Diatoms and their inﬂuence on the biologically mediated uptake … · 2016-01-12 · diatom blooms in a silicon-enriched near shore upwelling system off Oman. In the open western

Biogeosciences 3 1ndash13 2006wwwbiogeosciencesnetbg31SRef-ID 1726-4189bg2006-3-1European Geosciences Union

Biogeosciences

Diatoms and their influence on the biologically mediated uptake ofatmospheric CO2 in the Arabian Sea upwelling system

T Rixen1 C Goyet2 and V Ittekkot 1

1Center for Tropical Marine Ecology Fahrenheitstr 6 28 359 Bremen Germany2Universite de Perpignan Bat B EA 1947 BDSI 52 avenue Paul ALDUY 66 860 Perpignan France

Received 17 December 2004 ndash Published in Biogeosciences Discussions 25 January 2005Revised 4 May 2005 ndash Accepted 6 December 2005 ndash Published 9 January 2006

Abstract Sediment trap experiments have been carried outin order to study processes controlling shifts from diatom tonon-diatom dominated systems in the western Arabian SeaOne of our major problems was to link sediment trap recordsto surface ocean processes Satellite-derived observations onupper ocean parameters were helpful to reduce this prob-lem in the past and gain a new quality by combining it withresults obtained during the Joint Global Ocean Flux Study(JGOFS) in the Arabian Sea The new results imply that in-tense grazing can decline or impede the development of largediatom blooms in a silicon-enriched near shore upwellingsystem off Oman In the open western Arabian Sea diatomblooms recover within the offshore advecting upwelled wa-ter and lead to peak organic fluxes into the deep sea but onlyduring the later phase of the upwelling season During on-set of the upwelling season grazing favoured by eolian ironinputs causing the formation of thinner diatom shells seemsto prevent the development of a large diatom bloom withinthe silicon-enriched offshore advecting upwelled water Anincreased relevance of diatoms and diatom-grazing copepodsin the planktonic community as well as oligotrophic condi-tions seem to raise the ratio between organic carbon forma-tion and calcium carbonate carbon precipitation (rain ratio)in the surface water The decomposition of organic matterin the water column reduces the rain ratio within in the sink-ing matter especially in the oligotrophic region dominated bycyanobacteria and reduces also the variation of the carbon tonutrient uptake ratios seen in the surface water

1 Introduction

Marine organisms influence oceanatmospheric CO2 ex-change by photosynthesising organic matter and by precip-itating carbonate shells The subsequent export of organic

Correspondence toT Rixen(trixenuni-bremende)

carbon and calcium carbonate into the deep sea and the re-sulting net effect on the atmospheric CO2 concentration isdefined as the biological pump (Volk and Hoffert 1985) Thebiological pump is driven by nutrients such as phosphate andinorganic nitrogen (N = nitrate + nitrite + ammonia) whichare in addition to carbon required to build up organic mat-ter (Heinze et al 1991 Maier-Reimer et al 1996 Tyrrell1999 Hedges et al 2002) The efficiency of the biolog-ical pump can be enhanced by raising the uptake ratio ofcarbon to nutrients (CNP Redfield ratio) during the pro-duction of organic matter and by a reduction of the calciumcarbonate precipitation because the latter process increasesthe CO2 concentration in sea water (Redfield et al 1963Berger and Keir 1984 Heinze et al 1991) Depending onthe species and on environmental conditions the CP ratioof marine organisms varies between 40 andgt200 (Goldmanet al 1979 Burkhardt et al 1999 Geider and La Roche2002 Klausmeier et al 2004) Nevertheless despite suchvariations a global analysis suggests a mean Redfield ratioof 117plusmn1416plusmn11 (Anderson and Sarmiento 1994) whichis close to the constant Redfield ratio derived by a ldquoGeneralCirculation Modelrdquo (GCM 122161 Maier-Reimer 1996)and to those found by a basin-wide analysis in the ArabianSea (125141 Millero et al 1998)

Carbonate production is often parameterised by apply-ing a fixed ratio between the organic carbon production andthe precipitation of calcium carbonate carbon (rain ratioSarmiento et al 2002) Data-based estimates of rain ratiosare scarce and deviate between 33 and 125 (Sarmiento etal 2002 and references therein) A new estimate even sug-gests a mean global rain ratio of 16 (Sarmiento et al 2002)All these estimates fall within the range of global mean rainratios (23 to 266) which can be obtained by dividing theglobal mean organic carbon export derived from satellite data(2 to 16 1015 g C yrminus1 Falkowski et al 2000 Rixen et al2002) by estimates of the global mean carbonate carbon ex-port (Table 1)

copy 2006 Author(s) This work is licensed under a Creative Commons License

2 T Rixen et al Diatom and their role in the Arabian Sea upwelling system

Table 1 Contribution of carbonate-producing organism to the pelagic marine carbonate production

1015g C References

Total carbonate production 100 060ndash086 Milliman and Droxler (1996)Foraminifera 23ndash56 014ndash048 Schiebel (2002)Pteropods sim10 006ndash009 Schiebel (2002)Coccolithophorids 4ndash38 002ndash033 Schiebel (2002)

and references therein

Possible effects of changing Redfield and rain ratios onthe atmospheric CO2 concentration were quantified usingGCMs It was estimated that an increase of the CP ratioby 30 (1221 to 15861) could lower the atmospheric CO2concentration bysim72 ppm (Heinze et al 1991) A doublingof the global mean rain ratio from 4 to 8 could reduce theatmospheric CO2 concentration by 285 ppm (Heinze et al1991) and raising the rain ratio from 5 to 166 could how-ever reduce the atmospheric CO2 concentration by 70 ppm(Archer et al 2000) These changes explaining a major pro-portion of the glacialinterglacial variation of the atmosphericCO2 concentration were achieved by assuming that diatomsoutcompete the carbonate-producing coccolithophorids anddrive the export production Due to such a competition therain ratio is directly linked to biogenic opal production insome models used to study the feedback impact of the car-bonate production in the ocean on increasing CO2 concen-trations in the atmosphere (Heinze 2004)

Since diatom growth is believed to be limited by the avail-ability of silicon at lower latitudes (Dugdale and Wilkerson1998 Rixen et al 2000) changes of the global silicon cy-cle and re-organisation of the marine silicon cycle were pro-posed as being possible mechanisms to fertilise lower lati-tudes with silicon during glacial times (Froelich et al 1992Harrison 2000 Conley 2002 Ridgwell et al 2002) There-organisation of the marine silicon cycle is suggested to betriggered by an enhanced eolian iron input lowering the SiNuptake ratio of diatoms in the Southern Ocean and subse-quently enhancing the silicon export from higher to lowerlatitudes (Matsumoto and Sarmiento 2002)

In addition to coccolithophorids competing with di-atoms there are also carbonate-producing heterotrophs likeforaminifera and pteropods which feed on diatoms In theArabian Sea sediment trap experiments showed that coc-colithophorids contribute onlylt15 to the carbonate ex-port into the deep sea and that the peak flux of foraminiferainto the deep sea coincides with that of diatoms during thehighly productive upwelling season (Haake et al 1993b1993a Zeltner 2000) This suggests strongly that effects offoraminifera and pteropods on the rain ratio should be takeninto consideration as these organisms are important or eventhe main carbonate producer in the ocean (Table 1)

The total fluxes of calcium carbonate and organic car-bon measured by deep moored sediment traps in the Ara-bian Sea (Lee et al 1998 Honjo et al 1999 Rixen et al2002) and ratios between calcium carbonate dissolution andorganic carbon remineralization in the water column (Hupeand Karstensen 2000) have been used to calculate rain ratios(Rixen et al 2005) In this study these calculated rain ratiosranging between 2 and 35 will be compared with rain ratiosderived form total dissolved inorganic carbon concentrations(DIC) total alkalinity (TA) and CO2 partial pressure differ-ences (1pCO2) between the atmosphere and surface water(Goyet et al 1998b 1999 Millero et al 1998) Furthermoreplankton counts (Garrison et al 2000 Schiebel et al 2004)will be evaluated in conjunction with nutrient (Morrison etal 1998) and iron concentrations (Measures and Vink 1999)in order to study factors influencing shifts in the planktoniccommunity structure and associated changes of the composi-tion of sinking particles

2 Study area

The Arabian Sea is strongly influenced by the Asian mon-soon This climatic feature is driven by the sea-level pressuredifference between the Asian landmass and the Indian Ocean(Ramage 1971 1987) During the boreal winter the sea levelpressure over Asia exceeds that over the Indian Ocean due toa stronger cooling of the landmass Following the pressuregradient and deflected by the Coriolis force the wind blowsfrom the NE (NE monsoon) over the Arabian Sea This sit-uation reverses when the summer heating of the Asian land-mass leads to the formation of one of the strongest atmo-spheric lows on Earth This low attracts the SE trade windsand after crossing the equator the former SE winds blow asSW winds over the Arabian Sea due to associated changes ofthe Coriolis force The SW winds (SW monsoon) form a tro-pospheric jet (Findlater Jet) extending almost parallel to theArabian coast (Fig 1 Findlater 1977 Rixen et al 1996)The monsoon winds and the deserts surrounding the westernand northern parts of the Arabian Sea lead to dust inputs intothe Arabian Sea which are among the highest in the worldocean (Tegen and Fung 1994 1995)

Biogeosciences 3 1ndash13 2006 wwwbiogeosciencesnetbg31

T Rixen et al Diatom and their role in the Arabian Sea upwelling system 3

Wind speed [m s-1]

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

55˚ 60˚ 65˚ 70˚

10˚ 10˚

15˚ 15˚

20˚ 20˚

25˚ 25˚

Title Diatoms and their effect Rixen et al Figure 1

Fig 1 Mean SW monsoon wind speeds over the Arabian Sea dur-ing the SW monsoon Data are obtained from Rixen et al (1996)Red squares indicate the US JGOFS (Honjo et al 1999) and thered circles the long-term IndoGerman sediment trap sites in thewestern and central Arabian Sea (Rixen et al 2002) The white cir-cle indicate the central Arabian Sea surface mooring location (ONRDickey et al 1998) and the black circles show the US JOGFS wa-ter sampling sites (Morrison et al 1998)

