مقایسۀ تطبیقی رتبه‌بندی کارکردی کلان شهرهای ایران بر اساس داده‌های صفتی و رابطه‌ای

در مطالعه نظام‏های شهری دو رهیافت صفت‏مبنا و شبکه‏مبنا قابل‏تمایز است؛ رهیافت نخست صفات منتسب به شهرها را ناظر اصلی بر اهمیت نسبی آن‏ها می‏داند و معطوف به سطح تمرکز فعالیت‏ها یا کارکردها در نقاط شهری است؛درحالی‏که رهیافت دوم بر اساس تعاملات بین شهرها و با استفاده از داده‏های جریانی جایگاه هر شهر را در نسبت با موقعیت قرارگیری آن می‏سنجد. اگرچه بنیان‏های نظری هر دو رهیافت به‏خوبی مستند شده، از یک‏سو ماهیت ارتباط بین این دو تاکنون به‏طور وسیعی نامشخص باقی‏مانده و از سوی دیگر در مطالعه تعاملات بین نواحی تفاوت جغرافیایی انواع مختلفی از جریان‏ها کمتر موردتوجه بوده است. ازاین‏رو، هدف از پژوهش پیش‏رو سنجش و رتبه‏بندی جایگاه کلان‏شهرهای کشور مبتنی بر هر دو رهیافت صفتی و تعاملی با استفاده از داده‏های رابطه‏ای جریان هوایی و زمینی افراد، مقایسه میزان همبستگی نتایج به‏دست‏آمده از هر یک با یکدیگر، و تشریح شباهت‏ها و اختلافات ممکن بین این دو است. نتایج نشان داد نه‏تنها رتبه‏بندی کارکردی کلان‏شهرهای ایران مبتنی بر رهیافت تعاملی با استفاده از دو جریان هوایی و زمینی افراد اختلافات آشکاری نسبت به یکدیگر دارند، بلکه میزان پشتیبانی رتبه‏بندی کلان‏شهرها در رهیافت اندازه‏مبنا و شبکه‏مبنا نیز باهم متفاوت است؛ به‏طوری‏که ضریب تعیین رتبه‏بندی تعاملی مبتنی بر جریان هوایی افراد و رتبه‏بندی صفتی برابر 85/0 درصد به‏دست‏آمده که نشان‏دهنده پشتیبانی این دو از یکدیگر است؛درحالی‏که همبستگی نهایی دو رهیافت مبتنی بر جریان زمینی افراد (36/0) از ضعف چنین حمایتی حکایت دارد. این تفاوت موید آن است که تعاملات بین‏شهری که در قالب جریان هوایی افراد روی می‏دهد بیش از دیگر انواع جریان‏ها ناشی از کارکردهای درونی شهرهاست و می‏تواند سازمان‏یابی فضایی شهرها را در سطح کلان منعکس کند.

A Comparison of Ranking Functional Urban Regions of Iran Based on Interaction and Node Attribute Data

