Current Projects »
Application of Neuro fuzzy techniques to predict coastal physical sensitivity and vulnerability for marine cage culture:
Juan Moreno Navas: Ph.D.
In recent years, Geographical information systems (GIS) have increased their importance as a tool for spatial decision-making.
This is partly due to recent developments and significant improvements in their capabilities which have made GIS the main application in land
allocation and environmental management. This project focuses on improving the capability of GIS for Coastal Zone Management,
specifically on coastal allocation of aquaculture farms. Incorporation of GIS with a modelling approach that is robust has the potential
for creating a successful modelling tool.
There is a need to develop new modelling techniques that assess to select the best areas for marine cage culture in
coastal water with less expensive data and which are robust when data are uncertain and incomplete. The main project aim
is to develop a model using Neuro-fuzzy techniques in a GIS to predict coastal physical sensitivity, vulnerability, suitability
and the simultaneous consideration of both criteria. Suitability and vulnerability, can define sustainability in the conceptual
sense that a sustainable or an optimum marine use system includes maximum coastal suitability and minimum coastal vulnerability.
The overall output will be an environmental spatial model for application in coastal areas intended to facilitate policy decision,
taking into account the intrinsic characteristics of the target area.
In the early 1990s, acknowledging unique differences between potential contamination factors based on the natural environment and those based on specific contaminant properties and/or human land use and surface management activities, the EPA (USA) moved to distinguish between the sensitivity of an aquifer and the overall vulnerability to contamination of an area’s ground water resources. This concept will be adapted to an aquaculture framework. Sensitivity refers to the relative ease with which an aquaculture contaminant applied on or near the coast can pollute the zone of interest, based solely upon hydrodynamics/topographic factors. It is a function of the intrinsic characteristics of the zone in question. In contrast, vulnerability combines the hydrodynamics/topographic characteristics determining sensitivity with “human” vulnerability factors, specifically addressing specific coastal uses, management practices, and/or contaminant properties from aquaculture activities.
An idealized data flow of the project is shown below. A hydrodynamic model and particle tracking model (using, 3D MOHID) is coupled with a GIS, (Arc View 3.2) and calibrated using in situ data set. The calibrated model is then used to predict the evolution of the parameters selected. The data set from this model will be the input of the Aquaculture Neuro Fuzzy Systems, A.N.F.S will be designed and built in NEFCLASS software (an interactive simulation software to develop, train, and test a Neuro- Fuzzy System for Classification). The final products, GIS/based fuzzy maps (physical sensitivity and vulnerability) are achieved by combining predicted environmental parameters selected with environmental and infrastructure data.
The study area is Mulroy bay, a sea loch in the north-west of Ireland. The whole area is a proposed Special area of Conservation ( SAC). Aquaculture is intensive in the area, with up 16 operators currently licensed for mussel, oyster, clams, scallops, abalone and salmon production. A 3D hydrodynamic model coupled to a particle-tracking model has been developed to study the circulation patters, dispersion processes and residence time in this Irish loch.
Click on images to view enlarged versions (open in a new window)
