scieee Science in your language
[en] (orig)

A compilation of global bio-optical in situ data for ocean-colour satellite applications – version two

Abstract

A global compilation of in situ data is useful to evaluate the quality of ocean-colour satellite data records. Here we describe the data compiled for the validation of the ocean-colour products from the ESA Ocean Colour Climate Change Initiative (OC-CCI). The data were acquired from several sources (including, inter alia, MOBY, BOUSSOLE, AERONET-OC, SeaBASS, NOMAD, MERMAID, AMT, ICES, HOT and GeP&CO) and span the period from 1997 to 2018. Observations of the following variables were compiled: spectral remote-sensing reflectances, concentrations of chlorophyll a, spectral inherent optical properties, spectral diffuse attenuation coefficients and total suspended matter. The data were from multi-project archives acquired via open internet services or from individual projects, acquired directly from data providers. Methodologies were implemented for homogenization, quality control and merging of all data. No changes were made to the original data, other than averaging of observations that were close in time and space, elimination of some points after quality control and conversion to a standard format. The final result is a merged table designed for validation of satellite-derived ocean-colour products and available in text format. Metadata of each in situ measurement (original source, cruise or experiment, principal investigator) was propagated throughout the work and made available in the final table. By making the metadata available, provenance is better documented, and it is also possible to analyse each set of data separately. This paper also describes the changes that were made to the compilation in relation to the previous version (Valente et al., 2016). The compiled data are available at https://doi.org/10.1594/PANGAEA.898188 (Valente et al., 2019).

Read accessible full text

A compilation of global bio-optical in situ data for ocean-colour satellite applications – version two

Author: Valente, André,Sathyendranath, Shubha,Brotas, Vanda,Groom, Steve,Grant, Michael,Taberner, Malcolm,Antoine, David,Arnone, Robert,Balch, William M.,Barker, Kathryn,Barlow, Ray,Bélanger, Simon,Berthon, Jean-François,Beşiktepe, Şükrü,Borsheim, Yngve,Bracher,
Publisher: Copernicus Publications
Year: 2019
Source: https://repositorio.ulisboa.pt/bitstream/10451/41158/1/MARE_ULisboa_EarthSystemScienceData_11_2019_1037-1068.pdf
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019
h ps://doi.o g/10.5194/essd-11-1037-2019
© Au ho (s) 2019. This wo k is dis ibu ed unde
he C ea i e Commons A ibu ion 4.0 License.
A compila ion o global bio-op ical in si u da a o
ocean-colou sa elli e applica ions – e sion wo
And é Valen e1, Shubha Sa hyend ana h2, Vanda B o as1,2, S e e G oom2, Michael G an 2,3,
Malcolm Tabe ne 3, Da id An oine4,5, Robe A none6, William M. Balch7, Ka h yn Ba ke 8,9,10,
Ray Ba low11, Simon Bélange 12, Jean-F ançois Be hon13, ¸Sük ü Be¸sik epe14, Yng e Bo sheim15,
As id B ache 16,17, Vi o io B ando9,18, Elisabe a Canu i13, F ancisco Cha ez19, And és Cianca20,
He é Claus e4, Lesley Clemen son9, Richa d C ou 21, Robe F ouin22, Ca los Ga cía-So o23,24,
S ua W. Gibb25, Richa d Gould21, S an o d B. Hooke 26, Ma i Kah u22, Mil on Kampel27,
Holge Klein28, Susanne K a ze 29, Raphael Kudela30, Jesus Ledesma31, Hube Loisel32,
Pa icia Ma ai7, Da id McKee33, B ian G. Mi chell22, Ti any Moisan34,†, F ank Mulle -Ka ge 35,
Leonie O’Dowd36, Michael Ond usek37, T e o Pla 2, Alex J. Poul on38, Michel Repecaud39,
Thomas Sch oede 9, Timo hy Smy h2, Denise Smy he-W igh 40, Heidi M. Sosik41,
Michael Twa dowski42, Vincenzo Vellucci4, Kenne h Voss43, Je emy We dell26, Ma cel We nand44,†,
Simon W igh 45, and Giuseppe Zibo di13
1MARE – Ma ine and En i onmen al Sciences Cen e, Faculdade de Ciências,
Uni e sidade de Lisboa, Campo G ande, 1749-016 Lisbon, Po ugal
2Plymou h Ma ine Labo a o y, Plymou h, PL1 3DH, UK
3EUMETSAT, Eume sa -Allee 1, 64295 Da ms ad , Ge many
4So bonne Uni e si é, CNRS, Labo a oi e d’Océanog aphie de Ville anche, LOV,
06230 Ville anche-su -Me , F ance
5Remo e Sensing and Sa elli e Resea ch G oup, School o Ea h and Plane a y Sciences,
Cu in Uni e si y, Pe h, WA 6845, Aus alia
6Uni e si y o Sou he n Mississippi, S ennis Space Cen e , MS, USA
7Bigelow Labo a o y o Ocean Sciences, 60 Bigelow D ., Eas Boo hbay, ME 04544, USA
8ARGANS L d, Plymou h, UK
9CSIRO Oceans and A mosphe e, Pe h, Wes e n Aus alia, Aus alia
10Aus alian Resea ch Da a Commons, Caul ield Eas , Aus alia
11Baywo ld Cen e o Resea ch and Educa ion, Cape Town, Sou h A ica
12Uni e si é du Québec à Rimouski, Rimouski, Quebec, Canada
13Eu opean Commission, Join Resea ch Cen e, Isp a, I aly
14Dokuz Eylul Uni e si y, Ins i u e o Ma ine Science and Technology, Izmi , Tu key
15Ins i u e o Ma ine Resea ch, Be gen, No way
16Al ed Wegene Ins i u e Helmhol z Cen e o Pola and Ma ine Resea ch, B eme ha en, Ge many
17Ins i u e o En i onmen al Physics, Uni e si y B emen, B emen, Ge many
18CNR – ISMAR, Rome, I aly
19Mon e ey Bay Aqua ium Resea ch Ins i u e, Moss Landing, CA, USA
20PLOCAN – Oceanic Pla o m o he Cana y Islands, Ca e e a de Talia e, 35214 Telde, G an Cana ia, Spain
21Na al Resea ch Labo a o y, S ennis Space Cen e , MS, USA
22Sc ipps Ins i u ion o Oceanog aphy, Uni e si y o Cali o nia San Diego, CA, USA
23Spanish Ins i u e o Oceanog aphy (IEO), Co azón de Ma ía 8, 28002 Mad id, Spain
24Plen ziako I sas Es azioa/Euskal He iko Unibe si a ea (PIE/EHU), A ea za z/g, 48620 Plen zia, Spain
25En i onmen al Resea ch Ins i u e, No h Highland College,
Uni e si y o he Highlands and Islands, Thu so, Sco land, UK
26NASA Godda d Space Fligh Cen e , G eenbel , MD, USA
Published by Cope nicus Publica ions.
