the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Aug 2024
Eddy Covariance Evaluation of Ecosystem Fluxes at a Temperate Saltmarsh in Victoria, Australia Shows Large CO2 Uptake
Abstract. Recent studies highlight the important role of vegetated coastal ecosystems in atmospheric carbon sequestration. Saltmarshes constitute 30 % of these ecosystems globally and are the primary intertidal vegetation outside the tropics. Eddy covariance (EC) is the main method for measuring biosphere-atmosphere fluxes, but its use in coastal environments is rare. At an Australian temperate saltmarsh site on French Island, Victoria, we measured CO2 and water gas concentration gradients, temperature, wind speed and radiation. The marsh was dominated by a dense cover of Sarcocornia quinqueflora. Fluxes were seasonal, with minima in winter when vegetation is dormant. Net ecosystem productivity (NEP) during the growing season averaged 10.54 g CO2 m-2 day-1 decreasing to 1.64 g CO2 m-2 day-1 in the dormant period, yet the marsh remained a CO2 sink due to some sempervirent species. Ecosystem respiration rates were lower during the dormant period compared with the growing season (1.00 vs 1.77 μmol CO2 m-2 s-1) with a slight positive relationship with temperature. During the growing season, fluxes were significantly influenced by light levels, ambient temperatures and humidity. Evapotranspiration peaked at 0.27 mm h-1. We cautiously estimate the annual NEP budget at this marsh to be 753 (±112.7) g C m-2 y-1 which is similar to carbon uptake by temperate saltmarshes in Europe and within the range measured at some US saltmarshes. This value is higher than the value hypothesised for global saltmarshes of 382 g C m-2 y-1 but is only half the mean value estimated for global mangroves.
- Preprint
(85759 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 30 Oct 2024)
-
RC1: 'Comment on egusphere-2024-2182', Anonymous Referee #1, 12 Sep 2024
reply
Review “Eddy Covariance Evaluation of Ecosystem Fluxes at a Temperate Saltmarsh in Victoria, Australia Shows Large CO2 Uptake” egusphere_2024-2182 by Reef et al.
Summary:
The manuscript presents the first study using EC to measure CO2 exchange in an Australian temperate saltmarsh over nearly 2 years, from November 2019 to August 2021. CO2 fluxes over the saltmarsh were highly seasonal with greatest fluxes during the growing period and smallest fluxes during the dormant period. Investigated relationships with selected environmental drivers showed growing season daytime NEE was mostly dependent on radiation (short-wave incoming) followed by temperature (showing a temp optimum for CO2 uptake) and a threshold for VPD. Data coverage was compromised due to many factors and data was not gap-filled. Authors cautiously present an annual estimate for NEE (or NEP).
The study is certainly of interest and within the scope of Biogeosciences. Although the study is more descriptive it is novel being the first study of EC-measurements in a Southern Hemisphere/Australian temperate saltmarsh. Data coverage was compromised but this is often the nature with EC-measurements, hence gap-filling procedures are often crucial. The manuscript would need improvements on various aspects listed below and needs to address inconsistencies, and provide clarification, better descriptions and more rigorous approach around data processing and analysis.
Definition of season: Throughout the manuscript the definition and length of growing and dormant season is inconsistent. In L200-202 the growing season is Oct-May and dormant season is June-Sept while shortly after in L208-209 the growing season is Oct-Mar and dormant season is Apr-Sept, earlier in the introduction the growing season was spring + summer and dormant season was autumn + winter. Later in L288 season were highlighted as Nov-Mar for growing season and Aug-Sep for dormant season. This is confusing and authors should clarify and be consistent with season definitions throughout the manuscript.
Data gaps and processing:
The authors acknowledge well the circumstances that led to data loss but it would be good to present a better table or figure on data gaps/availability in percentage. Currently, Fig 3) does not show much and is difficult to read with the current ratio of panels. It would be better to show a timeseries figure that shows measured data combined in stacking panels, i.e. Fig2a) with Fig3 a) and b) combined, also including radiation, VPD and ET. Time periods that have been excluded due to big data gaps can be highlighted, as well as dormant and growing season within one panel.
L237/238 - are the authors referring to half-hourly measurement points? The numbers themselves do not mean much, it would be better to state the percentage of data coverage during the selected measurement period, either as a whole or split by dormant and growing season (i.e. 1 y 9 m is roughly 638 days, i.e. data coverage is xx days or xx %)
The authors need to describe in better detail what data processing steps have been done, i.e. for each correction are different options available and these should be stated and references cited with them. Current version L225-228 is insufficient. Furthermore, how as the u* threshold determined. This needs to be better described and clarified.
Why did the authors choose to not gap-fill data. While using non-gapfilled data is valid to explore relationships with environmental drivers, it makes strong outcomes of seasonal variability difficult if the data gaps are skewed towards certain periods of the day or year. Gap-filling bigger data gaps is problematic but gap-filling between time periods with most coverage should be possible and within the flux community sufficient approaches are available.
