• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.107-107
• /
• 2017

최근 과학 기술의 발달과 정보화로 인해 다양한 분야의 방대한 정보들이 디지털화되어 저장되고, 이를 효율적으로 저장하고 관리하는 기술에 대한 수요가 증대하고 있다. 특히, 수자원 분야에서도 지점 관측 형태의 직접 계측방식이 아닌 선, 면 형태의 간접 계측방식의 기술이 개발됨에 따라 짧은 시간 동안 많은 수의 양질의 자료를 취득할 수 있게 되었다. 이에 따라 과거에 비해 방대하고 연속적인 하천공간정보가 축척되었으며, 이를 저장하고 관리하고자 하는 필요성이 대두되고 있다. 하천공간정보 중 하천단면자료는 하천기본계획수립 시 핵심이 되는 자료중 하나일 뿐만 아니라 하천의 흐름 특성을 해석하기 위한 여러 물리 모형의 입력 자료로서, 하천의 물이 흐르는 하도공간을 흐름방향의 횡방향으로 절단하였을 때 생기는 면을 기록한 자료이다. 국가에서는 하천단면자료 계측을 위해 많은 예산을 투입하고 있으며, 실제로 많은 수의 하천에서 하천단면자료를 계측하고 있다. 하지만 계측된 하천단면자료는 컴퓨터를 이용하여 제도하는 응용프로그램용 파일(CAD)로 저장되어, 하천단면자료를 지속적으로 저장 관리 모니터링하기에는 어려운 실정이다. 본 연구에서는 이러한 하천단면자료를 지속적으로 저장 관리 모니터링 할 수 있는 시스템 개발을 목적으로 현재 개발 중인 하천공간정보 표준 자료모형과 하천공간정보를 전송할 수 있는 표준자료 전송 언어인 RiverML을 활용하였다. 본 연구에서 도입한 하천공간정보 표준 자료모형은 하천망을 기반으로 한 상호연계를 기본으로 하고 있어, 하천망을 구성하는 각 하천과 해당 하천의 단면자료 간의 관계를 규명하기 쉬우며, RiverML은 이와 유사하게 하천망을 기반으로 다양한 정보를 전달하기 쉬운 전송언어이다. 또한 국가에서 개발한 공간정보 오픈 플랫폼인 브이월드(Vworld)의 API를 활용하여 저장된 하천단면자료를 표출할 수 있게 하였다. 본 연구의 결과는 하천단면자료 저장 관리 모니터링에 유용하게 사용될 수 있을 뿐만 아니라, 지속적인 개발을 통해 하천에서 계측되는 공간정보뿐만 아니라 하천의 시설물, 생태환경, 문화 등의 하천공간정보과 연계하여 효율적인 시스템을 개발할 수 있을 것으로 사료된다.

• Smart Media Journal
• /
• v.11 no.2
• /
• pp.31-38
• /
• 2022

Promptly predicting changes in the salinity in rivers is an important task to predict the damage to agriculture and ecosystems caused by salinity infiltration and to establish disaster prevention measures. Because machine learning(ML) methods show much less computation cost than physics-based hydraulic models, they can predict the river salinity in a relatively short time. Due to shorter training time, ML methods have been studied as a complementary technique to physics-based hydraulic model. Many studies on salinity prediction based on machine learning have been studied actively around the world, but there are few studies in South Korea. With a massive number of datasets available publicly, we evaluated the performance of various kinds of machine learning techniques that predict the salinity of the Nakdong River Estuary Basin. As a result, LightGBM algorithm shows average 0.37 in RMSE as prediction performance and 2-20 times faster learning speed than other algorithms. This indicates that machine learning techniques can be applied to predict the salinity of rivers in Korea.

