• Korean Journal of Ecology and Environment
• /
• v.43 no.1
• /
• pp.161-167
• /
• 2010

In this study, we implemented an experimental approach of ecological model development in order to emphasize the importance of input variable selection with respect to time-delayed arrangement between input and output variables. Time-series modeling requires relevant input variable selection for the prediction of a specific output variable (e.g. density of a species). Inadequate variable utility for input often causes increase of model construction time and low efficiency of developed model when applied to real world representation. Therefore, for future prediction, researchers have to decide number of time-delay (e.g. months, weeks or days; t-n) to predict a certain phenomenon at current time t. We prepared a total of 3,900 equation models produced by Time-Series Optimized Genetic Programming (TSOGP) algorithm, for the prediction of monthly averaged density of a potamic phytoplankton species Stephanodiscus hantzschii, considering future prediction from 0- (no future prediction) to 12-months ahead (interval by 1 month; 300 equations per each month-delay). From the investigation of model structure, input variable selectivity was obviously affected by the time-delay arrangement, and the model predictability was related with the type of input variables. From the results, we can conclude that, although Machine Learning (ML) algorithms which have popularly been used in Ecological Informatics (EI) provide high performance in future prediction of ecological entities, the efficiency of models would be lowered unless relevant input variables are selectively used.

• Hwankyungkyoyuk
• /
• v.21 no.4
• /
• pp.1-11
• /
• 2008

The purpose of this study is to evaluate an environmental education model school and to devise proper plans for school management through the school eco-audit. The results of this study are as follows. Firstly, environmental education model school is analysed from view point of planning and executing. So that, environmental education model school put emphasis on executing rather than planning. It requires to emphasis planning, for example, establishing environment management system, ecological measures, environment reporting. Secondly, environmental education model school is in the incipient stage as the environment prevention. The preparation for planning is inefficient compared to the ecological pretest of school eco-audit. In the aspect of ecological measures and education, the participant of all the school staff is unsatisfactory. In conclusion, the school eco-audit could be presented as a key role for establishing basic organizational conditions, ecological measures and education for sustainable development, It is necessary to drive environmental education model school to achieve extended effect for sustainable society.

• Korean Journal of Environmental Biology
• /
• v.35 no.1
• /
• pp.37-46
• /
• 2017

Climate change is the biggest concern of the $21^{st}$ century. Greenhouse gas (GHG) emissions from various sectors are attracting attention as a cause of climate change. The DeNitrification-DeComposition (DNDC) model simulates GHG emissions from cropland. To study future GHG emissions using this simulation model, various factors that could change in future need to be considered. Because most problems are from the agricultural sector, DNDC would be unable to solve the factor-changing problem itself. Hence, it is necessary to link DNDC with separate models that simulate each element. Climate change is predicted to cause a variety of environmental disasters in the future, having a significant impact on the agricultural environment. In the process of human adaptation to environmental change, the distribution and management methods of farmland will also change greatly. In this study, we introduce some drawbacks of DNDC in considering future changes, and present other existing models that can rectify the same. We further propose some combinations with models and development sub-models.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.18 no.5
• /
• pp.1-11
• /
• 2015

Jeju Island is managed intensively in terms of environmental and ecological aspect because of its extraordinary ecosystem types comprising numerous rare, protected flora and fauna. To depict rapid change of habitat status in Jeju Island, the InVEST Habitat Quality model has been operated and compared analytically with the Eco-Natural map. The Habitat Quality map of Jeju Island is turned out to have similar inclination with Eco-Natural map. We compared the average habitat quality value in each Eco-natural map class in Jeju Island and the habitat quality value of first second third grade and non-included area decreased as 0.95 0.76, 0.53 and 0.37 in eco natural map respectively. Compared to biodiversity map based on biological investigation, the result of the InVEST habitat quality model can be simply obtained by land cover map with threat and sensitivity data. Further studies are needed to make explicit coefficients for Jeju Island and Korean peninsula, then the Habitat Quality model could be applied to past and future scenarios to analyze extent of habitat degradation in time series to help decision makers.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.44 no.2
• /
• pp.167-172
• /
• 2011

