• Journal of Ecology and Environment
• /
• v.46 no.4
• /
• pp.292-303
• /
• 2022

Background: Climate change significantly influences the geographical distribution of plant species worldwide. Selecting indicator species allows for better-informed and more effective ecosystem management in response to climate change. The Korean Peninsula is the northernmost distribution zone of warm temperate evergreen broad-leaved (WTEB) species in Northeast Asia. Considering the ecological value of these species, we evaluated the current distribution range and future suitable habitat for 13 WTEB tree species designated as climate-sensitive biological indicator species. Results: Up-to-date and accurate WTEB species distribution maps were constructed using herbarium specimens and citizen science data from the Korea Biodiversity Observation Network. Current northern limits for several species have shifted to higher latitudes compared to previous records. For example, the northern latitude limit for Stauntonia hexaphylla is higher (37° 02' N, Deokjeokdo archipelago) than that reported previously (36° 13' N). The minimum temperature of the coldest month (Bio6) is the major factor influencing species distribution. Under future climate change scenarios, suitable habitats are predicted to expand toward higher latitudes inland and along the western coastal areas. Conclusions: Our results support the suitability of WTEB trees as significant biological indicators of species' responses to warming. The findings also suggest the need for consistent monitoring of species distribution shifts. This study provides an important baseline dataset for future monitoring and management of indicator species' responses to changing climate conditions in South Korea.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.8
• /
• pp.347-354
• /
• 2022

Since the era of the 4th industrial revolution has been ripe, the use of artificial intelligence(AI) based on massive data is beginning to be actively applied in various fields. However, as the process of analyzing forest species is carried out manually, many errors are occurring. Therefore, in this paper, about 60,000 pieces of AI learning data were automatically analyzed for pine, larch, conifer, and broadleaf trees of aerial photographs and pseudo images in the metropolitan area, and an AI model was developed to distinguish tree species. Through this, it is expected to increase in work efficiency by using the tree species division image as basic data when producing forest change detection and forest field topics.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.26 no.5
• /
• pp.33-46
• /
• 2023

In this study, carbon storage in the aboveground biomass, litter layer, and soil layer was calculated for abandoned mining restoration areas to determine the level of carbon storage after the restoration project through comparison with the ecological reference. Five survey sites were selected for each abandoned mining restoration area in Boryeong-si, Chungcheongnam-do, and the ecological reference that can be a goal and model for the restoration project. The carbon storage in the restoration area was 0~21.3Mg C ha^-1, the deciduous layer 3.3~6.0Mg C ha^-1, and the soil layer(0-30cm) 8.3~35.1Mg C ha^-1, showing a significant difference in carbon storage by target site. The total carbon storage was between 6.1 and 35.3% of the ecological reference, with restoration area ranging from 14.0 to 62.4 Mg C ha^-1. The total carbon storage in the restoration area and the ecological reference differed the most in the aboveground biomass and was less than 12%. Based on these results, forest restoration area need to improve the carbon storage of forests through continuous management and monitoring so trees can grow and restore productivity in the early stages of the restoration project. The results of this study can be used as primary data for preparing future forest restoration indicators by identifying the storage of abandoned mining restoration areas.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.5
• /
• pp.999-1008
• /
• 2023

Early prediction of chronic diseases such as diabetes is an important issue, and improving the accuracy of diabetes prediction is especially important. Various machine learning and deep learning-based methodologies are being introduced for diabetes prediction, but these technologies require large amounts of data for better performance than other methodologies, and the learning cost is high due to complex data models. In this study, we aim to verify the claim that DNN using the pima dataset and k-fold cross-validation reduces the efficiency of diabetes diagnosis models. Machine learning classification methods such as decision trees, SVM, random forests, logistic regression, KNN, and various ensemble techniques were used to determine which algorithm produces the best prediction results. After training and testing all classification models, the proposed system provided the best results on XGBoost classifier with ADASYN method, with accuracy of 81%, F1 coefficient of 0.81, and AUC of 0.84. Additionally, a domain adaptation method was implemented to demonstrate the versatility of the proposed system. An explainable AI approach using the LIME and SHAP frameworks was implemented to understand how the model predicts the final outcome.

