Data mining for environmental analysis and diagnostic: a case study of upwelling ecosystem of Arraial do Cabo

Pereira, Gilberto Carvalho; Coutinho, Ricardo; Ebecken, Nelson Francisco Favila

Abstracts

The Brazilian coastal zone presents a large extension and a variety of environments. Nevertheless, little is known about biological diversity and ecosystem dynamics. Environmental changes always occur; however, it is important to distinguish natural from anthropic variability. Under these scenarios, the aim of this work is to present a Data Mining methodology able to access the quality and health levels of the environmental conditions through the biological integrity concept. A ten-year time series of physical, chemical and biological parameters from an influenced upwelling area of Arraial do Cabo-RJ were used to generate a classification model based on association rules. The model recognizes seven different classes of water based on biological diversity and a new trophic index (PLIX). Artificial neural networks were evolved and optimized by genetic algorithms to forecast these indices, enabling environmental diagnostic to be made taking into account control mechanisms of topology, stability and complex behavioral properties of food web.

Data Mining; Intelligent Systems; Environmental Diagnostic; Ecosystem Management; Biological Integrity

A zona costeira brasileira apresenta grande extensão e variedade de ambientes. Contudo, pouco se sabe sobre sua diversidade biológica e o funcionamento dos ecossistemas. Como mudanças ambientais são constantes, é muito importante distinguir entre variabilidade natural e antrópica. Nesse cenário, o objetivo deste trabalho é apresentar a metodologia para o desenvolvimento de um Sistema Inteligente de Gerenciamento Integrado do Ecossistema Costeiro (SIGIEC) capaz de acessar o nível de qualidade e saúde ambiental através do conceito de Integridade Biológica. Foram usadas séries temporais de dez anos de parâmetros físicos, químicos e biológicos para extrair conhecimento e gerar modelos de regras de associação para classificar sete diferentes tipos de condições ambientais, analisadas através da diversidade biológica e um novo índice trófico (PLIX). Redes neurais artificiais foram otimizadas por algoritmos genéticos para fazer predições desses índices e apresenta-se um diagnóstico ambiental baseado na análise dos mecanismos de controle da topologia, estabilidade e propriedades do comportamento complexo de redes alimentares.

Mineração de dados; Sistemas inteligentes; Diagnóstico ambiental; Gerenciamento de ecossistemas; Integridade biológica

