- Apply for Authority
- P-ISSN2765-2203
- E-ISSN2765-2211
- KCI Candidate
Abstract
The oriental stork (Ciconia boyciana) is listed as an endangered species internationally. Its resident population has been extirpated in South Korea since 1971. Its predicted historical habitat included forests (54%), rice paddy fields (28%), grasslands (17%), river-streams (less than 1%), and villages (less than 1%) based on pre-extirpation records in a previous study. However, habitat attributes of recently reintroduced oriental storks since 2015 remain unknown. To examine habitat use patterns and home ranges of recently reintroduced oriental storks, 2015-2017 tracking data of 17 individuals were used to analyze their spatial attributes with a Kernel Density Estimate method and breeding status. Their habitat use patterns from peripheral to core areas were highly associated with increasing rice paddy fields (26%) and decreasing forested areas (55%). Scale-dependent home ranges were 51% smaller for breeders than for non-breeders on average. Our study results highlight that the habitat use pattern of reintroduced oriental storks seems to be comparable to the historical pattern where the used area is likely to be more centralized for breeders than for non-breeders in South Korea. Furthermore, the direction of habitat management for oriental storks should focus on biodiversity improvement of rice paddy fields with chemical free cultivation and irrigation.
Abstract
Ecological research relies on the interpretation of large amounts of visual data obtained from extensive wildlife surveys, but such large-scale image interpretation is costly and time-consuming. Using an artificial intelligence (AI) machine learning model, especially convolution neural networks (CNN), it is possible to streamline these manual tasks on image information and to protect wildlife and record and predict behavior. Ecological research using deep- learning-based object recognition technology includes various research purposes such as identifying, detecting, and identifying species of wild animals, and identification of the location of poachers in real-time. These advances in the application of AI technology can enable efficient management of endangered wildlife, animal detection in various environments, and real-time analysis of image information collected by unmanned aerial vehicles. Furthermore, the need for school education and social use on biodiversity and environmental issues using AI is raised. School education and citizen science related to ecological activities using AI technology can enhance environmental awareness, and strengthen more knowledge and problem-solving skills in science and research processes. Under these prospects, in this paper, we compare the results of our early 2013 study, which automatically identified African cichlid fish species using photographic data of them, with the results of reanalysis by CNN deep learning method. By using PyTorch and PyTorch Lightning frameworks, we achieve an accuracy of 82.54% and an F1-score of 0.77 with minimal programming and data preprocessing effort. This is a significant improvement over the previous our machine learning methods, which required heavy feature engineering costs and had 78% accuracy.
Abstract
In this study, the distribution, bait preference, hosts, symbionts, habitat status, and cytochrome c oxidase subunit-I sequences of the ant Linepithema humile were analyzed. This species habitat comprised approximately 1.5 km along stone crevices, flower gardens, roadsides, and container yards in the vicinity of Busan station, Korea. Feeding activity of this species was confirmed in 9 species of 7 families of plants. Which was approximately 14% of total flora (63 species of 37 families) in the studied area. Particularly, it was observed that Dendranthema indicum and Camellia japonica were the most frequently visited. Further, we verified that they interacted with four species of aphids. We identified 22 species of ants (17 genera, 4 subfamilies) inhabiting the Busan station area, including L. humile, Tetramorium tsushimae, Nylanderia flavipes and the alien species Paratrechina longicornis. In areas where L. humile was the dominant species, the habitat of local native ants was reduced by approximately 30%. Bait preference was ranked in the following order: jelly, water with glucose (25%), and yogurt, with the highest preference for baits containing sugar and protein. The 630-bp cytochrome c oxidase subunit-I sequences of the local L. humile populations were 99% identical to those of L. humile in the National Center for Biotechnology Information database, and the Argentine ants that invaded Korea had a sequence identical to that of other invasive populations from China, Japan, and New Zealand. As L. humile rapidly increases due to domestic adaptation, continuous monitoring and control strategy for eradication are needed to protect domestic biodiversity.
