the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Jul 2026
Loess Soil Structural Changes Induced by Bio- and Synthetic Polymer Stabilizers: Geoelectrical Insights
Abstract. Traditional assessments of soil aggregate stability rely on destructive, ex situ protocols that fail to capture the continuous evolution of soil architecture. This study evaluates the divergent stabilization trajectories induced by chia seed mucilage (CSM) and anionic polyacrylamide (A-PAM) in loess soil over a 20-day incubation. By integrating Spectral Induced Polarization (SIP) with microbial respiration and tracer breakthrough experiments, we track the transition from initial aggregation to long-term structural outcomes. Results demonstrate that CSM-induced stabilization is strictly transient; rapid microbial degradation of the biopolymer results in progressive reversal of the induced structural modifications and a reversion to baseline hydraulic behaviour. In contrast, A-PAM provides persistent physicochemical reinforcement, establishing a dual-porosity regime characterized by immobile water domains and preferential flow pathways. SIP signatures effectively resolved these dynamics: high-frequency shifts in quadrature conductivity (σ'' ) provided geoelectrical evidence of micro-aggregate formation, while the development of a 'conductive reservoir' in in-phase conductivity (σ' ) identified hydraulically isolated ion accumulation. These findings establish SIP as a high-resolution, non-invasive proxy for monitoring the interplay between biochemical persistence and the mechanical stabilization of soil structure.
- Preprint
(1119 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-3473', Anonymous Referee #1, 15 Aug 2026
-
AC1: 'Reply on RC1', Sonya Altzitser, 26 Sep 2026
We sincerely thank the reviewer for the constructive feedback. We have carefully considered all the comments and will revise the manuscript accordingly. Below, we provide detailed responses to each point raised:
- We thank the reviewer for this important comment. We agree that the proposed changes in pore architecture were not directly imaged, and we will revise the manuscript to present them as interpretations supported by the data rather than as demonstrated mechanisms.
We respectfully note, however, that we did not infer these interpretations from SIP alone. The SIP signatures were interpreted together with several independent measurements:- Tracer transport (Fig. 4), measured on the same SIP columns: by day 20, A-PAML showed pronounced tailing and a marked increase in dispersivity (), indicating a change in flow-path structure. ML converged to control-like behaviour.
- Effluent chemistry (Fig. 7), measured on the same SIP columns: in A-PAML, the increase in σ′ was decoupled from the low effluent Ca²⁺. In the control and ML, σ′ tracked the rise in effluent Ca²⁺.
- Aggregate stability (Fig. 3), measured on the same treated soils in parallel incubations: A-PAM produced persistent aggregation, while the aggregation induced by mucilage was transient.
- CO₂ emission (Fig. 2), measured on the same treated soils in parallel incubations: mucilage stimulated microbial activity, consistent with its degradation, while A-PAM did not.
Nevertheless, we agree that the wording in several places overstates the certainty of these mechanisms. In the revised manuscript, we will soften the relevant statements in the Abstract, Discussion, and Conclusions. We will also add a statement acknowledging that the proposed structural mechanisms were not directly validated by pore-scale imaging, and that future studies combining SIP with direct pore-structure characterization are needed to confirm them.
-
We thank the reviewer for this valuable comment. We agree that the failure of the ADE alone cannot demonstrate a dual-porosity system. Following this suggestion, we fitted the breakthrough curves with the mobile–immobile model (MIM; van Genuchten and Wierenga, 1976), in addition to the ADE, and compared the two models quantitatively using the corrected Akaike Information Criterion (AICc) and the root-mean-square error (RMSE). In both models, the water content was fixed at the measured porosity of each column, and iodide was treated as a conservative tracer (R = 1).
The results confirm that the behaviour of A-PAML differs distinctly from that of the other treatments. After 20 days of incubation, all treatments showed some change in transport behaviour. For the control and ML, both models described the data well (R² ≥ 0.98). The MIM provided a modest refinement, with a large mobile water fraction (β = 0.91–0.92), indicating only minor physical non-equilibrium. In contrast, neither model captured the pronounced tailing of A-PAML at day 20. The MIM offered no meaningful improvement over the ADE (ΔAICc = 1.1), and the residual error of the best-fitting model was three to five times higher than in the control and ML treatments (RMSE of 0.066, compared with 0.012 and 0.022, respectively). A-PAML also showed the largest dispersivity (α = 2.66 cm, compared with 1.18 and 1.56 cm for the control and ML).
