Research On The Relationship Between Urban Ecological Environment And Human Activities Based On Multi-source Data

Rapid economic development will promote the deepening of urbanization,but extensive development methods will only bring serious damage to the development of cities.At present,the country pays more and more attention to the quality of urban ecological environment,and is committed to developing ecologically friendly civilization cities.At the same time,human activities play a vital role in the quality of the urban ecological environment.With the improvement of sensor technology,the multi-source data ubiquitous in the Internet has been fully mined,which has played an important role in promoting the study of urban ecological environment.Combining the travel modes of human activities,fully mining multi-source data,and analyzing the relationship between urban ecological environment and human activities are important topics in urban ecological environment planning.At present,there are few related studies,especially using machine learning algorithms to analyze There are even fewer studies to analyze the relationship between urban ecological environment and human activities.The main research contents include:(1)Based on the analysis of previous studies,it is proposed to use POI,OSM road network data,and remote sensing image data to construct human activity indicators(walkable measure index for residential areas,street vitality index,urban function mix index)Method of urban ecological environment index(remote sensing ecological index).By calculating the OD cost distance from various facilities to the residential area,combined with the distance attenuation law and the walking correction index to obtain the walkable measure index of the residential area.Construct street-related vitality indicators and obtain street vitality index through entropy weight method.A 500 * 500 meter grid is established,and the number and type of POI points in each grid are calculated in combination with information entropy,so as to obtain the urban function mixing index.Take the above three as indicators of human activity.The RSEI model is used to construct the remote sensing ecological index,which is used as the urban ecological environment index.(2)Using machine learning algorithms(LR,PLR,RFR,XGBR,SVR)and coupled models to model urban ecological environment index and human activity index.The conclusions are as follows:(1)The walkability measure index(WMI)in the central area of Nanchang is significantly better than the WMI in the fringe areas of Nanchang.The overall residential area WMI develops in a ring shape from the middle to the surrounding area.In addition,the WMI of the central urban areas of Nanchang,Anyi and Jinxian counties is also relatively high.The street vitality index(SVI)of the first ring road in Nanchang is the highest;the SVI of the second ring road area is beginning to weaken;the SVI of the third ring road is generally low.The overall SVI is unevenly distributed in space,showing a trend of higher in the eastern part of the Ganjiang River and lower in the western part of the Ganjiang River.Nanchang's urban functional mixing index(UFMI)is best within the second ring,and some areas between the second and third rings perform better.In most regions outside the Third Ring Road,the UFMI is too singular.The urban eco-environment index of Nanchang City is lower at the center of the city,and gradually increases to the periphery.Among them,the vegetation coverage and humidity in the central region are relatively small and the building index is large,resulting in a low remote sensing ecological index;the surrounding counties have a large number of forest lands and wetlands,which leads to remote sensing ecological index preference;Xinjian County and Jinxian County The main reason for the low remote sensing ecological index in some areas is that there is too much bare soil,which leads to an increase in the dryness component and a decrease in the humidity and greenness components.(2)There is a negative correlation between the indicators of human activities and the urban ecological environment index,the stronger the human activities,the more vulnerable the ecological environment of Nanchang.Among them,the strongest negative correlation is the urban ecological environment and UFMI,and the weakest is the urban ecological environment and MWI.By comparing five regression models,the best regression effect on the relationship between urban ecological environment and human activities is the XGBR,followed by the PLR.The areas with weaker prediction results are the northern part of the upper part of Xinjian County,the northern part of Nanchang County,and Wanli District.The predicted values are greater than the true values,indicating that the true ecological environment quality is worse than predicted;The predicted values of Shengmi,Houtian Township,Qianfang Town,Qili Township,and Zhaobu Township of Jinxian County are less than the true values,indicating that the ecological environment index in these areas isbetter than predicted.The regions with better prediction results are Anyi County,southern Xinjian County,East Lake District,West Lake District,Qingyunpu District,southern Nanchang County,and eastern and southern Jinxian County,indicating that the ecological environment index of these areas is closely related to human activities.(3)From the analysis of coupling coordination,proper human activities can maintain a good degree of coupling with the ecological environment.When the intensity of human activities is large or small,the degree of coupling will be reduced.In addition to the strong coupling degree in the main urban area,the centers of Anyi County,Jinxian County,and Nanchang County also show moderate coupling;in the more remote suburbs,the coupling degree is relatively low.At the same time,the regions with stronger human activities have a higher degree of coupling and coordination,and vice versa.In addition,19 streets in Nanchang's downtown area are lagging in the urban ecological environment,and other peripheral areas are lagging in human activities,which is consistent with the aggregate human activities in Nanchang.