GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Building Science

Geothermal Heat Pumps: Reputation vs. Reality

Data-backed analysis of energy savings and financial return on investment of ground-source heat pump heating and cooling reveals a system that doesn't stand up to the claims that it makes good sense in residential applications

Ground-source heat pumps use the earth as a source of heat in winter, and as a place to dump excess heat in summer.

Geothermal heat pumps have a reputation for being efficient and “green”; is it deserved? In contrast, feedback from owner experience, and substantiated by discussion on industry boards and forums, paints a less flattering picture of the realities of geothermal. This article explores the reasons for the yawning gap between the common perception of geothermal as an efficient, clean, and renewable heat pump system and the realities of operation that quickly dispel those myths.

Defining geothermal

Geothermal, broadly, refers to capturing and utilizing energy that naturally exists below the surface of the earth, and it is often described as clean and renewable. That notion more aptly describes utility scale operations that make use of heat venting naturally at or near the surface, but geothermal is also used to describe a range of small-scale heat pump systems employed for residential and commercial heating and cooling. This analysis is restricted to the latter, which based on our research should not be considered clean or renewable, though its reputation seems to benefit from the misguided association with the former.

Geothermal heat pumps are designed to make use of underground temperatures that are more moderate compared with air temperatures above ground. While temperatures of undisturbed subsoils vary slightly by latitude, the U.S. Department of Energy claims that “below the frostline, about ten feet down, the earth maintains a nearly constant temperature of 54 degrees.” That feature appears attractive compared with air-source heat pumps, which need to extract cooling from much warmer air in the summer, and heat from much colder air in the winter.

The untold realities, however, are how underground earth temperatures change with every geothermal exchange, and how slowly those temperatures return to ambient and optimal levels. The data in this paper demonstrate that underground temperatures are manipulated relatively rapidly and significantly by…

GBA Prime

This article is only available to GBA Prime Members

Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.

Start Free Trial


  1. Tim_O | | #1

    It would be interesting to know what the loop lengths are relative to the heat load of the house. It sort of looks like the data is showing that heat is either not spreading through the ground to even out, or the loop lengths are overwhelming the heat load of the soil. It would also be interesting to see the soil temperatures between two loops for example to see how far from the loops the average ground temperature starts to return. I thought geothermal systems operated at 32F in heating for quite a while since there is a lot of heat released as the ground freezes.

    With that said... The conclusions remain the same, it doesn't make any financial sense in low or high load homes. I even priced out renting an excavator or trencher (hard to find one that can go deep enough) and doing the whole install myself. By the time you buy all the fittings, pump station, lines, etc, you're already behind!

    It would be nice if the 30% federal tax credit counted towards all heat pump systems, not just geothermal. I know the IRA offers 30% up to $2000, which is significant, but on whole house furnace replacements, it's not 30% of the total from what I have seen on quotes.

    1. James M. Leaman, Ph.D. | | #3

      In a limited length piece, some detail is sacrificed for the focal message, which you allude to in your second paragraph. In our view the IRA tax credit on air-source heat pumps is a step in the right direction, and we can advocate for at least equal treatment (regarding limit) in future legislation.

      1. Tim_O | | #6

        I agree, and understand the point about keeping the message on point. When I look at that data, it looks like the loop lengths and total ground area do not support the heating load of the home. I'd guess the geothermal installer used typical "rule of thumb" install lengths, depths, etc for the system. In the theory of these systems, the ground is an "infinite" heat sink. This is showing it is not at all the case. Which then means, the further north you go, or the less efficient the home, the less efficient these systems get. That second graph with the 8 day average temperature is a real eye opener.

  2. iwatson | | #2

    Wow, great article! And it kinda makes sense when you think about it - the insulative properties of the ground that are responsible for "keeping temperature constant" also work against you when you start to draw energy out of the system as it becomes difficult for the ground to "recharge".

    It's becoming clearer and clearer that the trick to effective green building is simplicity.

    One quibble - while it may be pedantic I think it's important to clear communication to differentiate between "true" geothermal (i.e. systems that use the Earth's heat) versus systems that use earth heat (i.e. heat in the soil from the sun), and that the systems being discussed in this article should only be called ground-source heat pumps.