Biological productivity within the Arabian Sea is deter-mined by the interplay between the euphotic zone and mixedlayer depth (MLD) whose deepening is caused by wintercooling and wind mixing (Rixen et al 2002) The inter-play between the euphotic zone and the MLD regulating theavailability of light and nutrients is well known from the tem-perate ocean In the Arabian Sea this leads to early and lateNE monsoon blooms During the SW monsoon the FindlaterJet creates one of the most productive upwelling areas in theocean (Antoine et al 1996) and a hot spot for CO2 emissionalong the Arabian coast (Kortzinger et al 1997 Goyet etal 1998b 1998a Sabine et al 2000) Diatom blooms arecommon during the later phases of the NE and SW monsoon(Haake et al 1993b Rixen et al 2000)

3 Data base methods and results

Nutrients DIC TA temperature and salinity profiles mea-sured at the sampling sites S1ndashS15 during the US JGOFScruises ttn49 (18 July 1995ndash13 August 1995) and ttn50 (14August 1995ndash13 September 1995) were obtained from theUS JGOFS database (Figs 1 2) The mixed layer depthwas defined as the depth at which a pronounced temperaturedecrease and nutrient increase occurred within profiles Sub-sequently all data were averaged for the mixed layer depth

Sea surface temperature [oC]

19 20 21 22 23 24 25 26 27 28 29 30 31 32

50˚ 60˚ 70˚

5˚ 5˚

10˚ 10˚

15˚ 15˚

20˚ 20˚

25˚ 25˚

20

21

22

23

24

25

26

27

28

29

0 300 600 900 120020

21

22

23

24

25

26

27

28

29

356

358

360

362

364

366

368

356

358

360

362

364

366

368S15S14S13S12S11S10S9 S7 S8 S6 S5 S4 S3 S2 S1 S1

Su

rfac

e te

mp

erat

ure

[oC

]

Su

rfac

e sa

linit

y

Distance to the coast [km]

Title Diatoms and their effect Rixen et al Figure 2Fig 2 Upper panel Mean satellite-derived sea surface temper-atures during the cruise ttn49 Data are obtained from PhysicalOceanography Distributed Active Archive Center at Jet PropulsionLaboratory California Institute of Technology) Lower panel Sur-face temperatures and salinity averaged for the depth of the mixedlayer (data are from the US JGOFS databasehttpusjgofswhoiedujgdirjgofsarabian) at each sampling site (S1 to S15)

(Table 2) The resulting mean mixed layer temperatures andsalinities increase generally from the coastal upwelling zonetowards the open ocean (Fig 2) During both cruises slightlyreduced temperatures and salinities occurred between sta-tions S5 and S7 approximately 500 km offshore As shownby satellite-derived SSTs charts (Fig 2) this anomaly wasassociated with the filament that extended almost parallelto the transect perpendicular to the coast towards the openArabian Sea Filaments are cold water structures caused byan accelerated advection of upwelled water While mov-ing offshore the upwelled water get mixed with the warmerand saltier surface water that was formed during the preced-ing oligotrophic intermonsoon season (Fischer et al 2002Weller et al 2002) Mixing of two water masses is indicatedby a linear correlation between temperature (T) and salin-ity (S) if latent and sensible heat fluxes between ocean andatmospheres are negligible Temperature und salinity (TS)data derived from the sampling sites along the Oman tran-sect are not correlated (Fig 3) and TS data obtained from

wwwbiogeosciencesnetbg31 Biogeosciences 3 1ndash13 2006


20

22

24

26

28

355 360 365 370Salinity

Tem

per

atu

re [

oC

]


Fig 3 Salinity versus temperature Salinities and temperatureswere measured during cruise ttn49 (open circles) and ttn50 (blackcircles) and averaged for the depth of the mixed layer (data are fromthe US JGOFS database) The black line connects the data col-lected at the sites S1 and S2

the sites S2ndashS14 deviate from the line (mixing line) thatconnects TS data obtained from the sampling site S1 andS15 In order to calculate energy fluxes required to explainthese deviations from the mixing line (1SST) the time dur-ing which the surface water was in contact with the atmo-sphere must be known The ages of the surface water canbe calculated at each sampling site by dividing the distanceto the coast by the mean advection velocity and consideringthat the upwelled water was already a few days old prior toit left the coast Since the lower temperatures and the re-duced salinity indicate an accelerated advection of upwelledwater within the filament an age correction has been appliedThe distance from the coast towards the most offshore station(S13) and the station closest to the coast (S1) issim1100 kmandsim26 km respectively At these two stations the salinitywas 366 and 357 (Table 2) Based on these two points de-fined by the distance to the coast and the salinity a linear re-gression was developed which allowed us to derive correcteddistances from the salinity measured at each sampling site(distance [km]=11351timesSalinityminus404518) The time sincethe water mass was in contact with the atmosphere was ob-tained by dividing the corrected distance by the mean ad-vection velocity ranging between 02 and 08 m sminus1 (Rixenet al 2000) Additionally it was assumed that coastally up-welled water was already 4 days old prior to it left the coastThe energy fluxes required to explain1SST have been calcu-lated as follows (ρtimesMLDtimesCptimes1SST )(age of the water)whereas ldquoρrdquo is the sea water density in ldquokg mminus3rdquo MLD is the

00

06

12

18

0 300 600 900 1200

Ph

osp

hat

e [1

0-6m

ol k

g-1

]


a)

00

01

02

03

04

05

00

06

12

18

00 02 04 06 08 100

Ph

osp

hat

e [1

0-6m

ol k

g-1

]

Co

nsu

mp

tio

n [

10-6

mo

l kg

-1]

b)

Proportion of oligotrophic water

Rixen et al Figure 4Title Diatoms and their effect Fig 4 (a) Phosphate concentration measured during cruise ttn49and averaged for the depth of the mixed layer versus the distanceto the coast (data are from the US JGOFS database)(b) Meanphosphate concentration (black circles) and the mixing line (blackline) versus the proportion of oligotrophic water Bars reveal thedifference between the measured phosphate concentration and thephosphate concentration indicated by the mixing line This differ-ence is regarded as biological consumption

mixed layer depth in ldquomrdquo ldquocprdquo is the specific heat of the wa-ter (sim3980 J kgminus1 Kminus1) 1SST and the age of the water aregiven in ldquoCrdquo and seconds respectively The calculated datashow that a heat flux of 90 to 584 W mminus2 could have causedthe1SST at sampling site S7 (Table 3) Energy fluxes of 150to 200 W mminus2 determined at the US JGOFS surface moor-ing site which is close to S7 (Weller et al 1998) fall withinthis range implying that heat fluxes could have caused1SSTat station S7 and most probably also at the other water sam-pling sites for which no data on heat fluxes are available

Latent heat fluxes do not affect the relationship be-tween salinity and nutrient concentrations as changes inthe amount of water in the mixed layer increase boththe salinity and the nutrient concentration Since in ad-dition to that salinity and nutrient concentration are un-affected by sensible heat flux salinity instead of tem-perature was used to define the end-members and tocalculate the mixing ratios ldquoardquo and ldquobrdquo within thetwo-end-member mixing analysis at sampling sites S1ndashS13 (SalinityS1minusS13=aS1minusS13timesSalinityupwelled water+bS1minusS13Salinityoligotrophic water whereas aS1minusS13+bS1minusS13=1) Sta-tion S1 was defined as the upwelling and station S13 asthe oligotrophic end-member because there are hardly anydiscernible variations in the salinity between stations S13



Table 2 Station number position distance between the station and the coast mixed layer depth mean mixed layer temperature salinityphosphate total inorganic nitrogen silicon and total inorganic carbon concentration mean total alkalinity and partial pressure differencebetween the atmosphere and the ocean Missing data are indicated byminus9900 The position was obtained by averaging the position of allavailable hydrographic casts at one sampling sites

St Lon Lat Distance MLD Temp Sal PO4 TN Si DIC TA 1pCO2[ E] [ N] [km] [m] [C] [psu] mdashmdashndash[micromolkg]mdashmdashndash microeqkg microatm

Cruise ttn49

S1 5732 1852 2573 1000 2084 3566 155 1740 923 214650 233860minus29909S2 5803 1816 9301 1800 2298 3571 132 1540 764 211143 232918minus22265S3 5886 1767 19358 5300 2599 3600 069 558 313 204665 234928minus9937S4 5987 1717 31398 4200 2665 3610 053 412 273 204478 234866minus8325S5 6051 1679 39279 4800 2590 3602 060 481 251 204292 234880minus8941S6 6125 1643 48212 4000 2455 3582 113 1216 559 209147 234020minus15956S7 6200 1603 57355 3400 2574 3605 079 725 355 205673 235063minus11640S8 6281 1564 66995 7500 2689 3637minus9900 152 180 203058 236900 minus5744S9 6351 1523 75738 8200 2666 3629 044 201 125 203504 236273minus5462S10 6425 1482 84937 8300 2673 3632 039 096 116 202816 236669minus3980S11 6500 1444 94011 7500 2728 3647 032 051 099 202441 237523minus2776S12 6500 1324 101075 8400 2719 3643 027 004 054 202013 237347minus2738S13 6501 1205 109272 10800 2771 3660 024 017 069 201874 238250minus1972S14 6500 1081 118665 10400 2772 3657 027 029 110 202096 238127minus2393S15 6491 997 124793 9500 2788 3658 025 040 119 201780 238140minus2745

Cruise ttn50

S1 5730 1850 2326 1400 2095 3566 166 1942 1118minus9900 minus9900 minus9900S2 5804 1809 9634 1000 2393 3569 132 1509 571minus9900 minus9900 minus9900S3 5885 1767 19268 2000 2382 3574 107 1150 199minus9900 minus9900 minus9900S4 5977 1718 30375 2500 2489 3579 096 962 151minus9900 minus9900 minus9900S5 6050 1680 39248 5200 2564 3584 085 838 112minus9900 minus9900 minus9900S6 6125 1643 48132 5200 2665 3611 053 315 097minus9900 minus9900 minus9900S7 6198 1601 57300 5000 2652 3604 059 384 149minus9900 minus9900 minus9900S8 6277 1564 66657 4800 2610 3595 076 652 170minus9900 minus9900 minus9900S9 6350 1525 75605 7400 2654 3607 057 384 242minus9900 minus9900 minus9900S11 6500 1444 94020 7500 2693 3623 041 122 073minus9900 minus9900 minus9900S13 6500 1205 109155 10200 2742 3651 034 031 059minus9900 minus9900 minus9900S14 6500 1080 118662 10000 2764 3651 032 009 042minus9900 minus9900 minus9900S15 6490 997 124766 8400 2785 3657 030 013 027minus9900 minus9900 minus9900

