Next Article in Journal
Assessing the Ability of Image Based Point Clouds Captured from a UAV to Measure the Terrain in the Presence of Canopy Cover
Previous Article in Journal
Harvesting Design by Capital Return
Previous Article in Special Issue
The Concept of Landscape Structure, Forest Continuum and Connectivity as a Support in Urban Forest Management and Landscape Planning
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 10
Issue 3
10.3390/f10030282
forests-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Effects of Spatial Pattern of Forest Vegetation on Urban Cooling in a Compact Megacity

by
Wen Zhou
,
Fuliang Cao
* and
Guibin Wang
Co-Innovation Center for Sustainable Forestry in Southern China, Nanjing Forestry University; 159 Longpan Road, Nanjing 210037, China
*
Author to whom correspondence should be addressed.
Forests 2019, 10(3), 282; https://doi.org/10.3390/f10030282
Submission received: 25 February 2019 / Revised: 16 March 2019 / Accepted: 18 March 2019 / Published: 22 March 2019
(This article belongs to the Special Issue Planning and Optimization of Urban and Peri-urban Forests)
Browse Figures
Review Reports Versions Notes

Abstract

:
Urban forests can be an effective contributor to mitigate the urban heat island (UHI) effect. Understanding the factors that influence the cooling intensity of forest vegetation is essential for creating a more effective urban greenspace network to better counteract the urban warming. The aim of this study was to quantify the effects of spatial patterns of forest vegetation on urban cooling, in the Shanghai metropolitan area of China, using correlation analyses and regression models. Cooling intensity values were calculated based on the land surface temperature (LST) derived from remote sensing imagery and spatial patterns of forest vegetation were quantified by eight landscape metrics, using standard and moving-window approaches. The results suggested that 90 m × 90 m was the optimal spatial scale for studying the cooling effect of forest vegetation in Shanghai’s urban area. It also indicated that woodland performed better than grassland in urban cooling and the size, shape, and spatial distribution of woodland patches had significant impacts on the urban thermal environment. Specifically, the increase of size and the degree of compactness of the patch shape can effectively reduce the LST within the woodland. Areas with a higher percentage of vegetation coverage experienced a greater cooling effect. Moreover, when given a fixed amount of vegetation covers, aggregated distribution provided a stronger cooling effect than fragmented distribution and increasing overall shape complexity of woodlands can enhance the cooling effect on surrounding urban areas. This study provides insights for urban planners and landscape designers to create forest adaptive planning strategies to effectively alleviate the UHI effect.

1. Introduction

Urbanization has taken place at an unprecedented rate around the world in the past few decades. More than half of the world’s population now lives in towns and cities and the population will swell to about 5 billion by 2030 [1]. Much of this urbanization will take place in Africa and Asia, bringing huge social, economic, and environmental changes [2]. One of the noticeable impacts of rapid urbanization on the environment is the urban heat island (UHI) effect [3], which is a phenomenon that urban areas have higher temperatures compared to the surrounding rural areas [4,5,6]. Increased temperatures due to the UHI effect leads to reduced thermal comfort for urban dwellers and increased energy consumption for cooling urban indoor environments [7]. The UHI phenomenon also serves as a trap for outdoor atmospheric pollutants and affects the quality of life of urban dwellers [8,9]. Thus, UHI mitigation is urgent in order to improve the urban living environment [10].
Urban vegetation has consistently demonstrated a considerable cooling effect on mitigating urban warming, from courtyard scale to urban scale [11,12,13,14,15,16]. Many previous studies have suggested that increasing vegetation area and density is an effective strategy to reduce or alleviate the effects of urban warming [12,17,18,19,20]. Vegetated space, mainly through direct shading and evapotranspiration, can reduce temperatures and form an urban cool island (UCI) [21,22,23,24]. Additionally, vegetated spaces transform a small part of the absorbed solar radiation by photosynthesis, instead of converting that part to heat energy. For example, Yilmaz et al. (2007) found the mean temperature of the urban forest was lower than the urban area and the temperature contrast was mainly caused by the canopy and evapotranspiration effects of the urban forest [25].
The UCI usually refers to the phenomenon that vegetated spaces have lower temperature than that of the surrounding built-up areas [26]. Generally, there are two types of UCI based on the data source applied as follows: Air/atmospheric UCI and surface UCI. The air UCI related studies use atmospheric temperature obtained from fixed weather stations or mobile equipment [27,28]. Observations based on atmospheric temperature have demonstrated that the size and shape of vegetation patches, as well as the tree species, are significant factors to influence the cooling effects [18,29,30]. Those field-based observations have found that urban greenspaces are 1 °C–7 °C cooler than the surrounding areas [31,32,33,34,35]. However, the acquisition of air temperature data is normally restricted by field conditions and the amount of equipment needed, which has limited the research subjects to one individual greenspace or a small number of green sites [36,37,38,39]. Therefore, quantifiable cooling effects and statistical relationships cannot be established based on these site-specific studies, which does not allow recommendations to be generated on how to best incorporate forest vegetation into an urban area for reducing temperatures [28,40].
Previous studies regarding surface UCIs mostly use land surface temperature (LST) derived from infrared remote sensing imagery. Currently, various types of satellite images and developed GIS technologies have been applied to retrieve the LST and detailed land-use and land-cover (LULC) information [41,42,43]. Studies of Chen et al. (2012) and Schwarz et al. (2012) have demonstrated that LST derived from remotely sensed thermal infrared data are positively and significantly correlated with field-based atmospheric temperature [44,45]. Since the LST data across the study area is time synchronized and spatially continuous and also can be easily combined with LULC map [46,47], it has been widely applied to study the relationship between landscape patterns and the LST [48]. Relating the LST data to land cover patterns has enabled the development of urban level climate adaptive strategies. In this study, we focused on remotely sensed LST and, consequently, surface UCI. Additionally, most previous studies [36,40,49,50,51] have focused on providing qualitative descriptions of thermal patterns or simple correlations between the LST and the land cover characteristics and only few have attempted to explore the optimal spatial patterns of urban forests to cool the surrounding urban areas [26,52,53], especially at a fixed regional scale.
The application of landscape pattern metrics greatly advanced the approach of quantifying the relationship between vegetation spatial heterogeneity and its cooling effects [54,55]. Some previous studies have suggested that the cooling effect was scale dependent [56,57,58,59]. Specifically, the term “scale” here normally refers to the spatial resolution of remotely sensed imagery and the spatial extent [60]. Several studies have indicated that remotely sensed imagery with different spatial resolutions would influence the quantification of landscape patterns [61,62]. The study of Li et al. (2013) reported that remotely sensed imagery with a higher resolution could more accurately quantify the spatial pattern of greenspace [48]. In general, previous studies have primarily focused on the impacts of spatial resolution on the statistical relationship between urban green patterns and LST, but the optimal spatial extent for its study was undefined in current literature.
As past research illustrates, the influence mechanism of spatial patterns of forest vegetation on urban cooling was not fully explored. As a result, only sporadic recommendations and strategies have been obtained to guide urban forest and land-use planning to mitigate the UHI effects. Therefore, the main objectives of this study were the following: (1) To identify the optimal spatial extent for examining the cooling effect of urban forests in Shanghai; (2) to investigate the main patch characteristics that influence the surface thermal environment within urban woodlands; (3) to explore the optimal spatial composition and configuration of woodlands that can effectively cool the urban areas.

