Objective In order to explore the unique demand of the elderly for urban park green space and solve existing problems in this regard under the background of increasing population aging and uneven distribution of urban park green spaces, this research not only analyzes the spatio-temporal response relationship between the distribution of elderly population and the accessibility of park green space, but also explores the planning strategies for park green space in age-friendly cities.

Methods According to the data about aged population and urban park green space (above 1 hm²) in Handan City in 2010 and 2020, this research implements a relatively comprehensive analysis of the spatio-temporal change of population aging in each sub-district within the research area by introducing the composite aging index (CAI). Specifically, the research firstly allocates the elderly population within each sub-district in proportion to the number of households in each community concerned. Then the research analyzes the accessibility of park green space for the elderly under various travel modes in different years using the information obtained through the open street map (OSM) road network, the Amap bus route planning interface and the improved Gaussian two-step floating catchment area (GA2SFCA) method in combination with such factors as time cost and proportion of the elderly. After that, the research combines the changes in population aging and in the accessibility of park green space for the elderly to analyze the spatio-temporal characteristics thereof.

Results The results show that during the period from 2010 to 2020: 1) The population aging in the research area increased obviously with the decline of birth rate and mortality rate and the migration of young adults; all the sub-districts involved in the research witnessed an increasing aging index except for the Heping Sub-district, and the number of sub-districts with high-level aging increased significantly on a diffusion trend in space. 2) Despite the increase in the total amount of green space in community parks, the per capita supply of green space for the elderly remained at a low level in the research area, as the newly-created green space in community parks distributed mainly in sub-districts with low-level aging, thus failing to alleviate the green space pressure of sub-districts with high-level aging; by the end of 2020, there were still 81.98% of the community parks having accessibility levels below the mean level with a standard deviation of as high as 12.22 from the perspective of the elderly, showing that the gap of green space in community parks was still very obvious in the research area; although the average accessibility of comprehensive parks for the elderly increased greatly, the long travel time would reduce the elderly's travel intention because new comprehensive parks were typically far away from their residence areas; besides, the change in the accessibility of green space in comprehensive parks failed to match the spatial change of aging. The reason is that highly accessible areas of green space in comprehensive parks shifted from the center to the northwest, while the sub-districts with high-level aging or fast-growing aging population were mainly concentrated in the southeast of the research area. 3) The change in the comprehensive accessibility of park green space for the elderly shows that sub-districts with a significant improvement in comprehensive accessibility were mainly concentrated at the edge of the research area with less serious aging, indicating that these sub-districts varied greatly from those with relatively severe exacerbation of CAI in terms of spatial distribution; the supply level of park green space within the sub-districts in the southeast of the research area is in urgent need of improvement.

Conclusion Based on the above research results, the research finds that the accessibility of all kinds of park green spaces improved to different degrees during the period from 2010 to 2020 in the research area, but the space-related supply and demand relationship between park green space and the elderly population was not improved. The research puts forward the following optimization strategies based on the above research results : 1) Increase the supply of community park green space in the area with high-level aging, and take into account different age groups rather than just the elderly when planning park green space; 2) for areas without sufficient space for expansion of park green space, improve the utilization rate of existing park green space through multifunctional space design, and accordingly enhance the carrying capacity of existing green space; 3) make park green space more convenient for the elderly by improving transportation network and increasing the number of park entrances to densely-populated areas; 4) encourage the elderly to migrate to areas with adequate park green space through favorable policies; 5) give more consideration to the elderly during the formulation of planning scheme for green space under the background of increasingly severe population aging, so as to give full play to the leading role of planning in urban construction, especially in terms of green space. This research especially introduces the CAI for analysis of the accessibility of park green space for the elderly, which can provide a reference for subsequent researches on the accessibility of urban park green space targeting specific groups, especially in terms of the improvement of research method.