| Urban underlying surface is formed by coupling of various land cover types,and the contribution of different land cover types to land surface temperature is significantly different.Global warming and rapid urbanization have led to a surge in urban artificial surface area,resulting in great changes in urban microclimate,typical phenomena such as urban heat island effect.Rising temperatures will threaten the health of city dwellers,increase urban air pollution and increase building energy consumption.On the contrary,urban blue-green space usually shows the cold island effect(UCI),which can effectively cool the city.In this context,we should understand the distribution pattern of urban surface temperature and its influencing factors,identify the potential influencing mechanism of urban artificial surface and blue-green space on surface temperature,clarify the cold island effect and its influencing factors of urban blue-green space,and then give full play to the cooling effect of urban blue-green space to alleviate the heat island effect(UHI).It is of great practical significance to improve the urban living environment and promote the healthy and sustainable development of cities.Based on quantitative remote sensing inversion,this paper focuses on the cold island effect and its influencing factors of blue-green space in big cities in China,and explores the optimal size model that can give full play to the efficiency of urban bluegreen space cold island by applying the theories and methods of geographic information system,landscape ecology and statistics.The main conclusions are as follows:(1)The clustering and outlier analysis(Anselin Local Moran’s I)was applied to the spatial pattern analysis of urban LST,and on this basis,the geographic detector model was applied to the research on the driving factors of the spatial distribution of LST.The results showed that the LST of Wuhan was more prone to cluster phenomenon in summer,accounting for 23.81%of its area,and the lowest in winter,only 13.29%.The high temperature was mainly concentrated in built-up areas,showing a symmetrical distribution along the Yangtze River and extending from downtown to all sides.The low temperature areas were mainly concentrated in lakes and grassland areas.The spatial distribution of LST in Wuhan was the result of the joint action of various influencing factors,and there were interactions among various factors.The interaction of any two influence factors on the spatial distribution of LST was greater than the independent effect of a single influence factor.Overall,MNDWI,IBI,NDVI,and their interaction were the main drivers of the spatial distribution of LST in Wuhan.Different land cover types have different functions in regulating urban LST.The LST of water body and green land(forest land and grassland)tended to be 5.4℃ and 2.6℃ lower than that of construction land,respectively,in summer,confirming that there was indeed a significant difference in LST between urban blue-green space and artificial surface.(2)Hierarchical regression model(HRM)was used to explore the competitive influence mechanism of the cooling effect of blue-green space and the warming effect of artificial surface on urban LST,and to explore its differences in different climatic regions and cities with different development levels.The results showed that there was a critical point(threshold)for the impact of the ratio of blue-green space to artificial surface on LST in cities.When the ratio of blue-green space to artificial surface exceeds this value,it will replace artificial surface as the dominant factor of LST,bringing cooling effect.Otherwise,artificial surfaces will dominate the surface temperature,resulting in a warming effect.The overall hierarchical linear regression model(HRM)results for 28 large cities in China showed that the critical points of competition between artificial surfaces and water bodies and vegetation for surface temperature are 80%(R2=0.45,p<0.05)and 70%(R2=0.40,p<0.05),respectively,which means that water is a more powerful cooling source than vegetation in most cities.The critical point of cities in arid climate zone was higher than that in subtropical,temperate and sub cold climate zone,which reflected that their blue-green space is better in alleviating urban high temperature.The critical point of cities with higher economic development level was lower than that of cities with lower development level,which means that even areas with relatively low artificial surface coverage are prone to high temperature in these cities,and more attention should be paid to preventing heat island effect.(3)The cold island extent of blue-green space in China’s "Top Ten Furnaces" cities ranged from 72.45(Nanjing)to 98.61 m(Hangzhou),with an average of 87.21 m.Most of the blue-green space patches had short-distance cold island extent.The cold island intensity in each city ranged from 2.25(Nanning)to 2.91℃(Hangzhou),with an average of 2.49℃ and were significant differences in median and variability.The cold island efficiency thresholds ranged from 3.97(Chongqing)to 6.48(Fuzhou)ha,and showed three groups of distribution characteristics:the first group was 6.10±0.38 ha(Fuzhou,Wuhan,Changsha);the second group was 4.91±0.09 ha(Nanchang,Nanjing,Xi’an);the third group was 4.08±0.11 ha(Hangzhou,Nanning,Hefei,Chongqing).The cold island efficiency threshold of each group can provide a relatively universal reference for the planning and design of urban blue-green spatial patch size in the corresponding group.The cold island extent and cold island intensity of the water body were larger than those of the green space.The cooling island efficiency threshold of water bodies in most cities(80%)was greater than that of green spaces,which means that water bodies need to be larger than green spaces in order to fully exert their cooling effect.(4)The influence of the blue-green spatial landscape characteristics of China’s"Top Ten Furnace" cities on the extent and intensity of cold islands varied significantly at patch and landscape scales.At the patch scale,there was a very significant(p<0.01)positive correlation between the patch area and its cold island extent and cold island intensity,indicating that the larger the patch area,the greater its cooling extent and cold island intensity.The fractal dimension index(FRAC)and landscape shape index(LSI)of blue-green space in most cities(more than 70%)were also positively correlated with their cold island extent and cold island intensity(p<0.01),suggesting that the more complex the patch shape was,the greater the cold island extent and cold island intensity would be.The background temperature(BGT)of the patch had a very significant(p<0.01)negative correlation with its cold island extent and cold island intensity,indicating that the lower the background temperature of the patch,the greater its cold island extent and cold island intensity.The NDVI of green space and the MNDWI of water has positive effects on the extent and intensity of cold islands at the patch scale,and some cities has significant effects(p<0.05).The higher the index,the greater the cold island extent and cold island intensity.However,at the landscape scale,the influence of bluegreen space landscape features on its cold island extent and cold island intensity was no longer significant.The patch mean background temperature(BGT_MN)has a significant positive impact on the three-group distribution of the cold island efficiency thresholds in China’s "Top Ten Furnaces" cities(F=7.74,p<0.05).The higher the background temperature,the higher the cold island efficiency threshold.There was a significant positive correlation between the mean NDVI(NDVI_MN)and the cold island effect threshold(F=5.97,P<0.05).The results showed that NDVI had a significant positive effect on the cold island efficiency threshold of green space.The more luxuriant the vegetation,the better the growth,and the larger the cold island effect threshold. |
PDF Full Text Request