AGRAWAL, R.; SRIKANT, R.. Fast algorithms for mining associations rules VLDB-94, 1994.
BLOCK, H.D. The perceptron: A model for brain fuctioning. Revs. Mod. Phys., v.34, p. 123-135, 1962.
BRADSHAW, C.J.A.; PURVIS, M.; RAYCOV, R.; ZHOU, Q.; DAVIS, L.S. Predicting patterns in spatial ecology using neural networks: modelling colonization by New Zealand fur seal. In: DENZER, R., SWAYNE, D.A, PURVIS; M.; SCHIMAK, G. (Ed.). Environmental Software Systems Environmental Information and Decision Support, n. 167. Dordrecht: Kluwer Academic Publishers, 2000. p. 57-65.
COWEN, R. K.; K. LWIZA, M. M.; SPONAUGLE, S.; PARIS, C.; OLSON. B. D. B. Connectivity of marine populations: open or close? Science, v. 287, p. 857-859, 2000.
COSTANZA, R.; D'ARGE, R. DE GROOT; FARBER, S.;GRASSO,M.; HANNON, B.; LIMBURG, K.; NAEEM, K.; O'NEILL,R.V.; PARUELO,J.; RASKIN, R.G.; SUTTON, P.; VAN DER BELT, M. The value of the world's ecosystem services and natural capital. Nature, v. 387, p. 253-260, 1997.
CHRISTENSEN, J. Auditing conservation in an age of accountability. Conserv. Practice, v.4, p. 12-19, 2003.
EUROPEAN ENVIRONMENTAL AGENCY - EEA. ELUNIS - European Nature Information System Electronic Source: http://eunis.eea.eu.int/index.jsp, Last web site update: 04.12.2003.
FAYYAD, U. M.; PIATETSKY-SHAPIRO, G.; PADHRAIC, S., UTHURUSAMY, R. (Ed.). Advances in knoledge discovery and Data Mining, Cambridge; London: MIT Press, 1996. 611 p.
GOLDBERG, D.E. Genetic algorithm in search, optimization, and machine learning New York: Addison-Wesley, 1989.
HAN, J.; KAMBER, H. Data Mining - Concepts and Thechniques, Morgan Kaufmann Publishers, Academic Press, San diego, CA, USA, 550 p., 2001.
HIXON, M.A.; BOERSMA, P.; HUNTER, M. L.; MICHELI, FIORENZA, NORSE, E.; POSSINGHAM, H.P.; SNELGROVE, P.V.R. Oceans at Risk: Research priorities in marine conservation biology. In: SOULÉ, M.E.; G.H. ORIANS, G.H. (Ed.). Conservation biology: research priorities for the next decade. Washington, D.C.: Island Press; 2001. p. 125-154.
LEK, S.; DELACOSTE, M.; BARAN, P.; DINOPOULUS, I.; LAUGA, J.; AULAGNIER, S. Aplication of neural networks to nonlinear relationship in ecology,Ecol. Model, v.90, p. 39-52, 1996
LIU, B.; HSU, W.; MA, Y. Building an accurate classifier using association rules. KDD-98. New York, p. 27-31, 1998.
MAIER, H. R.; DANDY, G. C. The application of artificial neural networks to the prediction of salinity Adelaide: Department of Civil and Environmental Engineering, The University of Adelaide, 1993. 464 p. (Research Report No. R101).
MAIER, H. R.; GRAEME, C. D.; MICHAEL, D. B. Use of artificial neural networks for modelling cyanobacteria Anabena spp.in the river Murray, South Australia, Ecol. Model, v. 105, p. 257-272, 1998.
MARGALEF, R. Diversidad de especies en las comunidades naturales. Publ. Inst. Biol. Apl., Barcelona, v. 9, p. 15-27, 1951.
MOORE, T.; MORRIS, K.; BLACKWELL, G.; GIBSON, S. Extraction of beach landforms from dems using a Coastal Management Expert System. Annual Conference of GeoComputation, 2., 1997, New Zealand. Proc., 1997.
NELSON, S. M.; ROLINE, R. A. Effects of multiple stressors on hyporheic invertebrates in lotic system. Ecol. Indicat, v. 3, n. 2, p. 65-79, 2003.
OAKES, R. M.; GIDO, K. B.; FALKE, J. A.; BROCK, B. L. Modelling of stream fishes in the Great Plains, USA. Ecology Freshwat. Fish , v. 14, p. 361-374, 2005.
OLDEN, J. D.; POFF, N. L.; B. P. BLEDSOE, B. P. Incorporating ecological knowledge into ecoinformatics: an example of modeling hierarchically structured aquatic communities with neural network. Ecol. Informatics , v. 1, p. 33-42, 2006.
ÖZESMI, S. L.; ÖZESMI, U. An artificial neural network approach to spatial habitat modelling with interspecific interaction. Ecol. Model, v. 116, n. 1, p. 15-31, 1999.
PEREIRA, G.C. Data Mining for environmental analisys and diagnostic Ph.D. dissertation, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil. ( In portuguese). 2005.
PEREIRA,G. C.; COUTINHO, R.; EBECKEN, N. F. F. 2002. Biological response neural network prediction in coastal upwelling field Oil and Hydrocarbon Spill, 3., 2002. Proc., 2002. p. 301-310.
RECKNAGEL, F.; FRENCH, M.; HARKONEN, P.; YABUNAKA, K. Artificial neural network approach for modelling and prediction of algal blooms. Ecol. Model, 96, pp. 11-28, 1997.
RUMELHART, D. E.; HINTON, G. E.; WILLIAMS, R. J. "Learning representations by back-propagation error". Nature, v. 323, p. 533-536, 1986.
SCARDI, M. Artificial neural network as empirical models for estimating phytoplankton production. Mar. Ecol. Prog. Ser, v. 139, p. 289-299, 1996.
USEPA - U.S. Environmental Protection Agency. Guidelines for ecological risk assesment. Risk Assessment Forum. Washington, DC: Office of Research and Development. U. S. Environmental Protection Agency (EPA/630/R-95/002F), 1998.
VALANTIN, J. L. A dinâmica do plâncton na ressurgência de Cabo Frio-RJ Inst. Pesqui. Mar.,Rio de Janeiro. Coletânea de trabalhos, In: F.P. Brandini (editor) Memórias de III EBP Curitiba, 1988.
VOLLENWEIDER, R. A.; GIOVANARDI, F. G.; MONTANARI, RI-NALDI. Characterization of the trophic conditions of marine coastal waters, with special reference to the NW adriatic Sea: proposal for a trophic scale, turbidity and generalized water quality index. Environmentrics, v. 9, p. 329-357, 1998.

