Limnological conditions and the occurrence of taste-and-odor compounds were studied in two reservoirs in Spartanburg County, South Carolina, from May 2006 to June 2009. Lake William C. Bowen and Municipal Reservoir #1 are relatively shallow, meso-eutrophic, warm monomictic, cascading impoundments on the South Pacolet River. Overall, water-quality conditions and phytoplankton community assemblages were similar between the two reservoirs but differed seasonally. Median dissolved geosmin concentrations in the reservoirs ranged from 0.004 to 0.006 microgram per liter. Annual maximum dissolved geosmin concentrations tended to occur between March and May. In this study, peak dissolved geosmin production occurred in April and May 2008, ranging from 0.050 to 0.100 microgram per liter at the deeper reservoir sites. Peak dissolved geosmin production was not concurrent with maximum cyanobacterial biovolumes, which tended to occur in the summer (July to August), but was concurrent with a peak in the fraction of genera with known geosmin-producing strains in the cyanobacteria group. Nonetheless, annual maximum cyanobacterial biovolumes rarely resulted in cyanobacteria dominance of the phytoplankton community. In both reservoirs, elevated dissolved geosmin concentrations were correlated to environmental factors indicative of unstratified conditions and reduced algal productivity, but not to nutrient concentrations or ratios. With respect to potential geosmin sources, elevated geosmin concentrations were correlated to greater fractions of genera with known geosmin-producing strains in the cyanobacteria group and to biovolumes of a specific geosmin-producing cyanobacteria genus (Oscillatoria), but not to actinomycetes concentrations. Conversely, environmental factors that correlated with elevated cyanobacterial biovolumes were indicative of stable water columns (stratified conditions), warm water temperatures, reduced nitrogen concentrations, longer residence times, and high phosphorus concentrations in the hypolimnion. Biovolumes of Cylindrospermopsis, Planktolyngbya, Synechococcus, Synechocystis, and Aphanizomenon correlated with the greater cyanobacteria biovolumes and were the dominant taxa in the cyanobacteria group. Related environmental variables were selected as input into multiple logistic regression models to evaluate the likelihood that geosmin concentrations could exceed the threshold level for human detection. In Lake William C. Bowen, the likelihood that dissolved geosmin concentrations exceeded the human detection threshold was estimated by greater mixing zone depths and differences in the 30-day prior moving window averages of overflow and flowthrough at Lake Bowen dam and by lower total nitrogen concentrations. At the R.B. Simms Water Treatment Plant, the likelihood that total geosmin concentrations in the raw water exceeded the human detection threshold was estimated by greater outflow from Reservoir #1 and lower concentrations of dissolved inorganic nitrogen. Overall, both models indicated greater likelihood that geosmin could exceed the human detection threshold during periods of lower nitrogen concentrations and greater water movement in the reservoirs.