• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.363-363
• /
• 2023

Artificial Neural Networks (ANN) is a powerful tool for addressing various practical problems and it has been extensively applied in areas of water resources. In this study, Artificial Neural Networks (ANNs) were developed for flood forecasting at specific locations on the Han River. The Partial Mutual Information (PMI) technique was used to select input variables for ANNs that are neither over-specified nor under-specified while adequately describing the underlying input-output relationships. Historical observations including discharges at the Paldang Dam, flows from tributaries, water levels at the Paldang Bridge, Banpo Bridge, Hangang Bridge, and Junryu gauge station, and time derivatives of the observed water levels were considered as input candidates. Lagged variables from current time t to the previous five hours were assumed to be sufficient in this study. A three-layer neural network with one hidden layer was used and the neural network was optimized by selecting the optimal number of hidden neurons given the selected inputs. Given an ANN architecture, the weights and biases of the network were determined in the model training. The use of PMI-based input variable selection and optimized ANNs for different sites were proven to successfully predict water levels during flood periods.

• Journal of the Korean GEO-environmental Society
• /
• v.3 no.4
• /
• pp.11-17
• /
• 2002

The purpose of this study is the prediction of pollutant loads into Seomjin river watershed using neural networks model. The pollutant loads into river watershed depend upon the water quantity of inflow from the upstream as well as the water quality of the inflow into the river. For the estimation of pollutants into river, a neural networks model which has the features of multi-layered structure and parallel multi-connections is used. The used water quality parameters are BOD, COD and SS into Seomjin river. The results of calibration are satisfactory, and proved the availability of a proposed neural networks model to estimate short-term water quality pollutants into river system.

• Journal of Industrial Technology
• /
• v.27 no.B
• /
• pp.229-233
• /
• 2007

This research investigated the way of generating the flowing of water in case of artificial fluctuation of river width by the unidimensional numerical analysis in order to reconstruct vertical and expanse features of flowing, and the problem of existing numerical analysis in accordance with local enlargement and reduction of river through hydraulic model experiments with results of numerical analysis. The result revealed that when the local section change in the same river is exist, it showed 0.93m in the case of no change of local section in the hydraulic model experiments and numerical analysis, however, it presented 1.645m on the occasion of local section changes in the hydraulic model experiments and numerical analysis. In other words, there was a significant difference in the existing numerical analysis, when there was a local section change. As a result of the experimental section for the enlargement and reduction of local river width, due to the sensitive change for fluctuation of flood discharge, there was a significant difference between numerical analysis and hydraulic model experiments. In addition, the result of comparison between the enlargement and reduction of local river width confirmed that the result of numerical analysis with hydraulic model experiments showed larger generation of deviation in case of enlargement of section than in case of reduction of section.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.446-446
• /
• 2018

The urban rivers have unique hydraulic characteristics between natural rivers and artificial canals. These hydraulic characteristics determine the characteristics of urban rivers with small environmental capacity and fragile ecosystems. With the development and utilization of natural resources, the pollutants that have been produced enter the river through different channels, which seriously damages the urban river ecosystem. Therefore, how to restore contaminated water to a normal state and reproduce a natural, self-regulating ecosystem is one of the most concerned issues in recently. Eco-hydraulics is a cross-disciplinary subject of hydraulics biology and ecology. It is closely related to the protection of rivers, wetlands, and ecological self-repair. In this study, The basic principle of eco-hydraulics is concisely described and its approaches to protection and rehabilitation of river are introduced. The conception of establishing gardenesque eco-pond for urban use is suggested. The strategies including changing the hydrodynamic features of rivers, adjusting the breeds and species and constructing the gardenesque eco-pond for improving the exist ing urban rivers are proposed. It provides scientific information and guidance for the restoration of rivers and wetlands by studying the close relationship between river hydraulic characteristics, currents, and rivers and ecosystems.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2000.05a
• /
• pp.251-256
• /
• 2000

• Journal of environmental and Sanitary engineering
• /
• v.16 no.3
• /
• pp.22-27
• /
• 2001

It have been investigated how algal and bacterial symbiotic reaction influences on removal of organic carbon in river ecosystem. And artificial experimentation apparatus was made for algae'and bacteia'culture as lab scale. Investigating and researching minutely the change of concentration of organic carbon substrate and the change of population density of algae'and of bacteria'with this artificial experimentation apparatus, the next results could be obtained. 1. Successful decrease of DOC(dissolved organic carbon) could not be expected unless algal and bacterial biomass floe was nut formed effectively and unless biosorption was not proceeded effectively in the very culture system in which artificial synthetic wastewater was supplied continuously at constant rate. 2. In conditions of culture liquid of 1335 glucnse mg/L(type 1) and of 267 glucose mg:L(type 2), the algal dominant species was always Chlorella vulgaris in both types in which artificial synthetic wastewater were supplied continuously at constant rate and algae population density was around maximum 107 cells/mL. 3. It was around 108 ~ 107 cells/mL that the population density of heterotrophic bacterium. In culture medium systems type 1 and type 2 in which artificial wastewater were supplied continuously at constant rate, the same density appeared initially when using the population density of Escherichia coli w 3110 as indirect indicator. And this density decreased rapidly till the culturing date 35 days were passed away, while this density increased with gentle slope after same date and then the trend of change at type 2 was more severe than one at type 1. 4. When seeing such a change of population density of Escherichia coli w 3110, the growth of heterotrophic bacterium appeared as survival instinct pattern of broader requirement of nutrient at condition of low concentration of organic carbon substrate than condition of high concentration of same substrate.

