• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.42-42
• /
• 2022

지금까지도 유역에서의 녹조 모니터링은 현장채수를 통한 점 단위 모니터링에 크게 의존하고 있어 기후, 유속, 수온조건 등에 따라 수체에 광범위하게 발생하는 녹조를 효율적으로 모니터링하고 대응하기에는 어려운 점들이 있어왔다. 또한, 그동안 제한된 관측 데이터로 인해 현장 측정된 실측 데이터 보다는 녹조와 관련이 높은 NDVI, FGAI, SEI 등의 파생적인 지수를 산정하여 원격탐사자료와 매핑하는 방식의 분석연구 등이 선행되었다. 본 연구는 녹조의 모니터링시 정확도와 효율성을 향상을 목표로 하여, 우선은 녹조 측정장비를 활용, 7000개 이상의 녹조 관측 데이터를 확보하였으며, 이를 바탕으로 동기간의 고해상도 위성 자료와 실측자료를 매핑하기 위해 다양한Machine Learning기법을 적용함으로써 그 효과성을 검토하고자 하였다. 연구대상지는 낙동강 내성천 상류에 위치한 영주댐 유역으로서 데이터 수집단계에서는 면단위 현장(in-situ) 관측을 위해 2020년 2~9월까지 4회에 걸쳐 7291개의 녹조를 측정하고, 동일 시간 및 공간의 Sentinel-2자료 중 Band 1~12까지 총 13개(Band 8은 8과 8A로 2개)의 분광특성자료를 추출하였다. 다음으로 Machine Learning 분석기법의 적용을 위해 algae_monitoring Python library를 구축하였다. 개발된 library는 1) Training Set과 Test Set의 구분을 위한 Data 준비단계, 2) Random Forest, Gradient Boosting Regression, XGBoosting 알고리즘 중 선택하여 적용할 수 있는 모델적용단계, 3) 모델적용결과를 확인하는 Performance test단계(R², MSE, MAE, RMSE, NSE, KGE 등), 4) 모델결과의 Visualization단계, 5) 선정된 모델을 활용 위성자료를 녹조값으로 변환하는 적용단계로 구분하여 영주댐뿐만 아니라 다양한 유역에 범용적으로 적용할 수 있도록 구성하였다. 본 연구의 사례에서는 Sentinel-2위성의 12개 밴드, 기상자료(대기온도, 구름비율) 총 14개자료를 활용하여 Machine Learning기법 중 Random Forest를 적용하였을 경우에, 전반적으로 가장 높은 적합도를 나타내었으며, 적용결과 Test Set을 기준으로 NSE(Nash Sutcliffe Efficiency)가 0.96(Training Set의 경우에는 0.99) 수준의 성능을 나타내어, 광역적인 위성자료와 충분히 확보된 현장실측 자료간의 데이터 학습을 통해서 조류 모니터링 분석의 효율성이 획기적으로 증대될 수 있음을 확인하였다.

• Journal of Environmental Impact Assessment
• /
• v.32 no.6
• /
• pp.473-487
• /
• 2023

Korea's multi-purpose dams, which were constructed in the 1970s and 1980s, have a single outlet located near the bottom for hydropower generation. Problems such as freezing damage to crops due to cold water discharge and an increase the foggy days have been raised downstream of some dams. In this study, we analyzed the effect of water intake depth on the reservoir's water temperature stratification structure and outflow temperature targeting Hapcheon Reservoir, where hypolimnetic withdrawal is drawn via a fixed depth outlet. Using AEM3D, a three-dimensional hydrodynamic water quality model, the vertical water temperature distribution of Hapcheon Reservoir was reproduced and the seasonal water temperature stratification structure was analyzed. Simulation periods were wet and dry year to compare and analyze changes in water temperature stratification according to hydrological conditions. In addition, by applying the intake depth change scenario, the effect of water intake depth on the thermal structure was analyzed. As a result of the simulation, it was analyzed that if the hypolimnetic withdrawal is changed to epilimnetic withdrawal, the formation location of the thermocline will decrease by 6.5 m in the wet year and 6.8 m in the dry year, resulting in a shallower water depth. Additionally, the water stability indices, Schmidt Stability Index (SSI) and Buoyancy frequency (N²), were found to increase, resulting in an increase in thermal stratification strength. Changing higher withdrawal elevations, the annual average discharge water temperature increases by 3.5℃ in the wet year and by 5.0℃ in the dry year, which reduces the influence of the downstream river. However, the volume of the low-water temperature layer and the strength of the water temperature stratification within the lake increase, so the water intake depth is a major factor in dam operation for future water quality management.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_3
• /
• pp.1747-1761
• /
• 2023

