급격한 산업화와 도시화로 인해 해양 오염이 심각해지고 있으며, 이러한 해양 오염을 실효적으로 관리하기 위해 수질평가 지수(Water Quality Index, WQI)를 마련하여 활용하고 있다. 하지만 수질평가지수는 다소 복잡한 계산과정으로 인한 정보의 손실, 기준값 변동, 실무자의 계산오류, 통계적 오류 등의 불확실성(uncertainty)을 내포하고 있다. 이에 따라 국내ㆍ외에서 인공지능 기법을 활용하여 수질평가지수를 예측하기 위한 연구가 활발히 이루어지고 있다. 본 연구에서는 해양환경측정망 자료(2000 ~ 2020년)를 활용하여 우리나 라 전 해역 즉, 5개의 생태구에 대한 WQI를 추정할 수 있는 가장 적합한 인공지능기법을 도출하기 위해 총 6가지의 기법(RF, XGBoost, KNN, Ext, SVM, LR)을 실험하였다. 그 결과, Random Forest 기법이 다른 기법에 비해 가장 우수한 성능을 보였다. Random Forest 기법의 WQI 점수 예측값과 실제값의 잔차 분석 결과, 모든 생태구에서 시간적 및 공간적 예측 성능이 우수한 것으로 나타났다. 이를 통해 본 연구에서 개발한 Random Forest 기법은 높은 정확도를 바탕으로 우리나라 전해역에 대한 WQI를 예측 가능할 것으로 사료된다.

The management of algal bloom is essential for the proper management of water supply systems and to maintain the safety of drinking water. Chlorophyll-a(Chl-a) is a commonly used indicator to represent the algal concentration. In recent years, advanced machine learning models have been increasingly used to predict Chl-a in freshwater systems. Machine learning models show good performance in various fields, while the process of model development requires considerable labor and time by experts. Automated machine learning(auto ML) is an emerging field of machine learning study. Auto ML is used to develop machine learning models while minimizing the time and labor required in the model development process. This study developed an auto ML to predict Chl-a using auto sklearn, one of most widely used open source auto ML algorithms. The model performance was compared with other two popular ensemble machine learning models, random forest(RF) and XGBoost(XGB). The model performance was evaluated using three indices, root mean squared error, root mean squared error-observation standard deviation ratio(RSR) and Nash-Sutcliffe coefficient of efficiency. The RSR of auto ML, RF, and XGB were 0.659, 0.684 and 0.638, respectively. The results shows that auto ML outperforms RF, and XGB shows better prediction performance than auto ML, while the differences between model performances were not significant. Shapley value analysis, an explainable machine learning algorithm, was used to provide quantitative interpretation about the model prediction of auto ML developed in this study. The results of this study present the possible applicability of auto ML for the prediction of water quality.

Urban areas in watersheds increase the impervious surface, and agricultural areas deteriorate the water quality of rivers due to the use of fertilizers. As such, anthropogenic land use affects the type, intensity and quantity of land use and is closely related to the amount of substances and nutrients discharged to nearby streams. Riparian vegetation reduce the concentration of pollutants entering the watershed and mitigate the negative impacts of land use on rivers. This study analyzes the data through correlation analysis and regression analysis through point data measured twice a year in spring and autumn in 21 selected damaged tributary rivers within the Han River area, and then uses a structural equation model to determine the area land use. In the negative impact on water quality, the mitigation effect of riparian vegetation was estimated. As a result of the correlation analysis, the correlation between the agricultural area and water quality was stronger than that of the urban area, and the area ratio of riparian vegetation showed a negative correlation with water quality. As a result of the regression analysis, it was found that agricultural areas had a negative effect on water quality in all models, but the results were not statistically significant in the case of urban areas. As a result of the model estimated through the structural equation, BOD, COD, TN, and TP showed a mitigation effect due to the accumulation effect of river water quality through riparian vegetation in agricultural areas, but the effect of riparian vegetation through riparian vegetation was found in urban areas. There was no These results were interpreted as having a fairly low distribution rate in urban areas, and in the case of the study area, there was no impact due to riparian forests due to the form of scattered and distributed settlements rather than high-density urbanized areas. The results of this study were judged to be unreasonable to generalize by analyzing the rivers where most of the agricultural areas are distributed, and a follow-up to establish a structural equation model by expanding the watershed variables in urban areas and encompassing the variables of various factors affecting water quality research is required.

