Dear Dr Hartman

many thanks for submitting a revised version of your manuscript with the title "Biogeochemical variations at the PAP-SO in the northeast Atlantic Ocean from weekly to inter-annual time scales", following its online discussion in "Biogeosciences Discussions".

The paper requires a number of minor revisions and technical corrections before I can accept it for publication. Please make these revisions and corrections and submit a new version of the manuscript. I will review the further revisions myself.

Kind regards
Jan Kaiser
Guest Editor Biogeosciences

1) Please add page numbers to the next version.
2) Where have the SOO data been deposited? Are they publicly accessible?
3) The author affiliations should use the same format in each case - at the minimum, include the city and post code; the street address can be omitted.
4) Figures 2, 3 and 5: The presentation of the data as two separate time series is confusing because it suggests that the data refer to simultaneous or comparable time periods. Please use a single time-axis for each panel, starting in May 2003 and ending in June 2012, with a break from April 2005 to April 2010. Please also add an x-axis and x-axis ticks for the tick labels. The December data point in the NAO time series does not line up with the "Dec" label. Please indicate whether the x-axis ticks (to be added) refer to beginning of the corresponding month or the middle of the month. Please also check the location of the vertical blue lines - I guess they should represent the beginning of the year?
5) The "correction" of DIC concentrations for the CO2 air-sea flux has not been explained. Please add a corresponding paragraph. Do you use a constant or varying Revelle factor? How is the CO2 air-sea flux time series "mapped" onto the DIC concentration time series, e.g. to correct the DIC concentration change between time t1 and t2, do you use the average flux at t1 and t2, multiplied by (t2-t1)? If so, how is this correction partitioned between the concentrations at t1 and t2? And how were MLD variations over the period between t1 and t2 taken into account? Also, when the MLD was less than 30 m, presumably no correction was applied to the DIC concentration?
6) The p(CO2, air) calculations assumes a surface pressure of 1013.25 hPa. Is this reasonable? The global mean sea level pressure is 984 hPa, nearly 3 % lower. As CO2 is undersaturated, this would reduce the corresponding air-sea flux, but ideally the actual pressure should be used in the calculation and can be obtained, for example, from global reanalysis products (ECMWF, NCEP) if not available from the surface buoy observations.
7) Please define the 2003-2005 and 2010-2012 time periods. Neither of them span 3 complete years, so it's important to mention the start and end months.
8) p. 2, l. 9 (2/9): It is unclear why you don't mention the secular CO2 increase due to anthropogenic CO2 emissions here and instead refer to the similar temperatures etc. for both time periods.
9) 2/24: Delete "However" - there is no contradicting or contrasting statement.
10) 2/27: Please add "for the role of the ocean" after "now and".
11) 3/29: Change "from 30 m" to "at 30 m".
12) 4/18: "within deep chlorophyll maxima" or "within the deep chlorophyll maximum".
13) 5/1: "Note that a measurement error of an early version"
14) 5/2: "fluctuation"
15) 5/5: "accelerate the equilibration" or "accelerate attaining equilibrium"
16) 5/6: Please add a brief explanation why the time series was interrupted.
17) 5/11: "checks"
18) 5/16: "chlorophyll a concentrations"
19) 5/17: "precision for fluorescence measured by these fluorometers is"
20) 5/19: "approximation of the chlorophyll a concentration. The fluorescence/chlorophyll a concentration ratio"
21) 6/16: Please add the microcat data to the revised time series (item 4). This is an interesting and useful comparison.
22) 6/28: "value" or "average of the p(CO2) values"
23) 7/8: Please format this equation as follows:
F(CO2) = k K_0 [p(CO2, sea) - p(CO2, air)
Only F, k, K_0 and p should be in italics.
24) 7/12: p(CO2, sea) and p(CO2, air)
25) 7/14: "averaged CO2 dry mole fractions measured" - please also see item 6 above.
26) 7/18: "Total alkalinity was calculated ..."
27) 7/19: Does this uncertainty include the 1 % temperature and 2 % salinity uncertainties due to the use of the Argo data (p. 6)?
28) 7/23: The "K"s should all be in italics because they are physical quantities.
29) 7/24: "calculate DIC concentrations introduces an uncertainty of ±6.4 µmol kg-1" [or more? - see item 27]
30) 7/26: "monthly mean MLD"
31) 9/1: "The increase in chlorophyll a fluorescence during the 2011 spring blook was also larger"
32) 9/13: Replace "an opposite trend" with "opposite seasonal variations"
33) 9/31: See item 5 above.
34) 10/4-15: These paragraphs need to come before the use of air-sea flux corrected DIC concentrations.
35) 10/14: Please enclose (5.7±2.8) and (5.0±2.2) in brackets (so that the unit applies to the mean and the uncertainty).
36) 10/14: The start (2003: April, 2010: May) and end months (2005: Mar, 2012: June) of both periods differ. Please discuss the bias caused by these differences. Do the flux calculations include the uncertainties due to a) p(CO2) calibration, b) the temperature and salinity-substitution from Argo floats, c) uncertainty in wind speed (due to the use of the 1º x 1 º FNMOC data), d) uncertainty in sea level pressure, e) uncertainties in the gas exchange coefficient. At least the first four uncertainties (a to d) should be quantifiable.
37) 10/26: "inversely" suggests a "1/temperature" relationship - I guess this should be "seasonal variations opposite to temperature"?
38) 10/28: Delete "the" before "convective".
39) 11/22: Delete "the" before "p(CO2)".
40) 11/24: The reference refers to tropical cyclones and has no immediate relevance to the PAP site. Please use a more suitable reference.
41) 12/4: The missing data have been substituted by SOO data in Figure 2, so I suggest you use these to fill in the gap.
42) 12/8: This suggests an average over 3 complete years in each case, but see item 36 regarding start and end months.
43) 12/21: 2010-2012
44) 13/5: Please add brackets around (mean±uncertainty).
45) 13/7: "p(CO2)
46) 13/8: 2010-2012
47) 13/13: "at PAP-SO"
48) 13/22: "require" and "At PAP-SO"
49) 14/3: "effect of storms"
50) Figure 1: Legend (colour bar) and axis labels are missing. Please add white contour lines and contour labels to the bathymetry. Please use a white symbol for the PAP site. The quality of the figure is rather poor (jpeg?) - please use PDF, EPS or another vector graphic format instead.
51) Figure 2: y axis labels should be "p(CO2)", "chlorophyll a concentration" and "nitrate concentration" (or "NO3- concentration"). Does weekly nitrate concentration refer to "weekly average" or "measured once on a weekly basis"? Please clarify.
52) Figure 3: Please cross-refer to the corresponding section in the text where you explain how DIC was calculated (based on p(CO2) and salinity-based TA parameterisations).
53) Figure 4: Please add "concentration" to x and y axis labels. "Jan" in the legend should have a capital "J".