2. Study Area

Shanghai, the largest city and the economic center of China, is located on the coast of the East China Sea. The Shanghai metropolitan region covers a total area of approximately 6340.5 km2 and had a permanent resident population of about 24 million, as of 2015 [63]. It is largely situated on a broad flat alluvial plain with a few remnant hills in the southwest and the average elevation is about 4 m above the sea level [64]. Shanghai has a subtropical monsoon climate with a mean annual temperature of 17.1 °C and a mean annual precipitation of 1166.1 mm [65]. The regional vegetation mainly consists of evergreen broadleaved forests and evergreen broadleaved-deciduous broadleaved mixed forests [57]. Shanghai has experienced a remarkable increase in urban area over the past 50 years [66]. Currently, Shanghai is one of the most urbanized cities in China and the degree of urbanization was defined as the percentage of the total population living in urban areas [67]. Our study focused on the urbanized area within the Shanghai center city, an area of about 430 km2 (Figure 1).
There are several reasons of choosing this area as the study area. First, the center city (areas within the outer ring road of the city) is highly urbanized with very intensive and concentrated urban human activities. The impacts of various human activities incur many ecological and environmental problems, such as urban heat islands and water and air pollution. Therefore, the development of strategies to adapt to and mitigate the UHI effect is crucial and urgent. Second, the landscape pattern is highly heterogeneous and the distribution of vegetated areas is concentrated in some areas and scattered in others, which is favorable to study the influence of different spatial patterns of vegetated areas on LST. Last, we applied a moving-window method in this study. The moving-window method is described in Section 3.5. Therefore, the boundary effects were unavoidable. This means that only cells in which the entire (square) window is contained within the landscape are evaluated, which suggests that squared study area is the optimal choice to conduct this method.

3. Materials and Methods

3.1. Remote Sensing Data and the Pre-Processing

One Spot 6 image taken at 10:08 a.m., 25 July 2016, with four multiple spectral bands (6 m resolution) and one panchromatic band (1.5 m resolution) were used for LULC mapping, due to its high spatial resolution (1.5 m after image fusion). Additionally, a cloud-free Landsat-8 Thermal Infrared Sensor (TIRS) image (Row/Path: 038/118, 10:24 a.m., 20 July 2016) from the United States Geological Survey (https://glovis.usgs.gov/) was used for LST retrieval, since its thermal channel (30 m resolution) is widely applied to retrieve LST data. July 20th and July 25th both represented summer, the hot and humid season. Both Spot 6 images and the Landsat-8 TIRS image were clear and nearly free of clouds. Besides, the acquisition dates of the two types of remotely sensed images are only five days apart and, thus, we assumed that there was no change in the urban vegetated areas. These images were all geometrically rectified to same projection system (Datum: WGS 84 / UTM zone 51N) using ENVI software Version 5.3 (Exelis Inc.: Tysons Corner, VI, USA).

3.2. LULC Classification

Based on the Spot 6 fused image (combined multispectral and panchromatic bands), an urban LULC map was created by the object-oriented classification method, combined with manual visual interpretation. The study area was classified into five land-use types including the following: Woodland (trees with shrubs and grasses), grassland (shrubs and grasses), water, impervious surfaces (roads and buildings), and barren land (land without vegetation cover, mainly including exposed soil and landfill sites) (Figure 2a). Field surveys were also conducted to ascertain some doubtful pixels to improve the accuracy of the classified map. Next, 30 samples for each land cover type, 150 samples in total, were selected using a random stratified method to examine the accuracy of the classified map. Land use survey data derived from historical aerial photos, and a 1:250,000 digitalized land use map acquired in 2016 were used as the reference data. As a result, the overall accuracy of LULC classification was 94%.

3.3. Retrieval of LST

The Landsat 8 TIRS imagery was applied to retrieve LST. The TIRS image was rectified to a common UTM coordinate system based on the Spot 6 images. The LST was retrieved using the Mono-Window Algorithm (MWA) [68] with the following equation:
T s = [ a 10 ( 1 − C 10 − D 10 ) + ( b 10 ( 1 − C 10 − D 10 ) + C 10 + D 10 ) T 10 − D 10 T a ] / C 10 ,
where a 10 and b 10 are the coefficients and constant with values of −67.9542 and 0.45987, respectively (according to Qin et al. (2001) when the atmospheric temperature is in the range of 20–50 °C) [68], T10 represents the at-sensor brightness temperature, and Ta is the mean atmospheric temperature. The values C10 and D10 are defined, respectively, by Equations (2) and (3) as follows:
C 10 = τ 10 ε 10 ,
D 10 = ( 1 − τ 10 ) [ 1 + ( 1 − ε 10 ) τ 10 ] ,
where τ 10 and ε 10 represent the total atmospheric transmissivity and the land surface emissivity, respectively. The LST map was then resampled to 5 m × 5 m to match the LULC map. The result of the spatial distribution of LST is shown in Figure 2b.

3.4. Urban Green Pattern Metrics

The following eight landscape metrics were selected to quantify the spatial pattern of urban vegetated areas and to analyze the relationship between these patterns and corresponding LST reductions (Table 1): Patch area (PA), shape index (SI), percentage of landscape (PLAND), mean patch shape index (Shape_MN), largest patch index (LPI), mean area (Area_MN), number of patches (NP), and aggregation index (AI) [69]. PA and PLAND are composition metrics at patch level and class level, respectively. All others are spatial configuration related metrics and their corresponding application levels were shown in Table 1. The LULC map was converted to a grid format with a 5 m × 5 m cell size in order to conduct the standard and moving-window analysis in FRAGSTATS (3.3). The selection of grid size was mainly based on the resolution of Spot 6 images and, meanwhile, it has to be a divisor of the resolution of TIRS image (30 m) for later data extractions. Landscape metrics at the patch level were conducted by standard analysis in FRAGSTATS (3.3) and metrics at the class level in this study were computed by the moving-window method.

