|本期目录/Table of Contents|

[1]冯 瑞,杨丽萍*,侯成磊,等.基于随机森林的陕西省西安市近地表气温估算[J].地球科学与环境学报,2022,44(01):102-113.[doi:10.19814/j.jese.2021.08027]
　FENG Rui,YANG Li-ping*,HOU Cheng-lei,et al.Estimation of Near-surface Air Temperature in Xi'an City of Shaanxi Province, China Based on Random Forest[J].Journal of Earth Sciences and Environment,2022,44(01):102-113.[doi:10.19814/j.jese.2021.08027]
点击复制

基于随机森林的陕西省西安市近地表气温估算(PDF)
分享到：

《地球科学与环境学报》[ISSN:1672-6561/CN:61-1423/P]

卷:
第44卷
期数:
2022年第01期
页码:
102-113
栏目:
环境与可持续发展
出版日期:
2022-01-15

文章信息/Info

Title:
Estimation of Near-surface Air Temperature in Xi'an City of Shaanxi Province, China Based on Random Forest
文章编号:
1672-6561(2022)01-0102-12
作者:
冯 瑞1; 杨丽萍2*; 侯成磊3; 王 彤2; 张 静2; 肖 舜4
(1. 长安大学 地球科学与资源学院,陕西 西安 710054; 2. 长安大学 地质工程与测绘学院,陕西 西安 710054; 3. 山东农业工程学院 国土资源与测绘工程学院,山东 济南 250100; 4. 陕西师范大学 地理科学与旅游学院,陕西 西安 710119)
Author(s):
FENG Rui1;  YANG Li-ping2*;  HOU Cheng-lei3;  WANG Tong2;  ZHANG Jing2;  XIAO Shun4
(1. School of Earth Science and Resources, Chang'an University, Xi'an 710054, Shaanxi, China; 2. School of Geological Engineering and Geomatics, Chang'an University, Xi'an 710054, Shaanxi, China; 3. College of Land Resources and Surveying & Mapping Engineering, Shandong Agriculture and Engineering University, Jinan 250100, Shandong, China; 4. School of Geography and Tourism, Shaanxi Normal University, Xi'an 710119, Shaanxi, China)
关键词:
近地表气温;  随机森林;  估算;  地表温度;  热岛效应;  高程;  Landsat 8;  西安
Keywords:
near-surface air temperature;  random forest;  estimation;  land surface temperature;  heat island effect;  altitude;  Landsat 8;  Xi'an
分类号:
P407; TP79
DOI:
10.19814/j.jese.2021.08027
文献标志码:
A
摘要:
随着城市规模不断扩大以及人口激增,城市气候与热环境问题日益凸显,开展城市近地表气温遥感监测研究能够为改善城市气候、减缓热岛效应、打造适宜人居环境提供参考。针对传统气温监测方法在多因素复杂关系模拟中的局限性,以陕西省西安市为研究对象,运用可以集成多要素、学习复杂、非线性映射关系的随机森林(Random Forest,RF)模型,基于Landsat 8卫星遥感数据以及SRTM高程数据相关参数的综合分析,构建多种近地表气温估算的随机森林模型,通过性能对比评估优选最佳模型,估算了2016年5月16日西安市近地表气温,分析了近地表气温的空间分布特征。结果表明:在所有近地表气温影响因子中,高程对随机森林模型近地表气温估算的贡献度最大,其次是地表温度。所有随机森林模型训练集的判定系数(R2)均高于0.916,均方根误差(RMSE)均低于0.467 ℃,验证集判定系数均高于0.726,均方根误差均低于0.840 ℃; 训练集判定系数均高于验证集,均方根误差均低于验证集; 最优随机森林模型训练集判定系数为0.934,均方根误差为0.425 ℃,验证集判定系数为0.795,均方根误差为0.783 ℃; 气温估算精度判定系数为0.792,均方根误差为1.055 ℃。西安市中心城区气温高于郊县区,中心城区最低气温平均值、最高气温平均值及气温平均值分别高于郊县区1.54 ℃、0.01 ℃和1.76 ℃。综上所述,西安市近地表气温南低北高,空间差异明显,自中心城区、郊县区至南部山区逐渐降低,呈现出显著的城市热岛效应。
Abstract:
With the continuously expanding of the urban areas and the dramatic population growth, serious urban climate and thermal environment issues have occurred and attracted extensive attention. High-precision and large-scale remote sensing monitoring of near-surface air temperature in Xi'an city can provide reference for the improvement of urban climate, the alleviation of heat island effect, as well as the creation of a suitable living environment. Traditional near-surface air temperature monitoring approaches are confined when dealing with complicated relationships among multiple factors. Taking Xi'an city as the study area, the random forest(RF)models, which can integrate multiple factors and simulate complicated and non-linear mapping relationship, were adopted. Through comprehensive analysis of various parameters derived from Landsat 8 imagery and SRTM DEM data, several near-surface air temperature estimation RF models were constructed, and the optimal model was confirmed by performance comparison and evaluation. Furthermore, near-surface air temperature in Xi'an city on May 16, 2016 was estimated, and the spatial distribution pattern was discussed. The results show that among all the near-surface air temperature influencing factors, altitude has the highest contribution to the estimated temperature of the RF models, followed by land surface temperature(LST); both of them are crucial for the estimation of near-surface air temperature. For all the RF models, the training set R2 is higher than 0.916, and the RMSE is lower than 0.467 ℃; all the validation set R2 is higher than 0.726, and the RMSE is lower than 0.840 ℃; the training set R2 of all RF models is higher than that of the validation set, while the training set RMSE is lower compared with the validation set; the training set R2 of the optimal model is 0.934 with the RMSE being 0.425 ℃, and the validation set R2 is 0.795 with a RMSE value of 0.783 ℃; the R2 and RMSE of the air temperature estimation accuracy are 0.792 and 1.055 ℃, respectively. Near-surface air temperature in the central urban districts of Xi'an city is higher than that of suburban areas, and the mean minimum temperature, the mean maximum temperature, as well as the mean temperature are higher than those of suburban areas by 1.54 ℃, 0.01 ℃ and 1.76 ℃, respectively. The near-surface air temperature in the study area is low in the south and high in the north, and presents obvious spatial heterogeneity. Air temperature gradually decreases from the central urban districts through suburbs to the southern mountainous areas, indicating a prominent urban heat island effect.

