This is related to issues around flow through the site, as well as sensor technology. The size of the waste stream is directly related to food conversion ratios, so the more you can do to reduce food conversion ratios, the lesser amount of waste you will have.
One of the things we found in some initial trials with St. Andrews Biological Station is that better flow through the cages tends to lead to a much lower residence time, so you are pushing these nutrients over a much wider area.
There is also a lot of good science from the Department of Fisheries and Oceans Freshwater Institute, as well as the St. Andrews Biological Station, relative to the actual assimilative capacity of the natural environment.
So the key is to try to site these units so that the flow is maximized, and then you're using other pieces of technology to reduce food conversion to its potential, and thereby reducing the overall amount of waste.