Aquaculture Greenhouse: The Sand Filter

I recently found out that a new coworker is something of an expert on setting up large aquariums and their filtration systems. This is something of a boon to me, since I don’t know a lot about it and am currently doing lots of research to fill the cavernous gaps in my knowledge. I only got to talk to Mike for about 5 minutes on the subject, but he already found a big flaw in my plan. Fortunately, I think it is one I can fix.

As I mentioned in a previous post, I am planning on flooding the lower two feet of my future greenhouse and growing fish under the path between beds. The goal is to create a balanced aquatic ecosystem. Ideally, the organic compounds that leach out of the soil or come from the greywater will be enough to feed the phytoplankton in the water, which will then become the base of the food chain in my tank.

The problem Mike found is in my soil. My garden beds, which will extend 3’ above the water level, will be made entirely of organic matter, with minor amounts of sand and pea gravel thrown in for soil structure. Organic matter that is below the water line can be a problem, though, as it will undergo anaerobic decomposition and pollute the water with toxic chemicals. My solution was to fill the area under the soil with sand.

According to Mike, any sort of filter for biologically alive water will quickly fill with a sort of slime. It is a good thing, as it really helps filter the water. However, it will grow to fill the pore spaces to the point where no water will get through and the water will back up and drown my plants. The only way to stop this from happening is to stir the sand every couple of months to break it up. This is obviously not an option as the sand will be underneath 3’ of productive garden soil.

My first thought was to come up with some sort of tray system. The weight of the soil above makes horizontal trays too cumbersome to deal with, so I thought vertical trays every few feet might solve that problem. That still requires a structure around it to enable me to pull it out and there are lots of ways that could go wrong.

Another option would be to just put drains every few feet just above water level. That way, when it plugs, there is an overflow drain that would continue to function. This is still a very viable option, but I really don’t like it much. When the sand is completely plugged, it becomes not only unusable space under the soil, but also stagnant and anaerobic. I am not fond of that plan.

Then it occurred to me that the primary purpose of the sand was not as a filter, but as a spacer, keeping organic soil above the water line. What if I used something else as a spacer instead. The soil will still do most of the filtering necessary, so the sand isn’t really needed. The product that comes to mind most quickly is a system of grids that are used for rainwater retention systems. They are made of a stiff plastic and are designed to hold a lot of weight while still having a lot of open volume for water storage. If I could put those in there and a layer of landscape fabric on the top, I would have something that would most likely not plug and would increase the amount of water I could hold in there considerably. Of course, there are other structural features I could use. I could go all Roman aqueduct on it and use stone arches. That would be pretty cool and would last forever. It would probably be cheaper as well. Stacked cinderblocks would probably also work, but wouldn’t leave as much room for water.

So far, the open water solution is my best. It increases the volume of water for the fish to swim in from 1100 gallons to around 2500 gallons. But that leaves me with a new dilemma. What organisms do I allow to access that area. It will be very difficult to get in there to do any kind of maintenance, so anything that dies in there stays in there. Also, if I decide that a particular organism is a pest, I will really never be able to clean it out of the system without draining the pond and starting over. The cavern also gives the fish a good hiding place, which may or may not be a good thing.

I love this process.