Interactive comment on “ Intercomparison of carbonate chemistry measurements on a cruise in northwestern European shelf seas ”

The authors present an extensive carbonate chemistry dataset acquired during a cruise in the northwestern European shelf seas, during which the carbonate system was overdetermined by measurements of pCO2, pH, AT and CT. Overall, measurements compare surprisingly well with each other and the authors are to be congratulated for the high quality of the dataset. While I greatly appreciate the efforts to unravel systematic discrepancies between different methods to characterise marine carbonate chemistry and the explanations provided for the remaining discrepancies, the manuscript currently still has some major deficits that should be fixed before final publication in BG.

(1) My biggest criticism concerns the insufficient discussion (and citation) of previous work. Generally, rather few citations can be found in the discussion even though a lot of work has been done in this field. With respect to the intercomparison of different methods, several studies have provided similar levels of agreement between estimated carbonate chemistry parameters (e.g. McElligott et al. 1998, Luecker et al. 2000, so the conclusion that the "results show that it is possible to obtain good consistency between measurements" (P2810, L22-24) is neither surprising nor new. This impression can be avoided by referring to previous findings. Similarly, it remains unclear how much of the 'best practise suggestions' are novel. Please clarify which suggestions have been made before (and by whom).
(2) The authors seem be undecided if they consider the dataset to prove high consistency between the different datasets (e.g. P2794, L12 and P2808, L2-6) or to contain systematic discrepancies (e.g. P2794, L15 or P2808 L9-11). Some of my confusion between these two conclusions seems to arise from the differences between "raw" and corrected pCO2 values. Please distinguish between these two levels more clearly. Also, isn't it common practise to correct measured pCO2 levels for the differences between SST and temperature in the equilibrator? If so, why is the non-temperature corrected data discussed?
(3) In the abstract (P2794, L12-14) as well as the discussion, the authors state that the present dataset is "suitable to be used as a basis for evaluations of the impact of OA on ocean biogeochemistry" (P2828, L5-6). It remains unclear how the authors come to this conclusion. How does the dataset provide information on Ocean Acidification and its effects on marine biogeochemistry? On which timescales do you expect in-situ OA to be measurable, also in view of measurement uncertainties as well as temporal and spatial variability of carbonate chemistry? Is this study thought to serve as a baseline for future investigations? Please clarify.

Printer-friendly Version
Interactive Discussion Discussion Paper P2794-P2796: The introduction reads as a list of various related studies rather than an introduction to or a rationale for the presented manuscript. Please restructure/rewrite and put the different studies (more) in context with each other. P2794, L11-18: If diurnal cycles with differences as high as 41 µatm were observed, can you really state that the datasets "were all of really high quality"? What would be your definition of "really high quality"? P2794, L12-14 & P2796, L10-11: How can this study be used to the evaluation of OA impacts?
P2797, L17-19: Sentence sound a bit clumsy, maybe change to "The pCO2-1 and pCO2-2 systems undertook 6187 and 26671 measurements of surface water pCO2 during the cruise, respectively." P2798 L29-P2799, L3: This is a very long sentence, please consider rewording. P2801, L13-14: Please state the number of (CRM?) measurements used to gain these values. P2801, L24: Please state that this is the Matlab version. P2801, L24-29: Why were different borate constants, but not carbonate equilibria constants compared? P2803, L6-9: Please reword this sentence in order to improve grammar.
P2803, L24 -P2806, L7: It is not clearly stated which paragraph of this section refers to which level of correction of raw pCO2 measurements. Maybe it would help to change the order of the paragraphs, starting with raw data followed by corrected data. P2804, L-12-16: In addition to the average difference, the RMSE should be discussed here, as the average difference alone does not provide enough information on the performance in relation to "strong gradients of temperature, salinity and pCO2".

Interactive Comment
Full Screen / Esc

Printer-friendly Version
Interactive Discussion Discussion Paper P2804, L-15-16: Why do you judge this results to be only "reasonable" (I would say it is pretty good!)? If the average difference in your study is comparable to the one described by Körtzinger et al. (2000), wouldn't that mean that the "less ideal" settings of your intercomparison do not seem to negatively affect overall consistency? P2804, L22-25: Please state average and maximum differences. P2805, L14ff: Is this temperature corrected data or not? Did you check for autocorrelations between PAR and temperature differences (I would expect that temperature offsets could be influenced by exposure of the equilibrator to bright sunlight)?
P2805, L27 -P2806, L1: But pCO2-2 was the one being closer to the values calculated from AT and CT (cf. P2804, L18). Please comment on this in the manuscript.
P2806, L1-L7: Are these novel suggestions? If not, please reference the statements appropriately.
P2807, L8-15: There are much more differences between both studies (e.g. measurement quality/overall uncertainty, number of samples, sample volume, scientific background of conductors, etc.) which could be discussed here.  Table 1: How can the calculated uncertainty be identical for the two pCO2measurements if you know that there are systematic differences between the two systems (light exposure, length of inlet system etc.)? Figure 2: Caption should read "... here defined as the residual of pCO2-1 (a) or pCO2-2 (b) and another measured or calculated..." Figure 3: Is this corrected data? If so, how do you end up with a RMSE of only 10 µatm? Caption should read "Comparison between the pCO2 between instruments 1 and 2 (µatm; white circles) and the phototosynthetically active radiance (PAR (W m−2), in black circles) measured over five days".
Figure 5: Given that you identify more problems with equilibrator 2, you should also show these plots for pCO2-2. Please add panel identifications (a-d). Caption should read "Box-and-whisker plots of the residuals between measured pCO2 (a), pH (b), CT (c), AT (d) and the respective estimates calculated from different pairs of measured variables (denoted on the x axis) for the two sets of borate constants. ..." Interactive comment on Biogeosciences Discuss., 11, 2793, 2014.