The effect of groundwater depth on topsoil organic matter mineralization during a simulated dry summer in North-West Europe

Françoys, Astrid; Mendoza, Orly; Hu, Junwei; Boeckx, Pascal; Cornelis, Wim; De Neve, Stefaan; Sleutel, Steven

doi:https://doi.org/10.5194/egusphere-2024-559

Preprints

https://doi.org/10.5194/egusphere-2024-559

Preprints

04 Mar 2024

| 04 Mar 2024

The effect of groundwater depth on topsoil organic matter mineralization during a simulated dry summer in North-West Europe

Astrid Françoys, Orly Mendoza, Junwei Hu, Pascal Boeckx, Wim Cornelis, Stefaan De Neve, and Steven Sleutel

Abstract. With climate change expected to intensify the occurrence and severity of droughts, the control of groundwater table (GWT) depth and capillary rise on topsoil moisture may render a critical driver of biological activity. Consequently, GWT depth could influence topsoil carbon mineralization. In this study, undisturbed 200 cm long soil columns of three different textures (loamy sand, sandy loam and silt loam) were subjected to two artificial GWT depths (–165 cm and –115 cm) in the laboratory. We examined (1) moisture supply by capillary rise along the soil profile and specifically into the top 20 cm soil, and (2) consequently the effect of GWT on decomposition of an added ¹³C-enriched substrate (ryegrass) over a period of ten weeks, with limited water applications representing a dry summer. A 50 cm difference in GWT depth (–165 cm vs. –115 cm) resulted in different topsoil moisture for the sandy loam (31 % vs. 38 % Water-filled pore space (WFPS)) and silt loam (33 % vs. 43 % WFPS) soils. In the loamy sand soil, GWT-induced moisture differences appeared only up to 85 cm above the GWT. The expected acceleration of mineralization of the added ryegrass under a shallower GWT was not confirmed. In contrast, C mineralization pulses after the wetting events were even higher for the drier –165 cm GWT soils. For the silt loam soil, where capillary rise supply had the largest contribution to topsoil moisture, a lower mineralization rate of the stable C_ryegrass pool was also found with shallower GWT. These findings suggest that a potential capillary rise effect of increased topsoil moisture on ryegrass mineralization might have been counteracted by other processes. We postulate that the Birch effect might have been magnified following the rewetting of drier topsoils under deeper GWT levels, ultimately enhancing mineralization compared to where the soil remains consistently wetter under shallower GWT levels. Based on our findings, including the process of texture-specific capillary supply from the GWT can be required to adequately simulate moisture in the topsoil during droughts as they occurred over the past summers in North-West Europe, depending on GWT and texture combination. However, the net effect on topsoil C mineralization is complex and correct simulation of C mineralization may require further integration of specific processes connected to fluctuating soil moisture state, such as the Birch effect.

Received: 26 Feb 2024 – Discussion started: 04 Mar 2024

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 preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Astrid Françoys, Orly Mendoza, Junwei Hu, Pascal Boeckx, Wim Cornelis, Stefaan De Neve, and Steven Sleutel

Status: closed

RC1:
'Comment on egusphere-2024-559', Anonymous Referee #1, 02 Apr 2024

Overall, the paper is well written and complete. The introduction provides an adequate description of the context and introduces the aim of the research with useful references. The experimental design is well described. I would suggest including more information about the soils used in the experiment (e.g. soil classification). The statistical approach is well described, you report the methods and software used. The results are clearly presented and the conclusion seems to be strongly supported by the results. References are complete and up-to-date.

