High-frequency, year-round time series of the carbonate chemistry in a high-Arctic fjord (Svalbard)
Abstract. The Arctic Ocean is subject to high rates of ocean warming and acidification, with critical implications for marine organisms as well as ecosystems and the services they provide. Carbonate system data in the Arctic realm are spotty in space and time and, until recently, there was no time-series station measuring the carbonate chemistry at high frequency in this region, particularly in coastal waters. We report here on the first high-frequency (1 h), multi-year (5 years) dataset of salinity, temperature, dissolved inorganic carbon, total alkalinity, CO2 partial pressure (pCO2) and pH at a coastal site (11 m) in a high-Arctic fjord (Kongsfjorden, Svalbard). We show that (1) the choice of formulations for calculating the dissociation constants of the carbonic acid remains unsettled for Arctic waters, (2) the water column is generally somewhat stratified despite the shallow depth, (3) the saturation state of calcium carbonate is subject to large seasonal changes but never reaches undersaturation (Ωa ranges between 1.4 and 3.0) and (4) pCO2 is lower than atmospheric CO2 at all seasons, making this site a sink for atmospheric CO2 (16.8 mol CO2 m−2 yr−1).
Jean-Pierre Gattuso et al.
Status: final response (author comments only)
- RC1: 'Comment on essd-2023-92', Leif Anderson, 05 Apr 2023
- RC2: 'Comment on essd-2023-92', Yuanxu Dong, 27 Apr 2023
Jean-Pierre Gattuso et al.
High-frequency, year-round time series of the carbonate chemistry in a high-Arctic fjord (Svalbard) (0.9) https://doi.org/10.5281/zenodo.7714954
Jean-Pierre Gattuso et al.
Viewed (geographical distribution)
This manuscript presents a very valuable high-frequency data set of carbon system relevant parameters covering several years in the surface water of an Arctic fjord system. Never before has it been possible to observe the evolution of climate relevant parameters as pCO2 and saturation state of aragonite in all seasons with this high time resolution. These data will set a very useful reference point for other studies of the carbon system in the Arctic Ocean.
Hence, well deserve to be published, but can be improved by making some minor changes as specified below.
Section 2.4. The SeaFET sensor is pressure sensitive and it therefore valuable to give information on how long the profiling system was kept at the depth before recording. On line 2018 it is given as 24 hours but this information would be valuable also here.
Line 106. The uncertainties given, I guess, is a result of analytical imprecision of the input parameters, but no information of these imprecisions are given in 2.2. Please do that, and also give the accuracy, which is as important for the computation of the other C-parameters. The same should be done for the discreet pH measurements.
Section on lines 173-175. I cannot follow this text. From the figures 3 & 4 as well as tables 3&4 I only see marginal differences in the results when using Lueker et al (2000) and Papadimitriou et al. (2018). The authors need to better describe what they mean.
Fig. 6. Add the orange lines as noted in the legend. Also specify what the red dots in C, D and E are. I guess that for D and E it is the measured values in water samples and thus the blue dots in E must be computed; from salinity? For pCO2 the situation must be different as it was only measured by the ferry box. Please give information on this.
In line 207 it reads “salinity below 8 m ..” while it in line 211 reads “Temperature at 11 m ..”. I hope that all the high time resolution data, except that of pH, is from the ferry box. If not any comparison is prone to uncertainties in water masses variability. Please specify the depth of sampling in detail.
The first paragraph of 3.9 needs to be looked over. The data that is not available in May to July are pH, not the ferry box pCO2. Hence that information needs no be after how the pCO2 vertical profile is computed. Then it finishes off with a comment that temperature is partly driving the vertical gradient. However, Fig. 7 show that non-thermal drivers exert a greater control than temperature.
Paragraph staring on line 240. I have difficulties with the signs here. First in the paragraph above it reads that the 11 m pCO2 overestimate the surface water values by 17 uatm (clear from Fig. 10), but in this paragraph it reads “correcting for the underestimation of 17 uatm ..”. This in combination with the first presented air-sea flux of 20 mol/(m2yr) and the second -17 mol/(m2yr) does not make sense. To control if the pCO2 estimated from pH and salinity derived AT at 0-4 m depth is comparable to the measured it would be nice to see how the pCO2 estimated from pH and salinity derived AT at 11 m compare to the measured.
Abstract: Not all the parameters mentioned are determined every hour.
Line 5 of abstract. Specify that 11 m is the bottom/sampling depth.
Line 19. Spell out what fastest and largest changes the Arctic Ocean exhibit.
Line 25-26. Delete the first “projected” in the text “The projected decrease in pH is projected to be larger in ....”.
Fig 1. Add (C) to legend and change a, b, c and d to 1, 2, 3 and 4.
Line 63. Set the number of outliers in relation to the total number of determinations.
Table 1. Use the letter µ instead of mu for pCO2, as in the text.
Line 155. Fig. 3 should be Fig. 4.
Fig. 4. Wrong figure has been posted, is the same as Fig.3. Legend. One AT should be CT.
Table 3. Give information of what Q1 and Q3 stands for.
Line 2001. Add C after Fig. 5, and in next line insert 5 instead of ?? after Fig.
Line 210 mention that numerous streams add freshwater in June to August. Another important source is melting sea ice and calving glaciers that add freshwater to the fjord system.
In line 220 it reads that pH information should be given in Fig. 8, but it is not. It would have been nice to see that but if the data do not allow then delete pH here.
Fig 8. It states that the density gradient is given in C. But it cannot be, but maybe sigma. Please clarify.
Line 250. Change “Its it ranges..”