METHANOGENESIS IN THE SEDIMENTS OF AN AUSTRALIAN FRESH-WATER WETLAND - COMPARISON WITH AEROBIC DECAY, AND FACTORS CONTROLLING METHANOGENESIS

Sediment oxygen demands and water-atmosphere methane emissions of a hi ghly productive, freshwater wetland on the floodplain of the River Mur ray in south-eastern Australia were quantified over a 14 month period in 1993-1994. Total sediment oxygen demands ranged from 1.3 to 3.3 mmo l m(-2) h(-1), of which <3 to 90% was due to chemical oxygen demand. M ethane emissions ranged from <0.01 mmol m(-2) h(-1) in winter to 2.75 mmol m(-2) h(-1) in summer. Methanogenesis accounted for at least 60% of the combined aerobic and methanogenic carbon flux in sediments from Eleocharis sphacelata beds, and at least 30% and 40% of the combined flux in sediments from Myriophyllum sp. beds and Vallisneria gigantea beds, respectively. In vitro incubations, using additions of sulfate a nd of molybdate, failed to indicate unequivocally competition for subs trates between sulfate-reducing and methanogenic bacteria. However, in vitro methanogenesis was strongly inhibited by nitrate, suggesting an interaction between benthic methanogens and denitrifying or other nit rate-reducing bacteria. Fe3+ decreased in vitro methanogenesis by 16-4 9% during January, February and March 1994; oxidation of organic matte r at the expense of the reduction of ferric ions could be a significan t route for detritus processing in Eleocharis-bed sediments in the war mer months. Methanogenesis was increased consistently by additions of some low molecular weight substrates, such as acetate, but not by othe rs, such as methanol, propionate and trimethylamine. Complex polymeric substrates, such as cellulose, starch and aquatic plant matter, incre ased in vitro methanogenesis rapidly and markedly. Despite this, no re lationship between methane emissions and benthic cellulase activity wa s observed in the field. Methanogenesis was strongly temperature depen dent, being maximal at 30 to 40 degrees C and minimal at 5 degrees C, thereby explaining the strong seasonality observed in methane emission s in situ.