Muti-scale Effects Of Urban Forest On Urban Thermal Environment

Along with the development of urbanization, industrialization and climatewarming, urban thermal environment has deteriorated. It has been considered as oneof the major environmental problems in the21st century. Urban forest can helpdecrease Land surface temperature (LST) to improve urban thermal environment. Theresearch on the relationship between urban forest characteristics and urban thermalenvironment can provide useful information for urban planners about how toconstruct urban forest at the landscape level to improve urban thermal environmentand mitigate urban heat island.Base on the remote sensing technology and fieldmeasurement,the research has been conducted to explore the relationship betweenurban thermal environment and urban forest characteristics such as urban forestcommunity structures,urban forest patch characteristics and urban forest pattern inChangchun.So we can clarify the present situation of urban forest in Changchun andreveal the role of urban forest in the regulation of urban thermal environment. Thefollowing conclusion can be realized:(1) Status of urban forest in Changchun: Changchun was covered by33.08%ofurban forest and the number of trees was up to6.0349million. So Changchun wasconsidered as forest city. But urban forest amount varied greatly among differenttypes of urban forest and different districts and showed a definite gradient decreasingfrom suburban areas to urban center areas. The diameter, Canopy density, Basal areaand Leaf area Index class distribution were all skewed toward low values. Urbanforest patch was dominated by small patch with the size of0-0.5ha. Urban forestlandscape patterns in Changchun are less reasonable, with serious fragmentation ofpatches, the lack of large patch of urban forest and the low connections among urbanforest patch.(2) Relationship between urban forest community and urban thermalenvironment. The results showed that urban forest community structures had significant positive relationship with urban thermal environment effets. However, thestandardized coefficients (Std. coefficient) of multivariate regressions varied amongthe selected urban forest structures. The order of the Pearson correlation coefficientsbetween urban thermal environment and urban forest structures was canopy density>LAI> basal area> tree height> diameter> stem density for land surface temperature.The Pearson correlation coefficients for air temperature were lower than that for landsurface temperature, which suggested that the relationship between urban thermalenvironment and urban forest structures was stronger for land surface temperaturethan air temperature.But the change trend is basically the same and LAI and basalarea were still the main factors in affecting urban thermal environment. The canopydensity had a positive linear relationship with urban thermal effects. But LAI had apositive non-linear relationship with urban thermal effects, indicating that urbanthermal effects increased non-linearly with the increase of urban park area andperimeter.And the threshold value (0.22) existed in the regulation of urban thermalenvironment for LAI.When LAI was in the range of0.22-1.8, LAI can decreasetemperature with the maximum efficiency.(3) Relationship between urban forest patch characteristics and urban thermalenvironment.Urban forest patch characteristics can play an important role in decreasingtemperatrure and improve urban thermal environment. Urban forest patch area andperimeter had a nagative non-linear relationship with temperature inside urban forestpatch, indicating that the temperature inside urban forest patch increased non-linearlywith the increase of urban park area and perimeter. Urban forest patch area andperimeter had a positive non-linear relationship with PCI intensity and the range oftemperature influence, indicating that PCI intensity and the range of temperatureinfluence increased non-linearly with the increase of urban park area and perimeter.When urban forest patch area was in the range of2.8ha-25ha, urban forest patch areacan decrease temperature with the maximum efficiency. Urban park shape(perimeter/area) had a significant negative relationship with urban thermal environment effects. Given a fixed area of urban park, the ratio of area/perimeterreaches the lowest value when the urban park shape is round. So the rounder the urbanpark shape is, the better the cool island effect.(4) Relationship between urban forest pattern and urban thermal environment.In addition to urban forest amount, the configuration is also significantly correlated tourban thermal environment. PD, SPLIT and LSI had a positive linear relationship withLST at all scales, indicating that LST decreased with the increase of PD, SPLIT andLSI of urban vegetation. But the gradient of each linear model for PD, SPLIT and LSIincreased with the increase of the spatial scale. LPI, COHESION and AI had anegative relationship with LST at all scales, indicating that LST decreased with theincrease of LPI, CONHESION and AI of urban vegetation. But the gradient of eachlinear model of LPI, CONHESION and AI also increased with the increase of thespatial scale, which indicated that LST decreased more with the increase of LPI,COHESION and AI by one unit when the spatial become larger. It suggested that theeffects of the urban forest configuration on temperature should be considered at alarger scale (city\district) rather than a small scale. Besides the relationship betweenurban forest pattern and urban thermal environment also change with the chang ofurban forest cover in study area. When urban forest cover was in the range of20-40%,urban forest pattern can decrease temperature with the maximum efficiency.Based on the relationship between urban thermal environment and urban forestcharacteristics, the optimizing strategy for urban forest was approved to improveurban thermal environment such as different optimizing strategies for different scales.First, we should arrange the spatial distribution of patches in the districts with20-40%percent of its area covered by urban forest on the large scale (such as decreasing PD,SPLIT and increasing LPI, CONHESION and AI).Second we should regulate thepatch with different strategies for different patch size in medium scale.Finally, weshould designed urban forest community in optimal ways to increase Canopy densityand LAI for urban forest.Results from this research can expand our scientific understanding of the effects of urban forest characteristics on urban thermal environment at different spatial level,and provide insights for urban planners to mitigate Urban Heat Island by arrangingurban forest for the limited land resources for urban greening and optimizing urbanforest structures, which have important theoretical and management implications.