Korean Journal of Ecology and Environment (생태와환경)

The Korean Society of Limnology (한국수생태학회)
권호 표지

Community Patterning of Benthic Macroinvertebrates in Urbanized Streams by Utilizing an Artificial Neural Network

인공신경망을 이용한 도시하천의 저서성 대형무척추동물 군집 유형성 연구

  • Kim, Jwa-Kwan (Department of Environmental Engineering, Catholic University of Pusan) ;
  • Chon, Tae-Soo (Division of Biological Sciences, Pusan National University) ;
  • Kwak, Inn-Sil (Brain Science Institute, The Institute of Physical and Chemical Research)
  • 김좌관 (부산카톨릭대학교 산업환경시스템학부) ;
  • 전태수 (부산대학교 생명과학부) ;
  • 곽인실 (일본 이화학연구소 뇌연구센터)
  • Published : 2003.03.31
Download PDF
⟨ Previous Next ⟩

Abstract

Benthic macro-invertebrates were seasonally collected in the Onchen Stream in Pusan, from July 2001 to March 2002. Generally 4 phylum 5 class 10 order 19 family 23 species were observed in the study sites. Ephemeroptera, Plecoptera and various species appeared in headwater stream while Oligochaeta and Chironomidae were dominated in downstream sites. Community abundance patterns, especially the dominant taxa, Oligochaeta and Chironomidae, appeared to be different depending upon the sampling months. Oligochaeta was usually observed in July, December and March while Chironomidae was appeared in September. The biological indices, TBI(Trent Biotic Index), BS (Biotic Score), BMWP (Biological Monitoring Working Party)were calculated with the appeared communities of the sampling sites through the survey months. TBI showed 1 to 8, BMWP was 1 to 93 and CBI appeared 9 to 387 in the different sites. The biological indices decreased from headstream to downstream sites, We implemented the unsupervised Kohonen network for patterning of community abundance of the sampling sites. The patterning map by the Kohonen network was well represented community abundance of the sampling sites. Also, we conducted RTRN (Real Time Recurrent Neural Network) for predicting of the biological indices in the different sites. The results appeared that the predicting values by RTRN were well matched field data (correlation coefficient of TBI, BMWP and CBI were 0.957, 0.979 and 0.967, respectively).

부산의 대표적인 자연형 계곡이 있는 범어사에서 도시하천인 온천천을 거쳐 수영강 합류부에 분포하는 저서성 대형무척추동물을 계절별로 조사하였다. 전 조사지점에서 층 4문 5강 10목 19과 23종이 조사되었다. 범어사에서는 하루살이류 (Ephemeroptera), 강도래류(Ple-coptera)등 다양한 분류군이 출현한 반면에 나머지 조사지에서 빈모류 (Oligochaeta)와 깔따구류 (Chirono-midae)가 우점하였다. 두 우점분류군의 출현시기는 7월12월과 3월은 빈모류군이 우점을 차지하였으나 9월은 깔따구류가 우점을 차지하여 차이를 보였다. 생물학적 지수인 TBI (Trent Biotic Index), BS (Biotic Score), BMWP (Biological Monitoring Working Party)를 조사하였는데 전체적으로 범어사 지점에서 하류로 가면서 각 지수들이 낮아지는 경향이 나타났다. 전체적인 지수 분포는 TBI 1-8, BMWP 1-93 그리고 CBI는 9-387의 분포를 보였다. 비지도 학습법인 코호넨 신경망을 통하여 지점별 저서생물 출현의 유형화가 잘 반영되어 표출되었다. 또한 하천의 지속적인 관리를 위해 생물학적 지수를 회귀신경망을 통하여 예측하였는데 전체적으로 각 지수의 예측은 실제치와 잘 일치하여 나타났다(TBI, BMWP와 CBI의 상관계수(correlation coefficient)는 각각 0.957, 0.979와 0.967).