1038 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
27Remo e Sensing Di ision, Na ional Space Resea ch Ins i u e (INPE), Sao Jose dos Campos, B azil
28Ope a ional Oceanog aphy G oup, Fede al Ma i ime and Hyd og aphic Agency, Hambu g, Ge many
29Depa men o Ecology, En i onmen and Plan Sciences, S ockholm Uni e si y, 106 91 S ockholm, Sweden
30Uni e si y o Cali o nia San a C uz, San a C uz, CA, USA
31Ins i u o del Ma del Pe ú, Callao, Pe u
32Labo a oi e d’Océanologie e de Géosciences, Uni e si é du Li o al-Cô e-d’Opale, Uni e si é Lille,
CNRS, UMR 8187, LOG, 32 a enue Foch, Wime eux, F ance
33Physics Depa men , Uni e si y o S a hclyde, Glasgow, G4 0NG, Sco land, UK
34NASA Godda d Space Fligh Cen e , Wallops Fligh Facili y, Wallops Island, VA, USA
35Ins i u e o Ma ine Remo e Sensing/ImaRS, College o Ma ine Science,
Uni e si y o Sou h Flo ida, S Pe e sbu g, FL, USA
36Fishe ies and Ecosys em Ad iso y Se ices, Ma ine Ins i u e, Rin ille – O anmo e, Galway, I eland
37NOAA/NESDIS/STAR/SOCD, College Pa k, MD, USA
38Lyell Cen e o Ea h and Ma ine Science and Technology, He io -Wa Uni e si y, Edinbu gh, UK
39IFREMER Cen e de B es , Plouzane, F ance
40Ocean Biogeochemis y and Ecosys ems, Na ional Oceanog aphy Cen e,
Wa e on Campus, Sou hamp on, UK
41Biology Depa men , Woods Hole Oceanog aphic Ins i u ion, Woods Hole, MA, USA
42Ha bo B anch Oceanog aphic Ins i u e, Fo Pie ce, FL, USA
43Uni e si y o Miami, Co al Gables, FL, USA
44Royal Ne he lands Ins i u e o Sea Resea ch, Texel, he Ne he lands
45Aus alian An a c ic Di ision and he An a c ic Clima e and Ecosys ems Coope a i e Resea ch Cen e,
Hoba , Aus alia
†deceased
Co espondence: And é Valen e (ado [email p o ec ed])
Recei ed: 8 Feb ua y 2019 – Discussion s a ed: 18 Feb ua y 2019
Re ised: 24 June 2019 – Accep ed: 25 June 2019 – Published: 15 July 2019
Abs ac . A global compila ion o in si u da a is use ul o e alua e he quali y o ocean-colou sa elli e da a
eco ds. He e we desc ibe he da a compiled o he alida ion o he ocean-colou p oduc s om he ESA Ocean
Colou Clima e Change Ini ia i e (OC-CCI). The da a we e acqui ed om se e al sou ces (including, in e alia,
MOBY, BOUSSOLE, AERONET-OC, SeaBASS, NOMAD, MERMAID, AMT, ICES, HOT and GeP&CO)
and span he pe iod om 1997 o 2018. Obse a ions o he ollowing a iables we e compiled: spec al emo e-
sensing e lec ances, concen a ions o chlo ophyll a, spec al inhe en op ical p ope ies, spec al di use a enu-
a ion coe icien s and o al suspended ma e . The da a we e om mul i-p ojec a chi es acqui ed ia open in e -
ne se ices o om indi idual p ojec s, acqui ed di ec ly om da a p o ide s. Me hodologies we e implemen ed
o homogeniza ion, quali y con ol and me ging o all da a. No changes we e made o he o iginal da a, o he
han a e aging o obse a ions ha we e close in ime and space, elimina ion o some poin s a e quali y con ol
and con e sion o a s anda d o ma . The inal esul is a me ged able designed o alida ion o sa elli e-de i ed
ocean-colou p oduc s and a ailable in ex o ma . Me ada a o each in si u measu emen (o iginal sou ce, c uise
o expe imen , p incipal in es iga o ) was p opaga ed h oughou he wo k and made a ailable in he inal able.
By making he me ada a a ailable, p o enance is be e documen ed, and i is also possible o analyse each se o
da a sepa a ely. This pape also desc ibes he changes ha we e made o he compila ion in ela ion o he p e i-
ous e sion (Valen e e al., 2016). The compiled da a a e a ailable a h ps://doi.o g/10.1594/PANGAEA.898188
(Valen e e al., 2019).
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/
A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo 1039
1 In oduc ion
Cu en ly, he e a e se e al se s o in si u bio-op ical da a,
wo ldwide, sui able o alida ion o ocean-colou sa elli e
da a. Whe eas some a e managed by he da a p oduce s, o h-
e s a e in in e na ional eposi o ies wi h con ibu ions om
mul iple scien is s. Many ha e igid quali y con ols and a e
buil speci ically o ocean-colou alida ion. The use o only
any one o hese da ase s would limi he numbe o da a
in alida ion exe cises. I is, he e o e, i al o acqui e and
me ge all hese da ase s in o a single uni ied da ase o max-
imize he numbe o ma chups a ailable o alida ion, hei
dis ibu ion in ime and space, and, consequen ly, o educe
unce ain ies in he alida ion exe cise. Howe e , me ging
se e al da ase s oge he can be a complica ed ask. Fi s i is
necessa y o acqui e and ha monize all da ase s in o a single
s anda d o ma . Second, du ing he me ging, duplica es be-
ween da ase s ha e o be iden i ied and emo ed. Thi d, he
me ada a should be p opaga ed h oughou he p ocess and
made a ailable in he inal me ged p oduc . Ideally, he com-
piled da ase would be made a ailable as a simple ex able,
o acili a e ease o access and manipula ion. In his wo k
such uni ica ion o mul iple da ase s is p esen ed. This was
done o he alida ion o he ocean-colou p oduc s om he
ESA Ocean Colou Clima e Change Ini ia i e (OC-CCI), bu
wi h he in en o se e he b oade use communi y as well.
A me ged da ase is no wi hou d awbacks: i is likely o
be la ge and so no always easy o manipula e; because he
me ging is done on p e-exis ing, p ocessed da abases, i is no
possible o ha e ull con ol o he whole p ocessing chain;
he da ase would be a compila ion o obse a ions collec ed
by se e al in es iga o s using di e en ins umen s, sampling
me hods and p o ocols, which migh e en ually ha e been
modi ied by he p ocessing ou ines used by he eposi o-
ies o a chi es. To minimize hese po en ial d awbacks, we
ha e, o he mos pa , inco po a ed only da ase s ha ha e
eme ged om he long- e m e o s o he ocean-colou and
biological oceanog aphical communi ies o p o ide scien is s
wi h high-quali y in si u da a, and we implemen ed addi ional
quali y checks on he da a o enhance con idence in he qual-
i y o he me ged p oduc . Ne e heless, i is s ill ecognized
ha di e en and unp edic able unce ain ies may a ec da a
om he di e se sou ces as a esul o he applica ion o a
a ie y o ield/labo a o y ins umen s, me hods and da a e-
duc ion schemes.
In Sec . 2 he me hodologies used o ha monize and in e-
g a e all da a, as well as a desc ip ion o indi idual da ase s
acqui ed, a e p o ided. In Sec . 3 he geog aphic dis ibu-
ion and o he cha ac e is ics o he inal me ged da ase a e
shown. Sec ion 4 p o ides an o e iew o he da a.
2 Da a and me hods
2.1 P ep ocessing and me ging
The compiled global se o bio-op ical in si u da a de-
sc ibed in his wo k has an emphasis, hough no exclu-
si e, on open-ocean da a. I comp ises he ollowing a i-
ables: emo e-sensing e lec ance ( s), chlo ophyll acon-
cen a ion (chla), algal pigmen abso p ion coe icien (aph),
de i al and colou ed dissol ed o ganic ma e abso p ion co-
e icien (adg), pa icle backsca e ing coe icien (bbp), di -
use a enua ion coe icien o downwa d i adiance (kd)
and o al suspended ma e ( sm). The a iables s, aph,
adg, bbp and kd a e spec ally dependen , and his depen-
dence is, he ea e , implied. The da a we e compiled om 27
sou ces (MOBY, BOUSSOLE, AERONET-OC, SeaBASS,
NOMAD, MERMAID, AMT, ICES, HOT, GeP&CO, AWI,
ARCSSPP, BARENTSSEA, BATS, BIOCHEM, BODC,
CALCOFI, CCELTER, CIMT, COASTCOLOUR, ESTOC,
IMOS, MAREDAT, PALMER, SEADATANET, TPSS and
TARA): each one desc ibed in Sec . 2.2. The da a sou ces in
his wo k should also be iewed as g oups o da a ha we e
acqui ed om a speci ic sou ce, s anda dized wi h a speci ic
me hod and la e me ged in o he compila ion. The compiled
in si u obse a ions ha e a global dis ibu ion and co e he
pe iod 1997 o 2018. The lis ed a iables, wi h he excep-
ion o o al suspended ma e , we e chosen as hey a e he
ope a ional sa elli e ocean-colou p oduc s o he ESA OC-
CCI p ojec , which cu en ly ocuses on he me ging o ou
ocean-colou sa elli e senso s: he Medium Resolu ion Imag-
ing Spec ome e (MERIS) o ESA, he Mode a e Resolu ion
Imaging Spec o adiome e (MODIS) o NASA, he Sea-
iewing Wide Field-o - iew Senso (SeaWiFS) o NASA,
and he Visible In a ed Imaging Radiome e Sui e (VIIRS)
o NASA and he Na ional Oceanic and A mosphe ic Admin-
is a ion (NOAA) o c ea e a ime se ies o sa elli e da a.