  3. James M. Leaman, Ph.D. | | #4

    Thanks for your feedback. Does the second paragraph address your comment about "true" geothermal? Your terminology classification is helpful.

  4. Expert Member


    Thanks for this useful analysis. Geothermal has always seemed to one of those interesting ideas people want to pursue, independent of whether it made sense or not. Given that, it's good to see this more clear-eyed overview.

  5. jim_gregory | | #7

    While I agree that high-performance air source heat pumps are more cost-effective than ground source heat pumps for most heating and cooling applications, how certain can we be that the behavior of this particular installation are broadly applicable elsewhere? As the article mentioned, there are many factors--horizontal vs vertical loops, loop depth, loop spacing, soil type, etc--that effect GSHP performance. Have there been other GSHP studies that found similar results as this one?

    1. James M. Leaman, Ph.D. | | #10

      While the loop bed temps report data from a single installation, feedback from GSHP installers and users (on forums) confirm this behavior is typical of horizontal closed-loop systems, and it is surprising when first encountered. That gap between general public perception and reality is what prompted our study and writing. We hope this prompts others to collect data on similar and different loop configurations to add to a growing database that can be more broadly applied.

  6. Longstory | | #8

    This is a useful article to get folks to think about the real science behind ground source heat pumps. While the data here are limited to a single installation, it reminds us to think about the total impact of these systems, and why they may not work well.

    Having said that, I have a one house experience in my own home since 2010 which has 4 vertical 300 ft wells in a closed loop installation, two Florida Heat Pump (now Bosch) heat pumps, and two small industrial circulation pumps. My loop temperatures are easily measured via access probes in the piping going to and from the units. I have measured these in all seasons informally since the installation.

    The data I have collected informally over now 13 years agrees in part with your graphs, but the seasonal differences of loop temperatures in my system do not. My loop temperatures to and from the wells vary with the seasons only about 10-15 degrees from what I believe is the baseline of 55 degrees. We are in Zone 4 BTW, as ground temperatures at 300 feet does change with latitude I am told. Our cave systems here in TN also maintain an air temperature of around 55 degrees. Mammoth Cave, which in not far away, has a constant air temp of 54 degrees year round.

    My units achieve a delta of 10 to 15 degrees at the unit when operating in all seasons, and compared to the propane furnaces and air conditioners they replaced, greatly reduced the cost of heating and cooling the home. Maintenance costs have been routine twice yearly checks on the systems and the loop pressures, and only one repair (under warranty at 5 years) to an evaporator coil has been required. I feel this is due to the units being inside out of the elements.

    I cannot comment on the energy expended to drill my 4 wells and place piping in 2010, but in simple energy bills I am happy with my ROI (I got the 30% tax credit in 2010). So, in short, I learned from this that improperly installed systems do not work well. I do not think this single case study reflects the broader experience or the broader science, or as I was taught in school by one professor "beware of the doctor with one case".

    Thanks for stimulating the conversation. I am sure this will trigger confirmation biases on both sides of the issue. My bias is that I am getting a new vertical well ground source heat pump for my downsized home that is currently being constructed.


    1. James M. Leaman, Ph.D. | | #9

      Thanks QH. Your experience is consistent with the theory of our case in that you have less variance of loop temperatures because your loops are vertical (deeper), and the fact that you have any variance at all is consistent with the loops being closed. We are demonstrating the case of a horizontal closed-loop system, which is the least costly to install, both in dollars and ecological impact. We agree that we would all benefit from a larger database of cases, and we hope this will prompt more installations with loop temperature probes.

  7. Mike_Wolfe | | #11

    It doesn't seem from your bio that you are qualified to write this article. While there are infinite trade-offs in engineering, there is always a "best" option. Your article hits on some good points but I'm thinking about a broken clock. I mean, 10 feet down for the probes? Is that an attempt at humor? You also skip over the disastrous impacts of good designs vs. poor designs to objectively critique a technology. How can you judge the merits of a system based on a design that didn't consider that pipes might freeze if not buried deep enough? In my world that would have been a non-starter long before digging. Are you aware that mechanical engineers are aware of all of your concerns and have practical answers, pros, and cons? No one energy solution will get us to the future. We can't get to the best solutions if those that are not experts in a given field are allowed to opine with any sort of weight.