Wetlands, defined as lands periodically inundated or exposed during the year, are crucial for sustaining biodiversity and filtering environmental pollutants. The importance of mapping and monitoring their topographical changes is therefore paramount. This study focuses on the topographical variations at the Incheon River estuary wetland post-restoration, noting a lack of adequate prior measurements. Using a multi-temporal Sentinel-1 dataset from October 2014 to March 2023, we mapped long-term variations in water bodies and detected topographical change anomalies using a pixel-frequency approach. Our analysis, based on 196 Sentinel-1 acquisitions from an ascending orbit, revealed significant topography changes. Since 2020, employing the pixel-frequency technique, we observed area increases of +0.0195, 0.0016, 0.0075, and 0.0163 km² in water level sections at depths of 2-3 m, 1-2 m, 0-1 m, and less than 0 m, respectively. These findings underscore the effectiveness of the wetland restoration efforts in the area.

• Journal of Korea Water Resources Association
• /
• v.56 no.11
• /
• pp.705-720
• /
• 2023

In South Korea, ongoing incidents related to drinking water quality have eroded consumer trust. Specifically, beyond quality incidents, there have been complaints about taste, odor, and other issues stemming from the presence of chlorine. To address this, water service operators are employing various management strategies from both temporal (scheduling) and spatial (rechlorination) perspectives to ensure uniform and safe distribution of chlorine residuals. In this study, we focus on the optimal monthly management of chlorine residuals, based on water distribution network analysis. Water quality reaction coefficients, including bulk fluid and wall reaction coefficients, were estimated through lab-scale tests and EPANET water quality simulations, respectively, accounting for temperature variations in a large-scale water distribution network. Utilizing these estimated coefficients, we examined the monthly variations in chlorine residual distribution under different chlorine injection conditions. The results indicate that the efficient concentration for chlorine injection, which satisfies the residual chlorine limit range, varies with temperature changes. Consequently, it is imperative to establish a specific and quantitative chlorine injection plan that considers the accurate spatial distribution of monthly chlorine residuals.

• Korean Journal of Ecology and Environment
• /
• v.56 no.3
• /
• pp.268-279
• /
• 2023

Nowadays, artificial intelligence model approaches such as machine and deep learning have been widely used to predict variations of water quality in various freshwater bodies. In particular, many researchers have tried to predict the occurrence of cyanobacterial blooms in inland water, which pose a threat to human health and aquatic ecosystems. Therefore, the objective of this study were to: 1) review studies on the application of machine learning models for predicting the occurrence of cyanobacterial blooms and its metabolites and 2) prospect for future study on the prediction of cyanobacteria by machine learning models including deep learning. In this study, a systematic literature search and review were conducted using SCOPUS, which is Elsevier's abstract and citation database. The key results showed that deep learning models were usually used to predict cyanobacterial cells, while machine learning models focused on predicting cyanobacterial metabolites such as concentrations of microcystin, geosmin, and 2-methylisoborneol (2-MIB) in reservoirs. There was a distinct difference in the use of input variables to predict cyanobacterial cells and metabolites. The application of deep learning models through the construction of big data may be encouraged to build accurate models to predict cyanobacterial metabolites.

