Linking Sea&ndash;Air and Benthic Methane Fluxes Across Seasons in a Tropical Seagrass Meadow of Taiwan

Tseng, Hsiao-Chun; Gaol, Agnes Sonya Meilani Lumban; Lin, Fu-Hsuen; Chen, Jian-Jhih; Chou, Wen-Chen

doi:10.5194/egusphere-2026-1488

Preprints

https://doi.org/10.5194/egusphere-2026-1488

Preprints

17 Apr 2026

| 17 Apr 2026

Linking Sea–Air and Benthic Methane Fluxes Across Seasons in a Tropical Seagrass Meadow of Taiwan

Hsiao-Chun Tseng, Agnes Sonya Meilani Lumban Gaol, Fu-Hsuen Lin, Jian-Jhih Chen, and Wen-Chen Chou

Abstract. This study provides the first integrated assessment of diel, seasonal, and annual methane (CH₄) dynamics in Taiwan’s seagrass ecosystems, focusing on the Haikou seagrass meadow and adjacent coastal waters. From May 2022 to June 2023, field campaigns combined surface water sampling, in situ benthic chamber incubations and porewater profiling in both seagrass and bare sand habitats to evaluate CH₄ fluxes at the sediment–water and water–air interfaces. Results showed similar temperature and salinity patterns between seagrass and coastal waters, but seagrass habitats exhibited strong diurnal oxygen fluctuations that suppressed daytime CH₄ accumulation. Seagrass habitats consistently had higher CH₄ concentrations and sea-to-air fluxes than other coastal area, with nighttime emissions exceeding daytime values and autumn fluxes peaking under windy conditions. Sediment incubations identified benthic processes as the dominant CH₄ source. Seagrass sediments sustained relatively stable fluxes across seasons, while sandy sediment produced episodic pulses during storm events. Porewater profiles revealed elevated CH₄ in the upper 12 cm of sediments, especially in seagrass, with declines at depth due to substrate limitation and anaerobic oxidation. At the ecosystem scale, the Haikou seagrass meadow emitted approximately 78.3 mol CH₄ yr⁻¹ to the atmosphere from the water column, while sediments released 1,410.1 mol CH₄ yr⁻¹ into the water column, 94 % of which was oxidized before reaching the atmosphere or transported laterally. These fluxes are ecologically significant, emphasizing the dual role of seagrass meadows as carbon sinks and localized sources of CH₄ within tropical coastal ecosystems in Taiwan.

Received: 17 Mar 2026 – Discussion started: 17 Apr 2026

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

Hsiao-Chun Tseng, Agnes Sonya Meilani Lumban Gaol, Fu-Hsuen Lin, Jian-Jhih Chen, and Wen-Chen Chou

Status: final response (author comments only)

RC1:
'Comment on egusphere-2026-1488', Anonymous Referee #1, 13 May 2026

General comments:

The manuscript presents methane fluxes at the sediment-water and water-air interfaces from field campaigns conducted in Taiwan’s seagrass ecosystem. The study shows that methane production is primarily associated with benthic processes at the meadow and provides observations from a tropical seagrass environment for which relatively limited data are available.
The manuscript is generally well structured, and the comparison with previous studies is particularly valuable. However, several parts of the discussion and conclusions currently overinterpret the results or use wording that is stronger than supported by the data. In particular, the manuscript would benefit from a clearer distinction between methane concentrations and methane fluxes, more cautious interpretation of the seasonal dataset, and a more explicit discussion of the large variability associated with some benthic flux measurements.
The manuscript would also benefit from clearer information regarding the exact locations and distances between the sites. Without this information, it is difficult to fully evaluate the independence of the “seagrass” and “sand” stations, particularly given that methane can be transported laterally in shallow coastal waters and could therefore influence the measurements there.
Overall, the manuscript contains an interesting dataset and has the potential to make a valuable contribution, but needs substantial clarification and rephrasing before publication. The conclusion should also be revised accordingly.
Specific comments:

Abstract:

Due to the lack of winter measurements, the term “annual” should be removed. In addition, almost one full year separates autumn 2022 from summer 2023, making the extrapolation to an annual cycle uncertain. All statements here and in the text need to be rephrased.
Introduction:

