Study On Spatio-Temporal Heterogeneity Of Urban Population Agglomeration

Cities can be viewed as organisms produced and formed in the practice of human beings,containing many interrelated elements.The citizens are the masters of their city,they shaped their city.At the same time,the complexity of the city complicates the behavior patterns of the people who live in it.Most urban geographical and social sciences describe cities as dynamic systems,the main characteristic of which is their changeability.Changes occur at everywhere and every day in the city,as well as the population migration.In this sense,far from being stable,the population are constantly changing,and a dynamic oriented study on agglomeration and variation of population is needed in order to understand the evolution of city.The diurnal agglomeration of population is the most basic,most important,and most frequent temporal-spatial pattern of population in the city.In this respect,the diurnal variation of population agglomeration in city is the focus of study in this paper.On the basis of existing research,this paper puts forward a method of spatialization of urban population from multi-source data,and reveals the difference of urban diurnal population agglomeration and the motivation behind。The data results of diurnal urban population agglomeration are very valuable for analyzing the urban problems,which are spatialization of urban photovoltaic energy self-sufficiency rate,and the correlation between urban population agglomeration and urban factors and urban space vitality.The following work has been carried out:(1)The daytime and nighttime population were identified:(i)12:00 in working day was selected as the representative time for daytime population;(ii)0:00 was selected as the representative time for nighttime population.We believed that the population at a certain time did not refer to the population at that time on a specific day,but should be the mathematical expectation of the population at that time for many days.Time,population structure and space were associated and grouped with mobile phone signals,and each grouping was called a scene.This paper put forward a population spatialization model based on the scene of "time-behavior-space"(TBS model).The scene was the basic unit of model,and it was from the urban operation mode.each scene had a clear population,time(day and night),space and data perception.For example,during the working day,children between the age of 3 and 6should be in kindergarten,and those without mobile phone signal,while people who work in the metropolitan area could be sensed by the mobile phone signal.Each scene had a corresponding population spatialization model.All the scenes formed the agglomeration of the urban population.(2)Based on the TBS model,the daytime and nighttime population agglomeration data sets of 100-meter grid in the central urban area of Wuhan were calculated.Then,we carried out data analysis from the point of view of frequency distribution characteristics,spatial distribution characteristics,statistical characteristics,change differences and so on.Through analysis,a new urban powerlaw distribution was found,the relationship between population density and distribution frequency accorded with the Clark distribution,and the fitting measurement coefficient R2 for daytime and nighttime "density-frequency(Area)" functions were 0.96693 and 0.93593.The results also showed that there was a great difference in the spatial distribution of urban population at daytime and at nighttime in the central urban area of Wuhan.Not only the range of population distribution at daytime was wider,but also the spatial gradient change rate of population density in daytime was smoother than at nighttime.The distribution of population in the central urban area at daytime and at nighttime was coupled,indicated that the current urban development of Wuhan was in a state of equilibrium in terms of employment and residence.the urban central area was not a pure central business region,but also mixed with many residential functions,the citizens in urban central area were mainly migrates internally.This pattern had been influenced by the historical evolution oof urban structure in China.(3)A method for analyzing the spatial distribution of population in frequency domain was proposed.The population distribution grids were viewed as a discrete image,the analysis of image was carried out from the point of view of frequency domain from Fourier transform,and the spatial change of population agglomeration was analyzed through the characteristics of energy distribution and conjugation.The results showed that this method could clearly reflect the spatial change gradient of population distribution,and using the conjugation of Fourier transform and the orthogonality of geographical space,this method could effectively reveal the main direction of spatial change of population distribution.The analysis results revealed that the population distribution in the central urban area of Wuhan like a landscape cross.(4)Five spatial patterns for the diurnal agglomeration heterogeneity of urban population were put forward,they were called Axis Pattern,Sandwich Pattern,Degradation Pattern,Spiced salt Pattern and Single Pattern.At present,the axis pattern was a spatial model with the greatest difference between diurnal population agglomeration in the central urban area of Wuhan,which is far greater than that caused by population migration such as working and going out.(5)From the analysis of the urban driving force of population agglomeration,the correlation between the spatio-temporal difference of population agglomeration and urban spatial form,and the spatial expression of urban energy structure which is closely related to urban operation,this paper carried out the urban operation-driven analysis of the spatio-temporal difference of population agglomeration.In the research on the urban driving force of population agglomeration,a total of 17 detection factors were constructed according to the two categories of resources,environment and production,life.The results showed that urban factors had more influence on population agglomeration at daytime than at nighttime,and the basic necessities were still the most important factors in determining population distribution.The data also verify the conclusions of existing studies: the impact of production and life factors on population agglomeration was greater than that of resources and environment factors on population agglomeration.However,the analysis of this paper also found that the influence of some resources and environment factors was very close to or even slightly greater than that of production and life factors;In the interpretation of population agglomeration to urban space: a comprehensive evaluation index system of urban external space vitality was put forward and built.The geographical detection results showed that the explanation of urban external spatial vitality by night population agglomeration was better than that by daytime population agglomeration,and the explanation of urban dynamic space by multi-temporal population agglomeration data was better than that of single daytime or nighttime population agglomeration data.This provides theoretical and data support for the study of urban landscape pattern selection of population data,and has theoretical and guiding significance;In the spatial expression of urban energy structure: Based on the near linear relationship between population and the allometric growth of urban energy consumption,a spatial expression model of urban photovoltaic energy structure was proposed.Based on the data of urban buildings and remote sensing images,two basic parameters of photovoltaic energy utilization in the central urban area of Wuhan were determined by sampling method,the roof photovoltaic availability coefficient of residential buildings(fr.pv)was 0.45,the roof photovoltaic availability coefficient of non-residential buildings(fr.pv)was 0.71.The spatialization of photovoltaic energy self-sufficiency rate in the central urban area of Wuhan was completed.Based on the result,the influence of photoelectric conversion rate on the area distribution and peak value of urban energy self-sufficiency rate was calculated by random analysis method.The results showed that when photovoltaic conversion rate increased by 5%,the peak spatial area of self-sufficiency rate increased by 20% on average.This meant that photovoltaic conversion rate had a very important impact on urban photovoltaic energy utilization.How to perceive the spatio-temporal heterogeneity of urban population agglomeration,especially the diurnal variation of population agglomeration in city,has always been a hot and difficult point in geography,and the related research is still rare,the research related to the spatio-temporal heterogeneity of urban population agglomeration has not been launched yet.Aiming at this problem,this paper tries to explore the diurnal urban population spatialization,by studying the situation and law of urban population agglomeration at the daytime and nighttime,the relationship with the city,and exploring the application of spatio-temporal heterogeneity of urban population agglomeration to solve urban problems.it provides a theoretical basis for the study of urban deductive mechanism,and provides methods and data support for urban land spatial layout optimization,efficient allocation of urban resources and energy,and fine urban governance.