These results indicate that the transport behaviour of A-PAML reflects a more complex form of non-equilibrium than can be represented by a single-rate exchange between mobile and immobile domains, such as multi-rate mass transfer (Haggerty and Gorelick, 1995). Such models, however, cannot be reliably constrained with the present data. In the revised manuscript, we will present the MIM analysis and model comparison, and describe the A-PAML transport behaviour as non-ideal transport consistent with physical non-equilibrium. The dual-porosity interpretation will be presented as a hypothesis supported by the combined evidence, rather than as demonstrated by the transport data.
-
We thank the reviewer for this comment. We agree that the tracer experiments do not allow evaluation of the effect of polymer concentration on solute transport, and we will state this limitation more clearly in the revised manuscript.
It should be noted that the tracer experiments were not intended to assess concentration effects. Because stable flow could not be maintained in the A-PAMH columns, the tracer experiments were limited to the low-concentration treatments (ML and A-PAML). This allowed a comparison between the natural and synthetic polymers at equal concentration. For the same reason, MH was not included. The effect of polymer concentration was evaluated through other measurements: aggregate stability and CO₂ emission were measured for both concentrations of both polymers (Figs. 2, 3), and SIP was measured for both mucilage concentrations (Figs. 5, 6). The conclusions regarding concentration in the manuscript are based on these measurements, not on the tracer data.
In addition, the inability to maintain stable flow in the A-PAMH columns, owing to excessive backpressure, is itself an indication of a substantial reduction in permeability at the higher A-PAM concentration. This observation is consistent with a strong effect of A-PAM on the pore structure. In the revised manuscript, we will report this observation more explicitly and clarify the rationale for the choice of treatments in the tracer experiments. We will also note that the effect of polymer concentration on solute transport remains to be evaluated in future studies.
-
We agree that CO₂ emission is not a direct measurement of polymer degradation, and that the residual mucilage in the soil was not quantified in this study.
Regarding the contribution of calcite dissolution, all treatments were prepared from the same calcareous loess, and calcite dissolution is expected to occur in all of them, including the control. For this reason, the CO₂ results were interpreted in terms of differences between treatments relative to the control rather than absolute values, as stated in the manuscript (L204-L205). The excess CO₂ emission in the mucilage treatments, reaching a 3-fold increase in MH on days 3 and 7 (Fig. 2), is therefore attributed to the added mucilage. This interpretation is further supported by the effluent chemistry of the SIP columns: the control, ML, and MH treatments reached similar Ca²⁺ and total cation concentrations by day 20 (Fig. 7), consistent with a similar extent of calcite dissolution among these treatments.
Moreover, the attribution of the decreasing mucilage effect to microbial degradation does not rely on CO₂ emission alone. It is supported by several independent observations over the same period: the decline in aggregate stability (Fig. 3), the convergence of the ML breakthrough curve toward control-like behaviour by day 20 (Fig. 4), and the convergence of σ″ between ML and MH (Fig. 5). Together with the well-documented lability of mucilage polysaccharides (Nazari, 2021; Nazari et al., 2022), these observations are consistent with microbial degradation as the main driver of the transient stabilization.
In the revised manuscript, we will present this mechanism as consistent with the combined data rather than as demonstrated, and we will note that direct quantification of mucilage degradation (e.g., polysaccharide analysis or isotope labeling) would be valuable in future studies.
-
We thank the reviewer for raising this important point. We agree that pore-water chemistry strongly affects the SIP response, and that the substantial increase in cation concentrations during incubation must be considered when interpreting the SIP signals.
The experimental design was intended to account for this effect through reference treatments. The control soil underwent the same calcite dissolution and increase in pore-water ionic strength as the polymer-treated soils, but without polymer-induced structural changes. The N-PAM treatments served as an additional reference, with minimal change in effluent chemistry (Fig. 7). Accordingly, the SIP responses of the polymer treatments were interpreted as deviations from these references rather than in absolute terms.
For the quadrature conductivity, the effect of pore-water chemistry can be largely ruled out. Quadrature conductivity is known to increase with pore-water salinity (Revil and Skold, 2011). In the control and mucilage treatments, effluent cation concentrations increased by 2.4–3.9-fold over the incubation (Fig. 7), yet σ″ decreased in all treatments except N-PAM (Fig. 5b). The decline in σ″ is therefore opposite to the trend expected from pore-water chemistry, and must reflect changes at the mineral–water interface. In the N-PAM treatments, both effluent chemistry and σ″ remained nearly constant.