본 연구는 Bacillus subtilis를 활용한 바이오플락 양식 기술(Biofloc technology, BFT)을 이용하여 대농갱이(Leiocassis ussuriensis) 양식의 가능성을 확인하기 위해 90일 동안 생존, 성장지수와 사육수 수질의 변화를 관찰하였다. 대농갱이를 입식하기 전 BFT 수 제조를 위해 실험수조에 사료와 당밀을 첨가한 후 B. subtilis를 접종하여 40일간 수질을 안정화시켰다. 실험결과, 대농 갱이의 생존율은 대조구 92.7±3.2%와 BFT 실험구 95.8±3.3%로 조사되었다. 증체율은 대조구 118.1±9.0%와 BFT 실험구 197.7±15.6%을 보였고, 일간 성장율은 대조구 0.87±0.5%, BFT 실험 구 1.21±0.06%로 나타났다. 사료효율은 대조구가 43.7±2.6%이었고, BFT 실험구는 70.1±4.1%로 측정되어 BFT 실험구의 사료효율이 더 높은 것으로 조사되었다. 실험기간 동안의 수질 변화를 측정한 결과, pH는 대조구와 BFT 실험구 모두 감소되었고, MLSS는 대조구에서 변화를 보이지 않았지만, BFT 실험구에서는 90일째부터 유의한 증가를 보였다. NH4 +-N와 NO2 --N는 대조구 에서 실험 30일째부터 유의한 증가를 보였으나, BFT 실험구에서는 변화를 보이지 않았다. 결론 적으로 B. subtilis를 활용한 BFT 시스템을 대농갱이 양성 과정에 적용한 결과, 수질은 안정화 되는 경향을 보였고, 성장도와 사료효율은 대조구에 비해 높은 것으로 조사되어 긍정적인 효 과가 있는 것으로 확인되었다.

The effect of permanganate oxidation was investigated as water treatment strategy with a focus on comparing the reaction characteristics of NaOCl and sodium permanganate (NaMnO4) in algae (Monoraphidium sp., Micractinium inermum, Microcystis aeruginosa)-contained water. Flow cytometry explained that chlorine exposure easily damaged algae cells. Damaged algae cells release intracellular organic matter, which increases the concentration of organic matter in the water, which is higher than by NaMnO4. The oxidation reaction resulted in the release of toxin (microcystin-LR, MC-LR) in water, and the reaction of algal organic matter with NaOCl resulted in trihalomethanes (THMs) concentration increase. The oxidation results by NaMnO4 significantly improved the concentration reduction of THMs and MC-LR. Therefore, this study suggests that NaMnO4 is effective as a pre-oxidant for reducing algae damage and byproducts in water treatment process.

본 연구에서는 하수 재이용을 위한 역삼투막 공정에서 전처리 정밀여과막(MF) 손상에 대한 누출되는 다양한 수 질변화로써 막 손상 검지 방안을 제시하였다. 이를 위하여 역삼투막 유입수질 적합성 평가지표인 SDI (silt density index)를 3에서 5의 범위 내에서 막 손상 시 검지 감도를 정량화하기 위하여 전처리 분리막이 1에서 3가닥 파단에 따라 SDI는 1.92에 서 6.11까지 증가한 결과를 확인할 수 있었다. 일반적으로 3을 기준으로 역삼투막 유입수질로 설정하였을 때 분리막이 3가닥 까지 파단이 되어야만 막 손상 검지가 가능하다는 것을 의미하며 역삼투막의 오염은 잠재적으로 가속화되어 효율을 저하시 킬 수 있다. 또한 이때 누출되는 입자성과 유기물질에 대하여 0.45 μm 이상의 크기만 걸러주는 입자계수는 입도분포별 막 파 단 개수에 따라 일정한 패턴을 확인할 수 없었으며, TOC 농도는 약 2배의 변화패턴으로써 SDI와의 상관관계로써 TOC가 막 손상 수질지표로써 신뢰성이 높은 것으로 확인되었다. 수질분석결과와 더불어 USEPA에서 제시하는 막 손상 검지 방법 중 압력손실시험과 이를 기반으로 LRVDIT 모델의 적합성 평가를 한 결과 막 손상 또는 역삼투막 공정으로 유입되는 막오염물질 을 신속하게 확인할 수 있는 SDI 및 TOC를 포함한 LRVDIT 모니터링과 UCL 설정을 병행해야 한다.

The prediction of algal bloom is an important field of study in algal bloom management, and chlorophyll-a concentration(Chl-a) is commonly used to represent the status of algal bloom. In, recent years advanced machine learning algorithms are increasingly used for the prediction of algal bloom. In this study, XGBoost(XGB), an ensemble machine learning algorithm, was used to develop a model to predict Chl-a in a reservoir. The daily observation of water quality data and climate data was used for the training and testing of the model. In the first step of the study, the input variables were clustered into two groups(low and high value groups) based on the observed value of water temperature(TEMP), total organic carbon concentration(TOC), total nitrogen concentration(TN) and total phosphorus concentration(TP). For each of the four water quality items, two XGB models were developed using only the data in each clustered group(Model 1). The results were compared to the prediction of an XGB model developed by using the entire data before clustering(Model 2). The model performance was evaluated using three indices including root mean squared error-observation standard deviation ratio(RSR). The model performance was improved using Model 1 for TEMP, TN, TP as the RSR of each model was 0.503, 0.477 and 0.493, respectively, while the RSR of Model 2 was 0.521. On the other hand, Model 2 shows better performance than Model 1 for TOC, where the RSR was 0.532. Explainable artificial intelligence(XAI) is an ongoing field of research in machine learning study. Shapley value analysis, a novel XAI algorithm, was also used for the quantitative interpretation of the XGB model performance developed in this study.

