Interactive comment on “A large CO2 sink enhanced by
Contents
1. Comment 4 b P values alone e g 4685 1 do not provide any information need to know which test Reply Change4 Information on the statistical test for each p value are now provided in the MS Comment 5 c pCO2 correlated with DO 4683 26 need to know which correlation Beware that standard regression is inappropriate as both variables are subject to er rors In such cases one should use type 2 or geometric regression Reply 5 In all C3580 BGD 12 C3579 C3598 2015 Interactive Comment BY experimental science variables are subject to errors What we want to discuss here is that because of respiration photosynthesis pCO2 was minimum when DO was maxi mum Nothing more than this qualitative statement for instance we don t use the value of the slope of the regression in a quantitative way in order to calculate some kind of photosynthetic respiratory quotients Change 5 We used Spearman correlation and specified this in the text Comment 6 d Trends that mirrored 4672 9 need a stats background Reply 6 Same reply as for comment 5 see also figure R1 at the end of this comment very classical trend in estuaries Borges and Abril 2011 Change 6 No change in the revised MS Comment 7 e Authors used a PCA in section 4 2 that has not been introduced be fore Moreover PCA is a reduction technique it does not show any correlation maybe need to use a Multiple Linear Regression to show that Repl2. We also verify the drift before and after each sampling cam paign The measurements before and after each sampling campaign were consistent at precision level of about 3 ppmv The excellent agreement between the verifica tions before and after each sampling campaign and the excellent agreement between the standards and the equipment shows that the method is robust and the pCO2 span is little The precision and the accuracy of the pC02 measurements were about 3 and 5 ppmv respectively The precision of the pH measurements was about 0 01 after 7 verifications against NBS standards We performed a three point calibration pH 4 01 pH 7 00 and pH 10 01 As we have overdetermined the carbonate system pCO2 pH and TA and we have chosen to used direct pC02 measurements and DIC calculated from pCO2 and TA we use pH measurements only for quality check Change 24 This is now specified in the materials and methods Comment 25 2 3 3 It is nice to try a small intercomparison between pairs but it need more quantifications are the slopes statistically different from 1 Are the intercept statistically different from 0 How are the errors propagated to calculate DIC Is this 6 5 umol kg 1 between calculated from each pair Reply 25 With the biogeochemical analysis we make in this paper we do not use DIC values to perform any budget calculation We only use pCO2 values for the CO2 budget Our paper is quiet long enough so we find such detailed DIC qu
3. another example on how stats can help cluster or any other grouping technique to show this difference between areas Reply 29 we con sider such cluster analysis would require a long description and would unnecessarily increase the length of the MS Comment 30 4682 10 you haven t defined SD before Reply Change 29 we define SD in the revised MS Comment 31 4682 15 needs stats to highly or moderate associate something Reply 31 Same reply as for comment 5 Comment 32 4682 20 21 what do outflow and inflow means here Reply Change 32 We removed the words inflow and outflow in this context Comment 33 4682 28 well representative is hard to believe on the light of such high diurnal changes Were all done at the same time of the day To avoid these doubts it could be nice to have instead of one particular profile an average profile with mean and error bars Reply Change 33 These figures represent examples of depth profiles of Guanabara Bay in summer and winter conditions following the main channel of the Bay The profiles were shown to illustrate the spatial and temporal differences in the vertical column water structure corroborating with Kjerfve et al 1997 results and also the diurnal enhancement of surface temperature in the upper parts of the bay These profile are not supposed to be representative for any average conditions they are used here to illustrate the dynamics of stratification in t
