Ken Kingston

Abstract

This thesis investigatesth e application of complex adaptives ystemsa pproaches (e. g. Artificial Neural Networks and Evolutionary Computation) to the study of coastal hydrodynamica nd morphodynamicb ehaviour.T raditionally, nearshorem orphologicalc oastal systems tudiesh ave developeda n understandingo f thosep hysicalp rocesseso ccurringo n both short temporal, and small spatial scales with a large degree of success. The associated approachesa nd conceptsu sedt o study the coastals ystema t theses calesh ave Primarily been linear in nature.H owever,w hent hesea pproachetso studyingt he coastals ystema re extendedto investigating larger temporal and spatial scales,w hich are commensuratew ith the aims of coastal managementr, esults have had less success.T he lack of successi n developing an understandingo f large scalec oastalb ehaviouri s to a large extent attributablet o the complex behavioura ssociatedw ith the coastals ystem.I bis complexity arises as a result of both the stochastic and chaotic nature of the coastal system. This allows small scale system understandingto be acquiredb ut preventst he Largers caleb ehaviourt o be predictede ffectively. This thesis presentsf our hydro-morphodynamicc ase studies to demonstratet he utility of complex adaptives ystema pproachesfo r studying coastals ystems.T he first two demonstrate the application of Artificial Neural Networks, whilst the latter two illustrate the application of EvolutionaryC omputation.C aseS tudy #I considerst he natureo f the discrepancyb etweent he observedl ocation of wave breakingp atternso ver submergeds andbarsa nd the actual sandbar locations.A rtificial Neural Networks were able to quantitativelyc orrectt he observedlo cations to produce reliable estimates of the actual sand bar locations. Case Study #2 considers the developmenot f an approachf or the discriminationo f shorelinel ocation in video imagesf or the productiono f intertidal mapso f the nearshorer egion. In this caset he systemm odelledb y the Artificial Neural Network is the nature of the discrimination model carried out by the eye in delineating a shoreline feature between regions of sand and water. The Artificial Neural Network approachw as shownt o robustly recognisea rangeo f shorelinef eaturesa t a variety of beaches and hydrodynamic settings. Case Study #3 was the only purely hydrodynamic study consideredin the thesis.I t investigatedth e use of Evolutionary Computationt o provide means of developing a parametric description of directional wave spectra in both reflective and nonreflective conditions. It is shown to provide a unifying approach which produces results which surpassedth ose achievedb y traditional analysisa pproachese vent hough this may not strictly have been considered as a fidly complex system. Case Study #4 is the most ambitious applicationa nd addressetsh e needf or data reductiona s a precursorw hen trying to study large scalem orphodynamicd ata sets.I t utilises EvolutionaryC omputationa pproachesto extractt he significant morphodynamic variability evidenced in both directly and remotely sampled nearshorem orphologiesS. ignificantd atar eductioni s achievedw hilst reWning up to 90% of the original variability in the data sets. These case studies clearly demonstrate the ability of complex adaptive systems to be successfidly applied to coastal system studies. This success has been shown to equal and sometimess urpasst he results that may be obtained by traditional approachesT. he strong performance of Complex Adaptive System approaches is closely linked to the level of complexity or non-linearity of the system being studied. Based on a qualitative evaluation, Evolutionary Computation was shown to demonstrate an advantage over Artificial Neural Networks in terms of the level of new insights which may be obtained. However, utility also needs to consider general ease of applicability and ease of implementation of the study approach.I n this sense,A rtificial Neural Networks demonstratem ore utility for the study of coastals ystems.T he qualitative assessmenatp proachu sedt o evaluatet he cases tudiesi n this thesis, may be used as a guide for choosingt he appropriatenesso f either Artificial Neural Networks or Evolutionary Computation for future coastal system studies.

Keywords

Oceanography, Geology, Mineralogy, Sedimentology, Artificial Intelligence, Coastal Hydrodynamics

Document Type

Thesis

Publication Date

2003

Download

Share

COinS