The "Cartion Park" Model for ERW on Croplands
Why do different ERW measurement approaches seem (?) to measure different amounts of CDR?
Version 1.0 - by Dirk Paessler - 7. Dec 2022
Introduction
As we all know humans need to drive down their emissions by 90% to slow climate change. But we are not done then, we also need to remove CO₂ from the atmosphere with CDR (carbon dioxide removal) in the order of 10 gigatons CO₂e per year by 2050, says the IPCC (SR 15, Chapter 2). The IPCC also estimates that enhanced rock weathering on croplands will be able to do 1 Gt of this, but then they continue to say: “While the geochemical potential to remove and store CO₂ is quite large, limited evidence on the preceding topics makes it difficult to assess the true capacity, net benefits and desirability of EW and ocean alkalinity addition in the context of CDR.” (SR 15, Chapter 4)
We still need to understand this better! With Project Carbdown we have been working on this since 2020. Our goal is to find and improve ways to properly measure the effects of enhanced weathering on croplands. Which - turns out - is really hard to do.
There is a discrepancy between measurements of the basalt dissolution on the top of the soil (these can be in the order of 10-60% dissolution per year) and the amount of cations or alkalinity measured in the leachate waters (1-5% per year in the beginning, more after 2-5 years). This is a problem for MRV (monitoring, reporting and verification) of CDR.
Since we believe that both measurements are actually correct we need to understand what happens in between the two. Where are the remains of the dissolved rock?
Acknowledgements
We would like to thank Dr. Dimitar Z. Epihov, University of Sheffield, for our in-depth conversation about ERW and his explanations of how weathering cations are being “parked” in the soil’s cation exchange capacity. This thought concept was the base for following article. Thanks also to the following scientists for reading and commenting earlier versions of this text and helping to understand all this in the first place: Jens Hartmann, Philip Pogge von Strandmann, Elena Vorrat, Thorben Amann, Ingrid Smet, Noah Planavski, Johann Rietzler, Arthur Hofmann, Ralf Steffens.
Introducing a visual “thinking model” for rock weathering in soils
During our discussions with the scientists of Project Carbdown the following visual model has developed which helps us to talk about the complex interactions in the soil. Let me introduce our model by telling a simple story describing what happens in the soil when we add basalt dust on top in order to do Carbon Dioxide Removal (CDR).
The basic idea is to imagine the cation as cars (we will call them “cartions”) and the layers of the soil as a huge, multi story car park for the cartions. By adding and dissolving basalt on top of the building we add new cartions (new cars) on top that want a nice parking space to rest for some time. Some time later the cartions move on and eventually leave the car park at the bottom (into the groundwater or the lysimeter tank).
Apart from the cartions there is also a second group of major players in our story: At the top of the car park there are also “passengers” (the CO₂ as Bicarbonate HCO₃⁻ anions) and their only hope is to be taken down by a car (=being balanced by a cartion) to the exit. Depending on its charge (e.g. + or 2+) a cartion can take one or more passengers as HCO₃⁻ for a bit of the way -- step by step -- downward until the last friendly cartion finally takes the passenger with himself out of the system (=CDR happened!). If the passengers can’t find a ride, they will stay, can go back to the soil’s C-cycle, and eventually might even get respirated to the air (=no CDR!)).
The key point for MRV is: Different measurement approaches look at different locations or aspects in the carpark (e.g. number of entries, speeds, traffic jams, traffic on ramps, car types/manufacturers, number of exits) at different moments in time. This inevitably leads to different measurements of the same thing. We need to understand THIS!
Of course the actual processes are much more complex than shown in our story! Just the basics of this are already over 100 pages in the fundamental book “Soil Science” (Scheffer-Schachtschabel). But we need a less complex model to explain this to ourselves and to others.
So let’s get into the story:
Let’s assume we have an agricultural field that received a treatment with a few tons of basalt dust per hectare. What will happen to the basalt…?
Drops of rainwater “collect” ambient CO₂ (at 420 ppm) while falling down from the clouds, creating carbonic acid on the way.
In the soil the water collects even more CO₂ from the soil (the natural CO₂ concentration in soil is thousands of ppm).
The acidic water reaches the basalt grains and reacts with the rock’s minerals, thereby dissolving the rock. This adds cartions like Mg²⁺, Ca⁺ and others into the water between the soil grains. New cars enter the car park!
The new cartions together with a related amount of bicarbonate HCO₃⁻ (=associated CO₂) are now in the water. These cartions are now driving around in the car park, looking for a parking space, they want to dock with a soil grain.
The new cartions push out older cartions from the parking spaces because they are now higher concentrated or because they are stronger (have higher binding power) or have more mobility. The share of the parking spots that can be used for this is called the “exchangeable fraction”. This is the mobilizable part of parking spots of the soil’s “Cation Exchange Capacity” (CEC), which is an official lab parameter for soils. The CEC metric of a specific soil gives us the size/height of the car park (how many parking spots we have). The higher the CEC of a soil the larger is the car park and the cartions need much longer to reach the exit.
