Relationship Between Multi-scale Green Spatial Pattern And Near-ground Climate Effect

Urban near-ground climate problems such as urban heat islands and urban calm wind have become major hidden dangers for residents’health and urban sustainable development.Scientific green space system layout can improve the near ground climate,due to its ecological benefits.A green space system is a multi-scale and multi-level integrated system.The relationship between green space patterns and near-ground climate in different scales is one of the scientific bases for green space system planning.However,due to the limitation of near climate data,most studies mainly focus on the relationship between green space patterns and the near ground climate on the micro block scale.The purpose of this study is to explore the correlation between the spatial pattern of green space and the near-ground wind temperature field in the administrative region and the central area of a city,which is helpful to explore the principles and methods of green space system planning.In this paper,remote sensing(RS)and geographic information system(GIS)spatial analysis techniques are used,and multi-source data such as satellite image,automatic weather station hourly data,and land use planning data are introduced,combined with landscape pattern analysis,spatial measurement,and statistical analysis,The following contents are explored:(1)Research on multi-source data processing and integration methodThe multi-source data such as satellite image,green space system planning map,and near-ground climate data of automatic station are collected and processed by remote sensing interpretation,graphics vectorization,and spatial interpolation.On this basis,a data integration model of multi-source data is established by using model builder in Arc GIS,which provides a data basis for subsequent analysis.(2)Characteristics analysis on green space pattern and near-ground wind-temperature field of multi-scaleIn the administrative region,two horizontal pattern indexes of green space area and spatial location and three vertical structure indexes are selected to analyze the spatial pattern characteristics of green space;the wind temperature field characteristics are analyzed by the grade gradient partition of wind temperature field and the distribution frequency of wind speed/temperature value.In the central area,nine horizontal pattern indexes and three vertical structure indexes are used to analyze the spatial pattern characteristics of green space.Wind speed and temperature are selected to analyze the characteristics of the near-ground wind-temperature field.The results show that there are significant spatial differences between the green space spatial pattern and the near-ground wind-temperature field,and there are high/low-value clusters in different regions.(3)In the administrative region:the correlation between the spatial pattern of green space and the near-ground wind-temperature field in GuangzhouThe superposition of green space horizontal pattern map and wind-temperature field map shows that the ecological park with an area of more than 30 km~2 is consistent with the high-speed wind field and low-temperature area;The distribution of most green space is consistent with the low-temperature area,but the agricultural regional green space is associated with the high-temperature area.By dividing the 5km*5km grid and extracting index data,the results of correlation and regression analysis show that the proportion of arbor woodland in green space is positively correlated with wind speed and negatively correlated with temperature;the proportion of shrub woodland and grassland are positively correlated with wind speed and temperature.The proportion of arbor woodland and shrub woodland could explain 29.7%and 24.3%of winter temperature difference,12.9%and 12.8%of summer temperature difference,respectively,and the proportion of grassland could explain17.2%of winter temperature difference.(4)In the central area:the correlation between spatial pattern of green space and near-ground wind-temperature field in GuangzhouBy dividing 2km*2km grid,the correlation between horizontal/vertical pattern index of green space and mean value of wind/temperature was analyzed.The results showed that green space rate,maximum patch index,proximity,aggregation,patch number,patch density,shape index,the proportion of arbor woodland,shrub woodland,and grassland were significantly correlated with the wind-temperature field.The results of regression analysis showed that the per capita green area and the proportion of arbor and woodland could explain 11.9%and12.8%of the difference of wind speed in winter;the water area could explain 12.4%of the difference of wind speed in summer,and the average building height could explain 15.4%and24.5%of the difference of wind speed in winter and summer respectively;Greenspace coverage and aggregation degree can explain 12.0%and 13.0%of the temperature difference in summer,respectively;arbor woodland proportion and grassland proportion can explain21.1%and 11.0%of the temperature difference in summer and winter,respectively;shrub woodland proportion can explain 14.5%and 19.6%of the temperature difference in winter and summer,respectively.