参考文献/References:

[1] PRIHODKO L,GOWARD S N.Estimation of Air Temperature from Remotely Sensed Surface Observations[J].Remote Sensing of Environment,1997,60(3):335-346.
[2] 齐述华,王军邦,张庆员,等.利用MODIS遥感影像获取近地层气温的方法研究[J].遥感学报,2005,9(5):570-575.
QI Shu-hua,WANG Jun-bang,ZHANG Qing-yuan,et al.Study on the Estimation of Air Temperature from MODIS Data[J].Journal of Remote Sensing,2005,9(5):570-575.
[3] 祝善友,张桂欣,尹 球,等.基于多源极轨气象卫星热红外数据的近地表气温反演研究[J].遥感技术与应用,2009,24(1):27-31.
ZHU Shan-you,ZHANG Gui-xin,YIN Qiu,et al.The Study on the Retrieval of the Air Temperature Based on Multi-sources Polar Orbit Meteorological Satellite Data[J].Remote Sensing Technology and Application,2009,24(1):27-31.
[4] 祝善友,张桂欣.近地表气温遥感反演研究进展[J].地球科学进展,2011,26(7):724-730.
ZHU Shan-you,ZHANG Gui-xin.Progress in Near Surface Air Temperature Retrieved by Remote Sensing Technology[J].Advances in Earth Science,2011,26(7):724-730.
[5] CHANG Y,XIAO J F,LI X X,et al.Exploring Diurnal Thermal Variations in Urban Local Climate Zones with ECOSTRESS Land Surface Temperature Data[J].Remote Sensing of Environment,2021,263:112544.
[6] 冷 佩,廖前瑜,任 超,等.近地表气温遥感反演方法综述[J].中国农业信息,2019,31(1):1-10.
LENG Pei,LIAO Qian-yu,REN Chao,et al.A Review of Methods for Estimating Near-surface Air Temperature from Remote Sensing Data[J].China Agricultural Informatics,2019,31(1):1-10.
[7] 孙从建,李 伟,陈 伟,等.昆仑山提孜那甫河流域2012～2016年近地表气温时空分布特征[J].干旱区地理,2019,42(3):459-468.
SUN Cong-jian,LI Wei,CHEN Wei,et al.Spatiotemporal Distribution of Near-surface Temperature over the Tizinafu River Basin in the Kunlun Mountains from 2012 to 2016[J].Arid Land Geography,2019,42(3):459-468.
[8] 游 婷,张 华,王海波,等.夏季白天中国中东部不同类型云分布特征及其对近地表气温的影响[J].大气科学,2020,44(4):835-850.
YOU Ting,ZHANG Hua,WANG Hai-bo,et al.Distribution of Different Cloud Types and Their Effects on Near-surface Air Temperature During Summer Daytime in Central Eastern China[J].Chinese Journ al of Atmospheric Sciences,2020,44(4):835-850.
[9] 李 军,游松财,黄敬峰.中国1961～2000年月平均气温空间插值方法与空间分布[J].生态环境,2006,15(1):109-114.
LI Jun,YOU Song-cai,HUANG Jing-feng.Spatial Interpolation Method and Spatial Distribution Characteristics of Monthly Mean Temperature in China During 1961-2000[J].Ecology and Environment,2006,15(1):109-114.
[10] ALIDOOST F,STEIN A,SU Z B.Copula-based Interpolation Methods for Air Temperature Data Using Collocated Covariates[J].Spatial Statistics,2018,28:128-140.
[11] FASS A,SOMMER M,ROHDEN C V.Interpolation Uncertainty of Atmospheric Temperature Profi-les[J].Atmospheric Measurement Techniques,2020,13(12):6445-6458.
[12] 韩秀珍,李三妹,窦芳丽.气象卫星遥感地表温度推算近地表气温方法研究[J].气象学报,2012,70(5):1107-1118.
