• 한국어업기술학회:학술대회논문집
• /
• 한국어업기술학회 2003년도 춘계 수산관련학회 공동학술대회발표요지집
• /
• pp.129-131
• /
• 2003

Korea has experienced a Cochlodinium polykrikoides red tide 10 out of the last 10 years (1993-2002). This species of red tide has been experienced at least once by all the southern coast of the Korean peninsular and has been transported up the southeastern coast all the way to the northeastern coast since 1995. The impression is that red tide is spending and becoming more common not only in the nearest coastal water, but also in the offshore. (omitted)

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2009년도 춘계학술발표논문집
• /
• pp.839-841
• /
• 2009

Dinoflagellates are known to cause red tide outbreaks and even to produce toxin. Recently, red tide events have frequently occurred in several embayments of the Korean coast and have brought serious damage to inshore fisheries. Thus, the red tide research activities including the taxonomy as well as distribution of toxic dinoflagellates have received ever increasing attention in Korean waters. Therefore, it is necessary to conduct an extensive taxonomical study on red tide organisms in coastal zone of Korea. The present study is to clarify the fine structures of Prorocentrum spp. and to describe each species with taxonomical notes for plankton monitoring network.

• 한국수산과학회지
• /
• 제50권2호
• /
• pp.222-226
• /
• 2017

It is well known that typhoons strongly influence marine ecosystems. For example, red tides nearly disappear after the passage of typhoons, although the physical or biological mechanism underlying this has not been elucidated. Here, a particle tracking model is executed in a three-dimensional primitive equation model to understand the process of red tide extinction after the passage of a typhoon. Red tide organisms may be regarded as tracers because they have limited mobility and thus their behavior is governed entirely by currents. Initially, tracers are randomly scattered within a limited area, and their spatial and temporal behavior is tracked during and after the passage of a typhoon. This model suggests that the extinction of red tides is significantly influenced by momentum disturbances caused by the typhoon.

• 한국지리정보학회지
• /
• 제9권4호
• /
• pp.129-141
• /
• 2006

인공위성을 이용한 원격탐사 기술의 비약적인 발전과 함께 지리, 해양 정보 등 사회전반에서 사용되는 영상 데이터량이 급속히 증가하고 있다. 따라서 대용량 원격탐사 영상의 해석을 위해서는 육안 검사보다 영상처리 기술을 이용한 자동화 방법이 필요하다. 본 연구에서는 인공위성 원격탐사 영상의 적조영역에 대해 GLCM(Gray Level Co-occurrence Matrix)을 이용하여 질감 정보를 취득하고, 이 데이터로부터 주성분 분석을 통해 적조영역을 자동으로 검출하는 방법에 대해 제안하였다. 기존의 적조영역 검출은 원격탐사 영상의 해색(sea color) 한 가지 특징에 의한 방법이 대부분이었으나 본 연구에서 GLCM의 질감 정보 8가지를 이용해서 2개의 주성분 누적 영상으로 변환시켰다. 연구결과 2개의 주성분 누적 영상의 백분율 분산 값은 90.4%였으며, 이를 해색 한 가지만을 이용한 적조영역 검출방법과 비교했을 때 보다 나은 결과를 나타내었다.

• ALGAE
• /
• 제32권2호
• /
• pp.139-153
• /
• 2017

The ability of a red tide species to take up nutrients is a critical factor affecting its red tide dynamics and species competition. Nutrient uptake by red tide species has been conventionally measured by incubating nutrient-depleted cells for a short period at 1 or 2 light intensities. This method may be applicable to certain conditions under which cells remain in oligotrophic water for a long time and high nutrients are suddenly introduced. Thus, a new method should be developed that can be applicable to the conditions under which cells are maintained in eutrophicated waters in healthy conditions and experience light and dark cycles and different light intensities during vertical migration. In this study, a new repletion method reflecting these conditions was developed. The nitrate uptake rates of the red tide dinoflagellate Prorocentrum micans originally maintained in nitrate repletion and depletion conditions as a function of nitrate concentration were measured. With increasing light intensity from 10 to $100{\mu}E\;m^{-2}s^{-1}$, the maximum nitrate uptake rate ($V_{max}$) of P. micans increased from 3.6 to $10.8 pM\;cell^{-1}d^{-1}$ and the half saturation constant ($K_{s-NO3}$) increased from 4.1 to $6.9{\mu}M$. At $20{\mu}E\;m^{-2}s^{-1}$, the $V_{max}$ and $K_{s-NO3}$ of P. micans originally maintained in a nitrate repletion condition were similar to those maintained in a nitrate depletion condition. Thus, differences in cells under nutrient repletion and depletion conditions may not affect $K_{s-NO3}$ and $V_{max}$. Moreover, different light intensities may cause differences in the nitrate uptake of migratory phototrophic dinoflagellates.