3.5. Moving-Window Analysis and Window Size Chosen

A moving window with specified shape (e.g., square in this study) and size was passed over the entire study area and values were returned to the focal cell to output a new complete and continuous grid map for each selected metric at the class [70]. Since previous studies have found that the cooling effect of green areas was scale dependent, the selection of the optimal scale of the moving window is crucial. As the resolution of TIRS band is 30 m, the tested window sizes were chosen as integer multiples of 30 m. Moreover, the mean patch size of woodland in the study area is 3800 m2 (about the same size as a 60 m × 60 m window) and the data analysis showed that 91.2% of the woodland patches were smaller than 8100 m2 (same size as a 90 m × 90 m window). In order to avoid dividing urban green patches into small parts by moving the window, four window sizes that were all greater than 60 m × 60 m were chosen and tested. As a result, four window sizes/spatial extents were tested (90 m × 90 m, 180 m × 180 m, 360 m × 360 m, and 720 m × 720 m) to find the optimal scale to investigate the relationship between spatial patterns of forest vegetation and temperature difference—the difference between the mean LST of each window unit and the mean LST of the study area.

3.6. The Calculation of Cooling Intensity

In this study, the mean LST (Tm) of the study area, excluding all water bodies, was treated as the reference land surface temperature. Hence, we defined the cooling intensity of woodland at patch level as △Tg = Tg − Tm, where Tg is the mean LST of a woodland patch. The mean LSTs of different land-use types were captured using the zonal analysis tool in ArcMap software. Moreover, the cooling intensity of vegetated areas to the surrounding urban area was calculated as △T = Ti − Tm, where Ti is the mean LST of each window/spatial extent unit over entire study area. The mean LST of each window patch was calculated by extracting the raster center values after resampling the LST map to corresponding window sizes.

3.7. Statistical Analysis

Statistical analyses were performed using the SPSS 23.0 (IBM: New York, NY, USA). Pearson correlation analysis was carried out to examine the relationship between landscape metrics and temperature reduction. Sample analyses were conducted to investigate the impact of SI on the cooling intensity. Linear regression analysis was conducted to further reveal the relationship between SI and PLAND of woodlands with temperature reduction. Window units with different values of PLAND of woodland were classified into several classes including the following: Pland_10 (9%–11%), Pland_20 (19%–21%), PA_30 (29%–31%), and Pland_40 (39%–41%). Since the study area was divided into fixed window units (e.g., 90 m × 90 m in this study), it was really hard to find sufficient sample units that contained exactly the same amounts of vegetation cover and, therefore, the scope of the PLAND value was expanded.

4. Results

4.1. The Optimal Spatial Extent for Examining the Cooling Effects of Forest Vegetation

The percentages of landscape (PLAND) of woodland were calculated by the moving-window method and showed significant linear relationships with temperature differences under four tested window sizes/spatial extents (Figure 3). The correlation coefficients of temperature differences and PLAND decreased when the window sizes increased. It may be because the PLAND values were unevenly distributed under greater spatial extents. Since it showed a good fit in Figure 3a (R² = 0.501) and also contained sufficient samples with relatively even distribution, the window size of 90 m × 90 m was decided as the optimal spatial extent in this study and was applied to subsequent analysis in Section 4.3.

4.2. Effects of Area and Shape of Urban Woodland Patches on Cooling Intensity

The results suggested that the thermal pattern was closely related to the distribution of LULC types. Similar to previous studies [50,56], water had the greatest cooling effect (−5.23 °C) followed by woodland (−2.19 °C) and grassland (−0.34 °C) compared to the mean LST. Besides, the mean LST of impervious surfaces was 4.87 °C higher than the mean LST, which was demonstrated as the main contributor of urban warming (Table 2). This study was targeted for urban forest vegetation and grassland showed a relative weak cooling effect and is highly attached to woodland. The following sections of this paper focus only on woodland.
The temperature will represent mixed pixels when the patch size is smaller than the pixel size of the LST map (30 m × 30 m), therefore, all patches with an area smaller than 0.5 ha were excluded for analysis. As a result, 845 woodland patches were selected for sample analysis. Pearson correlation analysis was then carried out between the patch area (PA) of woodlands and their cooling intensities, and the result suggested that PA was significantly correlated with the cooling intensity (r = 0.65 **, p < 0.01). As shown in Figure 4, the mean LSTs of woodland patches were presented by nine PA categories, including 0.5–2 ha (n = 403), 2–4 ha (n = 282), 4–6 ha (n = 60), 6–8 ha (n = 24), 8–10 ha (n = 22), 10–12 ha (n = 18), 12–14 ha (n = 18), 14–16 ha(n = 10), and 16–18 ha (n = 8) in sample plot and indicated that the PA is an important factor to influence the LST. The greatest cooling intensity of woodland patch in the study area was −7.82 °C. In general, the cooling intensities of larger woodland patches were stronger and also more stable than smaller ones. In addition, the cooling intensities were varied among woodland patches with similar sizes, especially for small sizes. The results also indicated that there may exist a maximum threshold with regard to the area of woodland patches.
In order to better understand the effect of shape of woodland patches on cooling intensity by different sizes, this study classified the woodland patches into several patch area (PA) classes including the following: PA_01 (0.9–1.1 ha), PA_05 (4.9–5.1 ha), PA_10 (9–11 ha) and PA_15 (13–17 ha). The results indicated that the effect of SI on the cooling intensity were statistically significant for PA_01 (R2 = 0.54, p < 0.01), PA_05 (R2 = 0.72, p < 0.01), PA_10 (R2 = 0.63, p < 0.01), and PA_15 (R2 = 0.53, p < 0.01) and all showed significant linear relationships (Figure 5). The SI was negatively correlated with the cooling intensity.