14 and 15 (Table 2) A mixing ratio ldquobrdquo of zero indicatespure upwelled water whereas a mixing ratio ldquobrdquo of one im-plies that no upwelled water is present The mixing ratiosldquoardquo and ldquobrdquo determined for each station and the mean nu-trient concentration calculated at the station S1 and S13(see Table 2) were subsequently used to calculate a mixingline (PhosphateS1minusS13mixing=aS1minusS13timesPhosphateS1+bS1minusS13PhosphateS13 Fig 4b) The mixing line represents the con-centration that could be expected if mixing were the onlyfactor controlling the nutrient concentration Deviations ofmeasured phosphate inorganic nitrogen and silicon concen-trations from the mixing line can be attributed to biologicalconsumption In order to determine error ranges of the de-rived biological consumption caused by analytical methodsa relative percentage error of 05 and 0012 of nutrient

and salinity data (US JGOFS data base documentation) wasconsidered Within these error ranges random errors wereproduced by applying the standard fortran 77 random num-ber generator (F77-RNG) Subsequently the mixing analysiswas performed 500 times by using the produced errors Themean ratios of inorganic nitrogen and phosphate consump-tions (NP ratios) vary betweensim7 and 24 (Fig 5b) and fallwithin the range of NP ratios determined in marine particu-late matter and phytoplankton (sim3ndash34 Geider and La Roche2002 Klausmeier et al 2004) The resulting standard devi-ations of the NP ratios range between 01 and 04

The DIC and TA data were treated in the same way asnutrient data but prior to calculating the biological con-sumption the DIC concentrations were corrected for CO2emission The CO2 emission into the atmosphere have



Table 3 Sampling site the mean mixed layer temperature (deter-mined) the temperature as suggested by the mixing line indicatein Fig 4 (expected) the difference between the mixed layer tem-perature and the expected temperature (1SST) and energy fluxesrequired to explain the temperature deviation The energy flux hasbeen calculated by assuming a mean advection velocity of 08 and02 m sminus1

Sampling Temperature [C] Energy fluxes [W mminus2]site determined expected deviation 02 m sminus1 08 m sminus1

ttn 49

S12 2719 2686 032 76700 23298S11 2728 2720 008 16303 4909S10 2673 2602 070 183139 57077S9 2666 2584 082 214771 67379S8 2688 2639 049 109286 33650S7 2574 2397 176 261963 90695S6 2455 2222 233 623791 271435S5 2589 2373 216 477645 168688S4 2665 2437 227 386403 130107S3 2598 2356 242 612288 219543S2 2297 2137 160 259434 151544

ttn50

S11 2693 2537 155 399021 127506S9 2653 2409 244 772361 265124S8 2610 2318 291 725146 270388S7 2652 2389 263 584430 203671S6 2664 2440 224 468718 157517S5 2564 2238 326 1081012 454687S4 2488 2195 293 534635 249007S3 2382 2159 223 369189 194683S2 2392 2121 271 261293 168481

been derived from the1pCO2 data published for each wa-ter sampling site in the US JGOFS database (Table 2)The required wind-dependent gas transfer velocity ldquokrdquo wascalculated using eight different formulations (Liss and Mer-livat 1986 Wanninkhof 1992 Wanninkhof and McGillis1999 Nightingale et al 2000 Feely et al 2001) Satellite-derived wind speeds (see Fig 1) were taken from Rixen etal (1996) The standard deviation and the resulting meanCO2 emission at each sampling site along the transect aregiven in Fig 6 In order to estimate the amount of CO2 thatescaped from the surface water the CO2 emissions have beenmultiplied by the age of the water which has been was calcu-lated by using a mean advection velocity of 06 m sminus1 Theresulting mean CO2 loss was subtracted from the amount ofDIC held in the mixed layer A further correction for the pen-etration of anthropogenic CO2 was neglected because of theshort time scale covered by the cruises The CO2-correctedDIC and the TA consumption (DICconsumption TAcons) havebeen used for differentiating between the net organic carbonproduction (POCproduction) and the precipitation of calcium

50

100

150

200CP

Rat

io

a)

10

20

30NP

Rat

io

b)

0

5

10

15

00 02 04 06 08 10

223 POCPIC

Rat

io

c)



Fig 5 (a) CP (b) NP and(c) rain (POCPIC) ratios versus theproportion of oligotrophic water Error bars are calculated as de-scribed in the text The number in panel (c) indicates the rain ratioat the sampling site S11 which is not plotted as it is out of scale

carbonate (PICprecipitation) according to the equations

DICconsumption=POCproduction+ PICprecipitation (1)

PICprecipitation=(TAcons + POCproductiontimes 015)2 (2)

The factor 015 accounts for the increase of TA during theproduction of organic matter and the factor 2 due to thefact that for each mole of carbonate precipitated from thesea water TA decreases by two units (Broecker and Peng1982 Zeebe and Wolf-Gladrow 2001) With two equa-tions and two unknowns the production of organic carbon(POCproduction) can be determined as follows

POCproduction=(DICcorrectedminusTAcons2)108 (3)

Error ranges were determined by producing ldquoanalyticalrdquo er-rors artificially by applying the F77-RNG within the rangeof the standard deviation given for the determination ofDIC concentrations (plusmn12 10minus6 mol kgminus1) and TA (plusmn3210minus6eq kgminus1 Millero et al 1998 US JGOFS data basedocumentation) Subsequently the errors were used to re-calculate the POC and PIC production 500 times for eachof the different gas transfer velocity coefficients cited aboveThe resulting mean CP and rain ratios and their standard de-viations are given in Fig 5a The mean CP uptake ratiosvary between 80 and 150 and are well within the range ofpublished CP ratios observed in marine particulate matterand phytoplankton The same holds true for the rain ratioswhich vary betweensim23 andsim81 except at station S11



0

10

20

30

40

50

60

70

0 300 600 900 1200

CO

2 fl

ux

[mm

ol m

-2 d

ay-1

]

Distance [km]


Fig 6 Mean CO2 emission derived from1pCO2 data obtainedfrom the US JGOFS data satellite-derived wind speeds and thedifferent gas transfer velocity coefficients ldquokrdquo versus the distanceto the coast Error bars indicate the standard variation of resultsderived from the different wind speed data sets and the differentmethods to calculate ldquokrdquo

where the rain ratio is 223 (Fig 5c) However due to thelow carbon consumption the standard deviation at the sam-pling sites S10ndashS12 (proportion of oligotrophic watergt07)are so high that these data have to be treated with cautionwithin the following discussion

In order to calculate the new production rates (Dug-dale and Goering 1967 Eppley and Peterson 1979) thePOCproduction (see Eq 3) was integrated over the depth ofthe mixed layer and subsequently divided by the age of theupwelled water The comparison with primary and exportproduction rates measured at the same time and at the samestations (Buesseler et al 1998) show that the resulting newcarbon production rates are lower than the primary produc-tion rates as expected (Fig 7) Since new production ex-ceeds export production rates it is assumed that the biomassis growing in the mixed layer but under different environ-mental conditions

New production rates which are consistent with publisheddata on primary and export production rates and reliable rainand Redfield ratios suggest that our approach is suitable forstudying biogeochemical processes in the Arabian Sea dur-ing the upwelling season In the following discussion the re-sults obtained by our mixing analysis will furthermore belinked to plankton counts and sediment trap data in orderto investigate factors influencing changes in the planktoniccommunity structure and associated impacts on the rain ra-tio

0

20

40

60

80

100

120

140

160

180

200

220

00 02 04 06 08 10

[mm

ol C

m-2

day

-1]



Fig 7 Primary production rates (stars) export production rates de-rived from 234Th measurements (open diamonds Buesseler et al1998) and sediment trap data (grey diamonds Rixen et al 2003)Organic carbon (new) production calculated from the DIC and TAuptake is indicated by black circles

4 Discussion

41 Diatom blooms

The shoaling of the mixed layer and enhanced organic car-bon flux measured in the deep sea reveal that the cruise ttn49and the cruise German JGOFS cruise M325 took place atthe beginning of the upwelling season in the open westernArabian Sea (Fig 8) On month later during the cruise ttn50the high organic carbon fluxes indicate the peak upwellingseason At this time plankton counts (Garrison et al 2000)show that diatoms dominate the planktonic community fromapproximately 100 to 400 km offshore (Fig 9b) Flagellatessucceed diatoms in the central Arabian Sea after silicon con-centrations in the surface water reach their oligotrophic in-termonsoon value ofsim2micromol kgminus1 (compare Figs 9b andd) Plankton growth rates generally decrease with decreasingnutrient concentrations after the latter falls below a certainthreshold (Lalli and Parsons 1993) Assuming that such athreshold is close to a silicon concentration ofsim2micromol kgminus1

diatom growth rates falling below the high grazing rates(Smith et al 1998) could explain the declining diatom bloomin the open western Arabian Sea where silicon in the surfacewater is not consumed This observation confirms experi-mental data suggesting a silicon-threshold ofsim2micromol kgminus1

for the dominance of diatoms within the planktonic commu-nity (Egge and Aksnes 1992) Near the coast at siliconconcentration ofsim11micromol kgminus1 the relatively low contri-bution of diatoms to the planktonic community (Fig 9b d)has been attributed to intense grazing by copepods slowingdown the development of diatom blooms in the coastal re-gion off Oman (Smith 2001) During the early phase of theupwelling season the abundance of diatoms decreased dras-tically close to the Oman coast as shown by plankton counts



0

5

10

15

20

25

30

2

3

4

5

6

May Jun Jul Aug Sep

1995

PO

C F

lux

[mg

m-2

day

-1]

Car

bo

nat

eB

iog

enic

Op

al

-120

-80

-40

0

Wat

er D

epth

[m

]

ttn49 ttn50

M325M325MLD

Eup

thot

ic z

one

Title Diatoms and their effect Rixen et al Figure 8Fig 8 Depth of the mixed layer (MLD black line) and of the eu-photic zone (1 light level green line) determined at the ONRmooring site Data was redrawn from (Dickey et al 1998) TheMLD represents the water-depth at which the temperature is 1Clower than at the surface The organic carbon fluxes (black line)and carbonatebiogenic opal ratios (red line) measured in the west-ern Arabian Sea at the long-term IndoGerman sediment trap site(WAST) at 3000 m water-depth The horizontal black lines showthe time during which the cruises M325 have been carried out andthe cruises ttn49 and 50 have been performed along the transect offOman in the western Arabian Sea The other horizontal lined indi-cate the time at which the measured surface data could be reflectedin the deep ocean A delay of 14 days has been added to originaldates