• Asian-Australasian Journal of Animal Sciences
• /
• v.13 no.7
• /
• pp.897-900
• /
• 2000

The experiment was conducted to study the effect of GnRH administration at induced estrus on pituitary and ovarian response in buffalo heifers. Eight Murrah river buffaloes of 12 to 13 months of age were treated with $PGF_{{2}{\alpha}}$ and eCG combination. GnRH (Fertagyl) 200 ug was injected (iv) at estrus in four heifers (treated group) while saline (2 ml, iv) was injected in remaining four heifers (control group). Blood was collected through jugular catheter to estimate plasma FSH, LH and estradiol level. The pretreatment plasma FSH, LH and estradiol values ranged from $8.46{\pm}1.97ng/ml$ to $12.31{\pm}1.30ng/ml$, $0.87{\pm}0.21ng/ml$ to $1.19{\pm}0.29ng/ml$ and $19.09{\pm}2.38pg/ml$ to $20.24{\pm}1.00pg/ml$ respectively. The plasma estradiol concentration elevated significantly (p<0.05) within 24 hr after eCG administration and reached its peak levels of $154.09{\pm}17.28pg/ml$ and $181.95{\pm}31.82pg/ml$ at estrus in respectively treatment and control groups. The plasma FSH and LH concentrations did not increase during follicular development after eCG administration while initial significant (p<0.05) increases in both plasma FSH and LH concentrations occured within 5 and 10 min, reaching peak levels of respectively $110.06{\pm}23.56ng/ml$ and $13.15{\pm}3.13ng/ml$ within 90 min after GnRH injection was detected. A sharp and significant decline in plasma estradiol concentration ($59.27{\pm}8.78pg/ml$) associated with synchronized ovulation within 24 hours after GnRH injection was recorded. The observation suggest that the hypophysis of prepubertal buffaloes treated with eCG have gonadotrophins awaiting the releasing factor to evoke release of gonadotrophin during the follicular phase to induce synchronized ovulation.

• ALGAE
• /
• v.17 no.3
• /
• pp.171-185
• /
• 2002

The standing crop of genus Microcystis, microcystin concentrations and environmental factors were monitored at stations of the lower reaches of the Nakdong River in 1998 and 1999 during the periods of its occurrence. The Microcystis were observed from May to Octorber, and the cell density was highest at Station Seonam up to 250,000 cells${\cdot}ml^{-1}$ forming scum over the water surface. There were signigicant relationships between the standing crop of Microcystis and nitrate nitrogen, total phosphorus concentrations and Ph. Presumably these parameters were important in the succession to Microcystis dominated phytoplankton community in the summer period in the river. However, Ammonium nitrogen, phosphate phosphorus concentrations and N/P ratio were not critical factors. The Microcystis bloom was notable above $25^{\circ}C$ of surface water temperature. Microcystins were detected from May to November in the algal materials from the river. The 84.2% of algal materials with Microcystis exhibited the microcystin with the maximum of 1711.8 ${\mu}g{\cdot}g^{-1}$ dw. The microcystin concentrations in the algal materials were significantly related to the stading crop of Microcystis, which was the primary determinant factor in the toxin levle of algal materials. The concentrations were also significantly related to pH of the water column in the positive pattern.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.136-136
• /
• 2022

This study investigates the possibility of coupling empirical mode decomposition (EMD) for runoff prediction from machine learning (ML) models. Here, support vector regression (SVR) and convolutional neural network (CNN) were considered for ML algorithms. Precipitation (P), minimum temperature (Tmin), maximum temperature (Tmax) and their intrinsic mode functions (IMF) values were used for input variables at a monthly scale from Jan. 1973 to Dec. 2020 in the Grand river basin, Canada. The support vector machine-recursive feature elimination (SVM-RFE) technique was applied for finding the best combination of predictors among input variables. The results show that the proposed method outperformed the individual performance of SVR and CNN during the training and testing periods in the study area. According to the correlation coefficient (R), the EMD-SVR model outperformed the EMD-CNN model in both training and testing even though the CNN indicated a better performance than the SVR before using IMF values. The EMD-SVR model showed higher improvement in R value (38.7%) than that from the EMD-CNN model (7.1%). It should be noted that the coupled models of EMD-SVR and EMD-CNN represented much higher accuracy in runoff prediction with respect to the considered evaluation indicators, including root mean square error (RMSE) and R values.

