On May 1 of this year, I completed an experiment started one year ago. This experiment was designed to attempt to break a few of the elementary rules of keeping fish. The rules I broke were concerning tank size, and water changes.
I wanted to keep a small tank (approximately 1 gallon of water) healthy for over one year without doing any water changes. That is not to say that I didn’t top-off evaporation---I had to do that--- but since the time I set-up this tank, I have never removed any water from it.
The tank set-up does not contain any filtration other than a small powerhead with the intake connected to a flexible tube with is capped with a chunk of foam filter material. The powerhead is buried under a mound of natural gravel at one side of the tank, pointing up to create a little fountain, and the foam covered intake is under shallower gravel on the opposite side of the tank in order to keep the whole tank circulated.
Near the fountain
is a small piece of bog wood, and a plant
frequently sold under the name Brazilian sword
but more commonly
known as the white sail
houseplant. It is planted with its roots
under the water but with the majority of its leaves above the surface.
One other piece of vegetation is used in this set-up, and that is a common strain of a filamentous alga which is quite easy to harvest out of the tank by hand. This is a very efficient way to bind-up and remove fish waste from the tank, without doing water changes. The tank receives indirect sunlight by being located in a north facing window. There is no other lighting used for this tank.
Populating the tank, originally, were 3 zebra danios, one Otocinclus and a few pond snails. By the end of the first year of the experiment, there was only one danio due to a mistake I made regarding feeding the fish.
I started out feeding the fish about once a week, but then thought that the danios would eventually learn to eat snail eggs if I cut back on feeding them. I skipped feeding for a month or so near August and accidentally starved a danio---a fact that I did not notice for several days (a system like this is supposed to be left alone, and is therefore ignored most of the time). I believe it was the decaying fish, sending this little ecosystem into a nasty ammonia spike, which killed one other danio and the Otocinclus. Luckily, one fish survived to continue in the experiment.
After this event I went back to feeding the fish roughly every 1-3 weeks, and harvesting out a little of the algae whenever topping-off the evaporated water. I frequently lost a little bit of gravel which was pulled out attached to the algae, and this was also replaced (the natural mineral content of the lapis luster gravel buffered the pH up a little bit so I wanted to have it replaced occasionally).
When first starting out this tank, I monitored water chemistry with test kits several times a week, then it dropped down to once a month, or less. There were never any detectable traces of ammonia, nitrite or nitrate.
After one year, the only real deterioration that is observable in this set-up is that the brazilian sword has slowed down in its growth, and the leaves are going yellow. I assume this is a lack of iron, and will continue this experiment into the second year, altering my routine only to add very small amounts of iron based tablet fertilizer, and a second (younger) danio, in the hopes of keeping this experiment going for longer than the limited life of the one remaining, original danio.
As a second experiment, I’m hoping to eliminate the need to harvest
out algae by running a 20 long, and either using some sort of an algae
eating livebearer, or maybe an omnivore contained in a small cage
like a breeding trap, while some sort of small reproducing crustacean
(daphnia, fairy shrimp, or cyclops) are allowed to populate the tank
and are only eaten when they stray into the cage
(this will
hopefully prevent over-predation, and allow a continuing colony of the
food item to reproduce).