    1. Tim_O | | #12

      It is a horizontal loop system, 10 feet seems like where that would sit. Going deeper means even higher costs, which is sort of the point of the article... It seems from everything here, he has a very standard geothermal design. At 10 feet, pipes aren't freezing from being too shallow, they are freezing because the heat load is exceeding what the soil is able to provide.

      1. Mike_Wolfe | | #33

        So, your argument is that if the ROI is not reasonable to go deeper, then you would simply bury them at a depth in which they would freeze? You wouldn't then consider a different boring strategy, depth, or another technology entirely?

        1. Tim_O | | #39

          My argument is not that at all, in fact, no one made that argument. Again, they aren't freezing due to lack of depth, it is due to a lack of capacity. In fact, the frost line in Alaska is around 8feet. But this house was in West Virginia, which has a frost depth less than 3feet.

          Considering a different technology is exactly what the article is saying. There is no point dialing in boring strategy, depth, etc... when the initial math shows an order of magnitude cost difference to an air source heat pump, no point in dialing in the last few digits when the first one is off.

    2. Expert Member
      MALCOLM TAYLOR | | #13


      It's generally taken as an admission you don't have any strong arguments to refute things in an article when you resort to dismissing it by questioning the author's credentials, rather than dealing with the data that was provided.

      Using your logic GBA might as well close up shop, as much of its very good technical content was written by Martin Holladay, who is a journalist.

      1. Mike_Wolfe | | #34

        It is one thing for an investigative journalist to query and report on the expertise of others. It is another to write an article under the guise of expertise yourself.

        1. Expert Member
          MALCOLM TAYLOR | | #38


          That's another straw-man argument. Neither James or Martin are investigative journalists.

    3. dadiy | | #16

      Since when does technical expertise in a focused field translate into balanced policy judgments involving multiple dimensions, including but not limited to engineering, ecology, economics, law, politics?

      1. Mike_Wolfe | | #35

        See response to Malcolm Taylor

    4. Tom_K | | #17

      I can tell you're not an engineer, because in practice there's almost never one best option. There's usually a Pareto frontier of many equivalently good options with different tradeoffs. When there are sufficiently many options, the choice can come down to what the engineer is used to, or a coin flip.

      1. Mike_Wolfe | | #36

        That is an incorrect assumption, but I actually agree with you. I was simply making the point that there is no fit-all to engineering solutions. "Best" could involve (2) "Bests", and may indeed result in a coin flip. But in my experience, often tipped by client preferences, there is usually a best option. Some sites/buildings are good for PV, some are not. Some sites are good for horizontal geo, some are not. Depth, loop length, pipe size, drilling orientation, etc., etc., etc. can all be adjusted to gain or lose COP. I just found the article to be woefully ignorant. Post #8 makes all of my points in a different way. Incorrectly designing or installing a system is not a good way to judge the merits of a technology.

        1. this_page_left_blank | | #37

          You seem to be missing the larger picture here. The point of the article is to examine the widely held belief that ground source heat pumps are the gold standard of efficient, green energy. He pretty clearly demonstrated that systems as they are being installed are financially infeasible and very high cost in terms of carbon footprint, as well as not performing as promised. Fixing the efficiency problems that plague most installs only solves one of those three problems. If the the fix for the efficiency problem involves digger deeper holes and/or making bigger ground loops, that makes the other two problems that much worse. It does not actually make a ground source heat pump a more attractive option.

  8. LukeInClimateZone7 | | #14

    I think it also deserves mention, albeit only in the comments section, that the circulation pump energy overhead relative to air source heat pumps is not trivial. At least that was a casual observation I've seen in a few installs in mild climates. I'm convinced that ground source heat pumps are significantly less efficient than air source, though they still have compelling value propositions on occasion.

