the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
From small scale variability to mesoscale stability in surface ocean pH: implications for air-sea CO2 equilibration
Abstract. One important aspect of understanding ocean acidification is the nature and drivers of pH variability in surface waters on smaller spatial (i.e., areas up to 100 km2) and temporal (i.e., days) scales. However, there has been a lack of high-quality pH data at sufficiently high resolution. Here, we describe a simple optical system for continuous high-resolution surface seawater pH measurements. The system includes a PyroScience pH optode placed in a flow-through cell directly connected to the underway supply of a ship through which near-surface seawater is constantly pumped. Seawater pH is measured at a rate of 2 to 4 measurements min-1 and is cross-calibrated using discrete carbonate system observations (total alkalinity, dissolved inorganic carbon and nutrients). This setup was used during two research cruises in different oceanographic conditions: the North Atlantic Ocean (December–January 2020) and the South Pacific Ocean (February–April 2022). Our findings reveal fine-scale fluctuations in surface seawater pH across the North Atlantic and South Pacific Oceans. While temperature is a significant abiotic factor driving these variations, it does not account for all observed changes. Instead, our results highlight the interplay between temperature, biological activity, and water masses on pH. Notably, the variability differed between the two regions, suggesting differences in the dominant factors influencing pH. In the South Pacific, biological processes appeared to be mostly responsible for pH variability, while in the North Atlantic, additional abiotic and biotic factors complicated the correlation between expected and observed pH changes. Although surface seawater pH exhibited fine-scale variations, it remained relatively stable over a 24-hour cycle due to reequilibration with atmospheric CO2. Thus, for the regions and time periods studied, ocean basin-scale analyses based on discrete samples collected during traditional research cruise transects would still capture the necessary variability for global CO2 cycle assessments and their implications.
- Preprint
(2817 KB) - Metadata XML
-
Supplement
(1916 KB) - BibTeX
- EndNote
Status: open (extended)
-
CC1: 'Comment on egusphere-2024-2853', Christopher Sabine, 25 Nov 2024
reply
General Comments:
This is a well written and well documented manuscript, but I am concerned about the accuracy of the measurements and the robustness of the conclusions. Based on the final corrected pH plots in figure 3, it looks like the pH changes along the two transects were very minimal. The authors calculate that the largest influence on pH over a 24-hour cycle is temperature, with an occasional hint of a biological signal. This really isn’t a surprise. These variations are occurring in both time and space, so it is impossible to quantitatively separate the effects. The authors conclude, “…although the processes responsible for these pH variations are well-understood, high-resolution data highlight the challenge of identifying a dominant factor at the fine-scale due to their complex interplay.” This makes me question what is the point of this manuscript? Perhaps the authors can do a better job of explaining what is new and novel about this work.
Specific Comments:
Lines 256-257: How do the authors define unreasonable drift patterns? This approach sounds very subjective.
Lines 259-263: The authors are correcting the measured pH values to pH calculated from TA and DIC, but how do they know the calculated values are correct? Which constants were used for the calculations? Is there no direct measurement of pH to validate the corrections? What about the two-point calibrations described in the methods? Were these calibrations not helpful? How frequently were they done?
Figure 3: Some of the pH changes are very large and the final pH curves look nothing like the raw data. If all the calculated pH values are used to adjust the underway pH measurements, then what confidence do we have that the values in between the calibrations are correct? Are there any independent pH measurements that were not used for calibration that we can use to assess accuracy?
Line 265-279: It seems a bit circular to use TA to adjust the pH values, then turn around and use TA together with the pH to calculate the other parameters. How do the Lee et al. TA values compare to the measured values? Are these uncertainties smaller than if the authors simply took all the measured TA and DIC values to calculate the other carbon parameters?
Line 382: This is not a traditional use of the term water mass. One does not normally think of water masses as surface features because external forcing (warming/cooling, precipitation/evaporation, etc.) can make water properties quite variable, compared to traditional subsurface water masses that have stable properties that can be defined and tracked as they move into the ocean interior. I understand what the authors are trying to say, but I think a different term for waters with different properties needs to be used.
Line 392: This section is focused on the influence of T and S on pH, which seem to have signals that are less than 0.01. I am wondering how robust these signals are if the raw pH values had to be corrected by ~0.4 units (fig 3) and the uncertainties in the final values are around +/-0.01 (fig 4 and 5). How do the authors know these are signals and not just noise that they are interpreting?