For the in-phase conductivity, we agree that the chemical and structural contributions cannot be fully separated with the available data. Effluent chemistry was measured at only two time points (day 0 and day 20), and, as discussed in the manuscript, the effluent may not fully represent the pore water in hydraulically less accessible domains. In the revised manuscript, we will therefore present the σ′-based interpretations, namely the reduced formation factor in MH and the conductive reservoir in A-PAML, as hypotheses consistent with the data rather than as established mechanisms.
-
We agree that the information on replication and statistical analysis should be presented more clearly, and we will revise the manuscript accordingly.
Regarding replication, all replicates in this study were independent. CO₂ emission was measured in four separate jars per treatment, and wet sieving was performed on independently prepared soil samples. SIP and tracer experiments were conducted in triplicate, independently packed columns. Columns that developed leakage during the experiment were excluded, and results are reported for the remaining replicates.
Regarding the statistical treatment of the temporal data, the CO₂ emission data were analyzed for differences between treatments within each sampling day (Tukey HSD, P < 0.05), as presented in Fig. 2. In addition, we tested the temporal changes within each treatment across the sampling days (Tukey HSD, P < 0.05). These results confirm significant temporal changes in all treatments and will be added to the Supplementary Material.
For the aggregate stability data (Fig. 3), the results are presented as normalized distributions of the stability fractions, which do not allow error bars to be displayed clearly. We will therefore add a table to the Supplementary Material reporting the mean ± SD of each stability fraction for each treatment and sampling day.
For the SIP data, Fig. 5 presents the mean ± SD of the replicate columns, and this will be stated explicitly in the figure caption. In the revised text, the reported changes in σ′ and σ″ will be given with their SD.
-
We thank the reviewer and agree. Our conclusions are limited to the 20-day incubation under laboratory conditions. In the revised manuscript, we will clarify that "persistent" refers to the stability observed over the incubation period, relative to the transient effect of mucilage, and we will present the long-term environmental implications as perspectives for future research.
- We thank the reviewer for this important comment. We agree that the proposed changes in pore architecture were not directly imaged, and we will revise the manuscript to present them as interpretations supported by the data rather than as demonstrated mechanisms.
-
AC1: 'Reply on RC1', Sonya Altzitser, 26 Sep 2026
-
RC2: 'Comment on egusphere-2026-3473', Anonymous Referee #2, 25 Aug 2026
Review of "Loess Soil Structural Changes Induced by Bio- and Synthetic Polymer Stabilizers: Geoelectrical Insights" by Sonya Sara Altzitser, Yael Golda Mishael, Nimrod Schwartz
The present manuscript deals the geoelectrical monitoring of structural change induced by different treatments. It is intended for publication in SOIL. The study is of interest for the soil science community as soil structures influence many important processes. The manuscript is well written and, to the best of my knowledge, is using state-of-the-art experimental set-up and procedure. However, I do have some comments regarding the discussion and petrophysical interpretations that I develop in the detail list below.
Introduction : Given the topic of the paper, I think that the authors need to develop the literature of structural change monitoring using geophysical tools (e.g., Romero-Ruiz et al., 2018, 2022).
Section 2.2: Given the influence of partial saturation on SIP signals, could the author provide more detail their sample saturation degree in addition to the water content. My understanding is that it was fully water saturated but it would clearer by stating it (at the moment it is only implied in line 169).
Section 3.1: the physical limitations to microbial activity (i.e., soil respiration) and their relationship to aggregate (i.e., microstructural properties) through diffusive processes are discussed in the literature (e.g., Nuna, 2017, Fülöp et al., 2026).
Figure 5: I understand the use of fmax to plot the imaginary part of the conductivity, but I’m afraid that it hides potential information from the time series. What about the variation of f_max with time (since f_max could be related to aggregate size as discussed in lines 419-428) ? Given the heterogeneity in initial conductivities and the influence of the real part of the conductivity on the imaginary part, how is the phase evolving? Would it be possible to plot the pore water electrical conductivity as a function of time?
Lines 313-315: would the higher mucilage explain the higher respiration rate in figure 2?
Equation 4: Accounting for aggregate structures in the soil samples would required more refined petrophysical models (e.g., Day-Lewis et al., 2017; Romero-Ruiz et al., 2022). But then since equation 4 is not really used in the paper, I’m wondering if it should be removed and the conductivity variations only discuss in the text citing the literature.