(Fig 9a Schiebel et al 2004) This decline is accompaniedwith a transition from aChaetocerosdominated diatom as-semblage near the coast to one dominated byNitzchia bicap-itata further offshore The decline and the associated tran-sition of the diatom assemblage has been linked to decreas-ing nutrient concentrations and intense grazing (Smith 2001Schiebel et al 2004)

The results obtained by the plankton counts agree withthose derived from deep moored sediment traps and ourmixing analysis showing silicon consumption and carbonatebiogenic opal ratios which are lower and higher respectivelyduring the early than during the later phase of the upwellingseason in the open western Arabian Sea (Figs 9c 10a) Viceversa near the coast carbonate biogenic opal ratios reveal ahigher contribution of diatoms to the exported matter duringthe early than during the later phase of the upwelling season(Fig 9c) This seems to be caused by the export of the de-clining diatom bloom during the early phase of the upwellingseason (Fig 9a) and the delayed development of the diatombloom one month later (Fig 9b)

0250500750

100012501500

0

20

40

60

Dia

tom

s [1

03 va

lves

l-1]

Dia

tom

s [1

03 va

lves

l-1]

Co

cco

lith

[10

3 ce

lls l-1

]a)

0

30

60

90

Pla

nkt

on

Co

mp

[

] P

lan

kto

n C

om

p [

]

b)

1

2

3

4

5

6

7

Car

bo

nat

e

Bio

gen

ic O

pal

c)

02468

101214

0 300 600 900 1200

Si -

co

nce

n

[10-6

mo

l kg

-1]

d)

Distance [km]

Title Diatoms and their effect Rixen et al Figure 9Fig 9 (a) Number of diatom valves and coccolithophorid cells(Schiebel et al 2004)(b) contribution of nanoplankton (green)diatoms (red) flagellates (blue) and cyanobacteria (black) to thebiomass of photoautotrophic plankton (Garrison et al 2000)(c)Carbonate biogenic opal ratios (blue ndash peak upwelling season red ndashonset of the upwelling season) derived from deepest sediment trapsdeployed at the US JGOFS and IndoGerman sediment trap site(see Figs 1 and 2)(d) silicon concentrations (blue ndash ttn50 red ndashttn49) versus the distance to the coast The sediment traps havebeen deployed at water-depth ranging approximately between 2500and 3500 m except near the coast the sediment trap was deployed ata water-depth ofsim1000 m

42 The role of iron for the development of diatom blooms

Iron enrichment experiments in the coastal upwelling sys-tem off California revealed that iron fertilization favours thegrowths rates ofChaetocerosand other diatoms but leadsto the formation of thinner shells as indicated by a low up-take of silicon relative to nitrogen (Hutchins and Bruland1998) The iron concentration in the diatom dominated openwestern Arabian Sea range between 05 and 1 nmol lminus1 dur-ing the cruise ttn50 and reach values ofgt2 nmol lminus1 duringthe cruise ttn49 near the coast (Fig 10b) Off Californiaan increase in the iron concentration from 05 to 25 halvesSiN uptake ratio (Hutchins and Bruland 1998) Assuminga similar impact of iron on the SiN uptake off Oman sug-gests thatChaetocerosblooming during the onset of the up-welling season near the coast off Oman might built thinner


T Rixen et al Diatom and their role in the Arabian Sea upwelling system 9 34

Fig 10 (a)Silicon consumption(b) iron concentration versus thedistance to the coast during the cruises ttn49 and 50

shells than diatoms growing a month later in same region butin water revealing lower iron concentrations The extremelystrong diatom shell is an important protection against preda-tors (Hamm et al 2003) Since enhanced iron concentrationslead to the formation of thinner shells it is suggested thatcopepode grazing is favoured in iron-enriched environmentsAn iron-favoured grazing could explain the decline of thenear shore diatom bloom during the onset of the upwellingseason (Fig 9a) On the other hand a less efficient grazingdue to lower iron concentrations (Fig 10b) and thicker shellscould slow down but do not prevent the development of alarge diatom bloom during the later phase of the upwellingseason (Fig 9b) at which the silicon concentrations are al-most equal to those during early phase of the upwelling sea-son (Fig 10d)

43 Rain ratios

Apart from one exception as discussed before the rain ratiosderived from mixing analysis range betweensim23 and 81and are generally higher than those obtained from sedimenttrap data (2ndash35 Fig 11) The rain ratios derived from thesediment trap data have been adjusted to a water-depth of100 m whereas the mixed layer depth islt100 m in the west-ern Arabian Sea (Table 1) POCPIC ratios tend to decreasewithin increasing water depth because the decomposition oforganic matter is faster than the dissolution of carbonates inthe upper water column Consequently the difference be-tween rain ratios derived from sediment trap data and themixing analysis could be caused by differences in the water-depth Moreover one should consider that sediment trap datarepresent a larger area and a longer time interval than data de-rived form nutrient profiles obtained at a specific site withina relatively short period of time

However rain ratios obtained from sediment trap data andthe mixing analysis reveal a similar trend in the upwelling

2

4

6

8

10

Rai

n r

atio

a)

2

3

4

00 02 04 06 08 102

3

4

00 02 04 06 08 10

84

86

88

90

92

00 02 04 06 08 102

3

4

00 02 04 06 08 10

MS1amp2MS1amp2 MS4 MS3 WAST CAST MS5

upwelling area oligotrophic area

22

Rai

n r

atio

(tr

ap d

ata)

CN

rat

io (

trap

dat

a)

b)


Rixen et al Figure 11Title Diatoms and their effect Fig 11 (a)Rain ratios and silicon consumption derived from themixing analysis versus the proportion of oligotrophic water (samedata as shown in Figs 5 and 10a)(b) Rain (black circles) and CNratios (open circles) derived from sediment traps versus the pro-portion of oligotrophic water data The rain ratios are mean SWmonsoon values and have been derived from trap data (Rixen etal 2005) The CN ratios obtained from deepest trap deployed ateach mooring site have been averaged for the period of the cruisettn4 The proportion of oligotrophic water was determined by ad-joining to them the closest water-sampling site during the cruisettn49 Since lateral advection has to be considered additionally notthe closest station was used but its neighbour station to the west(coast) The grey circles indicate long-term mean rain and CN ra-tio obtained from the IndoGerman sediment trap site in the centralArabian Sea (CAST) but it does not include data from the SW mon-soon 1995 due to an instrumental error

region characterized by a proportion of oligotrophic water(pow) lt06 (Fig 11) Near the upwelling centres (powlt035) the rain ratios increase with an increasing proportionof oligotrophic water until they reach a maximum at a powof sim035 This increase is accompanied with an enhancedsilicon consumption (Fig 10a) indicating a slight recovery ofthe diatom (Nitzchia bicapitata)bloom after the dramatic de-cline ofChaetocerosdominated diatom bloom near the coastduring the early upwelling season (Fig 9a) At the tran-sition between the upwelling dominated area and the moreoligotrophic sites the rain ratios decrease bysim30 (sedimenttrap data) and 45 (mixing analysis) This decline is accom-panied with decreasing silicon consumption (Fig 10a) andduring the later phase of the upwelling season also with de-creasing contribution of diatoms to the plantonic communitystructure (Fig 9b)

The high abundance of coccolithophorids at approxi-mately 450 km offshore (Fig 9a) occurred in a water mass



0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io



Fig 12CN ratios derived from the mixing analysis (black circles)and sediments traps (grey circles same data as in Fig 11b) versusthe proportion of oligotrophic water

which is characterized by a pow ofsim02 The associated rainratio of sim5 is relatively high (Fig 11a) and implies in linewith sediment trap results that coccolithophorids are onlyof minor importance for the carbonate export in the Ara-bian Sea This in turn implies that changes of the rain ra-tios as discussed before are mainly caused by variations inthe growth and export rates of foraminifera Enhanced rainratios associated with even a slightly increased contributionof diatoms in the planktonic community structure could forexample results from predators such as copepods competingwith foraminifera for diatoms

High rain ratios seem to characterize the oligotrophic re-gion in which cyanobacteria dominate the planktonic com-munity (Figs 11 9b) These high rain ratios derived frommixing analysis are not mirrored by the sediment trap datawhich reveal the lowest values in this region (Fig 11) Sinceit is generally assumed that cyanobacteria are rapidly rem-ineralized in the upper water column and hardly sink intothe deep sea (Karl et al 1996 1997) a preferential decom-position of cyanobacterial biomass could explain the differ-ence between the rain ratios derived from the mixing anal-ysis and the those obtained from the sediment trap data Apreferential decomposition of more labile nitrogen contain-ing compounds in the water column (Lee and Cronin 1982Wakeham et al 1997 Lee et al 2000) is also indicated byCN ratios which are lower in the organic matter producedin the surface water than those measured in organic mat-ter intercepted by sediment traps in the oligotrophic regions(Fig 12)

44 CN and CP ratios

Contrary to the oligotrophic region CN ratios partly decreasewith depth impling a preferential decomposition carbon-

enriched material in the upwelling area (Fig 12) Suchcarbon-enriched organic matter could be transparent ex-opolymers (TEP) which are produced by diatoms and fos-ter the formation of fast-sinking marine snow (Passow et al1994 2001) The formation of TEP raises the CP ratiosespecially at the end of diatom blooms (Engel et al 2002)and along the transect at the transition from the upwellingtowards the oligotrophic system the CP ratios reveal a dropfrom valuesgt108 to 78 (Fig 5a) The huge decrease of theCP and CN ratios is not reflected in the CN ratios of theexported matter in the deep sea which show only asim7 de-crease at the transition from the upwelling towards the olig-otrophic system (Fig 11b) This implies that the preferentialdecomposition of TEP-like organic matter in the upper wa-ter column reduces the impact of the TEP formation on theRedfield ratio in the exported matter during diatom blooms

5 Conclusions

The evaluation of data collected during the US and GermanJGOFS expedition in 1995 suggest that an intense grazingdecline and impede the development of large diatom bloomsin a silicon-enriched near shore upwelling system off OmanWithin the offshore advecting upwelled water mass a largediatom blooms enhancing the organic carbon export into thedeep sea develop only during the later phase of the upwellingperiod During the onset of the upwelling season no largediatoms blooms occurs in the open western Arabian Sea de-spite increased concentrations of dissolved silicon in the sur-face water Enhanced iron concentration due to eolian dustinputs into the offshore advecting upwelled water might havefavoured grazing at this time because it leads to the formationof thinner diatom shells The decrease of the rain ratios dur-ing transition from upwelling towards the oligotrophic off-shore region seems to be caused by foraminifera competingwith diatom-grazing copepods for food The large variabil-ity of Redfield ratios and also of the rain ratios in the olig-otrophic region is not reflected in the sediment trap recordThis suggests that the decomposition of organic matter inthe water columns contributes to the relative uniform globalmean Redfield ratio and leads to an underestimation of rainratios derived from sediment traps records in oligotrophic re-gions dominated by cyanobacteria

