One of the major objectives of the MEARMED project was the development of a predictive model to simulate the environmental response at Mediterranean Sea cage farms to differing cage stocking levels and feeding regimes. This was realised by means of establishing semi-empirical relationships between modelled flux (g waste solids per sq.m bed per year) and benthic community descriptors. The general rationale behind the development of the model was that any general relationships established are useful as these allow prediction of benthic effects for different planning and monitoring scenarios.
It was found that for most benthic indices (i.e. species, abundance, abundance/species ratio, Shannon Wiener and Simpson diversity, Redox), the generalised relationships between the indices and the modelled flux are satisfactory. It was also found that for some other indices, such as biomass, accuracy of the predictions were low, and therefore it was important to specify an Envelope of Acceptable Precision (EAP). This EAP specifies what percentage of stations is expected to fall within the envelope. This approach is transparent in that no complex or potentially misleading statistics are used to describe model performance. It was concluded that despite the occasional inaccuracies, MERAMOD provides an acceptable level of predictive capability, which is desirable in practical applications as not always the highest level of detail is available for input data.
Dissemination and use potential:
The MERAMOD model will be disseminated by presentations at international scientific conferences, promotional material distributed to fish farming companies, aquaculture industry bodies and environmental consultancies, circulation of information via the MERAMED newsletter circulation list and promotion via the MERAMED web site. The use potential of this model is global, in particular where sea bass and bream farms are sited in the marine environment. Given the well-established ethos and scientific principles behind the model, its use is not restricted to these species or the Eastern Mediterranean environment.
Key innovative features of the result:
The key features of this model lie in its validation and its improvement on existing models. The benthic module which allows prediction of benthic effects has been established for a number of descriptors and this is uncommon in this type of model. In addition, validation of the trajectories of waste particles from surface to sea bed was satisfactory as this was tested over a range of distances from the farm, not just directly underneath the cages (a common fault with existing models). Given the measured variability of particle flux and benthic community structure, these validation aspects are key features. A further key feature lies in the development of a method which measured accurately a high number of settling velocities of waste faecal particles for sea bass and bream, previously absent in the literature.
Use of the result and its expected benefits:
The modelling package is likely to be used by regulators, fish farming companies, consultants and scientific researchers (based on the partners experience from dissemination of a salmonid based North Atlantic model). The expected benefits are testing of various scenarios relating to planning and monitoring of aquaculture operations. The model will provide information to supplement normal observational baseline or monitoring programmes. It will benefit the environment by allowing an additional stage in the consultation process so that adverse effects from aquaculture operations can be avoided.