Long-term increase of phytoplankton biomass in Chesapeake Bay, 1950-1994

An analysis of historical and recent data on chlorophyll a for Chesapeake Bay reveals that a significant increase of phytoplankton biomass has occurred during the last 40 to 50 yr. Concentrations of chlorophyll a in the surface mixing layer have increased 5- to 10-fold in the seaward regions of the estuary and 1.5- to 2-fold elsewhere, paralleling published estimates of increased loading of N and P to the estuary since World War II. The characteristic high variability of freshwater flow that occurs on seasonal to interannual time scales, however, drives fluctuations of chlorophyll a that are superimposed on this apparent upward trend, potentially obscuring the effects of overenrichment on chlorophyll a concentrations in nutrient-limited regions of the Bay. To resolve a time trend of chlorophyll from this variability, we developed regional models of mean, monthly chlorophyll a using autoregressive moving average (ARMA) procedures. The models were developed with water quality data from monitoring cruises of the Chesapeake Bay Program spanning 1984 to 1992. The approach was to: (1) determine the relationship of prominent variables, including freshwater flow, salinity, temperature, region, and time of year, to chlorophyll a for the 'modern' Bay; (2) predict chlorophyll a for the 'historical' Bay based on these models and actual, observed variables for periods from 1950 to 1984 for which chlorophyll a data were available; (3) compare chlorophyll a predicted from modern relationships, to those observed in the past, using the residuals to identify deviations below or above expected concentrations that would suggest an effect on chlorophyll a unrelated to flow variability. The results show that: (1) observed and predicted chlorophyll a concentrations matched reasonably well for the years that were used to develop the models, with some exceptions when the models failed to capture extremely high concentrations of chlorophyll a during blooms; (2) chlorophyll a concentrations in the 1950s, 1960s, and 1970s were predominantly lower than predicted by the models, particularly in the mesohaline and polyhaline regions of the estuary that are most susceptible to nutrient limitation; (3) chlorophyll a concentrations were lower in the 1960s than in the 1970s, probably as the result of low flow and a concomitant reduction of nutrient loading in the 'dry' 1960s as compared to the 'wet' 1970s; (4) interannual variability was high for both observed and predicted chlorophyll a concentrations in the 1970s, and this variability was most strongly expressed in the mesohaline to polyhaline Bay, reflecting the spatial and temporal heterogeneity of phytoplankton that prevailed in that period. These findings support the hypothesis that a significant increase of chlorophyll a has occurred in the lower Bay that cannot be accounted for by variability of freshwater flow and attendant properties.