4.3. Effects of the Spatial Pattern of Vegetated Areas on Urban Cooling

As the only composition metric in the class level, the percentage of landscape (PLAND) of woodland was confirmed to have a significant impact on urban cooling. Linear regression analysis of the relationship between PLAND and temperature reduction (Figure 3a) showed that if the PLAND of a woodland was greater than 23%, in a 90 m × 90 m spatial extent, a cooling effect will be present. Window units containing grassland and water were excluded during the sample analysis because they also provide considerable cooling effects on surrounding areas. Several pattern metrics, including Shape_MN, LPI, Area_MN, NP, and AI, were computed to explore the influence of shape, fragmentation and aggregation of woodlands on urban cooling when the amount of vegetation was relatively fixed. The results showed that cooling intensity has positive correlations with Shape MN, LPI, Area_MN, and AI, and has a negative correlation with NP, which demonstrated that the spatial configuration of urban woodlands has a significant influence on the urban thermal environment (Table 3).

5. Discussion

Linking the method of landscape ecology and remote sensing, this study confirmed that both the composition and configuration of urban vegetated areas significantly affect the magnitude of LST in the study area. It suggested that, by balancing the amounts of vegetated areas and optimizing their spatial configuration, the cooling effect can be greatly enhanced. The main recommendations based on this study, in terms of maximizing cooling effects of greenspace, are to increase tree vegetation canopy cover and optimize green spatial patterns. Since land for urban greening is usually limited, especially for highly urbanized areas, interspersing small vegetation patches (e.g., greater than 2 ha, in this study) into urban land is also an effective method to cool the city. Apart from cooling, increasing vegetated areas can also provide some other ecological benefits, such as maintaining biodiversity, since larger greenspaces could contribute more to the conservation of biodiversity than small ones [71].

5.1. Implications of Optimal Spatial Extent Selection

Previous studies have confirmed that the optimal spatial scale for investigating the cooling effect of urban greenspaces is not yet known [57,59]. Since the percentage of landscape of vegetation was demonstrated as one of the key factors influencing the cooling effect [55], it was used to achieve the selection of the optimal spatial scale. In this study, a moving-window method was applied to identify the scaling effect on the relationship between urban green patterns and LSTs using landscape metrics. For this study area, a 90 m × 90 m window size was selected as the optimal spatial scale for investigating the cooling effects of forest vegetation. By studying the cooling effect under a fixed spatial scale (e.g., 90 m × 90 m square unit, in this study), it allowed specific recommendations to be generated on how to best incorporate forest vegetation in an urban area for reducing temperatures. Since no two cities are completely identical in geographical location, climate conditions, and spatial morphology, the optimal spatial scale for one city may differ from another. However, the idea and method of assessing the optimal spatial scale is replicable.

5.2. Implications of Patch Characteristics for Forest Adaptive Planning Strategies

Similar to the previous studies, the cooling effect of water is strongest, followed by woodland and grassland, due to lower boundary layer resistance, while impervious land is the primary contributor to the UHI phenomenon [50,56]. Results showed that the patch area and the shape index both have relationships with cooling intensity at the patch level. Specifically, the increase of area can better reduce the LST within woodland patches. The results also suggested that small woodland patches were also capable of mitigating urban warming and woodlands larger than 2 ha were normally effective in urban cooling. Additionally, the mean LSTs of smaller vegetation patches were more dissimilar than of the larger ones. The results suggested that the lower shape index of woodland can effectively enhance the cooling intensity, which means that greenspaces with relative regular and compact shapes performed better in cooling its internal thermal environment than irregular and elongated shapes. There is no agreed statement on how the shape of urban greenspace would influence the cooling intensity. For instance, the study by Feyisa et al. (2014) suggested that regular and compact shaped parks have a stronger cooling intensity compared to irregular and elongated parks [46]. However, Chen et al. (2014) demonstrated that SI is positively correlated with temperature reduction, indicating that urban green patches with more complex shapes will perform better in urban cooling [26]. Therefore, we can reasonably infer that the selection of research areas and types of remote sensing data, as well as the acquisition time of the data, may all influence the outcomes.

5.3. Implications of Spatial Patterns for Forest Adaptive Planning Strategies

Previous studies have focused on how the patch characteristics of vegetated areas influence its interior thermal environment. However, fewer have investigated the effects of spatial patterns of forest vegetation on the surrounding urban areas. Based on the linear regression analysis, increasing the percentage of landscape of woodlands can effectively enhance the cooling effect. Specifically, an occupancy of woodland greater than 23%, within a 90 m × 90 m, area could create a cool island capable of counteracting the UHI effect. The results suggested that increasing the amount of vegetation cover was more effective in urban cooling than optimizing the spatial configuration, which is consistent with the findings of Zhou et al. (2011) [55]. However, the spatial configuration of woodlands also has significant impact on the magnitude of LST. When given a fixed amount of vegetation, aggregated and concentrated distribution across the urban landscape is more effective to reduce the LST than scattered and fragmented distribution. The results also suggested that a vegetation patch with an irregular and elongated shape performed better in cooling its surrounding urban areas. This is probably because when a patch area is fixed, an irregular vegetation patch is in contact with larger area of surrounding landscape, compared to a compact one. In addition, a mainland-island spatial configuration of a woodland within a 90 m × 90 m area can effectively enhance the cooling effect.

6. Conclusions

Most of the previous studies regarding the cooling effect of urban greenspaces focused either on a small amount of green sites or certain types of greenspace, such as urban parks [35,51], green roofs [72], and roadside greenspaces [36]. However, fewer have attempted to assess the effects of spatial patterns of forest vegetation on urban cooling, especially at urban regional scale. Hence, conclusions drawn from an individual greenspace or few examples of green sites cannot be simply transferred and applied to make planning strategies to counteract the UHI effect at an urban and larger scale. Understanding the influence mechanism of forest vegetation on urban cooling is difficult because, not only patch characteristics (e.g., size and shape), but also the spatial configuration of urban greenspaces would affect the cooling intensity, as well as the characteristics of the surrounding urban area and climate conditions [26,51,69,73]. In this study, we have applied eight landscape metrics at patch and class levels and identified the main factors that influence the cooling intensity of urban forests. By generating a more comprehensive understanding in how forest vegetation can better cool its interior thermal environment and the surrounding urban areas, it may be possible to maximize the cooling effect in urban planning from the incipient planning stage.
With rapid urbanization, sprawl invades urban vegetated areas and they are largely replaced by extensive impervious low albedo paving and building materials and, to a large extent, causes the UHI phenomenon. For those highly urbanized areas, the removal of paved surfaces and buildings are expensive and impractical [74], however, the amount and spatial configuration of vegetated areas can be altered and optimized through vegetation management, such as increasing the tree vegetation canopy cover, the greening of roofs, adding vegetation cover on parking lots, and even building walls. Strategies relating to increasing the overall patch shape complexity are effective for cooling urban environments. Moreover, optimizing the structure of vegetation planting within a vegetated space, such as increasing the compactness of vegetation communities, can effectively reduce the internal temperature. Further quantifiable research is needed to generate compromised recommendations regarding to internal and external cooling effect of urban forests. Therefore, understanding the effects of composition and spatial configuration of forest vegetation on urban cooling is becoming increasingly important for tackling the UHI effect, particularly for areas where urbanization is still in process. The results of this study provide urban planners and natural resource managers with some theoretical and practical information on how urban forest vegetation should be planned and managed to better contradict the UHI effect.

