Correlation among soil microorganisms, soil enzyme activities, and removal rates of pollutants in three constructed wetlands purifying micro-polluted river water

Three pilot horizontal subsurface flow constructed wetlands (CWs) were constructed to purify micro-polluted waters in Chongqing, P.R. China. The quantity of soil microorganisms (bacteria, fungi, and actinomyces) and enzyme activities (dehydrogenase, catalase, and urease) were monitored, and they together formed the soil biological indicators in the experimental CWs. All three CWs had acceptable removal abilities for pollutants and were adaptable to water quality fluctuations. The removal rates of pollutants, number of soil microbes, and activities of soil enzymes in planted wetland were higher than those in unplanted wetland. There were significant correlations between the removal of total nitrogen (TN-N) (p < 0.01) and temperature, as well as the number of soil microbes (p < 0.01) and activities of soil enzymes (p < 0.05). The I canonical variables can be used to represent the most information of the soil biological indicators in the three CWs. There were good linear relationships between the canonical variables of microbes and enzymes, and positive correlations could be observed in the planted CWs. The soil biological indicators in the planted CWs can be divided into two the same classes by the cluster analysis method, whereas there were two kinds of classification in the unplanted CW including two or three classes. Hydrophytes planted in the CWs can help make the relationships among the soil biological indicators closer. Soil biological activity can be improved more greatly by planting Arundo donax in summer and planting Acorus calamu in winter. There were negative correlations between the removal of ammonium (NH4+-N) and soil biological indicators in all the three CWs, as well as the permanganate index (COD) and total phosphorus (TP-P), but the soil biological indicators did not have obvious relationships with the removal of phosphorus or carbon in all CWs (p > 0.1). The relationship between the removal of TN-N and soil biological indicators was very significantly positive in the planted CWs (p < 0.05), and positive in the unplanted CW. Soil enzyme activity had a more significant relationship with the removal rate of TN-N than soil microorganism.