Sediment organic matter properties facilitate understanding nitrogen transformation potentials in East African lakes

East African lakes include the most productive and alkaline lake group in the world. Yet, they generally receive fewer nutrient inputs than the densely populated subtropical and temperate lakes in the northern hemisphere. In these lakes with insufficient supplies of inorganic nitrogen, the mineralization of benthic organic matter can play an important role in driving the nutrient cycle and nitrogen loss. Using a suite of stable 15N isotope dilution and tracer techniques, we examined five main processes of the sediment nitrogen cycle in 16 lakes and reservoirs of Tanzania and Kenya, East Africa: gross nitrogen mineralization, ammonium immobilization, dissimilatory nitrate reduction to ammonium (DNRA), and the dinitrogen (N2) production via denitrification and anaerobic ammonium oxidation (anammox). Gross nitrogen mineralization and ammonium immobilization showed the maximum values of 9.84 and 12.39 μmol N kg-1 h-1, respectively. Potential DNRA rates ranged from 0.22 to 8.15 μmol N kg-1 h-1 and accounted for 10 %-74 % (average 25 %) of the total dissimilatory nitrate reduction. Potential nitrate reduction rates in most lakes were dominated by denitrification with a contribution of 26 %-85 % and a mean of 65 %. We further found that the sediment nitrogen transformations were driven mainly by benthic organic matter properties and water column phosphate concentrations, reflecting microbial metabolic responses to the changing carbon and nutrients availability. For instance, autochthonous production of protein-like organic matter attributed to active sediment nitrogen mineralization, DNRA, and denitrification. In contrast, the high degree of humification caused by the inputs of terrestrial humic-like substances slowed down the sediment nitrogen transformations. The contribution of DNRA to total dissimilatory nitrate reduction was significantly positively correlated to sediment C: N ratios. These results indicate that predictions of sediment N supply and loss in East African lakes can be improved by incorporating sediment organic matter properties.