The Thresholds of Urban Parks’ Characteristics on Heat Mitigation Effect: Case Study of 120 Parks in Foshan, China

Abstract

In the context of the increasingly severe global thermal environmental crisis, this paper examines 120 urban parks in Foshan, a representative water-network city in China’s southern region. By using multi-methods integrating Boosted Regression Tree (BRT) modeling, Restricted Cubic Spline (RCS) threshold detection, and K-means clustering, the study explores cooling effectiveness of urban parks based on five metrics: Park Cooling Distance (PCD), Park Cooling Area (PCA), Park Cooling Efficiency (PCE), Park Cooling Intensity (PCI), and Park Cooling Gradient (PCG). Results reveal threshold-dependent cooling effects, with park area, blue-green ratio, and FAR as key determinant factors. Thresholds for optimal cooling efficiency occur at 2.31–4.03 km² park area, vegetation coverage exceeding NDVI 0.277, and blue-water ratio within 0.277–0.515. All the sampling parks could be clustered into four groups: high-efficiency, low-efficiency, medium-efficiency, and comprehensive cooling performance. The findings could provide reference to deeply understand the threshold mechanism in urban heat mitigation and evidence-based planning strategies for optimizing green infrastructure in water-network cities, supporting science-driven thermal environment optimization.