Well, I'm thinking out loud here..
I really don't know for sure..this is all hypotheses
assuming the rate of no3 reduction is constant and that it only happens during photosynthesis... AND that the water exiting the reactor is not completely stripped of no3, I would assume that during light hours, if the flow is, say 100 gph and the cheato removes, say 5ppm... I can only assume that if the flow were half that, the cheato would remove twice as much because it would have twice the time to remove the no3..
whether it is sitting in a sump or in a reactor, it's basically the same thing...
unless it is so efficient that it is stripping all no3, then maybe increasing flow would not give the cheato the same amount of reaction time and this may prevent it from stripping all the no3..
Then, I guess the length of the reactor plays a role as well....
There seem to be plenty of variables.. not to mention type and intensity of the light source..
But.. all things being equal, I personally don't see flow as being a huge factor..
I guess there is an interesting experiment in all this hypothesis..
I'm pretty sure this didn't help you at all!!