This is he second e sion o he compila ion o global
bio-op ical in si u da a desc ibed by Valen e e al. (2016).
A ack-change ile o he manusc ip o he i s e sion can
be ound in he Supplemen . The new e sion has mo e da a
and a highe empo al and spa ial co e age. The inc eases
in he numbe o obse a ions a e mainly o chla, s and
aph. In compa ison wi h Valen e e al. (2016), he obse a-
ions o chla and aph ha e doubled in numbe and p o ide a
be e spa ial co e age, especially in he Sou he n and A c ic
Ocean. The s alues also inc eased in numbe , bu no as
much in spa ial co e age, because mos o he new obse a-
ions came om ixed loca ions.
The p esen second e sion is a compila ion o da a
om sou ces used in he i s e sion (MOBY, BOUS-
SOLE, AERONET-OC, SeaBASS, NOMAD, MERMAID,
AMT, ICES, HOT and GeP&CO) plus da a om addi-
ional sou ces (AWI, ARCSSPP, BARENTSSEA, BATS,
BIOCHEM, BODC, CALCOFI, CCELTER, CIMT, COAST-
COLOUR, ESTOC, IMOS, MAREDAT, PALMER, SEA-
www.ea h-sys -sci-da a.ne /11/1037/2019/ Ea h Sys . Sci. Da a, 11, 1037–1068, 2019
1040 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
DATANET, TPSS and TARA). The main di e ences om
he i s e sion a e (1) some o he da a sou ces used in he
i s e sion we e upda ed (MOBY, AERONET, SeaBASS
and HOT), (2) new da a sou ces we e added, (3) a new a i-
able was compiled ( o al suspended ma e ), (4) he o ma o
he da abase was modi ied and (5) wo new lags we e added.
Conce ning he change in o ma , in Valen e e al. (2016)
he compila ion was p o ided as one unique wo-dimensional
able. Now, gi en i s inc eased size (136 250 ows and 1286
columns compa ed wi h 80 524 ows and 267 columns p e-
iously), he able has been b oken in o h ee smalle ables
ha ela e o each o he ia one unique key iden i ying each
ow. One addi ional able is also p o ided o help wi h da a
manipula ion. Despi e his change, he compila ion should
s ill be iewed concep ually as one unique able, and as such,
i is s ill desc ibed in ha way. In he p esen e sion, wo
lags we e added: lag_ ime and lag_chl_me hod. The i s
is because in he p esen e sion h ee da a sou ces we e
used (ESTOC, MAREDAT and TPSS) whe e in o ma ion on
ime (hou o he day) was no a ailable. The ime o hese
obse a ions was se o 12:00:00 UTC and he obse a ions
we e lagged wi h “1” in he column lag_ ime. A second
lag was necessa y, because in wo da a sou ces (ARCSSPP
and SEADATANET) he e was unce ain y on whe he he
compiled chlo ophyll concen a ions we e measu ed using
luo ome ic, spec opho ome ic o high-pe o mance liq-
uid ch oma og aphy (HPLC) me hods. The compiled chlo o-
phyll obse a ions om hese wo da a sou ces we e lagged
wi h “1” in he column lag_chl_me hod and we e ma ked as
chla_ luo .
Remo e-sensing e lec ance ( s) is a p ima y ocean-colou
p oduc de ined as s =Lw/Es, whe e Lw is he upwa d
wa e -lea ing adiance and Es is he o al downwa d i adi-
ance a sea le el. Ano he quan i y ha is o en equi ed is
he no malized wa e -lea ing adiance (nLw) (Go don and
Cla k, 1981), which is ela ed o emo e-sensing e lec ance
ia s =nLw/Fo, whe e Fo is he op-o - he-a mosphe e so-
la i adiance. I no di ec ly a ailable, emo e-sensing e-
lec ance was calcula ed h ough he equa ions desc ibed
abo e, depending on he o ma o he o iginal da a. The
o iginal da a we e acqui ed in an ad anced o m (e.g. ime-
a e aged, ex apola ed o su ace) om nine da a sou ces de-
signed o ocean-colou alida ion and applica ions (MOBY,
BOUSSOLE, AERONET-OC, SeaBASS, NOMAD, MER-
MAID, COASTCOLOUR, TARA, AWI), he e o e only e-
qui ing he con e sion o a common o ma . In he p ocessing
made by he space agencies, he quan i y s is no malized
o a single Sun- iewing geome y (Sun a zeni h and nadi
iewing) aking in accoun he bidi ec ional e ec s as de-
sc ibed in Mo el and Gen ili (1996) and Mo el e al. (2002).
Thus, o consis ency wi h sa elli e s p oduc , he la e no -
maliza ion was applied o he in si u s.
Chlo ophyll aconcen a ion is he con en ional measu e
o phy oplank on biomass and one o he mos widely used
sa elli e ocean-colou p oduc s (IOCCG, 2008). To alida e
sa elli e-de i ed chlo ophyll aconcen a ion, wo di e en
a iables we e compiled: one o hese ep esen s chlo o-
phyll ameasu emen s made h ough luo ome ic o spec-
opho ome ic me hods, e e ed o he ea e as chla_ luo
and he o he is he chlo ophyll concen a ion de i ed om
HPLC measu emen s, e e ed o he ea e as chla_hplc. The
chlo ophyll da a we e compiled om he ollowing 25 da a
sou ces: BOUSSOLE, SeaBASS, NOMAD, MERMAID,
AMT, ICES, HOT, GeP&CO, AWI, ARCSSPP, BAR-
ENTSSEA, BATS, BIOCHEM, BODC, CALCOFI, CCEL-
TER, CIMT, COASTCOLOUR, ESTOC, IMOS, MARE-
DAT, PALMER, SEADATANET, TPSS and TARA. One e-
qui emen o chla_ luo measu emen s was ha hey we e
made using in i o me hods (i.e. based on ex ac ions o
chlo ophyll a). Al hough his se e ely dec eased he num-
be o obse a ions, since in si u luo ome y (e.g. luo ome-
e s moun ed on CTDs) is widely a ailable in oceanog aphic
da abases, i was decided o exclude such da a because o
po en ial p oblems wi h he calib a ion o in si u luo om-
e e da a. The a iable chla_hplc was calcula ed by sum-
ming all epo ed chlo ophyll ade i a i es, including di-
inyl chlo ophyll a, epime s, allome s and chlo ophyllide a.
The wo chlo ophyll a iables a e e ained sepa a ely in he
da abase o acili a e hei use. HPLC measu emen s could
be conside ed o highe quali y, bu luo ome ic measu e-
men s a e mo e nume ous. Thus one op ion o use s is o use
chla_ luo only when he e a e no chla_hplc measu emen s
a ailable. To be consis en wi h sa elli e-de i ed chlo ophyll
alues, which a e de i ed om he ligh eme ging om he
uppe laye o he ocean, all chlo ophyll obse a ions in he
op 10 m ( eplica es a he same dep h, o measu emen s a
mul iple dep hs) we e a e aged i he coe icien o a ia-
ion among obse a ions was less han 50 %, o he wise hey
we e disca ded. The a e ages we e hen assigned o he su -
ace. The dep h o 10 m was chosen as a comp omise be-
ween clea oligo ophic and u bid eu ophic wa e s. O he
me hods, such as chlo ophyll dep h a e ages using local a -
enua ion condi ions (Mo el and Ma i o ena, 2001), equi e
obse a ions a mul iple dep hs, which, gi en ou decision o
use only in i o measu emen s, would ha e educed consid-
e ably he inal numbe o obse a ions.
Wi h ega d o he inhe en op ical p ope ies (aph, adg,
bbp), i no al eady calcula ed and p o ided in he con-
ibu ed da ase s, hey we e compu ed om ela ed a i-
ables ha we e a ailable: pa icle abso p ion (ap), de i al
abso p ion (ad), colou ed dissol ed o ganic ma e (CDOM)
abso p ion (ag) and o al backsca e ing (bb). The ollow-
ing equa ions we e used: adg =ad +ag, ap =aph +ad, and
bb =bbp +bbw. Fo he la e equa ion, he a iable bbw
was compu ed using bbw =bw/2, whe e bw is he sca e -
ing coe icien o seawa e de i ed om Zhang e al. (2009).
