Part I Introduction ............................................. 1
1 Ecological Applications of Fuzzy Logic ....................... 3
1.1 Fuzzy Sets and Fuzzy Logic .............................. 3
1.2 Fuzzy Approach to Ecological Modelling and Data
Analysis ................................................ 4
1.3 Fuzzy Classification: A Fuzzy Clustering Approach ....... 6
1.4 Fuzzy Regionalisation: A Fuzzy Kriging Approach ......... 9
1.5 Fuzzy Knowledge-Based Modelling ......................... 9
1.6 Conclusions ............................................ 12
References .................................................. 12
2 Ecological Applications of Qualitative Reasoning ............ 15
2.1 Introduction ........................................... 15
2.2 Why Use QR for Ecology? ................................ 16
2.3 What is Qualitative Reasoning? ......................... 17
2.3.1 A Working Example ............................... 18
2.3.2 World-view: Ontological Distinctions ............ 19
2.3.2.1 Component-based Approach ............... 19
2.3.2.2 Process-based Approach ................. 21
2.3.2.3 Constraint-based Approach .............. 22
2.3.2.4 Suitability of Approaches .............. 23
2.3.3 Inferring Behaviour from Structure .............. 23
2.3.4 Qualitativeness and Representing Time ........... 25
2.3.5 Causality ....................................... 27
2.3.6 Model-fragments and Compositional Modelling ..... 30
2.4 Tools and Software ..................................... 30
2.4.1 Workspaces in Homer ............................. 31
2.4.2 Building a Population Model ..................... 32
2.4.3 Running and Inspecting Models with VisiGarp ..... 35
2.4.4 Adding Migration to the Population model ........ 36
2.5 Examples of QR-based Ecological Modelling .............. 39
2.5.1 Population and Community Dynamics ............... 39
2.5.2 Water Related Models ............................ 41
2.5.3 Management and Sustainability ................... 42
2.5.4 Details in Qualitative Algebra .................. 42
2.5.5 Details in Automated Model Building ............. 43
2.5.6 Diagnosis ....................................... 43
2.6 Conclusion ............................................. 44
References .................................................. 44
3 Ecological Applications of Non-Supervised Artificial
Neural Networks ............................................. 49
3.1 Introduction ........................................... 49
3.2 How to Compute a Self-Organizing Map (SOM) with an
Abundance Dataset? ..................................... 50
3.2.1 A Dataset for Demonstrations .................... 50
3.2.2 The Self-Organizing Map (SOM) Algorithm ......... 52
3.3 How to Use a Self-Organizing Map with an Abundance
Dataset? ............................................... 56
3.3.1 Mapping the Stations ............................ 56
3.3.2 Displaying a Variable ........................... 58
3.3.3 Displaying an Abiotic Variable .................. 59
3.3.4 Clustering with a SOM ........................... 60
3.4 Discussion ............................................. 63
3.5 Conclusion ............................................. 65
References .................................................. 66
4 Ecological Applications of Genetic Algorithms ............... 69
4.1 Introduction ........................................... 69
4.2 Ecology and Ecological Modelling ....................... 70
4.3 Genetic Algorithm Design Details ....................... 72
4.4 Applications of Genetic Algorithms to Ecological
Modelling .............................................. 74
4.5 Predicting the Future with Genetic Algorithms .......... 78
4.6 The Next Generation: Hybrids Genetic Algorithms ........ 79
References .................................................. 80
5 Ecological Applications of Evolutionary Computation ......... 85
5.1 Introduction ........................................... 85
5.2 Ecological Modelling ................................... 86
5.2.1 The Challenges of Ecological Modelling .......... 86
5.2.2 Summary ......................................... 88
5.3 Evolutionary Computation ............................... 88
5.3.1 The Basic Evolutionary Algorithm ................ 90
5.3.2 Summary ......................................... 93
5.4 Ecological Modelling and Evolutionary Algorithms ....... 93
5.4.1 Equation Discovery .............................. 93
5.4.2 Optimisation of Difference Equations ............ 94
5.4.3 Evolving Differential Equations ................. 95
5.4.4 Rule Discovery .................................. 95
5.4.5 Modelling Individual and Cooperative
Behaviour ....................................... 97
5.4.6 Predator-Prey Algorithms ....................... 100
5.4.7 Modelling Hierarchical Ecosystems .............. 100
5.5 Conclusion ............................................. 102
References ................................................. 102