Publication Dates

Publication in this collection
11 Apr 2008
Date of issue
Mar 2008

History

Reviewed
16 Apr 2007
Received
09 June 2006
Accepted
04 July 2007

This work is licensed under a Creative Commons Attribution 4.0 International License.

[1] AGRAWAL, R.; SRIKANT, R.. Fast algorithms for mining associations rules VLDB-94, 1994.

[2] BLOCK, H.D. The perceptron: A model for brain fuctioning. Revs. Mod. Phys., v.34, p. 123-135, 1962.

[3] BRADSHAW, C.J.A.; PURVIS, M.; RAYCOV, R.; ZHOU, Q.; DAVIS, L.S. Predicting patterns in spatial ecology using neural networks: modelling colonization by New Zealand fur seal. In: DENZER, R., SWAYNE, D.A, PURVIS; M.; SCHIMAK, G. (Ed.). Environmental Software Systems Environmental Information and Decision Support, n. 167. Dordrecht: Kluwer Academic Publishers, 2000. p. 57-65.

[4] COWEN, R. K.; K. LWIZA, M. M.; SPONAUGLE, S.; PARIS, C.; OLSON. B. D. B. Connectivity of marine populations: open or close? Science, v. 287, p. 857-859, 2000.

[5] COSTANZA, R.; D'ARGE, R. DE GROOT; FARBER, S.;GRASSO,M.; HANNON, B.; LIMBURG, K.; NAEEM, K.; O'NEILL,R.V.; PARUELO,J.; RASKIN, R.G.; SUTTON, P.; VAN DER BELT, M. The value of the world's ecosystem services and natural capital. Nature, v. 387, p. 253-260, 1997.

[6] CHRISTENSEN, J. Auditing conservation in an age of accountability. Conserv. Practice, v.4, p. 12-19, 2003.

[7] EUROPEAN ENVIRONMENTAL AGENCY - EEA. ELUNIS - European Nature Information System Electronic Source: http://eunis.eea.eu.int/index.jsp, Last web site update: 04.12.2003.

[8] FAYYAD, U. M.; PIATETSKY-SHAPIRO, G.; PADHRAIC, S., UTHURUSAMY, R. (Ed.). Advances in knoledge discovery and Data Mining, Cambridge; London: MIT Press, 1996. 611 p.

[9] GOLDBERG, D.E. Genetic algorithm in search, optimization, and machine learning New York: Addison-Wesley, 1989.

[10] HAN, J.; KAMBER, H. Data Mining - Concepts and Thechniques, Morgan Kaufmann Publishers, Academic Press, San diego, CA, USA, 550 p., 2001.

[11] HIXON, M.A.; BOERSMA, P.; HUNTER, M. L.; MICHELI, FIORENZA, NORSE, E.; POSSINGHAM, H.P.; SNELGROVE, P.V.R. Oceans at Risk: Research priorities in marine conservation biology. In: SOULÉ, M.E.; G.H. ORIANS, G.H. (Ed.). Conservation biology: research priorities for the next decade. Washington, D.C.: Island Press; 2001. p. 125-154.

