Effects of urban tree canopy loss on land surface temperature magnitude and timing

被引:79
作者
Elmes, Arthur [1 ]
Rogan, John [1 ]
Williams, Christopher [1 ]
Ratick, Samuel [1 ]
Nowak, David [2 ]
Martin, Deborah [1 ]
机构
[1] Clark Univ, Grad Sch Geog, 950 Main St, Worcester, MA 01610 USA
[2] US Forest Serv, USDA, Northern Res Stn, SUNY ESF, 5 Moon Lib, Syracuse, NY 13215 USA
关键词
Urban heat island; Seasonal trends analysis; Urban forestry; Image segmentation; Land surface temperature; Urban tree canopy mapping; ASIAN LONGHORNED BEETLE; CENTRAL MASSACHUSETTS; EMISSIVITY; CLASSIFICATION; FOREST; RETRIEVAL; VEGETATION; IMAGERY; SEGMENTATION; ALGORITHM;
D O I
10.1016/j.isprsjprs.2017.04.011
中图分类号
P9 [自然地理学];
学科分类号
0705 ; 070501 ;
摘要
Urban Tree Canopy (UTC) plays an important role in moderating the Surface Urban Heat Island (SUHI) effect, which poses threats to human health due to substantially increased temperatures relative to rural areas. UTC coverage is associated with reduced urban temperatures, and therefore benefits both human health and reducing energy use in cities. Measurement of this relationship relies on accurate, fine spatial resolution UTC mapping, and on time series analysis of Land Surface Temperatures (LST). The City of Worcester, Massachusetts underwent extensive UTC loss and gain during the relatively brief period from 2008 to 2015, providing a natural experiment to measure the UTC/LST relationship. This paper consists of two elements to this end. First, it presents methods to map UTC in urban and suburban locations at fine spatial resolution (similar to 0.5 m) using image segmentation of a fused Lidar/WorldView-2 dataset, in order to show UTC change over time. Second, the areas of UTC change are used to explore changes in LST magnitude and seasonal variability using a time series of all available Landsat data for the study area over the eight-year period from 2007 to 2015. Fractional UTC change per unit area was determined using fine resolution UTC maps for 2008, 2010, and 2015, covering the period of large-scale tree loss and subsequent planting. LST changes were measured across a series of net UTC change bins, providing a relationship between UTC net change and LST trend. LST was analyzed for both monotonic trends over time and changes to seasonal magnitude and timing, using Theil-Sen slopes and Seasonal Trend Analysis (STA), respectively. The largest magnitudes of UTC loss occurred in residential neighborhoods, causing increased exposure of impervious (road) and pervious (grass) surfaces. Net UTC loss showed higher monotonic increases in LST than persistence and gain areas. STA indicated that net UTC loss was associated greater difference between 2008 and 2015 seasonal temperature curves than persistence areas, and also larger peak LST values, with peak increases ranging from 1 to 6 degrees C. Timing of summer warm period was extended in UTC loss areas by up to 15 days. UTC gain provided moderate LST mitigation, with lower monotonic trends, lower peak temperatures, and smaller seasonal curve changes than both persistence and loss locations. This study shows that urban trees mitigate the magnitude and timing of the surface urban heat island effect, even in suburban areas with less proportional impervious coverage than the dense urban areas traditionally associated with SUHI. Trees can therefore be seen as an effective means of offsetting the energy-intensive urban heat island effect. (C) 2017 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS). Published by Elsevier B.V. All rights reserved.
引用
收藏
页码:338 / 353
页数:16
相关论文
共 65 条
[1]   Shade trees reduce building energy use and CO2 emissions from power plants [J].
Akbari, H .
ENVIRONMENTAL POLLUTION, 2002, 116 (SUPPL. 1) :S119-S126
[2]  
[Anonymous], 2001, TREES RISK RECLAIMIN
[3]  
Berk A., 2003, MODTRAN4 VERSION 3 R
[4]   Object based image analysis for remote sensing [J].
Blaschke, T. .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2010, 65 (01) :2-16
[5]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[6]   Segmentation of Landsat Thematic Mapper imagery improves buffelgrass (Pennisetum ciliare) pasture mapping in the Sonoran Desert of Mexico [J].
Brenner, Jacob C. ;
Christman, Zachary ;
Rogan, John .
APPLIED GEOGRAPHY, 2012, 34 :569-575
[7]   Improvements in land surface temperature retrieval from the Landsat series thermal band using water vapor and air temperature [J].
Cristobal, J. ;
Jimenez-Munoz, J. C. ;
Sobrino, J. A. ;
Ninyerola, M. ;
Pons, X. .
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 2009, 114
[8]   Seasonal Variations of the Urban Heat Island at the Surface and the Near-Surface and Reductions due to Urban Vegetation in Mexico City [J].
Cui, Yu Yan ;
de Foy, Benjamin .
JOURNAL OF APPLIED METEOROLOGY AND CLIMATOLOGY, 2012, 51 (05) :855-868
[9]   Fusion of hyperspectral and LIDAR remote sensing data for classification of complex forest areas [J].
Dalponte, Michele ;
Bruzzone, Lorenzo ;
Gianelle, Damiano .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2008, 46 (05) :1416-1427
[10]   Land surface temperature and emissivity estimation from passive sensor data:: theory and practice-current trends [J].
Dash, P ;
Göttsche, FM ;
Olesen, FS ;
Fischer, H .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 2002, 23 (13) :2563-2594