The di use a enua ion coe icien o downwa d i adiance
(kd) did no equi e any con e sion and was compiled as
o iginally acqui ed. Obse a ions o inhe en op ical p ope -
ies (su ace alues) and he di use a enua ion coe icien
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/
A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo 1041
o downwa d i adiance we e acqui ed in o al om six
da a sou ces designed o ocean-colou alida ion and ap-
plica ions (SeaBASS, NOMAD, MERMAID, AWI, COAST-
COLOUR, TPSS), hus al eady subjec o he p ocessing ou-
ines o hese da ase s. Conce ning o al suspended ma e ,
hese da a we e compiled as o iginally a ailable om MER-
MAID and COASTCOLOUR.
The me ged da ase was compiled om 27 se s o in
si u da a, which we e ob ained indi idually ei he om
a chi es ha inco po a e da a om mul iple con ibu-
o s (SeaBASS, NOMAD, MERMAID, ICES, ARCSSPP,
BIOCHEM, BODC, COASTCOLOUR, MAREDAT, SEA-
DATANET) o om pa icula con ibu o s, measu emen
p og ammes o p ojec s (MOBY, BOUSSOLE, AERONET-
OC, HOT, GeP&CO, AMT, AWI, BARENTSSEA, BATS,
CALCOFI, CCELTER, CIMT, ESTOC, IMOS, PALMER,
TPSS, TARA) and we e subsequen ly homogenized and
me ged. Da a con ibu o s a e lis ed in Table 2. The e we e
me hodological di e ences be ween da ase s. The e o e, a -
e acquisi ion, and p io o any me ging, each se o da a
was p ep ocessed o quali y con ol and con e ed o a com-
mon o ma . Du ing his p ocess, da a we e disca ded i hey
had (1) un ealis ic o missing da e and geog aphic coo di-
na e ields, (2) poo quali y (e.g. o iginal lags) o me hod
o obse a ion ha did no mee he c i e ia o he da ase
(e.g. in si u luo escence o chlo ophyll concen a ion), and
(3) spu iously high o low da a. Fo he las i em, he ollow-
ing limi s we e imposed: [0.001–100] mg m−3 o chla_ luo
and chla_hplc; [0–0.15] s −1 o s; [0.0001–10] m−1 o
aph, adg and bbp; [0–1000] g m−3 o sm; and [aw(λ)−
10] m−1 o kd, whe e aw is he pu e wa e abso p ion coe -
icien s de i ed om Pope and F y (1997). Also du ing his
s age, h ee me ada a s ings we e a ibu ed o each obse -
a ion: da ase , subda ase and con ibu o . The da ase con-
ains he name o he o iginal se o da a and can only be
one o he ollowing: aoc, boussole, me maid, moby, nomad,
seabass, ho , ices, am , gepco, a csspp, awi, ba en ssea, ba s,
biochem, bodc, calco i, cc, ccel e , cim , es oc, imos, ma e-
da , palme , seada ane , pss o a a. The subda ase s a s
wi h he da ase iden i ie and is ollowed by addi ional in o -
ma ion abou he da a, as <da ase >_<c uise/s a ion/si e>
(e.g. seabass_ca 81). The con ibu o con ains he name o
he da a con ibu o . An e o was made o homogenize he
names o da a con ibu o s om he di e en se s o da a.
These h ee me ada a a e he link o ace each obse a ion
o i s o igin and we e p opaga ed h oughou he p ocessing.
Finally, his p ocessing s age ended wi h each se o da a be-
ing scanned o eplica e a iable da a and eplica e s a ion
da a, which when ound we e a e aged i he coe icien o
a ia ion was less han 50 %, o he wise hey we e disca ded.
Replica es we e de ined as mul iple obse a ions o he same
a iable, wi h he same da e, ime, la i ude, longi ude and
dep h. Replica e s a ion da a we e de ined as mul iple mea-
su emen s o he same a iable, wi h he same da e, ime, la -
i ude and longi ude. Fo he la e case, a sea ch window o
5 min in ime and 200 m in dis ance was gi en o accoun o
s a ion d i . A small numbe o obse a ions ha we e iden-
i ied as eplica es had di e en subda ase iden i ie s (i.e.
di e en c uise names). These obse a ions we e conside ed
suspicious i he alues we e di e en and disca ded. I he
alues we e he same, one o he obse a ions was e ained.
This possibly o igina ed om he same g oup o da a being
con ibu ed o an a chi e by wo di e en da a con ibu o s.
Once a se o da a was homogenized, i s da a we e in-
eg a ed in o a unique able. This inal me ging ocused
on he emo al o duplica es be ween he se s o da a. Al-
hough some duplica es a e known (e.g. MOBY, BOUS-
SOLE, AERONET-OC and NOMAD da a a e ound in
SeaBASS and MERMAID), o he s a e unknown (e.g. how
many da a o GeP&CO, ICES, AMT and HOT a e wi hin
NOMAD, SeaBASS and MERMAID). The e o e, dupli-
ca es we e iden i ied using he me ada a (da ase and sub-
da ase ) when possible and empo al–spa ial ma ches, as an
addi ional p ecau ion. Fo empo al–spa ial ma ches, se e al
h esholds we e used, bu ypically 5 min and 200 m we e
aken o be su icien o iden i y mos duplica ed da a, which
e lec ed small di e ences in ime, la i ude and longi ude,
be ween he di e en se s o da a. La ge h esholds we e
used in some cases as a cau iona y p ocedu e. This was he
case when sea ching o NOMAD da a in o he da ase s, be-
cause NOMAD includes a ew cases whe e me ging o a-
diome ic and pigmen da a was done wi h la ge spa ial–
empo al h esholds (We dell and Bailey, 2005). A la ge em-
po al h eshold was also used when in eg a ing obse a ions
om he h ee da a sou ces ha did no ha e ime a ailable
(ESTOC, MAREDAT and TPSS). In ega d o all da a, i du-
plica es we e ound, da a om he NOMAD da ase we e
selec ed i s , ollowed by da a om indi idual p ojec s o
con ibu o s (MOBY, BOUSSOLE, AERONET-OC, AMT,
HOT,GeP&CO, AWI, BARENTSSEA, BATS, CALCOFI,
CCELTER, CIMT, ESTOC, IMOS, PALMER, TPSS and
TARA), and inally o he emaining da ase s (SeaBASS,
MERMAID,ICES, ARCSSPP, BIOCHEM, BODC, COAST-
COLOUR, MAREDAT and SEADATANET). This p oce-
du e was chosen o p ese e he NOMAD da ase as a whole,
since i is widely used in ocean-colou alida ion. I should
be no ed ha , by his p ocedu e, da a om indi idual p ojec s
o con ibu o s may be lis ed unde NOMAD (e.g. some
PALMER da a a e ound in NOMAD wi h me ada a s ing
nomad_palme _l e ). A e gi ing p io i y o NOMAD, he
p io i y was gene ally gi en o da a om indi idual p ojec s
o con ibu o s, bu due o an inc emen al app oach, whe e
only new da a a e added o p e ious e sions o he com-
pila ion, some da a om indi idual p ojec s o con ibu o s
(BATS, CALCOFI, CIMT, PALMER and TPSS) added in
la e s ages may be ound unde o he da a sou ces. This
occu s mainly o BATS and CALCOFI, which ha e hei
ea lie chlo ophyll da a in SeaBASS wi h me ada a s ings
seabass_ba s* and seabass_cal*, and also CIMT, which has
some o i s da a unde COASTCOLOUR. A e all da a om
www.ea h-sys -sci-da a.ne /11/1037/2019/ Ea h Sys . Sci. Da a, 11, 1037–1068, 2019

1042 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
Table 1. The s anda d a iables, nomencla u es and uni s in he inal able.
