Evaluating the mismatch between urban greenery and cycling

This paper proposes a measure to comprehensively evaluate the mismatch between urban greenery and cycling in Chicago using multi-source data (e.g., Google Street View images, road network, and ridership records of bike sharing) and multiple analytical methods (e.g., AI, network analysis, and GIS). The non-linear relationships between the mismatch and urban environment factors were further explored by employing a machine learning model. The key findings revealed that: 1) road segments presented significant mismatch indicating that road segments with abundant urban greenery were not in the shortest cycling routes in Chicago; and 2) road segments in the city center areas and main streets presented a significant positive mismatch, while branch streets and streets in the outer areas showed mild positive/negative mismatch; 3) road segments with inappropriate ratios of urban environmental variables may have significant mismatch between urban greenery and cycling. This study provides a new angle for evaluating whether urban greenery can meet the demands of urban residents through active mobility. The findings offer strategic guidance for urban designers and decision-makers to implement practical measures for enhancing sustainable urban development.