Drivers of CO<sub>2</sub> emissions during the dry phase of Mediterranean and Temperate ponds

Frutos Aragón, Victoria; Brucet, Sandra; Marcé, Rafael; Bucak, Tuba; Davidson, Thomas Alexander; von Plüskow, Louisa-Marie; Lemmens, Pieter; Trochine, Carolina

doi:https://doi.org/10.5194/egusphere-2025-3725

Preprints

https://doi.org/10.5194/egusphere-2025-3725

Preprints

12 Aug 2025

| 12 Aug 2025

Status: this preprint is open for discussion and under review for Biogeosciences (BG).

Drivers of CO₂ emissions during the dry phase of Mediterranean and Temperate ponds

Victoria Frutos Aragón, Sandra Brucet, Rafael Marcé, Tuba Bucak, Thomas Alexander Davidson, Louisa-Marie von Plüskow, Pieter Lemmens, and Carolina Trochine

Abstract. Pond ecosystems play an important role in the global carbon cycle with the potential to act as both sinks and sources. Emissions of CO₂ during their dry phases remain largely overlooked, despite growing evidence that climate change-induced shifts in temperature and precipitation will likely result in longer and more frequent dry periods. Here we assess CO₂ emissions from dry pond sediments in relation to climatic region, seasonal changes, and hydroperiod duration. Specifically, we aimed to identify the key environmental drivers shaping CO₂ fluxes during the dry phase. We measured CO₂ emissions from air-exposed sediments in 30 ponds across Mediterranean and Temperate regions. Ponds acted as sources of CO₂ during dry phases, with emissions ranging from 127 to 4889 mg C m⁻² d⁻¹ (mean ± SD = 1398 ± 1201). Although mean emissions did not differ significantly between climate regions, hydroperiod length interacted with climate and season, showing a significant effect in summer, particularly in Mediterranean ponds, where longer hydroperiods led to higher emissions. Emissions were considerably higher in summer than in autumn, primarily driven by an interaction between sediment temperature and water content. The highest fluxes occurred at c. 27 °C and sediment water content between 27 % and 44 %. Additionally, ponds in better conservation status and with lower carbonate content emitted more CO₂. Our findings improve understanding of CO₂ emissions during increasingly common dry phases and highlight how climate modulates local sediment conditions, thereby influencing the magnitude of these emissions. This underscores the need for comprehensive assessments of carbon fluxes that incorporate dry-phase emissions, accounting for climate, hydroperiod, and both direct and indirect effects of local environmental drivers.

Received: 31 Jul 2025 – Discussion started: 12 Aug 2025

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.

Download & links

Preprint (PDF, 1332 KB)

Supplement (1001 KB)

Download & links

Victoria Frutos Aragón, Sandra Brucet, Rafael Marcé, Tuba Bucak, Thomas Alexander Davidson, Louisa-Marie von Plüskow, Pieter Lemmens, and Carolina Trochine

Status: open (until 06 Nov 2025)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2025-3725', Anonymous Referee #1, 26 Sep 2025 reply
Review of the manuscript “Drivers of CO₂ emissions during the dry phase of Mediterranean and Temperate ponds” by Frutos-Aragón et. al for Biogeosciences:
The study examines CO₂ fluxes of dry sediments from 14 temperate and 16 Mediterranean ponds in Europe during their dry phase in summer and/or autumn of 2022. These fluxes were measured using chambers. Additionally, sediment and water analyses were conducted to characterize the sites and to investigate the differences.
I read the manuscript with great interest and believe that the study is promising, but that some improvements are necessary.
General comments or questions:
In the abstract, the methodology for measuring CO₂ fluxes is missing.

How exactly and to what extent does your study fill the knowledge gap you mentioned in the introduction?

In the introduction, a concluding sentence on how the questions will be answered is missing.

The coordinates of the sampling or measurement points within the sites, the sampling dates, and the names of the studied ponds are missing.

The chamber description still lacks some information (see the detailed comments).

For me, it is unclear how often or at what frequency the CO₂ measurements were conducted in each season. Were there temporal replicates?

In the manuscript, there is no information on whether the data requirements for the statistical tests used (e.g., ANOVA or t-test) are met.

I miss a conclusion chapter.

The graphics are slightly pixelated, and the image quality could be improved.

Detailed comments and suggestions:
Abstract: The methodology is missing

L23ff: What do you mean by hydroperiod?

L26: c.?

L37f: It would be good to specify the CO₂ emission value of ponds.

L104f: Instead of spanning latitudes or longitudes, it would be better to add the actual coordinates of some sites and refer to the appendix.