Keywords

References

  1. 한국동식물도감, 동물편(수서곤충류) v.30 윤일병
  2. Standard methods for the examination of water and waste water(16th ed.) APHA;AWWA;WPCF
  3. Aquatic insects and oligochaetes of North and South California Brigham, A.R.(ed.);W.U. Bringham(ed.);A. Gnika(ed.)
  4. Canadian Special Publication of Fisheries and Aquatic Sciences 84 Guide to the freshwater aquatic microdrile Oligochaetes of North America Brinkhurst, R.O.
  5. Ecol. Model. v.90 Patternizing communities by using an artificial neural network Chon, T.-S.;Y.S. Park;K.H. Moon;E.Y. Cha
  6. Ecol. Model. v.132 Determining temporal pattern of comunity dynamics by using unsupervised learning algorithms Chon, T.-S.;Y.S. Park;J.H. Park
  7. Korean Journal of Ecology v.23 Pattern recognition of long-term ecological data in community changes by using artificial neural networks: Benthic macroinvertebrates and chironomids in a polluted stream Chon, T.-S.;I.S. Kwak;Y.S. Park
  8. Artificial Neural Networks in Ecology and Evolution Patterning of community changes in benthic macroinvertebrates collected from urbanized streams for the short time prediction by temporal artificial neural networks Chon, T.-S.;Y.S. Park;E.Y. Cha;Lek S.(ed.);J.F. Guegan(ed.)
  9. Transactions of the ASAE v.37 Neural network models for predicting flowering and physiological maturity of soybean Elizondo, D.A.;McClendon, R.W.;Hoongenboom, G.
  10. IEEE Transactions on Neural Networks v.5 Dynamic recurrent neural networks: theory and applications Giles, C.L.;Kuhn, G.M.;Williams, R.J.
  11. Neural Networks Haykin, S.
  12. Neurocomputing Hecht-Nielsen, R.
  13. Biological Indicators of Freshwater Pollution and Environmental Management Hellawell, J.M.
  14. Limnol. Oceanogr. v.13 A simple method of assessing the annual production of stream benthos Hynes, H.B.N.;M.J. Coleman
  15. The ecology of running waters Hynes, H.B.N.
  16. Self-organization and Associative Memory Kohonen, T.
  17. Digital Neural Networks Kung, S.Y.
  18. Am. Nat. v.125 Succession of species within a community: chronological clustering, with applications to marine and freshwater zooplankton Legendre, P.;S. Dallot;L. Legendre
  19. IEEE Acoustics, Speech and Signal Processing Magazine no.Aprin An introduction to computing with neural nets Lippmann, R.P.
  20. An Introduction to the Aquatic Insects of North America Merritt, R.W.;K.W. Cummins
  21. Fresh-water Invertebrates of the United States Pennak, R.W.
  22. Ecological Modelling v.96 Artificial neural network approach for modelling and prediction of algal blooms Recknagel, F.;M. French;P. Harkonen;K.-I. Yabunaka
  23. The ecology of aquatic insects Resh, V.H.;D.M. Rosenberg
  24. Parallel Distributed Processing: Explorations in the Microstructure of Cognition v.Ⅰ Learning internal representations by error propagation Rumelhart, D.E.;G.E. Hinton;R.J. Williams;D.E. Rumelhart(ed.);J.L. McCelland(ed.)
  25. Ecological Modelling v.108 Modelling the population dynamics of red deer (Cervus elaphus L.) with regard to forest development Stankovski, V.;M. Debeljak;I. Bratko;M. Adamic
  26. Trans. Amer. Fish. Soc. v.66 Rainbow trout and bottom fauna production in one mile of stream Surber, E.W.
  27. Ecological Modelling v.84 Predicting grassland community changes with an artificial neural network model Tan, S.S.;Smeins, F.E.
  28. Neural computing: Theory and practice Wasserman, P.D.
  29. Keys and Diagnoses.(Part 1. Larvae). Ent. Scand. no.SUP.19 Chironomidae of the Holactic Region Wiederholm, T.
  30. Illustrated Encyclopedia of Fauna and Flora of Korea, Aquatic insects v.30 Yun, I.B.
  31. Biostatistical Analysis Zar, J.H.
  32. Introduction to Artificial Neural Systems Zurada, J.M.