Urban Climate Monitoring Network Design: Existing Issues and a Cluster-based Solution

Dense sensor networks are being built to collect urban climate information in global cities, yet the performance of network design has rarely been accessed. Existing studies on urban climate network have three major issues: 1) focusing on summertime and lack of seasonal variation, 2) overlooking meteorological variables other than air temperature, 3) not incorporating the current network into the future design. In this study, we proposed a cluster-based design for urban climate monitoring network, and examined its potential applications in Beijing and Hong Kong by using weather simulation data as ground truth. Results show a robust design strategy is to train the cluster analysis with multiple meteorological variables that contain seasonal variations. Utilizing the cluster-based design strategy, we optimize the urban climate monitoring networks by rearranging sensor locations and expanding the network. Compared to the current network, the rearranged network has an improved performance by 22.7% in Beijing and by 10.7% in Hong Kong for the study period. With a sampling ratio of 6.3%, the expanded monitoring network has a mean bias of 0.58 °C and 0.44 °C for representing the air temperature variability across Beijing and Hong Kong, respectively. The proposed method is not sensitive to the cluster models, background climate of the city, and resolution of the weather simulation data. The design strategy is thus applicable for other cities and can provide useful guidance for the establishment of dense urban climate monitoring networks.