AcknowledgementsWe would like to thank all the scientiststechnicians and officers and their crews of the numerous researchvessels as well as the national funding agencies who made theJoint Global Ocean Flux Study in the Arabian Sea possibleParticularly we would like to appreciate the work of S HonjoS Manganini T Dickey K Buesseler C I Measures S VinkJ M Morrison D L Garrison L Codispoti and S L Smith whichin particular contributed to our study We would like to thank alsoC Lee S W A Naqvi and T Pohlmann for helpful discussionsFurthermore we are grateful to the Federal German Ministry forEducation Science Research and Technology (BMBF Bonn)



and the German Research Council (DFG Bonn) the Council ofScientific and Industrial Research (CSIR New Delhi) and the De-partment of Ocean Development (DOD New Delhi) for financialsupport of the Bilateral IndoGerman Program on BiogeochemicalFluxes in the northern Indian Ocean P Wessels and W H F Smithare acknowledged for providing the generic mapping tools (GMT)as well as B Aksen for secretarial assistance

Edited by A Watson

References

Anderson L A and Sarmiento J L Redfield ratios of rem-ineralization determined by nutrient data analysis Global Bio-geochem Cycles 8(1) 65ndash80 1994

Antoine D Andre J-M and Morel A Oceanic primary produc-tion ndash 2 Estimation at global scale from satellite (coastal zonecolor scanner) chlorophyll Global Biogeochem Cycles 10(1)57ndash69 1996

Archer D Winguth A M E Lea D and Mahowald N Whatcaused the glacialinterglacial atmosphericpCO2 cycles RevGeophys 38(2) 159ndash189 2000

Berger W H and Keir R S Glacial-Holocene Changes in At-mospheric CO2 and the Deep-Sea Record in Climate Processesand Climates Sensitivity edited by Hansen J E and TakahashiT Am Geophys Union Washington 337ndash351 1984

Broecker W S and Peng T-H Tracers in the sea Lamont-Doherty Geological Observatory Columbia University Pal-isades New York 690 pp 1982

Buesseler K Ball L Andrews J Benitez-Nelson C BelastockR Chai F and Chao Y Upper ocean export of particulate or-ganic carbon in the Arabian Sea derived from thorium-234 DeepSea Research II 45(10ndash11) 2461ndash2487 1998

Burkhardt S Zondervan I and Riebesell U Effect ofCO2concentration on C N P ratio in marine phytoplankton Aspecies comparison Limnology and Oceanography 44(3) 683ndash690 1999

Conley D J Terrestrial ecosystems and the biogeochemical silicacycle Global Biogeochem Cycles 16(4) 68-1ndash68-8 2002

Dugdale R C and Goering J J Uptake of new and regener-ated forms of nitrogen in primary productivity Limnology andOceanography 12 196ndash206 1967

Dugdale R C and Wilkerson F P Silicate regulation of new pro-duction in the equatorial Pacific upwelling Nature 391 270ndash273 1998

Egge J K and Aksnes D L Silcate as regulating nutrient inphytoplankton competition Marine Ecology Progress Series 83281ndash289 1992

Engel A Goldtwait S Passow U and Alldredge A L Tempo-ral decoupling of carbon and nitrogen dynamics in a mesocosmdiatom bloom Limnology and Oceanography 47(3) 753ndash7612002

Eppley R W and Peterson B J Particulate organic matter fluxand planktonic new production in the deep ocean Nature 282677ndash680 1979

Falkowski P Scholes R J Boyle E Canadell J CanfieldD Elser J Gruber N Hibbard K Hogberg P Linder SMackenzie F T Moore III B Pedersen T Rosenthal YSeitzinger S Smetacek V and Steffen W The Global Carbon

Cycle A Test of Our Knowledge of Earth as a System Science290 291ndash296 2000

Feely R A Sabine C L Takahashi T and Wanninkhof RUptake and storage of carbon dioxide in the ocean The globalsurvey Oceanography 14(4) 18ndash32 2001

Findlater J Observational Aspects of the Low-level Cross-equatorial Jet Stream of the Western Indian Ocean Pageoph115 1251ndash1262 1977

Fischer A S Weller R A Rudnick D L Eriksen C C LeeC M Brink K H Fox C A and Leben R R Mesoscaleeddies coastal upwelling and the upper-ocean heat budget inthe Arabian Sea Deep Sea Research Part II Topical Studies inOceanography 49(12) 2231ndash2264 2002

Froelich P N Blanc V Mortlock R A and Chillrud S NRiver fluxes of dissolved silica to te ocean were higher duringglacials GeSi in diatoms rivers and oceans Paleoceanography7(6) 739ndash767 1992

Garrison D L Gowing M M Hughes M P Campbell LCaron D A Dennett M R Shalapyonok A Olson R JLandry M R and Brown S L Microbial food web structurein the Arabian Sea a US JGOFS study Deep Sea Research II47(7ndash8) 1387ndash1422 2000

Geider R J and La Roche J Redfield revisited variability ofCNP in marine microalgae and its biochemical basis EuropeanJournal of Phycology 37 1ndash17 2002

Goldman J C McCarthy J J and Peavey D G Growth rate in-fluence on the chemical composition of phytoplankton in oceanicwaters Nature 279 210ndash214 1979

Goyet C Millero F J OrsquoSullivan D W Eischeid G McCueS J and Bellerby R G J Temporal variations of pCO2 insurface seawater of the Arabian Sea in 1995 Deep Sea ResearchI 45(4ndash5) 609ndash623 1998a

Goyet C Metzl N Millero F Eischeid G OrsquoSullivanD and Poisson A Temporal variation of the sea surfaceCO2carbonate properties in the Arabian Sea Marine Chemistry63(1ndash2) 69ndash79 1998b

Goyet C Coatanoan C Eischeid G Amaoka T Okuda KHealy R and Tsunogai S Spatial variation of total CO2 andtotal alkalinity in the northern Indian Ocean A novel approachfor the quantification of anthropogenic CO2 in seawater J Ma-rine Res 57 135ndash163 1999

Haake B Rixen T and Ittekkot V Variability of moonsonal up-welling signals in the deep western Arabian Sea SCOPEUNEPSonderband 76 85ndash96 1993a

Haake B Ittekkot V Rixen T Ramaswamy V Nair R R andCurry W B Seasonality and interannual variability of parti-cle fluxes to the deep Arabian Sea Deep Sea Research I 40(7)1323ndash1344 1993b

Hamm C E Merkel R Springer O Jurkojc P Maler CPrechtel K and Smetacek V Architecture and material prop-erties of diatom shells provide effective mechanical protectionNature 421 841ndash843 2003

Harrison K G Role of increased marine silica input on paleo-pCO2 levels Paleoceanography 15(3) 292ndash298 2000

Hedges J I Baldock J A Gelinas Y Lee C Peterson M Land Wakeham S G The biochemical and elemental composi-tion of marine plankton A NMR perspective Marine Chemistry78 47ndash63 2002

Heinze C Simulating oceanic CaCO3 export production



in the greenhouse Geophys Res Lett 31 L16308doi1010292004GL020613 2004

Heinze C Maier-Reimer E and Winn K Glacial pCO2 Reduc-tion by the World Ocean Experiments with the Hamburg CarbonCycle Model Paleoceanography 6(4) 395ndash430 1991

Honjo S Dymond J Prell W and Ittekkot V Monsoon-controlled export fluxes to the interior of the Arabian Sea DeepSea Research II 46(8ndash9) 1859ndash1902 1999

Hupe A and Karstensen J Redfield stoichiometry in Arabian Seasubsurface waters Global Biogeochem Cycles 14(1) 357ndash3722000

Hutchins D A and Bruland K W Iron-limited diatom growthand SiN uptake ratios in a costal upwelling regime Nature 393561ndash564 1998

Karl D Letelier R Tupas L Dore J Christian J and HebelD The role of nitrogen fixation in biogeochemical cycling in thesuptropical North Pacific Ocean Nature 388 533ndash538 1997

Karl D M Christian J R Dore J E Hebel D V LetelierR M Tupas L M and Winn C D Seasonal and interan-nual variability in primary production and particle flux at StationALOHA Deep Sea Research II 43(2ndash3) 539ndash568 1996

Klausmeier C A Lichtman E Daufresne T and Levin S AOptimal nitrogen-to-phosphorus stoichiometry of phytoplank-ton Nature 429 171ndash174 2004

Kortzinger A Duinker J C and Mintrop L StrongCO2emissions from the Arabian Sea during south-west mon-soon Geophys Res Lett 24(14) 1763ndash1766 1997

Lalli C and Parsons T R Biological Oceanography an introduc-tion 301 p Pergamon Press Ltd Oxford 1993

Lee C and Cronin C The vertical flux of particulate organic ni-trogen in the sea decomposition of amino acids in the Peru up-welling area and the equatorial Atlantic J Marine Res 40(1)227ndash251 1982

Lee C Wakeham S G and Hedges I J Composition and fluxof particulate amino acids and chloropigments in equatorial Pa-cific seawater and sediments Deep Sea Research Part I Oceano-graphic Research Papers 47(8) 1535ndash1568 2000

Lee C Murray D W Barber R T Buesseler K O Dymond JHedges J I Honjo S Manganini S J and Marra J Particu-late organic carbon fluxes compilation of results from the 1995US JGOFS Arabian Sea Process Study Deep Sea Research II45(10ndash11) 2489ndash2501 1998

Liss P S and Merlivat L Air-sea gas exchange rates Introductionand synthesis in The Role of Air-Sea Exchange in Geochemi-cal Cycling edited by Buat-Mernard P p 113ndash129 ReidelBoston 1986

Maier-Reimer E Dynamic vs apparent Redfield ratio in theoceans A case for 3D-models J Marine Syst 9 113ndash1201996

Maier-Reimer E Mikolajewicz U and Winguth A Futureocean uptake of CO2 interaction between ocean circulation andbiology Climate Dynamics 12 711ndash721 1996

Matsumoto K and Sarmiento J L Silicic acid leakage from theSouthern Ocean A possible explanation for glacial atmosphericpCO2 Global Biogeochem Cycles 16(3) 5-1ndash5-23 2002

Measures C I and Vink S Seasonal variations in the distribu-tion of Fe and Al in the surface waters of the Arabian Sea DeepSea Research Part II Topical Studies in Oceanography 46(8-9)1597ndash1622 1999