Author Contributions

Conceptualization, W.Z. and F.C.; methodology, W.Z. and G.W.; software, W.Z.; validation, W.Z. and F.C.; formal analysis, W.Z.; investigation, W.Z.; resources, W.Z. and G.W.; data curation, W.Z.; writing—original draft preparation, W.Z.; writing—review and editing, W.Z., F.C and G.W.; visualization, W.Z.; supervision, F.C and G.W; project administration, F.C.; funding acquisition, W.Z. and F.C.

Funding

This research was funded by the National Key Research and Development Program of China (2017YFD0600701) and the Doctorate Fellowship Foundation of Nanjing Forestry University.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. United Nations, Department of Economic and Social Affairs, Population Division. World Urbanization Prospects: The 2014 Revision, CD-ROM ed.; Department of Economic and Social Affairs, Population Division: New York, NY, USA, 2014; ISBN 9789211515176. [Google Scholar]
  2. UNFPA. The state of the world population 2007: Unleashing the potential of urban growth. Indian Pediatr. 2008, 45, 481–482. [Google Scholar]
  3. Imhoff, M.L.; Zhang, P.; Wolfe, R.E.; Bounoua, L. Remote sensing of the urban heat island effect across biomes in the continental USA. Remote Sens. Environ. 2010, 114, 504–513. [Google Scholar] [CrossRef] [Green Version]
  4. Kolokotroni, M.; Ren, X.; Davies, M.; Mavrogianni, A. London’s urban heat island: Impact on current and future energy consumption in office buildings. Energy Build. 2012, 47, 302–311. [Google Scholar] [CrossRef]
  5. Voogt, J.A.; Oke, T.R. Thermal remote sensing of urban climates. Remote Sens. Environ. 2003, 86, 370–384. [Google Scholar] [CrossRef]
  6. Li, X.; Li, W.; Middel, A.; Harlan, S.L.; Brazel, A.J.; Turner, B.L. Remote sensing of the surface urban heat island and land architecture in Phoenix, Arizona: Combined effects of land composition and configuration and cadastral-demographic-economic factors. Remote Sens. Environ. 2016, 174, 233–243. [Google Scholar] [CrossRef]
  7. White, M.A.; Nemani, R.R.; Thornton, P.E.; Running, S.W. Satellite evidence of phenological differences between urbanized and rural areas of the eastern United States deciduous broadleaf forest. Ecosystems 2002, 5, 260–273. [Google Scholar] [CrossRef]
  8. Gobakis, K.; Kolokotsa, D.; Synnefa, A.; Saliari, M.; Giannopoulou, K.; Santamouris, M. Development of a model for urban heat island prediction using neural network techniques. Sustain. Cities Soc. 2011, 1, 104–115. [Google Scholar] [CrossRef]
  9. Santamouris, M.; Paraponiaris, K.; Mihalakakou, G. Estimating the ecological footprint of the heat island effect over Athens, Greece. Clim. Chang. 2007, 80, 265–276. [Google Scholar] [CrossRef]
  10. Opdam, P.; Luque, S.; Jones, K.B. Changing landscapes to accommodate for climate change impacts: A call for landscape ecology. Landsc. Ecol. 2009, 24, 715–721. [Google Scholar] [CrossRef]
  11. Susca, T.; Gaffin, S.R.; Dell’osso, G.R.; Dell’Osso, G.R. Positive effects of vegetation: urban heat island and green roofs. Environ. Pollut. 2011, 159, 2119–2126. [Google Scholar] [CrossRef]
  12. Zoulia, I.; Santamouris, M.; Dimoudi, A. Monitoring the effect of urban green areas on the heat island in Athens. Environ. Monit. Assess. 2009, 156, 275–292. [Google Scholar] [CrossRef] [PubMed]
  13. Bruse, M.; Fleer, H. Simulating surface-plant-air interactions inside urban environments with a three dimensional numerical model. Environ. Model. Softw. 1998, 13, 373–384. [Google Scholar] [CrossRef]
  14. Shashua-Bar, L.; Pearlmutter, D.; Erell, E. The cooling efficiency of urban landscape strategies in a hot dry climate. Landsc. Urban Plan. 2009, 92, 179–186. [Google Scholar] [CrossRef]
  15. Mackey, C.W.; Lee, X.; Smith, R.B. Remotely sensing the cooling effects of city scale efforts to reduce urban heat island. Build. Environ. 2012, 49, 348–358. [Google Scholar] [CrossRef]
  16. Oliveira, S.; Andrade, H.; Vaz, T. The cooling effect of green spaces as a contribution to the mitigation of urban heat: A case study in Lisbon. Build. Environ. 2011, 46, 2186–2194. [Google Scholar] [CrossRef]
  17. Hamada, S.; Ohta, T. Seasonal variations in the cooling effect of urban green areas on surrounding urban areas. Urban For. Urban Green. 2010, 9, 15–24. [Google Scholar] [CrossRef]
  18. Georgi, J.N.; Dimitriou, D. The contribution of urban green spaces to the improvement of environment in cities: Case study of Chania, Greece. Build. Environ. 2010, 45, 1401–1414. [Google Scholar] [CrossRef] [Green Version]
  19. Eliasson, I. The use of climate knowledge in urban planning. Landsc. Urban Plan. 2000, 48, 31–44. [Google Scholar] [CrossRef]
  20. Dimoudi, A.; Nikolopoulou, M. Vegetation in the urban environment: Microclimatic analysis and benefits. Energy Build. 2003, 35, 69–76. [Google Scholar] [CrossRef]
  21. Rahman, M.A.; Moser, A.; Rötzer, T.; Pauleit, S. Within canopy temperature differences and cooling ability of Tilia cordata trees grown in urban conditions. Build. Environ. 2017, 114, 118–128. [Google Scholar] [CrossRef]
  22. Rahman, M.A.; Armson, D.; Ennos, A.R. Effect of urbanization and climate change in the rooting zone on the growth and physiology of Pyrus calleryana. Urban For. Urban Green. 2014, 13, 325–335. [Google Scholar] [CrossRef]
