Climate change has altered the phenology or seasonal sequencing of events in a variety of ecosystems, including Narragansett Bay, RI (USA). The timing and magnitude of the winter-spring bloom have changed dramatically over the last three decades, resulting in a 60% decline in water column chlorophyll a concentrations at a station in the mid-bay. This large decline has been linked to climate variability (North Atlantic Oscillation) and a long-term warming trend. I hypothesized that the decline in chlorophyll a, a proxy for primary production, has led to less organic matter deposition to the benthos and changes in sediment biogeochemical cycling and benthic-pelagic coupling. In order to test this hypothesis, I measured benthic metabolism and the net flux of N2 gas across the sediment water-interface at seven stations in Narragansett Bay. In addition, I conducted a series of experimental manipulations using sediment cores and nine large (4 m^2) benthic mesocosms.

A comparison between sediment oxygen demand and the fluxes of dissolved inorganic nitrogen (DIN) across the sediment-water interface from the 1970/80s and measurements from this study showed significant decreases in mid and upper Narragansett Bay. In the more eutrophic Providence River estuary, the decline in oxygen is less clear and, while nitrogen fluxes remain unchanged, dissolved inorganic phosphate (DIP) fluxes appear to have declined significantly. Three decades ago, approximately 45% of the net primary production was consumed in the benthos. In comparison, only about 30% is remineralized in the benthos today. In the 1970s, summer sediment nutrient regeneration supplied 50% to 200% of the N and P needed by phytoplankton. In 2005 and 2006, upper and mid-bay summer nutrient regeneration supplied less than 13% of the N and less than 5% of the phosphate phytoplankton demand.

Net sediment denitrification in mid-Narragansett Bay also decreased significantly from rates measured in the late 1970s. In the summer of 2006, high rates of net sediment nitrogen fixation (-5 to –650 micro-mol N2-N m^-2 h^-1) were measured at four sites. This is particularly remarkable since nitrogen fixation in marine sediments has previously been considered inconsequential. Over the course of the summer (June, July, and August), sediment nitrogen fixation added an additional 90 million moles of N to Narragansett Bay. In comparison, net sediment denitrification in the summer of 2005 removed only 26 million moles of N. Thus, over the last three decades, the sediments of Narragansett Bay switched from being a sink for nitrogen to being a source.

A series of experiments were conducted to study the effects of enriching the overlying water with nitrate, ammonium, or phosphate on net sediment N2 fluxes. In addition, separate experiments added organic matter to the sediment surface. Inorganic enrichment of the overlying water caused no change in the net sediment flux of N2. Organic matter enrichment first increased nitrogen fixation, but ultimately switched the sediments from being a net source to a net sink of nitrogen. A clear threshold was observed where nitrogen fixation took place when organic matter deposition fell below about 0.3 g C m^-2 d^-1. A conceptual model for the possible mechanisms regulating the net flux of N2 gas in estuarine sediments was also developed.