• ALGAE
• /
• 제19권1호
• /
• pp.7-14
• /
• 2004

Blue-Min was initially developed as an adsorbent for harmful gas removal and recently improved to apply to livestock, agriculture and aquaculture as an assistant feed. In the Blue-Min treatment, growth of harmful algae (Cochlodinium polykrikoides and the others causing the red-tide in the ocean) were inhibited below 10% in comparison with control and coagulation removal of harmful alge with Blue-Min treatment was more efficient than that of yellow loess treatment. It would be expected that the Ble-Min can be useful for the extirpator against the red-tide organisms and restrain the toxic algal growth around the fish aquaculture using the assistant feed. Recently, its utility has become to be diverse as it was revealed that aquaculture productivity increase by its application and, in addition, that it improve the water quality or sediment conditions in the aquaculture of Chinese White Shrimp. When Blue-Min was treated with the proper dose, the growth inhibition of Microcystis aeruginosa and lsochrysis galbana, which are typical red-tide organisms in freshwaters and food organisms in aquaculture, respectively, were less than that of marine red-tide organisms, while their growth slightly increased with low concentration treatiment. In addition, polyunsaturated fatty acids (PUFA) content of I. galbana slightly increase with the Blue-Min treatment. Through our research, the Blue-Min has diverse and comples function against various biological organisms and is proved as a biological activator or depressor.

• 한국전자통신학회논문지
• /
• 제14권6호
• /
• pp.1161-1170
• /
• 2019

본 연구에서는 심층 신경망을 이용하여 Cochlodinium polykrikoides 적조 발생을 예측하는 모델을 제안한다. 적조 발생 예측을 위해 8개의 은닉층을 가진 심층 신경망을 구축하였다. 위성 재분석 자료와 기상수치모델 자료를 이용하여 과거 적조 발생해역의 해양 및 기상인자 총 59개를 추출하여 신경망 모델 학습에 활용하였다. 전체 데이터셋 중 적조 발생 사례는 적조 미발생 사례에 비해 매우 적어 불균형 데이터 문제가 발생하였다. 본 연구에서는 이를 해결하기 위해 과표집화(Over sampling) 기반 데이터 증식(Data augmentation) 기법을 적용하였다. 과거자료를 활용하여 모형의 정확도를 평가한 결과 약 97%의 정확도를 보였다.

• 대한원격탐사학회지
• /
• 제32권6호
• /
• pp.673-679
• /
• 2016

본 연구에서는 Geostationary Ocean Color Imager (GOCI)의 수출광량(Water-leaving radiance) 자료와 흡광계수(Absorption Coefficients) 자료를 이용하여 한국 남해에 발생한 Cochlodinium polykrikoides 적조를 탐지하는 방법을 제안한다. C. polykrikoides의 분광스펙트럼을 분석하여, 412 nm에서 555 nm구간 파장대의 수출광 및 흡광 특성을 확인하였고 이를 이용하여 적조지수를 개발하였다. 본 연구에서 제안한 탐지기법은 광학적으로 복잡한 남해에서 발생한 적조를 탐지하였으며, 이러한 결과를 바탕으로 향후 적조 방재에 활용될 수 있을 것이라 판단된다.

• The Korean Journal of Ecology
• /
• 제9권3호
• /
• pp.149-160
• /
• 1986

Relationship between the causative organi는 of red-tide and environmental factors had been ecologically dealt wtih. The surveys were conducted at seven station in Jinhae Bay from July to September 1981. The water temperature and salinity had wide range, i.e. 23.3~29.3$^{\circ}C$ and 19G.78~31.29$\textperthousand$, but several chemical factors remarkably fluctuated; dissolved oxygen 102.9~210.4%, COD 2.10~8.96mg$O_2$/l, pH 8.1~9.4, $NO_3$-N trace~1, 052$\mu$g/l, $PO_4$-P 0.6~58.9$\mu$g/l and chlorophyll-a 2.18~290.5mg/$m^3$ in the observed area. The red-tide was mainly caused by two dinoflagellata taxa throughout major outbreaks occurred in July through September. Leading species of red-tide were Gymnodinium nagasakiense belong to the ajor species. During the surveyed period, cell nubers of the causative organisms of the red-tide extensively varied from 3${\times} /10^4$ cells/l to $1.5\times10^7 $cells/l with moths and stations; Prorocentrum spp. 0.3~12.5$\times\10^5$ cells/l in July; Gymnodinium nagasakiense 0.2~5.9$\times10^6 cells/l, 1.1~4.7$\times10^6$ cells/l, and 0.2~15.1$\times10^6$ cells/l in July, August, and September, respectively. Gymnodinium nagasakiense red-tide seemed to be caused by the high water temperature in summer, unusually low salinity due to heavy rains, and the concentrated nutrients for phytoplankton supplied with the municipal sewages from the urban areas and the wastewaters from the industrial complexes.

• 한국전자통신학회논문지
• /
• 제16권6호
• /
• pp.1363-1372
• /
• 2021

유해조류의 대번식으로 정의되는 적조는 광역적으로 발생·확산되는 특성을 가진다. 이는 기존의 조사 방법만으로는 탐지의 한계가 있다. 따라서 본 연구에서는 적조를 원격탐사 기법을 활용하여 탐지하였다. 또한 단순히 chlorophyll의 농도가 아닌 광특성을 이용하여 탐지의 정확도를 높이고자 하였다. 적조는 해수신호가 복잡한 남해안에서 주로 발생하며 남해안의 주 적조 종은 Cochlodinium polykirkoides이다. 따라서 기계학습 기법을 활용하여 시각적인 판단에 국한되지 않고 연구자의 관찰과 경험에 의존해 발견하지 못했던 특징을 반영하여 객관성을 확보하고자 하였다. 본 연구에서는 기계학습 모델 중에서 서포트백터머신과 랜덤포레스트를 사용하였고 두 모델의 성능 평가 지표로 정확도 등을 산출한 결과 각각 85.7% 80.2%의 정확도를 보였다.