Va iable/column Desc ip ion and uni s
idx Unique key iden i ying each ow
ime GMT, <YYYY-MM-DD>T<HH:MM:SS>Z
la Decimal deg ee, −90 :90, sou h nega i e
long Decimal deg ee, −180 :180, wes nega i e
dep h_wa e Sampling dep h (m) – all assigned o ze o
chla_hplc To al chlo ophyll aconcen a ion de e mined om he HPLC me hod (mg m−3)
chla_ luo Chlo ophyll aconcen a ion de e mined om luo ome ic o
spec opho ome ic me hods (mg m−3)
s_<band>Remo e-sensing e lec ance (s −1)
aph_<band>Algal pigmen abso p ion coe icien (m−1)
adg_<band>De i al plus CDOM abso p ion coe icien (m−1)
bbp_<band>Pa icle backsca e ing coe icien (m−1)
kd_<band>Di use a enua ion coe icien o downwa d i adiance (m−1)
sm To al suspended ma e (g m−3)
e opo1 Wa e dep h om ETOPO1 (m)
chla_hplc_da ase Me ada a s ing o chla_hplc
chla_hplc_subda ase Me ada a s ing o chla_hplc
chla_hplc_con ibu o Me ada a s ing o chla_hplc
chla_ luo _da ase Me ada a s ing o chla_ luo
chla_ luo _subda ase Me ada a s ing o chla_ luo
chla_ luo _con ibu o Me ada a s ing o chla_ luo
s_da ase Me ada a s ing o s
s_subda ase Me ada a s ing o s
s_con ibu o Me ada a s ing o s
aph_da ase Me ada a s ing o aph
aph_subda ase Me ada a s ing o aph
aph_con ibu o Me ada a s ing o aph
adg_da ase Me ada a s ing o adg
adg_subda ase Me ada a s ing o adg
adg_con ibu o Me ada a s ing o adg
bbp_da ase Me ada a s ing o bbp
bbp_subda ase Me ada a s ing o bbp
bbp_con ibu o Me ada a s ing o bbp
kd_da ase Me ada a s ing o kd
kd_subda ase Me ada a s ing o kd
kd_con ibu o Me ada a s ing o kd
sm_da ase Me ada a s ing o sm
sm_subda ase Me ada a s ing o sm
sm_con ibu o Me ada a s ing o sm
lag_ ime “1” i obse a ion wi hou ime (se o 12:00:00 UTC)
lag_chl_me hod “1” i obse a ion as unknown chlo ophyll me hod
a gi en sou ce we e ee o duplica es, hey we e me ged
consecu i ely by a iable in he inal able. Du ing his p o-
cess, we also sea ched o ows (s a ions) ha we e sepa-
a ed om each o he by ime di e ences less han 5 min
and ho izon al spa ial di e ences o less han 200 m. When
such ows we e ound, he obse a ions in hose ows we e
me ged in o a single ow. The compiled me ged da a we e
compa ed wi h he o iginal se s o ce i y ha no e o s oc-
cu ed du ing he me ging. As a inal s ep, a wa e column
(s a ion) dep h was eco ded o each obse a ion, which was
he closes wa e column dep h om he ETOPO1 global
elie model (Na ional Geophysical Da a Cen e ETOPO1;
Aman e and Eakins, 2009). Fo obse a ions whe e he clos-
es wa e dep h was abo e sea le el (e.g. da a collec ed e y
nea he coas ), i was gi en he alue o ze o.
Da a p ocessing hus included wo majo s eps: p ep o-
cessing and me ging. The i s s ep was ela ed o each se
o con ibu ing da ase s in pa icula and aimed o iden i y
p oblems and con e he da a o in e es o a s anda d o -
ma . The second s ep deal wi h he in eg a ion o da a in o
one unique ile and included he elimina ion o duplica ed
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/
A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo 1043
Table 2. O iginal se s o da a and da a con ibu o s in he inal able.
Da a sou ce Desc ip ion Da a con ibu o s
Ma ine Op ical
Buoy (MOBY)
Daily obse a ions o emo e-sensing e lec ance,
measu ed by a ixed moo ing sys em, loca ed wes
o he Hawaiian island o Lanai. Da a compiled be-
ween 1997 and 2012. Da a we e ob ained om he
MOBY websi e. Compiled s anda d a iable: s.
Paul DiGiacomo, Kenne h Voss
Bouée pou
l’acquisi ion d’une
Sé ie Op ique
à Long Te me
(BOUSSOLE)
High- equency (15 min) obse a ions o emo e-
sensing e lec ance, om a ixed moo ing sys em,
loca ed in he wes e n Medi e anean Sea. Mea-
su emen s o chlo ophyll aconcen a ion a e also
a ailable a he moo ing loca ions. Remo e-sensing
e lec ance and chlo ophyll ada a we e compiled
be ween 2003–2012 and 2001–2012, espec i ely.
Da a we e p o ided by Da id An oine. Compiled
s anda d a iables: s and chla_hplc.
Da id An oine, Vicenzo Vellucci
AE osol RObo ic
NETwo k-Ocean
Colo
(AERONET-OC)
Daily obse a ions o emo e-sensing e lec ance,
measu ed by modi ied sun pho ome e s. Da a
compiled be ween 2002 and 2012. Si es in-
cluded Abu_Al_Bukhoosh (∼25◦N, ∼53◦E),
COVE_SEAPRISM (∼36◦N, ∼75◦W), Glo-
ia (∼44◦N, ∼29◦E), Gus a _Dalen_Towe (∼
58◦N, ∼17◦E), Helsinki Ligh house (∼59◦N,
∼24◦E), LISCO (∼40◦N, ∼73◦W), Lucinda
(∼18◦S, ∼146◦E), MVCO (∼41◦N, ∼70◦W),
Palg unden (∼58◦N, ∼13◦E), Venice (∼45◦N,
∼12◦E) and Wa eCIS_Si e_CSI_6 (∼28◦N, ∼
90◦W). Da a we e ob ained om he AERONET-
OC websi e. Compiled s anda d a iable: s.
Robe A none (Wa eCIS), Sam Ahmed (LISCO),
Vi o io B ando (Lucinda), Dick C ou (Wa e-
CIS), Hui Feng (MVCO), Alex Gile son (LISCO),
Rick Gould (Wa eCIS), B en Holben (COVE-
SEAPRISM), Susanne K a ze (Palg unden),
Thomas Sch oede (Lucinda), Heidi M. Sosik
(MVCO), Giuseppe Zibo di (Abu Al Bukhoosh,
Glo ia, Gus a Dalen Towe , Helsinki Ligh house and
Venice)
SeaWiFS
Bio-op ical A chi e
and S o age Sys em
(SeaBASS)
Global a chi e o in si u ma ine da a om mul-
iple con ibu o s. Bio-op ical global da a be ween
1997 and 2012 we e ex ac ed om he SeaBASS
websi e. Pigmen da a we e mos ly ex ac ed us-
ing he Pigmen sea ch ool, which p o ides da a
di ec ly om he a chi es. Radiome ic da a we e
ex ac ed using he Valida ion ool, which only
p o ides in si u da a wi h ma chups o ocean-
colou senso s. Compiled s anda d a iables: s,
chla_hplc, chl_ luo , aph, adg, bbp and kd.
Robe A none, Ke in A igo, William Balch,
Ray Ba low, Mike Beh en eld, Suk u Besik epe,
Emmanuel Boss, Ch is B own, Douglas Capone,
Ken Ca de , F ancisco Cha ez, Alex Chekalyuk,
Jay-Chung Chen, Dennis Cla k, He e Claus e,
Lesley Clemen son, Jo ge Co edo , Glenn Co a,
Y es Dandonneau, Heidi Die ssen, Da id Es-
linge , Pio Fla au, Robe F ouin, Ca los Ga cia,
Alex Gile son, Joaquim Goes, Gwo-Ching Gong,
Ad iana Gonzalez-Sil e a, Rick Gould, La y Ha d-
ing, Jon Ha e, S an B. Hooke , Chuanmin Hu,
Mil on Kampel, Sung-Ho Kang, Ga y Ki kpa ick,
Oleg Kopele ich, Sam Laney, Pie e La ouche,
Jesus Ledesma, Zhongping Lee, Rica do Le elie ,
Ma lon Lewis, S e en Loh enz, Ma y Luz Canon,
An onio Mannino, John Ma a, Chuck McClain,
Ch is ophe Menkes, Ma k Mille , G eg Mi chell,
Ru Mo ison, James Muelle , F ank Mulle -Ka ge ,
Ruben Neg i, James Nelson, No man Nelson,
Ma y Jane Pe y, Da id Phinney, John Po e ,
Collin Roesle , Da id Siegel, Mike Sie acki, Je -
ey Sma , Raymond Smi h, Heidi Sosik, James Spin-
hi ne, Da iusz S amski, Rick S ump , Aji Sub-
amaniam, Chuck T ees, Michael Twa dowski,
Kenne h Voss, Ma cel We nand, Ronald Zane eld,
E ic Ze le , Giuseppe Zibo di, Richa d Zimme man
www.ea h-sys -sci-da a.ne /11/1037/2019/ Ea h Sys . Sci. Da a, 11, 1037–1068, 2019
1044 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
Table 2. Con inued.