Cartions stay in the parking spots
until more “sticky” cartions replace them (e.g. Mg²⁺ or Ca⁺ from the basalt kick out older/weaker occupants) or
until the concentrations of new cartions go up in the water and seek equilibrium or
until the pH in the soil/water gets lower. This can happen when rain (pH 4.x) creates "flush events" where suddenly more H⁺ is in water and H⁺ replaces (pushes out) cartions on soil’s parking spaces (H⁺ is stronger than everybody else). => A strong rain event moves a lot of stuff in the car park around and downward!
The more H⁺ is in the water, the more H⁺ sticks to the parking places (seeks equilibrium) => In soils with low pH the number of usable parking spots is much lower, the CEC is heavily degraded (which also degrades plant life).
Every time the cartions are parked out they get flushed just a bit further down in the soil and will be re-parked many many times (thousands of re-parkings!)=> this can take a lot of time (it can take weeks and months to move a cartion even over a small vertical distance of only 30-50 cm)! To give an impression of the orders of magnitude here: We are talking about movement on molecular level combined with a huge surface area (1 gram of soil has a surface in the order 0,5-50 square meters) which needs to be traversed by the cartions.
Only as long as a cartion is in the water it keeps a negatively loaded Bicarbonate HCO₃⁻ (water can’t be electrically loaded) and the evil “C” can’t go back to the natural C cycle, which means either the plants or respiration (=> CDR has happened!!!). A cartion parked in the car park (bound to soil particles) does not balance a bicarbonate and thus hasn’t done its CDR job yet (!).
The mineralization detour: About 10% of the Bicarbonate in the water precipitates as carbonate minerals and re-releases 50% of the CO₂ initially captured. With enough water it may be re-dissolved, captures the same amount of CO₂ again and everything moves on.
The slow clay dead end: Our car park has a built-in “scrapyard” for superfluous cartions. Some of the Mg²⁺ and other cartions form new clays ("neoformation"), a process that also releases some of captured CO₂ again and reduces the CO₂ drawdown effect. We definitely will create clays when using basalt, the question is how much. Clay formation keeps the water undersaturated, so that the basalt can keep dissolving (good for CDR). The reaction speed is different: the CEC is very fast, while clay formation is much slower. For a single season of basalt addition, clays will matter less than the CEC. Over several years, though, clays will probably be more important than the CEC in affecting cartion behavior and alkalinity. The balance between clay formation and basalt dissolution is one of the unanswered questions of measurement/modeling.
If we add/dissolve enough rock (and not all goes into clay) finally all parking places in the car park eventually become full and a steady flow of cartions from the basalt treatment should come out at the bottom (an estimated 10% already show up at the car park exit earlier).
When finally the cartion leaves the carpark (reaches our lysimeter at the bottom or the groundwaters) it takes one Bicarbonate HCO₃⁻ with it => we measure Total Alkalinity of the leachate water and can finally prove CDR here! Bingo!
Ultimately these Bicarbonates HCO₃⁻ flow with groundwater into rivers (there might be some back&forth with secondary mineral formation etc. in the process). They end up in the ocean where they boldly fight acidification of the sea water, become beautiful sea shells and proudly end up like the white limestone rock of Dover in a few million years!
So to summarize: Most of the parking places are full from the start (there's always enough cartions knocking around to fill most sites). The new cartions (from basalt dissolution) kick off old cartions from the CEC, and replace them. The cartions kicked out of the parking places (CEC) move a bit further down in the soil, and kick other cartions out of other parking places, and replace them. So it's really like a whole row of car parks, and each car has to drive from one car park to the next, then park there, then move on to the next car park, etc. Eventually, a car will make it through all the car parks, and make it out the end… The speed at which the new cartions make it through the multiple exchanges will mostly depend on basalt amount, CEC capacity of the soil, pH and water flow rate.
The open questions
Now with this “picture” in mind we can now describe the areas of scientific work that we need to work on and need to understand better:
#1: Clay formation
How many clay minerals are created, and how many cations (especially Mg) do we lose there, which can’t add to the CDR effect? => Unknown (Method to measure: Li isotopes)
Carpark analogy: How many cartions end up in our scrapyard and are lost for CDR?
#2: AEC (Anion Exchange Capacity)
What about the parking spots of the Anion Exchange Capacity? Are these relevant regarding weathering and/or for the CDR process? => Unknown
Carpark analogy: Do we need to think about a second car park building for anions in our story? => As far as we know it seems to be neither a limitation to CDR nor a way of measuring it
#3: Kinetics
How fast do cations from the basalt dissolution and their respective anion “partners” move downwards or horizontally, depending on water flow directions? => unknown
Carpark analogy: How fast does the constant re-parking go and how fast can a cartion move from one level to the next?