Figure 1: Since graphs a) and b) are already zoomed in, it would be helpful to include an overview graph with all sites, countries, climate regions, etc.

It would be interesting to categorize the sites in Figure 1 according to the dry or wet phase definition written in L109 ff.

Method chapter: Please use the same description style for each instrument. Currently, it differs. You do not have to repeat it if the instrument is already mentioned.

L117: What do you mean by “dry fluxes”? I also recommend using N instead of n for the number of observations/samples throughout the manuscript.

L123ff: Why were 40-year averages of annual temperatures and precipitation used instead of the more common 30-year averages for the climatic description of sites?

L130: You have finally defined what you mean by "hydroperiod length," although the term was already mentioned before. It would have been helpful to define it the first time it was mentioned.

L157: Why did you use a filter size of 0.7 µm to obtain the dissolved fraction instead of, for example, 0.45 µm?

L165ff: This subchapter is missing important information about the chamber measurement (e.g., whether it is a through-flow or static system), the chamber type (transparent or opaque), the chamber size (area, height, and volume), and the chamber material. In addition, did you use any additional materials during the measurement, such as tubes or a pump? What time of day did you conduct the chamber measurements at the four to eight spots per pond? How many measurement days were there per season at each pond, and how often were they conducted? Did you correct the measured CO₂ concentrations for water vapor?

L167: I am not familiar with the mentioned sensor. What is its precision, compared to a Licor, Los Gatos, or Picarro gas analyzer, for example? At what frequency does this sensor measure during the five-minute closing time?

L174f: Explain why one measurement had a different sampling technique and measurement time. Was it a one-hour measurement time or a one-hour closing time?

L175: What was the size and the material of the vials or syringes? When during the one-hour closing time did you measure, and how much volume did you take? Were your chambers equipped with an overpressure valve?

L175f: What did you measure with the gas chromatograph? This is unclear to me, as it was never mentioned again in the manuscript.

L178: Why did the CO₂ sensor did not require a calibration?

L178f: What data did you use for the 3-point average, and why was there background noise?

L180: Why did you only use the last two to three minutes of each five-minute measurement period? What is the reference for Equation 1?

L182: Why did you choose to use the carbon unit for the fluxes? Additionally, you used hourly units here, but daily units throughout the manuscript.

L190: Milli-Q is a brand name, not a water type.

L191: Why did you use 48 hours?

L195: Please cite the references that used this as a proxy.

L220: Table S2 is mentioned before Table S1. Please reconsider the order of the tables.

L221f: This sentence could be moved to the next subchapter, “Statistical (or data) analysis”.

L224ff: Have you had tested your data for normality?

L259: The R version is important for repeatability.

Figure 2: The unit in the y-axis label is missing a bracket. Why does the boxplots have different widths? Does the x-axis label mean Pond ID? To better illustrate the differences and support your results, I would reconsider the representation and categorize by other environmental variables.

L263: Instead of "overall," I would say "on average" that all your ponds were a CO₂ source, as you have also measured a few negative CO₂

L265: It would be better to call them CO₂ "fluxes" instead of "emissions" since you have measured some negative values.

L270: Both are red lines. Please be more precise in the description. What is the mean value of?

Table 2: I don't think the abbreviation T-FCO₂ is necessary or correct here. I would rather use the term "CO₂ fluxes" over the four columns on the right, including the unit. Where T-FCO₂ and the unit are now, I would put the mean ± SD. I also recommend including the number of observations.

Figure 3: In the graph, you used n.s., but in the caption, you explained the abbreviation NS. In the supplementary figures, you wrote that an absence indicates no significance. Be consistent.

L299: Here, you wrote ".01"; in L302, you wrote "0.01" for p. Be consistent throughout the entire manuscript.

Figure 4: What are the R² values of the linear regression lines?

L305: The caption of Figure 4 mentions a dashed line, but I cannot see one in the figure.

Table 3: I miss the p values you mentioned in L302f.

Figure 5: I'm not sure what to say about these trend lines. I don't trust them because they look like a point cloud with temperature differences and the influence of the edge effect.

L345: Sometimes there is a space between the number and "Celsius," and sometimes there isn't. Be consistent throughout the manuscript.

Table 5: In the text, you always used SD, but in the tables, you used sd. Please be consistent and explain every abbreviation in the table caption. Use the same rounding for all emission values shown here. Since seasonality affects CO₂ fluxes in your studied ponds and in most ecosystems outside the tropics, it would be helpful to know what season the reference values were measured.