Line 414: I do not understand this statement about the waters having lower thermal inertia. What do the authors mean? Lower than what? Water has a low thermal inertia compared to the air, but the sentence is trying to explain why there is a diurnal temperature signal in the surface water. The deeper waters do not have a higher heat capacity, they are just removed from the forcing.
Line 635-637: This sentence seems to convey the essence of this manuscript: However, although the processes responsible for these pH variations are well-understood, high-resolution data highlight the challenge of identifying a dominant factor at the fine-scale due to their complex interplay. What is the take home message that you are trying to convey? It sounds like there isn’t much point in doing these high-resolution measurements.
Citation: https://doi.org/10.5194/egusphere-2024-2853-CC1 -
RC1: 'Comment on egusphere-2024-2853', Anonymous Referee #1, 21 Jan 2025
reply
General Comments
The authors present their work discussing the need for fine spatial and temporal pH data, thus requiring the high resolution measurements they made on two transect cruises. However, then ultimately conclude that less resolved traditional research cruises are able to capture the pH variability we need. The authors mention the use of these fine scale measurements in coastal and dynamic environments, which makes sense. But the authors have failed to convey what is the need for the same fine scale measurements in the less dynamic open ocean environments, and what the measurements would be used for. Many processes are at play, but what is the goal of disentangling these processes in the open ocean? The authors should try to convey their message without devaluing their work in the conclusions.
With regards to scientific significance, the authors present underway pH measurements that are not very common to my knowledge. Importantly, they calibrate their underway measurements with discrete measurements. The authors highlight the use of underway pH measurements, though ultimately conclude they are not necessary to be this fine scale, somewhat mitigating the significance of their work. With regards to scientific quality, I do appreciate the details the authors present regarding their underway pH measurements and associated discrete measurements. However, their corrections and offsets are quite large. They compare very small changes in pH to determine which influences (temperature, biological activity, or water masses) are dominant, but their changes are much smaller than the offsets they present. This leads me to question if these small changes are really meaningful within all the noise. With regards to presentation quality, the authors generally present their work in a clear manner. The results section discussing the influences to pH is a bit confusing to follow. I will discuss more about this in the specific comments section below.
I am concerned with the results of this manuscript, which compare observed (corrected) underway pH values with pH calculated in a variety of manners. The authors suggest very small residuals, but this does not account for the uncertainty in their underway pH measurements. Specifically, in Fig. 4 and 5, almost all of the various pH calculations fall within the pH uncertainty. So I am not sure specific conclusions can be made from this pH data.
Specific Comments
The authors indicate they are looking at fine scales of up to 100 km2 and days across ocean basins (line 106). However, later in the manuscript they begin discussing pH changes on the order of seconds to minutes (lines 402-418). The controls of pH are being convoluted between larger scale processes and split-second thermodynamics. In reality, the instant thermodynamics of the carbon system are not controlling the pH at longer time scales. The whole thing is quite confusing.
Lines 140-156: What are the precision and accuracy estimates for the pH optode? A lot of the discussion is based on looking at small changes in pH, but there is no mention for how accurate the pH optode is by itself.
Lines 181-186: Why was the two point calibration conducted so far from the pH points of interest? All surface ocean pH will be around 8.1 (+/- 0.1 roughly). So why was one of the calibration points at pH 2-4 and 10-11? Also, how accurate is this calibration? The authors also mention making a pH offset adjustment with a CRM, though CRMs only give certified values for TA and DIC. The authors should give more details of this process.
Fig. 3b: Why did the raw pH measurements shift from being always below corrected values to always above corrected values?
Lines 259-261: The reader needs more information here about how pH was calculated from TA and DIC. You give more details later in the manuscript, but it needs to be mentioned when calculations are first introduced.
Line 277: The constants of Lueker et al. (2000) are most commonly used in “best-practice” calculations of open ocean conditions. I am curious by your choice to use Sulpis et al. (2020) instead, as this set of constants has been shown to perform worse than Lueker with regards to internal consistency analyses. I suspect your readers will have the same question. Can you please justify your choice or switch to using the Lueker constants?
Lines 280-287: For the derived pH parameters, which TA and DIC are you using? The underway measurements or the derived TA and calculated DIC (from pH and derived TA)? You need to be clearer. This process seems convoluted. You already used pH to calculate DIC and are now using DIC to calculate pH. I’m not sure this is a sound rationale. Also, your pHTA,fCO2 is calculated using fCO2 that was already calculated using pH. Instead, could you use inputs of measured underway pH and derived TA as your inputs, and then simply adjust your output conditions to be the temperature and salinity changes you are after?
