Development and Applications of a Predictive Model for Geosmin in North Han River, Korea

Prediction of Geosmin at Different Depths of Lake Using Machine Learning Techniques

Geosmin is a major concern in the management of water sources worldwide. Thus, we predicted concentration categories of geosmin at three different depths of lakes (i.e., surface, middle, and bottom), and analyzed relationships between geosmin concentration and factors such as phytoplankton abundance and environmental variables. Data were collected monthly from three major lakes (Uiam, Cheongpyeong, and Paldang lakes) in Korea from May 2014 to December 2015. Before predicting geosmin concentration, we categorized it into four groups based on the boxplot method, and multivariate adaptive regression splines, classification and regression trees, and random forest (RF) were applied to identify the most appropriate modelling to predict geosmin concentration. Overall, using environmental variables was more accurate than using phytoplankton abundance to predict the four categories of geosmin concentration based on AUC and accuracy in all three models as well as each layer. The RF model had the highest predictive power among the three SDMs. When predicting geosmin in the three water layers, the relative importance of environmental variables and phytoplankton abundance in the sensitivity analysis was different for each layer. Water temperature and abundance of Cyanophyceae were the most important factors for predicting geosmin concentration categories in the surface layer, whereas total abundance of phytoplankton exhibited relatively higher importance in the bottom layer.

Forecasting a 2-methylisoborneol outbreak in a brackish lake

2-Methylisoborneol (2-MIB) is the primary cause of the earthy and musty odor produced by cyanobacteria, which deteriorates the quality of fishery products and tap water. Despite the need for controlling outbreaks, few studies have been conducted on 2-MIB in brackish lakes, where capture fisheries are active. This study examined the association between water quality and the outbreak of 2-MIB in a brackish lake using statistical analysis of long-term monitoring data and developed forecasting models for 2-MIB outbreaks. We investigated Lake Ogawara, which is a brackish lake with a cool-temperate climate in Japan, where 2-MIB outbreaks frequently occur between August and December. Logistic regression analyses were performed using the outbreak or non-outbreak of 2-MIB (2-MIB(+ / -)) as the dependent variable and water quality parameters as the independent variables. The results suggested that the density of 2-MIB-producing cyanobacteria was higher when (1) dissolved inorganic nitrogen concentrations were low under the relaxation of phosphorus limitation and/or (2) salinity or micronutrient concentrations were high. In addition, we successfully developed forecasting models with a high predictive power that determined 2-MIB(+ / -) in August-December using only two water quality parameters: dissolved inorganic phosphate and pH in April and total nitrogen/total phosphorous and salinity in May.

Degradation of geosmin and 2-methylisoborneol in water with UV/chlorine: Influencing factors, reactive species, and possible pathways

Geosmin (GSM) and 2-methylisoborneol (2-MIB) are two common odor compounds in drinking water. In this paper, the performance of UV/chlorine was compared with that of chlorine and UV to degrade GSM (100 ng L^-1) and 2-MIB (100 ng L^-1) in water. UV/chlorine was further exploited, and a steady-state kinetic model was used to conduct a detailed study on efficiency, rate, reactive species and pathway. The results showed that UV/chlorine greatly could improve the removal ratio to 90% within 5 min, from approximately 20% with only UV or dark chlorine in 60 min. The removal ratio and rate depended on UV light intensity, free chlorine dosage, reaction time and water quality parameters (e.g. pH, concentrations of HCO₃^- and Cl^-). Among these factors, the first two obviously could accelerate the rate and increase the ratio. The degradation was also significantly improved in an acidic condition, while alkaline conditions and HCO₃^- had inhibitory effects, and Cl^- created no difference. Contributions of OH and Cl to the degradation of 2-MIB and GSM were further revealed, and OH was found to be the most important reactive species. In the UV/chlorine process, 6 degradation byproducts of 2-MIB, including 1 alcohol, 2 ketones, and 3 olefins, were identified, and 14 degradation byproducts of GSM, including 6 ketones, 1 aldehyde, 2 alcohols, 3 naphthenes, and 2 olefins, were found by using solid-phase microextraction coupled to gas chromatography-mass spectrometry. The possible degradation pathways of GSM and 2-MIB in UV/chlorine thus were proposed.

Predicting Taste and Odor Compounds in a Shallow Reservoir Using a Three–Dimensional Hydrodynamic Ecological Model

The objective of this study was to establish a technique to predict the occurrence of algal bloom and the algal-derived taste and odor compounds 2-methylisoborneol (2-MIB) and geosmin using a three-dimensional (3D) model that could reflect the complex physical properties of a shallow reservoir. Water quality, phytoplankton, and taste and odor compounds monitoring was conducted at the Jinyang Reservoir in 2016. In June, there was a potential for a high concentration of 2-MIB (maximum 80 ng/L) to occur owing to the appearance of Pseudanabaena sp.; additionally, from July to August, there was potential for a high concentration of geosmin (maximum 108 ng/L) to occur, because of the presence of Anabaena sp. A 3D hydrodynamic model was coupled with an ecological model to predict cyanobacteria bloom and the presence of taste and odor compounds. Cyanobacteria producing either 2-MIB or geosmin were distinguished to enhance the accuracy of the modeled predictions. The results showed that the simulations of taste and odor compounds spatial distribution and occurrence time were realistic; however, the concentration of geosmin was overestimated when Microcystis sp. was blooming. The model can be used as a management tool to predict the occurrence of algal taste and odor compounds in reservoir systems and to inform decision-making processes concerning dam operation and water treatment.