4. debate about influence on filter no filter sampling for TA and DIC Could you discuss some of this in your method Did you test for this influence Reply Change 23 We measured TA on filtrated water so we removed acid neutralizing particles We did not test the difference between filtrated and non filtrated We calculated DIC from pCO2 and TA Comment 24 Which are pH and pCO2 accuracy precision How are the number of the verification 4679 26 How often did you calibrate the sensors pCO2 span for long range is the response still linear How pCO2 data fits with SOCAT standards Is there any plan to submit the dataset somewhere Biogeosciences strongly promotes the full availabil ity of the data sets reported in the papers that it publishes in order to facilitate future data comparison and compilation as well as meta analysis This can be achieved by uploading the data sets in an existing database and providing the link s in the paper Reply 24 We used 3 gas mixture standards 410 1007 and 5035 in ppmv to calibrate C3585 BGD 12 C3579 C3598 2015 Interactive Comment BY the LICOR before each sampling We used N2 passing through fresh soda lime to set the zero and we used the standard at 1007 ppmv to set the span We used the second and third standards at 410 and 5035 ppmv to check linearity The number of verifications after each calibration was about 7 the licor signal was stable and linear in the range of calibration
5. on explaining this caveat You are claiming diurnal variation is important but you are still missing part of the day when even less data is available Reply Change 45 The night sampling was not proper performed mainly due to a problem of security We heard about some problems with boat assaults in the upper parts of the bay and related to drugs traffic and we thought that anchor our boat in a point 24 hours maybe was not a good idea Then we thought about the sampling with boat in movement from dawn to dusk The sampling design was within our possibilities Indeed nighttime was less sampled than daytime However we believe that logically the early morning predawn corresponds to the maximum surface water pCO2 because of the long night period before We also have a quiet precise sampling between 5 00 and midday so we could observe rapid pCO2 decrease around 9 9 30 and then pCO2 becomes quiet stable at low values during the whole day Then we have used multiple night end members except for S2 to diurnally integrate the fluxes Comment 46 4685 23 26 strange phrase please rewrite more clear Reply Change 46 We modified as suggested Comment 47 4685 26 4686 4 really long phrase for an important info Also it will be nice to see this graph and have more quantification as it s an important result also present in the abstract Reply Change 47 We rephrased this part in the text Also we attached this graph figure R1 in the final o
6. read this paragraph first it will be easier to under stand the whole study area section Right now it is difficult to know where the wide entrance which is the inner regions Reply Change 20 figure 1 was improved as recommended so it is location is easier now Comment 21 4678 10 caution here and everywhere else where you talk about areas C3584 BGD 12 C3579 C3598 2015 Interactive Comment BY For example in 4684 10 75 of surface area of the Bay is wrong is 75 of the sampled surface area of the Bay 67 of the total area Another example is 4697 9 11 the polluted sampled area is only 10 but the no sample 10 might be also polluted due to its location and that will make 20 of the bay Modify Table 2 and Figure 6 a accordingly all entire sampled bay Reply Change 21 We agree that 75 and 10 are related to the sampled area We performed the modifications in the considered section as well as in table 2 and figure 6 a Comment 22 4678 29 is this probe also calibrated Is the same as in 4679 16 Do you use discrete sample to calibrate for DO temperature salinity or only to converse chl a Reply Change 21 We calibrated the DO probe against saturated air The temperature was not calibrated we used 2 thermometers with very consistent results at 0 01 C one from the probe and from the pHmeter The salinity probe was calibrated against certified material Comment 23 There is an ongoing
