The growth of five duckweed species (Lemna gibba L., Lemna minor L., Lemna trisulca L., Spirodela polyrhiza (I.) Schleiden and Wolffa arrhiza (L.) Hork. ex Wimm.). and one waterfern (Azolla filiculoides Lamk.) was assessed in laboratory scale experiments comparing domestic sewage and two types of artificial waste water with a standard mineral growth medium (1/10 Huttner). Across the three waste waters, organic loading varied between 300 and 442 mg l(-1) GOD, total Kjeldahl-N varied between 14 and 52 mg N l(-1) and total-P between 7 and 9 mg P l(-1). All species yielded less on the two artificial waste waters than on the mineral medium. In contrast, in domestic waste water Lemna gibba and Spirodela polyrhiza performed equally well compared td the mineral medium (RGR on Huttner 0.27 +/- 0.01 and 0.15 +/- 0.02, respectively). In a Subsequent 12 days batch experiment with Lemna gibba in presented domestic waste water, a detailed mass balance was established for the two macronutrients nitrogen (N) and phosphorus (P). The duckweed was,subjected to a harvesting regime: every fourth day the density was reduced to the initial 17 g dry weight m(-2). Total-P was reduced to 2.4 +/- 0.2 mg P l(-1) (77% removal) and Kjeldahl-N to 6.2 +/- 0.4 mg N l(-1) (94% removal). The P-balance accounted for 93% of the inputs and 56% was removed by the duckweed. Also periphyton contributed substantially to P removal (18%). For N, however, 50% of the inputs was unaccounted. Although a nitrification inhibitor was applied, it is concluded that this must have been ineffective, since most of. the N-loss occurred in the first part of the experiment, when pH was still too low (pH < 8) for substantial NH3-volatilization. Hence, nitrification/denitrification is held responsible for 50% of the nitrogen removal from the system, the-harvested duckweed accounted for 34%. Tn short, denitrification was the major pathway for N-removal and duckweed absorption for P-removal. (C) 1998 Elsevier Science Ltd. All rights reserved.
