• Journal of Korean Society of Water and Wastewater
• /
• v.35 no.4
• /
• pp.259-275
• /
• 2021

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.

• Journal of Environmental Impact Assessment
• /
• v.26 no.5
• /
• pp.291-302
• /
• 2017

The research was carried out in order to find climate factors which determine the distribution of Eurya japonica, and the potential habitats (PHs) under the current climate and climate change scenario by using species distribution models (SDMs). Four climate factors; the warmth index (WI), the minimum temperature of the coldest month (TMC), summer precipitation (PRS), and winter precipitaion (PRW) : were used as independent variables for the model. Seventeen general circulation models under RCP (Representative concentration pathway) 8.5 scenarios were used as future climate scenarios for the 2050s (2040~2069) and 2080s (2070~2099). Highly accurate SDMs were obtained for E. japonica. The model of distribution for E. japonica constructed by SDMs showed that minimum temperature of the coldest month (TMC) is a major climate factor in determining the distribution of E. japonica. The area above the $-5.7^{\circ}C$ of TMC revealed high occurrence probability of the E. japonica. Future PHs for E. japonica were projected to increase respectively by 2.5 times, 3.4 times of current PHs under 2050s and 2080s. It is expected that the potential of E. japonica habitats is expanded gradually. E. japonica is applicable as indicator species for monitoring in the Korean Peninsula. E. japonica is necessary to be monitored of potential habitats.

• Ecology and Resilient Infrastructure
• /
• v.3 no.3
• /
• pp.189-200
• /
• 2016

The species distribution model would be a useful tool for understanding how invasive alien species spread over the country and what environmental variables contribute to their distributions. This study is focused on the potential distribution of two invasive alien species, the common ragweed (Ambrosia artemisiifolia) and knotgrass (Paspalum distichum) in the Korean Peninsula. The maximum entropy (Maxent) model was used for the prediction of their distribution by inferring their climatic environmental requirements from localities where they are currently known to occur. We obtained their presence data from the Global Biodiversity Information Facility and the Korean plant species databases and bioclimatic data from the WorldClim dataset. As a results of the modelling, the potential distribution predicted by global occurrence data was more accurate than that by native occurrence data. The variables determining the common ragweed distribution were precipitation of the driest month and annual mean temperature. Both annual and the coldest quarter mean temperatures were critical factors in determining the knotgrass distribution. The Maxent model could be a useful tool for the prediction of alien species invasion and the management of their expansion.

• Korean Journal of Environment and Ecology
• /
• v.37 no.6
• /
• pp.450-463
• /
• 2023

The objective of this study is to derive the core habitat of the Kirengeshoma koreana Nakai utilizing Habitat Suitability Index (HSI) and Maximum Entropy (MaxEnt) models. Expert-based models have been criticized for their subjective criteria, while statistical models face difficulties in on-site validation and integration of expert opinions. To address these limitations, both models were employed, and their outcomes were overlaid to derive the core habitat. Five variables were identified through a comprehensive literature review and spatial analysis based on appearance coordinates. The environmental variables encompass vegetation zone, forest type, crown density, annual precipitation, and effective soil depth. Through surveys involving six experts, importance rankings and SI (Suitability Index) scores were established for each variable, subsequently facilitating the creation of an HSI map. Using the same variables, the MaxEnt model was also executed, resulting in a corresponding map, which was merged to construct the definitive core habitat map. Out of 16 observed locations of K. koreana, 15 were situated within the identified core habitat. Furthermore, an area historically known to host K. koreana but not verified in the present, Mt. Yeongchwi, was found to lack a core habitat. These findings suggest that the developed models exhibit a high degree of accuracy and effectively reflect the current ecological landscape.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.29 no.7
• /
• pp.802-811
• /
• 2023

Marine spatial planning (MSP) is a crucial element for rational allocation and sustainable use of marine areas. Particularly, Fishing Activity Protected Areas constitute essential zones accounting for 45.6% designated for sustainable fishing activities. However, the current assessment of these zones does not adequately consider future demands and potential values, necessitating appropriate evaluation methods and predictive tools for long-term planning. In this study, we selected key fish species (Scomber japonicus, Trichiurus lepturus, Engraulis japonicus, and Larimichthys polyactis) within the Fishing Activity Protected Area to predict their distribution and compare it with the current designated zones for evaluating the ability of the prediction tool. Employing the Intergovernmental Panel on Climate Change (IPCC) 6th Assessment Report scenarios (SSP1-2.6 and SSP5-8.5), we used species distribution models (such as MaxEnt) to assess the movement and distribution changes of these species owing to future variations. The results indicated a 30-50% increase in the distribution area of S. japonicus, T. lepturus, and L. polyactis, whereas the distribution area of E. japonicus decreased by approximately 6-11%. Based on these results, a species richness map for the four key species was created. Within the marine spatial planning boundaries, the overlap between areas rated "high" in species richness and the Fishing Activity Protected Area was approximately 15%, increasing to 21% under the RCP 2.6 scenario and 34% under the RCP 8.5 scenario. These findings can serve as scientific evidence for future evaluations of use zones or changes in reserve areas. The current and predicted distributions of species owing to climate change can address the limitations of current use zone evaluations and contribute to the development of plans for sustainable and beneficial use of marine resources.

