Two of the Three Geothermal Heat Maps Are Going the Wrong Way
I’m getting very close to the end of the geothermal/hydrothermal branch of the ocean heat source project. There’s a lot of heat coming through the seafloor, but can that heat actually explain the warming acceleration we’re seeing in the abyssal ocean?
It isn’t enough to show that the ocean floor is producing a huge amount of heat. Heat entering through the seafloor can get carried sideways, upward, into another basin, or away from the part of the ocean I’m trying to explain. So, instead of just comparing watts, I’ve been putting different geothermal heat maps into a 3D ocean model and following the heat month by month for 20 years.
And now, two of the three geothermal heat maps are giving me basically the same answer. The real changing ocean circulation is moving the heat in the opposite direction from what I need.
Three heat sources
I’m testing three versions of heat entering through the ocean floor. These sources are three separate ways of describing where heat from below might enter the ocean.
1. Conductive heat
This is the background heat slowly leaking out of the solid Earth and through the seafloor. Globally, that source is about:
≈ 1.68 × 1013 W
≈ 16.8 trillion watts
This map gives me a baseline geothermal source without concentrating huge amounts of heat in the younger parts of the ocean crust.
2. Lithospheric heat
The second map is what I’ve been calling lithospheric heat. Internally, the model is based on GDH1. The basic idea is that ocean crust cools as it gets older. Younger seafloor tends to have a larger heat flux, and older seafloor has had more time to cool. So, this map lets the amount of heat vary across the seafloor according to crustal age. The native heat map used in my transport model contains about:
≈ 2.889 × 1013 W
≈ 28.89 trillion watts
This source has a very different geography from the basic conductive map, which is why I needed to test it separately.
3. Hydrothermal remainder
The third source is the hydrothermal remainder. This is the extra heat in the lithospheric heat estimate after removing the conductive heat. It gives me a way to test the part of the heat budget that could be hot water circulating through the ocean crust instead of heat conducting through it.
My earlier estimate for that component was:
≈ 1.22 × 1013 W
≈ 12.2 trillion watts
Why I’m running each heat map twice :(
For each geothermal heat map, I run two versions of the ocean. The first uses the actual month by month ocean circulation from April 2004 through March 2024. The second uses a repeating seasonal climatology that has winter, spring, summer and fall. Instead of allowing the circulation to change from year to year, it repeats the same typical seasonal cycle over and over.
Everything else stays the same, and the difference is the circulation. So, I can subtract the two:
ΔH = HACTUAL − HCLIMATOLOGY
And for the monthly heat delivery into my target:
ΔPcirculation = PACTUAL − PCLIMATOLOGY
If that number is positive, the real changing circulation is delivering more of the geothermal heat to the target. If it is negative, the real circulation is delivering less.
Why the acceleration matters
Again, this project is not just about whether the abyssal ocean contains geothermal heat, because of course it does. The thing I’m trying to explain is that the warming is accelerating. So, the delivery of heat to the abyssal ocean has to increase with time. That is why I’m tracking a dimensionless delivery fraction, ff, and asking whether it changes fast enough.
df/dt > 0
A positive value means the circulation is progressively delivering a larger fraction of the steady geothermal source into the target. A negative value means it is doing the opposite.
First test: conductive heat
The conductive experiment was the first surprise. The real circulation definitely changed where the heat ended up. About 12% of the modeled conductive heat inventory was spatially rearranged compared with the repeating seasonal ocean.
≈ 1.27 × 1021 J
≈ 12.0% of the modeled inventory
So, yes, circulation matters, but it mattered in the wrong direction. In about 89.6% of the months, the real circulation produced less positive conductive heat burden in the target than the climatology did.
And the trend over time was negative:
df/dt ≈ −3.56 × 10−4 yr−1
I also checked that trend using Newey West uncertainty, because one month of ocean circulation is obviously not completely independent of the month before it. The entire 95% interval stayed negative.