IntroductionThe positions of cities within an urban system have been studied on a variety of geographical scales ranging from the metropolitan and regional level to the national level. The two main approaches can be distinguished: attributebased and interactionbased. First, focuses on the concentration of activities or functions in a node and characterize cities’ importance by using data on the internal attributes of nodes such as the population size, economic profiles, and the presence of transport and communication functions. Second, one could rank cities from an interaction perspective by using flow data and concentrates on the degree to which nodes interact with each other in the system of flows. Although the fundamentals of both approaches are well documented, the nature of the relationship between these two approaches has remained hitherto largely unexplored. Even though many studies employ either interaction or nodeattribute data to study the positions of cities in the urban system, relatively little is known about the relationships between these two different types of data. This study aims to examine the extent to which the positions of cities using the interaction and node attribute data correlate with each other, and how possible (dis)similarities between the two can be explained. MethodologyAlthough there are several types of flow that could be used for studying interaction, we have concentrated on flows of people travelling between distinct metropolitan areas for two reasons. First, face to face relationships continue to be important for the development of urban systems. Second, it is the less frequent journeys undertaken over greater spatial distances rather than daily (commuting) journeys that are pertinent to the development of urban systems on the higher spatial scale. The analysis has been conducted separately for aerial and terrestrial journeys. In this study, we have employed data on longdistance mobility which has been collected in Comprehensive Transportation Studies of Iran in 2016 .As reported in this study, the origin destination survey was carried out in 56 study areas which are according to political divisions. In this survey, a longdistance journey was defined as a journey to a destination more than 100 km away. With respect to node attributes, information on sociodemographic and economic was obtained from the Population and Housing Census data collected by Statistical Center of Iran (2016).According to the enactment of Ministry of Roads and Urban Development, each city which has more than 500 thousand population is known as metropolitan; based on, in 2016 Iran has had 15 metropolitan, But in this study, the metropolitan areas were operationalized via the concept of functional urban regions (FURs) to represent the spatial units that are functionally interrelated in economic terms, because these can be compared with one another more easily. However, the delimitation of such areas is constrained by the availability of data in at least two respects. First, the functional interdependencies should ideally be derived from interaction data such as daily commuter flows. Second, because the flow data is only available for 56 defined study area of CTSI, the metropolitan areas necessarily is constrained to these areas. Results and discussionThe results show that not only ranking of cities by interaction data differ for types of flow but also the relationships between interaction and node attributes differ for these types of flow. Tehran, Mashhad, Esfahan and Ahvaz there are in the highest orders based on aerial flow data while Tehran, Esfahan, Qom and Arak are the most important metropolitan areas based on aerial flow data terrestrial flow data. In addition, Tehran, Mashhad, Esfahan and Kerman has acquired the first to forth position respectively based on attributebased data. This division indicates that nodes do not necessarily hold an important position on both aspects simultaneously. The difference between transportation modes in representation of rankings of cities by using interactionbased data _aerial and terrestrial_ has been approved: The choosing destinations in terrestrial transportation from each origin gravitate to nearer distances and the number of passengers is affected by distances more intensely which is known as “Distance Decay” factor. We also find that the differences between the two rankings can be explained to some extent by the fact that corporeal interaction is influenced by the “physical barriers” which means that the top ranking metropolitan areas are those located centrally in our study area, Iran such as Tehran, Esfahan and Qom. ConclusionIn this study, we have considered to what extent the rankings of metropolitan areas using interaction and nodeattribute data are correlated. Data on longdistance passenger mobility for aerial and terrestrial journeys and the attributes of the metropolitan areas have been used to generate the rankings of 15 metropolitan areas.The results shows that node attributes data tend to overestimate the importance of metropolitan areas that are not situated on central area of Iran like Kerman and Shiraz. These metropolitan areas function as central nodes in their regional economies and hold high positions on the economic attributes, but may have weaker relationships with other metropolitan areas. This contrast suggests that the physical barriers imposed by distance play a part in limiting the interaction between metropolitan areas as far as corporeal travel is concerned. Oppositely, the results shows that interaction data based on terrestrial flow tend to underestimate the importance of metropolitan areas that are not situated on central area of Iran like Zahedan and Orumia. This result may suggest that none of node attribute data or flow data are not sufficient to explain the positions of metropolitan areas on the two overall rankings, at least for the current data. Nevertheless, compared with terrestrial flows, aerial flows and nodeattributes are more strongly correlated. Aerialinteraction and nodebased data show a correlation factor of 0.85 suggesting that they are good proxies for one another. However, since different types of flow tend to have different characteristics, terrestrialinteraction and nodebased data show a correlation factor of 0.36. It can be concluded that ranking of urban regions by means of nodebased attributes can be better explained by aerial flow data than terrestrial flow data.KeywordsUrban system, nodeattribute approach, interaction approach, metropolitan areas of Iran, comparative analysis