• Korean Journal of Agricultural Science
• /
• v.49 no.2
• /
• pp.193-202
• /
• 2022

The recurrent neural network (RNN) algorithm has been widely used in water-related research areas, such as water level predictions and water quality predictions, due to its excellent time series learning capabilities. However, studies on water quality predictions using RNN algorithms are limited because of the scarcity of water quality data. Therefore, most previous studies related to water quality predictions were based on monthly predictions. In this study, the quality of the water in a reservoir in Nonsan, Chungcheongnam-do Republic of Korea was predicted using the RNN-LSTM algorithm. The study was conducted after constructing data that could then be, linearly interpolated as daily data. In this study, we attempt to predict the water quality on the 7th, 15th, 30th, 45th and 60th days instead of making daily predictions of water quality factors. For daily predictions, linear interpolated daily water quality data and daily weather data (rainfall, average temperature, and average wind speed) were used. The results of predicting water quality concentrations (chemical oxygen demand [COD], dissolved oxygen [DO], suspended solid [SS], total nitrogen [T-N], total phosphorus [TP]) through the LSTM algorithm indicated that the predictive value was high on the 7th and 15th days. In the 30th day predictions, the COD and DO items showed R² that exceeded 0.6 at all points, whereas the SS, T-N, and T-P items showed differences depending on the factor being assessed. In the 45th day predictions, it was found that the accuracy of all water quality predictions except for the DO item was sharply lowered.

• Journal of Korean Society of Water and Wastewater
• /
• v.14 no.2
• /
• pp.174-180
• /
• 2000

In this study two water quality models, widely used in Korea, WASP5 and SWRRB were applied to Lake Paldang. The simulated results were compared with the measured data. The simulation results using WASP5 showed that this model could reasonably predict the concentrations of $NO_3$-N, Organic N, and Organic P. In order to investigate the effect of pollution by non-point source SWRRB was simulated and the concentrations of nutrients were predicted. The results from WASP5 and SWRRB are not directly comparable because their input data are different and output values are differently presented. Therefore, if these two simulation models were applied simultaneously, many valuable data and information could be obtained due to their own applicabilities and advantages.

• Journal of Korean Society on Water Environment
• /
• v.26 no.2
• /
• pp.200-207
• /
• 2010

The sewage treatment plants to be built to improve the water quality of the Mankyung River will total 11, of which combined capacity will reach $39,850m^3/day$, and saying in detail, 5 at Gunsan city, 2 at Iksan city, 1 at Kimje city and 3 at Wanju gun, The scenario for water quality improvement was developed, considering the conditions of plant operation ratio and the accomplishment of the water quality target (BOD 4.4 mg/L, T-P 0.356 mg/L) at the end of the watershed of Mankyung B was predicted, making use of QUAL2E model. As a result of prediction using QUAL2E model based on scenarios with 70% and 100% of operation ratio, respectively, at 11 plants in 2010, the water quality at the watershed of Mankyung B was estimated at 4.322 mg/L which was lower than the target of BOD 4.4 mg/L, indicating the target water quality was achieved, when it comes to 70% of operation ratio, But in case of T-P, it was estimated at 0.565 mg/L, which was higher than the target. When it comes to 100% of operation ratio, T-P also was 0.563 mg/L which exceeded the target, 0.356 mg/L. As indicated above, the effect of water quality improvement appeared very insignificant, which was attributable to the limit of small scale sewage treatment plant in total reduction capacity. Hence, the measures for additional reduction in a bid to achieve the target water quality of T-P at the designated location need to be taken, and the measures to build the Sewage treatment facilities at the place where the pollution is significantly caused by T-P appeared to be required as well.

• Journal of Korean Society on Water Environment
• /
• v.25 no.2
• /
• pp.311-321
• /
• 2009

Long-Term Hydrologic Impact Assessment (L-THIA) was modified to improve runoff and pollutant load prediction for Korean watersheds with changes in land use classification and event mean concentration produced from observed data in Korea. The L-THIA model was linked with SCE-UA, which is one of the global optimization techniques, to automatically calibrate direct runoff. Modified L-THIA model was applied to Gumho River Basins to analyze spatial distribution of nonpoint source pollution. The results of model calibration during 1991~2000 and validation during 1981~1990 for direct runoff represented high model efficiency of 0.76 for calibration and 0.86 for validation. As a results of spatial analysis of nonpoint source pollution, the BOD was mainly loaded from urban area but SS, TN, and TP from agricultural area which is mainly located along the stream. Modified L-THIA model improve its accuracy with minimum imput data and application efforts. From this study, we can find out the L-THIA model is very useful tool to predict direct runoff and pollutant loads from the watershed and spatial analysis of nonpoint source pollution.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1271-1271
• /
• 2001