L485: The data cannot be reviewed because they are unavailable. I recommend that the authors make the data easily accessible to everyone, not just upon request, for reasons of repeatability and reusability.

The information about the R packages used is missing and can be added to the references or the methods chapter.

Why did you use two appendices (A and B)?

Table A2: Could you add more lines to better separate the categories in the left column and put each variable name in its own row? This would make the table easier to read.

Figure S1: Variable names, including units, are sometimes split into two lines. This makes the table difficult to read. Please reconsider this.

Figure S2: In the manuscript, you called it "Pond ID"; here, on the x-axis, it is "Pond Code." Be consistent. The y-axis is missing a unit.

Figure S3: Please include the number of observations for each box plot.

Figure S4: There is a typo in the x-axis label: months.

Figure S5. The sentence about the absence of an asterisk can be removed from the caption.

Reply
Citation: https://doi.org/10.5194/egusphere-2025-3725-RC1
RC2:
'Comment on egusphere-2025-3725', Ji-Hyung Park, 09 Oct 2025 reply
Please note that this referee report is provided by Associate Editor to expedite the delayed review process.

General comments
The manuscript presents two-season field measurements of CO₂ emissions from air-exposed sediments in 30 ponds across Mediterranean and temperate climate zones. Based on the significant relationship between hydroperiod lengths and CO2 emissions in Mediterranean ponds in summer, the authors suggest that longer hydroperiods play a critical role in creating temporary conditions for higher CO2 emissions. Using various statistical approaches, they further identified key drivers of sediment CO2 emissions, including temperature and sediment contents of water and carbonate. The key findings from the well-designed study are novel and invite further study to elucidate the large temporal variability in CO2 emissions from ponds, which have been understudied compared to other freshwater systems. Despite the novelty and significance of the key findings, the manuscript shows weakness in linking and interpreting these findings, as well as a lack of detail in several areas, as described below. I hope my comments will help the authors improve the logical flow and clarity of the manuscript.

1. Hypotheses and data interpretation
Although hydroperiods and sediment water contents are suggested as the primary controls on sediment CO2 emissions, descriptions across Introduction, Results, and Discussion appear not consistent, and in some cases contradictory. First, hypotheses (2) and (3) need to provide more interrelated and mechanistic predictions. Higher sediment contents might be influenced more directly by more recent precipitation events (like 1-month or 1-week antecedent precipitation) than the yearly hydroperiod as considered here. Please provide a more detailed explanation of the relationship between hydroperiods and water contents. Any rationale for using hydroperiod rather than other drought indices would also be helpful. Second, the findings shown in Fig. 4 indicate the significant relationship between hydroperiods and CO2 emissions only for Mediterranean ponds in summer, and the significance appears controlled by a few sites with very long hydroperiods. However, this hydroperiod effect is emphasized too much across the R & D sections, with some of them having inconsistent connotations: for instance, refer to L 275-276 (“Mediterranean ponds exhibited higher air and sediment temperatures, shorter hydroperiods, typically drying in summer. They also showed lower sediment water content, and reduced macrophyte coverage, consistent with an earlier drying period.”). Please check the consistency of descriptions across R & D (sections 4.1 and 4.2 appear to address two separate stories regarding the hydroperiod effect) to provide a more coherent explanation for the relationship between hydroperiods and sediment water contents.

2. Realigning paragraphs
Although the manuscript was easy to follow on a sentence-by-sentence level, the use of very long or several scattered short paragraphs made it difficult to grasp the overall logical structure. In the Introduction, for example, the page-long initial paragraph is followed by five short paragraphs. A thorough revision of the manuscript is recommended to reorganize the long and short paragraphs in accordance with a coherent logical flow.

3. Clarity of tables and figures
There are numerous missing or inaccurate details that could be improved through careful revision. Please refer to the specific comments below.

Specific comments
Title: A slight change would enhance clarity: for example, Drivers of CO2 emissions during the dry phase “in” Mediterranean and Temperate ponds or Drivers of CO2 emissions “from” Mediterranean and Temperate ponds “during the dry phase”

Line (L) 17: sources of carbon (or CO2)?

L 17 “remain largely overlooked”: This statement overlooks the decadal research on this topic.

L 25: “the” interaction

L 35-38: Please provide some estimates of CO2 and CH4 emissions from ponds to describe their role more quantitatively.

L 85: Please define “hydroperiod length”.

L 96: Without the above-mentioned definition, it is difficult to understand “shorter hydroperiods leading to lower emissions due to reduced sediment water content”.

L 99: Can you illustrate “conservation status” using an example?