6 Ecological Applications of Adaptive Agents ................. 109
6.1 Introduction .......................................... 109
6.2 Adaptive Agents Framework ............................. 110
6.3 Individual-Based Adaptive Agents ...................... 112
6.4 State Variable-Based Adaptive Agents .................. 114
6.4.1 Algal Species Simulation by Adaptive Agents .... 116
6.4.1.1 Embodiment of Evolutionary
Computation in Agents ................. 116
6.4.1.2 Adaptive Agents Bank .................. 117
6.4.2 Pelagic Food Web Simulation by Adaptive
Agents ......................................... 121
6.5 Conclusions ........................................... 122
Acknowledgements ........................................... 122
References ................................................. 123
7 Bio-Inspired Design of Computer Hardware by Self-
Replicating Cellular Automata .............................. 125
7.1 Introduction .......................................... 125
7.2 Cellular Automata ..................................... 126
7.3 Von Neumann's Universal Constructor ................... 128
7.4 Self-Replicating Loops ................................ 131
7.5 Self-Replication in the Embryonics Project ............ 132
7.5.1 Embryonics ..................................... 132
7.5.2 The Tom Thumb Algorithm ........................ 136
7.5.2.1 Construction of the Minimal Cell ...... 136
7.5.2.2 Growth and Self-Replication ........... 140
7.5.2.3 The LSL Acronym Design Example ........ 141
7.5.2.4 Universal Construction ................ 144
7.6 Conclusions ........................................... I45
Acknowledgements ........................................... 146
References ................................................. 146
Part II Prediction and Elucidation of Stream Ecosystems ....... 149
8 Development and Application of Predictive River Ecosystem
Models Based On Classification Trees and Artificial
Neural Networks ............................................ 151
8.1 Introduction .......................................... 151
8.2 Study Sites, Data Sources and Modelling Techniques .... 152
8.2.1 The Zwalm River Basin .......................... 152
8.2.2 Data Collection ................................ 153
8.2.3 Classification Trees ........................... 154
8.2.4 Artificial Neural Networks ..................... 155
8.2.5 Model Assessment ............................... 156
8.3 Results ............................................... 157
8.3.1 Classification Trees ........................... 157
8.3.1.1 Model Development and Validation ...... 157
8.3.1.2 Application of Predictive
Classification Trees for River
Management ............................ 158
8.3.2 Artificial Neural Networks ..................... 160
8.3.2.1 Model Development and Validation ...... 160
8.3.2.2 Application of Predictive Artificial
Neural Networks for River
Management ............................ 162
8.3.2.2.1 Prediction of
Environmental Standards .... 162
8.3.2.2.2 Feasibility Analysis of
River Restoration
Options .................... 163
8.4 Discussion ............................................ 164
Acknowledgements ........................................... 165
References ................................................. 165
9 Modelling Ecological Interrelations in Running Water
Ecosystems with Artificial Neural Networks ................. 169
9.1 Introduction .......................................... 169
9.2 Materials and Methods ................................. 170
9.2.1 DataBase ....................................... 170
9.2.2 Data Pre-Processing ............................ 170
9.2.3 Artificial Neural Network Types ................ 171
9.2.4 Dimension Reduction ............................ 171
9.2.5 Quality Measures ............................... 171
9.3 Data Exploration with Unsupervised Learning Systems ... 172
9.4 Correlations and Predictions with Supervised
Learning Systems ...................................... 175
9.4.1 Correlations and Predictions of Environmental
Variables ...................................... 177
9.4.2 Dependencies of Colonisation Patterns of
Macro-Invertebrates on Water Quality and
Habitat Characteristics ........................ 177
9.4.2.1 Aquatic Insects in a Natural Stream,
the Breitenbach ....................... 177
9.4.2.2 Anthropogenically Altered Streams ..... 180
9.4.3 Bioindication .................................. 181
9.5 Assessment of Model Quality and Visualisation
Possibilities: Hybrid Networks ........................ 182
9.6 Conclusions ........................................... 183
Acknowledgements ........................................... 185
References ................................................. 185
10 Non-linear Approach to Grouping, Dynamics and
Organizational Informatics of Benthic Macroinvertebrate
Communities in Streams by Artificial Neural Networks ....... 187
10.1 Introduction .......................................... 187