7. 2 C3579 C3598 2015 Interactive Comment BY R2 0 994 Slope 1 008 0 006 slope significantly different from 0 p lt 0 0001 The slopes are not statistically different from 1 p 0 20 and the intercepts are not signifi BGD cantly different from 0 p 0 86 The method used is one equivalent to an Analysis of 12 C3579 C3598 2015 Covariance ANCOVA according to the GraphPad Guide User Manual 6 0 Interactive comment on Biogeosciences Discuss 12 4671 2015 Interactive Comment C3594 a n aa sg Im Fu 0 1000 2000 3000 4000 CO ppmv Fig 1 see caption in response text C3595 BGD 12 C3579 C3598 2015 Interactive Comment PAR Accum Prec7 Wind DO Chla pCO Salinity Temp 0 11 0 83 0 87 0 87 0 83 0 02 0 68 PAR Accum 0 11 0 29 047 0 43 0 46 0 55 0 27 Prec 7 _0 83 0 29 0 88 0 83 0 91 0 08 0 66 Wind po 0874 047 0 88 0 76 0 93 0 36 0 76 0 87 0 43 0 83 0 76 0 85 0 06 0 60 Chla 0 83 0 46 0 91 0 93 0 85 0 38 0 86 pCO2 0 02 0 55 0 08 0 36 0 06 0 38 0 43 Salinity 0 68 027 0 66 0 76 0 60 0 86 0 43 Temp Correlations significant at p lt 0 05 Correlations significant at p lt 0 01 Fig 2 see caption in response text C3596 BGD 12 C3579 C3598 2015 Interactive Comment Inputs of carbon to Guanabara Bay mmol C m d Comment CO air water
8. Biogeosciences Discuss 12 C3579 C3598 2015 www biogeosciences discuss net 12 C3579 2015 Biogeosciences Author s 2015 This work is distributed under Discussions the Creative Commons Attribute 3 0 License ssa99y uado Interactive comment on A large CO sink enhanced by eutrophication in a tropical coastal embayment Guanabara Bay Rio de Janeiro Brazil by L C Cotovicz Jr et al L C Cotovicz Jr et al lecjunior id uff br Received and published 15 July 2015 Comment on A large CO2 sink enhanced by eutrophication in a tropical coastal em bayment Guanabara Bay Rio de Janeiro Brazil by L C Cotovicz Jr et al Comment 1 General comments The present paper under review for biogeosciences describes the surface water pCO2 and ancillary parameters in a semi enclosed estuar ine embayment The paper gives interesting results adding more studies of the carbon cycle in coastal area and recognize the sampling strategy effort done by the authors It was also been really quick from the last sampling period to the submission so con gratulate the first author for this effort There are not mistakes in the bibliography list C3579 BGD 12 C3579 C3598 2015 Interactive Comment BY and spelling attesting to the close attention the authors paid to writing Reply1 Thank you very much for this general positive evaluation of our paper Comment 2 However have three major concerns about the cur
9. ality check secondary in comparison with other topics eutrophication stratification etc Nevertheless as requested we provide to the reviewer the information on the quality of our data that in Figure R4 and we added few sentence in the material and methods Change 25 We included this information on the DIC quality check in materials and methods Comment 26 Results 3 1 Are the sampling period statistically different than the C3586 BGD 12 C3579 C3598 2015 Interactive Comment BY climatology Without stats is difficult to compare for example it seems to me sum mer period is not warmer and dryer that the average from Fig 2 there are 2 months with more rain that climatological and 3 dryer same with temperature Reply 26 The temperature variation of the sampling period was not different from the climatological regime p gt 0 05 test t except for Dec 2013 and Jan 2014 p lt 0 001 test t Es pecially for these two months the period was significantly warmer than the average Changes 26 We corrected the part of the text with this consideration We included in the graphs the standard deviation bars Comment 27 The driest months is August 2013 otherwise state driest months dur ing summer period precipitation consistent with historical data is that true It seems Jul more rain and Aug really dry 4681 24 26 We don t have a table for sea sonal variation so it s difficult to foll
10. an nutrient from sewage influence the whole bay but not carbon Reply Change 59 Because carbon is mineralized and evaded to the atmosphere in the sewage network and in the high polluted channels This is stated in two places in MS as well as the discussion about eutrophication and the conclusion Tables Comment 60 1 Abbrevation of salinity Sal has not been used during the ms DO wrong Reply Change 60 Modified as suggested Comment 61 2 How can U10 be different if they came only from 2 meteorological station please clarify Reply Change 61 U10 are different because they come from 2 different meteorological stations see material and method 4680 4 For the sectors 1 2 and 3 we used the data from meteorological station located at the Santos Dumont Airport whereas for sector 4 and 5 we used the data from the Galeao Airport located at the Governador Island Figure 1 Figures Comment 62 1 Dotted lines don t seem dotted in my file Missing South and West in lat long to be able to locate the map Reply Change 62 We change to the black line instead of dotted line We inserted South and West in lat long of the C3592 BGD 12 C3579 C3598 2015 Interactive Comment BY figure Comment 63 2 Dez Mensal Reply Change 63 We modified the figure and we modified the caption as suggested We also included the standard deviation bars Comment 64 3 Scales are different also in 5 Reply Change 64 the sca