HAN Xiu-zhen,LI San-mei,DOU Fang-li.Study of Obtaining High Resolution Near-surface Atmosphere Temperature by Using the Land Surface Temperature from Meteorological Satellite Data[J].Acta Meteorologica Sinica,2012,70(5):1107-1118.
[13] 周 佳,赵亚鹏,岳天祥,等.结合HASM和GWR方法的省级尺度近地表气温估算[J].地球信息科学学报,2020,22(10):2098-2107.
ZHOU Jia,ZHAO Ya-peng,YUE Tian-xiang,et al.Near Surface Air Temperature Estimation by Combining HASM with GWR Model on a Provincial Scale[J].Journal of Geo-information Science,2020,22(10):2098-2107.
[14] 徐伟燕,孙 睿,金志凤,等.基于MODIS数据的近地表气温估算[J].气象与环境科学,2015,38(1):1-6.
XU Wei-yan,SUN Rui,JIN Zhi-feng,et al.Estimation of Near-surface Air Temperature Based on MODIS Data[J].Meteorological and Environmental Sciences,2015,38(1):1-6.
[15] 陈命男.上海城市地温的遥感反演及气温拟合研究[D].上海:复旦大学,2012.
CHEN Ming-nan.Research on Interpreting Land Sur-face Temperature and Fitting Air Temperature from Remote Sensing in Shanghai[D].Shanghai:Fudan University,2012.
[16] LIN X H,ZHANG W,HUANG Y,et al.Empirical Estimation of Near-surface Air Temperature in China from MODIS LST Data by Considering Physiographic Features[J].Remote Sensing,2016,8(8):629-643.
[17] MOHAMMADI C,FARAJZADEH M,RAHIMI Y G,et al.Comparison of Univariate and Multivariate Geographically Weighted Regression for Estimating Air Temperature over Iran[J].Arabian Journal of Geosciences,2018,11(13):360-376.
[18] 徐永明,覃志豪,万洪秀.热红外遥感反演近地层气温的研究进展[J].国土资源遥感,2011(1):9-14.
XU Yong-ming,QIN Zhi-hao,WAN Hong-xiu.Advances in the Study of Near-surface Air Temperature Retrieval from Thermal Infrared Remote Sensing[J].Remote Sensing for Land and Resources,2011(1):9-14.
[19] 徐永明,覃志豪,沈 艳.基于MODIS数据的长江三角洲地区近地表气温遥感反演[J].农业工程学报,2011,27(9):63-68.
XU Yong-ming,QIN Zhi-hao,SHEN Yan.Estimation of Near-surface Air Temperature from MODIS Data in the Yangtze River Delta[J].Transactions of the CSAE,2011,27(9):63-68.
[20] PAPE R,LÖFFLE R.Modelling Spatio-temporal Near-surface Temperature Variation in High Mountain Landscapes[J].Ecological Modelling,2004,178(3):483-501.
[21] YOO C,IM J,PARK S,et al.Estimation of Daily Ma-ximum and Minimum Air Temperatures in Urban Landscapes Using MODIS Time Series Satellite Data[J].ISPRS Journal of Photogrammetry and Remote Sensing,2018,137:149-162.
[22] 高 亮,杜 鑫,李强子,等.融合土地覆盖和土壤水分产品的近地表空气温度空间化方法[J].地球信息科学学报,2020,22(10):2023-2037.
GAO Liang,DU Xin,LI Qiang-zi,et al.A Near-surface Air Temperature Spatialization Method Integrating Landuse and Soil Moisture Products[J].Journal of Geo-information Science,2020,22(10):2023-2037.
[23] 邢立亭,李 净,焦文慧.基于MODIS和随机森林的兰州市日最高气温和最低气温估算[J].干旱区研究,2020,37(3):689-695.
