Walkability of greenways from the perspective of the elderly: A case study of Huangpu River waterfront greenway

• A walkability evaluation framework based on multi-sourced data for greenways was proposed, specifically from the viewpoint of older adults. • Street view images collected by wearable GoPro were utilized to supplement traditional GIS data in the evaluation. • It was found that older adults value safety facilities, green view index, and seating facilities most while walking on greenways. • The results could provide insights for developing elderly-friendly greenways. Under the background of global aging, outdoor public spaces play a crucial role in promoting physical activity among older individuals, helping to advance the process of healthy aging. Among these spaces, urban greenways are particularly effective in encouraging walking behaviors and improving both physical and mental health. However, current assessments of greenway’s walkability have mainly focused on the general population, with scant attention paid to the viewpoint of older adults. Hence, this study proposed a novel walkability assessment framework from the perspective of the elderly, which employs multi-sourced data in conjunction with Fully Convolutional Network (FCN) to measure walkability, taking the renowned 45-km-long Huangpu River waterfront Greenway in Shanghai as a case study. Apart from using traditional GIS data to quantify the walkability factors, locally collected street view images using wearable GoPro were also utilized in the evaluation. According to the results of overall walkability scores, the Hongkou Riverwalk shows the best walkability for older people, followed by the Huangpu Riverwalk, the Xuhui Riverwalk, the Yangpu Riverwalk, and the East Coast Riverwalk is the worst. It is also noted that among the walkability factors, older adults place the highest value on safety facilities, green view index and seating facilities. Moreover, four greenway patterns were classified to characterize different walkability, including “safety-dominant” pattern, “comfort-dominant” pattern, “interestingness-dominant” pattern, “usefulness-dominant” pattern. Last but not least, the implications and limitations were also discussed in this paper, some of which would be the direction of our future work.