References:
Day‐Lewis, F. D., Linde, N., Haggerty, R., Singha, K., & Briggs, M. A. (2017). Pore network modeling of the electrical signature of solute transport in dual‐domain media. Geophysical Research Letters, 44(10), 4908-4916.
Fülöp, O., Nunan, N., Gueye, M., & Jougnot, D. (2026). Electrical conductivity measurements as proxies for diffusion-limited microbial activity in soils under controlled laboratory conditions. Soil, 12(1), 703-714.
Nunan, N. (2017). The microbial habitat in soil: Scale, heterogeneity and functional consequences. Journal of Plant Nutrition and Soil Science, 180(4), 425-429.
Romero‐Ruiz, A., Linde, N., Keller, T., & Or, D. (2018). A review of geophysical methods for soil structure characterization. Reviews of Geophysics, 56(4), 672-697.
Romero‐Ruiz, A., Linde, N., Baron, L., Breitenstein, D., Keller, T., & Or, D. (2022). Lasting effects of soil compaction on soil water regime confirmed by geoelectrical monitoring. Water Resources Research, 58(2), e2021WR030696.Citation: https://doi.org/10.5194/egusphere-2026-3473-RC2 -
AC2: 'Reply on RC2', Sonya Altzitser, 26 Sep 2026
We sincerely thank the reviewer for the positive assessment of our work and for the constructive and insightful feedback. The reviewer's comments, and particularly the suggested literature, helped us strengthen both the framing of the study and the interpretation of our geoelectrical results. We have carefully considered all the comments and will revise the manuscript accordingly. Below, we provide detailed responses to each point raised:
-
We sincerely thank the reviewer for pointing us to these studies, which are highly relevant to our work and strengthen both the framing and the interpretation of our results. We will expand the Introduction to review the use of geophysical tools for monitoring changes in soil structure. Romero-Ruiz et al. (2018) showed the potential of geophysical methods for non-invasive soil structure characterization, and Romero-Ruiz et al. (2022) demonstrated that geoelectrical monitoring can capture lasting structural changes in soil. We will also highlight that SIP, although recognized as promising for this purpose, has so far received little attention in this context, which our study aims to address. Furthermore, the finding of Romero-Ruiz et al. (2022) that increased aggregate connectivity is reflected in bulk electrical conductivity directly supports our interpretation of the in-phase conductivity results, and we will incorporate it into the Discussion accordingly.
-
The reviewer is correct: all SIP and tracer columns were fully water-saturated throughout the experiments. The columns were saturated from the bottom with 5 mM CaCl₂ to displace air, followed by saturated flow for three pore volumes. We agree that this should be stated explicitly rather than implied, and we will revise the Methods to clearly state that all SIP measurements and tracer experiments were conducted under fully saturated conditions.
-
We thank the reviewer for this valuable comment and for pointing us to these studies, which provide an important mechanistic framework for interpreting our respiration results. In the manuscript, we proposed that the stable aggregates formed under A-PAM treatment alter pore architecture, reduce oxygen diffusion, and restrict microbial access to carbon substrates and electron acceptors, thereby limiting respiration. The studies suggested by the reviewer directly support this interpretation. Nunan (2017) and Fülöp et al. (2026) show that microbial decomposition largely depends on the diffusion of substrates toward decomposers, and that aggregation can physically protect organic matter from microbial access. Moreover, Fülöp et al. (2026) suggested that near saturation, respiration may be limited by oxygen diffusion, which is relevant to our study since respiration was measured at a moisture content close to saturation. We will incorporate these studies into the revised Discussion to strengthen our interpretation of the consistently lower CO₂ emissions under A-PAM treatment.
Furthermore, the framework proposed by Fülöp et al. (2026), linking electrical conductivity to diffusion-limited respiration through the tortuosity of the water phase, offers an interesting perspective on our results. The MH treatment, which showed the highest respiration rates, also showed the σ′ increase that we attributed to reduced tortuosity of the conductive pathways. Although respiration and SIP were measured in separate experimental setups, and a direct quantitative link therefore cannot be established here, we will briefly note this qualitative consistency in the revised Discussion as a direction for future research.
-
We thank the reviewer for these insightful suggestions. We have analysed the temporal evolution of both the relaxation time (τ), derived from the peak frequency of σ″, and the phase, but did not include these results in the original manuscript.
