Rocks. Squeezing things in is harder than squeezing things out.

Contents

Polarities & pressures; Predictabilities: Averages aren't reality; Injection infection; Phase flops; Surprise; Fractured reservoir corollaries; Geothermal corollaries; Embracing the different paths, but understanding the difference.

Squeezing things into rocks, is not perhaps a technical term. ??What am I thinking?? Fluids, and also fractures.

Polarities & Pressures

There are certain things in life that have a polarity to them.? A direction.? It’s much easier to swim down a river than to swim up it.?? Just ask a salmon.

It’s much easier to collect the water from a roof into a water butt than to put it back up there again.? It’s much easier to blow a feather down a set of stairs than to blow it back up again.? When we vacuum the dust up off a carpet, we kind of know where it is going to go – into the vacuum cleaner bag.?? When we reverse the vacuum and blow the dust out, we don’t know where each speck is going to settle.

All these things are illustrations of things that have a direction to them.? It’s easier going one way that it is going the other or knowing what’s going to happen one way than going the other.

The one thing we learn very early on about rocks is that they are heavy.? If we find lifting a boulder hard work, imagine what several km of rock on top of our head would feel like. ?

In fact we know this also of water.? It’s heavy too.?? Lugging a bucket of water around all day would be wearying.? So again, imagine not just a bucket, but some km of water on top of us.? No joke.

The deep earth experiences just that.? Where there are pores connected to the surface in some way, they carry all that weight of water entrained in the rock above them in the deep earth.? All sedimentary and volcanic rocks are entrained in this way in the early stages of their surface deposition before they are buried further. The rocks surrounding those pores at depth are in contrast carrying the weight of all that rock above them.? And if for some reason, the pores are blocked from connecting to the surface, perhaps by impermeable layers, then the rocks and the water share that load of stuff on top of them.?

That load ?- it’s heavy.?? It imparts pressure.? Lots.

Now fluids in the subsurface move around typically because of pressure or chemical gradients.? Pressure can make things do stuff that is counterintuitive.? We point our hose up and the pressure in our hosepipe makes the water go up, before gravity overcomes it, and it goes down again.? Temperature might indirectly have an effect too by influencing the density of fluids, with denser cooler ones sinking under gravity. Salinities can have a similar effect.? But it is pressure that rules the roost.? Things at high pressure like to move towards things at low pressure.? That’s why we have wind.

Pressure gradients or differences in the subsurface are imparted by various things – rocks compacting at different rates, with different levels of water escape to the surface in different places.?? It’s like standing on a plastic ketchup bottle with the cap closed or standing on it with it half open.? Tectonic forces can impart pressure gradients.? So can simple topography.? The weight of a water column where there is a high mountain can impart a pressure gradient relative to an adjacent basin where there is not.

What we know though, from this weight of water and rock which the deep earth always experiences, is that the pressure from above, down there, must always be fantastically huge compared to what we experience at the surface.?

Here on our fragile edge of space, we have some weight to deal with, but it is just the weight of the air above us, not the weight of rock and water down there.? That means if we ever drill a hole that connects the surface to the deep subsurface, and if the fluids in those rocks are bearing even a small portion of the rock load above them, then they will find that hole if there is a connection.? They will want to zip to it through any permeability passage they can find.? Like bees to honey.? That might be a convoluted passage such that it takes time, but there is no ambiguity about where they are going to.? To the source of lower pressure.? To the hole. A low pressure “wave” will emanate from it and call the fluids home.? Whatever the permeability pathway, whatever it’s nature, it will be that low pressure hole seeking. ??The direction of travel will be unambiguous.

Now think about going the other way.? Trying to pump water or anything else from our surface back into those rocks at depth. ??Even had one of those long thin party balloons that your kids want you to inflate and no matter how hard you try it just won’t blow up?? Imagine that a millionfold. ?Pumping fluids back into rock takes a lot of pressure.

More than that, it is unpredictable.? We no longer have a point of low pressure surrounded by rocks of higher pressure whose entrained fluids all want to find the unambiguously low-pressure point. We have a single point of higher-pressure disseminating fluid into a variable permeability maze of slightly varying pressures within the rock layers.? We don’t know the intricate geometries of every fracture, of every pore “throat” (the narrow connecting passages between pores), so we can never know precisely how that will play out. ?They can choose one of many different paths, each one of which takes a certain pressure to reach into and displace any existing fluids there.