Citation: https://doi.org/10.5194/egusphere-2024-559-RC1
- AC1: 'Reply on RC1', Astrid Françoys, 12 Jun 2024
  
  We would like to thank you for the time and effort you invested in reviewing our draft paper. We greatly appreciate your positive evaluation. According to the WRB soil classification map of Flanders (scale 1:40,000, converted from the Belgian Soil Map: DOI: 10.13140/2.1.4381.4089), the soil profiles originating from Kruisem and Bottelare are classified as Eutric Retisols, while the soil profiles from Oosterzele are Eutric Cambisols (Siltic). Obviously, soil profile descriptions would provide a more precise assessment. We do believe that already the major affecting physicochemical specifications for moisture dynamics are supplied, e.g. detailed texture values, OC, pH, bulk densities (Table 1) and pF curves (Fig. B1).
  
  Citation: https://doi.org/10.5194/egusphere-2024-559-AC1
RC2:
'Comment on egusphere-2024-559', Anonymous Referee #2, 29 Apr 2024

27 replace 'can be' with 'is'

37 remove 'content'

50 replace 'moisture' with 'water'

66 replace 'moisture' with 'water'

114 with respect to the difficult-to-interpret results of this study, I wonder whether the lab room temperature may have had an effect. As stated, the two GWT tables were imposed for two consecutive time periods. The C turnover is highly sensitive to temperature. Let's assume a more or less standard Q10 of 1.8 for an Arrhenius-type equation. That means for a 10 degrees increase in soil temeprature you would have 1.8 times the amount of C turnover. The lab temperature has a standard deviation of plusminus 0.5 degrees. For a 1 degree increase you end up with 18 % more respiration. Now, if the average temperature in the lab for the two consecutive incubation periods differs...

144 replace 'moisture' with 'water'

145 replace 'moisture' with 'water'. Well, there is a little issure with terminology: there is either 'water content' or 'moisture'. Please correct for the entire manuscript...

207 well, the more correct term would probably be 'potential' instead of 'height'

209 I suggest to replace 'moisture transport' with 'water flux'.

209 and 210 replace 'transport' with flux

243 I would prefer to replace 'moisture transport' eith 'variably-saturated water flux'

245 that first sentence can be deleted

246 Are measured (saturated) hydraulic conductivities available? In combination with the measured soil water retention functions this would allow to estimate the infiltration depth, or even a soil water content profile over depth. To get an idea of how much the upper 20 cm with the labeled C were actually affected by the irrigation for the different soils...

Figure 2 The y-Axis of graphs should basically always start from zero.

295 A difference in heads is usually referred to as a 'hydraulic gradient'.

315 well, now you end up with two factors affecting (or not affecting) the water content in the upper 20 cm with labelled C. The ground water table AND the irrigation. The irrigation will affect the respiration of the different soils differently. This might be difficult to interprete in the end.

Fig. 3 Maybe a log10 scale applied to the y-axis helps this figure

Fig. 4 I am quite sure a log scale applied to the y-axis benefits this figure

379/380 delete or move to results

387 I assume '-30' and '-10 cm' are the coordinates in the column (i.e. the depth) and not the difference in pressure head?!

406 No, that is not unlikely. This could be a result of macropores.

409 well, there are very small lab tensiometers on the market.

421 Did you observe macropores during sampling?

435 these 'bell-shaped' curves were usually determined for disturbed soil samples. And in the wet range the decrease of respiration is rather related to an oxygen deficit and is actually not related to water content. For undisturbed samples the moisture sensitivity function of respiration actually looks different.

438 better state that the microbes show a more direct reaction to the water-filled pore-space than to the pressure head

448 I suggest to delete 'unwanted asynchrony'

453 which 'artefact'?

457 please replace 'refute' with 'reject'

458 replace 'it needs .. in mind' with 'against the background'

500 but not only the GWTs were modified, simultaneously precipitation was simulated

516 now, if Richards equation is used there will always be water flux in all directions. It might be constructive to advocate Richards-based models in contrast to bucket models...

Citation: https://doi.org/10.5194/egusphere-2024-559-RC2
- AC2: 'Reply on RC2', Astrid Françoys, 12 Jun 2024
  
  We would like to thank the anonymous reviewer for the time invested in our draft paper, and for giving us the chance to provide a revised version. In the attached PDF we provide a systematic and detailed point-by-point overview of your remarks (Italic), followed by the answer after the ‘-->’ symbol. We denote any rephrased or added text in green, and underline specifically altered parts.
  