• Fisheries and Aquatic Sciences
• /
• v.27 no.2
• /
• pp.87-99
• /
• 2024

Global climate change, followed by an increase in anthropogenic activities in aquatic ecosystems, and species invasions, has resulted in a decline in aquatic organism biodiversity. The Batanghari River, Sumatra's longest river, is polluted by mercury-containing illegal gold mining waste (PETI), industrial pollution, and domestic waste. Several studies have provided evidence suggesting a decline in fish biodiversity within the Batanghari River. However, a comprehensive evaluation of the present status of biodiversity in this river is currently lacking. The species under investigation were identified through various molecular-based identification methods, as well as morphological identification, which involved the use of neighbor-joining (NJ) trees. All collected specimens were initially identified using morphological techniques and subsequently confirmed with molecular barcoding analysis. Morphological and DNA barcoding identification categorized all specimens (1,692) into 36 species, 30 genera and 16 families, representing five orders. A total of 36 DNA barcodes were generated from 30 genera using a 650-bp-long fragment of the mitochondrial cytochrome oxidase subunit I (COI) gene. Based on the Kimura two-parameter model (K2P), The minimum and maximum genetic divergences based on K2P distance were 0.003 and 0.331, respectively, and the average genetic divergence within genera, families, and orders was 0.05, 0.12, 0.16 respectively. In addition, the average interspecific distance was approximately 2.17 times higher than the mean intraspecific distance. Our results showed that the COI barcode enabled accurate fish species identification in the Batanghari River. Furthermore, the present work will establish a comprehensive DNA barcode library for freshwater fishes along Batanghari River and be significantly useful in future efforts to monitor, conserve, and manage fisheries in Indonesia.

• Journal of Korean Society of Forest Science
• /
• v.105 no.2
• /
• pp.247-252
• /
• 2016

This study was conducted to evaluate the technical efficiency of Annual Chestnut production in South Korea. In this study, technical efficiency is the maximum possible production for which a certain amount of costs is inputted. For analysis on the technical efficiency we used output-oriented BCC Model, and then we analyzed correlation among input costs, production, gross income, net income, and market price per unit in order to determine the cause of variation in the technical efficiency. As study materials, we used statistics for the forestry production costs for 7 years from 2008 to 2014. The study results showed that the maximum possible production and actual production in 2008, 2009, and 2010 were 1,568 kg, 1,745 kg, and 1,534 kg by hectares in the order which were the same values. Consequently, the technical efficiency of those was all evaluated as 1.00. On the other hand, actual production from 2011 to 2014 was 1,270 kg 1,047 kg, 1,258 kg, and 1,488 kg by hectares in the order and the maximum possible production was 1,524 kg, 1,467 kg, 1,635 kg, and 1,637 kg by hectares in the analysis. From those values, the technical efficiency was evaluated in the following order:0.83, 0.71, 0.75, 0.91. The lowest value of the technical efficiency was 0.71 in 2012, and the values of this increased gradually since 2013. It is indicated that the cause of variation in the technical efficiency was related to the relationship between production and market price, and there was a negative correlation with r = -0.821 (p<0.05). The level of maximum available production per unit area was between 1,488kg in lower limit and 1,745 kg in upper limit, and the average was turned out as 1,548 kg.

• Korean Journal of Environment and Ecology
• /
• v.18 no.4
• /
• pp.369-381
• /
• 2004

This study was achieved to present the way to restore the Seonheulgot pasture-grassland damaged by landuse and interference for a long time to evergreen broad-leaved forests as the native vegetation structure. As a result of analyzing ecological succession tendency of structure in survey area, we established the optimal restoration model. The total of survey sites were 26, and the classified plant community types were four types by M.I.P of dominant woody species. Finally we classified the four types based on diameter of dominant woody species in canopy layer. The six community types are as follows: Community I was runner-shrub forest, community II was evergreen broad-leaved shrub forest, and community III was evergreen broad-leaved forest of small diameter. Community IV and V were evergreen broad-leaved forest of middle diameter. Community Ⅵ was evergreen broad-leaved forest of large diameter. The number of constituent species was 24 in community I, 28 in community II as the shrub forest, 16 as the evergreen broad-leaved forest of small diameter, 29 in community III, 30 in community IV as the evergreen broad-leaved forest of middle diameter and 27 in community Ⅵ as the evergreen broad-leaved forest of large diameter. The range of Shannon's index of all communitys was from 0.8763 to 1.2630 and the Similarity index between the community composed of middle diameter woody species and large diameter woody species. The ecological succession of community I, II, and III were changed from pasture-grassland to broad-leaved forest and the structure of community IV, V, and Ⅵ was similar to evergreen broad-leaved forest in warm temperate region. We suggest the restoration planting model evergreen broad-leaved forest of in Seonheulgot pasture-grassland, as follows: The target restoration vegetation were Castanopsis cuspidata var. sievoldii community and Queycus glauca community. Castanopsis cuspidata var. sievoldii and Quercus glauca should be dominant woody species in canopy layer, the number of trees was 10 per 100$m^2$, and Castanopsis cuspidata var, sievoldii, Quercus glauca, Camellia japonica, and Eurya japonica should be dominant woody species in the understory layer, the number of trees was 14 per 100$m^2$.