    1. James M. Leaman, Ph.D. | | #22

      LukeInClimateZone7, good additional you have comparable data points that you'd be willing to share?

  9. AntonioO | | #15

    It would be great if the author could point readers to more details of the GSHP equipment and ASHP equipment under comparison (brands, heating/cooling capacity, COP values, total loop volume-as another reader mentioned) and more details of the houses (square footage, blower door test results, window/wall ratios for example) for the two compared systems.


  10. AdamT | | #18


    Where was the probe house located? I'm surprised that ground temperatures got so high, even 10 feet down.


    1. Expert Member
      DCcontrarian | | #19

      The ground is getting warm because they are pumping heat into it.

    2. James M. Leaman, Ph.D. | | #21

      Adam, West-central Virginia, in the Shenandoah Valley.

      1. AdamT | | #29


  11. Expert Member
    DCcontrarian | | #20

    I look at the charts and say "those loops are undersized."

    Clearly, if the loops were bigger, the seasonal fluctuation would be less, and there is some loop size where the ground-source heat pump becomes more efficient than the air-source. But that doesn't really solve the problem, it just points out two other issues with ground-source:

    1. It's hard to engineer the loops, because it's hard to predict what the heat capacity and conductivity of the soil is going to be. Anyone who's ever dug a hole knows you never know what you're going to find. It's hard to measure during construction or even after. I look at this article and see that it took at least a year to determine that the loops were undersized.

    2. The loops are the expensive part of the system, so there is a big temptation toward undersizing. A system may not pencil out at all once the loops are right-sized. And there is no benefit to oversized loops, so if there's doubt about what the proper size is the pressure is to go small.

    I haven't heard much about Dandelion lately, but this was the problem they were trying to solve. Their business model was to focus on a geographically small area, where they could get detailed knowledge about soil composition and behavior.

    1. James M. Leaman, Ph.D. | | #23

      DC_Contrarian, you comments add helpful additional context. For reference, the probe house loops were sized on equipment specification and Manual J load calculations, but lots of other variables are at play, including user behavior. In terms of soil composition, optimal clay soil was backfilled in the loop beds. And then you rightfully highlight the tradeoffs with bigger loops, adding cost in both dollars and ecological impact.

      1. Expert Member
        DCcontrarian | | #27

        I don't think you can really chalk it up to user behavior. I would think good engineering would be to size the loops for basically whatever the heat pump can put out. Looking at those graphs I'd say they're not even close to what I would consider properly sized.

        It sounds like the key is in the "equipment specification." Clearly what the specification is calling for is inadequate. What I would wonder is whether the local conditions are just particularly unfavorable, or whether the specification is, to be charitable, overly optimistic.

        If the issue is local conditions, then the question becomes how to predict and account for them.

        Here in DC, in the city most single family homes are built on 5,000 square foot lots, to the extent ground-source is popular it's done with vertical wells. The people I've spoken to seem to be satisfied with the performance, although I haven't spoken with anyone with a deep technical understanding. I'll temper that statement with saying that they're usually installed in houses that are super-insulated and well sealed. The comments I hear are along the lines of "my house is super comfortable in all weather, no drafts or cold spots," which likely is due more to the insulation level than the heat source.

        1. this_page_left_blank | | #28

          The installed size of loop already didn't make any financial sense. If the solution is bigger loops, it just puts an exclamation point on the fact that ground source heat pumps are rarely a good choice.

          1. Expert Member
            DCcontrarian | | #30

            Yep. And it's not something that a technological breakthrough is going to fix.

  12. Expert Member
    DCcontrarian | | #24

    "These recorded loop bed temperatures align with stories from geothermal installers who relayed the experience of excavating horizontal loops in the winter, only to find them encased in a tube of frozen ground. Geothermal equipment manufacturers are aware of this phenomenon and claim that heat can still be extracted from loop temperatures as low as 16°F. "

    Problem is, ice is a much better insulator than water. If the tube is surrounded by ice the fluid has to be much colder to get the same heat flow. Which leads to more ice.

    Once you start getting ice, it's really game over.

    Air-source heat pumps have an analogous problem, they run a defrost cycle when icing is detected.

