| 24 Nov 2025
Soil processes govern alkalinity and cation retention in enhanced weathering for carbon dioxide removal
Abstract. Avoiding the most damaging consequences of climate change will almost certainly require pairing rapid emission cuts with large‑scale carbon dioxide removal (CDR). Among the proposed CDR pathways, enhanced weathering (EW) accelerates natural mineral dissolution to convert atmospheric CO₂ into long‑lived bicarbonate and carbonate reservoirs. Despite the many reported data from EW experiments, large uncertainty remains about the realisable CDR potential of applying rock materials to agricultural land. One of the relevant sinks for CO₂ is the transfer to bicarbonate alkalinity, and various EW studies have reported a wide range of results for this process. Intercomparison of these data is problematic due to the different experimental set-ups, environmental conditions as well as combinations of rock materials and soil types. In order to assess and compare the realisable CDR potential of various EW combinations, a large greenhouse experiment was set up in which 4 different soil types (7 different soil batches) were treated with 13 different feedstock materials. The experiment included growing perennial ryegrass (Lolium perenne) and was conducted over two years with high irrigation rates (> 2000 mm a-1) and elevated temperatures (>19 °C) to speed up the weathering process. Alkalinity production was highly variable among the treatments and some even showed a loss of alkalinity compared to their controls. Consistent with expected dissolution kinetics, alkalinity production rates followed the trend: steel slag > limestone / carbonate-rich metabasalt > peridotite > basanite. Matrix analyses of soil properties versus feedstock revealed that alkalinity production from acidic soils was highest. At higher pH-levels (> 7 pH), carbonate mineral saturation likely constrains further dissolution, potentially favouring carbonate formation. Detailed analyses of cation pools (exchangeable, carbonates, oxides and clay) revealed large changes within the first year where 10–50 times more cations were retained than exported via leachate, making the realised CDR potential as alkalinity relatively small compared to the CDR potential of cations retained. Understanding the dynamics of transfers between cation pools and their potential saturation are important to develop models and enable projections. Data reported from EW studies so far are insufficient to enable calibration of models, specifically if projections in CDR-realisations should span decades.
Competing interests: Mathilde Hagens is a member of the editorial board of Biogeosciences. Tom Reershemius and Bruno Casimiro have received research funding from UNDO Carbon, a for-profit EW company. Dirk Paessler is the CEO and sole shareholder of the Carbon Drawdown Initative. This paper is not providing a direct financial benefit for the Carbon Drawdown Initiative.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Review report:
Hammes et al., reported results of a large greenhouse enhanced weathering (EW) mesocosm experiments to contribute to the understanding of the potential of carbon dioxide removal of EW. The experiments included a set of different combinations of EW feedstocks (13) and soil types (7). Over a period of 2 years, sample collection and analysis were conducted to study the export of Alkalinity in leachate, and the cations retained in different fractions of the soil. The authors observed that there was no measurable change between control and treated pot, or even a reduction of alkalinity export. This led them further investigate the potential traps for released cations within the soil system. The authors discussed the key controls of the exported Alkalinity and retained cations in the experiments, including mineralogy of the feedstock, soil characteristics, and hydrological properties. They concluded that in-soil retention dominated over the Alkalinity export in leachate, and that the partitioning of the cations in soil pools is a critical component for calculating CDR of EW. The experiments were well designed, and the number of experiments and replicates is the highlight of this study. This study has great potential to generate a data set to look into some of the areas that urgently require improved knowledge in in EW. However, due to a few concerns which I will provide more details below, I could not recommend publication of this manuscript in its current state.
In general:
1. Data quality and statistics
This study generated a large dataset with the amount of experiments and replicates conducted. However, there lacks information to allow readers to assess the quality of the data. Firstly, in the sections describing the analytical methods, there lacks information about the evaluation of the quality of the instrumental analysis. Secondly, the uncertainties of the data were not reported in texts or all figures. Where there shows variability in data, there is no information about how such variability was assessed. Thirdly, the handling of such large datasets would require an approach to process the data statistically. Given the number of replicate and experiments, do the measurements show significant variation among replicates of the same experiment? I suggest that a section on the statistics of the data be reported in this study, with information about uncertainty.
2. The study used rainwater as irrigation water, and supplied the pots through a rainwater cistern (Line 150). Although this ensured that the contribution of cations to the leachate by the rainwater is the same between experiment, the influence of the temporal variation in this contribution on the experiments could not be ruled out as rainwater chemistry is not constant with time. This requires justification with analysis of rainwater chemistry, and the contribution of cations from the rainwater needs to be considered when discussing cations in the leachate.
