Abstract: Objective · To explore the spatial distribution and spatial-temporal clustering of coronavirus disease 2019 (COVID-19) in Jingzhou City. Methods · Data of COVID-19 cases in Jingzhou City from January 1 to March 12, 2020 were collected. Trend surface analysis, spatial autocorrelation and spatial-temporal scanning analysis were conducted to understand the spatial-temporal distribution of COVID-19 at town (street) level in Jingzhou City, and the spatial-temporal clustering characteristics of local cases and imported cases were compared. Results · Trend surface analysis showed that the incidence rate of COVID-19 in Jingzhou City was slightly “U” from west to east, slightly higher in the east, and inverted “U” from south to north, slightly higher in the south. Global autocorrelation showed that the incidence rate of COVID-19 in Jingzhou City was positively correlated (Moran's I=0.410, P=0.000). Local spatial autocorrelation analysis showed that the highly clustered areas and hot spot areas were mainly in Shashi District, Jingzhou District and the main urban area of Honghu City (Xindi Street) (P<0.05). Five clusters were found by spatial-temporal scanning of imported cases. The cluster time of the main cluster was from January 18 to February 3, 2020, and it was centered on Lianhe Street, covering 15 towns (streets) in Shashi District and Jingzhou District (LLR=174.944, RR=7.395, P=0.000). Five clusters were found by spatial-temporal scanning of local cases. The cluster time of the main cluster was from January 20 to February 24, 2020, which was located in Xindi Street, Honghu City (LLR=224.434, RR=16.133, P=0.000). Conclusion · Obvious spatial-temporal clustering of COVID-19 was found in Jingzhou City, and Shashi District, Jingzhou District and Honghu City were the most prevalent areas.

Key words: coronavirus disease 2019 (COVID-19), spatial autocorrelation, spatial-temporal scanning, trend surface analysis