  Citation: https://doi.org/10.5194/egusphere-2024-559-AC2
RC3:
'Comment on egusphere-2024-559', Anonymous Referee #3, 16 May 2024

Abstract:
L21 this is the first location where wetting events are mentioned. Please describe these events earlier in the abstract (L15)
L24 this sentence is quite long and complicated, consider rewording for clarity or breaking up into 2 shorter sentences.
L30 the last two words in the abstract are the birch effect. For anyone who is not familiar with it this is a very confusing ending to an abstract. I suggest either explaining the birch effect at the top of the abstract or describing the process (i.e. enhanced mineralization after wetting of dry soil) instead of saying birch effect.
Intro:
In general, I think the phenomenon of capillary rise in drying soils is known and as the authors note has been thoroughly studied in the context of water availability etc. What is new here is the effects of this phenomenon on C mineralization, given the effects of water and changes in water content on it. The intro should therefore, I think, focus more on this part and here’s a good place to describe the birch effect and fit it into your hypotheses, and less on what is known and why your method has merits over other methods.
L34 a word is missing before less. Becomes? Is?
L34 In particular, (add comma)
Methods:
In general, the methods are appropriate for the proposed research questions and are well described. A few comments:
L105 a layer of silt clay loam was added between the two 1m segments to ensure connectivity. I understand the reasoning for this given that two cores are artificially places on top of the other, but doesn’t this bias the whole capillary rise measurement?
L114-L115 Because two GWT treatments were consecutively applied to each core (all in the same order of GWTs) it means that the second GWT treatment is not independent from the first because it carries over the effects of having been subjected to the first GWT treatment. Can you comment on this. If random cores were given the reverse order of treatments and shown that this does not affect the results that would have been compelling.
L121 I was disappointed that the authors did not report their results for the mineralization of native SOC and only show the 13CO2 results. This would have given a fuller picture of C dynamics in their systems. Also, because the added 13C litter is likely to ‘exist’ as particulate organic matter and a large proportion of SOC ‘exists’ as mineral associated organic matter, quantifying their relative mineralization under different GWT treatments could have greatly improved your hypotheses and results, especially given the lack of difference in 13C mineralization between soils. Since that data surely exists and does not require repeating or doing new experiments, I highly encourage the authors to include this data. The authors` concerns regarding soil properties effects on SOC mineralization can be partially addressed by normalizing CO2 to C content, and besides the mineralization of the 13C litter is likely also impacted by various properties.
L132 Why did you use VWC sensors which have to be soil-calibrated and then have to be converted to matric potential s using a retention curve instead of using water potential sensors?
L180 delta13C of CO2 undergoes fractionation at different diffusivities (e.g. water contents). https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2008JG000766
Given that your labeled material wasn’t very highly enriched, such differences can have an effect on your calculations. Were the parallel incubations of end members (L179-180) done at the same GWTs as the experimental incubations?
Results:
Consider renaming the treatments to something friendlier on the eye of the readers (e.g. instead of -165 cm and -115 cm GWT, GWT-deep and GWT-shallow)
Discussion:
L445-450 I agree that the asynchrony between 13C mineralization and water content differences are difficult to overcome. However, the differences in cumulative mineralization (fig5) and rate (fig4) become significant only as WFPS differences become significant. So I am not convinced that this surprising result is because the Birch effect was a more dominant process than water content. Could it be that litter 13C mineralization was favored at lower WFPS because it was occluded within pores that still retained water at lower WFPS especially in silt loam (e.g., https://doi.org/10.1016/j.soilbio.2022.108777), while other C sources were preferentially mineralized at higher WFPS because they were in larger pores? I again encourage the authors to look at the native C mineralization to provide a clearer picture of C mineralization in your experiments.
Regardless, if the Birch Effect turns out to be such an important aspect of the discussion of the C mineralization results, it should be explicitly termed, explained, and integrated in your hypotheses in the introduction.