Throughout the entire incubation, the A-PAML treatment exhibited relaxation times two to three times shorter than all other treatments, which remained relatively stable over time. This shows that the shift toward shorter polarization length scales, presented in the manuscript only for day 0, persisted over the full 20 days, further supporting our interpretation of stable micro-aggregate formation under A-PAM treatment.
The phase decreased over time in all treatments. The decline was steepest in MH and A-PAML, intermediate in the control and ML, and smallest in N-PAM, mirroring the magnitude of the σ′ increase in each treatment. Normalizing the polarization response by σ′ therefore does not change the treatment differences, and the phase trends are consistent with our interpretation of the σ′ and σ″ results.
To allow readers to evaluate these trends directly, the full electrical spectra for all treatments and measurement days will be provided in the Supplementary Material of the revised manuscript.
Regarding pore-water electrical conductivity, a continuous time series unfortunately cannot be provided, since the columns were not equipped with internal pore-water sampling, and EC could be measured only in effluent collected under flow conditions at day 0 and day 20.
-
Yes, the higher respiration rates observed in the MH treatment are consistent with its higher mucilage concentration. As mucilage is a readily available carbon source rich in labile polysaccharides, the five-fold higher application rate in MH provided substantially more substrate for microbial activity, resulting in a dose-dependent respiration response.
This observation also connects to the concentration-dependent threshold hypothesis proposed in the manuscript. Previous studies attributed the persistence of structural improvements after mucilage decomposition to microbial byproducts (Morel et al., 1991; Traoré et al., 2000; Marsico et al., 2018). The higher microbial activity in MH may therefore have promoted the production of such byproducts, including extracellular polymeric substances, which could act as secondary binding agents. This mechanism may help explain why MH retained partial aggregate stability throughout the incubation while ML did not, and it is consistent with the hypothesis presented in our conclusions regarding EPS-mediated stabilization. We will clarify this link between respiration, mucilage concentration, and aggregate stability in the revised Discussion.
-
We agree that Eq. 4 assumes a single conducting continuum and cannot quantitatively describe aggregated soils with distinct inter- and intra-aggregate pore domains. However, we would prefer to keep the equation, since our discussion of σ′ relies on separating its electrolytic, geometric, and surface contributions. These are the reduced formation factor proposed for MH, the reduced surface conductivity proposed for N-PAM, and the pore-water conductivity driving σ′ in the control and ML treatments. We believe the equation helps the reader follow these arguments. In the revised manuscript, we will present Eq. 4 explicitly as a conceptual framework rather than a quantitative model, and clearly state its limitations for aggregated systems.
Moreover, the studies suggested by the reviewer provide valuable support for our interpretation of the A-PAML results. Day-Lewis et al. (2017) showed that in dual-domain media, the bulk electrical response is affected by solutes in the less mobile domain, which are not captured by fluid sampling. This is consistent with our proposed 'conductive reservoir', in which ions accumulate within hydraulically isolated intra-aggregate pores, contributing to σ′ but not appearing in the effluent. We will cite Day-Lewis et al. (2017) and Romero-Ruiz et al. (2022) in the revised Discussion to support this interpretation, and note that applying such dual-domain petrophysical models to aggregated soils is a promising direction for future quantitative work.
-
-
AC2: 'Reply on RC2', Sonya Altzitser, 26 Sep 2026
-
RC3: 'Comment on egusphere-2026-3473', Anonymous Referee #3, 30 Aug 2026
Altzitser et al. present a study investigating structural changes in loess following the addition of chia seed mucilage and polyacrylamide through measuring aggregate stability, CO2 respiration, tracer breakthroughs, and spectral induced polarization (SIP). The use of SIP is especially interesting as a non-destructive method for monitoring soil change. The manuscript is generally well-written and the experiment includes an interesting combination of methods. However, I generally share the concerns of the other two reviewers and also think that aspects of the experimental design, statistical analysis, and soil characterization require further clarification. Therefore, I believe major revisions are necessary before publication.
Major comments:
1. Interpretation of CO2 measurements
The authors primarily attribute the loss of mucilage-induced stabilization to microbial degradation of the mucilage. The higher CO2 production in the mucilage treatments could indicate increased microbial activity following labile carbon addition, but I do not think it directly demonstrates degradation of the mucilage itself. As the authors also acknowledge, it could be influenced by calcite dissolution. Additionally, the CO2 measurements were made in separate incubations under slightly different physical/hydrological conditions, which could influence microbial activity and degradation. As a result, I think the authors should be a bit more cautious using the respiration measurements to explain structural changes observed in the columns, unless there is additional evidence of degradation available.
