Objective The pattern evolution of public green space is deeply influenced by the way of urban growth, thus producing positive and negative performance in several dimensions, which is much more complicated in mountainous areas. A layout tailored to local conditions can positively adapt to urban patterns, effectively coordinate artificial and natural systems, and provide sustainable ecological well-being for humanity. If there is a lack of in-depth understanding of specific environmental characteristics, the layout of public green spaces is prone to functional aphasia and negative adaptation. Coordinating the landscape connectivity and resident accessibility of public green space is a challenge to urban development in mountainous towns, also a guarantee for dynamically balancing the development needs of “people” and “ecology”. In terms of landscape connectivity, current research mainly focuses on regional scales such as habitat protection, hydrological connectivity assessment, and environmental suitability assessment. However, there is still a gap in research on smaller scales such as urban public green space, and exploring the evolution laws and importance of landscape connectivity of different public green spaces can reveal their role and positioning in maintaining the overall ecological pattern of cities and towns. Besides, the research on green space accessibility mostly focuses on large cities in plains, with insufficient attention paid to the accessibility issues caused by the native terrain and spatial growth mode of mountainous towns. How to form a public green space layout that balances ecological connectivity and resident accessibility in the process of urban spatial growth under mountainous conditions is an urgent problem to be solved.

Methods Taking Jingning County in Lishui City, Zhejiang Province as an example, this research uses the Markov Model to deduce the change of land use structure and the process of urban spatial growth from 2007 to 2021. By summarizing and analyzing the distribution location, green space form, and surrounding environment, the research identifies the unique types of public green spaces under this construction method. On this basis, in combination with the morphological spatial pattern analysis method and the connectivity index method, the research obtains the landscape connectivity of green space from 2012 to 2022. Moreover, the research also uses spatial syntax to obtain the accessibility of residents from 2012 to 2022. From the perspective of coordinating ecological connectivity with resident accessibility, the evolution law of public green space pattern in response to urban spatial growth is revealed, providing a theoretical basis for the positive “adaptive” layout of public green spaces.

Results The research identifies the spatial growth mode of the built area in Jingning County that is featured by axial extension and finger-shaped expansion, as well as three unique types of public green spaces formed during this growth process, namely narrow green belt, sporadic regular green space, and irregular mountainous green space. The pattern evolution of public green space in Jingning County has resulted in three typical negative performance: 1) An increase in ecological space area without significant improvement in structural connectivity, and a decrease in functional connectivity; 2) a mismatch between the spatial distribution of green space and the “core” of accessibility (highly accessible areas); 3) poor local accessibility of linear green space.

Conclusion Further research has found that at the level of landscape connectivity, the landscape connectivity of public green spaces is not related to the size of ecological space area, but is significantly and positively correlated with the proportion of their core ecological areas. The accessibility level of public green spaces is related to their own pattern characteristics and spatial distribution. With the construction and layout of public green spaces in Jingning and the improvement of surrounding road network, the phenomenon of “misalignment” between the highly accessible areas of urban roads and the layout of public green spaces has been alleviated to some extent. This research finally proposes several optimization strategies for public green spaces in mountainous towns that are under axial and finger-shaped spatial growth. At the level of landscape connectivity optimization, the research proposes to expand the core area within public green spaces; at the level of accessibility optimization, the research proposes a planning strategy for the highly accessible core layout of roads in nearby areas and the division of key construction sections for linear green spaces; additionally, the research also proposes the method for differentiated application of the aforesaid strategies in new and old urban areas, with a view to promoting the positive adaptation of public green spaces to the axial and finger-shaped spatial growth mode, and improving the ecosystem service functions of public green space while maintaining the characteristics of mountainous towns. This research reveals the positive and negative adaptation of public green spaces to urban spatial growth from the perspective of coordinating ecological connectivity with resident accessibility. As the mutual adaptation between subjects is influenced by local economic and cultural characteristics, future research may consider adding other response dimensions.