Abstract: A phytoplankton collection survey was conducted in a southern Chinese reservoir from January 2022 to April 2022 to investigate the dynamics of dominant cyanobacterial species and their drivers before and after cyanobacterial blooms. To facilitate water environment monitoring, three sampling sites were set up based on the water diversion and supply characteristics of the reservoir. The principal coordinates analysis (PCoA) was used to explore the dissimilarity of dominant cyanobacterial species before and after the bloom. Furthermore, the relationship between dominant cyanobacterial species and driving factors was analyzed using variation partitioning analysis (VPA) and linear fitting. The results showed that: (1) Three dominant genera were identified, namely Pseudanabaena, Cylindrospermopsis raciborskii, and Raphidiopsis curvata. (2) Pseudanabaena were the most dominant genus before the water bloom, accounting for 72.17% of the total phytoplankton. During the water bloom, all three dominant cyanobacteria reproduced rapidly, and their relative content was 33.06%, 30.71%, and 28.65%, respectively. (3) The VPA results and linear fits indicated that the relative water column stability (RWCS), water temperature (WT), and total phosphorus (TP) were the main factors affecting the cyanobacterial blooms in the reservoir. This study showed that the water diversion process of the reservoir had a significant impact on cyanobacteria, and the response characteristics and mechanism of dynamic water diversion on cyanobacterial blooms should be further studied.