Originally Posted by
Paultergeist
Filtration Study Proposal: I have been contemplating a small-scale study to compare and contrast a couple of very different bio-filtration strategies. It seems to me that there are some very interesting ideas being discussed regarding bio-filtration, and a deeper insight into how these filtration methods operate could really be useful to me in my pond-planning. The three systems which I am going to try to compare are (a) conventional plastic media (b) Anoxic Filtration and (c) the Bakki shower. I thought I would describe the background behind what I intend to do, which the hope that perhaps others may have some good suggestions.
Due to economic constraints, this little study is rather primitive. The fish “tanks” used are plastic storage containers, each holding about 35 gallons of water. I will be using inexpensive “feeder” goldfish as the test subjects – probably 10 fish per tank. A measured and equal amount of food will be added to each tank, and I will be tracking water parameters on a daily basis (unless I have to leave town). I am going to try to get a flow rate of about two pond volumes per hour, subject to how well the cheap little power-heads perform. There are lots of various and obvious limitations to what I can manage within the space of my garage and my budget, but I hope to at least form a much more-educated opinion as to what sort of filtration methodology I want to use in my own up-coming pond build.
Conventional commercial filtration: I was on the verge of referring to this strategy as “inert plastic aerobic (submerged) filtration,” but that sounded cumbersome. With this strategy, I am attempting to emulate the sort of bio-filtration one might get from beads, K1/K3, poly-strapping, bio-balls, etc. My assumption is that there are no appreciable chemical interactions between the plastic substrate and the water (hence the use of the word “inert”) but that the function of the plastic media is solely to provide surface area for bacterial colonization.
Most of the opinions I have read on this type of filtration state that the bio-media should be well-aerated. I am therefore attempting to fluidize the K1 in order to keep the media fairly aerobic, and hopefully swirl it around a bit. From previous experience, I fully expect the media (once colonized) to cycle through the “nitrification pathway,” by which ammonia/ammonium is converted to nitrite, and nitrite is converted on to nitrate. Nitrate, however, is not widely reported to be reduced (i.e. “denitrification”) by this type of filtration. While far less toxic than the ammonia precursor, the nitrate concentrations when running these types of systems is often said to be cumulative, and is typically controlled via dilution (water changes).
My implementation of this system will be a plastic trash can with some K1 in it – I am going to attempt to fluidize the K1 with an air pump as well. The outlet of the K1 bucket will flow into another bucket – this bucket will contain a power head to return water to the fish tank. (I might toss in a handful of plastic bioballs which I just so happen to have as well).
Anoxic filtration: In using the term “Anoxic Filtration,” I am referring to strategy promoted/developed by Kevin Novak. There seems to be a continued level of controversy surrounding this filtration system. I have read opinions from both adopters of this system as well as skeptics. Most impressive – at least in my opinion – of the claims made with regard to this system is that implementation of this filtration system can reportedly result in no detectable nitrate levels. It has been stated that this absence of nitrate is due (at least partly) to the filtration media (“biocenesis baskets”) acting directly upon ammonia. The removal of nitrate via bio-filtration would seem to complete the nitrogen cycle, and thus may allow for reduced water changes, which makes this strategy intriguing.
Although Kevin Novak does not reference this specific pathway, one plausible explanation that might account for the performance claims associated with this system is a pathway known as the “anammox reaction.” In a nutshell, the anammox reaction is a biochemical process where (ammonia + nitrite) or, alternatively (ammonia + nitrate) are used by bacteria in an energy-yielding pathway. While the reaction does not seem to be poisoned by oxygen, the presence of significant concentrations of oxygen would favor aerobic processes (which would yield much more energy for the bacteria), so the anammox reaction most commonly occurs in nature at deep water ocean depths where the oxygen concentration is low.
My experimental system will consist of 2-3 “biocenesis pots” (plant baskets containing kiln-fired clay-based kitty litter with laterite). These biocenesis pots will be in a dedicated filter tank, with water entering the filter tank via a diffuser (pipe with holes drilled into it) and returned to the fish tank via power-head.
Bakki Shower: With a small footprint and low assembly costs, Bakki showers almost sound too good to be true. If what I have been reading is correct, some users are dispensing with using a Settling Chamber (SC) entirely with this method. While the direct off-gassing of ammonia is well-documented (air-stripping) for this type of installation, some users are reporting that showers are getting rid of nitrite and nitrate as well. These systems also reportedly cycle (establish bacteria) very quickly. In fairness to the other systems, I should note that the Bakki is requiring more wattage (a Danner Mag-Drive pump) in order to raise the water to the height required for a shower.
My experimental implementation will be three (3) 5-gallon buckets – each bucket approximately half-full with lava rock – arranged in a vertical stack such that water is pumped into the top-most bucket and passes through the three lava rock layers before returning to the test tank. The Danner pump will be submerged in the fish tank itself, pushing water up a length of flexible hose to the top of the upper-most bucket.
Objectives:
1. Track ammonia, nitrite, and nitrate levels during the cycling (establishing) of the filter media used.
2. Specifically, determine behavior of nitrate levels over time: (accumulation, leveling off, reduction, complete absence, etc).
3. Investigate the ability of the three systems to respond to sudden increases in fish waste and/or bio-load (i.e. deliberately over-feed).
4. Compare (roughly) presence of VOCs between the three systems. This would probably be accomplished with a 2-liter soda bottle-construction “Phoam Phractionator” (a.k.a. Protein Skimmer) and I’ll just make crude assessments based on foam produced.
5. Anything else I can think of.
I really do not know if this sort of comparison has already been done many times. I haven’t been able to find much in the way of qualitative performance comparisons for this sort of systems. If I am merely retracing the steps of others, please let me know. Any constructive suggestions are appreciated.