Da a sou ce Desc ip ion Da a con ibu o s
NASA bio-Op ical
Ma ine Algo i hm
Da ase (NOMAD)
High-quali y global da ase o coinciden bio-
op ical in si u da a. The da ase was buil upon he
SeaBASS a chi e. The cu en e sion ( e sion 2.0
ALPHA, 2008) was used, wi h an addi ional se
o columns o emo e-sensing e lec ance co ec ed
o he bidi ec ional na u e o he ligh ield, p o-
ided by NOMAD c ea o s. Da a compiled be ween
1997 and 2007. Compiled s anda d a iables: s,
chla_hplc, chl_ luo , aph, adg, bbp and kd.
Robe A none, Ke in A igo, William Balch, Ray Ba -
low, Mike Beh en eld, Ch is B own, Douglas Capone,
Ken Ca de , F ancisco Cha ez, Dennis Cla k,
He e Claus e, Jo ge Co edo , Glenn Co a, Da id Es-
linge , Pio Fla au, Robe F ouin, Rick Gould,
La y Ha ding, S an B. Hooke , Oleg Kopele ich,
Ma lon Lewis, An onio Mannino, John Ma a,
Ma k Mille , G eg Mi chell, Ti any Moisan, Ru Mo i-
son, F ank Mulle -Ka ge , James Nelson, No man Nel-
son, Da id Siegel, Raymond Smi h, Timo hy Smy h,
James Spinhi ne, Da iusz S amski, Rick S ump ,
Aji Sub amaniam, Kenne h Voss
MERIS Ma ch-up
In si u Da abase
(MERMAID)
Global da abase o in si u bio-op ical da a ma ched
wi h concu en MERIS Le el 2 sa elli e ocean-
colou p oduc s. The Ex ac ma chup ool o ac-
qui e da a was used. Da a we e compiled be ween
2002 and 2012. Access has been g an ed h ough
a signed se ice le el ag eemen . Compiled s an-
da d a iables: s, chla_hplc, chl_ luo , aph, adg,
bbp and kd.
Simon Belange , Jean-F ancois Be hon, Vanda B o-
as, Elisabe a Canu i, Pie e Y es Deschamps, An-
nelies Homme som, Ma i Kah u, Holge Klein, Hu-
be Loisel, Da id McKee, G eg Mi chell, Michael On-
d usek, Michel Repecaud, Da id Siegel, Ga in Til-
s one, Giuseppe Zibo di
A lan ic Me idional
T ansec (AMT)
Mul idisciplina y p og amme ha makes biologi-
cal, chemical and physical oceanog aphic measu e-
men s du ing an annual oyage be ween he Uni ed
Kingdom and des ina ions in he Sou h A lan ic. I
has compiled obse a ions o chlo ophyll aconcen-
a ion be ween 1997 (AMT5) and 2005 (AMT17).
Da a we e p o ided by he B i ish Oceanog aphic
Da a Cen e (BODC). Compiled s anda d a iables:
chla_hplc and chl_ luo .
Ray Ba low, S ua Gibb, Vic o ia Hill, Pa ick Hol-
ligan, Ge ald Moo e, Leonie O’Dowd, Alex Poul on,
Emilio Sua ez
In e na ional
Council o he Ex-
plo a ion o he Sea
(ICES)
Da abase o se e al collec ions o da a ela ed o
he ma ine en i onmen . I has compiled obse a-
ions o chlo ophyll aconcen a ion in he no he n
Eu opean seas, be ween 1997 and 2012. Da a we e
p o ided by he ICES da abase on he ma ine en i-
onmen (2014, Copenhagen, Denma k). Compiled
s anda d a iables: chla_hplc and chl_ luo .
No a ailable
Hawaii Ocean
Time-se ies (HOT)
Mul idisciplina y p og amme ha makes epea ed
biological, chemical and physical oceanog aphic
obse a ions nea Oahu, Hawaii. Measu emen s o
chlo ophyll aconcen a ion be ween 1997 and 2012
we e ex ac ed om he p ojec websi e. Compiled
s anda d a iables: chla_hplc and chl_ luo .
Bob Bidiga e, Ma hew Chu ch, Rica do Le elie , Jas-
mine Naho niak
Geochemis y, Phy-
oplank on, and
Colo o he Ocean
(GeP&CO)
P og amme o in si u da a collec ion aboa d a me -
chan ship om F ance o New Caledonia, be ween
1999 and 2002. Measu emen s o chlo ophyll a
concen a ion we e ob ained om he p ojec web-
si e. Compiled s anda d a iables: chla_hplc and
chla_ luo .
Y es Dandonneau
ARCSSPP A c ic Sys em Science P ima y P oduc ion
da abase. A ailable om he NODC FTP si e.
Compiled s anda d a iable: chla_ luo .
Pa icia Ma ai
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/
A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo 1045
Table 2. Con inued.
Da a sou ce Desc ip ion Da a con ibu o s
AWI Se e al 2007–2012 c uises in he A lan ic, Pa-
ci ic and Sou he n Ocean om As id B ache ’s
g oup a AWI. P o ided by As id B ache . A ail-
able om PANGAEA. Compiled s anda d a i-
ables: chla_ luo , s and aph.
As id B ache
BARENTSSEA Da a collec ion om c uises o he Ins i u e o Ma-
ine Resea ch (No way) mainly a ound he Ba en s
Sea. P o ided by Knu Yng e Bø sheim. Compiled
s anda d a iable: chla_ luo .
Knu Yng e Bø sheim
BATS Da a collec ion om he Be muda A lan ic Time-
se ies S udy. A ailable om BATS websi e. Com-
piled s anda d a iables: chla_ luo and chla_hplc.
No a ailable
BIOCHEM The Fishe ies and Oceans Canada da abase o bio-
logical and chemical da a. Mos ly da a om Gul o
S Law ence. A ailable om BIOCHEM websi e.
Compiled s anda d a iable: chla_ luo .
Diane A chambaul , Hughes Benoi , Es he Bon-
neau, Eugene Colbou ne, Alain Gagne, Y es Gagnon,
Tom Hu lbu , Ca he ine Johnson, Pie e Joly, Mau-
ice Le asseu , Pa ick Ouelle , Jacques Plou de,
Luc Sa oie, Michael Sca a , Philippe Schwab,
Michel S a , F ançois Villeneu e
BODC B i ish Oceanog aphic Da a Cen e. Mainly Eu o-
pean seas. P o ided by BODC. Compiled s anda d
a iables: chla_ luo and chla_hplc.
No a ailable
CALCOFI C uise da a om he Cali o nia Coope a i e
Oceanic Fishe ies In es iga ions p og amme. A ail-
able om CalCOFI websi e. Compiled s anda d
a iable: chla_ luo .
Ral Goe icke
CCELTER C uise da a om Cali o nia Cu en Ecosys em
Long Te m Ecological Resea ch. A ailable om
CCELTER websi e. Compiled s anda d a iable:
chla_ luo .
Ral Goe icke
CIMT Sampling om he Cen e o In eg a ed Ma ine
Technology (Cali o nia). A ailable om CIMT
websi e. Compiled s anda d a iable: chla_ luo .
Raphael Kudela
COASTCOLOUR Quali y-con olled compila ion o bio-op ical da a
in se e al coas al si es. A ailable om PANGAEA.
Compiled s anda d a iables: chla_ luo , chla_hplc,
s, aph, adg, bbp and sm.
No a ailable
ESTOC Sampling om he Es ación Eu opea de Se ies
Tempo ales del Oceano Cana y Islands. P o ided
by And és Cianca. Compiled s anda d a iable:
chla_ luo .
Oc a io Llinas and And es Cianca
IMOS Aus alian Na ional Re e ence S a ions. A ailable
om he Aus alian Ocean Da a Ne wo k (AODN).
Compiled s anda d a iable: chla_hplc.
