No abstract

No abstract

Natural aquatic systems undergo fluctuating redox conditions due to microbial activities, varying water saturation levels, and nutrient dynamics. With fluctuating redox conditions, trace metals can mobilize or sequester in response to changes in iron and sulfur speciation and the concentrations and lability of organic carbon. This study examined the effect of redox fluctuations on trace metal mobility in samples collected from two different natural aquatic systems: riparian wetlands and a stream. The wetland soils contained low sulfur and total Fe contents as compared to stream sediments. The mineral composition at both sites was dominated by quartz. We incubated water-saturated soils under three cycles of anoxic–oxic conditions (τanoxic/τoxic = 3) spanning 24 days and monitored the change in dissolved and bioavailable metal concentrations. For both natural systems, reduction of iron oxides under anoxic conditions caused Co and Zn releases. In contrast, oxidation of sulfides mobilized Cu under oxic conditions in both sites. In wetland soils, dissolution of Fe (hydr)­oxides increased Ni solubility; however, in stream sediments, Ni release occurred when sulfides or organic matter were oxidized. For stream sediments, each subsequent redox cycle increased the bioavailability of trace metals. Redox fluctuations in wetland soils increased bioavailable Zn and Cu and decreased bioavailable Ni and Co. This study illustrates that different trace metals display distinct bioavailability patterns during redox fluctuations in natural environments. The biogeochemical cycling of nutrients in systems with redox fluctuations may be influenced by these trace metal availability patterns in addition to the availability of electron donors and acceptors.

Although artificial intelligence (AI) such as machine learning (ML) and deep learning (DL) has been recognized as an emerging and promising tool, its application becomes challenging with incomplete data collection. Herein, in the absence of the influent phosphorus load and chemical dosage data for phosphorus removal, we employed ML/DL models to predict effluent phosphorus using nine-year data from a small-scale wastewater treatment plant. Attempts were made to select essential model input features from 42 variables by using Pearson correlation analysis to reveal internal correlations among variables. First, five ML regression models were used to predict the effluent phosphorus load, and a maximum coefficient of determination (R 2) of 0.637 was achieved with the support vector machine model. Then, the DL model named long short-term memory could predict phosphorus load in one-day advance with an R 2 value of 0.496. Finally, on the basis of the historical data, an anomaly alarm design was proposed to minimize the chance of exceeding the discharge permit and achieved a maximum accuracy of 79.7% to predict the phosphorus concentration after comparing seven ML classification models. This study provides an example of applying AI for process improvement and potential cost reduction with incomplete data sets.

No abstract