For SPS, all you need to do is make sure your major ions (alk, calcium, mag) are at appropriate and stable levels, that your nitrates and phosphates remain very low, and that you have appropriate flow and lighting. The corals don't care how you do that.
I would say that you're probably fine using a canister filter in the kind of set-up that you're describing. Corals produce very little waste, and unless you're over-feeding them, you're not going to have much material in the tank to break down in to nitrates or phosphates to begin with, which is the major claimed draw-back of canister filters. The major benefit of sumps over canister filters is that they're a good place to put equipment that you don't want to see, and they add extra water volume, but I don't think there's anything extra special about them from a filtration point of view. If you think about what normally goes in a sump - heaters, skimmers, extra live-rock, dosing and auto-top off equipment - none of those things NEED to be in a sump, and how is keeping a chamber of a sump filled with live-rock any different than filling the body of a canister filter with live-rock?
In this application, where no fish are being kept, and there isn't even a skimmer planned, I don't think it matters where you keep your live-rock, be it a sump or a canister. The rubble in the canister will be as much of a detritus trap as it would be anywhere else, and it's still going to get populated by all the pods, sponges, worms, and bacteria that it would if it were in a sump.
The only thing I'd say is that you still might run in to a problem with nitrates and phosphates rising over time, but not because of the canister per-se. Only because in a system like this it looks like the only nutrient export you'll have is through water changes and what little de-nitrification you'll get from the anaerobic cores of the live-rock rubble in the canister. Since rubble has much more surface area than large chunks of live-rock that could more easily fit in the sump, I don't think there will be much true de-nitrifcation happening in your rock rubble (this would be true of the rubble no matter what vessel it's placed in). That means that the bulk of your nutrient export will be through the corals/algae absorbing it, and water changes. You'll need to feed the system very lightly for this to be sufficient to keep levels in check long term, or do larger/more frequent water changes. However, with such a light bio-load, you could always run it for a few months to see if this even became a problem, and if it did, there's several hang-on-back skimmers/reactors that could house whatever nutrient reducing media you wanted. You would have the same problem if you had a sump filled with nothing but live-rock rubble, you'd just have a more convenient place to put the equipment.
|