Wingecarribee and Fitzroy Falls Reservoirs are part of the Shoalhaven system within the Sydney metropolitan water supply network and are both shallow, well mixed, polymictic water bodies. After construction in 1973, land use activities within the catchments changed from low intensity grazing to more intensive practices such as dairying, particularly on the northern foreshore of Wingecarribee Reservoir, and vegetable farming around Fitzroy Falls Reservoir. The change in land use practice raised concerns about the possible impact on water quality of the reservoirs. This study sought to differentiate between the influence of catchment effects, inter-reservoir transfers and seasonality to enable an assessment of long term trends in water quality to be made. Data collected on a monthly basis by the Water Board between October 1973 and December 1985 were assessed and representative or 'indicator' variables for the different influences were selected. Classification of the phytoplankton to create sample clusters followed by discriminant analysis of the abiotic data were performed to rank the 'indicator' variables The classification analyses indicated that time of sampling was more important than site differences in determining clustering and that both Wingecarribee and Fitzroy Falls Reservoirs were well mixed and similar. The influence of chronological changes following the initial filling of the reservoirs, seasonality and inter-reservoir transfers were detected in the clustering patterns. Changes attributable to land use activities within the catchments were not detected. The reduction in concentrations of variables from high values following the initial filling of the reservoirs was the main influence regulating clustering between 1974 to 1982 and the influence of seasonality on clustering was most apparent during 1976 and 1978. The assessment of the influence of inter-reservoir transfers was restricted because of limited data. Seasonal periodicities and ongoing changes in the phytoplankton and physical and chemical data over the study period were assessed as a means of interpreting and verifying the analysis results.