물 자연 그리고 사람 - 물로 더 행복한 세상을 만들어가겠습니다.
HOME성과논문실적

논문실적

Spatial Patterning of Population Proliferations in Benthic Macroinvertebrates in Association with EnvironmentalFactors in the River System in Korea Based on Geo-Self-Organizing Map (Geo-SOM) 게시글의 제목, 학술지명, 저자, 발행일, 작성내용을 보여줌
Spatial Patterning of Population Proliferations in Benthic Macroinvertebrates in Association with EnvironmentalFactors in the River System in Korea Based on Geo-Self-Organizing Map (Geo-SOM)
학술지명 2019 응용곤충학회 추계학술발표회 저자 김호준,박연정,전태수,임주백,장용혁,최재한,송행섭,곽인실,주기재,이황구,박정호
발표일 2019-10-24

Populations proliferate unexpectedly due to disturbances prevalent in ecosystems nowadays. Recently population
size increased rapidly in a number of aquatic species in streams and rivers. However, population eruptions are a
complex manner caused by numerous environmental factors including pollutions sources. A machine learning
technique was used for data analysis in this study. Benthic macroinvertebrate communities were collected by the
Surber net and dredge across the dam areas along the main tributary in the river system in Korea including Nakdong
River, Han-River, Geumgang River and Young-san River monthly and bimonthly from March, 2018 to May, 2019.
Population densities during the survey periods were trained by Geo-Self-Organizing Map (Geo-SOM). The Geo-
SOM training illustrated geographic areas specifically presenting high population densities in Chironomids, Bryozoa,
and Oligochaetes, being concurrently associated with environmental factors pertaining to the sampled dam areas.
Some species including Tanytarsus sp. 1, Pectinatella magnifica, and Limnodrilus hoffmeisteri were locally abundant
in relation to high levels of nutrients (e.g., TN, TP), and chemical (e.g., pH, DO) and physical (e.g., water speed,
substrate size) factors. Clusters and nodes that were grouped by high levels of population densities in the Geo-SOM
were presented in an organized manner to serve as a source of prognosing population proliferations in association
with places and seasons in field conditions. Feasibility of extracting complex information from population density
changes by machine learning was further discussed regarding prediction and management of aquatic species.

목록