• Korean Journal of Ecology and Environment
• /
• v.33 no.2 s.90
• /
• pp.136-144
• /
• 2000

M. aeruginosa isolated from Lake Paldang was cultured in CB medium, and then each wet plants put into the cultured medium at a rate of 0.5 g and 2.5 g wet wt/l. There was slight inhibition by the input of cattail and iris of each 0.5 g wet wt/l cultured medium, but showed no reduction in algal growth in other flasks. Among the applied plants, ginkgo, pine needles, big cone pine, waterreed and water chestnut had an effect on inhibition of algal growth at the input of 2.5 g wet wt/l. Plants which were dried for 3 days at $50^{\circ}C$ introduced into the testing flask for 10days cultured at dose rates of 2.5 g/l. When chlorophyll a concentration was remarkably high as $802.6\;{\mu}g/l$ after five days, there was noticeably less chlorophyll compared with control at a rate of 98% by big cone pine, 96% by ginkgo, 95% by pine needles and 86% by rice straw, respectively. To examine the effect of plant extracts on algal growth, big cone pine and water chestnut were put to the amount of 1.25 g liquid extracts/l. Chlorophyll a concentration and cell density decreased to the extent of average 43% as compared with the beginning of experiment, but when concentration of chlorophyll a increased a most high, the inhibition of algal growth by liquid extracts did not affect at all. When a quantity of plant equivalent to 2.5 g liquid extracts/l inhibited the growth of algae by 95% after nine days.

• ALGAE
• /
• v.18 no.3
• /
• pp.207-215
• /
• 2003

Dynamics of phytoplankton communities were investigated at five stations in Youngsan River from December 1995 to November 1996. Phytoplankton were identified to the total of 466 taxa, which were composed of 136 genera, 438 species, 27 varieties and 1 forma. The standing crops ranged 1,163-50,765 cells ${\cdot}ml^{-1}$ during the investigation periods. The variation of classes was in order to Chrysophyceae - Bascillariophyceae - Chlorophyceae at St. 1 and was only Bacillariophyceae at the other stations. The dominant species were 12 taxa including Aulacoseira ambigua, A. garnulata, Cyclotella meneghiniana, C. pseudostelligera, Chlorella vulgaris, Dinobryon sertularia, Flagilaria elliptica, Monoraphidium contortum, Micractinium pusillum, Nitzschia palea, Scenedesmus grahneisii and Stephanodiscus hantzschii f. tenuis. The standing crops of dominant species ranged from 10.6% to 94.7%. The genus Dinobryon in Chrysophyceae was dominant species at St. 1 in December 1995, but not recorded in the other months. The dominant species were composed with the planktonic diatoms from winter to spring and were the tychoplanktonic and the benthic species at St. 2-5 from summer to fall. The relationships between total standing crops and water temperature, pH, $NH_4$, $NO_2$, $NO_3$ and $PO_4$ showed low positive or negative coefficients. Stephanodiscus hantzschii f. tenuis had low positive or negative coefficients with water temperature, $NH_4$, $NO_2$, $NO_3$ and $PO_4$. The fluctuations of standing crops in the Youngsan river phytoplankton community were not caused by a single factor but controlled by the complex factors with interaction between phytoplankton community and envirowmental factors.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.14 no.3
• /
• pp.148-157
• /
• 1981