• Korean Journal of Ecology and Environment
• /
• v.37 no.1 s.106
• /
• pp.12-25
• /
• 2004

Physicochemical parameters, plankton biomass, and sediment were surveyed from 1998 to 2000 at two months interval in a eutrophic coastal lagoon(Lake Hwajinpo, Korea). The lake is separated from the sea by a narrow sand dune. Littoral zone is well vegetated with leafing-leaved aquatic plants. The lake basin is divided into two subbasins by a shallow sill. It has intrusion of seawater by permeation and stormy waves. Stable chemoclines are formed by salinity difference at 1m depth all the year round. DO was often very low (< 1 mg$O_2\;L^{-1}$) at hypolimnion. Temperature inversions were observed in November. Nitrate and ammonium concentrations were very low(< (1.1 mgN $L^{-1}$), even though TN was usually 2.0 ${\sim}$ 3.5 mgN $L^{-1}$. TN/TP was generally lower than the Redfield ratio. Transparency was 0.2 ${\sim}$ 1.7 m, and COD, TP, and TN of sediment were 3.1 ${\sim}$ 40.3 mg$O_2\;g^{-1}$, 0.91 ${\sim}$ 1.39 mgP $g^{-1}$, and 0.34 ${\sim}$ 3.07 mgN $g^{-1}$, respectively. Phytoplankton chlorophyll- a concentrations were mostly over 40 mg $m^{-3}$. Two basins showed different phytoplankton communities with Oscillatoria so., Trachelomonas sp., Schizochlamys gelatinosa, and Anabaena spiroides dominant in South basin, and with Trachelomons sp., Schroederia so., schizochlamys gelatinosa, and Trachelomonas sp. dominant in the North basin. The seasonal succession of phytoplankton was very fast, possibly due to sudden changes in physical conditions, such as wind, turbidity, salinity and light.

• Journal of Bio-Environment Control
• /
• v.22 no.4
• /
• pp.366-370
• /
• 2013

The effects of the early tree shapes with renewal by top-working on Y-shaped peach tree, and the influence on the maintaining fruit quantity with gradual renewal of interstock cultivar were determined. In the comparison of the places of top-working tree, top working tree on the inside of the main branch of interstock cultivar had higher graft union rate and branch growth than those of top working tree on the outside. Tree width, basal diameter of shoot, and number of bearing shoots were smaller in top working tree than in replanted tree. Although labor time was not different to control top working tree and replanted tree, labor time was much required to manage top working tree with interstock cultivar. Accumulated fruit production was 2,384 kg/10a in top working tree and 2,025 kg/10a in replanted tree for three years. However, top working tree had no loss of fruits because interstock cultivar of top-working tree had 3,727 kg/10a of fruits. No variation on fruit quality was observed between top working tree and replanting tree. In terms of economic value of top-working tree, labor to manage interstock cultivar, fertilizer price, fruit bagging, and grading and packing price increased. However, fruit production increased, and price of seedling, rental equipment, pulling-out trees, and repairing supporting system decreased. Therefore, gradual renewal of topworking tree has effects on the maintaining fruit quantity, supplementation on fruit loss, and renewal cultivar.

• Journal of Korea Foresty Energy
• /
• v.23 no.1
• /
• pp.1-18
• /
• 2004

Effects of nitrogen and phosphorus fertilization on aboveground biomass and distribution of nutrient content in tree components were determined in adjacent 41-year-old plantations of Pinus rigida Miller and Larix kaempferi Gordon on a similar soil in Yangpyeong, Gyeonggi Province. Fertilizer was experimentally applied at the three levels [control(C), 200 N kg/ha+25 P kg/ha(LNP), and 400 N kg/ha+50 P kg/ha(HNP)] in June, 1996. The results obtained from this study were summarized as follows: Total aboveground biomass of P. rigida and L. kaempferi stands was 149, 145ton/ha for control, 166, 149ton/ha for LNP, and 152, 170ton/ha for HNP, respectively, and did not differ between the tree species. Aboveground biomass of P. rigida and L. kaempferi plantations were distributed as follows: stemwood>branch>stembark>foliage. Biomass of understory vegetation was 1,111kg/ha for P. rigida and 907kg/ha for L. kaempferi, and comprised 0.6～0.7% of the total aboveground biomass of the plantations. The concentrations of all nutrients(N, P, K, Ca, Mg) were generally higher in L. kaempferi than in P. rigida because of high nutrient uptake characteristics of Larix spp. Nutrient concentrations in tree components of P. rigida and L. kaempferi plantations were not significantly different among treatments, whereas they were significantly different between species and among tree components. The nutrient concentration in tree components of both tree species decreased in order of foliage>branch>stembark>stemwood. The contents of all nutrients were higher in the fertilized plots than in the control plots, generally the highest in LNP for P. rigida and in HNP for L. kaempferi. Total nutrient(N+P+K+Ca+Mg) contents of aboveground biomass were 703kg/ha for P. rigida and 869kg/ha for L. kaempferi, respectively.

