基于FSDAF模型的干旱区典型绿洲城市夏季地表热岛效应时空演变研究

doi:10.12118/j.issn.1000–6060.2021.01.13

摘要/Abstract

摘要：

全球城市化进程加快引起的干旱区绿洲城市热岛效应变化及其生态环境问题已成为当前国内外城市气候、环境和生态等领域的研究热点之一。基于遥感热红外通道反演的地表温度（Land surface temperature,LST）是开展城市地表热岛（Surface urban heat island,SUHI）效应监测研究的关键参数。然而,受热红外遥感数据“时空矛盾”的制约,目前在单一星载卫星传感器下尚不存在同时具有高时间分辨率和高空间分辨率的热红外遥感数据源,因而制约了在干旱区绿洲城市范围内开展高精度地表热岛效应监测研究。针对上述问题,以干旱区典型绿洲城市——乌鲁木齐市为研究区域,以Landsat系列影像和MODIS地表温度产品为基础数据源,基于FSDAF （Flexible spatiotemporal data fusion method）时空融合模型分析了2001—2018年乌鲁木齐市在城市扩张背景下夏季地表热岛效应的时空变化特征以及夏季LST与城市地表参数之间的关系。研究结果表明：（1）乌鲁木齐市夏季热岛强度（SUHI intensity,SUHII）在不同的郊区范围内均呈现增加的趋势。在较小的郊区范围内,SUHII₁从2001年的1.24 ℃增加到了2018年的2.83 ℃;在较大的郊区范围内,SUHII₂从2001年的1.44 ℃增加到了2018年的2.88 ℃;（2）在研究区各土地利用类型中,裸地的夏季LST最高,水体最低;（3）研究区地表反照率和不透水面的增加与城市夏季LST升高呈正相关,而植被指数与植被覆盖度则与LST呈负相关关系;（4）在干旱区绿洲城市,城区内部植被面积的增加有助于缓解城市热岛效应,而仅郊区植被的增加则会导致SUHII的进一步增强。

关键词: 地表热岛, 时空融合模型, 干旱区绿洲城市

Abstract:

The urban heat island effect and ecological and environmental problems of an oasis city in an arid area resulting from the acceleration of the global urbanization process have become one of the current research hotspots worldwide. As a critical parameter in investigating the surface urban heat island (SUHI), the land surface temperature (LST) retrieved from thermal infrared images is, to some extent, inaccurate because there is no single sensor that can capture real-time LST at both high spatial and temporal resolutions. This paper used the flexible spatiotemporal data fusion method to generate high spatiotemporal resolution summer daytime LST data in 2001, 2011, and 2018. Spatiotemporal variations of the summer daytime SUHI over Urumqi, an oasis city in the arid area of western China, were assessed based on several urban surface biophysical variables. The results show that the SUHI intensity (SUHII) in Urumqi during the study period was calculated using two indicators of SUHII₁ (the urban and smaller rural area difference in the average LST) and SUHII₂ (the urban and larger rural area difference in the average LST). Significantly increasing trends of SUHII in the study area were observed. SUHII₁ increased from 1.24 °C in 2001 to 2.36 °C in 2011 and 2.83 °C in 2018, whereas SUHII₂ increased from 1.44 °C to 2.58 °C and 2.88 °C in the same periods. The highest and lowest summer daytime LST values were observed over areas of bare soil and water, respectively. The distribution of the summer daytime LST correlated positively with the albedo, the impervious surface area, and it correlated negatively with the enhanced vegetation index and fractional vegetation cover. The results emphasize the role of bare soils in aggravating the SUHI in cities in arid areas. Finally, we find that in oasis cities in arid areas, such as Urumqi, although increasing the amount of vegetation covering the urban area may be an effective way to mitigate the SUHI, more profuse vegetation coverage within a larger rural area will increase the SUHII during the summer daytime.

Key words: surface urban heat island, spatiotemporal fusion model, oasis city in arid area

王爽,王承武,张飞云. 基于FSDAF模型的干旱区典型绿洲城市夏季地表热岛效应时空演变研究[J]. 干旱区地理, 2021, 44(1): 118-130.

WANG Shuang,WANG Chengwu,ZHANG Feiyun. Spatiotemporal variations of the summer daytime surface urban heat island of oasis city in arid area based on FSDAF model[J]. Arid Land Geography, 2021, 44(1): 118-130.