10.2 Grouping Through Self-Organization .................... 190
10.2.1 Static Grouping ................................ 190
10.2.2 Grouping Community Changes ..................... 203
10.3 Prediction of Community Changes ....................... 207
10.3.1 Multilayer Perceptron with Time Delay .......... 207
10.3.2 Elman Network .................................. 211
10.3.3 Fully Connected Recurrent Network .............. 214
10.3.4 Impact of Environmental Factors Trained with
the Recurrent Network .......................... 218
10.4 Patterning Organizational Aspects of Community ........ 221
10.4.1 Relationships among Hierarchical Levels in
Communities .................................... 221
10.4.2 Patterning of Exergy ........................... 227
10.5 Summary and Conclusions ............................... 233
Acknowledgements ........................................... 234
References ................................................. 234
11 Elucidation of Hypothetical Relationships between Habitat
Conditions and Macroinvertebrate Assemblages in
Freshwater Streams by Artificial Neural Networks ........... 239
11.1 Introduction .......................................... 239
11.2 Study Site ............................................ 240
11.3 Materials and Methods ................................. 240
11.3.1 Data ........................................... 240
11.3.2 Neural Network Modelling ....................... 241
11.3.3 Sensitivity Analysis ................................ 242
11.4 Results and Discussion ................................ 243
11.4.1 Elucidation of Hypothetical Relationships ........... 243
11.4.2 Discovery of Contradictory Relationships ............ 247
11.4.3 Limitations of the Method ........................... 248
11.5 Conclusions ........................................... 249
References ................................................. 250
Part III Prediction and Elucidation of River Ecosystems ....... 253
12 Prediction and Elucidation of Population Dynamics of the
Blue-green Algae Microcystis aeruginosa and the Diatom
Stephanodiscus hantzschii in the Nakdong River-Reservoir
System (South Korea) by a Recurrent Artificial Neural
Network .................................................... 255
12.1 Introduction .......................................... 255
12.2 Description of the Study Site ......................... 256
12.3 Materials and Methods ................................. 257
12.3.1 Data Collection and Analysis ................... 257
12.3.2 Modelling the Phytoplankton Dynamics ........... 259
12.3.3 Neural Network Validation and Knowledge
Discovery on Algal Succession .................. 261
12.4 Results and Discussion ................................ 261
12.4.1 Limnological Aspects and Plankton Dynamics in
the Lower Nakdong River ........................ 261
12.4.2 Configuring the Neural Network Architecture
for Predictability ............................. 263
12.4.3 Elucidation of Ecological Hypothesis ........... 265
12.4.3.1 Microcystis aeruginosa ................ 267
12.4.3.2 Stephanodiscus hantzschii ............. 267
12.5 Implications of Ecological Informatics for
Limnology ............................................. 268
12.6 Conclusions ........................................... 269
Acknowledgements ........................................... 270
References ................................................. 270
13 An Evaluation of Methods for the Selection of Inputs for
an Artificial Neural Network Based River Model ............. 275
13.1 Introduction .......................................... 275
13.2 Methods ............................................... 277
13.2.1 Unsupervised Input Preprocessing ............... 277
13.2.2 Supervised Input Determination ................. 280
13.3 Case Study ............................................ 282
13.4 Model Development ..................................... 282
13.4.1 Performance Measures and Model Validation ...... 283
13.4.2 Data Division .................................. 283
13.4.3 Determination of Model Inputs .................. 284
13.5 Results and Discussion ................................ 284
13.6 Conclusions ........................................... 290
Acknowledgements ........................................... 291
References ................................................. 291
14 Utility of Sensitivity Analysis by Artificial Neural
Network Models to Study Patterns of Endemic Fish Species ... 293
14.1 Introduction .......................................... 293
14.2 Contribution of Environmental Variables ............... 294
14.3 Application to Ecological Data ........................ 295
14.4 Results ............................................... 296
14.4.1 Predictive Power ............................... 296
14.4.2 Sensitivity Analysis ........................... 298
14.5 Discussion ............................................ 302
14.6 Conclusions ........................................... 304