Citation: https://doi.org/10.5194/egusphere-2024-559-RC3
- AC3: 'Reply on RC3', Astrid Françoys, 13 Jun 2024
  
  We appreciate the detailed feedback, which provided us with valuable insights. In the attached PDF, we provide a point-by-point response to each remark. Our responses begin after the arrows, with rephrased or added text highlighted in green and specifically altered parts underlined.
  
  Citation: https://doi.org/10.5194/egusphere-2024-559-AC3

Status: closed

RC1:
'Comment on egusphere-2024-559', Anonymous Referee #1, 02 Apr 2024

Overall, the paper is well written and complete. The introduction provides an adequate description of the context and introduces the aim of the research with useful references. The experimental design is well described. I would suggest including more information about the soils used in the experiment (e.g. soil classification). The statistical approach is well described, you report the methods and software used. The results are clearly presented and the conclusion seems to be strongly supported by the results. References are complete and up-to-date.

Citation: https://doi.org/10.5194/egusphere-2024-559-RC1
- AC1: 'Reply on RC1', Astrid Françoys, 12 Jun 2024
  
  We would like to thank you for the time and effort you invested in reviewing our draft paper. We greatly appreciate your positive evaluation. According to the WRB soil classification map of Flanders (scale 1:40,000, converted from the Belgian Soil Map: DOI: 10.13140/2.1.4381.4089), the soil profiles originating from Kruisem and Bottelare are classified as Eutric Retisols, while the soil profiles from Oosterzele are Eutric Cambisols (Siltic). Obviously, soil profile descriptions would provide a more precise assessment. We do believe that already the major affecting physicochemical specifications for moisture dynamics are supplied, e.g. detailed texture values, OC, pH, bulk densities (Table 1) and pF curves (Fig. B1).
  
  Citation: https://doi.org/10.5194/egusphere-2024-559-AC1
RC2:
'Comment on egusphere-2024-559', Anonymous Referee #2, 29 Apr 2024

27 replace 'can be' with 'is'

37 remove 'content'

50 replace 'moisture' with 'water'

66 replace 'moisture' with 'water'

114 with respect to the difficult-to-interpret results of this study, I wonder whether the lab room temperature may have had an effect. As stated, the two GWT tables were imposed for two consecutive time periods. The C turnover is highly sensitive to temperature. Let's assume a more or less standard Q10 of 1.8 for an Arrhenius-type equation. That means for a 10 degrees increase in soil temeprature you would have 1.8 times the amount of C turnover. The lab temperature has a standard deviation of plusminus 0.5 degrees. For a 1 degree increase you end up with 18 % more respiration. Now, if the average temperature in the lab for the two consecutive incubation periods differs...

144 replace 'moisture' with 'water'

145 replace 'moisture' with 'water'. Well, there is a little issure with terminology: there is either 'water content' or 'moisture'. Please correct for the entire manuscript...

207 well, the more correct term would probably be 'potential' instead of 'height'

209 I suggest to replace 'moisture transport' with 'water flux'.

209 and 210 replace 'transport' with flux

243 I would prefer to replace 'moisture transport' eith 'variably-saturated water flux'

245 that first sentence can be deleted

246 Are measured (saturated) hydraulic conductivities available? In combination with the measured soil water retention functions this would allow to estimate the infiltration depth, or even a soil water content profile over depth. To get an idea of how much the upper 20 cm with the labeled C were actually affected by the irrigation for the different soils...

Figure 2 The y-Axis of graphs should basically always start from zero.

295 A difference in heads is usually referred to as a 'hydraulic gradient'.

315 well, now you end up with two factors affecting (or not affecting) the water content in the upper 20 cm with labelled C. The ground water table AND the irrigation. The irrigation will affect the respiration of the different soils differently. This might be difficult to interprete in the end.

