[hoppycalif]

I'm not sure that the water temperature gradient in the unheated substrate can create any water currents. In a cold house the bottom of the substrate would be slightly above room temperature and the top would be slightly below tank water temperature. That is in the wrong direction to create any convection currents.

But, the substrate is suffused with water, so it will have a heat conductivity close to that of water, which is pretty good. That means there can't be a big temperature gradient in the substrate - heat is just lost through the substrate to the bottom glass of the tank, another fairly good heat conductor.

My guess is that the plant roots are never cold, with or without the substrate heater.

 

[hoppycalif]

Convection occurs because a substance has a lower density at high temperature than at low temperature. This causes the low density, high temperature substance to be buoyant, to try to float in the high density low temperature substance. So, if you can add heat to the bottom of a mass of water the heated water will rise, creating convection currents.

Without a substrate heater, or a heater under the tank, an aquarium will never have warmer water under the substrate relative to the tank. (An exception would be for those using water chillers to keep the tank water below the room temperature.) Therefore, none of us with conventional tanks will ever have convection water currents in our substrate.

It isn't difficult to calculate the probable temperature gradient in a substrate, given the room temperature and the tank water temperature. I'm not in the mood to do it right now, but I may decide to do that some time.

EDIT: Thinking about this some more: in nature the sun provides almost all of the heat for bodies of water. And, the ground is a heat sink, absorbing heat from the sun or from whatever is sitting on the ground that is heated by the sun. Therefore, I don't think natural bodies of water will experience "substrate" convection currents either.