RESUMEN
To better understand the controls on the opportunistic human pathogen Vibrio vulnificus in warm tropical waters, we conducted a year-long investigation in the Ala Wai Canal, a channelized estuary in Honolulu, HI. The abundance of V. vulnificus, as determined by quantitative PCR (qPCR) of the hemolysin gene (vvhA), varied spatially and temporally by nearly 4 orders of magnitude (≤3 to 14,000 mL-1). Unlike in temperate and subtropical systems, temperatures were persistently warm (19 to 31°C) and explained little of the variability in V. vulnificus abundance. Salinity (1 to 36 ppt) had a significant, but nonlinear, relationship with V. vulnificus abundance with the highest vvhA concentrations (>2,500 mL-1) observed only at salinities from 7 to 22 ppt. V. vulnificus abundances were lower on average during the summer dry season, when waters were warmer but more saline. The highest canal-wide average abundances were observed during a time of modest rainfall, when moderate salinities and elevated concentrations of reduced nitrogen species and silica suggested a groundwater influence. Parallel quantification of the vcgC gene suggested that C-type strains, which are responsible for most human infections, comprised 25% of the total V. vulnificus on average, but their relative contribution was greater at higher salinities, suggesting a broader salinity tolerance. Generalized regression models suggested that up to 67% of sample-to-sample variation (n = 202) in log-transformed V. vulnificus abundance was explained using the measured environmental variables, and up to 97% of the monthly variation in canal-wide average concentrations (n = 13) was explained with the best subset of four variables. IMPORTANCE Our data illustrate that, in the absence of strong seasonal variation in water temperature in the tropics, variation in salinity driven by rainfall becomes a primary controlling variable on V. vulnificus abundance. There is thus a tendency for a rainfall-driven seasonal cycle in V. vulnificus abundance which is inverted from the temperature-driven seasonal cycle at higher latitudes. However, stochasticity in rainfall and its nonlinear, indirect effects on V. vulnificus concentration means that high abundances can occur at any location in the canal at any time of year, making it challenging to predict concentrations of this pathogen at a high temporal or spatial resolution. Much of the variability in canal-wide average concentrations, on the other hand, was explained by a few variables that reflect the magnitude of freshwater input to the system, suggesting that relative risk of exposure to this pathogen could be predicted as an average for the system.