11. average range of 0 7 m micro meso tidal Generally the tidal currents inside the bay are lower than 0 5 m s except near the bay entrance Kjerfve et al 1997 With an average water tidal renewal of about one month the tide tidal currents effect in exchanging with open waters are relatively small Change8 We mention the small impact of tidal exchange in the revised MS on the material and methods section Comment 10 b What is the influence of mangrove This is really interesting point of the view of Kon et al works Reply 10 Indeed mangrove surrounding waters generally show large pCO2 supersaturation eg Kone and Borges 2008 In Guanabara Bay C3582 BGD 12 C3579 C3598 2015 Interactive Comment BY the extension of mangrove forest is not so large and the volume of water exchanged with the mangrove sediments is probably moderate This is what our data suggest as we could not find supersaturated pCO2 conditions near of the mangrove 2 5 3 km This indicates little export low tidal probably associated with a rapid consumption of mangrove derived DIC by the phytoplankton Change9 we mentioned in the MS the low mangrove influence as suggested Comment 11 c Is there any other biological activity in the bay seagrass cultures macrophytes Reply 11 Primary production in Guanabara Bay is dominated by phy toplankton We not could find researches conducted in Guanabara Bay that accounts information about other b
12. f this review Furthermore the figure 07and related discussion in the MS also provides information about this relationship Comment 48 4686 5 more than 2000 in S4 during April 2014 is not slight Re C3590 BGD 12 C3579 C3598 2015 Interactive Comment BY ply Change 48 we removed slight Discussion Comment 49 4689 23 you haven t defined PCA yet not explain how you did Reply Change 49 We explained PCA in the material and methods section Please see the reply change 7 Comment 50 4690 13 of what explain Reply Change 50 We modified this part to and showed the higher influence over the factor 2 which explained 19 of data variance Comment 51 4690 19 395 units Reply Change 51 Units are inserted Comment 52 4690 28 discuss or take care of some organism are able to uptake bicarbonate by the enzyme carbonic anhydrase and by the proton pump mechanism Reply Change 52 we insert in the text a sentence about this process according to Kirk 2011 Kirk J T O Light and photosynthesis in aquatic ecosystems 3 ed Cambridge University Press New York 2011 Comment 53 4691 25 4692 16 move to another section of the methods entitle temperature and biological control in pCO2 calculations Clarify which means you are using each survey independently or winter summer as a whole Explain the ratio T B deltapCO2bio deltapCO2temp Reply Change 53 We moved this text to the method
13. flux 26 49 All bay average This study CO air water flux 33 102 Sectors 3 4 and 5 This study strong and permanent annual CO sink area Organic carbon load from 43 All bay average FEEMA 1998 majority of sewage organic carbon seems to be mineralized in sewage network River DIC DOC and TOC Undocumented inputs Internal Processes mmol C m d Comment NCP 51 225 143 Sectors 4 and 5 This study NPP 60 300 170 Sectors 2 3 and 5 Rebello et al 1988 Total Respiration Undocumented Degassing Burial Export mmol C m d Comment CO air water flux 54 177 Sector 2 This study permanent CO2 degassing in a restricted area Total organic carbon burial 27 114 Sectors 3 4 and 5 Carreira et al 2002 Monteiro et al 2011 DIC and TOC export to the Undocumented coastal area Annual average according to the k600 model parameterizations of Wanninkhof 1992 and Abril et al 2009 The lower value refers to the model of Wanninkhof 1992 whereas the higher value refers to the model of Abril et al 2009 Range and annual average in parenthesis Fig 3 see caption in response text C3597 BGD 12 C3579 C3598 2015 Interactive Comment SS 2000 1000 DIC calculated from TA pH 0 0 1000 2000 3000 DIC calculated from TA pCO Fig 4 see caption in response text C3598 BGD 12 C3579 C3598 2015 Interactive Comment
