Abstract:Cylindrospermopsis raciborskii is a bloom-forming cyanobacterium that has attracted global attention following Microcystis and is currently expanding rapidly in southern China. To investigate the distribution patterns and driving factors of C. raciborskii in this area, a field survey was conducted during the dry season in 100 reservoirs across seven river basins of Guangdong Province. The abundance of C. raciborskii was quantified using Real-time quantitative PCR targeting the rpoC1 gene. The results showed that C. raciborskii was detected in all collected samples, with abundances ranging from 4.98×10? to 4.23×10? cells/L. The abundance varied by up to four orders of magnitude among reservoirs, indicating substantial spatial variability in population size. Principal component analysis revealed that spatial variation in environmental factors among reservoirs across river basins in Guangdong Province was primarily driven by nitrogen, phosphorus, and water temperature. At the basin scale, the mean abundance of C. raciborskii in the Pearl River Delta and the western Guangdong coastal river basins was significantly higher than that in the Xijiang, Hanjiang, and Beijiang basins. These two basins were also the most severely affected by C. raciborskii blooms, with heavy blooms occurring in 10 reservoirs. According to the Australian risk classification system for cyanobacterial blooms based on C. raciborskii abundance, 53% of the surveyed reservoirs were categorized as low risk, whereas 23% were classified as Alert Level 2. The binary classification machine learning models of Extreme Gradient Boosting (XGBoost) and Random Forest (RF) together with multiple linear stepwise regression analysis indicated a significant positive relationship between the C. raciborskii abundance and trophic state index (TSI). These results suggest that eutrophication level is the primary driver of the spatial variation in C. raciborskii abundance among reservoirs in Guangdong Province.