2. Statistical analysis and replication
I could not find a description of the statistical analyses used in the manuscript. For example, in Fig 2, significant differences are shown among the treatments but it is not stated which test was used or how pairwise differences were determined. Replication is also unclear for some of the other measurements. Although replication is reported for the CO₂ and tracer experiments, the number of replicate columns used for the SIP measurements is also not clear. I think the manuscript would benefit from a statistical analysis section clearly describing tests used and the number of independent replicates for each part of the experiment.
3. Soil characterization
The description of the soil used seems quite limited considering how much of the discussion relies on mineralogy and chemistry. The soil is described mainly as a sandy clay loam collected from an uncultivated loess site, but later interpretations depend on calcite dissolution, surface conductivity, changes in pore-water chemistry, and mineral surface properties. I think more thorough characterization of the soil would add needed support to these interpretations. At minimum, it would be helpful to report pH, carbonate content, SOC, and particle-size distribution, and ideally CEC if available. If these properties were not measured, I think the authors should acknowledge this limitation and be more cautious in discussion of underlying mechanisms.
Minor comments:
1. Lines 124 and 206–207: There seems to be an inconsistency in the reported CO2 sampling dates. The methods state that samples were collected on days 1, 3, 4, 10, 14, 17, and 20, whereas the results refer to a significant difference on day 7. Please check and correct the sampling dates throughout the manuscript.
2. Lines 140–142: The sand fraction is estimated from the control based on the assumption that no highly stable aggregates are present and is then subtracted from all treatments. I think this assumption should be better justified, particularly since the highly stable fraction is important for the subsequent interpretation. It would also be useful to clarify why the sand fraction was not determined independently.
3. Lines 148–149 and 190–192: It is stated that A-PAMH was excluded from the tracer and SIP experiments because a stable flow could not be maintained due to excessive backpressure and leakage. I think this observation itself is potentially important and should be discussed when generalizing the A-PAML results to A-PAM more broadly.
Citation: https://doi.org/10.5194/egusphere-2026-3473-RC3 -
AC3: 'Reply on RC3', Sonya Altzitser, 26 Sep 2026
We thank the reviewer for the careful evaluation of our manuscript and the constructive comments. We will revise the manuscript accordingly. Below, we provide detailed responses to each point raised:
Major comments:
-
We thank the reviewer for this comment, and we agree that elevated CO₂ emission indicates stimulated microbial activity but does not directly quantify mucilage degradation, since the residual mucilage in the soil was not measured in this study.
Regarding calcite dissolution, all treatments were prepared from the same calcareous loess, so calcite dissolution is expected to contribute to CO₂ emission in every treatment, including the control. As stated in the manuscript, the CO₂ results were therefore interpreted as differences between treatments relative to the control, rather than as absolute values. On this basis, the excess emission in the mucilage treatments, which reached a 3-fold increase in MH on days 3 and 7 (Fig. 2), is attributed to the added mucilage. This interpretation is consistent with the effluent chemistry of the SIP columns: the control, ML and MH treatments reached similar Ca²⁺ and total cation concentrations by day 20 (Fig. 7), suggesting a similar extent of calcite dissolution among these treatments.
Regarding the separate incubations, the respiration jars and the SIP columns were prepared from the same treated soils, following the same protocol, and both were incubated under static conditions. The jars were kept at a moisture content of 30% (w/w), close to saturation given the porosity of the packed soil (0.41–0.45). We agree, however, that the two systems are not identical. The columns were fully water-saturated and were flushed at packing. The jars retained some air-filled porosity and a headspace, and may therefore have had higher oxygen availability. For this reason, we regard the respiration data as an indicator of the relative biodegradability of the two polymers, rather than a measure of the degradation rate within the columns.
Importantly, the attribution of the transient mucilage effect to microbial degradation does not rest on the respiration data alone. Over the same period, aggregate stability in the mucilage treatments declined (Fig. 3), the ML breakthrough curve converged toward control-like behaviour by day 20 (Fig. 4), and σ″ converged between ML and MH (Fig. 5). Together with the well-documented lability of mucilage polysaccharides (Nazari, 2021; Nazari et al., 2022), these independent observations are consistent with microbial degradation as the main driver of the transient stabilization.