Millero F J Degler E A OrsquoSullivan D W Goyet C and Eis-cheid G The carbon dioxide system in the Arabian Sea DeepSea Research Part II Topical Studies in Oceanography 45(10ndash11) 2225ndash2252 1998

Milliman J D and Droxler A W Neritic and pelagic carbonatesedimentation in the marine environment ignorant is not blissGeologische Rundschau 85 496ndash504 1996

Morrison J M Codispoti L A Gaurin S Jones B Mang-hanani V and Zheng Z Seasonal variation of hydrographicand nutrient fields during the US JGOFS Arabian Sea ProcessStudy Deep Sea Research II 45(10ndash11) 2053ndash2101 1998

Nightingale P D Malin G Law C S Watson A J Liss P SLiddicoat M I Boutin J and Upstill-Goddard R C In situevaluation of air-sea gas exchange parameterizations using novelconservative and volatile tracers Global Biogeochem Cycles14 373ndash387 2000

Passow U Alldredge A L and Logan B E The role of particu-late carbohydrate exudates in the flocculation of diatom bloomsDeep Sea Research I 41(2) 335ndash357 1994

Passow U Shipe R F Murray A Pak D K Brzezinski M Aand Alldredge A L The origin of transparent exopolymer par-ticles (TEP) and their role in the sedimentation of particulatematter Continental Shelf Research 21(4) 327ndash346 2001

Ramage C S Monsoon Meteorology Academic Press New YorkLondon 1971

Ramage C S Monsoon Climates in The Encyclopedia of Cli-matology edited by Oliver J E and Fairbridge R W VanNostrand Reinhold Company New York 1987

Redfield A C Ketchum B H and Richards F A The Influenceof organisms on the composition of sea-water in The sea editedby Hitt M N pp 26ndash77 Wiley amp Sons New York 1963

Ridgwell A J Watson A J and Archer D Modeling the re-sponse of the oceanic Si inventory to perturbation and con-sequences for atmospheric CO2 Global Biogeochem Cycles16(4) 19-1ndash19-15 2002

Rixen T Haake B and Ittekkot V Sedimentation in the westernArabian Sea the role of coastal and open-ocean upwelling DeepSea Research II 47 2155ndash2178 2000

Rixen T Guptha M V S and Ittekkot V Sedimentation inReport of the Indian Ocean Synthesis Group on the Arabian SeaProcess Study edited by Watts L Burkill P H and Smith Spp 65ndash73 JGOFS International Project Office Bergen 2002

Rixen T Guptha M V S and Ittekkot V Deep ocean fluxes andtheir link to surface ocean processes and the biological pumpProgress in Oceanography 65 240ndash259 2005

Rixen T Haake B Ittekkot V Guptha M V S Nair R Rand Schlussel P Coupling between SW monsoon-related sur-face and deep ocean processes as discerned from continous parti-cle flux meausurements and correlated satellite data J GeophysRes 101(C12) 28 569ndash28 582 1996

Sabine C L Wanninkhof R Key R M Goyet C and MilleroF J Seasonal CO2 fluxes in the tropical and subtropical IndianOcean Marine Chemistry 72(1) 33ndash53 2000

Sarmiento J L Dunne J Gnanadesikan A Key R M Mat-sumoto K and Slater R A new estimate of the CaCO3 toorganic carbon export ratio Global Biogeochem Cycles 16(4)54-1ndash54-12 2002

Schiebel R Planktic foraminiferal sedimentation and the marinecalcite budget Global Biogeochem Cycles 16(4) 13-1ndash13-21



2002Schiebel R Zeltner A Treppke U F Waniek J Bollmann J

Rixen T and Hemleben C Distribution of diatoms coccol-ithophores and planktic foraminiferes along a trophic gradientduring the SW monsoon in the Arabian Sea Marine Micropale-ontology 51 345ndash371 2004

Smith S Roman M Prusova I Wishner K Gowing M Codis-poti L A Barber R Marra J and Flagg C Seasonal re-sponse of zooplankton to monsoonal reversals in the ArabianSea Deep Sea Research Part II Topical Studies in Oceanogra-phy 45(10ndash11) 2369ndash2403 1998

Smith S L Understanding the Arabian Sea Reflections on the1994ndash1996 Arabian Sea Expedition Deep Sea Research Part IITopical Studies in Oceanography 48(6ndash7) 1385ndash1402 2001

Tegen I and Fung I Modeling of mineral dust in the atmosphereJ Geophys Res 99(D11) 22 897ndash22 914 1994

Tegen I and Fung I Contribution to the atmospheric mineralaerosol load from land surface modification J Geophys Res100(D9) 18 707ndash18 726 1995

Tyrrell T The relative influences of nitrogen and phosphorus onoceanic primary production Nature 400 525ndash531 1999

Volk T and Hoffert M I The carbon cycle and atmospheric CO2natural variation archean to present edited by Sundquits E Tand Broecker W S p 99ndash110 AGU Washington 1985

Wakeham S G Lee C Hedges J I Hernes P J and PetersonM L Molecular indicators of diagenetic status in marine or-ganic matter Geochimica et Cosmochimica Acta 61(24) 5363ndash5369 1997

Wanninkhof R Relationship between gas exchange and windspeed over the ocean J Geophys Res 97 7373ndash7381 1992

Wanninkhof R and McGillis W M A cubic relationship betweengas transfer and wind speed Geophys Res Lett 26 1889ndash19831999

Weller R A Baumgartner M F Josey S A Fischer A S andKindle J C Atmospheric forcing in the Arabian Sea during1994ndash1995 observations and comparisons with climatology andmodels Deep Sea Research Part II Topical Studies in Oceanog-raphy 45(10ndash11) 1961ndash1999 1998

Weller R A Fischer A S Rudnick D L Eriksen C C DickeyT D Marra J Fox C and Leben R Moored observations ofupper-ocean response to the monsoons in the Arabian Sea dur-ing 1994ndash1995 Deep Sea Research Part II Topical Studies inOceanography 49(12) 2195ndash2230 2002

Zeebe R E and Wolf-Gladrow D CO2 In Seawater EquilibriumKinetics Isotopes 346 p Elsevier Science B V Amsterdam2001

Zeltner A Monsoonal influenced changes of coccolithophorecommunities in the northern Indian Ocean ndash alteration duringsedimentation and record in surface sediments in TubingerMikropalaontologische Mitteilungen pp 103 Institut und Mu-seum fur Geologie und Palaontologie der Universitat TubingenTubingen 2000



Table 1 Contribution of carbonate-producing organism to the pelagic marine carbonate production

1015g C References

Total carbonate production 100 060ndash086 Milliman and Droxler (1996)Foraminifera 23ndash56 014ndash048 Schiebel (2002)Pteropods sim10 006ndash009 Schiebel (2002)Coccolithophorids 4ndash38 002ndash033 Schiebel (2002)

and references therein

Possible effects of changing Redfield and rain ratios onthe atmospheric CO2 concentration were quantified usingGCMs It was estimated that an increase of the CP ratioby 30 (1221 to 15861) could lower the atmospheric CO2concentration bysim72 ppm (Heinze et al 1991) A doublingof the global mean rain ratio from 4 to 8 could reduce theatmospheric CO2 concentration by 285 ppm (Heinze et al1991) and raising the rain ratio from 5 to 166 could how-ever reduce the atmospheric CO2 concentration by 70 ppm(Archer et al 2000) These changes explaining a major pro-portion of the glacialinterglacial variation of the atmosphericCO2 concentration were achieved by assuming that diatomsoutcompete the carbonate-producing coccolithophorids anddrive the export production Due to such a competition therain ratio is directly linked to biogenic opal production insome models used to study the feedback impact of the car-bonate production in the ocean on increasing CO2 concen-trations in the atmosphere (Heinze 2004)

Since diatom growth is believed to be limited by the avail-ability of silicon at lower latitudes (Dugdale and Wilkerson1998 Rixen et al 2000) changes of the global silicon cy-cle and re-organisation of the marine silicon cycle were pro-posed as being possible mechanisms to fertilise lower lati-tudes with silicon during glacial times (Froelich et al 1992Harrison 2000 Conley 2002 Ridgwell et al 2002) There-organisation of the marine silicon cycle is suggested to betriggered by an enhanced eolian iron input lowering the SiNuptake ratio of diatoms in the Southern Ocean and subse-quently enhancing the silicon export from higher to lowerlatitudes (Matsumoto and Sarmiento 2002)

In addition to coccolithophorids competing with di-atoms there are also carbonate-producing heterotrophs likeforaminifera and pteropods which feed on diatoms In theArabian Sea sediment trap experiments showed that coc-colithophorids contribute onlylt15 to the carbonate ex-port into the deep sea and that the peak flux of foraminiferainto the deep sea coincides with that of diatoms during thehighly productive upwelling season (Haake et al 1993b1993a Zeltner 2000) This suggests strongly that effects offoraminifera and pteropods on the rain ratio should be takeninto consideration as these organisms are important or eventhe main carbonate producer in the ocean (Table 1)

The total fluxes of calcium carbonate and organic car-bon measured by deep moored sediment traps in the Ara-bian Sea (Lee et al 1998 Honjo et al 1999 Rixen et al2002) and ratios between calcium carbonate dissolution andorganic carbon remineralization in the water column (Hupeand Karstensen 2000) have been used to calculate rain ratios(Rixen et al 2005) In this study these calculated rain ratiosranging between 2 and 35 will be compared with rain ratiosderived form total dissolved inorganic carbon concentrations(DIC) total alkalinity (TA) and CO2 partial pressure differ-ences (1pCO2) between the atmosphere and surface water(Goyet et al 1998b 1999 Millero et al 1998) Furthermoreplankton counts (Garrison et al 2000 Schiebel et al 2004)will be evaluated in conjunction with nutrient (Morrison etal 1998) and iron concentrations (Measures and Vink 1999)in order to study factors influencing shifts in the planktoniccommunity structure and associated changes of the composi-tion of sinking particles