Over the full 20 year window, the modeled circulation effect was:
Δf ≈ −0.0071
But, my earlier geothermal energy calculation said I would need approximately:
Δf ≈ +0.496
So the modeled circulation effect was:
≈ −1.43%
The minus sign here means the heat moves in the opposite direction from what’s required. And, when I looked specifically at the deepest 4 to 6 kbar layer, the same thing happened.
in the 4–6 kbar abyssal layer
≈ −5.52 × 1020 J
About 79% of those deep cells contained less conductive tracer under the real circulation. That closed one possibility: A steady conductive heat source cannot be made to look like the increasing abyssal heat source I need just by using the real changing ocean circulation.
Lithospheric heat map
Of course, I didn’t want to generalize from that result. The lithospheric heat map puts heat into different parts of the seafloor. And, if the source geography changes, the ocean might carry the heat differently. So, I repeated the entire 20 year experiment, and circulation again made a large difference.
This time it rearranged about:
≈ 2.37 × 1021 J
≈ 13.0% of the modeled inventory
Again, the circulation absolutely matters. Yet, once again, it moved the heat in the wrong direction. About 88.3% of the monthly differences were negative. The trend was:
df/dt ≈ −2.37 × 10−4 yr−1
The Newey West 95% uncertainty range was also entirely negative:
−2.92 × 10−4
to
−1.83 × 10−4 yr−1
Over the same 20 year window:
Δf ≈ −0.00473
The earlier geothermal calculation says this source would need:
Δf ≈ +0.2878
So the modeled circulation effect is:
≈ −1.64%
Again, wrong sign.
Deepest ocean makes the result even clearer
Under the actual changing circulation, the lithospheric tracer ended up with more heat in the shallower layers and less heat in the deepest layers compared with the repeating seasonal ocean. In the 4 to 6 kbar abyssal layer:
ACTUAL ≈ 4.93 × 1021 J
CLIMATOLOGY ≈ 5.65 × 1021 J
≈ −7.24 × 1020 J
And, about 78% of the abyssal cells contained less lithospheric tracer under the real changing circulation. The real circulation is not gradually pushing this steady lithospheric heat source farther into the abyss, it is pulling heat away from the deepest target.
Major ocean basins
I checked the Atlantic, Indian and Pacific separately because I wanted to make sure the global answer wasn’t being driven by one weird basin, and it wasn’t. All three major basins had a negative trend. So, I’m seeing the same basic direction globally and across the major basins.
The geographic pattern also didn’t help. The places where changing circulation added or removed lithospheric heat had almost no useful spatial relationship with the actual abyssal warming acceleration.
Using the real circulation actually made the geographic match slightly worse than using the repeating seasonal circulation.
So where does that leave us?
Right now, both conductive and lithospheric heat moved in the wrong direction when tracking with real ocean circulation. Right now I’m evaluating the last geothermal source geography in this branch.
Next: hydrothermal heat
The hydrothermal ACTUAL run is now beginning. It starts with zero tracer heat in March 2004. Then, I add the same static hydrothermal source every month and allow the real observed circulation to move it through the ocean until March 2024. I’m also saving a compact 3D copy of the tracer every month, so I’ll be able to watch the hydrothermal heat move through the ocean just like I did with the lithospheric source.
Then I’ll run the matched climatology experiment.
And subtract them:
ACTUAL − CLIMATOLOGY
We’re getting close…
This geothermal mechanism branch started with the knowledge that there is an enormous amount of heat coming through the ocean floor. We calculated that already, but what if the ocean is moving that heat into the abyss?
The answer so far is interesting… The ocean does substantially rearrange geothermal heat. But, for two very different geothermal source maps, the changing circulation has moved the heat in the opposite direction from what would be required to explain the accelerating abyssal warming.
So, we now know changing circulation is not helping these static geothermal sources by turning them into an increasingly strong abyssal heat source.
One heat map remains, and if the hydrothermal residual gives the same answer, I’ll be very close to closing the geothermal/hydrothermal mechanism as the dominant explanation for the acceleration I’m trying to explain.
Which means, yes… I may finally have to give electromagnetic induction heating my full attention.