This is not to say nothing can be done.? ?We know enough about porous and fractured rocks to get a feel for how things typically happen on average, so things can work to some level.? But we can never know the detail of what is actually going to happen, going the other way.?? And the thing about permeability, is where it is good, fluids don’t bother with the hard work of fighting against where it is poor.? The very definition of permeability involves how much flow occurs for a given pressure difference.? The more flow for the less pressure difference, the higher the permeability.?? So those fluids aren’t going to bother with travelling into low permeability parts of a rock if they have a high permeability pathway to access at an easier lower pressure. ?They may well prefer to travel a long way on a highway than a short distance on a B-road.

Predictabilities: ?Averages aren’t reality

What does this permeability pathway complexity mean?? It means that you might have a great idea of where you want your fluids pumped into the rock to go, but if they happen to find a high permeability pathway that’s different, that’s where they are going to go.? ?It’s like herding cats.?? It’s one thing calling cats home when it’s feeding time – they know where they want to go and you can rely on them coming.?? It’s another thing directing where they go after they’ve been fed.?? They’ll find their own path.?

The problem is that while we have a decent idea of how things average out in rocks, fluid flow is not dictated by average character of rocks.?? It’s dictated by the precise minima and maxima of certain parameters in any given location, and we can never know that in detail.? We can estimate it.? We can estimate the variability present within volumes.? But we can’t know the absolute value at any one place far from our wellbore.

So, if we hold to an idea in our head that putting different fluids back into an oil or gas field is just the same thing as producing fluids out of it, but in reverse, then we suffer a misconception.? One of them is rolling a boulder downhill, one of them is rolling it uphill.?? You can guess from what we’ve visited already, which is which.

Now I don’t want to overplay the seriousness of this.? It isn’t a show-stopper to doing things, it just means we have to be realistic whenever the subject at hand is putting fluids back into rocks, or predicting fluid movement in rocks.? Whatever that fluid is.? It’s harder, riskier to predict. ?It’s different to extraction.? Very different. They are not magically interchangeable or mutually reversible.

It also means if we are doing anything in the subsurface that involves inducing desired chemical reactions – for whatever purpose – we have to fully understand that we are totally at the mercy of these permeability pathways in terms of what in-situ reactants we can access with our in-fluid reactants.?? We won’t be accessing everything, and there may be large, large volumes we simply can’t. Furthermore, the reaction products might affect permeability in difficult to predict ways that might hinder further mixing of reactants. ?

The best permeability pathways will win out, and they may or may not be going the places we want. ??That leads to uncertainty in performance in complicated injectivity projects. The moment we have a good permeability pathway available, - if it can take the volume - anything less good is out of bounds for fluids being injected at pressure.

Injection infection

Such problems can be exacerbated if we also want to maintain the pressures in the reservoir or we want to “sweep” oil towards the hole, so that we inject water to try and “craft” this.? ?If the porosity and permeability is very evenly distributed in a rock – for example in a nice thick sandstone, then this can work well.? The moment however we get localised higher permeability pathways – which can be thin beds or channels, or fractures, the risk is our injected water doesn’t sift evenly through the sandstone, but zips along the permeability highway direct to the producing well, and we start to produce our injected water instead.? Not the desired result.?

Exactly the same thing can happen in geothermal wells, when we want to produce hot water from deep below, but we might also want to maintain the pressure in a reservoir, which we do by re-injecting the cool water from which heat has already been extracted for use.? Ideally, we want to re-inject that water far away from any production well so that the cool doesn’t interfere with the warm.?? However, if our cold-water injection happens to find a high permeability pathway to the producing well, we suddenly start producing cooler water rather than warm water.?

If we are exploiting a rock where there are lots of fractures (and all rocks are fractured to some extent), this can be especially risky, since the fractures can have dramatically higher permeabilities than pore throats within the matrix, and so are often preferred by any fluid with slightly higher pressures, such as those injected will have.? ?If production of oil or gas in petroleum applications or hot water in geothermal applications also involves the artificial enhancing of existing fractures (e.g. EGS in geothermal applications) or production of new fractures (fracking) then this risk is increased.??