• Ecology and Resilient Infrastructure
• /
• v.7 no.4
• /
• pp.345-352
• /
• 2020

Securing the lead time for evacuation is crucial to minimize flood damage. In this study, downstream water levels for heavy rainfall were predicted using measured water level observation data. Multiple regression analysis and artificial neural networks were applied to the Seom River experimental watershed to predict the water level. Water level observation data for the Seom River experimental watershed from 2002 to 2010 were used to perform the multiple regression analysis and to train the artificial neural networks. The water level was predicted using the trained model. The simulation results for the coefficients of determination of the artificial neural network level prediction ranged from 0.991 to 0.999, while those of the multiple regression analysis ranged from 0.945 to 0.990. The water level prediction model developed using an artificial neural network was better than the multiple-regression analysis model. This technique for forecasting downstream water levels is expected to contribute toward flooding warning systems that secure the lead time for streams.

• Journal of Korean Society on Water Environment
• /
• v.34 no.5
• /
• pp.447-461
• /
• 2018

To realize river restoration that ecological characteristics of the river are reflected, we classified the river into four reaches of valley stream, upstream, midstream, and downstream based on substrate as well as riverbed gradient obtained from the relationship between distance from the river mouth, and above sea level. Considering that the rivers of Korea have been dominated by various and intense artificial interferences over a lengthy period, we determined cross sectional range of the river based on the geological map and clarified transformation degree by reach. Vegetation profile diagram was prepared by depicting horizontal range and vertical stratification of major vegetation appearing in a belt transect of 10 m breadth installed between weirs constructed in both sides of the river. Restoration models by river reach were prepared based on breadth of waterway, bare ground, herb, shrub, and tree dominated vegetation zones on vegetation profiles wherein a flooding regime was reflected. Species composition information collected from vegetation established in each zone was systematized to use for restoring each reach ecologically. Further, background that longitudinal reaches and horizontal zones were divided, was discussed by comparing with case studies in foreign countries. In addition, necessity of ecological restoration of the river was discussed based on degree of integrity of Korean rivers, ecological significance of riparian vegetation, and importance of reference information for ecological restoration of the river.

• Fisheries and Aquatic Sciences
• /
• v.11 no.1
• /
• pp.36-40
• /
• 2008

We analyzed the nucleotide sequences of about 500 bp of the mitochondrial NADH dehydrogenase subunit 3 (ND3) gene to estimate the genetic variation of Korean masu salmon (Oncorhynchus masou) populations. DNA samples were collected from 104 river-only specimens and 52 anadromous specimens from three hatcheries and one river. There are no records of artificial release into the river. We amplified the ND3 gene by polymerase chain reaction, targeting areas that included parts of the cytochrome oxidase III gene and the NADH dehydrogenase subunit 4L gene, and defined 14 haplotypes based on 12 variable nucleotide sites in the examined region. Among the haplotypes, ten were specific to river-only specimens within hatchery populations. Haplotype diversity of river-only populations in hatcheries was higher than that of anadromous and wild populations. Pairwise population $F_{ST}$ estimates and neighbor-joining tree analyses inferred that anadromous and river-only populations were distinct. These results suggest that sequence polymorphism in the ND3 region may be a useful marker for analyzing the genetic variation and population structure of masu salmon.

• Journal of Environmental Science International
• /
• v.11 no.9
• /
• pp.881-895
• /
• 2002

Sediments and river waters form the channel of Hwasun-cheon were studied in sedimentological size and geochemical analyses of metallic elements for the purposes of identification of depositional environments and geochemical characteristics. The sizes of sediments are assigned to pebble to coarse sand in mean size and polymodal in distribution. And the sediments are poorly to very poorly sorted and positively skewed. According to the grain size distributions of the sediments, the Hwasun-cheon belongs to gravel-bed river on the basis of the grain size distribution of the sediments. The behaviors of metallic elements in the sediments mainly depend on not grain size distribution but the geology connected with geomorphological reliefs near the stream. Contamination indices(CI) of Zn, Cu and Pb are 2.83 to 6.96 with average 4.31 in the sediments. Hwasun-cheon is assigned to general stream type in accordance with water quality of physical factors and chemical characteristics by Piper's diagram. Though meaningful values of BOD, T-N, T-P were locally depicted near Masan-ri, Hwasun-eup and Jiseok-cheon areas, artificial metal concentration do not represent in the most area of the stream. Sediments and river water are considered that the relatively more or less high metallic contents in the stream are originated from coal mine and urban area.