Alexandrium catenella (formerly A. tamarense Group 1, or A. fundyense) is the leading cause of Paralytic Shellfish Poisoning in North and South America, Europe, Africa, Australia and Asia. The quantification of A.catenella via sxtA, a gene involved in Paralytic Shellfish Toxin synthesis, may be a promising approach, but has not been evaluated in situ on blooms of A. catenella, in which cell abundances may vary from not detectable to in the order of 106 cells L-1. In this study, we compared sxtA assay performance to a qPCR assay targeted to a species-specific region of ribosomal DNA (rDNA) and an established fluorescent in situ hybridization (FISH) microscopy method. Passing-Bablok regression analyses revealed the sxtA assay to overestimate abundances when <5 cell equivalents A. catenella DNA were analysed, but otherwise was closer to microscopy estimates than the rDNA assay, which overestimated abundance across the full range of concentrations analysed, indicative of a copy number difference between the bloom population and a culture used for assay calibration a priori. In contrast, the sxtA assay performed more consistently, indicating less copy number variation. The sxtA assay was generally reliable, fast and effective in quantifying A. catenella and was predictive of PST contamination of shellfish.