (30 cm)	-0.02	-0.10	0.09	0.15	-0.02	0.03	-0.15	-0.02	-0.09	0.11	-0.002
土壤含水率 Soil moisture content (40 cm)	-0.03	0.04	0.10	0.15	-0.02	0	0	0	-0.05	0.15	0.034

模型特征组合 Model feature combination	输入特征 Input characteristic	特征个数 Feature number
组合1 Combination 1	Rn、Sg、RH、T、WS	5
组合2 Combination 2	Rn、Sg、RH、T、WS、ST_5 cm	6
组合3 Combination 3	Rn、Sg、RH、T、WS、SV_5 cm	6
组合4 Combination 4	Rn、Sg、RH、T、WS、ST_5 cm、SV_5 cm	7
组合5 Combination 5	Rn、Sg、RH、T、WS、ST_5 cm、ST_10 cm、ST_20 cm、ST_30 cm、ST_40 cm	10
组合6 Combination 6	Rn、Sg、RH、T、WS、SV_5 cm、SV_10 cm、SV_20 cm、SV_30 cm、SV_40 cm	10
组合7 Combination 7	Rn、Sg、RH、T、WS、ST_5 cm-ST_40 cm、SV_5 cm-SV_40 cm	15