This study explored the usefulness and implications of the Bayesian hyperparameter optimization in developing species distribution models (SDMs). A variety of machine learning (ML) algorithms, namely, support vector machine (SVM), random forest (RF), boosted regression tree (BRT), XGBoost (XGB), and Multilayer perceptron (MLP) were used for predicting the occurrence of four benthic macroinvertebrate species. The Bayesian optimization method successfully tuned model hyperparameters, with all ML models resulting an area under the curve (AUC) > 0.7. Also, hyperparameter search ranges that generally clustered around the optimal values suggest the efficiency of the Bayesian optimization in finding optimal sets of hyperparameters. Tree based ensemble algorithms (BRT, RF, and XGB) tended to show higher performances than SVM and MLP. Important hyperparameters and optimal values differed by species and ML model, indicating the necessity of hyperparameter tuning for improving individual model performances. The optimization results demonstrate that for all macroinvertebrate species SVM and RF required fewer numbers of trials until obtaining optimal hyperparameter sets, leading to reduced computational cost compared to other ML algorithms. The results of this study suggest that the Bayesian optimization is an efficient method for hyperparameter optimization of machine learning algorithms.

본 연구는 한국 남서해안 갯벌 중 유부도, 곰소만, 임자도, 여자만 갯벌의 저서돌말류 종조성 및 생물량을 알아보기 위해 2016년 8월과 9월에 현장조사를 실시하였다. 유부도 갯벌에서 저서돌말류는 48속 181종, 곰소만 갯벌에서 51속 194종, 임자도 갯벌에서 64속 224종, 여자만 갯벌에서 56속 188종이 출현하였으며, 총 70속 274종이 출현하였다. 분류군 가운데 중심돌말목보다 우상돌말목이 다양하게 출현하였고 우점종은 중심돌말목의 Paralia sulcata와 Navicula spp.였다. 유부도 갯벌의 생물량은 정점별로 18.8~136.1 mg m-2, 곰소만 갯벌 31.9~215.7 mg m-2, 임자도 갯벌 2.9~120.2 mg m-2, 여자만 갯벌 10.1~147.7 mg m-2, 4개 지역의 총 생물량 범위는 2.9~215.7 mg m-2였다. 엽록소 a의 분해 산물인 phaeopigment 농도로는 유부도 갯벌에서 1.7~470.8 mg m-2, 곰소만 갯벌에서 52.3~277.2 mg m-2, 임자도 갯벌에서 0.6~78.9 mg m-2, 여자만 갯벌에서 39.1~346.3 mg m-2으로 나타났다. 조사 대상지역의 종 다양성과 생물량은 높은 경향을 나타내었다. 한국의 남서해안은 다양한 저서돌말류가 서식하며 엽록소 a의 높은 농도는 일차생산력을 결정하는 주요인자다.

Samples were collected from benthic and subaerophytic habitats of fresh and brackish water in Korea from April 2017 to April 2018. Accordingly, three genera and seven species were newly recorded in Korea. Three unrecorded genera were Cephalothrix, Cyanophanon and Toxifilum, and seven species were Anagnostidinema pseudacutissimum, Calothrix elenkinii, Cephalothrix komarekiana, Cyanophanon mirabile, Cyanosarcina chroococcoides, Leptolyngbya ectocarpi and Toxifilum mysidocida. Anagnostidinema pseudacutissimum and Cephalothrix komarekiana had been previously reported to inhabit freshwater, but these were found to be present in brackish water during the period of the present study. Also, Cyanosarcina chroococcoides previously reported to inhabit moorland waters and swamps, were found to be present in subaerophytic habitat.

본 연구에서는 동해 북부지역 거진, 중부지역 죽변 및 남부지역 감포의 조간대에 분포하고 있는 해조상의 지리적 종조성과 우점종 및 서식 특성을 2014년 5월과 8월에 조사하였다. 해조상의 지리적 종조성을 보면 거진, 죽변 및 감포는 모두 온대성 해조상의 특성으로서 거진은 구멍갈파래(Ulva pertusa), 염주말(Chaetomorpha moniligera)이 우점하였고, 죽변은 염주말(C. moniligera), 개서실(Chondria crassicaulis)이 우점하였다. 그리고 감포는 지충이(Sargassum thunbergii), 구멍갈파래(U. pertusa), 솜털류(Ectocarpus sp.)가 우점하였다. 지역별 해조상을 대표적으로 나타낼 수 있는 C/P, R/P와 (R+C)/P의 비율에서 거진은 각각 0.85, 2.10, 2.94, 죽변은 각각 0.58, 3.15, 3.73 그리고 감포는 각각 0.80, 3.91, 4.71로 전체 지역이 온대성 특성의 해조상이 뚜렷하게 나타났다. 군집분석의 결과에 의하면, 이들 지역은 거진의 A 그룹과 죽변과 감포 지역의 B 그룹으로 나누어졌으며, 이것은 지리적 종조성에서 죽변과 감포는 거진보다 온대성 해조상 특성이 강하게 반영되어 나타났다.