XING Li-ting,LI Jing,JIAO Wen-hui.Estimation of Daily Maximum and Minimum Temperature of Lan-zhou City Based on MODIS and Random Forest[J].Arid Zone Research,2020,37(3):689-695.
[24] 陈孟奇.基于MODIS数据的淮河流域近地表气温估算及年积温时空演变研究[D].南京:南京师范大学,2020.
CHEN Meng-qi.Study on Near-surface Air Temperature Estimation and Sptial Temporal Evolution of the Accumulated Temperature in Huai River Basin Based on MODIS Data[D].Nanjing:Nanjing Normal University,2020.
[25] 秦艳丽,时 鹏,何文虹,等.西安市城市化对景观格局及生态系统服务价值的影响[J].生态学报,2020,40(22):8239-8250.
QIN Yan-li,SHI Peng,HE Wen-hong,et al.Influence of Urbanization on Landscape Pattern and Ecosystem Service Value in Xi'an City[J].Acta Ecologica Sinica,2020,40(22):8239-8250.
[26] 孙 娴,魏 娜,郝 丽,等.近60年西安市极端气温事件变化特征分析[J].西北大学学报(自然科学版),2014,44(6):997-1000.
SUN Xian,WEI Na,HAO Li,et al.Characteristics of Extreme Temperature Events in Xi'an for the Latest 60 Years[J].Journal of Northwest University(Natural Science Edition),2014,44(6):997-1000.
[27] 金丽娜,李雄飞.2014～2017年西安市城市热岛、冷岛精细化时空特征分析[J].沙漠与绿洲气象,2021,15(1):97-102.
JIN Li-na,LI Xiong-fei.Spatial and Temporal Characteristics of Urban Heat Island and Cold Island in Xi'an City from 2014 to 2017[J].Desert and Oasis Meteorology,2021,15(1):97-102.
[28] 聂敬娣,张俊华,黄 波.城市热岛效应对人体健康影响研究综述[J].生态科学,2021,40(1):200-208.
NIE Jing-di,ZHANG Jun-hua,HUANG Bo.A Review of the Human Health Consequences of Urban Heat Island Effect[J].Ecological Science,2021,40(1):200-208.
[29] 白 琳,徐永明,何 苗,等.基于随机森林算法的近地表气温遥感反演研究[J].地球信息科学学报,2017,19(3):390-397.
BAI Lin,XU Yong-ming,HE Miao,et al.Remote Sensing Inversion of Near-surface Air Temperature Based on Random Forest[J].Journal of Geo-information Science,2017,19(3):390-397.
[30] 胡德勇,乔 琨,王兴玲,等.单窗算法结合Landsat 8热红外数据反演地表温度[J].遥感学报,2015,19(6):964-976.
HU De-yong,QIAO Kun,WANG Xing-ling,et al.Land Surface Temperature Retrieval from Landsat 8 Thermal Infrared Data Using Mono-window Algorithm[J].Journal of Remote Sensing,2015,19(6):964-976.
[31] 胡德勇,乔 琨,王兴玲,等.利用单窗算法反演Landsat 8 TIRS数据地表温度[J].武汉大学学报(信息科学版),2017,42(7):869-876.
HU De-yong,QIAO Kun,WANG Xing-ling,et al.Comparison of Three Single-window Algorithms for Retrieving Land-surface Temperature with Landsat 8 TIRS Data[J].Geomatics and Information Science of Wuhan University,2017,42(7):869-876.
[32] ROUSE J W,HAAS R H,SCHELL J A,et al.Monitoring Vegetation Systems in the Great Plains with ERTS[C]∥FREDEN S C,MERCANTI E P,BECKER M A.The Third Earth Resources Technology Sa-tellite-1 Symposium.Washington DC:NASA,1974:309-317.
[33] 徐涵秋.利用改进的归一化差异水体指数(MNDWI)提取水体信息的研究[J].遥感学报,2005,9(5):589-595.
XU Han-qiu.A Study on Information Extraction of Water Body with the Modified Normalized Difference Water Index(MNDWI)[J].Journal of Remote Sensing,2005,9(5):589-595.
