Water column to substrate nutrient exchange

Vladimir Zhurov

Lifetime Members
Lifetime Member
Aug 25, 2006
London, ON

I am curious how fast is exchange of nutrients between water column and substrate. It will depend on type/depth of substrate, type of nutrient, water currents and such, but I guess there must be some research that was done.

From more practical point. Say we start a new tank with plain sand (approx. 1+ mm diameter) and start dosing water column. How soon will substrate become fertile?




Lifetime Charter Member
Lifetime Member
Jan 24, 2005
Re: Water column to substrate nutrient exchange

If I understand what Tom has been saying, rooted plants prefer to take nutrients from the water column if they are availble, and thus usually don't develop as much root mass as they would in a nutrient-poor environment.

Given that, it apparently doesn't matter much what substrate is used, as long as it holds the plant in place.


Tom Barr

Staff member
Jan 23, 2005
Re: Water column to substrate nutrient exchange

It would depend in your case on how much organic matter is present in the substrate, that will bind labile forms of nutrients and be released by bacteria.

No bacteria, no breakdown of waste, no reduction etc.

Substrate grain size, so called "darcian velocity" (See hydrology) will play a large role in the flow rates.

You can also measure the flux by redox and O2 meters.
Tracer dyes
Steril jar test, with the nutrients measured in the water column vs time with the substrate in question.

But as you can see with the jar test, it's not very useful for testing how long the substrate will become "dirty sand" since that depends on plant biomass, decay, plant types, wood, and other inputs into the tank like fish etc.

Hard to sand how long and how much the substrate wuill hold before leeching back will occur. More CEC, the better it will fill up with nutrients in general.

As the plants and bacteria remove the nutrients from the CE sites, they are available for nutrient binding once again, so it is a continual process, not a static one that reaches a set point really.

Tom Barr