  23. Grimmond, C.S.B.; Oke, T.R. An evapotranspiration-interception model for urban areas. Water Resour. Res. 1991, 27, 1739–1755. [Google Scholar] [CrossRef]
  24. Taha, H.; Akbari, H.; Rosenfeld, A.; Huang, J. Residential cooling loads and the urban heat island-the effects of albedo. Build. Environ. 1988, 23, 271–283. [Google Scholar] [CrossRef]
  25. Yilmaz, S.; Toy, S.; Irmak, M.A.; Yilmaz, H. Determination of climatic differences in three different land uses in the city of Erzurum, Turkey. Build. Environ. 2007, 42, 1604–1612. [Google Scholar] [CrossRef]
  26. Chen, A.; Yao, X.A.; Sun, R.; Chen, L. Effect of urban green patterns on surface urban cool islands and its seasonal variations. Urban For. Urban Green. 2014, 13, 646–654. [Google Scholar] [CrossRef] [Green Version]
  27. Song, J.; Li, X. A study on the cooling effects of greenery on the surrounding areas by computer simulation for green built environment. In Proceedings of the Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Beijing, China, 6–9 September 2010; Volume 6330, pp. 653–661. [Google Scholar]
  28. Bowler, D.E.; Buyung-Ali, L.; Knight, T.M.; Pullin, A.S. Urban greening to cool towns and cities: A systematic review of the empirical evidence. Landsc. Urban Plan. 2010, 97, 147–155. [Google Scholar] [CrossRef]
  29. Fintikakis, N.; Gaitani, N.; Santamouris, M.; Assimakopoulos, M.; Assimakopoulos, D.N.; Fintikaki, M.; Albanis, G.; Papadimitriou, K.; Chryssochoides, E.; Katopodi, K.; et al. Bioclimatic design of open public spaces in the historic centre of Tirana, Albania. Sustain. Cities Soc. 2011, 1, 54–62. [Google Scholar] [CrossRef]
  30. Rahman, M.A.; Armson, D.; Ennos, A.R. A comparison of the growth and cooling effectiveness of five commonly planted urban tree species. Urban Ecosyst. 2015, 18, 371–389. [Google Scholar] [CrossRef]
  31. Jauregui, E. Influence of a large urban park on temperature and convective precipitation in a tropical city. Energy Build. 1990, 15, 457–463. [Google Scholar] [CrossRef]
  32. Upmanis, H.; Eliasson, I.; Lindqvist, S. The influence of green areas on nocturnal temperatures in a high latitude city (Goteborg, Sweden). Int. J. Climatol. 1998, 18, 681–700. [Google Scholar] [CrossRef]
  33. Spronken-Smith, R.A.; Oke, T.R. The thermal regime of urban parks in two cities with different summer climates. Int. J. Remote Sens. 1998, 19, 2085–2104. [Google Scholar] [CrossRef]
  34. Kardinal Jusuf, S.; Wong, N.H.; Hagen, E.; Anggoro, R.; Hong, Y. The influence of land use on the urban heat island in Singapore. Habitat Int. 2007, 31, 232–242. [Google Scholar] [CrossRef]
  35. Chang, C.R.; Li, M.H.; Chang, S.D. A preliminary study on the local cool-island intensity of Taipei city parks. Landsc. Urban Plan. 2007, 80, 386–395. [Google Scholar] [CrossRef]
  36. Park, J.; Kim, J.H.; Lee, D.K.; Park, C.Y.; Jeong, S.G. The influence of small green space type and structure at the street level on urban heat island mitigation. Urban For. Urban Green. 2017, 21, 203–212. [Google Scholar] [CrossRef]
  37. Bacci, L.; Morabito, M.; Raschi, A.; Ugolini, F. Thermohygrometric Conditions of Some Urban Parks of Florence (Italy) and Their Effects on Human Well-Being. Trees 2003, 6, 49. [Google Scholar]
  38. Chen, Y.; Wong, N.H. Thermal benefits of city parks. Energy Build. 2006, 38, 105–120. [Google Scholar]
  39. Doick, K.; Hutchings, T. Air temperature regulation by urban trees and green infrastructure. For. Comm. 2013, 1–10. [Google Scholar]
  40. Lin, W.; Yu, T.; Chang, X.; Wu, W.; Zhang, Y. Calculating cooling extents of green parks using remote sensing: Method and test. Landsc. Urban Plan. 2015, 134, 66–75. [Google Scholar] [CrossRef]
  41. Tran, H.; Uchihama, D.; Ochi, S.; Yasuoka, Y. Assessment with satellite data of the urban heat island effects in Asian mega cities. Int. J. Appl. Earth Obs. Geoinf. 2006, 8, 34–48. [Google Scholar] [CrossRef]
  42. Schwarz, N.; Lautenbach, S.; Seppelt, R. Exploring indicators for quantifying surface urban heat islands of European cities with MODIS land surface temperatures. Remote Sens. Environ. 2011, 115, 3175–3186. [Google Scholar] [CrossRef]
  43. Small, C. Comparative analysis of urban reflectance and surface temperature. Remote Sens. Environ. 2006, 104, 168–189. [Google Scholar] [CrossRef]
  44. Chen, X.; Su, Y.; Li, D.; Huang, G.; Chen, W.; Chen, S. Study on the cooling effects of urban parks on surrounding environments using Landsat TM data: A case study in Guangzhou, southern China. Int. J. Remote Sens. 2012, 33, 5889–5914. [Google Scholar] [CrossRef]
  45. Schwarz, N.; Schlink, U.; Franck, U.; Großmann, K. Relationship of land surface and air temperatures and its implications for quantifying urban heat island indicators—An application for the city of Leipzig (Germany). Ecol. Indic. 2012, 18, 693–704. [Google Scholar] [CrossRef]
  46. Feyisa, G.L.; Dons, K.; Meilby, H. Efficiency of parks in mitigating urban heat island effect: An example from Addis Ababa. Landsc. Urban Plan. 2014, 123, 87–95. [Google Scholar] [CrossRef]
  47. Weng, Q. Thermal infrared remote sensing for urban climate and environmental studies: Methods, applications, and trends. ISPRS J. Photogramm. Remote Sens. 2009, 64, 335–344. [Google Scholar] [CrossRef]