14. he bay Comment 34 4683 27 temporally and spatially correlation need to sets of correlation values Also defined what R2 is exactly N in table 1 is different that n here Please clarify unify Reply Change 34 We performed the Spearman correlation The r refers to the Spearman s rank correlation coefficient We corrected the N now the N in the text is the same with that on the table Comment 35 4684 2 unify the use of DO AOU or Saturation not defined and use C3588 BGD 12 C3579 C3598 2015 Interactive Comment BY the units accordanly right now is DO with not correct also in table 1 Values of 3750 ppmv are not showed in fig the scale didn t arrive so high Reply Change 35 As stated in the material and methods we report DO in of saturation The values in the scale were corrected Comment 36 4684 6 small and protected embayment means S4 S5 Re ply Change 36 No This is a small region of the sector 2 We deleted this part of the text to avoid confusion Comment 37 4684 15 here and elsewhere also in figures unify the date format Can you clarify rewrite brown red bloom understand what you mean but I m sure some biologist will have some concerns about that AYZ Reply Change 37 We unified the data format Biologist Bastiaan Knoppers has written this section and considers it is ok in its present form for such paper on carbon biogeochemistry Comment 38 4684 23 open waters m
15. hemical analysis See Table R8 in the final of this response Concerning water residence time we agree this affects the CO2 dynamics and air water exchanges However in the case of Guanabara Bay residence time is not always easy to estimate if one wants to differentiate the residence time in each sectors According to Kjerfve et al 1997 the average renewal time of 50 of the bay water volume is 11 4 days with important spatial variations water renewal increasing seaward So the average residence time 99 renewal of water in the whole bay is probably around one month Although we don t know the exact value of residence time in each sector we know that sectors S4 and S5 present the longest residence times compared to S1 which is connected to the ocean So residence times vary between few days to one week in sector 1 and one month or more in sectors 4 and 5 We think that this semi quantitative characterisation of residence time together with stratification still allows explaining most of CO2 spatial dynamics in Guanabara Bay Change8 we calculated NCP insert it in materials and methods results and discussion as well as in the new table Reference to residence time was made in a semi quantitative approach as described above Comment 9 3 Some question had when read the abstract and have not been answer are a Have the tide tidal currents an effect in exchanging with open water Reply 9 The tide in the bay is semidiurnal with an
16. ight be misleading talking inside the bay Reply Change 38 We changed to the bay waters Comment 39 4684 28 In the latter what Sampling period S1 Reply Change 39 In the latter referred to the back and forth track performed in S1 We rephrased this sentence Comment 40 4685 1 talking about night time for early morning seems contradiction Maybe use predawn Reply Change 40 We replaced nighttime to predawn Comment 41 4685 4 here and elsewhere please unify time threshold sometimes is 9 00 9 30 10 00 Reply Change 41 We unified the threshold as 9 30 AM Comment 42 4685 5 you didn t sample September 2014 Reply Change 42 We changed to Sep 2013 Comment 43 4685 9 a reference is missing here Reply Change 43 We modified this part to which apparently corresponded to the start hour of photosynthetic activity by phytoplankton It s an apparently hour and we don t have a reference for this C3589 BGD 12 C3579 C3598 2015 Interactive Comment BY specify Rebello et al 1988 for example used 10 00 AM as the start of phytoplankton activity Comment 44 Unify how you write night time if you are going to keep using it Re ply Change 44 We unified to nighttime because is a common term in other studies Comment 45 4685 20 diurnal variability know is logistically difficult or sometimes impossible to do a proper night sampling However would like to read something