[34] 查 勇,倪绍祥,杨 山.一种利用TM图像自动提取城镇用地信息的有效方法[J].遥感学报,2003,7(1):37-40.
ZHA Yong,NI Shao-xiang,YANG Shan.An Effective Approach to Automatically Extract Urban Land-use from TM Imagery [J].Journal of Remote Sensing,2003,7(1):37-40.
[35] LIANG S L.Narrowband to Broadband Conversions of Land Surface Albedo I Algorithms[J].Remote Sensing of Environment,2001,76(2):213-238.
[36] BREIMAN L.Random Forests[J].Machine Learning,2001,45(1):5-32.
[37] 李永丽,王 浩,金喜子.基于随机森林优化的自组织神经网络算法[J].吉林大学学报(理学版),2021,59(2):351-358.
LI Yong-li,WANG Hao,JIN Xi-zi.Self-organizing Neural Network Algorithm Based on Random Forest Optimization[J].Journal of Jilin University(Science Edition),2021,59(2):351-358.
[38] VLADIMIR S,ANDY L,CHRISTOPHER T,et al.Random Forest:A Classification and Regression Tool for Compound Classification and QSAR Modeling[J].Journal of Chemical Information and Computer Sciences,2003,43(6):1947-1958.
[39] MIN M,LI J,WANG F,et al.Retrieval of Cloud Top Properties from Advanced Geostationary Satellite Ima-ger Measurements Based on Machine Learning Algorithms[J].Remote Sensing of Environment,2020,239:111616.
[40] 方匡南,吴见彬,朱建平,等.随机森林方法研究综述[J].统计与信息论坛,2011,26(3):32-38.
FANG Kuang-nan,WU Jian-bin,ZHU Jian-ping,et al.A Review of Technologies on Random Forests[J].Statistics and Information Forum,2011,26(3):32-38.
[41] CHOU J S,PHAM A D.Nature-inspired Metaheuristic Optimization in Least Squares Support Vector Regression for Obtaining Bridge Scour Information[J].Information Sciences,2017,399:64-80.
[42] LIU Y,BI J W,FAN Z P.Multi-class Sentiment Classification:The Experimental Comparisons of Feature Selection and Machine Learning Algorithms[J].Expert Systems with Applications,2017,80:323-339.
[43] KOHAVI R.A Study of Cross-validation and Bootstrap for Accuracy Estimation and Model Selection[C]∥IJCAI.Proceedings of the 14th International Joint Conference on Artificial Intelligence.Montreal:IJCAI,1995:1137-1143.

相似文献/References:

[1]张梦涵,魏 进*,卞海丁.基于机器学习的边坡稳定性分析方法——以国内618个边坡为例[J].地球科学与环境学报,2022,44(06):1083.[doi:10.19814/j.jese.2022.09019]
　ZHANG Meng-han,WEI Jin*,BIAN Hai-ding.Slope Stability Analysis Method Based on Machine Learning—Taking 618 Slopes in China as Examples[J].Journal of Earth Sciences and Environment,2022,44(01):1083.[doi:10.19814/j.jese.2022.09019]

备注/Memo

备注/Memo:
收稿日期:2021-08-16; 修回日期:2021-11-03
基金项目:国家自然科学基金项目(41371220,42071345); 陕西省重点研发计划项目(2020ZDLSF06-07); 中央高校基本科研业务费专项资金项目(300102269112)
作者简介:冯 瑞(1996-),女,陕西延安人,理学硕士研究生,E-mail:fengruichd@163.com。*通讯作者:杨丽萍(1968-),女,陕西铜川人,副教授,理学博士,E-mail:zylpyang@chd.edu.cn。
更新日期/Last Update: 2022-02-25