图/表 15

图1

表1

表2

表3

图2

图3

表4

图4

图5

图6

图7

图8

图9

图10

表5

参考文献 21

[1]	Zhou X F, Chen H. Impact of urbanization-related land use land cover changes and urban morphology changes on the urban heat island phenomenon[J]. Science of the Total Environment, 2018,635:1467-1476. doi: 10.1016/j.scitotenv.2018.04.091
[2]	Ren P, Zhang X, Liang H, et al. Assessing the impact of land cover changes on surface urban heat islands with high-spatial-resolution imagery on a local scale: Workflow and case study[J]. Sustainability, 2019,11(19):5188. doi: 10.3390/su11195188
[3]	Huang B, Wang J, Song H, et al. Generating high spatiotemporal resolution land surface temperature for urban heat island monitoring[J]. IEEE Geoscience and Remote Sensing Letters, 2013,10(5):1011-1015. doi: 10.1109/LGRS.2012.2227930
[4]	Liu H, Weng Q. Enhancing temporal resolution of satellite imagery for public health studies: A case study of west Nile virus outbreak in Los Angeles in 2007[J]. Remote Sensing of Environment, 2012,117:57-71. doi: 10.1016/j.rse.2011.06.023
[5]	Wu P, Shen H, Zhang L, et al. Integrated fusion of multi-scale polar-orbiting and geostationary satellite observations for the mapping of high spatial and temporal resolution land surface temperature[J]. Remote Sensing of Environment, 2015,156:169-181. doi: 10.1016/j.rse.2014.09.013
[6]	郑明亮, 黄方, 张鸽. 基于TsHARP模型和STITFM 算法的地表温度影像融合研究[J]. 遥感技术与应用, 2018,33(2):275-283.
	[ Zheng Mingliang, Huang Fang, Zhang Ge. Land surface temperature data fusion of Landsat ETM and MODIS by combining TsHARP and STITFM mode[J]. Remote Sensing Technology and Application, 2018,33(2):275-283. ]
[7]	魏然, 单杰. 城市地表温度影像时空融合方法研究[J]. 武汉大学学报·信息科学版, 2018,43(3):428-435.
	[ Wei Ran, Shan Jie. Spatial and temporal fusion of urban land surface temperature image mapping[J]. Geomatics and Information Science of Wuhan University, 2018,43(3):428-435. ]
[8]	Zhu X, Hekmer E H, Gao F, et al. A flexible spatiotemporal method for fusing satellite images with different resolutions[J]. Remote Sensing of Environment, 2016,172:165-177. doi: 10.1016/j.rse.2015.11.016
[9]	李振杰, 金莉莉, 何清, 等. 乌鲁木齐市城区和郊区气温分布及廓线特征[J]. 干旱区地理, 2019,42(6):1273-1281.
	[ Li Zhenjie, Jin Lili, He Qing, et al. Characteristics of temperature distribution and profiles in Urumqi City and its suburb[J]. Arid Land Geography, 2019,42(6):1273-1281. ]
[10]	Qin Z, Karnaeli A, Berliner P. A mono-window algorithm for retrieving land surface temperature from Landsat TM data and its application to the Israel-Egypt border region[J]. International Journal of Remote Sensing, 2001,22(18):3719-3746. doi: 10.1080/01431160010006971
[11]	Zhang Y, Balzter H, Liu B, et al. Analyzing the impacts of urbanization and seasonal variation on land surface temperature based on subpixel fractional covers using Landsat images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2016,10(4):1344-1356. doi: 10.1109/JSTARS.4609443
[12]	杨敏, 杨贵军, 陈晓宁, 等. 基于FSDAF 方法融合生成高时空分辨率地表温度[J]. 国土资源遥感, 2018,30(1):54-62.
	[ Yang Min, Yang Guijun, Chen Xiaoning, et al. Generation of land surface temperature with high spatial and temporal resolution based on FSDAF method[J]. Remote Sensing for Land and Resources, 2018,30(1):54-62. ]
[13]	Peng S, Piao S, Ciais P, et al. Surface urban heat island across 419 global big cities[J]. Environmental Science & Technology, 2011,46, 696-703. doi: 10.1021/es2030438 pmid: 22142232
[14]	Dallimer M, Tang Z, Bibby P R, et al. Temporal changes in greenspace in a highly urbanized region[J]. Biology Letters, 2011,7(5):763-766. doi: 10.1098/rsbl.2011.0025
[15]	Zhou D, Zhao S, Zhang L, et al. Remotely sensed assessment of urbanization effects on vegetation phenology in China’s 32 major cities[J]. Remote Sensing of Environment, 2016,176:272-281. doi: 10.1016/j.rse.2016.02.010
[16]	Cai Y T, Zhang M, Lin H. Estimating the urban fractional vegetation cover using an object-based mixture analysis method and Sentinel-2 MSI imagery[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020,13:341-350. doi: 10.1109/JSTARS.4609443
[17]	Yang F, Matsushita B, Fukushima T. A pre-screened and normalized multiple endmember spectral mixture analysis for mapping impervious surface area in Lake Kasumigaura Basin, Japan[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2010,65(5):479-490. doi: 10.1016/j.isprsjprs.2010.06.004
[18]	Wang J, Wu Z, Wu C, et al. Improving impervious surface estimation: An integrated method of classification and regression trees (CART) and linear spectral mixture analysis (LSMA) based on error analysis[J]. GIScience & Remote Sensing, 2018,55(4):583-603.
[19]	Mackey C W, Lee X, Smith R B. Remotely sensing the cooling effects of city scale efforts to reduce urban heat island[J]. Building and Environment, 2012,49:348-358. doi: 10.1016/j.buildenv.2011.08.004
[20]	Liang S. Narrowband to broadband conversions of land surface albedo I: Algorithms[J]. Remote Sensing of Environment, 2001,76(2):213-238. doi: 10.1016/S0034-4257(00)00205-4
[21]	Knorn J, Rabe A, Radeloff V C, et al. Land cover mapping of large areas using chain classification of neighboring Landsat satellite images[J]. Remote Sensing of Environment, 2009,113:957-964. doi: 10.1016/j.rse.2009.01.010