Dear Dr Hartman

many thanks for submitting your revised manuscript, which I am happy to accept for publication in Biogeosciences.

Thank you for checking the mean sea level pressure at the PAP site. The global mean value of 984 hPa I referred to be is based on Trenberth & Guillemot (1984) - sorry, I didn't mention the full reference in my previous comments.

Trenberth K. E. and Guillemot C. J. (1994) The total mass of the atmosphere. Journal of Geophysical Research 99: 23079-23088 doi:10.1029/94JD02043

Choosing 1 atm as reference pressure appears to be reasonably close to the actual pressure at the PAP site so as not to cause a significant bias over a period of months (except for aliasing effects). Over shorter time periods, this may require correction. For the purposes of the manuscript, could you please (as a technical correction) give the actual sea level pressure (±1 standard deviation) at an appropriate place in the manuscript, ideally for the period of interest rather than 2014? Where the pressure data are not available from surface buoy measurements, you could use one of several global reanalysis products (ERA-Interim, NCEP).

Best wishes
Jan Kaiser


Note:
I acknowledge that axis labels in a number of a already published manuscripts and some textbooks are not as clear as they could be. I'll leave it up to you whether you would like to adopt best practice as documented for example in the IUPAC Green Book, p. 3 (http://media.iupac.org/publications/books/gbook/IUPAC-GB3-2ndPrinting-Online-22apr2011.pdf), and also in the BIPM SI Brochure, the IUPAP Red Book, and ISO literature.

Just for information with respect to the uncertainty notation, I also refer to the IUPAC Green Book, p. 154, which shows that both the mean value and its uncertainty are enclosed in brackets, preceding the unit.

IUPAC Green Book:
Cohen et al. (Eds.) (2008), International Union for Pure and Applied Chemistry, Physical Chemistry Division: Quantities Units and Symbols in Physical Chemistry, 3rd, 2nd printing ed., Royal Society of Chemistry, Cambridge.