17. iological activities like seagrass macrophytes etc The bay presented poorly coverage of consolidate substrates and does not provided ideal con ditions for biological settlement other than plankton Change 11 this is now mentioned in the study site section Comment 12 d What happens when the summer stratification break Reply 12 As illustrated in figure 3 summer thermal stratification is diurnal and is weaker at night due to convective cooling in contrast haline stratification maintains at night We never observed in the field occurrence of summer stratification break along the sampling campaigns Probably the haline stratification can be broken during strong storms which are not common for the region The sectors 4 and 5 that showed the most prominent stratification are located in a confined and more protected region of the Bay making the stratification break more difficult In general the stratification is weaker and driven by the strong PAR incidence thermal stratification Change12 No change in the MS related to this comment use page number line number or section number as in the friendly printed version of BDG to locate the comments Comment 13 Specific comments Abstract Here and elsewhere in the ms Why the use of pCO2 units Community normally use watm Reply Change 12 There is no C3583 BGD 12 C3579 C3598 2015 Interactive Comment BY definitive convention for pCO2 unit and ppmv or uatm ca
18. les in the figures are different to better shows the results and in the figure caption we took care to alert the reader about this Comment 65 4 Rainbow scales have been under debate not to use them In g it s really difficult to see where is super high region 2 Reply Change 65 We tested various color scales and used the best we found to show our data Comment 66 5 Letters are small difficult to read In the small inner graphs shadow when is night time Legend gt lt 9 30 is confusing better earlier later pre post dawn Reply Change 66 We increased the letters of the figure We shadowed the small inner graphs We modified to before 9 30 AM after 9 30 AM Comment 67 6 Unify data format and 9 00 or 9 30 Reply Change 67 We unified the data format to 9 30 Additional figure captions Figure R1 Relationship between pCO2 ppmv vs DO for the continuous mea surements of Guanabara Bay N 9002 Figure R2 Spearman correlation matrix for PAR umol m 2 s 1 accumulated precip itation of 7 days Accum Prec 7 mm wind velocity Wind cm s 1 dissolved oxygen DO chlorophyll a Chl a ug L 1 pCO2 ppmv salinity temperature Temp C and CH4 nmol L 1 in the Guanabara Bay The values were calculated with averages for each sampling campaign Figure R3 Summary of the documented carbon fluxes in the Guanabara Bay Figure R4 Linear Regression between DIC calculated from pH TA and pCO2 TA C3593 BGD 1
19. n be used For instance Nature journals impose the use of ppmv Comment 14 4672 21 25 important message in a too long and confusing phrase It s not clear how the embayment is classic in contrast of what Here and elsewhere emitter is a synonym of source Please stick to sink source to not confuse readers It is scientific writing not literature do not be afraid of repetition also in 4696 15 Reply Change 14 We have modified the text of this section in the abstract Also we replaced the term emitter to source Comment 15 Introduction 4673 25 no necessary comma before reference Re ply Change 15 Modified as suggested Comment 16 4673 28 you don t use GPP again so avoid excess of abbreviation Reply Change 16 Modified as suggested Comment 17 4674 2 you haven t defined what LOICZ stand for In one of its manual there s a nice example on how to calculate residence times Reply Change 17 We defined the LOICZ abbreviation Comment 18 Materials and methods 4676 11 moderate stratified in wintertime do not look like that in your winter profiles completely mixed Reply Change 18 We refer to completely mixed in the revised MS Comment 19 Locate Ilha do Governador Santos Dumont Geleao in Fig 1 Also add where the sea is it took me a while to understand the situation for non knowers of the area Reply Change 19 We added all this information in Fig 1 Comment 20 4677 24 if
20. ow last part I miss this table in other part of the ms i e 4686 8 to allow easy follow 4682 5 rainy season remain the reader when is that Reply Change27 We modified the text to The other sampled months had air temperature and precipitation consistent with historical data Fig 2 despite of some deviations from the historical average It s difficult to analyse the accumulated precipitation in this context because we didn t use the accumulated precipitation of the month instead we used the accumulated precipitation of 7 days before each sampling campaign Comment 28 4682 7 you haven t defined what s exactly eutrophic hypertrophic and which are the threefold between both classifications Readers might be familiars with eutrophic but hypertrophic is more rear Reply Change28 This classification was based on the classical Nixon s classification Nixon 1995 of 4 trophic states from oligotrophic lt 100 g Cm 2 yr 1 to hypertrophic gt 500 gCm 2 yr 1 according to the annual phytoplankton primary production In this way Guanabara Bay presented an overall classification of eutrophic whereas sectors 4 and 5 can reach the clas sification of hypertrophic because the primary production can be higher than 500 gCm 2yr 1 Rebello et al 1988 Change 27 We cite Nixon 1995 in the revised MS C3587 BGD 12 C3579 C3598 2015 Interactive Comment BY Comment 29 4682 6 9 here is