2 Study area

The Arabian Sea is strongly influenced by the Asian mon-soon This climatic feature is driven by the sea-level pressuredifference between the Asian landmass and the Indian Ocean(Ramage 1971 1987) During the boreal winter the sea levelpressure over Asia exceeds that over the Indian Ocean due toa stronger cooling of the landmass Following the pressuregradient and deflected by the Coriolis force the wind blowsfrom the NE (NE monsoon) over the Arabian Sea This sit-uation reverses when the summer heating of the Asian land-mass leads to the formation of one of the strongest atmo-spheric lows on Earth This low attracts the SE trade windsand after crossing the equator the former SE winds blow asSW winds over the Arabian Sea due to associated changes ofthe Coriolis force The SW winds (SW monsoon) form a tro-pospheric jet (Findlater Jet) extending almost parallel to theArabian coast (Fig 1 Findlater 1977 Rixen et al 1996)The monsoon winds and the deserts surrounding the westernand northern parts of the Arabian Sea lead to dust inputs intothe Arabian Sea which are among the highest in the worldocean (Tegen and Fung 1994 1995)



Wind speed [m s-1]

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

55˚ 60˚ 65˚ 70˚

10˚ 10˚

15˚ 15˚

20˚ 20˚

25˚ 25˚







19 20 21 22 23 24 25 26 27 28 29 30 31 32

50˚ 60˚ 70˚

5˚ 5˚

10˚ 10˚

15˚ 15˚

20˚ 20˚

25˚ 25˚

20

21

22

23

24

25

26

27

28

29

0 300 600 900 120020

21

22

23

24

25

26

27

28

29

356

358

360

362

364

366

368

356

358

360

362

364

366


Su

rfac

e te

mp

erat

ure

[oC

]

Su

rfac

e sa

linit

y






20

22

24

26

28

355 360 365 370Salinity

Tem

per

atu

re [

oC

]




00

06

12

18

0 300 600 900 1200

Ph

osp

hat

e [1

0-6m

ol k

g-1

]


a)

00

01

02

03

04

05

00

06

12

18

00 02 04 06 08 100

Ph

osp

hat

e [1

0-6m

ol k

g-1

]

Co

nsu

mp

tio

n [

10-6

mo

l kg

-1]

b)









Cruise ttn49


Cruise ttn50









ttn 49

S12 2719 2686 032 76700 23298S11 2728 2720 008 16303 4909S10 2673 2602 070 183139 57077S9 2666 2584 082 214771 67379S8 2688 2639 049 109286 33650S7 2574 2397 176 261963 90695S6 2455 2222 233 623791 271435S5 2589 2373 216 477645 168688S4 2665 2437 227 386403 130107S3 2598 2356 242 612288 219543S2 2297 2137 160 259434 151544

ttn50

S11 2693 2537 155 399021 127506S9 2653 2409 244 772361 265124S8 2610 2318 291 725146 270388S7 2652 2389 263 584430 203671S6 2664 2440 224 468718 157517S5 2564 2238 326 1081012 454687S4 2488 2195 293 534635 249007S3 2382 2159 223 369189 194683S2 2392 2121 271 261293 168481


50

100

150

200CP

Rat

io

a)

10

20

30NP

Rat

io

b)

0

5

10

15

00 02 04 06 08 10

223 POCPIC

Rat

io

c)












0

10

20

30

40

50

60

70

0 300 600 900 1200

CO

2 fl

ux

[mm

ol m

-2 d

ay-1

]

Distance [km]






0

20

40

60

80

100

120

140

160

180

200

220

00 02 04 06 08 10

[mm

ol C

m-2

day

-1]




4 Discussion

41 Diatom blooms






0

5

10

15

20

25

30

2

3

4

5

6

May Jun Jul Aug Sep

1995

PO

C F

lux

[mg

m-2

day

-1]

Car

bo

nat

eB

iog

enic

Op

al

-120

-80

-40

0

Wat

er D

epth

[m

]

ttn49 ttn50

M325M325MLD

Eup

thot

ic z

one




0250500750

100012501500

0

20

40

60

Dia

tom

s [1

03 va

lves

l-1]

Dia

tom

s [1

03 va

lves

l-1]

Co

cco

lith

[10

3 ce

lls l-1

]a)

0

30

60

90

Pla

nkt

on

Co

mp

[

] P

lan

kto

n C

om

p [

]

b)

1

2

3

4

5

6

7

Car

bo

nat

e

Bio

gen

ic O

pal

c)

02468

101214

0 300 600 900 1200

Si -

co

nce

n

[10-6

mo

l kg

-1]

d)

Distance [km]








43 Rain ratios



2

4

6

8

10

Rai

n r

atio

a)

2

3

4

00 02 04 06 08 102

3

4

00 02 04 06 08 10

84

86

88

90

92

00 02 04 06 08 102

3

4

00 02 04 06 08 10



22

Rai

n r

atio

(tr

ap d

ata)

CN

rat

io (

trap

dat

a)

b)







0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io






44 CN and CP ratios



5 Conclusions






Edited by A Watson

References
























































































Wind speed [m s-1]

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

55˚ 60˚ 65˚ 70˚

10˚ 10˚

15˚ 15˚

20˚ 20˚

25˚ 25˚







19 20 21 22 23 24 25 26 27 28 29 30 31 32

50˚ 60˚ 70˚

5˚ 5˚

10˚ 10˚

15˚ 15˚

20˚ 20˚

25˚ 25˚

20

21

22

23

24

25

26

27

28

29

0 300 600 900 120020

21

22

23

24

25

26

27

28

29

356

358

360

362

364

366

368

356

358

360

362

364

366


Su

rfac

e te

mp

erat

ure

[oC

]

Su

rfac

e sa

linit

y






20

22

24

26

28

355 360 365 370Salinity

Tem

per

atu

re [

oC

]




00

06

12

18

0 300 600 900 1200

Ph

osp

hat

e [1

0-6m

ol k

g-1

]


a)

00

01

02

03

04

05

00

06

12

18

00 02 04 06 08 100

Ph

osp

hat

e [1

0-6m

ol k

g-1

]

Co

nsu

mp

tio

n [

10-6

mo

l kg

-1]

b)









Cruise ttn49


Cruise ttn50









ttn 49

S12 2719 2686 032 76700 23298S11 2728 2720 008 16303 4909S10 2673 2602 070 183139 57077S9 2666 2584 082 214771 67379S8 2688 2639 049 109286 33650S7 2574 2397 176 261963 90695S6 2455 2222 233 623791 271435S5 2589 2373 216 477645 168688S4 2665 2437 227 386403 130107S3 2598 2356 242 612288 219543S2 2297 2137 160 259434 151544

ttn50

S11 2693 2537 155 399021 127506S9 2653 2409 244 772361 265124S8 2610 2318 291 725146 270388S7 2652 2389 263 584430 203671S6 2664 2440 224 468718 157517S5 2564 2238 326 1081012 454687S4 2488 2195 293 534635 249007S3 2382 2159 223 369189 194683S2 2392 2121 271 261293 168481


50

100

150

200CP

Rat

io

a)

10

20

30NP

Rat

io

b)

0

5

10

15

00 02 04 06 08 10

223 POCPIC

Rat

io

c)












0

10

20

30

40

50

60

70

0 300 600 900 1200

CO

2 fl

ux

[mm

ol m

-2 d

ay-1

]

Distance [km]






0

20

40

60

80

100

120

140

160

180

200

220

00 02 04 06 08 10

[mm

ol C

m-2

day

-1]




4 Discussion

41 Diatom blooms






0

5

10

15

20

25

30

2

3

4

5

6

May Jun Jul Aug Sep

1995

PO

C F

lux

[mg

m-2

day

-1]

Car

bo

nat

eB

iog

enic

Op

al

-120

-80

-40

0

Wat

er D

epth

[m

]

ttn49 ttn50

M325M325MLD

Eup

thot

ic z

one




0250500750

100012501500

0

20

40

60

Dia

tom

s [1

03 va

lves

l-1]

Dia

tom

s [1

03 va

lves

l-1]

Co

cco

lith

[10

3 ce

lls l-1

]a)

0

30

60

90

Pla

nkt

on

Co

mp

[

] P

lan

kto

n C

om

p [

]

b)

1

2

3

4

5

6

7

Car

bo

nat

e

Bio

gen

ic O

pal

c)

02468

101214

0 300 600 900 1200

Si -

co

nce

n

[10-6

mo

l kg

-1]

d)

Distance [km]








43 Rain ratios



2

4

6

8

10

Rai

n r

atio

a)

2

3

4

00 02 04 06 08 102

3

4

00 02 04 06 08 10

84

86

88

90

92

00 02 04 06 08 102

3

4

00 02 04 06 08 10



22

Rai

n r

atio

(tr

ap d

ata)

CN

rat

io (

trap

dat

a)

b)







0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io






44 CN and CP ratios



5 Conclusions






Edited by A Watson

References
























































































20

22

24

26

28

355 360 365 370Salinity

Tem

per

atu

re [

oC

]




00

06

12

18

0 300 600 900 1200

Ph

osp

hat

e [1

0-6m

ol k

g-1

]


a)

00

01

02

03

04

05

00

06

12

18

00 02 04 06 08 100

Ph

osp

hat

e [1

0-6m

ol k

g-1

]

Co

nsu

mp

tio

n [

10-6

mo

l kg

-1]

b)









Cruise ttn49


Cruise ttn50









ttn 49

S12 2719 2686 032 76700 23298S11 2728 2720 008 16303 4909S10 2673 2602 070 183139 57077S9 2666 2584 082 214771 67379S8 2688 2639 049 109286 33650S7 2574 2397 176 261963 90695S6 2455 2222 233 623791 271435S5 2589 2373 216 477645 168688S4 2665 2437 227 386403 130107S3 2598 2356 242 612288 219543S2 2297 2137 160 259434 151544

ttn50

S11 2693 2537 155 399021 127506S9 2653 2409 244 772361 265124S8 2610 2318 291 725146 270388S7 2652 2389 263 584430 203671S6 2664 2440 224 468718 157517S5 2564 2238 326 1081012 454687S4 2488 2195 293 534635 249007S3 2382 2159 223 369189 194683S2 2392 2121 271 261293 168481


50

100

150

200CP

Rat

io

a)

10

20

30NP

Rat

io

b)

0

5

10

15

00 02 04 06 08 10

223 POCPIC

Rat

io

c)












0

10

20

30

40

50

60

70

0 300 600 900 1200

CO

2 fl

ux

[mm

ol m

-2 d

ay-1

]

Distance [km]






0

20

40

60

80

100

120

140

160

180

200

220

00 02 04 06 08 10

[mm

ol C

m-2

day

-1]




4 Discussion

41 Diatom blooms






0

5

10

15

20

25

30

2

3

4

5

6

May Jun Jul Aug Sep

1995

PO

C F

lux

[mg

m-2

day

-1]

Car

bo

nat

eB

iog

enic

Op

al

-120

-80

-40

0

Wat

er D

epth

[m

]

ttn49 ttn50

M325M325MLD

Eup

thot

ic z

one




0250500750

100012501500

0

20

40

60

Dia

tom

s [1

03 va

lves

l-1]