Benthic macroinvertebrates are important ecological and environmental indicators as primary or secondary consumers, and therefore are widely used in the evaluation of aquatic environments. However, there are no comprehensive river ecosystem monitoring surveys that link the major physicochemical water quality items with benthic macroinvertebrates in urban streams. Therefore, this study investigated the distribution characteristics of benthic macroinvertebrates and physicochemical water quality items (17 items) in Yangjaecheon and Yeouicheon from 2019 to 2020. At the same time, by applying Spearman’s rank correlation analysis and nonmetric multidimensional scaling (nMDS) analysis in the water quality data and biotic index, we tried to provide basic data for diagnosing the current status of river ecosystems in major urban rivers in Seoul. Based on the study results, a total of 39 species and 3,787 individuals were identified in Yangjaecheon, the water quality (based on BOD, TOC, and TP) of Yangjaecheon was higher than Grade Ib (good), and the BMI using benthic macroinvertebrates appeared as Grade C (normal) at all the sites. In Yeouicheon, a total of 51 species and 4,199 individuals were identified, the water quality (based on BOD, TOC, TP) was higher than Grade Ib (good) similar to Yangjaecheon, and the BMI of both Upstream and Saewon bridge was Grade B (good), while Yeoui bridge was Grade C (normal). Overall, analysis results for the distribution of benthic macroinvertebrates by a nonmetric multidimensional scaling method showed no significant difference between the two streams (p=0.1491). Also, significant environmental variables related to benthic macroinvertebrates distribution were determined as water temperature and DO. On the other hand, the results of the correlation analysis between biotic index and major water quality items confirmed that R1 and BMI could be used for on-site urban river water quality evaluation.

Understanding the characteristics of reservoir water quality is fundamental in reservoir ecosystem management. The water quality of reservoirs is affected by various factors including hydro-morphology of reservoirs, land use/cover, and human activities in their catchments. In this study, we classified 83 major reservoirs in South Korea based on nine physicochemical factors (pH, dissolved oxygen, chemical oxygen demand, total suspended solid, total nitrogen, total phosphorus, total organic carbon, electric conductivity, and chlorophyll-a) measured for five years (2015~2019). Study reservoirs were classified into five main clusters through hierarchical cluster analysis. Each cluster reflected differences in the water quality of reservoirs as well as hydromorphological variables such as elevation, catchment area, full water level, and full storage. In particular, water quality condition was low at a low elevation with large reservoirs representing cluster I. In the comparison of eutrophication status in major reservoirs in South Korea using the Korean trophic state index, in some reservoirs including cluster IV composed of lagoons, the eutrophication was improved compared to 2004~2008. However, eutrophication status has been more impaired in most agricultural reservoirs in clusters I, III, and V than past. Therefore, more attention is needed to improve the water quality of these reservoirs.

저서동물은 저서환경특성을 나타내는 중요한 지시자로 알려져 있다. 본 연구에서는 무안만 조 하대의 환경 및 저서동물의 분포특성을 조사하였으며, 수질평가지수(WQI)와 저서생물지수 (AMBI)를 이용하여 저서생태계 건강성을 평가하였다. 현장채집은 2019년 하계 무안만 조하대 의 10개 정점에서 이루어졌다. 무안만 조하대는 상부지역이 하부지역에 비해 세립한 입도특성 을 나타내고 있었으며, 높은 유기물 함량을 보였다. 일부 정점에서 오염지표종인 Musculista senhousia, Theora fragilis and Lumbrineris longifolia과 같은 종들도 우점을 나타내고 있었다. 군집분석결과 무안만 조하대는 상부, 중부, 하부 그룹으로 구분되었으며, 유기물 함량과 저서 건강성 평가지수(WQI 및 AMBI)와의 상관결과와 일치하였다. 본 연구결과, 무안만 조하대의 저서생태계는 양호한 것으로 평가되었다. 하지만 저서동물이 균등하게 분포하지 않고, 기회 종이 출현하고 있어 조하대의 유기물 부하량이 증가하고 있는 것으로 보인다.

빈번한 가뭄의 피해는 수자원의 양극화를 초래하고 있다. 국내에서는 2011년부터 2017년 사이에 국지적이고 주 기적인 이상 가뭄이 지속되어 10년 가뭄 빈도로 설계된 소규모 저수지의 저수율이 감소하였다. 이러한 저수율의 감소 는 수질을 떨어트릴 수 있고 이미 확보된 용수의 사용마저 제한할 수 있다. 따라서 가뭄에 대비하고 극복하기 위해서 는 수질관리와 가뭄 빈도의 재산정이 필요하다. 이 연구는 저수지에서 수질 변화의 원인인 잠재적 오염물질을 추정하고, 가뭄 기간 동안 저수지의 저수율 감소와 그에 따른 수질 변화를 검토한 것이다.