[12] LEK, S.; DELACOSTE, M.; BARAN, P.; DINOPOULUS, I.; LAUGA, J.; AULAGNIER, S. Aplication of neural networks to nonlinear relationship in ecology,Ecol. Model, v.90, p. 39-52, 1996

[13] LIU, B.; HSU, W.; MA, Y. Building an accurate classifier using association rules. KDD-98. New York, p. 27-31, 1998.

[14] MAIER, H. R.; DANDY, G. C. The application of artificial neural networks to the prediction of salinity Adelaide: Department of Civil and Environmental Engineering, The University of Adelaide, 1993. 464 p. (Research Report No. R101).

[15] MAIER, H. R.; GRAEME, C. D.; MICHAEL, D. B. Use of artificial neural networks for modelling cyanobacteria Anabena spp.in the river Murray, South Australia, Ecol. Model, v. 105, p. 257-272, 1998.

[16] MARGALEF, R. Diversidad de especies en las comunidades naturales. Publ. Inst. Biol. Apl., Barcelona, v. 9, p. 15-27, 1951.

[17] MOORE, T.; MORRIS, K.; BLACKWELL, G.; GIBSON, S. Extraction of beach landforms from dems using a Coastal Management Expert System. Annual Conference of GeoComputation, 2., 1997, New Zealand. Proc., 1997.

[18] NELSON, S. M.; ROLINE, R. A. Effects of multiple stressors on hyporheic invertebrates in lotic system. Ecol. Indicat, v. 3, n. 2, p. 65-79, 2003.

[19] OAKES, R. M.; GIDO, K. B.; FALKE, J. A.; BROCK, B. L. Modelling of stream fishes in the Great Plains, USA. Ecology Freshwat. Fish , v. 14, p. 361-374, 2005.

[20] OLDEN, J. D.; POFF, N. L.; B. P. BLEDSOE, B. P. Incorporating ecological knowledge into ecoinformatics: an example of modeling hierarchically structured aquatic communities with neural network. Ecol. Informatics , v. 1, p. 33-42, 2006.

[21] ÖZESMI, S. L.; ÖZESMI, U. An artificial neural network approach to spatial habitat modelling with interspecific interaction. Ecol. Model, v. 116, n. 1, p. 15-31, 1999.

[22] PEREIRA, G.C. Data Mining for environmental analisys and diagnostic Ph.D. dissertation, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil. ( In portuguese). 2005.

[23] PEREIRA,G. C.; COUTINHO, R.; EBECKEN, N. F. F. 2002. Biological response neural network prediction in coastal upwelling field Oil and Hydrocarbon Spill, 3., 2002. Proc., 2002. p. 301-310.

[24] RECKNAGEL, F.; FRENCH, M.; HARKONEN, P.; YABUNAKA, K. Artificial neural network approach for modelling and prediction of algal blooms. Ecol. Model, 96, pp. 11-28, 1997.

[25] RUMELHART, D. E.; HINTON, G. E.; WILLIAMS, R. J. "Learning representations by back-propagation error". Nature, v. 323, p. 533-536, 1986.

[26] SCARDI, M. Artificial neural network as empirical models for estimating phytoplankton production. Mar. Ecol. Prog. Ser, v. 139, p. 289-299, 1996.

[27] USEPA - U.S. Environmental Protection Agency. Guidelines for ecological risk assesment. Risk Assessment Forum. Washington, DC: Office of Research and Development. U. S. Environmental Protection Agency (EPA/630/R-95/002F), 1998.

[28] VALANTIN, J. L. A dinâmica do plâncton na ressurgência de Cabo Frio-RJ Inst. Pesqui. Mar.,Rio de Janeiro. Coletânea de trabalhos, In: F.P. Brandini (editor) Memórias de III EBP Curitiba, 1988.

[29] VOLLENWEIDER, R. A.; GIOVANARDI, F. G.; MONTANARI, RI-NALDI. Characterization of the trophic conditions of marine coastal waters, with special reference to the NW adriatic Sea: proposal for a trophic scale, turbidity and generalized water quality index. Environmentrics, v. 9, p. 329-357, 1998.

Brasil

Brasil

Data mining for environmental analysis and diagnostic: a case study of upwelling ecosystem of Arraial do Cabo

Abstracts

Publication Dates

History