Chapter 10
Prediction and Elucidation of Population
Dynamics of the Blue-green Aigae Microcystis
aeruginosa and the Diatom Sfephanodiscus
hantzschii in the Nakdong River-Reservoir
System (South Korea) by a Recurrent Artificial
Neural Network
K.-S. Jeong . F. Recknagel . G.-I. Joo
10.1
Introduction
Ecological modeling is an interdisciplinary branch in ecology. A model
synthesized from adequate laboratory and field data can explain observed patterns
and predict future ecosystem behaviors (Odum 1983; Krebs 1994). For
accomplishing both objectives, it is necessary to use models that adequately
address the uncertainty and complexity of ecosystems. Artificial Neural Networks
(ANN) have been demonstrated to successfully model non-linear and complex
phenomena. They have been used in aquatic ecology (e.g. Recknagel, 1997;
Brosse et al., 1999), medicine, linguistics, and social sciences (Bullinaria 1997;
BIom et al. 1999; Carson et al. 1999; Young et al. 2000).
ANN can function both as predictors and classifiers of temporal and spatial
ecosystem patterns (e.g. Lek et al. 1996; Recknagel et al. 1997; Chon et al. 2000).
Applications of scenario and sensitivity analyses have demonstrated explanatory
capabilities of ANN (e.g. Recknagel and Wilson 2000; Jeong et al. 2001a). Highly
complicated freshwater ecosystems can thus be elucidated to a certain extent by
the ANN approach.
ANN may have good applicability in lotic and lentic freshwater ecosystems,
wh ich are distinguished primarily by the degree of water flow (see Burt 1992).
Phytoplankton often is the major primary producer in these systems; it can exhibit
very different population and community aspects in rivers compared to lakes
(Reynolds 1992). Rivers with a high degree of flow regulation have even greater
complexity and different community features (Stober and Nakatani 1992; Joo et
al. 1997). Great efforts have been undertaken in modeling phytoplankton
dynamics by heuristic and deterministic approaches (e.g. Kamp-Nielson 1978;
Prediction and Elucidation of Population
Dynamics of the Blue-green Aigae Microcystis
aeruginosa and the Diatom Sfephanodiscus
hantzschii in the Nakdong River-Reservoir
System (South Korea) by a Recurrent Artificial
Neural Network
K.-S. Jeong . F. Recknagel . G.-I. Joo
10.1
Introduction
Ecological modeling is an interdisciplinary branch in ecology. A model
synthesized from adequate laboratory and field data can explain observed patterns
and predict future ecosystem behaviors (Odum 1983; Krebs 1994). For
accomplishing both objectives, it is necessary to use models that adequately
address the uncertainty and complexity of ecosystems. Artificial Neural Networks
(ANN) have been demonstrated to successfully model non-linear and complex
phenomena. They have been used in aquatic ecology (e.g. Recknagel, 1997;
Brosse et al., 1999), medicine, linguistics, and social sciences (Bullinaria 1997;
BIom et al. 1999; Carson et al. 1999; Young et al. 2000).
ANN can function both as predictors and classifiers of temporal and spatial
ecosystem patterns (e.g. Lek et al. 1996; Recknagel et al. 1997; Chon et al. 2000).
Applications of scenario and sensitivity analyses have demonstrated explanatory
capabilities of ANN (e.g. Recknagel and Wilson 2000; Jeong et al. 2001a). Highly
complicated freshwater ecosystems can thus be elucidated to a certain extent by
the ANN approach.
ANN may have good applicability in lotic and lentic freshwater ecosystems,
wh ich are distinguished primarily by the degree of water flow (see Burt 1992).
Phytoplankton often is the major primary producer in these systems; it can exhibit
very different population and community aspects in rivers compared to lakes
(Reynolds 1992). Rivers with a high degree of flow regulation have even greater
complexity and different community features (Stober and Nakatani 1992; Joo et
al. 1997). Great efforts have been undertaken in modeling phytoplankton
dynamics by heuristic and deterministic approaches (e.g. Kamp-Nielson 1978;