Line 44: The authors should not present methane as having a single GWP value of “27–80 times greater” than CO2. Methane has a GWP of ~27 over a 100-year timescale and ~80 over a 20-year timescale. In addition, these values refer to GWP rather than SGWP (sustained-flux GWP).
Line 47: The statement: “More broadly, the coastal ocean is a global hotspot for CH₄ emissions, accounting for 75% of oceanic CH₄ emissions” does not appear to be stated explicitly in Weber et al 2019 and should be rephrased more carefully. In addition, the estimate “with continental shelves alone contributing approximately 13 Tg CH₄ yr⁻¹ from Bange et al. 1994 is outdated. The manuscript should instead use the more recent estimates shown in Weber et al. (“Extrapolation of rate measurements from active seafloor seeps across areas of likely seepage suggests that global CH4 ebullition from continental shelf sediments (0–200 m) likely falls between 18 and 48 Tg yr−1, with a most likely rate of ~35 Tg yr−1).
Methods:

Line 89: For transparency, the four sites should be indicated in Figure 1c. Since the study compares “seagrass meadow” and “bare sand” sites, the reader needs to know how far they are located from each other and from the surrounding coastal features. Methane can be transported laterally, meaning that methane measured at the sand site could potentially be influenced by emissions originating from the seagrass meadow.
Sections 2.2 and 2.3: The authors mention that the incubation experiments lasted between 2 and 5 hours “depending on sampling needs”, but these needs are not explained. Please clarify what controlled the duration of the incubations.
It is also unclear why natural water movement needed to be simulated above the chamber if the chamber was fully closed. Please clarify the role of the magnetic agitator and whether it simulated internal circulation within the chamber. Figure 2c is unclear. Zooming on one of the photos to describe the chamber in more detail could help.
Sections 2.2 and 2.3 could likely be merged for clarity. After reading section 2.2, the reader is left wondering how the physical and chemical parameters were measured. In addition, the paragraph beginning with “water samples” belongs more naturally in section 2.3.

It is also unclear how the porewater sampling was conducted relative to the chamber system. Finally, a pH sensor is mentioned but pH is not discussed later in the manuscript.
Results and discussion:

Section 3.1

The section refers to “surface water conditions”, whereas the methods indicate that temperature, salinity, and dissolved oxygen were measured within the chamber. Please clarify whether the values presented correspond to chamber measurements or ambient surface waters.
Since the values discussed in the text are not always directly consistent with Table 1 or figure 3, please clarify whether the reported values correspond to seasonal averages from all replicates. The source of the freshwater influence responsible for the lower salinity in the seagrass area should also be clarified. Again, the exact locations of the stations would help interpret these observations.
Please discuss in more detail why autumn 2022 showed lower daytime methane concentration compared to the other seasons. This pattern is likely related to storm-induced wind and enhanced mixing, as suggested by the elevated wind speeds during this period.
In addition, with only three seasonal campaigns available, it seems too strong to state: “Dissolved CH₄ concentrations in seagrass waters exhibited clear seasonal and diurnal variability…”. Please rephrase.
I am also confused by the interpretation in Lines 153-156. If methane production is primarily associated with seagrass sediments, then the lower methane concentration in coastal water is not simply related to water depth and potential oxidation. It may instead reflect the absence of methane production in sandy sediment. Alternatively, methane observed in coastal waters could partly originate from lateral transport from the seagrass area. Again, this is difficult to evaluate without precise site locations.
Section 3.3

I am not convinced that the term “relatively stable” accurately describes methane production and fluxes in seagrass sediment, especially considering the 156% increase observed in autumn. Similarly, the description of “episodic peaks” in sandy sediment may be overstated given that only three seasons are available.
The authors state that “CH₄ production rates and fluxes were higher in seagrass sediments than in sandy sediment” but then note that “autumn 2022 presented an exception”. However, this exception is huge (sandy autumn flux = 388.9 ± 382.8 and seagrass autumn flux = 105.3 ± 284.2), and strongly influences the annual average, which weakens the broader claim that seagrass is the dominant hotspot.

In addition, the extremely large standard deviations reported in autumn 2022 should be discussed explicitly, as they indicate substantial temporal and spatial variability and uncertainty in the seasonal extrapolations.
Section 3.4

The two statements in the first sentence are not fully supported by figure 6. Spring 2022 shows a distinct pattern, with stronger methane concentration in the sand. This makes the following interpretation too strong. Similarly, the description of a “steady decline” below 12 cm is overstated, as some profiles clearly show increases at depth.
The reference to Tang et al. 2025 should be explained in more detail. This study is highly relevant to the present work but is only mentioned briefly here. It should be placed into clearer context, particularly regarding differences in sampling periods. Tang et al. sampled in March-May 2023, whereas the present study was conducted in June 2023.
The rest of the paragraph is a bit hard to follow because it mixes processes derived from literature and analogies with the present study. Some references are presented in a way that suggests direct observations in the present study. For example, lines 235-236 mention bubbles observed by Sun et al 2022, although these observations were made in a different environment.
Section 3.5

