Freshwater wetlands of the Florida Everglades are oligotrophic (water total phosphorus < 10 ug/L in unenriched marsh), with published values for average aboveground net primary productivity of sawgrass ranging from 255-606 g C m^-2 yr-¹ and periphyton from 17-10,371 g C m^-2 yr^-1. High temporal and spatial variability in these estimates, derived primarily by harvesting and small-scale gas exchange techniques, have been attributed to the underlying geology and hydrology of the landscape and phenology of dominant species. However, some of this variability has also been attributed to the variety of techniques used to measure production, and only recently have eddy covariance and whole ecosystem chamber methods been employed. Because accurate productivity estimates are key to understanding ecosystem response to changes in climate and water management in the Everglades, we are conducting a whole ecosystem chamber-based study of ecosystem productivity in conjunction with eddy covariance estimates of carbon flux at two sites of contrasting hydroperiod in Everglades National Park. Site TS1b is an infrequently flooded wet prairie site while Site SRS2 is a permanently flooded marsh; both are part of the Florida Coastal Everglades Long Term Ecological Research program that supplies ancillary data about the sites. We performed monthly measurements of CO₂ flux in 5 replicate plots using an LI-840 infrared gas analyzer connected to a clear, polycarbonate chamber. Measurements were taken every second for one minute in light and dark.
We found mean annual NEP values at TS1b (515 ± 183 g C m^-2 yr^-1) and SRS2 (1150 ± 155 g C m^-2 yr^-1) to be similar to previously published values calculated by harvesting, indicating an effect of increased water delivery on ecosystem productivity. However, temporal changes showed the opposite pattern where NEP at TS1b decreased significantly when inundated during the wet season (P < 0.01). NEP and ER at SRS did not show significant intraannual pattern. Sawgrass was found to be the major contributor to carbon sequestration at both TS and SRS. Together with estimates from associated eddy covariance towers, these CO₂-based estimates indicate a net flux of inorganic carbon to the marsh ecosystem. This is in contrast to NEP estimates from continuous dissolved oxygen measurements that have indicated net heterotrophy in the marsh ecosystem.