#4: CDR happens… when?
When are our beloved cartions finally set free and can then be balanced by Bicarbonate HCO₃⁻ (=CDR happens)? How much alkalinity is already present in pore water? How much CO₂ is fixed and when? => unknown
Carpark analogy: When are passengers so tightly tied to their cartions that they won’t break up again later?
#5: What role does DOC (dissolved organic carbon) play?
Do we need to take this into account for our long term perspective of CDR or is it a transient thing of the C cycle, mostly unrelated to CDR?
Carpark analogy: This could be considered a conversion of cartions into something else in a process that can also be reversed.
#6 (after #1-#5 have been answered):
How do we properly explain that 50% basalt dissolution per year fit together with 1-5% cation flow at 20-30 cm depth? Is this just a matter of (a long) time until both converge?
Carpark analogy: Why do we measure much more cartions at the entrance of the car park at the top than at the exit at the bottom, or didn’t we just not wait long enough at the exit?
#7 (finally it is about MRV):
How fast are #1-#4? What kind of theoretical CO₂ sink potential does this mean in x years when ERW is scaled up? How much of that can be achieved realistically? How much CDR can actually be put into trustworthy certificates for field x with y tons of basalt?
These are the questions that are guiding us during the planning and setup of our greenhouse experiment.
Science is when you know what you don’t know
OK, this might look like we don’t know anything about enhanced weathering and there is no way we can make this work to save the climate. No, that’s not the case! Enhance Weathering itself has been studied by scientists for decades and it is actually vey well understood, but mostly on a geological timescale. Throwing rocks onto fields will almost always draw down CO₂ over time (unless you chose the wrong field/soil)! Given the super-slow reaction time of the climate system it is not even a problem if ERW on a field might take 50 years, it’s still fast enough to slow climate change.
What is new is that we now look at a (slow!) geological process on a timescale of months and on scales of centimeters and meters. We are realizing that what we understand on a large scale is much more complicated on the small scale. To fully understand enhanced weathering we need this new understanding, especially so we can learn how to optimize the processes to maximize the effects for the climate.
But this shouldn’t stop us from scaling up the weathering infrastructure. As I said, we know it works.
PS: A few additional notes about CEC
During the writing of this we found some more aspects of the CEC that are important to mention, but didn’t fit well into the storyline, but which could be interesting to the readers anyways.
The effective CEC is highly dependent on pH (high pH = higher CEC = more parking spots) of the soil.
The number of new cartions from the basalt is very small compared to the number of cartions that are already parked. The parking spots aren’t created equal, they attract different cartions. The parking/resting times of individual cartions can go from minutes/hours to aeons.
Some parking spaces are VIP spots for long-term-parking which can’t be used for our process, these are not part of the exchangeable fraction: When pH falls under ca. pH=5 cartions like Al³⁺ and H⁺ can block most of the CEC, lowering the exchangeable fraction (parking spots are permanently blocked), which also inhibits plant growth. We don’t want both.
Soils with lots of plant stuff (organics) have a HUGE, hard to assess CEC.
For anions there is an analogous thing, the AEC, which is often overlooked and underestimated (little scientific work in this field, grade of importance for CDR seems to be unclear).
The ability to kick out a parked, old cartion is largely due to the concentration of the “new” cartion - so a high concentration provides an activity gradient that pushes the new cartion (or anion for the AEC) in, and replaces an old one. The ultimate demonstration of this is how you measure the CEC: you add a solution that is very rich in Na+ (or NH4+), so something like Na acetate, and there is so much Na+, that it kicks off everything else, which can then be measured. In natural solutions, the concentration is never that high, but the principle is the same. If the new cartion has a different charge (e.g. Ca2+ compared to Na+), then this process can leave the CEC with a negative charge - this will also then attract other cartions to make up the charge balance.
Some cartions “like” adsorbing more than others. The term for that is “mobility”. A mobile cartion (Na is the most mobile - which is why seawater is Na-rich) likes staying in solution, while an immobile one (very immobile cartions are Al or Fe) likes going into or onto solids. In terms of CO₂ drawdown, Ca and Mg are still fairly mobile - about 3-5 times less than Na. Which means some of them are adsorbed, but a lot less than Al or Fe.
The interesting thing that we don’t know (yet) is to what extent equilibrium makes a difference here: in natural soils, we study the CEC that is in equilibrium with its surroundings - the rainwater, soil composition, etc probably hasn’t changed that much for 100s or 1000s of years. In enhanced weathering, we’re changing that, so the CEC is reacting more strongly than normal.
New cartions exchange with old cartions on the CEC from top to bottom. So there will always be retardation of new cartions by the CEC, whether all CEC sites are full or not. The first cartions that come through to the lysimeter/groundwater will always be old cartions from the CEC, that have been replaced by new cartions from the basalt dissolution. This is why e.g. isotope measurements in leachate waters “see” very few new cartoons for months after basalt application.