Water pollutants in drainage mainly consist of organic compounds. Hence, total organic carbon (TOC), chemical oxygen demand (COD), and biochemical oxygen demand (BOD) were generally used as the indices of pollution. However, these values are determined by special analyzer (TOC), titration method (COD), or microbe culture (BOD). Therefore, the development of simple and easy methods for the determination of water pollution is required. The authors reported the evaluation of water pollution by near infrared (NIR) spectroscopy in a model system with food components (Takamura et al. (200) Near Infrared Spectroscopy: Proceedings of 9th International Conference, pp. 503-507). In this study, the relationship between NIR spectra and drainage was investigated in order to develop a method for evaluation of drainage by NIR. Drainage was obtained in Nara Purification Center. The ranges of TOC, COD, and BOD were 0-130, 0-100 and 0-200, respectively. NIR transmittance spectra were recorded on NIR Systems Model 6250 Research Composition Analyzer in the wavelength range of 680-1235 and 1100-2500 nm with a quartz cell (light path: 0.5, 1, 2, 4 and 10mm) at 10-40. Statistical analysis was performed using NSAS program. A partial least squares (PLS) regression analysis was used for calibration. As the result, a good correlation between the raw NIR spectra and OC was obtained in the calibration. The best light path was 10 and 0.5mm in the wavelength range of 680-1235 and 110-2500nm, respectively. In the calibration, correlation coefficients(R) were 096-0.97 in the both range. In the prediction, however, a good correlation (R=0.89-0.96) was obtained only in the range of 6801235 nm, Similar results were obtained in the cases of COD and BOD. These results suggest the possibility that NIR spectroscopy can be used to evaluate drainage.

• Journal of Korean Society of Rural Planning
• /
• v.7 no.1 s.13
• /
• pp.27-36
• /
• 2001

Estimation of current and future loading from watershed is necessary for the sound management of water quality of an estuary lake. Pollution sources of point and non-point source pollution were surveyed and Identified for the Koheung watershed. Unit factor method was used to estimate potential pollutant load from the watershed of current conditions. Flow rate and water qualify of base flow and storm-runoff were monitored in the main streams of the watershed. Estimation of runoff pollutant loading from the watershed into the lake in current conditions was conducted by GWLF model after calibration using observed data. Prospective pollutant loading from the reclaimed paddy fields under cultivation conditions was estimated using the modified CREAMS model. As a result, changes of pollutant loading into estuary lake according to non-cultivation and cultivation conditions of reclaimed tidal land were estimated.

• Journal of Korean Society of Water and Wastewater
• /
• v.25 no.5
• /
• pp.667-679
• /
• 2011

In this study, the influences of pollutant from Dae-po Stream and So-gam Stream located at the downstream of Nak-dong River on the water quality at Mul-geum water intake station were analyzed using RAMS model. Field measurements of velocity by ADCP, and water quality distribution of BOD and TP by water sampling were carried out to present the input and verification data for numerical simulations. The comparison between RAM2 and ADCP measurement, which aimed for the analysis of 2-D velocity distribution around Mul-geum water intake station showed that two results matched well along the spanwise direction. The prediction of pollutant concentration by RAM4 agreed fairly well with the measured data except for the points nearby right banks in the vicinity of tributary pollutant source. Flushing effect by the increase of mainstream discharge in Nak-dong River was analyzed to provide the damage mitigation in preparation for the accidental water pollution. With increasing mainstream discharge, high velocity and increased water quantity induced increasing dilution effect, thereby decreasing the inflow pollutant concentration rapidly.

• Proceedings of the Korean Society of Soil and Groundwater Environment Conference
• /
• 2003.09a
• /
• pp.422-425
• /
• 2003

This study was carried out to apprehend the variation of quality of mine drainage in the abandoned Samtan coal mine. After closure of coal mine, although still pumping, water level in underground was raised to loom and the concentration of some elements such as Fe and Mn was elevated. At present, the worst pollution source in this area is too the acidic leachate drained from uncovered mine waste impoundment. The flow rate of mine drainage from the adit is ave. about 20,000t/d. If water were flooded and deteriorated due to stopping pumping, the impact of the mine drainage on the stream around the abandoned mine would be more severe. Therefore, It is considered that the prediction of water quality of mine drainage from the adit after stopping pumping will be very important with a view to establishing countermeasures.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 1999.10c
• /
• pp.708-714
• /
• 1999

For the assessment of the level of stream pollution, SWAT-WASP model linked with GIS was applied to a respresentative rural watershed and evaluated for its applicability through calibration and verfication using observed data. Using daily water yields, sediment yields and nutrient discharge simulated by SWAT model, WASP input file was build. Point source pollutant and water quality change in stream was considered in WASP model. For the model applicatiion , digital maps were constructed for watershed boundary, ladn-use , soil series , digital elevation, and topographic data of Bok-Ha watershed using GRASS. The model application results showed that the simulated runoff was in a good agreement with the observed data and indicated reasonable applicability of the model.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.29 no.1
• /
• pp.24-35
• /
• 2023

Rapid industrialization and urbanization have led to severe marine pollution. A Water Quality Index (WQI) has been developed to allow the effective management of marine pollution. However, the WQI suffers from problems with loss of information due to the complex calculations involved, changes in standards, calculation errors by practitioners, and statistical errors. Consequently, research on the use of artificial intelligence techniques to predict the marine and coastal WQI is being conducted both locally and internationally. In this study, six techniques (RF, XGBoost, KNN, Ext, SVM, and LR) were studied using marine environmental measurement data (2000-2020) to determine the most appropriate artificial intelligence technique to estimate the WOI of five ecoregions in the Korean seas. Our results show that the random forest method offers the best performance as compared to the other methods studied. The residual analysis of the WQI predicted score and actual score using the random forest method shows that the temporal and spatial prediction performance was exceptional for all ecoregions. In conclusion, the RF model of WQI prediction developed in this study is considered to be applicable to Korean seas with high accuracy.