  48. Li, Z.-L.; Tang, B.-H.; Wu, H.; Ren, H.; Yan, G.; Wan, Z.; Trigo, I.F.; Sobrino, J.A. Satellite-derived land surface temperature: Current status and perspectives. Remote Sens. Environ. 2013, 131, 14–37. [Google Scholar] [CrossRef] [Green Version]
  49. Tan, M.; Li, X. Integrated assessment of the cool island intensity of green spaces in the mega city of Beijing. Int. J. Remote Sens. 2013, 34, 3028–3043. [Google Scholar] [CrossRef]
  50. Keramitsoglou, I.; Kiranoudis, C.T.; Ceriola, G.; Weng, Q.; Rajasekar, U. Identification and analysis of urban surface temperature patterns in Greater Athens, Greece, using MODIS imagery. Remote Sens. Environ. 2011, 115, 3080–3090. [Google Scholar] [CrossRef]
  51. Cao, X.; Onishi, A.; Chen, J.; Imura, H. Quantifying the cool island intensity of urban parks using ASTER and IKONOS data. Landsc. Urban Plan. 2010, 96, 224–231. [Google Scholar] [CrossRef]
  52. Fan, H.Y.; Yu, Z.W.; Yang, G.Y.; Liu, T.Y. How to cool hot-humid (Asian) cities with urban trees? An optimal landscape size perspective. Agric. For. Meteorol. 2019, 265, 338–348. [Google Scholar] [CrossRef]
  53. Li, X.; Zhou, W.; Ouyang, Z.; Xu, W.; Zheng, H. Spatial pattern of greenspace affects land surface temperature: Evidence from the heavily urbanized Beijing metropolitan area, China. Landsc. Ecol. 2012, 27, 887–898. [Google Scholar] [CrossRef]
  54. Evju, M.; Sverdrup-Thygeson, A. Spatial configuration matters: A test of the habitat amount hypothesis for plants in calcareous grasslands. Landsc. Ecol. 2016, 31, 1891–1902. [Google Scholar] [CrossRef]
  55. Zhou, W.; Huang, G.; Cadenasso, M.L. Does spatial configuration matter? Understanding the effects of land cover pattern on land surface temperature in urban landscapes. Landsc. Urban Plan. 2011, 102, 54–63. [Google Scholar] [CrossRef]
  56. Weng, Q.; Lu, D.; Schubring, J. Estimation of land surface temperature-vegetation abundance relationship for urban heat island studies. Remote Sens. Environ. 2004, 89, 467–483. [Google Scholar] [CrossRef]
  57. Li, J.; Song, C.; Cao, L.; Zhu, F.; Meng, X.; Wu, J. Impacts of landscape structure on surface urban heat islands: A case study of Shanghai, China. Remote Sens. Environ. 2011, 115, 3249–3263. [Google Scholar] [CrossRef]
  58. Li, X.; Zhou, W.; Ouyang, Z. Relationship between land surface temperature and spatial pattern of greenspace: What are the effects of spatial resolution? Landsc. Urban Plan. 2013, 114, 1–8. [Google Scholar] [CrossRef]
  59. Liu, H.; Weng, Q.H. Scaling Effect on the Relationship between Landscape Pattern and Land Surface Temperature: A Case Study of Indianapolis, United States. Photogramm. Eng. Remote Sensing 2009, 75, 291–304. [Google Scholar] [CrossRef]
  60. Wu, J. Effects of changing scale on landscape pattern analysis: Scaling relations. Landsc. Ecol. 2004, 19, 125–138. [Google Scholar] [CrossRef]
  61. Vannier, C.; Vasseur, C.; Hubert-Moy, L.; Baudry, J. Multiscale ecological assessment of remote sensing images. Landsc. Ecol. 2011, 26, 1053–1069. [Google Scholar] [CrossRef]
  62. Townsend, P.A.; Lookingbill, T.R.; Kingdon, C.C.; Gardner, R.H. Spatial pattern analysis for monitoring protected areas. Remote Sens. Environ. 2009, 113, 1410–1420. [Google Scholar] [CrossRef]
  63. Shanghai Municipal Bureau Statistics Shanghai Statistical Yearbook 2016. Available online: www.stats-sh.gov.cn/tjnj/nj16.htm?d1=2016tjnj/C0201.htm (accessed on 30 November 2018).
  64. Huang, W.; Li, J.; Guo, Q.; Mansaray, L.R.; Li, X.; Huang, J. A satellite-derived climatological analysis of urban heat Island over Shanghai during 2000–2013. Remote Sens. 2017, 9, 641. [Google Scholar] [CrossRef]
  65. Qiu, T.; Song, C.; Li, J. Impacts of urbanization on vegetation phenology over the past three decades in Shanghai, China. Remote Sens. 2017, 9, 970. [Google Scholar] [CrossRef]
  66. Li, J.; Wang, X.; Wang, X.; Ma, W.; Zhang, H. Remote sensing evaluation of urban heat island and its spatial pattern of the Shanghai metropolitan area, China. Ecol. Complex. 2009, 6, 413–420. [Google Scholar] [CrossRef]
  67. Du, H.; Song, X.; Jiang, H.; Kan, Z.; Wang, Z.; Cai, Y. Research on the cooling island effects of water body: A case study of Shanghai, China. Ecol. Indic. 2016, 67, 31–38. [Google Scholar] [CrossRef]
  68. Qin, Z.; Karnieli, 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. Int. J. Remote Sens. 2001, 22, 3719–3746. [Google Scholar] [CrossRef]
  69. McGarigal, K.; Cushman, S.A.; Neel, M.C.; Ene, E. FRAGSTATS v3: Spatial Pattern Analysis Program for Categorical Maps. Computer Software Program Produced by the Authors at the University of Massachusetts, Amherst. 2002. Available online: http://www.umass.edu/landeco/research/fragstats/fragstats.html (accessed on 25 July 2018).
  70. McGarigal, K.; Cushman, S.A. The gradient concept of landscape structure. In Issues and Perspectives in Landscape Ecology; Cambridge University Press: Cambridge, UK, 2005; pp. 112–119. ISBN 9780511614415. [Google Scholar]
  71. Beninde, J.; Veith, M.; Hochkirch, A. Biodiversity in cities needs space: A meta-analysis of factors determining intra-urban biodiversity variation. Ecol. Lett. 2015, 18, 581–592. [Google Scholar] [CrossRef]
  72. Klein, P.M.; Coffman, R. Establishment and performance of an experimental green roof under extreme climatic conditions. Sci. Total Environ. 2015, 512–513, 82–93. [Google Scholar] [CrossRef] [PubMed]