3. The influence of the biomass (the planted ryegrass) and biological activity (the added worms) needs more discussion than the very short mentioning in Line 485-486. Due to various types of treatment, the biomass and biological activity would vary, and that the feedstock and soil types are not the only variables in the experiments. Biomass the biological activity might affect the uptake of the cations by plants, the condition of the soil (e.g. organic matter degradation changing soil pH), and the physical properties of the soils as worms move. These will all have an impact on both the Alkalinity export in leachate and store in soil.
4. It was useful to know that the reported study was a part of a larger experiment. It is however not necessary, especially when the information about the larger experiment gets mixed up with that of the reported study. I suggest that the authors focus on providing information that is relevant to this study, and could include those from the wider experiment in the supplementary information.
Section 2.3: The description of the feedstock here is not clear. It seems that the authors were giving information about the wider experiments they were doing, which this study is part of, as well as the information for this specific study. The writing of it makes it very difficult to follow which was reported in this study, and generated some inconsistencies. For instance, Line 193-194: the authors described different application rates for basanite. However, only 2 rates could be found in Table 1. And in Line 195: the authors stated that the focus was 6 representative feedstocks. However, the experiments using the 7 feedstocks from EW suppliers were included in Table 1, reported and discussed in the manuscript.
5. Discussion
The current discussion is insightful. But it could be discussed more with the data generated in this study.
Line 476: The authors reported the measurement of the cations in leachate. They could take the next step of looking into these data to test this mechanism.
Section 4.3.2. The authors should provide some evidence to argue against or for this mechanism. Some analysis of the soil porosity would be useful.
More comments (minor-major points) are given by section of the manuscript below:
Section 1.
Line 103: What did the authors mean by “comprehensive”? comprehensive geochemical analysis? Comprehensive experiment designs? The manuscript does not reflect either. The authors should be clear here.
Line 109: The years should be spelled out here.
Line 123: The stated hypothesis is not what the work as described by the authors earlier in the same paragraph was testing, and focussed on discussing. A more clearly described hypothesis is needed here. It also lacks clear description of the objectives of the study. The writing of this paragraph could be improved by stating the hypothesis first, then describing the work done to test it.
Section 2.
Line 145: I could not see the cause-effect relationship here: all experiments experienced identical climatic conditions does not really allow for accelerated weathering conditions. It is not clear whether the authors meant to describe the comparability of the experiments without having climatic condition as a variable, or they meant to describe the conditions they set to allow for accelerate weathering. If the former, the sentence needs to be rewritten to reflect that. If latter, the conditions should be given.
Line 147: The irrigation is up to 4000 mm a-1, instead of 2000 in the manuscript, based on Table S5.
Table S1. The authors gave a very brief description of the analytical methods for each parameter. However, there is not any information given to assess the quality of the data. Such information includes, uncertainty, quality control, etc. If such information was published in separate literature, please provide reference.
Line 182: CEC should be spelled out in full, cation exchange capacity, first before using abbreviation.
Line 183: Inconsistent significant number for CEC.
Line 187: PSD was now clear. The abbreviation should be given following its full name first before using elsewhere.
Line 190: ‘representative materials’ is not very clear language. Please give information about what they are representative of.
Line 195: “ a short description of these materials is given below…” I could not see the information the authors said that was given below.
Line 199: More information should be given about the sampling containers , such as materials, whether it is cleaned with water or acid etc.
Section 2.4.3.: Information should be given about the quality control of the analyses: repeated measurements of reference materials, for instance, and reporting this result is needed.
Section: 2.5.1: Information should be given about the quality control of the leaching procedures, ICP-OES and ICP-MS measurements.
Figure 2: Application rate for EW 1-7 in legend is missing. Uncertainties should be plotted.
Figure 3-7: What are the numbers right after the symbol (beginning of the texts) in the legends? What are the dots (all the datapoints?) and lines (average of the replicates?)? How is the variability (shaded area) assessed? standard deviation? or? Such information should be described and given.
Line 299: 4 times should be spelled out instead of using 4x.
Line 333: The authors reported CO2 sequestration rate, but did not give any information about how it was calculated.
Figure 8: Application rate should be given especially the limestone treatment was applied at a different rate from others.
3.2 and figure 8: I do not think that the liquid and solid phase could be compared this way, i.e. by normalizing both to per pot. The amount of the Alkalinity in the leachate was measured in the unit of per volume of the leached liquid, while the cations in the solid phase were measured in the unit of per mass of the solid. Also, Alkalinity in water includes the cations and anions, and therefore is not really directly comparable with the sum of cations in solid phase. Plotting them on the same figure and make comparison are not really comparing the same thing.
Section: 4.3.1
Line 466: it is discussed in more details as the 2nd point, why is it also mentioned here?
Line 585: I struggle to see why point 3 was put down here as one of the indices to support the argument originate form background soil carbonate.