Lesley Clemen son
MAREDAT Quali y-con olled global compila ion o chla
HPLC. A ailable om PANGAEA. Compiled s an-
da d a iable: chla_hplc.
Ray Ba low, Robe Bidiga e, He e Claus e,
Denise Cummings, Giacomo DiTullio, Ch is Galli-
enne, Ral Goe icke, Pa ick Holligan, Da id Ka l,
Michael Land y, Michael Lomas, Michael Lucas,
Jean-Claude Ma y, Walke Smi h, Denise Smy he-
W igh , Rick S ump , Emilio Sua ez, Koji Suzuki,
Ma ia Ve ne , Simon W igh
www.ea h-sys -sci-da a.ne /11/1037/2019/ Ea h Sys . Sci. Da a, 11, 1037–1068, 2019
1052 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
Figu e 2. The dis ibu ion o (a) s a 44Xnm and (b) s a 55Xnm. Da a we e i s sea ched o a 445 and 555 nm and hen wi h a sea ch
window o up o 8 nm o include da a a 547 nm. The black boxes delimi he pe cen iles 0.25 and 0.75 o he da a and he black ho izon al
lines show he ex ension o up o pe cen iles 0.05 and 0.95. The ed line ep esen s he median alue and he black ci cles he alues below
(and abo e) he pe cen ile 0.05 (0.95). The numbe o measu emen s o each da ase is epo ed on he igh axis o he g aph.
2.2.23 MARine Ecosys em DATa (MAREDAT)
The MAREDAT da abase is a global assemblage o pig-
men s measu ed by HPLC (Peloquin e al., 2013a) om
he combina ion o 136 independen ield da ase s, solici ed
om in es iga o s and da abases. The da abase p o ides
high-quali y measu emen s o axonomic pigmen s includ-
ing chlo ophyll aand b, 19’-bu anoyloxy ucoxan hin, 19’-
hexanoyloxy ucoxan hin, alloxan hin, di inyl chlo ophyll a,
ucoxan hin, lu ein, pe idinin, p asinoxan hin, iolaxan hin
and zeaxan hin. The da abase is a ailable h ough PAN-
GAEA (h ps://doi.o g/10.1594/PANGAEA.793246; Pelo-
quin e al., 2013b). Fo his wo k only measu emen s o o al
chlo ophyll a lagged wi h high quali y we e used. The ime
o day was una ailable and was se o 12:00:00 UTC. These
obse a ions we e lagged wi h “1” in he column lag_ ime.
The compiled a iable was chla_hplc.
2.2.24 Palme s a ion Long-Te m Ecological Resea ch
(PALMER)
PALMER is a moni o ing s a ion loca ed in wes e n
An a c ic Peninsula. The Palme s a ion in es iga es he
ma ine ecology o he Sou he n Ocean wi h a ocus on
he pelagic ma ine ecosys em, including sea ice habi-
a s, egional oceanog aphy and nes ing si es o seabi d
p eda o s. The PALMER da a include measu emen s o
me eo ological, oceanog aphic, sea ice, p eda o s, nu i-
en s and biogeochemis y, pigmen s, p ima y p oduc ion,
zooplank on and mic obe pa ame e s. This wo k used
he measu emen s o chlo ophyll analysed by HPLC and
luo ome y aken a he Palme s a ion (h ps://doi.o g/
10.6073/pas a/0624c7d161d3b5486d7ba06c2e50ee21;
Scho ield e al., 2018a; and h ps://doi.o g/10.6073/pas a/
dea95430a6ad84ecea023ee1ced650d3; Scho ield e al.,
2018b) and om he annual c uises o he coas o he
wes e n An a c ic Peninsula (h ps://doi.o g/10.6073/pas a/
Figu e 3. Tempo al dis ibu ion o chlo ophyll aconcen a ion
(chl), emo e-sensing e lec ance ( s), algal pigmen abso p ion co-
e icien (aph), de i al plus CDOM abso p ion coe icien (adg),
pa icle backsca e ing coe icien (bbp), he di use a enua ion co-
e icien o downwa d i adiance (kd) and o al suspended ma e
( sm) in he inal able. All chlo ophyll da a we e conside ed, bu o
a gi en s a ion, HPLC da a we e selec ed i a ailable. Colou s indi-
ca e he numbe o s a ions a ailable o each a iable, as a unc ion
o mon h and hemisphe e o da a acquisi ion (N – No he n Hemi-
sphe e; S – Sou he n Hemisphe e). The emp y (whi e) squa es indi-
ca e no da a o ha mon h.
4d583713667a0 52b9d2937a26d0d82e; Scho ield e
al., 2018c; and h ps://doi.o g/10.6073/pas a/c479b922
d42ace1ce37 9a977e214952; Scho ield e al., 2017). The
compiled a iables we e chla_hplc and chla_ luo .
2.2.25 SeaDa aNe a chi e (SEADATANET)
SeaDa aNe is a Pan-Eu opean in as uc u e o ocean and
ma ine da a managemen . I aims o de elop a s anda dized
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/

A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo 1053
Figu e 4. Ranges o emo e-sensing e lec ance band a ios
(412 : 443 and 490 : 555) o all da a. The poin s om he NOMAD
da ase a e shown in blue o e e ence. To maximize he numbe
o a ios pe da ase a sea ch window up o 12 nm was used, when
he ou wa eleng hs (412, 443, 490, 555) we e no simul aneously
a ailable. The e ec o di e en sea ch windows was negligible in
he a io dis ibu ion.
sys em o managing la ge and di e se da ase s collec ed by
oceanog aphic c uises and au oma ic obse a ion sys ems.
Fo his wo k, disc e e chlo ophyll aconcen a ion obse -
a ions wi h an access es ic ion se o academic and un-
es ic ed we e acqui ed om he SeaDa aNe pla o m wi h
guidance om he help desk. Only da a om he Ins i u e o
Ma ine Resea ch – No wegian Ma ine Da a Cen e (NMD),
No way, which comp ised mos o he acqui ed da a, we e
used. All chlo ophyll obse a ions we e om disc e e sam-
ples measu ed by luo ome ic, spec opho ome ic o HPLC
me hods, bu he exac me hod was no gi en. Thus, he ob-
se a ions we e ma ked as chla_ luo , al hough some we e
possibly om HPLC measu emen s, and we e lagged wi h
“1” in he column lag_chla_me hod. The compiled a iable
was chla_ luo .
2.2.26 Da a p o ided by T e o Pla and Shubha
Sa hyend ana h (TPSS)
In his wo k, he TPSS da a sou ce e e s o a g oup o ob-
se a ions ha we e p o ided o his compila ion by T e o
Pla and Shubha Sa hyend ana h. This is a collec ion o bio-
op ical in si u da a collec ed du ing c uises p edominan ly in
he no hwes e n A lan ic bu also om he Indian Ocean,
Sou h Paci ic and cen al A lan ic (see Sa hyend ana h e al.,
2009, o addi ional de ails ega ding he c uises). I com-
p ises measu emen s o phy oplank on pigmen s and algal
pigmen abso p ion coe icien s. The ime o day was un-
a ailable and was se o 12:00:00 UTC. These obse a ions
we e lagged wi h “1” in he column lag_ ime. The compiled
a iables we e chla_hplc, chla_ luo and aph.
2.2.27 Bio-op ical da a om Ta a expedi ions (TARA)
The Ta a expedi ions consis o se e al c uises a ound he
wo ld, some wi h du a ions o se e al yea s, designed o
s udy and unde s and he dis ibu ion o plank onic o gan-
isms in he wo ld ocean. The disc e e obse a ions o emo e-
sensing e lec ance and chlo ophyll aconcen a ion om
HPLC measu emen s aken du ing he Ta a Oceans (2009–
2013) and Medi e anean (2014) expedi ions we e consid-
e ed in his wo k. These da a we e p o ided o he ESA OC-
CCI p ojec by Emmanuel Boss and we e a ailable in he
SeaBASS a chi e. The emo e-sensing e lec ances we e co -
ec ed o he bidi ec ional na u e o he ligh ield (Mo el
and Gen ili, 1996; Mo el e al., 2002). The compiled a i-
ables we e chla_hplc and s.