Community structures and species composition in benthic crustacean around Oryukdo of Busan was investigated using a SCUBA diving in 2011. A total 115 species, 51 families, and 4 taxa were collected during the sampling periods. Species were included 52 species in amphipoda, 49 in decapoda, 8 in isopoda and 6 in thoracica. On cluster analysis, the benthic crustacea community was classified into three groups based on the species composition: group A occurred in spring, group B in autumn and winter and group C in summer. Species composition in benthic crustacean corresponded with the water temperature.

An Individual-Based Model (IBM) was developed by employing natural and toxic survival rates of individuals to elucidate the community responses of benthic macroinvertebrates to anthropogenic disturbance in the streams. Experimental models (doseresponse and relative sensitivity) and mathematical models (power law and negative exponential distribution) were applied to determinate the individual survival rates due to acute toxicity in stressful conditions. A power law was additionally used to present the natural survival rate. Life events, covering movement, exposure to contaminants, death and reproduction, were simulated in the IBM at the individual level in small (1 m) and short (1 week) scales to produce species abundance distributions (SADs) at the community level in large (5 km) and long (1~2 years) scales. Consequently, the SADs, such as geometric series, log-series, and log-normal distribution, were accordingly observed at severely (Biological Monitoring Working Party (BMWP⁄10), intermediately (BMWP⁄40) and weakly (BMWP›50) polluted sites. The results from a power law and negative exponential distribution were suitably fitted to the field data across the different levels of pollution, according to the Kolmogorov-Smirnov test. The IBMs incorporating natural and toxic survival rates in individuals were useful for presenting community responses to disturbances and could be utilized as an integrative tool to elucidate community establishment processes in benthic macroinvertebrates in the streams.

마산만 조간대에서 서식하는 해조류 군집의 종조성 및 분포패턴을 2007년도에 계절별로 조사하였다. 조사기간 동안 총 42종의 해조류가 서식하였으며, 이 중 녹조류 5종, 갈조류 8종, 그리고 홍조류가 29종으로 나타났다. 우점종은 Enteromorpha linza, Ulva pertusa, Sargassum thunbergii, Gelidium amansii, G. divaricatum, Corallina pilulifera, Gracilaria textorii, 그리고 Polysiphonia morrowii으로 나타났다. 특히 우점종들은 수직분포가 뚜렷하였는데, 조간대 상부에는 U. pertusa, G. divaricatum가, 중부에는 E. intestinalis, E. linza, S. thunbergii, G. divaricatum이 우점하였다. 한편 하부에는 U. pertusa, Undaria pinnatifida, S. thunbergii, G. amansii, G. divaricatum, C. pilulifera이 주로 출현하였다. 조사기간 동안 출현한 출현종수는 계절 및 정점에 따라 변하였다. 겨울에 가장 많은 38종이 출현하였으며, 가을에는 28종으로 가장 적었다. 정점별로는 정점 4, 6에서 가장 많았으며, 정점 1, 2에서 가장 낮게 나타났다. 해조류 군집의 공간적 분포를 파악하기 위한 군집분석을 이용한 수지도 및 MDS 분석 결과에서는 해조류의 종조성 및 현존량이 각 정점의 지형적인 차이 및 조류, 해수의 움직임과 같은 물리해양학적인 특징에 의한 것으로 나타났다.

Benthic macroinvertebrate communities were collected in various streams in agricultural and urban areas in Korea. Community structure and species abundance distribution in benthic macroinvertebrates were comparatively analyzed to reveal ecological states pertaining to the sample sites. Species richness in the total macroinvertebrate communities decreased in the gradient along with increase in the level of pollution. Some selected taxa were present in communities collected in the streams in agricultural areas. The rank-abundance curves were also characteristic in revealing community structure according environmental impact. Patterns of species abundance were efficient in revealing ecological states of the sample sites where physico-chemical indicators may not be easily differentiated in stressful conditions. The feasibility of ecological informatics was further discussed in revealing ecological water quality.

Benthic macroinvertebrates were collected with a Surber sampler in streams across different levels of pollution in Korea. Abundance data in different species were analyzed according to the models used in species abundance distribution (SAD), covering geometric series, log series and log normal model. Different patterns in SADs were observed according to the states pertaining to the sample sites. While the log normal patterns were broadly applied to the collected samples, the other models (e.g., geometric series) were selectively observed. The slopes of rank abundance also accordingly reflected the degree of pollution. Application of the theoretical models was additionally discussed in revealing the states of communities. Efficiency of SADs was proposed in community monitoring in response to stressful conditions in aquatic ecosystems.