  13. Expert Member
    DCcontrarian | | #25

    One of the defenses you'll hear of poor ground-source performance is that since the heat pumps are a niche market, it takes longer for innovation to arrive and be accepted. Right now the leading edge in efficiency innovation is the air-to-air market; innovation like variable speed compressors and coil temperature modulation have become routine there yet haven't really reached the ground-source market.

    Part of the defense is that the big sunk (literally) cost in a ground source system is the ground loop, which could potentially last centuries. Heat pumps last a few decades at most, so as advanced technology becomes available it can be retrofitted into existing ground loops.

    This article really shows the fault in that thinking. If the loop can't perform, it can't perform. No new technology above ground is going to change the facts in the ground. Physics is physics.

    1. Expert Member
      DCcontrarian | | #26

      Relatedly, the description of freezing of the coil is troubling. Ground freezing can exert enormous forces, it can heave roads and lift boulders out of the ground. That's just the kind of thing that could snap a ground loop. So much for your multi-century lifespan.

      I would argue that any freezing is a sign of a design flaw in a ground loop.

      1. Tim_O | | #31

        My curiosity is still regarding how wide spread the freezing ground is. Is it just encasing the tubing, or is it 3 feet from the tubing? It might help answer the question as to how to fix the issue. If it's only localized, then more piping in the same trenches might be all that's needed to help spread the heat load.

        At the end of the day, no one wants to really put the research and money into it that it needs because it isn't a worthwhile investment.

        1. James M. Leaman, Ph.D. | | #32

          This is a good question. Finding answers would require multiple temperature probes at increasing distance from each pipe, or documentation by technicians digging up loops. I agree that either of these seem drastic and further expense on a very heavy investment. For reference, the probe house in this case placed temperature probes roughly 6 inches from piping, but at different depths and horizontal locations.

  14. mrkawfey | | #48

    Let me edit the headline for this article for you:
    Geothermal Heat Pumps: Reputation vs. Reality
    Data-backed analysis of energy savings and financial return on investment of horizontal field-loop ground-source heat pump heating and cooling reveals a system that doesn't stand up to the claims that it makes good sense in residential applications.

    The biggest problem with this article is that you paint all geothermal with a broad brush and your conclusions about your test case have ZERO bearing on deep well systems. In fact, I'm not sure than any people with serious knowledge about heat pump technology needed this analysis to know that field loops are a mediocre choice at best.

    New designs for air source heat pumps with variable speed pumps and compressors as well as new refrigerant technology have been increasing the COP for ASHPs for quite some time. Field loop systems are somewhere in between well-based systems and ASHP systems in terms of basic energy transfer/storage/retrieval. As the cost of ASHPs comes down (which it will because it's an easy, non-invasive swap) and the performance goes up, it's only natural that the cost and impact of field loop systems become less viable every year. This was always inevitable.

    BUT none of this applies to GSHPs with wells (open or closed). The ground at these depths has very little communication to the above ground world and the thermal inertia is several orders of magnitude larger than what you have with soil at the depth of a field loop system. It's this thermal storage capacity that makes well-based systems so much more efficient than any other heat pump.

    There is already a lot of data (papers and studies) about the performance of the ground energy in these types of systems. They confirm exactly what we have experienced with ours in northern NY. We have 2 wells at about 400ft each. We are now on our second heat pump so the initial cost of the wells should be considered as shared by both systems (and will be further shared when we are on our 3rd or 4th or whatever number, system). The ground temperatures drop throughout the winter and they rise during the summer. The COP of the system is reflected in the delta-T changes, but will always be better than ASHP. Additionally, every year the average ground temperature changes as well. In a heating climate, you are taking more energy out of the ground in the winter than you put back in the summer.

    This reminds me of a study that was done in China where they looked at this saturation of ground temperatures over many years in different climates and modeled a system improvement where the ground loop fluid would be passed though rooftop solar collectors before being sent back down the hole. This is an inexpensive improvement that completely eliminated the year-to-year temperature creep.