Dia

tom

s [1

03 va

lves

l-1]

Co

cco

lith

[10

3 ce

lls l-1

]a)

0

30

60

90

Pla

nkt

on

Co

mp

[

] P

lan

kto

n C

om

p [

]

b)

1

2

3

4

5

6

7

Car

bo

nat

e

Bio

gen

ic O

pal

c)

02468

101214

0 300 600 900 1200

Si -

co

nce

n

[10-6

mo

l kg

-1]

d)

Distance [km]








43 Rain ratios



2

4

6

8

10

Rai

n r

atio

a)

2

3

4

00 02 04 06 08 102

3

4

00 02 04 06 08 10

84

86

88

90

92

00 02 04 06 08 102

3

4

00 02 04 06 08 10



22

Rai

n r

atio

(tr

ap d

ata)

CN

rat

io (

trap

dat

a)

b)







0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io






44 CN and CP ratios



5 Conclusions






Edited by A Watson

References


























































































Cruise ttn49


Cruise ttn50









ttn 49

S12 2719 2686 032 76700 23298S11 2728 2720 008 16303 4909S10 2673 2602 070 183139 57077S9 2666 2584 082 214771 67379S8 2688 2639 049 109286 33650S7 2574 2397 176 261963 90695S6 2455 2222 233 623791 271435S5 2589 2373 216 477645 168688S4 2665 2437 227 386403 130107S3 2598 2356 242 612288 219543S2 2297 2137 160 259434 151544

ttn50

S11 2693 2537 155 399021 127506S9 2653 2409 244 772361 265124S8 2610 2318 291 725146 270388S7 2652 2389 263 584430 203671S6 2664 2440 224 468718 157517S5 2564 2238 326 1081012 454687S4 2488 2195 293 534635 249007S3 2382 2159 223 369189 194683S2 2392 2121 271 261293 168481


50

100

150

200CP

Rat

io

a)

10

20

30NP

Rat

io

b)

0

5

10

15

00 02 04 06 08 10

223 POCPIC

Rat

io

c)












0

10

20

30

40

50

60

70

0 300 600 900 1200

CO

2 fl

ux

[mm

ol m

-2 d

ay-1

]

Distance [km]






0

20

40

60

80

100

120

140

160

180

200

220

00 02 04 06 08 10

[mm

ol C

m-2

day

-1]




4 Discussion

41 Diatom blooms






0

5

10

15

20

25

30

2

3

4

5

6

May Jun Jul Aug Sep

1995

PO

C F

lux

[mg

m-2

day

-1]

Car

bo

nat

eB

iog

enic

Op

al

-120

-80

-40

0

Wat

er D

epth

[m

]

ttn49 ttn50

M325M325MLD

Eup

thot

ic z

one




0250500750

100012501500

0

20

40

60

Dia

tom

s [1

03 va

lves

l-1]

Dia

tom

s [1

03 va

lves

l-1]

Co

cco

lith

[10

3 ce

lls l-1

]a)

0

30

60

90

Pla

nkt

on

Co

mp

[

] P

lan

kto

n C

om

p [

]

b)

1

2

3

4

5

6

7

Car

bo

nat

e

Bio

gen

ic O

pal

c)

02468

101214

0 300 600 900 1200

Si -

co

nce

n

[10-6

mo

l kg

-1]

d)

Distance [km]








43 Rain ratios



2

4

6

8

10

Rai

n r

atio

a)

2

3

4

00 02 04 06 08 102

3

4

00 02 04 06 08 10

84

86

88

90

92

00 02 04 06 08 102

3

4

00 02 04 06 08 10



22

Rai

n r

atio

(tr

ap d

ata)

CN

rat

io (

trap

dat

a)

b)







0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io






44 CN and CP ratios



5 Conclusions






Edited by A Watson

References


























































































ttn 49

S12 2719 2686 032 76700 23298S11 2728 2720 008 16303 4909S10 2673 2602 070 183139 57077S9 2666 2584 082 214771 67379S8 2688 2639 049 109286 33650S7 2574 2397 176 261963 90695S6 2455 2222 233 623791 271435S5 2589 2373 216 477645 168688S4 2665 2437 227 386403 130107S3 2598 2356 242 612288 219543S2 2297 2137 160 259434 151544

ttn50

S11 2693 2537 155 399021 127506S9 2653 2409 244 772361 265124S8 2610 2318 291 725146 270388S7 2652 2389 263 584430 203671S6 2664 2440 224 468718 157517S5 2564 2238 326 1081012 454687S4 2488 2195 293 534635 249007S3 2382 2159 223 369189 194683S2 2392 2121 271 261293 168481


50

100

150

200CP

Rat

io

a)

10

20

30NP

Rat

io

b)

0

5

10

15

00 02 04 06 08 10

223 POCPIC

Rat

io

c)












0

10

20

30

40

50

60

70

0 300 600 900 1200

CO

2 fl

ux

[mm

ol m

-2 d

ay-1

]

Distance [km]






0

20

40

60

80

100

120

140

160

180

200

220

00 02 04 06 08 10

[mm

ol C

m-2

day

-1]




4 Discussion

41 Diatom blooms






0

5

10

15

20

25

30

2

3

4

5

6

May Jun Jul Aug Sep

1995

PO

C F

lux

[mg

m-2

day

-1]

Car

bo

nat

eB

iog

enic

Op

al

-120

-80

-40

0

Wat

er D

epth

[m

]

ttn49 ttn50

M325M325MLD

Eup

thot

ic z

one




0250500750

100012501500

0

20

40

60

Dia

tom

s [1

03 va

lves

l-1]

Dia

tom

s [1

03 va

lves

l-1]

Co

cco

lith

[10

3 ce

lls l-1

]a)

0

30

60

90

Pla

nkt

on

Co

mp

[

] P

lan

kto

n C

om

p [

]

b)

1

2

3

4

5

6

7

Car

bo

nat

e

Bio

gen

ic O

pal

c)

02468

101214

0 300 600 900 1200

Si -

co

nce

n

[10-6

mo

l kg

-1]

d)

Distance [km]








43 Rain ratios



2

4

6

8

10

Rai

n r

atio

a)

2

3

4

00 02 04 06 08 102

3

4

00 02 04 06 08 10

84

86

88

90

92

00 02 04 06 08 102

3

4

00 02 04 06 08 10



22

Rai

n r

atio

(tr

ap d

ata)

CN

rat

io (

trap

dat

a)

b)







0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io






44 CN and CP ratios



5 Conclusions






Edited by A Watson

References
























































































0

10

20

30

40

50

60

70

0 300 600 900 1200

CO

2 fl

ux

[mm

ol m

-2 d

ay-1

]

Distance [km]






0

20

40

60

80

100

120

140

160

180

200

220

00 02 04 06 08 10

[mm

ol C

m-2

day

-1]




4 Discussion

41 Diatom blooms






0

5

10

15

20

25

30

2

3

4

5

6

May Jun Jul Aug Sep

1995

PO

C F

lux

[mg

m-2

day

-1]

Car

bo

nat

eB

iog

enic

Op

al

-120

-80

-40

0

Wat

er D

epth

[m

]

ttn49 ttn50

M325M325MLD

Eup

thot

ic z

one




0250500750

100012501500

0

20

40

60

Dia

tom

s [1

03 va

lves

l-1]

Dia

tom

s [1

03 va

lves

l-1]

Co

cco

lith

[10

3 ce

lls l-1

]a)

0

30

60

90

Pla

nkt

on

Co

mp

[

] P

lan

kto

n C

om

p [

]

b)

1

2

3

4

5

6

7

Car

bo

nat

e

Bio

gen

ic O

pal

c)

02468

101214

0 300 600 900 1200

Si -

co

nce

n

[10-6

mo

l kg

-1]

d)

Distance [km]








43 Rain ratios



2

4

6

8

10

Rai

n r

atio

a)

2

3

4

00 02 04 06 08 102

3

4

00 02 04 06 08 10

84

86

88

90

92

00 02 04 06 08 102

3

4

00 02 04 06 08 10



22

Rai

n r

atio

(tr

ap d

ata)

CN

rat

io (

trap

dat

a)

b)







0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io






44 CN and CP ratios



5 Conclusions






Edited by A Watson

References
























































































0

5

10

15

20

25

30

2

3

4

5

6

May Jun Jul Aug Sep

1995

PO

C F

lux

[mg

m-2

day

-1]

Car

bo

nat

eB

iog

enic

Op

al

-120

-80

-40

0

Wat

er D

epth

[m

]

ttn49 ttn50

M325M325MLD

Eup

thot

ic z

one




0250500750

100012501500

0

20

40

60

Dia

tom

s [1

03 va

lves

l-1]

Dia

tom

s [1

03 va

lves

l-1]

Co

cco

lith

[10

3 ce

lls l-1

]a)

0

30

60

90

Pla

nkt

on

Co

mp

[

] P

lan

kto

n C

om

p [

]

b)

1

2

3

4

5

6

7

Car

bo

nat

e

Bio

gen

ic O

pal

c)

02468

101214

0 300 600 900 1200

Si -

co

nce

n

[10-6

mo

l kg

-1]

d)

Distance [km]








43 Rain ratios



2

4

6

8

10

Rai

n r

atio

a)

2

3

4

00 02 04 06 08 102

3

4

00 02 04 06 08 10

84

86

88

90

92

00 02 04 06 08 102

3

4

00 02 04 06 08 10



22

Rai

n r

atio

(tr

ap d

ata)

CN

rat

io (

trap

dat

a)

b)







0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io






44 CN and CP ratios



5 Conclusions






Edited by A Watson

References


























































































43 Rain ratios



2

4

6

8

10

Rai

n r

atio

a)

2

3

4

00 02 04 06 08 102

3

4

00 02 04 06 08 10

84

86

88

90

92

00 02 04 06 08 102

3

4

00 02 04 06 08 10



22

Rai

n r

atio

(tr

ap d

ata)

CN

rat

io (

trap

dat

a)

b)







0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io






44 CN and CP ratios



5 Conclusions






Edited by A Watson

References
























































































0

5

10

15

20

25

00 02 04 06 08 10

CN

rat

io






44 CN and CP ratios



5 Conclusions






Edited by A Watson

References

























































































Edited by A Watson

References






























































































































































Documents

Diatoms and their inﬂuence on the biologically mediated uptake … · 2016-01-12 · diatom blooms in a silicon-enriched near shore upwelling system off Oman. In the open western