• Journal of Korean Society of Forest Science
• /
• v.109 no.3
• /
• pp.259-270
• /
• 2020

Extreme weather events, such as heat and drought, have occurred frequently over the past two decades. This has led to continuous reports of cases of forest damage due to physiological stress, not pest damage. In 2014, pine trees were collectively damaged in the forest genetic resources reserve of Sogwang-ri, Uljin, South Korea. An investigation was launched to determine the causes of the dieback, so that a forest management plan could be prepared to deal with the current dieback, and to prevent future damage. This study aimedto 1) understand the topographic and structural characteristics of the area which experienced pine tree dieback, 2) identify the main causes of the dieback, and 3) predict future risk areas through the use of machine-learning techniques. A model for identifying risk areas was developed using 14 explanatory variables, including location, elevation, slope, and age class. When three machine-learning techniques-Decision Tree, Random Forest (RF), and Support Vector Machine (SVM) were applied to the model, RF and SVM showed higher predictability scores, with accuracies over 93%. Our analysis of the variable set showed that the topographical areas most vulnerable to pine dieback were those with high altitudes, high daily solar radiation, and limited water availability. We also found that, when it came to forest stand characteristics, pine trees with high vertical stand densities (5-15 m high) and higher age classes experienced a higher risk of dieback. The RF and SVM models predicted that 9.5% or 115 ha of the Geumgang Pine Forest are at high risk for pine dieback. Our study suggests the need for further investigation into the vulnerable areas of the Geumgang Pine Forest, and also for climate change adaptive forest management steps to protect those areas which remain undamaged.

• Journal of Korean Society of Forest Science
• /
• v.101 no.1
• /
• pp.36-45
• /
• 2012

Education on the environment and plants offered by arboretum for today's people not only contribute to foster a better natural environment in urban region but also provide visitors with decent refreshment environment and beyond. In the study, the author undertook the observation on usage behavior and satisfaction model of arboretum visitors expect and investigated the facilities and programs to be offered by arboretum in order to propose the opinion regarding the service. For observation size of variables in a multiple regression analysis of variables is influencing satisfaction rankings walks the line of flow, the educational effect on the environment, cleanliness of the facility, visits pay, natural beauty, diversity of trees, accessibility and friendliness of staff, expansion of facilities in the arboretum and appeared as a complement. In case of visitor attribute, the residents living near the facility showed the highest visit frequency of more than 5 times, especially as part of taking a walk. This proves that the visit to arboretum is considered as part of everyday life, and thus a new program and walk path as well as movement route are needed to be developed for the visitors. In the question relating to the facilities and operation programs in Daegu Arboretum, particularly the requests by visitors, they responded that the establishment of cultural event, beautiful natural scenery, refreshment and convenience facilities is the most critical issue. In addition, the management on withered trees and bare lands is an urgent issue as well. In this sense, the Operation and Management Strategies based upon the visitor behaviors and model of satisfaction are needed to deal with the adoption of diverse events and festivals joined by local residents, ombudsman program, environmental program development for students and teachers within the region, negligent bare lands and withered tree replacement, and cafeteria facility improvement and supplement as well as the bench marking of other facilities than arboretums located in other regions. These items are thought to be sufficiently dealt with by Daegu Arboretum having no more external resources. It is recognized that the visitor satisfaction begins from a minor thing, and a small difference determines a great satisfaction, and thus the software approach rather than hardware one is in need.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.18 no.1
• /
• pp.42-54
• /
• 2016

It is highly anticipated that warming temperature resulting from global climate change will affect the phenological pattern of kiwifruit, which has been commercially grown in Korea since the early 1980s. Here, we present the potential impacts of climate change on the variations of flowering day of a gold kiwifruit cultivar, Haegeum, in the Jeonnam Province, Korea. By running six global climate models (GCM), the results from this study emphasize the uncertainty in climate change scenarios. To predict the flowering day of kiwifruit, we obtained three parameters of the 'Chill-day' model for the simulation of Haegeum: $6.3^{\circ}C$ for the base temperature (Tb), 102.5 for chill requirement (Rc), and 575 for heat requirement (Rh). Two separate validations of the resulting 'Chill-day' model were conducted. First, direct comparisons were made between the observed flowering days collected from 25 kiwifruit orchards for two years (2014-15) and the simulated flowering days from the 'Chill-day' model using weather data from four weather stations near the 25 orchards. The estimation error between the observed and simulated flowering days was 5.2 days. Second, the model was simulated using temperature data extracted, for the 25 orchards, from a high-resolution digital temperature map, resulting in the error of 3.4 days. Using the RCP 4.5 and 8.5 climate change scenarios from six GCMs for the period of 2021-40, the future flowering days were simulated with the 'Chill-day' model. The predicted flowering days of Haegeum in Jeonnam were advanced more than 10 days compared to the present ones from multi-model ensemble, while some individual models resulted in quite different magnitudes of impacts, indicating the multi-model ensemble accounts for uncertainty better than individual climate models. In addition, the current flowering period of Haegeum in Jeonnam Province was predicted to expand northward, reaching over Jeonbuk and Chungnam Provinces. This preliminary result will provide a basis for the local impact assessment of climate change as more phenology models are developed for other fruit trees.