• Journal of Environmental Impact Assessment
• /
• v.25 no.1
• /
• pp.77-87
• /
• 2016

Predicting possible distributed area of Schisandra chinensis which has long term cultivation period among non-timber forest products is needed to be studied to deal with climate change. Hence, distribution of Schisandra chinensis in the 2050s and 2070s was predicted under two scenario, RCP 4.5 and RCP 8.5, with ensemble of 5 climate models used in IPCC AR5. According to estimation using RCP 4.5, distribution of Schisandra chinensis in 2050s appeared to decrease 43% of current area and appeared to decrease 57% in 2070s respectively. Moreover, According to estimation using RCP 8.5, distribution of Schisandra chinensis in 2050s appeared to decrease 55% of current area and appeared to decrease 85% in 2070s. As a final outcome, Schisandra chinensis was estimated to extinct in the future except Gangwon-do and Gyeongsangbuk-do when analyzing change between current distributed area and future distributed area. As a result, those areas were classified as vulnerable areas to climate change. Therefore, Gangwon-do and Gyeongsangbuk-do were thought to be ideal for growing Schisandra chinensis. The result from this study can be used to provide basic information for selecting proper area of Schisandra chinensis considering climate change effect.

• The Journal of Engineering Geology
• /
• v.12 no.2
• /
• pp.203-214
• /
• 2002

The statistical model was introduced to satisfy various experimental condition on the sorption of heavy metals (Pb, Cu, Cd, and Zn) by bentonite. The Box-Behnken model designed statistically was applied to determine relative impact among three variables such as pH, HCO$_3$ contents and heavy metal concentrations on the sorption. The SAS program was used to obtain the statistical solution. The statistical surface response analysis indicates that initial concentration of heavy metals and pH have an important effect on the sorption, and bicarbonate is not a serious variable. The sorption capability about heavy metals of bentonite is in the order of Pb>Cu>Zn>Cd. The precipitation as hydroxyl and carbonate complexes of heavy metals was thermodynamically analyzed as major mechanism of sorption under alkaline pHs and high bicarbonate solution. It was found that there is a little difference between the model prediction on the precipitation of heavy metals and the results of batch sorption experiment. The thermodynamic data of the programs have to revise to obtain the best fit condition between the model prediction and the experimental results.

• Korean Journal of Plant Taxonomy
• /
• v.52 no.3
• /
• pp.184-195
• /
• 2022

The invasive alien species Brassica tournefortii Gouan (Brassicaceae) is herein reported for the first time in Korea, from Gunsan-si, Gochang-gun, and Jeju-si. Brassica tournefortii can easily be distinguished from B. juncea and B. napus by its dense stiff hairs at the base of the stem and leaves, basally and distally branched stems, partially dehiscent fruits, and seeds that become mucilaginous in the presence of moisture. Although some taxonomists have classified this species as belonging to Coincya Rouy based on its fruit and seed characteristics, the existence of one vein on the fruit valves and our maximum likelihood analysis using internal transcribed spacer sequences placed it in Brassica. Distribution data, photographs, and a description of B. tournefortii are presented herein. Moreover, potential changes in the distribution of B. tournefortii were predicted under different climate scenarios, but our analysis showed that the probability of the spreading of this species is low. Nevertheless, continuous monitoring is necessary for an accurate assessment. The results of the present study can be used to conduct an invasion risk assessment and can assist with the effective management of this invasive alien species.

• Journal of Environmental Impact Assessment
• /
• v.32 no.6
• /
• pp.503-515
• /
• 2023

The disconnection and fragmentation of ecological spaces that occur during the development process pose a significant threat to biodiversity. Urban center areas with high development pressure are particularly susceptible to low connectivity due to a scarcity of ecological space. This issue tends to be more pronounced in larger cities.To address this challenge, continuous efforts are needed to assess and improve the current state of ecological space connectivity at the level of individual projects and urban management. However, there is a lack of discussion regarding the analysis and improvement of ecological connectivity in metropolitan cities In line with this objective, this study evaluated the connectivity of ecological spaces in the city centers of Seoul, Busan, Daegu, Incheon, Gwangju, Daejeon, and Ulsan. The evaluation revealed that city centers exhibited lower connectivity of ecological spaces compared to their peripheries or the overall city. In addition, in the ecological network analysis that reflected regional characteristics, such as the species distribution model conducted on Daejeon, 510 optimal paths connecting forests of more than 1ha were derived. This study is significant as an example of deriving an ecological network based on regional characteristics, including quantitative figures necessary for establishing goals to improve urban ecological connectivity and biodiversity. It is anticipated that the results can be utilized to propose directions for enhancing ecological connectivity in environmental impact assessments or urban management and to establish an evaluation framework.

• Korean Journal of Environment and Ecology
• /
• v.34 no.6
• /
• pp.589-600
• /
• 2020

Effective conservation and management of protected areas require monitoring the settlement of invasive alien species and reducing their dispersion capacity. We simulated the potential distribution of invasive alien plant species (IAPS) using three representative species distribution models (Bioclim, GLM, and MaxEnt) based on the IAPS distribution in the forest genetic resource reserve (2,274ha) in Uljin-gun, Korea. We then selected the realistic and suitable species distribution model that reflects the local region and ecological management characteristics based on the simulation results. The simulation predicted the tendency of the IAPS distributed along the linear landscape elements, such as roads, and including some forest harvested area. The statistical comparison of the prediction and accuracy of each model tested in this study showed that the GLM and MaxEnt models generally had high performance and accuracy compared to the Bioclim model. The Bioclim model calculated the largest potential distribution area, followed by GLM and MaxEnt in that order. The Phenomenological review of the simulation results showed that the sample size more significantly affected the GLM and Bioclim models, while the MaxEnt model was the most consistent regardless of the sample size. The optimal model overall for predicting the distribution of IAPS among the three models was the MaxEnt model. The model selection approach based on detailed flora distribution data presented in this study is expected to be useful for efficiently managing the conservation areas and identifying the realistic and precise species distribution model reflecting local characteristics.