In this paper seven of the ten Water Control Zones (WCZs) in Hong Kong’s coastal waters with monthly or bi-weekly sampling data of 17 parameters collected at 37 monitoring stations from 1988 to 1999 were selected to analyze the spatial and temporal variations of chlorophyll-a and its influencing factors. Cluster analysis was employed to group the monitoring stations based on the structure of the data set. Multiple step regression was employed to determine the significant influencing factors of chlorophyll-a level. The results suggest that all the monitoring stations could be grouped into two clusters. Cluster Ⅰ with frequent red tide incidents comprises two WCZs which are semi-enclosed bays. Cluster Ⅱ with less red tide occurrence comprises the other five WCZs in an estuarine environment in the west. For both clusters, the organic contents indicator, BOD5, was a common significant influencing factor of the chlorophyll-a level. Nitrogen and light penetration condition related to turbidity, total volatil In this paper seven of the ten Water Control Zones (WCZs) in Hong Kong’s coastal waters with monthly or bi-weekly sampling data of 17 parameters collected at 37 monitoring stations from 1988 to 1999 were selected to analyze the spatial and temporal variations of chlorophyll- a and its influencing factors. Cluster analysis was employed to group the monitoring stations based on the structure of the data set. Multiple step regression was employed to determine the significant influencing factors of chlorophyll-a level. Cluster Ⅱ with less red tide incidents comprise two WCZs which are semi-enclosed bays. Cluster Ⅱ with less red tide occurrence includes the other five WCZs in an estuarine environment in the west. For both clusters, the organic contents indicator, BOD5, was a common significant influencing factor of the chlorophyll-a level. Nitrogen and light penetration condition related to tur bidity, total volatil