L 117: Did 23 sites also include semi-permanent and permanent ponds? In the latter case, the described bare sediment would be contradictory to the definition of permanent ponds (L 110).

L 130 “water presence”: Do you mean rainy days or literal water presence in ponds?

L 154 (throughout the manuscript): not Chlorophyll a, but chlorophyll a

L 167: Please provide key details on the chamber design, including the used material, size, ventilation, etc.

L 174-177: It would provide useful information for assessing the accuracy of sensor data if you compare sensor and additional GC measurements.

L 195: Please provide a relevant reference for this carbonation estimation.

L 224: How did you test the normal distribution of your datasets?

L 265: Are these negative values from partially water-flooded sediments where phytoplankton take up CO2? Please elaborate on the site characteristics and discuss the meaning of these values (if outside measurement error ranges).

L 284-286: Please clarify whether you are talking about the proportion of each component based on unit mass of sediment or DOC.

L 288-293: These sentences are good examples of unnecessary separation mentioned before.

L 300: Given the significance of the hydroperiod effect, it would be helpful to elaborate more as to how “the effect of hydroperiod was season-specific and climate-dependent” as displayed in Fig 4.

L 301: Was the summer trend also significant for the temperate sites?

L 320-330: In a sense, this part seems secondary, but covers the bulk of section 3.2. More space could be saved for more relevant drivers.

L 354 “all ponds emitted CO₂ during the dry phase”: This statement is contradictory to the result descriptions (Fig. 2).

L 357: “shaped” or “was shaped by”?

L 445: It would help readers to compare the magnitudes of plant uptake vs. CO2 emissions if you provide some literature values estimating plant C uptake.

L 417: Fig 5 shows the generally highest levels of CO2 emissions across the highest temperature ranges.

L 460 “ponds with more permanent hydroperiod”: This is quite confusing, given your descriptions of your sites. Did you mean simply “longer hydroperiod”?

Fig 1 caption: Countries “are”

Fig 2: Please complete the vertical axis title with the second parenthesis.

Table 2: If this displays the same data as Fig 2, please think about removing or revising it to avoid double presentation.

Fig 4: Please indicate the significance levels for the depicted regressions. It would be easier to find out the significance if only significant regressions were shown as regression lines.

Tables 3, 4, 5: Please explain in the caption the abbreviations including SE, df, CL, AIC, BIC, and CI.

5: What is ORQ? Are all the depicted trends statistically significant?

Table 5: What about showing the employed models in a separate column?

Table “6” (page 18): Please also correct the unnecessary values below the decimal point.

Reply
Citation: https://doi.org/10.5194/egusphere-2025-3725-RC2

Victoria Frutos Aragón, Sandra Brucet, Rafael Marcé, Tuba Bucak, Thomas Alexander Davidson, Louisa-Marie von Plüskow, Pieter Lemmens, and Carolina Trochine

Supplement

https://doi.org/10.5194/egusphere-2025-3725-supplement

Victoria Frutos Aragón, Sandra Brucet, Rafael Marcé, Tuba Bucak, Thomas Alexander Davidson, Louisa-Marie von Plüskow, Pieter Lemmens, and Carolina Trochine

Viewed

Total article views: 1,400 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
1,337	47	16	1,400	23	17	17

HTML: 1,337
PDF: 47
XML: 16
Total: 1,400
Supplement: 23
BibTeX: 17
EndNote: 17

Views and downloads (calculated since 12 Aug 2025)

Month	HTML	PDF	XML	Total
Aug 2025	627	16	5	648
Sep 2025	661	16	7	684
Oct 2025	49	15	4	68

Cumulative views and downloads (calculated since 12 Aug 2025)

Month	HTML	PDF	XML	Total
Aug 2025	627	16	5	648
Sep 2025	661	16	7	684
Oct 2025	49	15	4	68

Viewed (geographical distribution)

Total article views: 1,409 (including HTML, PDF, and XML) Thereof 1,409 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 20 Oct 2025

Short summary

Pond sediments emit substantial CO₂ during dry phases, often overlooked despite climate change increasing dry periods. We measured emissions in 30 ponds across Mediterranean and Temperate regions, finding higher emissions in summer and in ponds with longer hydroperiods, especially in Mediterranean climates. Emissions peaked at moderate sediment water content and warm temperatures, highlighting the need to include dry-phase emissions in carbon flux assessments.


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

Drivers of CO2 emissions during the dry phase of Mediterranean and Temperate ponds

Supplement

Viewed

Viewed (geographical distribution)

Drivers of CO₂ emissions during the dry phase of Mediterranean and Temperate ponds