A few thoughts on filtration

Originally written for the NZ Native Fish Study Group - by Dave Cooper

I would like to attempt to convey an understanding of the basic principles of filtration in an aquarium (or pond), and a simple overview of what is available in NZ and how to use it. These words are, of course, only my opinion but as I am a professional aquarist with day to day hard experience with most types of filters, and many other aspects of fish-keeping, I think it is a reasonably informed one.

Essentially, an aquarium filter is a system to force the contaminated water through the medium (or collection of media) and remove some, or all, of the contaminants along the way. They include the humble undergravel filter (UGF), in which the gravel itself is the media; the internal box filter; the ‘hang on the back’ power filter; the external canister filter; the wet/dry trickle filter; the fluidised bed, and a few more besides. The more primitive of them, such as the UGF, and internal box filter usually use an air pump as motor power. Air pumps are noisy and inefficient, but cheap. Most filters nowadays are powered by their own electric water pump and are referred to as power filters.

It is worth mentioning here that air pumps do not actually aerate the water by infusing air into the water but by circulating it and agitating the surface. Gas exchange, the absorption of oxygen and release of CO2, takes place at the surface, and circulation and surface agitation are ways to improve it. Power filters, properly set up, will also achieve this very well, making extra aeration from an air pump superfluous. This is a very common misunderstanding.
So, all filters are essentially gadgets for passing water over a media, the differences are largely in their efficiency, flexibility, reliability, flow rate etc. Your choice of filter for any particular job will depend upon a combination of many factors but for general use I find the external canister filter most useful. In particular, I have a strong preference for those canister filters of the German company EHEIM. Not only did Eheim invent the canister filter but they are extremely well made with a well earned reputation for reliability. Their unique (and patented) “in at the bottom, out at the top” design makes them very efficient. They are also user friendly and therefore more likely to be serviced regularly. They do cost more than most but the best usually does.

So, what are these contaminants that we hope our filter will reduce or remove? Well, they are mostly the result of fish metabolism and the breakdown of uneaten food, plant parts etc (nitrogenous wastes and phosphate) but can also include, at various times, medications (of many kinds), chlorine, copper, smoke, suspended solids, oils and any number of metals. All those old jokes are true, fish really do do nasty things in the water!

As nitrogenous wastes are the real biggie, I’ll deal with them first. The fact of life is that fish excrete ammonia, not only via their faeces and urine but also across their gills. Ammonia is toxic to fish at very low levels; as low as 0.1ppm, depending on the species, can be lethal and any at all is definitely detrimental. Interestingly, ammonia is more lethal at a higher pH and is one of the reasons why banded kokopu are better kept at a low pH as they seem to be sensitive to ammonia.

As luck would have it, ammonia is oxidised by a group of bacteria (Nitrosomonas sp.) into Nitrite. Unfortunately, nitrite is also toxic to fish. However, as it happens nitrite is relished by bacteria of the genus Nitrobacter who convert it into nitrate which is relatively non toxic (200 - 400ppm not being uncommon). These bacteria are referred to as nitrifying bacteria, the process is known as nitrification or “the nitrogen cycle” and is what happens in any biological filter (and nearly all aquarium filters are largely, if not wholly, biological filters).

These nitrifying bacteria are aerobic (need oxygen) and this is why a bio-filter should be running continuously or they die from lack of oxygen. There is also another group of bacteria called de-nitrifying bacteria that will break nitrates into nitrogen and oxygen. These de-nitrifying bacteria are anoxic, meaning they need a low oxygen environment and usually this only occurs in some specialist bio-media such as sintered glass (of which SIPORAX is my preferred brand).

The main factors affecting the number of bacteria present in your bio-filter, and therefore its efficiency, are available surface area and available food source. The more surface area your bio-media has, and the less it clogs up, the better your filter will work. Once again, sintered glass leads the field (although the options are numerous and they all work to one degree or another). SIPORAX has an available surface area of about 2102m per litre of media. Most other media fall in the 4 - 10m2 per litre category so you can see the difference is considerable.

The food source for these bacteria is basically the waste from your fish, so you can see that the population will fluctuate with your fish population and waste content of the water (bio-load). This is why you should stock an aquarium gradually and avoid sudden jumps in the bio-load.

When washing out a bio-filter it is important you wash it only in water from the aquarium as chlorinated tap water, or water of a different pH or temperature, can damage or destroy your culture of nitrifying bacteria.

Chemical filtration is also commonly used in aquaria, often in conjunction with biological filtration, and usually takes the form of activated carbon and/or ion exchange resins. Activated carbon is very useful at removing temporary pollutants such as medications, copper etc and at removing colours and odours from the water. However, it will not remove nitrogenous wastes and has the nasty habit of releasing its pollutants back into the water when it is exhausted or when the pH of the water is changed (especially if towards the acidic). Also it needs to be replaced often (every two weeks according to some authors) and the quality is variable. For these reasons, I prefer to use carbon for specific purposes, and short periods only. It is also important to remember to remove the carbon from your filter if you need to medicate your fish for any reason, as it will remove the medication.

Ion exchange resins are available in a number of different forms and can be used to remove a range of contaminants. Zeolite, the most commonly used resin, will remove nitrogenous wastes and can be recharged when full by soaking in salt water. However a good bio-filter will achieve this much more effectively. Exchange resins are best used only for specific applications in my opinion, although zeolite can be useful in short term applications or in times of variable bio-load when a bio-filter would be unable to keep up.

The third type of filtration media is mechanical. As the name implies, this is simply a physical trap for suspended solids. Usually this takes the form of dacron ‘filter wool’ or a foam cartridge, although a sand filter or diatom filter are other examples of mechanical filtration.

It is usual to have the mechanical media placed in a filter so that the water passes through it before the bio-media, as this will help prevent clogging of the bio-media. It is worth noting, especially for the do-it-yourself types, that not all foam is suitable. Some, indeed most, foams you are likely to get (except, of course, foams from an aquarium store) are biodegradable and so hardly much use as they will soon ‘melt’ into sludge.

So, as well as being reliable and easy to use, a good filter should provide mechanical and biological functions and at least the facility to add chemical media if required. Some media will carry out more than one function. Foam, for instance, when kept relatively unclogged, can become a quite reasonable bio-media, likewise activated carbon. Bio-media, if allowed to, can also be a reasonable mechanical media but these are not the functions for which they were designed and such use will inevitably lead to a reduction in the efficiency of its primary purpose.

One last factor to consider in turn-over rate; the general rule of thumb is to allow two turn-overs of the total volume of the systems, through the filter per hour (once every two hours in ponds). However, this will vary according to the individual aquarium. Those keeping the native torrentfish may like to allow 4 - 5 turn-overs per hour, while those keeping mudfish, perhaps only once.

Finally, I would like to point out that an article of this nature can only be a ‘once over lightly’ and the ‘advanced hobbyist’ reading it will probably have more to say about what I have left out than what I have included. For most of you I hope I have provided a practical overwrite of the topic and passed on a few handy hints without further confusing you.

Oh yeah, I nearly forgot. If you don’t want, for whatever reason, to have a filter or have to use a less than ideal one then remember:
Keep your stocking rate low
Feed sparingly. Good quality food only.
Do heaps of partial water changes.