A chlorophyll time series for Narragansett Bay: Assessment of the potential effect of tidal phase on measurement

An 18-yr chlorophyll time series for Narragansett Bay based on weekly sampl es collected without regard to tidal phase revealed a long-term decrease in mean annual levels. The potential influence of neglecting tidal phase in t he sampling strategy on measured chlorophyll and its apparent long-term dec rease is evaluated. A two year data set (1995-1996) is used as a proxy for the 1973-1990 time series together with an observed relationship between co ntinuous measurements of in situ chlorophyll fluorescence and accompanying tidal phase. The deviations in chlorophyll from long-term means relative to deviations from mean low water at the time of sample collection are also a nalyzed, as is the potential influence of tidally-induced advective increas es or dilution on measured chlorophyll levels. The analyses, which compare the magnitude and trends in tidally adjusted and directly measured chloroph yll, indicate that semi-diurnal intratidal variations in chlorophyll had li ttle apparent effect on the long-term and seasonal patterns and trends dedu ced from the chlorophyll measurements. Neither tidal advection of the chlor ophyll gradient, nor bloom magnitude appear to compromise application of th e model. The 18-yr decline in annual mean chlorophyll observed between 1973 -1990 in Narragansett Bay is considered to be a bonafide portrayal of actua l events, and not an artifact of failure to consider tidal phase in the wee kly sampling strategy. The results also suggest that intratidal variability in chlorophyll does not seriously confound its meaningful measurement and usefulness as a representative index of phytoplankton. abundance at the per manent monitoring station established for Narragansett Bay. Nonetheless, th ere is need to refine and to inco orate temporal sampling strategies more c losely attuned to the tempo of growth, grazing, and nutrient recycling whic h accompany estuarine phytoplankton dynamics.