• Journal of Korean Society of Forest Science
• /
• v.87 no.1
• /
• pp.90-97
• /
• 1998

This study was carried out to compare the characteristics of relative conductivity(RC) in the twenty-one deciduous broad-leaved species. The results obtained were summarized as follows : 1. The RC ranged from $3.18{\times}10^{-15}$ to $1.23{\times}10^{-10}m^2$ in deciduous broad-leaved species, and appeared different values in various portions of a tree and species. 2. In eight diffuse-porous species, the ranges of average RC were $3.49{\times}10^{-13}$ to $6.35{\times}10^{-11}m^2$ in stems. $4.89{\times}10^{-13}$ to $2.99{\times}10^{-11}m^2$ in branches, and $1.33{\times}10^{-13}$ to $9.42{\times}10^{-12}m^2$ in junction part between stem and branch. In thirteen ring-porous species, the RCs were $7.14{\times}10^{-13}$ to $4.76{\times}10^{-11}m^2$ in stems, $2.93{\times}10^{-13}$ to $3.91{\times}10^{-11}m^2$ in branches, and $1.14{\times}10^{-13}$ to $9.20{\times}10^{-12}m^2$ in junction part. 3. The RC and diameter of stem have no interrelation in eight diffuse-porous species. The RC of stem increased with decreasing diameter of stem in eleven ring-porous species except Fraxinus mandshurica and Maackia amurensis. 4. In five deciduous broad-leaved species, the RC of 1-year-old shoots(the leader and adjacent laterals ) ranged from $4.87{\times}10^{-13}$ to $8.41{\times}10^{-11}m^2$ in the leaders, and $7.93{\times}10^{-14}$ to $2.01{\times}10^{-11}m^2$ in the adjacent laterals. The RCs were much greater in the leaders than in the adjacent laterals because of the hydraulic dominance of the leader shoot.

• Horticultural Science & Technology
• /
• v.33 no.2
• /
• pp.155-165
• /
• 2015

Well-feathered (over 10 feathers) 'Fuji'/M.9 apple trees were planted at $3.0{\times}1.0m$ and trained to slender spindle with 2.5 m height or to tall spindle with 3.5 m height, and the vegetative growth, productivity, and fruit quality of two training systems were compared for 8 years. The canopy volume of the tall spindle trees surpassed that of the slender spindle trees 4 years after planting and was 25% larger than that of the slender spindle trees 5 years after planting. The accumulated yield over 8 years for the tall spindle system was 14% higher than that of the slender spindle system. Alternate bearing and incidence of marssonina blotch were observed in both treatments after 5 years of planting. There was often vegetative imbalance in the trees however, the degree of yield loss and vegetative imbalance of the tall spindle trees was lower than those of the slender spindle trees. Soluble solid content and fruit red color of the tall spindle trees were higher than that of the slender spindle trees in 5 year after planting, resulting from increased light penetration in the canopy due to even distribution of lateral branches and from fruit bearing in different height locations of the trees. In conclusion, increasing the tree height to about 3.5 m using slender spindle 'Fuji'/M.9 apple trees planted with over 333 trees per 10a led to better light penetration, yield and fruit quality compared to a conventional wide training system with the slender spindle.