Terminology:
NEP is the Net ecosystem productivity and is not equal to net ecosystem uptake, only when NEP is positive. The authors should clearly define and be consistent throughout the manuscript what negative or positive NEP is, either uptake or emission. Generally net ecosystem CO2 uptake as NEP is positive, while shown as NEE it is negative. Most of the time positive NEP in the manuscript is uptake but then in the discussion NEP uptake off other studies are indicated with negative symbol and also the cited net emissions (L488 to L493).
L127 – instead of photosynthetic flux of CO2 it would be better to say GPP of saltmarshes is the CO2 uptake by all plants via photosynthesis. While GPP is explained, ecosystem respiration is not, hence it should be stated all potential CO2 sources contributing to ecosystem respiration from the saltmarsh.
L268: not once is GEP mentioned further in the manuscript and thus, the definition here is not relevant.
Relationships with environmental drivers:
Earlier studies have identified that NEE varies between saltmarshes that was linked to differences in species, salinity, hydrology and biochemical conditions. Authors should consider discussing highlighting such differences with regards to measured NEE in this study and reported ones in previous studies.
The authors mention that the saltmarsh experiences semi-diurnal tides. This is an important environmental driver for the ecosystem and consequently for CO2 fluxes and thus it would be important to assess how a change in water table would influence CO2 exchange. I’m surprised that authors did not consider this in their study.
Authors mentioned heatwaves or high temperatures – did authors consider how daily or diurnal NEE patterns compare between different temperature conditions, i.e. hot days vs mild/warm days?
Can authors clarify why they chose certain environmental drivers and correlations, i.e. Fig 5b – the rationale for the NEP – temperature relationship? I would like to see the relationship between ecosystem respiration and temperature that the authors mention but not show, as this is one of the most important relationships used in a method to estimate gross fluxes, GPP and ER, from NEE. This relationship is most certainly not linear, but exponential – commonly described with the Lloyd & Taylor type Arrhenius function (also see L266)
Given the Gaussian relationships shown for temperature (or VPD), it would be good to explore and look into potentially confounding effect of environmental drivers as well.
Annual Budget:
Given the substantial data gaps and the decision to not gap fill the data, presenting an annual NEP estimate, even cautiously, is not warranted. It is not where along the Nov 2019 to Aug 2021 timeline the data coverage is.
It is not clear or understandable what is meant with the 24-hour carbon balance integration, for what period of time, what data? From Fig.6 it seems only the 90 + 18 days have been some sort of gap filled that should describe the daily range of NEE? If only a certain days are getting gapfilled why not others? As mentioned above there are approaches available to gap fill data, either using monthly diurnal values or ANN, etc.
However, it is not clear how the annual budget was calculated. It is not possible to derive an annual budget based on 108 days. Thus, it is not possible to follow the steps that would allow any reproducibility.
The same applies to the data analysis in general. Authors state all has been done in R, but do not cite any significant or important package used, or R for itself.
I recommend to authors to look into better gapfilling procedures if some sort of annual budget is desired and also to improve the environmental driver analsyis. If the entire data analysis is only based on 90 or 18 days, it needs to be acknowledged that the significance of results is limited.
Minor comments:
The highlighted background colours are not explained (Fig2a). Does it show data gaps, availability or growing/dormant season?
Fig 5 – better figure caption is necessary and should include what data has been used, i.e. daytime data, and only the growing season, non-gapfilled, but are these half-hourly or daily values? In the text it states daytime NEE which has been set to >12 W .m-2 but the figure clearly shows data below that threshold. What is actually shown and the correlations based on? Authors should clarify what is uptake, i.e. positive NEP, but then the data also includes negative values, hence shown data it is not strictly uptake, but rather NEP – Net ecosystem productivity.
Discussion:
L402 to 411 – I question the relevance of non-flux processes with regards to seasonality. This is not really relevant for the carbon or CO2 flux processes investigated here nor the defined objectives of the study.
In relation to NEE estimates from saltmarshes elsewhere, the discussion would improve to higlight the potential differences between these ecosystems that might explain differences in NEE (or NEP).
In the introduction it is highlighted that saltmarshes are defined by a strong seasonality but then in the discussion and conclusion it is somehow a novelty that Australian saltmarshes have a seasonality. This is confusing, can the authors explain in more detail?
ET – the authors acknowledge the distinct surface areas within the footprint that will affect any measurements on ET and thus estimates on WUE. These are also visible in Fig. 1c) and 1d) – Hence ET values, especially during the summer period might include a greater contributions of water bodies and thus would bias any WUE estimates currently presented. It might be possible the authors attempt to account for or sector out water bodies by i.e. excluding certain wind-sectors were water bodies are from the footprint .
Citation: https://doi.org/10.5194/egusphere-2024-2182-RC1
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 158 | 25 | 10 | 193 | 4 | 2 |
- HTML: 158
- PDF: 25
- XML: 10
- Total: 193
- BibTeX: 4
- EndNote: 2
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1