The calculation of the annual methane emission of 78.3 moles is based only on three seasons which do not form a continuous annual cycle, excludes winter conditions and includes a storm in autumn. This limitation should be acknowledged more clearly and the wording should be moderated.
The statement “Seagrass sediments acted as persistent CH₄ hotspots…” (Line 243) is overclaiming and in conflict with the values (sediment-to-water fluxes are 88.2 μmol/m2/d in seagrass and 150.4 in the sand). The term “hotspot” implies a stronger methane source, yet sandy emits more methane annually. This is further complicated by the later statement “On an annual basis, sandy areas contributed approximately twice the CH₄ flux to the overlying water column compared to seagrass meadows” (Line 245) which directly contradicts the previous statement even with the rest of the sentence. The manuscript would benefit from distinguishing more clearly between 1) relatively persistent baseline emissions in seagrass sediments and 2) larger but more episodic fluxes in sandy sediments.
Line 251: It is the first time that missing winter data is acknowledged. This should be mentioned earlier.
Line 253: This sentence puts too much emphasis on oxidation. Lateral transport may imply that methane is eventually emitted to the atmosphere outside the meadow rather than oxidized locally. I suggest rephrasing to “This imbalance between sediment production and atmospheric emission suggests that a substantial fraction (here ~94%) of sediment-derived CH₄ was either oxidized within the water column or transported laterally before reaching the atmosphere.”
The comparison with other studies in Table 3 is very useful and valuable. However, including water depth for each study would improve the comparison.
Conclusions:

The conclusion should reflect the corrections previously suggested. In addition, it would benefit from a clearer distinction between methane concentrations and methane fluxes, as these concepts are occasionally mixed throughout the discussion.
Technical corrections:

Line 54: Remove “the” before “CH4 generated in sediments”

Figure 1: The squares representing the zoomed areas are difficult to see. I suggest making them thicker and red, and adding labels a, b and c to each panel. This figure would also benefit from locations, following Tang et al. 2025

Line 83: Sediment samples? Water samples? Please clarify

Lines 83-87: The seasons are introduced at the beginning of the section, followed later by “different seasons”. I suggest switching both sentences for clarity.

Lines 144-145: Cite Figure 3b after the salinity range, and Table 1 after mentioning seasonal averages, since Figure 3 only shows night and day values.

Lines 176-181: These lines simply repeat Table 2 and could be shortened.

Line 224: Wind disturbance should also be mentioned as a contributing factor

Figure 6: Please explain why error bars are absent from some porewater profiles

Line 240: Refer to Table 1

Line 243: Refer to Table 2

Line 248: Consider adding “may also intensify”

Line 304: Correct typo “approximarely”

Citation: https://doi.org/10.5194/egusphere-2026-1488-RC1
- AC1: 'Reply on RC1', Hsiao-Chun (Jean) Tseng, 25 May 2026
  
  We thank the reviewer for the constructive comments. We are currently revising the manuscript and will carefully address all comments in the revised version and provide a detailed point-by-point response.
  
  Citation: https://doi.org/10.5194/egusphere-2026-1488-AC1
RC2:
'Comment on egusphere-2026-1488', Anonymous Referee #2, 22 May 2026

The paper presents a unique dataset and well-structured fieldwork analyses of methane fluxes across seasons and compares seagrass meadows with sandy environments in coastal Taiwan. This type of research is important for improving our understanding of the role of seagrass meadows in methane emissions and coastal methane cycling. The fieldwork and analytical approaches are generally well-designed however, the manuscript requires major revisions to adequately represent the study's scope and significance.

Comments:
Introduction: The Introduction is currently too broad, contains substantial redundancy, and lacks a sufficiently clear explanation of the study's motivation, knowledge gaps, and objectives. The section requires restructuring. In addition, several citations do not refer the original studies appropriately. The entire Introduction is a single paragraph and should be divided into logically structured paragraphs.

Line 27-30: The information is somewhat redundant in the opening paragraph. It may be better to move more directly toward the relevance of seagrass meadows.
Line 39: Consider including numerical values to better demonstrate how exceptional these systems are.
Line 40: This process is essentially photosynthesis, therefore, repetition of the previous sentence.
Line 41: maybe even millennia
Line 44: SGWP: abbreviation does not seem necessary
Line 48: Consider replacing Bange et al. with a more recent reference.
Line57-58: Redundant information
Line 70: “Intensive field campaign” should be supported with numerical details

Materials and methods:
Line 86- Sampling during late May and mid-June does not sufficiently represent two distinct seasons. If seasonal comparisons are intended, additional regional information describing seasonal environmental variability should be provided.
Line 121- More detailed information is required regarding the atmospheric CH4 data, how these data were obtained, including data source, spatial and temporal resolution, and interpolation methods, if applicable. Since atmospheric CH4 is a key component in flux calculations, this methodological aspect warrants a more comprehensive description.