References ................................................. 304
Part IV Prediction and Elucidation of Lake and Marine
Ecosystems ............................................ 307
15 A Comparison between Neural Network Based and Multiple
Regression Models in Chlorophyll-a Estimation .............. 309
15.1 Introduction .......................................... 309
15.1.1 Eutrophication in Water Bodies and Relevant
Models ......................................... 309
15.1.2 Artificial Neural Networks ..................... 310
15.1.3 The Use of Artificial Neural Networks in
Environmental Modelling ........................ 311
15.2 Data and Lakes ........................................ 311
15.3 Methodology ........................................... 313
15.3.1 Artificial Neural Network Approach ............. 314
15.3.1.1 Training Method ....................... 314
15.3.1.2 Data Pre-Processing ................... 316
15.3.1.3 Improving Generalisation .............. 316
15.3.2 Multiple Regression Modelling Approach ......... 317
15.4 Results ............................................... 317
15.5 Conclusions and Recommendations ....................... 320
15.5.1 Conclusions .................................... 320
15.5.2 Recommendations ................................ 321
Acknowledgments ............................................ 322
References ................................................. 322
16 Artificial Neural Network Approach to Unravel and
Forecast Algal Population Dynamics of Two Lakes Different
in Morphometry and Eutrophication .......................... 325
16.1 Introduction .......................................... 325
16.2 Materials and Methods ................................. 326
16.2.1 Study Sites and Data ........................... 326
16.2.2 Methods ........................................ 327
16.3 Results ............................................... 330
16.3.1 Forecasting Seasonal Algal Abundances and
Succession ..................................... 330
16.3.2 Relationships between Algal Abundances and
Water Quality Conditions ....................... 331
16.3.3 Relationships between Algal Abundances,
Seasons and Water Quality Changes .............. 336
16.4 Discussion ............................................ 340
16.4.1 Forecasting Seasonal Algal Abundances and
Succession ..................................... 340
16.4.2 Relationships between Algal Abundances,
Seasons and Water Quality Changes .............. 341
16.5 Conclusions ........................................... 344
Acknowledgements ........................................... 344
References ................................................. 344
17 Hybrid Evolutionary Algorithm* for Rule Set Discovery in
Time-Series Data to Forecast and Explain Algal Population
Dynamics in Two Lakes Different in Morphometry and
Eutrophication ............................................. 347
17.1 Introduction .......................................... 347
17.2 Materials and Methods ................................. 348
17.2.1 Study Sites and Data ........................... 348
17.2.2 Hybrid Evolutionary Algorithms ................. 349
17.2.2.1 Structure Optimisation of Rule Sets
Using GP .............................. 351
17.2.2.2 Parameter optimization of Rule Sets
Using a General Genetic Algorithm ..... 356
17.2.2.3 Forecasting by Rule Sets .............. 357
17.3 Case Studies Lake Kasumigaura and Lake Soyang ......... 358
17.3.1 Parameter Settings and Measures ................ 358
17.3.2 Results and Discussion ......................... 359
17.4 Conclusions ........................................... 366
References ................................................. 366
18 Multivariate Time-Series Prediction of Marine Zooplankton
by Artificial Neural Networks .............................. 369
18.1 Introduction .......................................... 369
18.2 Generalisation ........................................ 371
18.3 Automatic Termination of Training ..................... 374
18.4 Case Study: Zooplankton Prediction .................... 378
18.5 Conclusions ........................................... 381
Acknowledgement ............................................ 382
References ................................................. 382
19 Classification of Fish Stock-Recruitment Relationships in
Different Environmental regimes by Fuzzy Logic Combined
with a Bootstrap Re-sampling Approach ...................... 385
19.1 Introduction .......................................... 385
19.2 Fuzzy Stock-Recruitment Model ......................... 386
19.2.1 Traditional Stock-Recruitment Model ............ 386
19.2.2 Fuzzy Stock-recruitment Model .................. 388
19.2.2.1 Fuzzy Membership Function (FMF) ....... 389
19.2.2.2 Fuzzy Rules ........................... 390
19.2.2.3 Fuzzy Reasoning ....................... 391
19.3 Hybrid Optimal Learning and Bootstrap Re-sampling
Algorithms ............................................ 393