Fig. 3 Maybe a log10 scale applied to the y-axis helps this figure

Fig. 4 I am quite sure a log scale applied to the y-axis benefits this figure

379/380 delete or move to results

387 I assume '-30' and '-10 cm' are the coordinates in the column (i.e. the depth) and not the difference in pressure head?!

406 No, that is not unlikely. This could be a result of macropores.

409 well, there are very small lab tensiometers on the market.

421 Did you observe macropores during sampling?

435 these 'bell-shaped' curves were usually determined for disturbed soil samples. And in the wet range the decrease of respiration is rather related to an oxygen deficit and is actually not related to water content. For undisturbed samples the moisture sensitivity function of respiration actually looks different.

438 better state that the microbes show a more direct reaction to the water-filled pore-space than to the pressure head

448 I suggest to delete 'unwanted asynchrony'

453 which 'artefact'?

457 please replace 'refute' with 'reject'

458 replace 'it needs .. in mind' with 'against the background'

500 but not only the GWTs were modified, simultaneously precipitation was simulated

516 now, if Richards equation is used there will always be water flux in all directions. It might be constructive to advocate Richards-based models in contrast to bucket models...

Citation: https://doi.org/10.5194/egusphere-2024-559-RC2
- AC2: 'Reply on RC2', Astrid Françoys, 12 Jun 2024
  
  We would like to thank the anonymous reviewer for the time invested in our draft paper, and for giving us the chance to provide a revised version. In the attached PDF we provide a systematic and detailed point-by-point overview of your remarks (Italic), followed by the answer after the ‘-->’ symbol. We denote any rephrased or added text in green, and underline specifically altered parts.
  
  Citation: https://doi.org/10.5194/egusphere-2024-559-AC2
RC3:
'Comment on egusphere-2024-559', Anonymous Referee #3, 16 May 2024

Abstract:
L21 this is the first location where wetting events are mentioned. Please describe these events earlier in the abstract (L15)
L24 this sentence is quite long and complicated, consider rewording for clarity or breaking up into 2 shorter sentences.
L30 the last two words in the abstract are the birch effect. For anyone who is not familiar with it this is a very confusing ending to an abstract. I suggest either explaining the birch effect at the top of the abstract or describing the process (i.e. enhanced mineralization after wetting of dry soil) instead of saying birch effect.
Intro:
In general, I think the phenomenon of capillary rise in drying soils is known and as the authors note has been thoroughly studied in the context of water availability etc. What is new here is the effects of this phenomenon on C mineralization, given the effects of water and changes in water content on it. The intro should therefore, I think, focus more on this part and here’s a good place to describe the birch effect and fit it into your hypotheses, and less on what is known and why your method has merits over other methods.
L34 a word is missing before less. Becomes? Is?
L34 In particular, (add comma)
Methods:
In general, the methods are appropriate for the proposed research questions and are well described. A few comments:
L105 a layer of silt clay loam was added between the two 1m segments to ensure connectivity. I understand the reasoning for this given that two cores are artificially places on top of the other, but doesn’t this bias the whole capillary rise measurement?
L114-L115 Because two GWT treatments were consecutively applied to each core (all in the same order of GWTs) it means that the second GWT treatment is not independent from the first because it carries over the effects of having been subjected to the first GWT treatment. Can you comment on this. If random cores were given the reverse order of treatments and shown that this does not affect the results that would have been compelling.
L121 I was disappointed that the authors did not report their results for the mineralization of native SOC and only show the 13CO2 results. This would have given a fuller picture of C dynamics in their systems. Also, because the added 13C litter is likely to ‘exist’ as particulate organic matter and a large proportion of SOC ‘exists’ as mineral associated organic matter, quantifying their relative mineralization under different GWT treatments could have greatly improved your hypotheses and results, especially given the lack of difference in 13C mineralization between soils. Since that data surely exists and does not require repeating or doing new experiments, I highly encourage the authors to include this data. The authors` concerns regarding soil properties effects on SOC mineralization can be partially addressed by normalizing CO2 to C content, and besides the mineralization of the 13C litter is likely also impacted by various properties.
L132 Why did you use VWC sensors which have to be soil-calibrated and then have to be converted to matric potential s using a retention curve instead of using water potential sensors?
L180 delta13C of CO2 undergoes fractionation at different diffusivities (e.g. water contents). https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2008JG000766
Given that your labeled material wasn’t very highly enriched, such differences can have an effect on your calculations. Were the parallel incubations of end members (L179-180) done at the same GWTs as the experimental incubations?
Results:
Consider renaming the treatments to something friendlier on the eye of the readers (e.g. instead of -165 cm and -115 cm GWT, GWT-deep and GWT-shallow)
Discussion:
L445-450 I agree that the asynchrony between 13C mineralization and water content differences are difficult to overcome. However, the differences in cumulative mineralization (fig5) and rate (fig4) become significant only as WFPS differences become significant. So I am not convinced that this surprising result is because the Birch effect was a more dominant process than water content. Could it be that litter 13C mineralization was favored at lower WFPS because it was occluded within pores that still retained water at lower WFPS especially in silt loam (e.g., https://doi.org/10.1016/j.soilbio.2022.108777), while other C sources were preferentially mineralized at higher WFPS because they were in larger pores? I again encourage the authors to look at the native C mineralization to provide a clearer picture of C mineralization in your experiments.
Regardless, if the Birch Effect turns out to be such an important aspect of the discussion of the C mineralization results, it should be explicitly termed, explained, and integrated in your hypotheses in the introduction.