Results and discussion: This section should be separated into two sections. In the current structure, most subsections begin with a presentation of results, followed by broad, loosely connected discussion paragraphs. Many discussion points are insufficiently linked to the study's actual findings. A clearer structure that presents all results first, followed by a dedicated discussion section that directly refers to those findings, would substantially improve the manuscript.
Line 172. “Collectively” is vague. Which specific results support this statement? Flux estimates, spatial variability, or seasonal differences?
Line 193-211. This section lacks a direct reference to the results presented and reads more like general background information. It should be rewritten to connect to the study's findings. There are also several overstatements.
Line 222-224: There is insufficient evidence presented to support this statement.
Line 225-237: No clear reference is made to the current study results.
Line 243-245: What evidence supports the conclusion that seagrass meadows are a persistent CH₄ source? Additionally, “severe weather conditions” should be defined more clearly.
Line 252-254: More information is required for the overall circulation in the area, and more comments on the lateral transport of methane.
Line 255: This statement is not adequately supported by evidence.
Line 257-263: The discussion again lacks direct reference to the observed results.
Line 264: This appears to introduce a new subsection or topic and may require restructuring.
Line 277- 280: The interpretation appears overstated relative to the scope of the presented data.
Line 281“Taken together” is unclear. Which findings are being synthesized here? If this serves as a concluding statement, it may fit better in the conclusions.
Line 293-295: This also contains overstatements.
Line 304: “78.3 moles” requires a complete unit description (e.g., mol m⁻² yr⁻¹).
Line 310-311: The statement seems redundant.

Citation: https://doi.org/10.5194/egusphere-2026-1488-RC2
- AC2: 'Reply on RC2', Hsiao-Chun (Jean) Tseng, 25 May 2026
  
  We thank the reviewer for the constructive comments. We are currently revising the manuscript and will carefully address all comments in the revised version and provide a detailed point-by-point response.
  
  Citation: https://doi.org/10.5194/egusphere-2026-1488-AC2
RC3:
'Comment on egusphere-2026-1488', Knut Ola Dølven, 27 May 2026

See attached PDF for my review of the paper.

Citation: https://doi.org/10.5194/egusphere-2026-1488-RC3
- AC3:
  'Reply on RC3', Hsiao-Chun (Jean) Tseng, 27 May 2026
  
  We thank Professor Dølven for the detailed suggestions, particularly regarding different aspects of the flux calculations, as well as for the overall positive feedback. We are currently revising the manuscript and will carefully address all comments in the revised version, along with providing a detailed point-by-point response.
  
  Citation: https://doi.org/10.5194/egusphere-2026-1488-AC3
  - RC4: 'Reply on AC3', Knut Ola Dølven, 27 May 2026
    
    Great. Looking forward to that.
    I just realized that Nordham et al., (2025) was missing from the bibliography in the compiled PDF. The full reference to Nordham et al., (2025) is:
    Nordam, T., Dissanayake, A., Brakstad, O., Hakvåg, S., Øverjordet, I., Litzler, E., Nepstad, R., Drews, A., and
    
    Röhrs, J. (2025). Fate of dissolved methane from ocean floor seeps. Environmental science & technology, 59.
    
    Citation: https://doi.org/10.5194/egusphere-2026-1488-RC4

Hsiao-Chun Tseng, Agnes Sonya Meilani Lumban Gaol, Fu-Hsuen Lin, Jian-Jhih Chen, and Wen-Chen Chou

Viewed

Total article views: 434 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
319	88	27	434	14	18

HTML: 319
PDF: 88
XML: 27
Total: 434
BibTeX: 14
EndNote: 18

Views and downloads (calculated since 17 Apr 2026)

Month	HTML	PDF	XML	Total
Apr 2026	146	44	13	203
May 2026	173	44	14	231

Cumulative views and downloads (calculated since 17 Apr 2026)

Month	HTML	PDF	XML	Total
Apr 2026	146	44	13	203
May 2026	173	44	14	231

Viewed (geographical distribution)

Total article views: 434 (including HTML, PDF, and XML) Thereof 434 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 30 May 2026

Short summary

Seagrass habitats showed higher CH₄ concentrations and emissions than nearby coastal waters. Sediments were the main CH₄ source, with stable emissions from seagrass sediments and storm-driven pulses from sandy areas. Although large amounts of CH₄ were produced in sediments, about 94 % was oxidized in the water column before reaching the atmosphere or transported laterally.


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