A three-dimensional ecological model (EMT-3D) was applied to Tokyo Bay to simulate 4,4'-bis (2-sulfostyryl)biphenyl (DSBP). The simulated results were in good agreement with the observed values, with a correlation coefficient of R=0.8431 and a coefficient of determination of $R^2$=0.7108. The sensitivity analysis indicated that the photolysis rate is the most important factor. Therefore, the parameters must be considered carefully in modeling. The mass balance results showed that the standing stock of DSBP in water and in particulate organic carbon was 621.2 and 19.5 kg, respectively, and the effluent flux to the open sea was 2.63 and 0.055 kg/day, respectively.

• Journal of Environmental Impact Assessment
• /
• v.21 no.6
• /
• pp.841-857
• /
• 2012

Preliminary ecological environmental assessments including physico-chemical constituents, water quality, fish fauna analysis, physical habitat health, and ecological health assessment were conducted as a primary step for Jeungchon micro-habitat ecosystem restoration in 2012. Water chemistry analysis of conductivity, dissolved oxygen, chlorophyll-a and etc. indicated that there were no significant differences(p < 0.05) among 6 sites between the headwaters and downstream. Multi-metric model analysis of Qualitative Habitat Evaluation Index(QHEI) showed that brooklets were at "good condition" as a mean QHEI of 158.7(n = 6) and the longitudinal differences of the model values between the sites were minor(QHEI range: 153 - 165). Total fish species and the number of individuals were 12 and 481, respectively, and dominant species were Zacco platypus(49.5%) and Zacco koreanus(36.8%). Tolerance guild analysis showed that the proportion of sensitive species($S_S$) had a negative linear function[$S_S=86.35-0.31(D_H)$; $R^2$ = 0.892, p < 0.01] with a distance from the headwaters, while the proportion of tolerant species($T_S$) had a positive linear function($R^2$ = 0.950, F = 90.28, p < 0.001) with the distance. Trophic feeding guild analysis showed that the proportion of insectivore species($I_n$) had a negative linear function($R^2$ = 0.934, p < 0.01) with a distance from the headwaters, while the proportion of omnivore species($O_m$) had a positive linear function($R^2$ = 0.958, p < 0.001) with the distance. Index of Biological Integrity(IBI) model, based on fish assemblages, showed a "fair condition" as a mean IBI of 23(n = 6), and there was a distinct differences of ecological health between the headwaters(S1 = 30; "good condition") and the downstreams(S6 = 14; "poor condition"). Overall, the preliminary environmental impact assessments suggest that water quality, physical habitat conditions(QHEI model), and ecological health(IBI model) were maintained well, even if the state was not an excellent conditions.

• Journal of Wetlands Research
• /
• v.21 no.1
• /
• pp.16-26
• /
• 2019

Wetlands that provide diverse ecosystem services, such as habitat provision and hydrological control of flora and fauna, constitute ecosystems through interaction between wetlands existing in a wetlandscape. Therefore, to evaluate the wetland functions such as resilience, it is necessary to analyze the ecological connectivity that is formed between wetlands which also show hydrologically dynamic behaviors. In this study, by defining wetlands as ecological nodes, we generated ecological networks through the connection of wetlands according to the dispersal model of wetland species. The characteristics of these networks were then analyzed using various network metrics. In the case of the dispersal based on a threshold distance, while a high local clustering is observed compared to the exponential dispersal kernel and heavy-tailed dispersal model, it showed a low efficiency in the movement between wetlands. On the other hand, in the case of the stochastic dispersion model, a low local clustering with high efficiency in the movement was observed. Our results confirmed that the ecological network characteristics are completely different depending on which dispersal model is chosen, and one should be careful on selecting the appropriate model for identifying network properties which highly affect the interpretation of network structure and function.