Citation: https://doi.org/10.5194/egusphere-2024-559-RC3
- AC3: 'Reply on RC3', Astrid Françoys, 13 Jun 2024
  
  We appreciate the detailed feedback, which provided us with valuable insights. In the attached PDF, we provide a point-by-point response to each remark. Our responses begin after the arrows, with rephrased or added text highlighted in green and specifically altered parts underlined.
  
  Citation: https://doi.org/10.5194/egusphere-2024-559-AC3

Astrid Françoys, Orly Mendoza, Junwei Hu, Pascal Boeckx, Wim Cornelis, Stefaan De Neve, and Steven Sleutel

Viewed

Total article views: 488 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
338	92	58	488	22	16

HTML: 338
PDF: 92
XML: 58
Total: 488
BibTeX: 22
EndNote: 16

Views and downloads (calculated since 04 Mar 2024)

Month	HTML	PDF	XML	Total
Mar 2024	73	19	2	94
Apr 2024	56	10	16	82
May 2024	50	14	6	70
Jun 2024	73	17	23	113
Jul 2024	28	13	6	47
Aug 2024	13	2	3	18
Sep 2024	23	4	0	27
Oct 2024	13	1	1	15
Nov 2024	7	7	1	15
Dec 2024	2	5	0	7

Cumulative views and downloads (calculated since 04 Mar 2024)

Month	HTML	PDF	XML	Total
Mar 2024	73	19	2	94
Apr 2024	56	10	16	82
May 2024	50	14	6	70
Jun 2024	73	17	23	113
Jul 2024	28	13	6	47
Aug 2024	13	2	3	18
Sep 2024	23	4	0	27
Oct 2024	13	1	1	15
Nov 2024	7	7	1	15
Dec 2024	2	5	0	7

Viewed (geographical distribution)

Total article views: 520 (including HTML, PDF, and XML) Thereof 520 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 13 Dec 2024

Short summary

To assess the impact of groundwater table (GWT) depth on soil moisture and C mineralization, we designed a laboratory setup using 200 cm undisturbed soil columns. Surprisingly, the moisture increase induced by a shallower GWT did not result in enhanced C mineralization. We presume this capillary moisture effect was offset by increased C mineralization upon rewetting, particularly noticeable in drier soils when capillary rise affected the topsoil to a lesser extent due to a deeper GWT.


Total:	0
HTML:	0
PDF:	0
XML:	0