21. rent version of the manuscript 1 Statistical analysis need to be better explained and improved Re ply Change2 In the revised MS we have improved and explained in more details the statistical analyses as recommended We added a new section in the material and methods section called 2 4 Statistical Analysis However we did not follow all the numerous and detailed comments by the reviewer 1 when he she asked for more sta tistical analysis of our data Indeed in few cases in particular for more quantitative biogeochemical analysis in the discussion section we found that statistical analysis would not put additional value to our MS To the contrary as our MS was already quiet long we had to make some choices in our revisions In fact we found more impor tant to strengthen our biogeochemical analysis as asked by reviewer WJ Cai rather than adding more statistical analysis that we believed would add limited value to our MS the few Reviewer s comments that were not accounted for are all detailed below Nevertheless important effort has been made for the description of statistics and we recognize that these significant changes were necessary and greatly improved our MS Comment 3 a There is no section on the M amp M on how the stats have been carried on which program Reply Change3 We included a new section Statistical Analysis at the end of the Material and Methods which includes the stats description program etc
22. s sections as suggested We explained the ratio T B in the new text and the means were made for winter summer as a whole Comment 54 4693 26 how does the turbidity influence this same question in 4698 5 Reply Change 54 We did not fully understood this comment In Guanabara Bay water are clear and the phytoplankton biomass itself creates most of the turbidity Comment 55 4694 11 and elsewhere plankters plankton Reply Change 55 Modified as suggested Comment 56 4694 26 During this period space missing Reply Change 56 Modified as suggested C3591 BGD 12 C3579 C3598 2015 Interactive Comment BY Comment 57 4697 9 10 However this region occupies only about 10 of the surface area of the bay The other 10 you could not sample seems logical to be also polluted understand the logistic problem but that will be now 20 of the bay just discuss the possible influence of the no sampled area Reply Change 57 S2 represents only 10 of the sampled area If we consider the unsampled region of sector 2 it still represents less than 10 of the area of the bay because if we includes the unsampled area of this region in this context we also must include the other unsampled regions of the other sectors We inserted in the text small discussion about this no sampled area Comment 58 4697 15 at daytime at nighttime Reply Change 58 Modified as suggested Conclusions Comment 59 4698 20 How c
23. y 7 We agree with the reviewer that PCA does not show correlations but rather can shows patterns and find ing patterns reducing the dimensions of the dataset through linear combinations We used this analysis to make data easy to explore and visualize We agree that Spear man Correlation matrix could also have been used in order to investigate the physical and biological controls CO2 dynamics We did calculated this correlation matrix for the same the same variables of the PCA analysis See Table R2 in the response to review ers and we obtained the same conclusion as with the PCA In the interest of brevity we have chosen to keep the PCA analysis in Figure 8 of our MS but not to include the correlation matrix If reviewer 1 and or associated editor ask us to do so we can include in the final version of the MS Change7 we do not use the term correlation when we talk about the PCA Comment 8 2 The paper talks a lot about residence times and net ecosystem produc tion think the authors have everything they need to calculate both of these parame ters and will increase the robustness of this really good paper Reply8 Following the comment of both reviewers we have used our pCO2 diurnal data to calculate net com C3581 BGD 12 C3579 C3598 2015 Interactive Comment BY munity production NCP We also added a new table that compares various carbon fluxes in Guanabara Bay responding to the comment of W J Cai on biogeoc
Download Pdf Manuals
Related Search