-
We thank the reviewer for this comment. The same point was raised by Reviewer 1, and we refer the reviewer to our detailed response there (RC1, comment 6). In brief, we will add a statistical analysis section to the Methods specifying the tests used and the number of independent replicates for each experiment, including the SIP columns, which were run in triplicate.
-
We thank the reviewer for this comment. We agree that the soil description should be expanded, since several of our interpretations rely on the soil's mineral composition. The loess soil used in this study consists of 60.5% sand, 17.5% silt, and 21.9% clay, with 2.9% organic matter and 19.5% CaCO₃. These properties will be added to the Materials section of the revised manuscript.
These data provide direct support for several of the mechanisms discussed in the manuscript. The high carbonate content (19.5% CaCO₃) establishes calcite as a major and readily available mineral phase. This is consistent with our attribution of the increase in effluent Ca²⁺ and σ′ in the control and mucilage treatments to calcite dissolution, and of the decline in σ″ to the loss of reactive mineral surfaces. The clay content (21.9%) supports the relevance of surface conductivity and EDL polarization to the interpretation of the SIP signals.
Minor comments:
-
Thank you for noticing. The sampling day listed as "day 4" in the Methods is a typographical error and should read day 7. The sampling dates will be corrected throughout the revised manuscript to match those presented in Fig. 2 (days 1, 3, 7, 14, 17 and 20).
-
We thank the reviewer for this comment. The control soil received no stabilizer, and its aggregate stability remained low throughout the experiment (Fig. 3). After full chemical dispersion with sodium hexametaphosphate, the material retained on the sieve in the control was therefore assumed to consist of primary sand particles.
The sand fraction was determined from the control rather than taken from the particle-size analysis, because the two do not measure the same quantity. Particle-size analysis reports all sand particles (>50 µm), whereas only particles coarser than the 60-mesh sieve (250 µm) are retained in the wet-sieving procedure. In the control, only a small fraction of the sample mass was retained on the sieve after full dispersion, although sand makes up 60.5% of the soil. This indicates that most of the sand in this loess is finer than 250 µm and passes the sieve, consistent with the fine-grained nature of loess. Subtracting the sand content obtained from particle-size analysis would therefore greatly overestimate the retained sand mass. The control, processed under identical conditions with the same sieve, sample mass, and dispersion protocol, provides a direct measure of the sand mass retained by this procedure.
Furthermore, because the retained mass in the control was small, the correction has only a minor effect on the calculated stability fractions, and the approach is conservative. If part of the material retained in the control consisted of naturally stable aggregates rather than sand, it would also be subtracted from the amended treatments. The highly stable fraction in these treatments would then be slightly underestimated rather than overestimated.
-
We thank the reviewer for this comment. We agree that this observation is important in its own right. The inability to maintain stable flow in the A-PAMH columns, owing to excessive backpressure, indicates a substantial reduction in permeability at the higher A-PAM concentration. This is consistent with a strong effect of A-PAM on pore structure, although it may also reflect other processes such as polymer swelling or pore clogging. It also suggests that the transport behaviour observed in A-PAML cannot be directly extrapolated to higher application rates, and it is in line with previous reports that excessive A-PAM doses can be less beneficial than optimal ones (Soltani et al., 2022).
In the revised manuscript, we will report this observation explicitly in the Results and discuss its implications. We will also specify throughout the Discussion and Conclusions that the tracer and SIP findings refer to A-PAML, rather than generalizing them to A-PAM as a whole.
-
-
AC3: 'Reply on RC3', Sonya Altzitser, 26 Sep 2026
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 195 | 59 | 29 | 283 | 22 | 22 |
- HTML: 195
- PDF: 59
- XML: 29
- Total: 283
- BibTeX: 22
- EndNote: 22
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript addresses an interesting topic and combines aggregate stability, microbial respiration, tracer transport, soil chemistry, and SIP measurements. In my opinion, the manuscript has potentially interesting observations, but the central mechanistic conclusions are insufficiently supported by the experimental evidence. The interpretation of SIP measurements, dual-porosity development, and the proposed conductive reservoir relies substantially on indirect inference. In addition, the limited tracer treatments, short experimental duration, and unresolved influence of soil solution chemistry substantially weaken the study. These are fundamental issues that cannot be adequately addressed through minor or moderate revision. I therefore recommend rejection in its present form.