With a good knowledge of a stress field in a rock, and the rheology of a uniform unit, it is surprising how accurately these artificial fractures can be engineered.?? However, what is always slightly less known is their interaction with geometries of pre-existing fractures and faults, and that is often where things can get more complicated than anticipated, and fluids can go where not originally intended.

The process of engineering those fractures involves pressuring up, and precisely how those pressures interact with pre-existing fracture permeability pathways is never known perfectly.? In worst case scenarios there can be chain reactions of “popped” communication with faults at greater depth, and release of inherent tectonic stresses. That is to say seismic events induced by the newly elevated pressures communicating along fractures and faults.? This always happens to some minor, very small, perfectly acceptable degree and is always closely monitored, but occasionally there are larger surprises.?

These miscellaneous varieties of problems relating to sneaky permeability pathways, can force shutdowns of expensive wells and severely impact the economics of projects.

Phase Flops

Things get even more tricky if we start thinking about different phases moving around in the subsurface.? Historically the first prime example of this has been water breakthrough in oil and gas fields. That gets into the whole science of relative permeabilities – when we have two different phases in the pores of a rock, or even three, such as oil, gas, and water, then one of them might travel through the permeability with greater ease than the others, for a given pressure difference.? Sometimes what this means is that the water around an oil or gas accumulation finds the lower pressure of the hole through the available permeability first, ahead of lots of oil and/or gas still in the accumulation, and obviously that’s not what is wanted.?? The water “cones” into the wellbore a bit like an inverted whirlpool.

That can mean we want to limit the pressure difference imposed by the producing hole, so that these differences aren’t exacerbated.? We don’t suck so hard, and the water isn’t so “tempted” by the hole. Things happen then more slowly, and production rates might be lower, but it is more controlled, and cumulatively over time we get more out.? In? contrast, once the water breaks through, there is much less that can be done to reverse it, and the total cumulative production will be reduced as a result. ??

Surprise

So, in summary, we can see that putting things into rocks at higher pressure – fluids and fractures – is inherently less predictable than using a lower pressure hole to retrieve fluids from them.? We can infer behaviour from the average properties of rocks we have encountered in boreholes, but we cannot know the detailed intricacies of permeability pathways in vast volumes of rock around them, and it can always hold surprises and hard to workaround constraints.?

The pressures we use to inject fluids means they find things in the rocks, like our unherdable cats, in the ways that suit them best.? That’s not always the way we would have them go. Most of the time those surprises are manageable, but sometimes they cause problems, and in the worst case scenarios, abandonment of wells. ??While extraction also has uncertainties arising, and chances of stranded resource being left behind because of permeability variations, we at least have much more control over the direction of pressure gradients with respect to a producing borehole, even at a distance from it. ?The lower pressures are eventually “felt” and induce migration towards it.?

Where fluids different to water like oil and gas have found their way into these lower permeability zones in the first place, it is in part due to sheer vastness of geological time.? That is not something any injecting well can ever replicate.

As already stated, what I’m not wanting to do here is overdramatise the scale of the problem, but it is certainly important to dispel any misconceptions that extracting fluids out of a volume of rock has the same level of control as putting fluids back into a volume of rock.? There will be large parts of that volume that are unreachable from our well simply by virtue of the permeability pathways that exist in three dimensions, and the timescales we wish to operate on.? Push back is not the same as pull out.?

We can attempt to increase how much of that volume we “see” by increasing injection pressures – but that also runs the risk of “popping” small fractures or faults or other permeability pathways we don’t know about, that just sends the fluids further away down those pathways, rather than soaking further into the rocks where we want them to go.?

Putting things into rocks – fluids, and fractures, is then an inherently more uncertain affair than taken things out.?? The pressure difference between our surface and the rocks at depth imposes a polarity that matters.? ??Going against that polarity is much harder than going with it.? Going into the wind is much harder than going against it.? Deep Earth’s geopressure wind blows upwards. ?Taking stuff up is easier than pushing stuff down.

Fractured reservoir corollaries

One corollary of this is not to lose sight of the inherent uncertainty that fractured reservoirs bring.? They work marvellously in many areas of the world, whatever we are using them for, be it hydrocarbon of geothermal – but the permeability contrasts within them are inherently stronger and that means things can suddenly go further and faster than we anticipated, as certain pressure difference thresholds between pathways are “popped”.?? Every reservoir is a fractured reservoir to some extent, but some way more than others.