19.3.1 Hybrid Optimal Learning Algorithms ............. 394
19.3.2 Bootstrap re-sampling Procedure ................ 396
19.4 Two Real Data Analyses ................................ 397
19.4.1 West Coast Vancouver Island Herring Stock ...... 397
19.4.1.1 Data Prescription and Preliminary
Analyses .............................. 397
19.4.1.2 Fuzzy-SR Model Analysis ............... 398
19.4.1.3 Bootstrap Re-sampling Analysis ........ 400
19.4.2 Southeast Alaska Pink Salmon ................... 402
19.4.2.1 Data Prescription and Preliminary
Analysis .............................. 402
19.4.2.2 Fuzzy-SR Model Analysis ............... 403
19.4.2.3 Bootstrap Re-sampling Analysis ........ 404
19.5 Summary and Discussion ................................ 404
Acknowledgements ........................................... 406
References ................................................. 406
20 Computational Assemblage of Ordinary Differential
Equations for Chlorophyll-a Using a Lake Process Equation
Library and Measured Data of Lake Kasumigaura .............. 409
20.1 Introduction .......................................... 409
20.2 Methods and Materials ................................. 410
20.2.1 LAGRAMGE: Computational Assemblage of ODE ...... 410
20.2.2 Domain Knowledge Library for Lake Ecosystems ... 411
20.2.3 Task Specification ............................. 412
20.2.4 Data of Lake Kasumigaura ....................... 415
20.2.5 Experimental Framework ......................... 416
20.3 Results and Discussion ................................ 418
20.3.1 Experiment 1 ................................... 418
20.3.2 Experiment 2 ................................... 422
20.3.3 Experiment 3 ................................... 424
20.4 Conclusions
References ................................................. 427
Part V Classification of Ecological Images at Micro and
Macro Scale ................................................ 429
21 Identification of Marine Microalgae by Neural Network
Analysis of Simple Descriptors of Flow Cytometric Pulse
Shapes ..................................................... 431
21.1 Introduction .......................................... 431
21.2 Materials and Methods ................................. 435
21.2.1 Pulse Shape Extraction ......................... 435
21.2.2 Data Filtering ................................. 435
21.2.3 Data Transformation ............................ 435
21.2.4 Principal Component Analysis ................... 436
21.2.5 Neural Network Analysis ........................ 438
21.2.6 Hardware and Software .......................... 439
21.3 Results ............................................... 439
21.4 Discussion ............................................ 441
21.5 Conclusions ........................................... 441
Acknowledgement ............................................ 441
References ................................................. 442
22 Age Estimation of Fish Using a Probabilistic Neural
Network .................................................... 445
22.1 Introduction .......................................... 445
22.2 Traditional Methods of Age Estimation ................. 445
22.3 Approaches to Automation in Fish Age Estimation ....... 447
22.4 The Application of a Probabilistic Neural Network to
Fish Age Estimation ................................... 448
22.5 Results ............................................... 452
22.6 Discussion ............................................ 454
Acknowledgements ........................................... 456
References ................................................. 456
23 Pattern Recognition and Classification of Remotely Sensed
Images by Artificial Neural Networks ....................... 459
23.1 Introduction .......................................... 459
23.2 Neural Networks in Remote Sensing ..................... 460
23.2.1 Classification Applications .................... 460
23.2.2 Regression Applications ........................ 461
23.3 The Neural Networks Used in Remote Sensing ............ 461
23.3.1 Feedforward Neural Networks .................... 462
23.3.1.1 Multi-Layer Perceptron (MLP) .......... 463
23.3.1.2 Radial Basis Function (RBF) ........... 464
23.3.1.3 Probabilistic Neural Networks (PNN) ... 465
23.3.1.4 Generalised Regression Neural
Networks (GRNN) ....................... 466
23.3.1.5 Other Network Types ................... 467
23.4 Current Status ........................................ 468
23.4.1 An Example of Neural Networks for
Classification ................................. 469
23.4.2 Concerns with neural Networks .................. 471
23.5 Conclusions ........................................... 472
Acknowledgments ............................................ 473
References ................................................. 473
Index ......................................................... 479
Appendix ...................................................... 483
|
