Control the aquarium experiment – reasoning

Big Fish article, highlight, know Tags: , ,

What does it help to say we can grow all the plants without any algae if we have no idea on how we did it? And if we suspect we know how we did how can we confirm our suspicions if we have nothing to compare with? Before looking for the results it pays to understand the design and method of exposure. After all, understanding the conditions of the experiment is as important as understanding and interpreting the results of the experiment.

Shall we experiment to find the answer?

When I say experiment I actually mean a controlled experimental study as opposed to random observations or an observational study. I will explore in detail the differences between different study designs in another article; all types of study designs are informative but we need to understand the limitations.

I blame one thing and one thing alone:

To rationally and systematically improve our knowledge of the planted aquarium we can create two aquariums that are as similar as possible in everything except one thing. Plants in one aquarium will act as a control and plants in the other, with the changed variable will be the exposed group. For example, we have aquarium A where we provide light, CO2, substrate, water and a full EI for the plants. Aquarium B will have the same dimensions, same water quality, same substrate , same light, same CO2, same plants but there will be full EI and 5ppm PO4 /week additionally added.  Because EI is the standard that most people use we will consider plants in Aquarium A the control, and plants in Aquarium B the exposed. After a pre-determined time, we can compare the plants in the two aquariums (and aquariums themselves) and see if there are any differences. The differences observed are most likely to be due to the extra 5ppm PO4. The simultaneity and similarity of the two growing setups allows us to eliminate many temporal and spatial confounders and effect modifiers. More on this later, just know that because the aquariums are similar and ran at the same time we can more easily compare them.

How similar is similar?

The two aquariums should be the same in as much as possible. Some things are more likely to influence the outcome of the experiment then others, but we cannot quantify the influence of every small aspect. For example, if one aquarium is on the left and the other on the right it is less likely to affect the growth of plants than if one aquarium receives more sunshine then the other.

For a quick example of one of my setups  I will publish another article “My controlled experimental aquarium – setup and procedure”. Subscribe to the newsletter to know when it is published.

By creating this setup, we are able to apply the scientific method on some popular questions regarding planted aquariums. I will be the first in saying the setup and instruments used are not lab grade quality. While this imposes certain limitations, the scientific method and critical thinking still applies. The setup can help establish certain working theories.

Leave a Reply

Your email address will not be published. Required fields are marked *