    I realize that I am beginning to ramble, but the point is: There is so much more information I could share that you are missing about deep-well systems and their performance that you should scrub every mention of them from this "study". They will always be the most energy efficient version of these systems and as responses to climate change become more urgent, cost will become less relevant. I don't feel like doing your homework for you, but if you have specific questions I'll answer what I can.

    I forgot to mention that the choice of thermal grout will have a huge effect on efficiency and different grouts are formulated for different performance criteria. The effect on efficiency is large enough to be a first-order driver and can mask or overwhelm other signals when studying total system performance. Once again though, this has been thoroughly studied and technical documentation is available for those that seek to find it. Given the field loop and backfill choice in this single system seems to have been poorly designed, the only real conclusion you can safely draw from all the data presented is: "The field loop seems to have been poorly designed." Extrapolating any other conclusion to GSHP systems in general shows a lack of understanding about the contributions to efficiency made by the various components of the system design.

  15. user_736053 | | #40

    This article and the conclusions it draws seem to miss the mark completely! The data recorded and used to reach the conclusion is very limited and only based on a shallow, loop-bed-style system. Here in the Northeastern USA, we use deep vertical wells. It is easy for a well drilling rig to drill the well(s) for the system while on site for the drinking well. The "Environmental Impact" of drilling a well is quite exaggerated by the author, considering how quickly multiple wells can be drilled often needing little casing and with very little site disturbance. There are companies that use very compact well drilling equipment allowing them to squeeze into small urban lots. (See Dandelion)

    The ground temperatures at 300 feet certainly fluctuate much less than those measured at 10 feet or shallower. When the liquid travels 600 feet (300 down/300 up) it generates a minor temperature change of the ground around the well.

    The article only briefly touches on the most predominant style of ground source heat pump used in our region and therefore over generalizes in its conclusions. In fact, some drinking wells (such is the case with our own) can be used with a pond as an open loop system. This style of system is completely ignored as an option in the article and has very little environmental impact. We (and other homes in our region) have just ONE well which supplies both drinking water and the water for the heat pump. In our case, there is NO CHANGE to the ground temperature around our well, because we are not dumping or re-circulating a liquid back into it. This system did not require ANY additional wells and has a tested COP of 4.7 which has been confirmed during operation. (Try to get that out of an air source heat pump in the northeast during a late January cold snap.)

    So, while the author has provided some conclusions from a recorded dataset, I feel that the conclusions are is very bias and the data very limited in its scope. Therefore, I feel that it is naive to apply the criticisms to all GSHP systems in general and specifically to those in operation the real world of the Northeastern USA.

    1. Tim_O | | #41

      Can you elaborate on roughly what a 300 foot well would cost to drill with a rig already onsite and what tonnage that would support? Because from what I can see, that's around $10k+ alone.

      Otherwise, building a Pond is not exactly a low cost exercise... If you already have the pond and your well already supports the volume a geothermal system takes, that's an interesting concept. What is the volume of water your geothermal system adds to your pond?

      1. user_736053 | | #42

        Yes, a 300 foot deep well might cost $10k, but if you are building a house (and well) to last 100 years or more, that is a very minor investment; about the price of a fancy french door, which might last 20 years or so.

        We would only recommend an open loop if someone already has a pond on site. That said, a large percentage of homes in our area (southern VT) have ponds. And other clients may fancy adding a pond as part of their landscaping plan, so thinking ahead can be benificial.

        An open loop GSHP might add 8-10 gallons per minute of pure spring water to a pond. BUT considering we have months when our system only runs 6 minutes a day, the full output can be rather limited. In the warmest month our system ran 3 1/2 hours a day for cooling, which equates to about 1,900 gallons. For reference, a tiny pond (150 sq ft) holds about 3,300 gallons and a large (1/2 acre pond) holds about 66,000 gallons.

        I (and the article) also forgot to mention that the FEDERAL TAX credits for a Ground Source heat pump are 30%! This incentive can save homeowners significant investment on their HVAC system, in the neighborhood of $15,000. The federal incentives for air source heat pumps caps out at $2,000! That is a significant difference.

