The effect of nutrient additions on the partitioning of nutrients in an experimental coastal Mediterranean system

The hypothesis that the importance of dissolved organic matter (DOM) as a reservoir of C, N, and P declines, relative to that of the particulate pool, with increasing nutrient inputs was tested using mesocosms exposed to a gradient of nutrient inputs in the Spanish Mediterranean. The nutrient additions included a treatment equivalent to the loading in the coastal ecosystem studied (5 mmol N m^–2 d^–1), and mesocosms receiving half , 2-, 4-, 8-, and 16-fold this value, as well as a mesocosm to which no nutrients were added. Nutrients were added at ratios of 20 N (as ammonium) : 7 Si : 1 P. The initial concentration of dissolved inorganic nutrients was very low (dissolved inorganic nitrogen < 0.05 M, phosphate < 0.01) and comprised, together with the particulate pool < 25% of the total N and P in the system, with the bulk N and P in the system present as DOM (> 75%). Particulate and dissolved organic matter was depleted in N (C/N ratio > 15) and, particularly, P (C/P ratio > 1000), indicative of a strongly nutrient, particularly phosphorus, deficient ecosystem. Experimental nutrient additions lead to a parabolic change in C/N and C/P ratios in the dissolved organic matter with increasing nutrient inputs, which approached the Redfield stoichiometry at nutrient inputs > 8 fold above the ambient loading. The relative size of the dissolved inorganic nutrient pools (about 20% of the N and P) did not vary, but there was a tendency towards an increase in the relative size of the particulate pool at the expense of a decrease in the relative importance of DOM as a reservoir of N, P and C, with increasing nutrient inputs. The production of nutrient-depleted organic matter at low nutrient inputs likely prevents efficient recycling, leading to the dominance of nutrients in DOM in the system.