It can work for us.? It might be great if we are producing oil from a huge-fractured anticline in the Zagros with a big oil column.? It might however be a disaster if we are injecting cool water into fault zones that finds our geothermal hot water producer, or if we connect prematurely our water column below a thin oil column.?

If we are also introducing artificially enhanced or new fractures via EGS (enhanced geothermal system) or fracking into such a fractured reservoir, it also affects this chance of accidentally creating unwanted permeability pathways that circumvent production strategies. Often the potential prize is significant and worth these risks, but it is good to be aware of them.

Geothermal corollaries

It's worth noting that a key driver for EGS fracture enhancement in geothermal contexts is increasing fluid flow in deeper hotter rocks, that happen to be tighter, for geothermal power applications.? ?It so happens that a wide range of diagenetic reactions start to kick in around 3 km depth and so EGS is concerned i) occasionally with tight but hot rock shallower than that or ii) more normally rocks below those depths, that are tight precisely because of their depth below 3km, where porosity destroying chemical processes set in pervasively. ???

This use of EGS is also driven by the fact that a typical Earth geothermal gradient of 25-30 deg C/km, with an average surface temperature of 15 deg C, puts the 3km “floor” of nice easy porosity in the 90 to 105 deg C bracket, whereas typically for any hope of commercial power generation, temperatures of ~ 140 deg C are needed, which is more of a starting ~ 4km affair (115-135 deg C) in a normal geotherm.? So geothermal power in areas of unelevated geothermal gradient is looking at depths where diagenetic reactions in rocks are very evolved – and that means porosities lowered.? We see this reflected in geothermal well success rates.?? They tail off almost step-wise significantly below 3km depth.? Hence that is EGS’s home turf.?

The nice thing about less than 3km is that compaction is the dominant porosity reducing process so good porosities are often present in sedimentary rocks.? To about 2.5 km they are also reachable by cheaper investigative slimhole “finder” wells.? The catch of course is twofold – firstly that predicting permeability variations in sedimentary rocks is no picnic either – yet this something exploration geophysics, facies model and sequence stratigraphic understanding, coupled with prior well and new finder well investigations, can help.?

The second catch is that that these shallower depths are cooler.? Yet we have in that 2-3km window a geothermal direct use heat “goldilocks” spot, where the porosity and permeability is still good, and the temperatures are decent too - in the 65-105 deg C bracket.?? That is more than enough for a whole lot of significant direct heat use applications.?

What are we saying here?? We are saying that if our objective is not geothermal power but geothermal heat, there is a lot to be said for keeping it simple and going for that hydrothermal stuff between 2 km and 3 km driven by still decent primary sedimentary porosity and permeability, and with no real need for EGS and all the issues surrounding fractures and fracturing.

The difficulty will always be mapping that useful primary poroperm at 2-3km depths.? But here’s a thought to take away – we only have to map that carefully once for that information to be available for all posterity, and likewise incrementally added to by whatever data gathering comes next.? That ever increasing information bank on subsurface permeability, at a 2-3 km depth especially useful to us for heat, has value for centuries upon centuries of future application.

Embracing the different paths, but understanding the difference

So whatever we are doing in the subsurface, if it introduces new things – we need to guard ourselves against the misconception that things are just going to replicate what happened as we took stuff out – and also the misconception that permeability pathways – including fractures - are always well behaved, or fully known.? That makes putting stuff in more complicated than taking it out.? The two processes won’t be talking to exactly the same volumes. Putting stuff in isn't taking stuff out run backwards.

Beep beep beep beep – this geology is reversing? Nope, not a thing…

Image credits in order of appearance:

All images are from Pixabay suppliers:

1.????? RaquelxBaires (Raquel Baires)

2.????? 12252435

3.????? gentlegiant27153

4.????? PublicDomainPictures

5.????? Wal_172619

6.????? ArtTower (Brigette Werner)

7.????? SwidaAlba (Kang-Rui Leng)

8.????? A_Different_Perspective (Silke)

9.????? Simon

10. JuliaSchSch

11. TimHill (Tim Hill)

12. Anrita1705 (Anrita)

13. Natalia_Kollegova