THE CARBON BALANCE OF GRAZED AND NONGRAZED SPARTINA-ANGLICA SALT-MARSHES AT SKALLINGEN, DENMARK

1 Canopy-level CO2 exchange and biomass were measured monthly for 9 co nsecutive months during 1990 on grazed and non-grazed portions of a Da nish salt marsh dominated by Spartina anglica. The empirical measureme nts were used to calibrate a model of canopy photosynthesis and soil r espiration that was subsequently integrated to provide estimates of an nual net ecosystem production (NEP). 2 Standing total and green biomas s on the non-grazed site reached maxima during September of 1.9 +/- 0. 2 and 1.1 +/- 0.2 kg m(-2) (+/- 1 SD, n = 3), respectively. Correspond ing maxima (0.8 +/- 0.1 and 0.5 +/- 0.05 kg m(-2)), were achieved duri ng July at the grazed site. 3 Sediment organic matter as a percentage of dry weight was 31.6% and 5% at the non-grazed and grazed sites, res pectively. Accounting for the higher sediment bulk density at the graz ed site meant that concentrations of organic carbon were 0.027 and 0.0 21 g cm(-3) at non-grazed and grazed sites, respectively. Thus, the fr action of NEP removed by grazing animals significantly reduced the acc umulation of sediment organic matter. 4 Estimate of NEP varied between 17 and 38 mol m(-2) year(-1) of carbon at the grazed site and between 5 and 38 mol m(-2) year(-1) at the non-grazed site, depending on assu mptions about canopy geometry. Soil respiration, 36 mol m(-2) year(-1) , did not differ between sites. Canopy respiration was less (21 mol m( -2) year(-1)) at the grazed site than at the non-grazed site (28 mol m (-2) year(-1)), and accounted for 22-29% and 28-41% of gross photosynt hesis at grazed and non-grazed sites, respectively. The decrease in ca nopy respiration compensated the grazed canopy for losses to herbivore s. 5 A canopy photosynthesis model was used to calculate a theoretical envelope of canopy biomass, termed the production window, inside of w hich NEP is positive and outside of which NEP is negative. The product ion window is bisected by the optimum canopy biomass, which is the the oretical biomass at which NEP is maximized. The seasonality of NEP is determined by the trajectory of actual canopy biomass through the prod uction window. It was shown that canopy biomass on the grazed site was close to the theoretical optimum biomass late in the season, while th e non-grazed canopy biomass was supraoptimal.