Any bacteria that does become water borne is easily removed by skimming. This keeps the system healthy and stops the accumulation of unwanted forms of bacteria ie cyano.
Capn...have you discovered something different than the AA articled discussed, where the rate of removal was only 28-39% at max and best case scenario? Sources?
http://www.advancedaquarist.com/2011/3/aafeature
It also points to much lower bacteria counts in a carbon dosed reef tank, and that skimming is not effective in removal of excess bacteria.
The reason we want lots of flow through the reactors is to bring more nitrates and phosphates to the bacteria and not the other way around.
We really want to bring them what they can consume with the turn rate in the reactor, but based on the science that I have found, it is the oxygen that seems to be the limiting factor, have you found something else?
http://scialert.net/fulltext/?doi=jfas.2012.150.161&org=10
This article seems to be suggesting that the increase in oxygen levels corresponds to an increase of bacteria in the system.
My experience agrees somewhat, but do we want the flow so high that bacteria leave the reactor in large quantities from being sloughed off or do we want to contain them as much as possible? Only asking here, because I have run high flow and low flow and have recently been just running them in the bottom of a canister filter and have found no difference in keeping my parameters in check. High flow did however cause the pellets to lose size much faster.
If you dose the system with carbon then you are increasing the bacteria level in the system and this can cause some problems that switching to the pellet reactors solved
Explain, having done liquid and solid carbon dosing since about 2005/6, I see very little difference in the end result, but find the pellets easier.