Simon J. Cripps
Rogaland Research Institute, P.O. Box 2503 Ullandhaug, N-4004 Stavanger, Norway
and
Aquaculture Department, Swedish University of Agricultural Sciences, S-901 83 Umeaa,
Sweden.
ABSTRACT
Waste material in the effluents from intensive aquaculture facilities occurs in low concentrations and at high water flow rates compared with, for example, domestic sewage. Aquaculture effluents are therefore difficult to treat. Much of the waste phosphorous and biodegradable organic matter and some of the waste nitrogen is associated with the particulate phase. Treatment techniques suitable for aquaculture application therefore usually aim to separate the suspended particles from the primary flow by sedimentation or mechanical microscreening. The aim of this study was to experimentally determine the distribution of nutrients with respect to particle size. From this information, a model was developed to enable the more efficient targeting of treatment effort on those particles that were likely to cause the greatest environmental loading.
Representative samples of the study effluent were carefully taken from a Swedish hatchery containing about 1 million salmonids of up to 25 g weight each. The particulate phase of the effluent was serially divided into seven particle size fractions using 200, 100, 85, 65, 47, 25 and 5 µm pore size membranes. Concentration ranges of the different fractions from unfiltered to 5 µm filtration were respectively: 131.7 - 67.9 µg/l total phosphorus; 0.70 - 0.47 µg/l total nitrogen; and 6.9 - 1.8 mg/l suspended solids. From these results a model was proposed. This indicated that no single filtered fraction contained a disproportionate concentration of any of these parameters. The pore size of microscreen chosen to treat this effluent should then be as small as possible, being governed by operation factors such as hydraulic loading. The total particulate phosphorus concentration in the suspended solids increased significantly with decreased mean particle diameter. Though, as suspended solids concentrations were lower in the smaller pore size fractions, overall total phosphorus concentrations were unaffected.
The model was verified by comparison with published commercial microscreen effluent treatment results. It is proposed that the site-specific model could be used to estimate: the minimum achievable removal efficiencies at a given microscreen pore size between 200 - 5 µm; the pore size required at a facility; and changes in treatment efficiency if the pore size is altered. Such estimates could be used to optimise treatment efficiency and so reduce the negative environmental impacts of aquaculture effluents. A 65 µm pore size microscreen would, for example, be expected to remove at least 30 % of the total phosphorus, >23 % of the total nitrogen and >30 % of the suspended solids. The model was also useful in proposing waste management strategies. Further work is required to establish if the model is generally applicable to other aquaculture facilities, or is specific to site or some other parameter such as feed quality or culture species.