• Journal of Korean Society on Water Environment
• /
• v.22 no.5
• /
• pp.796-804
• /
• 2006

The objective of this research was to develop ecological multi-metric models using natural fish assemblages for a diagnosis of current stream health condition, and apply the model to nationwide lotic ecosystems of the Geum River, the Youngsan River, and the Sumjin River. The ecological stream health model was based on the index of biological integrity (IBI), which was originally developed in North American streams by Karr (1981), and the Rapid Bioassessment Protocol (RBP), which was scientifically established by the US EPA (1999). The metric numbers and metric attributes were partially changed for the regional applications, so the scoring criteria was modified for the assessment. Overall, metric values, based on the IBI calculations, reflected conventional water quality characteristics, based on nutrient regime, and agreed with results of staticeco-toxicity tests. Some stations impaired in terms of stream health were identified by the IBI approach, and also major key stressors affecting the stream health were identified by additional evaluations of physical habitats. Our preliminary results suggested that biological integrity in stream ecosystems was largely disturbed by habitat degradation as well as chemical pollutions. This new approach would be used as a key tool for ecological restorations and species conservations in the degraded aquatic ecosystems in Korea and applied for elucidating major causes of ecological disturbances. Ultimately, this approach provides us an effective management strategy of stream ecosystems through establishments of ecological networks in various watersheds.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.35 no.5
• /
• pp.56-63
• /
• 2007

This study used the nearest feature model to connect forest patches within the western Geum River Basin. Due to many different forest patch sizes, 3 alternative methods were tested to determine the best way to establish an ecological network with forest patches. Alternative 1 used all forest patches to determine whether patches were large enough. Alternative 2 used forest patches over 10 ha in size. Alternative 3 used natural conservation indices to select forest patches containing better qualities in the natural conservation level. As a result 635 out of 724 patches of over 10 ha were selected for comparison. Alternative 1 showed that forest patches of less than 10 ha were outliers interrupting the establishment of the ecological network. They generated an unnecessary ecological network to link core areas to comparison features. The ecological network was improved by using forest patches greater than 10 ha in size(Alternative 2). Each comparison feature was much more hierarchically connected to core areas in Alternative 2 than in Alternative 1. Forest patches filtered by natural conservation indices were useful for obtaining the best ecological network. Alternative 3 clearly showed the connections in the ecological network between core areas and forest.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.28 no.6
• /
• pp.1-15
• /
• 2001

The purpose of this study is to figure out the structural characteristics of urban plant community and suggest restoration model of Quercus mongolica in the case of Korean national Capital Region. The investigation areas were selected from urban area of Mt. Nam at Chung-Gu, suburban areas of Mt. Bong at Eunpyoung-GU, Mt. Sungju at Buchon City and non-urban areas of Mt. Suri at Kunpu City and Mt. Chonma t namyangju-City. After the main study field had been classified into the evaluation of the ecological characteristics and the modeling of the vegetation. We analyzed to evaluate the ecological characteristics of the forest structure -- successional stage, naturalness, multi-layer structure of the forest and species diversity, and the plant community structures. We have proposed vegetation restoration model based on the selection of proper plants, the number of individuals, diameter short area of breast height, the shortest distance between plants in non-urban area. As for successional stage, It was judged that the ecological succession may not be followed like the present stage of the surveyed areas in urban, suburban and non-ruban areas. As for the retention of naturalness and multi-layer structures of vegetation, In Quercus mongolica community, Robinia pseudo-acacia and Ailanthus altissima occurred in each layers at Mt. Nam, Mt. Bong and Mt. Sungju, and Eupatorium rugosum occurred in herbaceous layer at Mt. Nam. Consequently, the ecological restoration plan following the structure of the vegetation in Mt. Chonma seemed to be advisable in Q. mongolica community, there were less number of species and individuals in urban areas than those of non-urban areas. Planting of trees following the simulated native plant community of non-urban areas seemed to be required to promote the plants in urban areas. Considering the number of individuals up to three layers in each 400$m^2$ area, it was composed of twenty nine in canopy layer, forth nine in understory layer, 367 in shrub layer and 33.7% herbaceous ground cover in the Q.mongolica community. The suggested restoration model in this study is nan applicable model for the introduction in the cities, and this study shows that continuous experiments and field investigation on this model should be performed in the future.