        I also want to make it clear that I DO NOT sell, represent, or install Ground Source Heat Pumps.

        1. Expert Member
          MALCOLM TAYLOR | | #49


          I'm having a bit of trouble with the idea of pumping 8-10 gallons of ground water a minute just to service a heat pump, and as an ancillary effect fill a surface pond.

          I'm also not sure it's useful to dismiss the high cost of equipment by comparing it to luxury items that might be included in a house - like 20k front doors. That logic could be used to justify any expense, reasonable or not.

          1. user_736053 | | #51

            Hello Malcolm, What trouble are you having? Do you think the flow rate is too high or too low? Do you have an issue with us using the earth as a heat sink? Is the fact that we are providing safe clean water to anyone downhill of us troubling you? Or do you feel we should be burning fossil fuels to do this task, instead of solar produced electricity?

            We have a spring on our site that is pouring about 25 gallons a minute into the pond (24/7) [36,000 gallons/day]already and we also have a 40 gallon a minute overflow (24/7) [58,000 gallons/day] from our well that is also dumping into our pond. A very occasional 9 gallons a minute sporadically [<1,900 gallons/day] is literally a drop in the bucket by comparison.

            I certainly recognize the initial cost of the equipment and installation seems significant. But when you consider the Federal tax credits, the State incentives and you calculate the annual operating cost savings, the math all works, even when compared to air source heat pumps. The ROI might be more than 10 years, but, typically our clients are retiring or keeping the house for the rest of their lives. Most even plan to pass the home down to their families, and are looking forward 25-100 years. With forward thinking like this, it is certainly easy to amortize a $10k drilled well over 100 years.

          2. Expert Member
            Michael Maines | | #52

            I have heard many times that "pump-and-dump" systems can ruin ecosystems and contaminate the groundwater. I don't know much about the specifics, that's what I've been hearing for many years now.

          3. Expert Member
            MALCOLM TAYLOR | | #54

            User ...679,

            Because of your spring, you have a singularly unique situation where pumping ground water just to use it for heating may make sense. Is that also true of the rest of your clients?

            As a general proposition bringing ground water to the surface for anything except household use is a bad idea. Most aquifers struggle to provide enough water for the nearby inhabitants. if they also get used for heating, and that water is simply dumped unused into ponds they will soon be depleted.

          4. Expert Member
            DCcontrarian | | #55

            Heat pollution is a real concern too. If you cause a stream to flow in the winter that otherwise wouldn't, you're affecting the environment.

      2. Longstory | | #46

        Coincidentally, my wife and I are completing a 2000 sq ft what we hope is a pretty good house and we just had ground source vertical 300 ft wells placed. The rig was able to get them about 20 ft from the rear of the house to make piping to the heat pump in the garage easier. The wells are in Zone 4 Middle TN. Soils are some topsoil over clay and limestone. The loops are plastic piping and the drill holes were then filled with a thermal grout to make heat transfer more consistent.

        The drill contractor spent about 4 days on the two wells and they are ready to connect now. Their bid was 15,000 for the job. The HVAC contractor is installing a 3 ton unit, two stage, for 24,000 including all ductwork. All work will qualify for the 30% tax credit this year.

        As I mentioned earlier, I had a similar but larger system in our prior home, and was pleased with that ROI, particularly the drop in electricity usage.

        How much energy went into the drilling I have no way to calculate, but I know that vertical wells can and do work.

    2. Deleted | | #43


  16. DavidJones | | #44

    An excellent article.

    I don't think its news to any builders who do energy modeling that the upcharge from air to source heat pumps to ground source heat pumps is never justified by energy saving.

    To the list of negatives for the ground source heat pumps, I would add maintenance risk. Geothermal systems are complex and therefore more likely to fail. Their parts are not readily available off the shelf. Working on ground source heat pumps requires more skill specialized skills than an air source heat pump. I sleep well at night knowing the air source heat pumps in my homes could be completely replaced in single day if needed and at a lower cost than a major service on a ground source system.

  17. user-869671 | | #45

    This is a really informative article , but the annual utility costs used in the ROI chart seem really low to me.

