I am attempting to keep plants warm in a greenhouse. Therefore, I assume that heating the soil should be most efficient, as that's what plants seem to care about the most (i.e. the temperature of the roots appears to be more important than the temperature of the leaves, though I can't seem to find much evidence around this). If my assumption is true, this is convenient because greenhouses have lousy insulation (and are somewhat difficult to insulate retrospectively), so much of the heat loss happens via the air.

So, which is more efficient for heating the soil?

A) An electric wire element buried in the soil (i.e. a long wire connected directly to A/C power that generates heat).

B) A tank of water containing a heat resistive element and circulation pump with a hose running through the soil.

Intuition may suggest that the pumps are wasting energy, and some heat is lost via evaporation if the tank is not air-tight and insulated, but perhaps the water system is perhaps a little more versatile because you can take energy from other sources (such as the sun) to heat the water. Then again, you could use PV solar energy to power the wire element (though it could take a large array).

I'm attempting to prove which approach is more efficient, and a greenhouse is a difficult environment to control, so I'm struggling to get scientific results. Any insight into which approach generates the most energy loss would be appreciated.

Edit: From the comments, a heat pump sounds like the most efficient option. Caveat: This seems like by far the most expensive in terms of equipment and I'd probably have to buy off the shelf since it's beyond my skill level.

  • $\begingroup$ If I understand correctly you are only considering in option b) you are considering only a heat resistive element and a circulation pump. You are not considering a heat pump? Additionally, it would be nice to know the temperatures that you want to bring the soil/plants. $\endgroup$
    – NMech
    Commented Jul 27, 2021 at 12:03
  • $\begingroup$ Yes, correct understanding, I have edited to reflect. Yes, now you mention it a heat pump would also work, and sounds more efficient than option B. I need to bring the soil temp to 25ºC/77ºF (the ambient air tends to help maintain the soil temp, so raising the air temp may also be required if it's efficient to do so). $\endgroup$ Commented Jul 27, 2021 at 13:09
  • 1
    $\begingroup$ The water seems like a lot of extra hassle unless you actually make a solar still. $\endgroup$
    – DKNguyen
    Commented Jul 27, 2021 at 13:38
  • $\begingroup$ Believe me, it is (a lot of hassle). Speaking from experience. $\endgroup$ Commented Jul 27, 2021 at 16:18
  • $\begingroup$ Is the greenhouse in the sun? Typically that's enough. If you must heat, cost of use will be heat pump < gas furnace < electric resistance heating (tho in some parts of the US natural gas may have a similar cost to run as the heat pump). The cost to install is probably the reverse. Heat pumps also don't work well in extreme cold, which I assume is when you have the problem. It seems to me there is something here you aren't telling us. Regardless, if you are trying to actually heat the ground you're fighting a losing battle. Put plants in pots and heat the air. $\endgroup$
    – Tiger Guy
    Commented Jul 28, 2021 at 4:22

1 Answer 1


At these temperatures 25ºC (and provided that the outside environment temperature is not subzero or very close to zero ºC), the heat pump would be significantly more economic/efficient.

Option A - wire in the soil.

The main problem with option A is that the resistive elements could warm up the soil, but depending on the moisture of the soil you might have different conductivity coefficients (usually moisture improves conductivity). If the soil is too dry then you might have temperature buildup locally which will not be very good for the plants.

Another problem I forsee, is that if the greenhouse is not insulated then if you put the resistive elements too deep, then approximately half of the energy will be wasted (it will heat up the ground underneath).

Also if you put the wires at shallower depths you may run in problems if you have plants that need digging every year, and you may hurt the plants.

Option B

As mentioned earlier, the heat pump would be the way to go. It is much more efficient (COP could easily reach 4 or higher, i.e. for every electrical kWh it is possible to move about 4 or more kWh).

From the question I was not sure if you were planning to build a open or closed circuit.

IMHO, the closed circuit (i.e. warm the water and circulate it in a closed circuit in buried hoses and then return it - a sort of reverse geothermal installation) would be better off.

If you are planning on an open circuit (i.e. warming up the water and then watering the plants) then the main problems I forsee is with the mass/transfer distribution. I.e. there is a limit on the heat that you may transfer to the plants. So depending on the losses you may be able (or not) to provide enough energy to control the temperature in the greenhouse.

Solar collectors vs PV

Depending on the climate that you are (and more specifically solar irradiance and cloud coverage), it may be more economic to use a system with solar collectors

enter image description here

Figure : Basic solar collector setup eia.gov)

If you are for example located anywhere near the Mediterranean you are probably familiar and it actually makes a lot of sense to have a system where you heat the water when there is sun. If there is no sun you can have a heating element inside it (whether resistive or connected to a heat pump does not make a difference).

The good thing about solar collectors vs PV, is that they use the indirect radiation also (so even if its partially cloudy you can get a preheating), and additionally, the energy conversion is a better than PV system (typically in the order of 60%, while PV is in the order of 20-30% depending on technology).

  • $\begingroup$ Excellent answer. I'll get working on the solar collector. I suspected this would be much better than PV. Good tip about the local heat build up with option A; I had not thought about that! $\endgroup$ Commented Jul 27, 2021 at 16:21
  • $\begingroup$ And, when I get bored of DIY-ing this, I'll go out and buy a heat pump!! $\endgroup$ Commented Jul 27, 2021 at 16:22
  • $\begingroup$ How does a heat pump not also be affected by soil moisture affecting conduction and producing hot spots? It's still a heat releasing conduit in the ground is it not? $\endgroup$
    – DKNguyen
    Commented Jul 27, 2021 at 20:56
  • 1
    $\begingroup$ @DKNguyen the heat pump by itself its only warming the water in storage tank. I expect that there is a three way valve for adjusting the temperature of the fluid in the hose. So the temperature of the soil always approaches the target temperature of the water (you can either give or take away heat). With the resistive element, you can only give heat. Is that what you are asking? $\endgroup$
    – NMech
    Commented Jul 27, 2021 at 21:51
  • $\begingroup$ @NMech Sort of. The soil also approaches the temperature of the heating element so unless you're running the heat element hotter than the temperature you want the soil to be (which you might) you can always use closed-loop control to tone the power down. I don't imagine things heating up so fast or getting so out of control you would actually need to actively remove heat. I imagine a water system would need this monitoring too to work in the way you described since you could also run a water system hotter than the objective temperature then tone it down. $\endgroup$
    – DKNguyen
    Commented Jul 27, 2021 at 21:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.