This study investigated the operation of three full-scale Lemna System surface fl ow municipal wastewater treatment plants, built according to the Lemna Corporation design. These plants consist of two ponds, the first aerated and the second for duckweed, with a barrier grid in the latter to ensure uniform plant distribution across its area. According to designers duckweed improves the efficiency of wastewater treatment. The three treatment plants are situated in central Poland and they differ in the occurrence of duckweed, two of them, located in Raków and Bąkowiec, operate without duckweed. and the third in Falęcin Stary, Lemna minor covers ca. 90% of second pond surface. The efficiency of Lemna System wastewater treatment was found not to differ between the plants with and without duckweed. The aerated pond played the main role in reduction of pollutants in the investigated Lemna Systems
In this study, the aim was to model the toxic effect of copper (Cu) and analyse the removal of Cu in aqueous Saharan and non-Saharan mediums by Lemna minor. Two separate test groups were formed: with Saharan dust (S) and without Saharan dust (WS). These test groups were exposed to 3 different Cu concentrations (0.05, 0.50 and 5.00 ppm). Time, concentration, and group-dependent removal efﬁ ciencies were compared using the non-parametric Mann-Whitney U test and statistically signiﬁ cant differences were found. The optimum removal values were tested at the highest concentration 79.6% in the S medium and observed on the 4th day for all test groups. The lowest removal value (16%) was observed at 0.50 ppm on the 1st day in the WS medium. When the S medium and WS medium were compared, in all test groups Cu was removed more successfully in the S medium than the WS medium contaminated by Cu in 3 different concentrations of (0.05 ppm, 0.50 ppm, 5.00 ppm). The regression analysis was also tested for all prediction models. Different models were performed and it was found that cubic models show the highest predicted values (R2). The R2 values of the estimation models were found to be at the interval of 0.939–0.991 in the WS medium and 0.995–1.000 in the S medium.