卫星数据	数据获取日期（年-月-日）	数据获取时间（UTC）	已知时刻（t₁）基础输入LST影像	预测时刻（t_m）基础输入LST影像	验证数据
Landsat TM	2001-05-23	04:36	√
MOD11A1	2001-05-23	04:30	√
MOD11A1	2001-07-10	04:30		√
Landsat TM	2001-07-10	04:36			√
MOD11A1	2001-09-12	04:30		√
Landsat TM	2001-09-12	04:36			√
Landsat TM	2011-06-04	04:45	√
MOD11A1	2011-06-04	04:20	√
MOD11A1	2011-08-07	04:20		√
Landsat TM	2011-08-07	04:44			√
Landsat 8	2018-05-22	04:55	√
MOD11A1	2018-05-22	04:25	√
MOD11A1	2018-06-23	04:55		√
Landsat 8	2018-06-23	04:55			√
MOD11A1	2018-08-26	04:55		√
Landsat 8	2018-08-26	04:55			√
MOD11A1	2018-09-11	04:55		√
Landsat 8	2018-09-11	04:55			√

实验	年中天数/d	数据获取日期	预测时刻（t_n）的LST影像
实验	年中天数/d	数据获取日期	2001年	2011年	2018年
融合实验	169~176	6月18—25日		√	√
	177~184	6月26日—7月3日		√	√
	185~192	7月4—11日	√	√	√
	193~200	7月12—19日	√	√	√
	201~208	7月20—27日	√	√	√
	209~216	7月28日—8月4日	√	√	√
	217~224	8月5—12日	√	√	√
	225~232	8月13—20日	√	√	√
	233~240	8月21—28日	√	√	√
	241~248	8月29日—9月5日	√	√	√
	249~256	9月6日—9月13日	√	√	√

类型	验证数据				验证数量	分类数量	正确数量	生产者精度/%	用户精度/%
类型	植被	水体	裸土	建设用地	验证数量	分类数量	正确数量	生产者精度/%	用户精度/%
植被	172	0	13	15	209	200	172	82.30	86.00
水体	0	20	0	0	21	20	20	95.24	100.00
裸土	7	1	135	37	185	180	135	72.97	75.00
建设用地	30	0	37	933	985	1000	933	94.72	93.30

日期（年-月-日）	2001-6-10	2001-9-12	2011-8-7	2018-6-23	2018-8-26	2018-9-11
R²	0.93	0.88	0.94	0.91	0.90	0.86
AAD/K	2.16	2.44	2.15	2.27	2.30	2.59
RMSE/K	2.38	2.66	2.38	2.43	2.47	2.76

	2001年		2011年		2018年
	Area 1	Area 2	Area 1	Area 2	Area 1	Area 2
EVI	-0.79	-0.81	-0.81	-0.82	-0.84	-0.85
FVC	-0.67	-0.68	-0.73	-0.77	-0.74	-0.80
ISA	0.85	0.82	0.88	0.86	0.88	0.89
albedo	0.86	0.88	0.86	0.89	0.91	0.92