  18. Erikas | | #47

    This is a good article, it definitely shows how little is known about the 'efficiency' claims of residential geothermal installations, and I have no reason to doubt the veracity of the factual statements that the author has made... but it's important to note that the supporting evidence of the lack of efficiency comes back to a single, closed-loop installation that, to my eye at least, seems to have an undersized field relative to the heating/cooling demand of the house. That isn't to say that the installation was non-standard, but maybe the lesson learned here (on efficiency at least) is that the closed loop sizing rules used here aren't working.

    As a case-in-point, I've had a residential geothermal installation operating at my house for 14 years now and, although I need to calculate the data on actual efficiency from my data logging (this isn't my day job!), I can say with certainty that, although the well temperatures fluctuate by season, they still remain considerably warmer in the winter and considerably colder in the summer than ambient temperatures. Here's a graph showing my well temperature (inbound and outbound at the entrance of the pipes to my basement) for the past year:

    The warmest the well 'supply' gets in summer is a bit over 60F; the coldest the well 'supply' gets in winter is a bit above 45F.

    Although it's not plotted on the same graph (again, not enough time in the day), here are outdoor temperatures registered during the same period of time:

    You'll see almost the entire heating season is below 45F and almost the entire cooling season has been above 60F. (During the shoulder seasons, the heat pumps don't typically run much for obvious reasons.)

    I suspect, if you ran the efficiency calculations on a comparable home with air-source heat pumps, you would likely see a double digit percentage efficiency gain. It all depends on the size and configuration of your geothermal field... if it's properly sized for the application, you can see real efficiency gains... in my experience at least.

    Some other key points:
    I don't disagree with the author that geothermal may not be cost effective. Even just the heat pumps are incredibly expensive -- I just replaced one for over $20K not including well or piping -- in part because prices increased by 30% as soon as a 30% tax credit was available. I can't imagine that geothermal is cost-effective over any reasonable period of time. That said, although the author is correct that HVAC equipment like the heat pumps typically last 20 years, note that the field itself will likely last for far longer and that's where much of the initial installation cost is.

    Open loop systems don't require two separate wells, one for supply and one for return, although that's certainly a common installation. You can use a single vertical well pulling water from the bottom and returning to the top of the column in an open loop configuration... assuming open loop systems are allowed in your jurisdiction (they are here in New Hampshire). That cuts the vertical drilling cost in half obviously... for a loss of efficiency as you aren't tapping into the same area of water under ground.

    I won't speak to the embodied energy necessary in drilling a well (or a field for that matter), but the author is definitely correct that there is some there. In my case, running a drill rig every day for a week probably doesn't have a huge impact compared to other items in construction (concrete), but there's certainly some impact.

    My installation is a very unusual installation in that it's neither really closed loop nor open loop; the performance closely mimics as vertical well closed loop installation even though it's technically an open loop system. It was designed as an open loop vertical well system with a single well. When we drilled the well, we never actually managed to hit water. As we live on top of a mountain of granite (it's not called the Granite State for no reason), I hypothesized that, if we didn't hit water, maybe it would hold water. I filled it with community water and, sure enough, the water stayed inside the well. We therefore just keep cycling the same water through that vertical column much like a closed loop system from a performance perspective.

    Bottom line, my data is basically the equivalent of a vertical closed loop... it just isn't a pure closed loop.

    Sorry for all of the information here, but just wanted to respond with another 'real world' installation that has very different performance characteristics than the 'probe house.'

  19. Expert Member
    DCcontrarian | | #50

    What I would be interested in knowing is whether these results are typical, or whether they are highly dependent upon local conditions. If it's dependent upon local conditions it seems like it would be fairly simple to do a test, similar to a perc test for a septic system, that measures the heat dissipation properties of the site.

    1. Expert Member
      Michael Maines | | #53

      Same here. While the data is very interesting and confirms concerns I had, a sample size of one is not exactly scientifically rigorous.

  20. vpc2 | | #56

    Great to see this clearly presented. Thanks! Way too much pressure to use this kind of heating and cooling and is not energy efficient in the least, but they keep "pounding the drum" anyway.

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |