Kyung Hwa Cho

Linking land-use type and stream water quality using spatial data of fecal indicator bacteria and heavy metals in the Yeongsan river basin

This study reveals land-use factors that explain stream water quality during wet and dry weather conditions in a large river basin using two different linear models-multiple linear regression (MLR) models and constrained least squares (CLS) models. Six land-use types and three topographical parameters (size, slope, and permeability) of the watershed were incorporated into the models as explanatory variables. The suggested models were then demonstrated using a digitized elevation map in conjunction with the land-use and the measured concentration data for Escherichia coli (EC), Enterococci bacteria (ENT), and six heavy metal species collected monthly during 2007-2008 at 50 monitoring sites in the Yeongsan Watershed, Korea. The results showed that the MLR models can be a powerful tool for predicting the average concentrations of pollutants in stream water (the Nash-Sutcliffe (NS) model efficiency coefficients ranged from 0.67 to 0.95). On the other hand, the CLS models, with moderately good prediction performance (the NS coefficients ranged 0.28-0.85), were more suitable for quantifying contributions of respective land-uses to the stream water quality. The CLS models suggested that industrial and urban land-uses are major contributors to the stream concentrations of EC and ENT, whereas agricultural, industrial, and mining areas were significant sources of many heavy metal species. In addition, the slope, size, and permeability of the watershed were found to be important factors determining the extent of the contribution from each land-use type to the stream water quality. The models proposed in this paper can be considered useful tools for developing land cover guidelines and for prioritizing locations for implementing management practices to maintain stream water quality standard in a large river basin.

Evaluating causes of trends in long-term dissolved reactive phosphorus loads to lake erie

Renewed harmful algal blooms and hypoxia in Lake Erie have drawn significant attention to phosphorus loads, particularly increased dissolved reactive phosphorus (DRP) from highly agricultural watersheds. We use the Soil and Water Assessment Tool (SWAT) to model DRP in the agriculture-dominated Sandusky watershed for 1970-2010 to explore potential reasons for the recent increased DRP load from Lake Erie watersheds. We demonstrate that recent increased storm events, interacting with changes in fertilizer application timing and rate, as well as management practices that increase soil stratification and phosphorus accumulation at the soil surface, appear to drive the increasing DRP trend after the mid-1990s. This study is the first long-term, detailed analysis of DRP load estimation using SWAT.

Meteorological effects on the levels of fecal indicator bacteria in an urban stream: a modeling approach.

Gwangju Creek (GJC) in Korea, which drains a highly urbanized watershed, has suffered from substantial fecal contamination, thereby limiting the beneficial use of the water in addition to threatening public health. In this study, to quantitatively estimate the sinks and sources of fecal indicator bacteria (FIB) in GJC under varying meteorological conditions, two FIB (i.e., Escherichia coli and enterococci bacteria) were monitored hourly for 24h periods during both wet and dry weather conditions at four sites along GJC, and the collected data was subsequently used to develop a spatiotemporal FIB prediction model. The monitoring data revealed that storm washoff and irradiational die-off by sunlight are the two key processes controlling FIB populations in wet and dry weather, respectively. FIB populations significantly increased during precipitation, with greater concentrations occurring at higher rainfall intensity. During dry weather, FIB populations decreased in the presence of sunlight in daytime but quickly recovered at nighttime due to continuous point-source inputs. In this way, the contributions of the key processes (i.e., irradiational die-off by sunlight, settling, storm washoff, and resuspension) to the FIB levels in GJC under different meteorological conditions were quantitatively estimated using the developed model. The modeling results showed that the die-off by sunlight is the major sink of FIB during the daytime in dry weather with a minor contribution from the settling process. During wet weather, storm washoff and resuspension are equally important processes that are responsible for the substantial increase of FIB populations.

The modified SWAT model for predicting fecal coliforms in the Wachusett Reservoir Watershed, USA

This study assessed fecal coliform contamination in the Wachusett Reservoir Watershed in Massachusetts, USA using Soil and Water Assessment Tool (SWAT) because bacteria are one of the major water quality parameters of concern. The bacteria subroutine in SWAT, considering in-stream bacteria die-off only, was modified in this study to include solar radiation-associated die-off and the contribution of wildlife. The result of sensitivity analysis demonstrates that solar radiation is one of the most significant fate factors of fecal coliform. A water temperature-associated function to represent the contribution of beaver activity in the watershed to fecal contamination improved prediction accuracy. The modified SWAT model provides an improved estimate of bacteria from the watershed. Our approach will be useful for simulating bacterial concentrations to provide predictive and reliable information of fecal contamination thus facilitating the implementation of effective watershed management.

Prediction of contamination potential of groundwater arsenic in Cambodia, Laos, and Thailand using artificial neural network

The arsenic (As) contamination of groundwater has increasingly been recognized as a major global issue of concern. As groundwater resources are one of most important freshwater sources for water supplies in Southeast Asian countries, it is important to investigate the spatial distribution of As contamination and evaluate the health risk of As for these countries. The detection of As contamination in groundwater resources, however, can create a substantial labor and cost burden for Southeast Asian countries. Therefore, modeling approaches for As concentration using conventional on-site measurement data can be an alternative to quantify the As contamination. The objective of this study is to evaluate the predictive performance of four different models; specifically, multiple linear regression (MLR), principal component regression (PCR), artificial neural network (ANN), and the combination of principal components and an artificial neural network (PC-ANN) in the prediction of As concentration, and to provide assessment tools for Southeast Asian countries including Cambodia, Laos, and Thailand. The modeling results show that the prediction accuracy of PC-ANN (Nash-Sutcliffe model efficiency coefficients: 0.98 (traning step) and 0.71 (validation step)) is superior among the four different models. This finding can be explained by the fact that the PC-ANN not only solves the problem of collinearity of input variables, but also reflects the presence of high variability in observed As concentrations. We expect that the model developed in this work can be used to predict As concentrations using conventional water quality data obtained from on-site measurements, and can further provide reliable and predictive information for public health management policies.

Contamination by arsenic and other trace elements of tube-well water along the Mekong River in Lao PDR

Arsenic and other trace element concentrations were determined for tube-well water collected in the Lao PDR provinces of Attapeu, Bolikhamxai, Champasak, Savannakhet, Saravane, and Vientiane. Water samples, especially from floodplain areas of central and southern Laos, were significantly contaminated not only with As, but with B, Ba, Mn, U, and Fe as well. Total As concentrations ranged from <0.5 μg L(-1) to 278 μg L(-1), with over half exceeding the WHO guideline of 10 μg L(-1). 46% of samples, notably, were dominated by As(III). Samples from Vientiane, further north, were all acceptable except on pH, which was below drinking water limits. A principal component analysis found associations between general water characteristics, As, and other trace elements. Causes of elevated As concentrations in Lao tube wells were considered similar to those in other Mekong River countries, particularly Cambodia and Vietnam, where young alluvial aquifers give rise to reducing conditions.

Development of early-warning protocol for predicting chlorophyll-a concentration using machine learning models in freshwater and estuarine reservoirs, Korea

Chlorophyll-a (Chl-a) is a direct indicator used to evaluate the ecological state of a waterbody, such as algal blooms that degrade the water quality in lakes, reservoirs and estuaries. In this study, artificial neural network (ANN) and support vector machine (SVM) were used to predict Chl-a concentration for the early warning in the Juam Reservoir and Yeongsan Reservoir, which are located in an upstream region (freshwater reservoir) and downstream region (estuarine reservoir), respectively. Weekly water quality data and meteorological data for a 7-year period were used to train and validate both the ANN and SVM models. The Latin-hypercube one-factor-at-a-time (LH-OAT) method and a pattern search algorithm were applied to perform sensitivity analyses for the input variables and to optimize the parameters of the two models, respectively. Results revealed that the two models well-reproduced the temporal variation of Chl-a based on the weekly input variables. In particular, the SVM model showed better performance than the ANN model, displaying a higher prediction accuracy in the validation step. The Williams-Kloot test and sensitivity analysis demonstrated that the SVM model was superior for predicting Chl-a in terms of prediction accuracy and description of the cause-and-effect relationship between Chl-a concentration and environmental variables in both the Juam Reservoir and Yeongsan Reservoir. Furthermore, a 7-day interval was determined as an efficient early warning interval in the two reservoirs. As such, this study suggested an effective early-warning prediction method for Chl-a concentration and improved the eutrophication management scheme for reservoirs.

Characteristics of wet and dry weather heavy metal discharges in the Yeongsan Watershed, Korea

A comprehensive water quality monitoring program was conducted in the Yeongsan (YS) River, Korea from 2005 to present to investigate wet and dry weather pollutant discharge in an attempt to establish point and non-point pollution management strategies. As part of this monitoring program, 11 heavy metal species were measured during dry and wet weather conditions in the YS River, where Gwangju City (GJ), a subcatchment of the YS River, was further monitored to clarify the responsibility of different metal species discharged into the mainstream. Monthly grab water samples showed that greater amounts of metals along the YS River were discharged during the wet summer months due largely to storm runoff. In addition, further monitoring results revealed that GJ, a highly urbanized area, was a significant contributor of the heavy metals being discharged into the YS River during both wet and dry weather. The most abundant metal species discharged from GJ were manganese, aluminum and iron with different contributions of wet and dry weather flows to the total discharge load. Wet weather flow was a significant contributor to the annual dissolved metal loads, accounting for 44-93% of the annual load depending on the metal species, with the exception of chromium and cadmium (9% and 27%, respectively). Mostly, metal loads during wet weather were shown to be proportional to the rainfall depth and antecedent dry period. A substantial fraction of metals were also associated with solids, suggesting that sedimentation might be an appropriate management practice for reducing the metal load generated in GJ. Overall, although dissolved metal concentrations in YS River were at an acceptable level for aquatic community protection, continual metal discharge throughout the year was considered to be a potential problem in the long-term due to gradual water quality degradation as well as continuous metal accumulation in the system.

Modeling transport of Escherichia coli in a creek during and after artificial high-flow events: Three-year study and analysis

Escherichia coli is the leading indicator of microbial contamination of natural waters, and so its in-stream fate and transport needs to be understood to eventually minimize surface water contamination by microorganisms. To better understand mechanisms of E. coli release and transport from soil sediment in a creek the artificial high-water flow events were created by releasing 60-80 m(3) of city water on a tarp-covered stream bank in four equal allotments in July 2008, 2009 and 2010. A conservative tracer difluorobenzoic acid (DFBA) was added to the released water in 2009 and 2010. Water flow rate, E. coli and DFBA concentrations as well as water turbidity were monitored with automated samplers at three in-stream weirs. A one-dimensional model was applied to simulate water flow, and E. coli and DFBA transport during these experiments. The Saint-Venant equations were used to calculate water depth and discharge while a stream solute transport model accounted for release of bacteria by shear stress from bottom sediments, advection-dispersion, and exchange with transient storage (TS). Reach-specific model parameters were estimated by evaluating observed time series of flow rates and concentrations of DFBA and E. coli at all three weir stations. Observed DFBA and E. coli breakthrough curves (BTC) exhibited long tails after the water pulse and tracer peaks had passed indicating that transient storage (TS) might be an important element of the in-stream transport process. Comparison of simulated and measured E. coli concentrations indicated that significant release of E. coli continued when water flow returned to the base level after the water pulse passed and bottom shear stress was small. The mechanism of bacteria continuing release from sediment could be the erosive boundary layer exchange enhanced by changes in biofilm properties by erosion and sloughing detachment.

Determination of the optimal parameters in regression models for the prediction of chlorophyll-a: A case study of the Yeongsan Reservoir, Korea

Statistical regression models involve linear equations, which often lead to significant prediction errors due to poor statistical stability and accuracy. This concern arises from multicollinearity in the models, which may drastically affect model performance in terms of a trade-off scenario for effective water resource management logistics. In this paper, we propose a new methodology for improving the statistical stability and accuracy of regression models, and then show how to cope with pitfalls in the models and determine optimal parameters with a decreased number of predictive variables. Here, a comparison of the predictive performance was made using four types of multiple linear regression (MLR) and principal component regression (PCR) models in the prediction of chlorophyll-a (chl-a) concentration in the Yeongsan (YS) Reservoir, Korea, an estuarine reservoir that historically suffers from high levels of nutrient input. During a 3-year water quality monitoring period, results showed that PCRs could be a compact solution for improving the accuracy of the models, as in each case MLR could not accurately produce reliable predictions due to a persistent collinearity problem. Furthermore, based on R(2) (goodness of fit) and F-overall number (confidence of regression), and the number of explanatory variables (R-F-N) curve, it was revealed that PCR-F(7) was the best model among the four regression models in predicting chl-a, having the fewest explanatory variables (seven) and the lowest uncertainty. Seven PCs were identified as significant variables, related to eight water quality parameters: pH, 5-day biochemical oxygen demand, total coliform, fecal indicator bacteria, chemical oxygen demand, ammonia-nitrogen, total nitrogen, and dissolved oxygen. Overall, the results not only demonstrated that the models employed successfully simulated chl-a in a reservoir in both the test and validation periods, but also suggested that the optimal parameters should cautiously be considered in the design of regression models.