以分形理论为基础,借助于地理信息系统软件,对南京市区交通网络进行了长度-半径维数和分枝维数的测算,结果表明南京市区的交通网络具有分形特征,长度-半径维数为1.574,而分枝维数为1.3934,两者之间存在一定差异。通过进一步对主城区各行政区交通网络的盒子维数的测算,揭示出各行政区的交通网络也具有分形特征,分维值介于1.3568与1.4991之间,其大小与经济发展状况并没有很好的对应关系。如何以经济状况为基础数据指导城市各行政区的交通发展,以及两者之间是否在其他区域尺度上存在相关关系等,仍是需要有更多研究基础才能深入探讨的问题。

It is very important to research the fractal characteristics of the distribution of transportation network. At present the researches in this field mainly focus on the regional road network and have not attached enough importance to the urban transportation network. There are several different calculation methods of fractal dimension value. Length-radius dimension and dendrite-radius dimension of transportation network of Nanjing city are calculated with the help of the software ArcGis in this paper. Xinjiekou, the transportation hinge of Nanjing city is chosen as the calculation center for fractal characteristics analysis, and the radius value changes between 1 to 9km. The results show that the transportation network of the city has fractal characteristics. The value of length-radius dimension is 1.574. The fractal dimension value of dendrite-radius is 1.3934. It shows that the connectivity of the road of Nanjing city should be strengthened. The box dimension of different districts of the main city has also been calculated at the scale of 50 to 500m in this research and the results show that the value is between 1.3568 and 1.4991. The respective value of each district is 1.4991(Xiaguan),1.4902 (Gulou), 1.4401(Qinhuai), 1.4393(Jianye), 1.4129(Baixia)and 1.3568(Xuanwu). The fractal dimension of transportation network of these districts does not have very good relationship with economic index. It can be concluded that besides urban transportation network, there are many other factors that influence the economic development of the districts, such as the existence of water region and upland, which influence the road construction.How to plan the transportation network of the districts with the present economic condition is still a difficult problem. It has also been brought forward that the accuracy of the fractal dimension value is relative because the order of the road has not been considered. How to assign the weight value to the roads with different orders is very important for future research. There is still no answer to the questions such as in which scale the fractal dimension of the transportation network does not have good relationship with regional economy index, and whether they have good relationship in other regional scale. More researches should be done to get the answer to the relative questions.