Lines 305-323: Again, I am curious what the uncertainty of the actual optode is. This needs to be considered as an uncertainty contribution.
Lines 310-323: Have you considered that the uncertainty in pHTA,DIC calculations may not be random? There is a known pH-dependent pH offset (see Williams et al. 2017) where the error in calculated pH from TA and DIC is dependent on the pH. It seems you may be underestimating the uncertainty in pHTA,DIC.
Lines 333-344: This information is more introductory than methods.
Lines 402-418: There are a lot of technical details here about essentially what happens to a CO2 molecule in the span of seconds to minutes. However, the timescales the authors say they are working with is on the order of days. What is the relevance of these split-second reactions? The temperature is not changing fast enough to observe changes in the dissociation of the carbon species at this scale. You may see temperature changes between day and night, but not within seconds to minutes.
Lines 415: These waters don’t have a higher heat capacity, they are just insulated from the air.
Section 3.1: This section is quite difficult to follow. The authors are trying to discuss differences in pH between two basins, from day to night cycles, and throughout cycles 1-n during a cruise. Perhaps the authors should choose to discuss these types of comparisons individually, instead of all together. I think the more interesting results are between basins and between day and night. The specific cycles themselves are a bit too in the weeds.
Lines 452-464: This is again a lot of details about instantaneous thermodynamics. These processes are occurring on much quicker timescales than days, so I am not sure of its relevance to this manuscript. For open ocean pH, with all the uncertainties of measurements and changing dynamics, there is no way to observe these fine scale processes. I suggest the discussions of thermodynamics be made more concise or removed.
Section 3.2: Can the authors point to any studies discussing the generally expected levels of biological activities in these open ocean regions? They mention strong biological day-night signals, but would you actually expect to see much change in an open ocean area?
Lines 578-584 (and throughout results): There is a lot of background information that would be better suited for the introduction.
Lines 582-584: Now the authors are discussing time frames from the LGM, which is significantly longer than anything discussed in this paper. This seems irrelevant.
Technical Corrections
Line 19: Is it Dec 2019-Jan 2020? List both years since they are different.
Lines 32-35: These two sentences say the same thing, and could be combined for conciseness.
Line 40-41: “High resolution studies” of what exactly?
Line 87: To be clear, increased atmospheric CO2 only boosts oceanic CO2 uptake if the atmospheric CO2 > oceanic CO2. If atmospheric CO2 increases in a region, but is still less than oceanic CO2, the flux will still be towards the atmosphere.
Line 96: Might be helpful to include a range of latitudes for the “average North Atlantic latitude” since you mention a separate region above 55 N later in the sentence. Also, since you first mention N Atlantic has longer equilibration times, you should list the 18 month time-frame first for sentence structure.
Line 101: Surface temperatures?
Line 117: Give date ranges for both cruises for consistency.
Line 118: You need to define pH total (which should be denoted as pHT) before including in the figure caption. You should also make it clear you are measuring pHT throughout the manuscript.
Line 122: Cruises datasets should be cited or give appropriate links to the available data. Do the cruises have cruise reports you could link to?
Line 134: Again, listing of dates needs to be consistent for both cruises.
Line 135: “Discrete carbonate chemistry and nutrient samples” – you should list the parameters like you did for the first cruise in line 128 OR mention they collected the same samples as the cruise above. In general, try to make sure you have parallel structure in your sentences.
Line 141: Here is your first mention of pH on the total scale – you need to define what this is.
Fig. 2: What is the scale on the top and right sides of Fig. 3b?
Line 346: Missing end parentheses.
Fig. 4: The mean pH lines seem unnecessary, and they make it harder to see the other expected pH lines in the figures.
Fig. 4-5: You need to define what observed pH is (you do this in Fig. 6 but it needs to be when first used). Is this the underway pH corrected using the discrete TA and DIC data?
Fig. 6: You need to define what ∆pH is.
Line 378: Again, is pHobs the corrected underway pH measurements? Try to make this clearer.
Line 467: Are you referring to Fig. 8?
Citation: https://doi.org/10.5194/egusphere-2024-2853-RC1
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
266 | 64 | 32 | 362 | 45 | 9 | 8 |
- HTML: 266
- PDF: 64
- XML: 32
- Total: 362
- Supplement: 45
- BibTeX: 9
- EndNote: 8
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1