This experiment was carried out to evaluate the sanitary quality of sea water at popular swimming beaches such as Haeundae, Kwanganri and Songjeong which are located in the suburbs of Busan Korea. One hundred and seventy five sea water samples were collected from the 25 stations during the both summer seasons in 1979 and in 1980. The sampling stations were established in the above 3 swimming beaches and the estuary of Suyeong river (See Fig.1.). To evaluate the sanitary quality, coliform group, fecal coliform, viable cell count and Vibrio parahaemolyticus were observed. The median value of coliform group MPN of the samples ranged 2,300 to 6,500/100m1 in Kwanganri, 670 to 910/100m1 in Haeundae and 430 to 460/100m1 in Songjeong swimming beach. Geometric mean of fecal coliform MPN was 3,200/100m1 in Kwanganri, 500/100m1 in Haeundae and 360/100m1 in Songjeong swimming beach. Composition of coliform was $23\%$ Escheri-chia coli group, $20\%$ Citrobacter freundii group, $41\%$ Klebsiella aerogenes group and $16\%$others. The density of Vibrio parahaemolyticus in Kwanganri was much higher than those of in other beaches by 5 to 6 times. Plate count of the samples in the estuary of Suyeong river was varied from 300 to $8.5\times10^6/ml$ during the study period. No remarkable difference was observed in bacterial density by tide in swimming beaches but the bacterial density of sea water at ebb tide was obviously higher than those of the samples at flood tide in the estuary of Suyeong river.

• ALGAE
• /
• v.17 no.2
• /
• pp.95-104
• /
• 2002

Population dynamics of Anabaena and the anatoxin-a concentration were monitored with physicochemical parameters at 3 sites in the lower Naktong River from May to September in 2000. Total 4 species of Anabaena (A. flosaquae, A. smithii, A. ucrainica and A. mucosa) were identified with morphological characterisitcs. Anabaena flos-aquae was most abundant among the populations. The standing crop of Anabaena ranged from 10 to 11,220 cells · $ml^{-1}$ and biomass of Anabaena more 1,000 cells · $ml^{-1}$ was obseved once at St. Mulgeum and St. Seonam, twice at St. Hagueon out of total 9 samplings. There were not significant correlations between the standing crop of Anabaena and other physicochemical parameters such as temperature, nitrate, total nitrogen, phosphate, total phophorus and N/P ratios. The frequency of trichomes with akinetes was low and ranged from 0 to 4% in the total Anabaena population and A. smithii showed highest frequency of 2.8% among all species. The population at St. Seonam showed highest frequency of 1.4% among all sampling sties. The population in September showed the highest frequency of 3.0% among all sampling period. The frequency of trichomes with heterocysts was low and ranged from 1 to 87% inthe total Anabaena population and A. smithii showed highest frequency of 55.1% among all species. The population at St. Mulgeum showed highest frequency of 17.6% among all sampling sites. The population in August showed the highest frequency of 21.4% among all sampling period. The frequency of trichomes with akinetes and/or heterocysts was not related to all the physicochemical parameters of temperature, nitrate, total nitrogen, phosphate, total phosphorus and N/P ratios. The anatoxin-a concentations were determined in algal materials dominated by Microcystis and Anabaena from June though August by derivatization using 7-fluoro-4-nitro-2, 1,3-benzoxadiazole (NBD-F) and HPLC analysis with fluorimetric detection. All the concentrations were below the detection limit of 0.1 ㎍ · $l^{-1}$ in the present study.

• Korean Journal of Microbiology
• /
• v.35 no.1
• /
• pp.41-46
• /
• 1999

In detecting pathogeuic viruses in water sample, polymerase-chain-reaction (PCR) amplification was uscd. In order lo obtain the intact viral particlc, five concentration techniques were compared and an improved procedure was developed with some modifications. Among them, adsorption-elution~EG precipitation and flocculatio~~iultracentriEugation were more efficient than others with thc detection limit of 10 PFU $ml^{-1}$. By the additional step removing inhibitory compounds for PCR reaction, the purity of the concentrated sample was improved and the detection limit was lowered by one order (to 1 PFU $ml^{-1}$. To examine the availability of the optimized procedure for field surveys, the distributions of enterovirus in Han River were estimated using the novel procedure. Seventy-five percentage (618) of sewagc samples and twenty percentage (2110) of river water samples were positive for enterovirus. These results indicate that adsorption-elutionPEG precipitation by PCR method is useful for the prompt and handy monitoring of viral contaminaiton in water environment and pathogenic viruses are widely distributed in water environments of Seoul.