  73. Ng, E.; Chen, L.; Wang, Y.; Yuan, C. A study on the cooling effects of greening in a high-density city: An experience from Hong Kong. Build. Environ. 2012, 47, 256–271. [Google Scholar] [CrossRef]
  74. Rosanna, M.M.; Chapin, F.S. Urban Land Use Planning. Am. Cathol. Sociol. Rev. 2007. [Google Scholar] [CrossRef]
Forests 10 00282 g001 550
Figure 1. Location map of Shanghai and the study area. The false color Spot 6 image is displayed with RGB composition of band 5 (near infrared), band 4 (red), and band 3 (green).
Figure 1. Location map of Shanghai and the study area. The false color Spot 6 image is displayed with RGB composition of band 5 (near infrared), band 4 (red), and band 3 (green).
Forests 10 00282 g001
Forests 10 00282 g002 550
Figure 2. (a) LULC map of the study area on July 25th, 2016, based on Spot 6 images. (b) Spatial distribution of LST of the study area on 20 July 2016, derived from a Landsat 8 TIRS image.
Figure 2. (a) LULC map of the study area on July 25th, 2016, based on Spot 6 images. (b) Spatial distribution of LST of the study area on 20 July 2016, derived from a Landsat 8 TIRS image.
Forests 10 00282 g002
Forests 10 00282 g003 550
Figure 3. (a–d) Relationship between the percentage of landscape (PLAND) of woodland and the temperature difference (°C) in (a) 90 m × 90 m, (b) 180 m × 180 m, (c) 360 m × 360 m, and (d) 720 m × 720 m window sizes.
Figure 3. (a–d) Relationship between the percentage of landscape (PLAND) of woodland and the temperature difference (°C) in (a) 90 m × 90 m, (b) 180 m × 180 m, (c) 360 m × 360 m, and (d) 720 m × 720 m window sizes.
Forests 10 00282 g003
Forests 10 00282 g004 550
Figure 4. Land surface temperature (LST) by patch area (PA) of woodland in sample plot.
Figure 4. Land surface temperature (LST) by patch area (PA) of woodland in sample plot.
Forests 10 00282 g004
Forests 10 00282 g005 550
Figure 5. (a–d) Relationship between the cooling intensity (°C) and SI of (a) PA_01, (b) PA_05, (c) PA_10, (d) PA_15 of woodland.
Figure 5. (a–d) Relationship between the cooling intensity (°C) and SI of (a) PA_01, (b) PA_05, (c) PA_10, (d) PA_15 of woodland.
Forests 10 00282 g005
Table
Table 1. Landscape metrics used to describe urban forest patterns.
Table 1. Landscape metrics used to describe urban forest patterns.
Landscape MetricsAbbreviationApplication LevelsDescriptionUnits
Patch areaPAPatchThe area of the patchHectares
Shape indexSIPatchThe most straightforward measure of overall shape complexityNone
Percentage of LandscapePLANDClassThe proportion of total area occupied by a particular patch type; a measure of landscape composition and dominance of patch typesPercent
Mean patch shape indexShape_MNClassMean value of shape index of a particular patch typeNone
Largest patch indexLPIClassThe area (m2) of the largest patch of the corresponding patch type divided by total landscape area (m2), multiplied by 100 (to convert to a percentage)Percent
Mean areaArea_MNClassThe sum of area across all patches of the corresponding patch type divided by the number of patches of the same typeHectares
Number of patchesNPClassThe number of patches of the corresponding patch type/landscapeNone
Aggregation indexAIClassThe number of like adjacencies involving the corresponding class, divided by the maximum possible number of like adjacencies involving the corresponding class, which is achieved when the class is maximally clumped into a single, compact patch; multiplied by 100 (to convert to a percentage)Percent
Table
Table 2. Results of the mean LST of different land-use types and the cooling effect (°C).
Table 2. Results of the mean LST of different land-use types and the cooling effect (°C).
LULC TypeMean LSTStandard Deviation of LSTsTemperature Reduction Compared with The mean LST of Study AreaTemperature Reduction Compared with the Mean LST of Impervious Surface
Woodland (n = 7049)40.232.05−2.19−7.06
Grassland (n = 4098)42.081.96−0.34−5.21
Impervious surface (n = 2786)47.291.524.870
Barren land (n = 495)43.520.891.1−3.77
Water (n = 1250)37.191.01−5.23−10.1
Table
Table 3. Pearson correlations between configuration metrics and cooling intensity (°C).
Table 3. Pearson correlations between configuration metrics and cooling intensity (°C).
Pattern MetricsCooling Intensity
Pland_10
(n = 580)		Pland_20
(n = 365)		Pland_30
(n = 260)		Pland_40
(n = 190)
Shape_MN0.38 **0.23 **0.45 **0.39 **
LPI0.23 **0.31 **0.43 **0.10 *
Area_MN0.20 **0.26 **0.39 **0.47 **
NP−0.19 **−0.12 *−0.25 **−0.27 **
AI0.24 **0.29 **0.34 **0.35 **
* Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).

Share and Cite

MDPI and ACS Style

Zhou, W.; Cao, F.; Wang, G. Effects of Spatial Pattern of Forest Vegetation on Urban Cooling in a Compact Megacity. Forests 2019, 10, 282. https://doi.org/10.3390/f10030282

AMA Style

Zhou W, Cao F, Wang G. Effects of Spatial Pattern of Forest Vegetation on Urban Cooling in a Compact Megacity. Forests. 2019; 10(3):282. https://doi.org/10.3390/f10030282

Chicago/Turabian Style

Zhou, Wen, Fuliang Cao, and Guibin Wang. 2019. "Effects of Spatial Pattern of Forest Vegetation on Urban Cooling in a Compact Megacity" Forests 10, no. 3: 282. https://doi.org/10.3390/f10030282

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop