Abstract: A comprehensive evaluation index system is of significant importance for the planning, construction, evaluation and management of green cities. The formulation of region differentiated index systems is fundamental for promoting the development of green cities in accordance with the local conditions, and is crucial for the Chinese cities with substantial economic, lifestyle and environmental differences. Firstly, we classified cities with three perspectives, including urban production development, living status, and environmental background. Next, a self-organizing feature map (SOFM) neural network was constructed to classify 288 cities into 26 types. Then, the SMART principle was applied to build a green city index system with three first-level indicators, including green production, green lifestyle, and green environment, 11 secondary indicators, and 70 third-level indicators. The third-level indicators were further sub-classified into basic and characteristic indicators. Finally, we provided the detailed applicable indicator classification for each of the 288 selected cities. The index system described here fully considers the construction foundation and development needs of different cities, and provided new ideas and methods for the establishment of green city index system, with both theoretical innovation and practical value.

Key words: climate change, radial growth, Picea jezoensis var. komarovii, Changbai Mountain., Larix olgensis