3 Resul s
In his wo k se e al se s o bio-op ical in si u da a we e ac-
qui ed, homogenized and me ged in o a single able. The a-
ble comp ises in si u obse a ions be ween 1997 and 2018,
wi h a global dis ibu ion, and includes he ollowing a i-
ables: emo e-sensing e lec ance ( s), chlo ophyll acon-
cen a ion (chla), algal pigmen abso p ion coe icien (aph),
de i al and colou ed dissol ed o ganic ma e abso p ion
(adg), pa icle backsca e ing coe icien (bbp), di use a -
enua ion coe icien o downwa d i adiance (kd) and o al
suspended ma e ( sm). All obse a ions in he able we e
p ocessed in such a way ha hey can be compa ed di ec ly
wi h sa elli e-de i ed ocean-colou da a. The able consis s
o 136 250 ows and 1286 columns. Each ow ep esen s a
unique s a ion in space and ime, sepa a ed om he es by
a leas 5 min and 200 m. Fo each obse a ion in a gi en s a-
ion, he e a e h ee me ada a s ings: da ase , subda ase and
con ibu o . The columns o he able ake he o m desc ibed
in Table 1. The da a con ibu o s a e indica ed in Table 2. Re-
ga ding spec al a iables, all o iginal wa eleng hs we e p e-
se ed, which equi es a la ge numbe o unique wa eleng hs
o be main ained in he da abase. No band shi ing was pe -
o med ( hough some a chi ed da a in some da a sou ces may
ha e been me ged wi h nea by wa eleng hs) and no mini-
mum numbe o wa eleng hs pe obse a ion was imposed.
This allows u he manipula ion o he able o di e en
pu poses. In he ollowing pa ag aphs, he able is analysed
and he inal g oup o obse a ions is desc ibed o each con-
ibu ing da ase ; howe e , he numbe s epo ed he e do no
e lec he o iginal numbe s in each da ase , since duplica es
ac oss con ibu ing da ase s we e emo ed (e.g. NOMAD
and o he s we e emo ed om MERMAID).
Obse a ions o emo e-sensing e lec ance a e a ailable a
611 unique wa eleng hs (i.e. columns), be ween 404.7 and
1022.1 nm (Fig. 1). In o al he e a e 59 781 obse a ions
(i.e. ows) wi h emo e-sensing e lec ance in he able. The
o al numbe o obse a ions a e pa i ioned pe con ibu -
ing da ase s as ollows: AERONET-OC (31 574), BOUS-
www.ea h-sys -sci-da a.ne /11/1037/2019/ Ea h Sys . Sci. Da a, 11, 1037–1068, 2019
1054 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
Figu e 5. Global dis ibu ion o emo e-sensing e lec ance pe da ase in he inal able. The da a sou ces a e iden i ied wi h di e en
colou s. Poin s show loca ions whe e a leas one obse a ion is a ailable. C osses show si es om which ime se ies da a o emo e-sensing
e lec ance a e a ailable.
Figu e 6. Compa ison o coinciden obse a ions o chlo o-
phyll aconcen a ion de i ed wi h di e en me hods (chla_ luo
and chla_hplc). The da a we e ans o med p io o eg ession anal-
ysis o accoun o hei log-no mal dis ibu ion.
SOLE (17 364), MOBY (5466), NOMAD (3326), MER-
MAID (885), SeaBASS (698), AWI (54), COASTCOLOUR
(307) and TARA (107). Da a om AERONET-OC, BOUS-
SOLE and MOBY co espond o con inuous ime se ies,
and, hence, he highe numbe o obse a ions. Da a dis-
ibu ion a 44Xand 55Xnm is p o ided in Fig. 2a and
b, espec i ely. Da a we e i s sea ched o a 445 and
555 nm and hen wi h a sea ch window up o 8 nm o in-
Figu e 7. Numbe o obse a ions pe chlo ophyll aconcen a ion
acqui ed wi h di e en me hods (chla_ luo and chla_hplc).
clude also da a a 547 nm. Median alues a 44Xnm ange
om 0.003 m−1(AERONET-OC) and 0.009 m−1(MOBY),
whe eas a 55Xnm he median alues lie be ween 0.001 m−1
(AWI) and 0.007 m−1(COASTCOLOUR). The obse a ions
a e une enly dis ibu ed be ween each mon h o he yea in
bo h hemisphe es, wi h a highe co e age in summe mon hs
(Fig. 3). The e a e ewe da a in he Sou he n Hemisphe e
han in he No he n Hemisphe e (Fig. 3). Fo addi ional
analysis, s band a ios we e plo ed agains each o he
(490 : 555 e sus 412 : 443, Fig. 4). Mos poin s a e wi hin
he bounda ies o he NOMAD da ase , bu some sca e ed
poin s we e ound. These poin s we e e ained in he able o
allow u he manipula ion wi h di e en quali y con ol c i-
e ia. Complemen a y analysis o emo e-sensing e lec ance
da a is made when o he a iables a e concu en ly a ail-
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/
A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo 1055
Figu e 8. Global dis ibu ion o chlo ophyll aconcen a ion pe in e al o he obse ed alue. All chlo ophyll da a we e conside ed, bu o
a gi en s a ion, HPLC da a we e selec ed i a ailable.
Figu e 9. Global dis ibu ion o chlo ophyll aconcen a ion pe da ase in he inal able. All chlo ophyll da a we e conside ed, bu o a
gi en s a ion, HPLC da a we e selec ed i a ailable. C osses show si es om whe e da a o chlo ophyll a e a ailable in a speci ic geog aphic
loca ion.
able and discussed below (see Figs. 11 and 16). The geo-
g aphic dis ibu ion o emo e-sensing e lec ance obse a-
ions (Fig. 5) shows a highe numbe o obse a ions in some
coas al egions, such as hose o No h Ame ica and no h-
e n Eu ope. The cen al egions o he ocean show a lowe
numbe o obse a ions, wi h he A lan ic Ocean ha ing he
highes densi y in ela ion o he o he oceans. The bes geo-
g aphic co e age is p o ided by he NOMAD da abase. Da a
om SeaBASS a e ewe in numbe bu a e s ill impo an .
Da a om MERMAID a e mainly loca ed along he coas s o
Eu ope, No h Ame ica and he cen al egion o he No h
A lan ic Ocean. The obse a ions om COASTCOLOUR
a e concen a ed in 17 coas al si es a ound he wo ld, while
AWI da a a e a ailable o he A lan ic, Paci ic and Sou he n
Ocean. TARA da a a e sp ead ac oss se e al egions, wi h
he highes da a densi y in he Medi e anean Sea.
www.ea h-sys -sci-da a.ne /11/1037/2019/ Ea h Sys . Sci. Da a, 11, 1037–1068, 2019
1056 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
Figu e 10. The chlo ophyll a(mg m−3) da a pa i ioned in o 5◦×5◦boxes showing (a) numbe o obse a ions, (b) a e age alue and
(c) s anda d de ia ion in each box. All chlo ophyll da a we e conside ed, bu o a gi en s a ion, HPLC da a we e selec ed i a ailable. In he
s anda d de ia ion plo , g ey colou boxes ep esen ze o s anda d de ia ion (i.e. one obse a ion).
Fo chlo ophyll aconcen a ion, wo ypes o obse -
a ions we e compiled, one measu ed by luo ome ic o
spec opho ome ic me hods (chla_ luo ) and he o he
measu ed by HPLC me hods (chla_hplc). A compa ison
o bo h measu emen s (Fig. 6), when a ailable a he same
s a ion, shows good ag eemen (T ees e al., 1985). As
s a ed be o e, he analysis was done on he inal me ged
able; hus no da a we e il e ed and he good ela ion can
be explained in pa by he quali y con ol implemen ed by
he da a p o ide s and cu a o s o eposi o ies such as NO-
MAD and SeaBASS (We dell and Bailey, 2005). The o al
numbe o ows wi h concu en chla_ luo and chla_hplc
is 5344, wi h con ibu ions om SeaBASS (39 %), TPSS
(18 %), NOMAD (13 %), PALMER (9 %), BATS (6 %),
COASTCOLOUR (5 %), MERMAID (4 %), HOT (4 %), and
AMT+GeP&CO+BODC+CCELTER+CALCOFI (2 %).
The chla_ luo obse a ions a e a ailable in 61 525 s a ions
( ows), wi h alues anging om 0.001 o 100 mg m−3
(Fig. 7). They a e om NOMAD (2350), SeaBASS
(18 122), MERMAID (3711), ICES (5421), HOT (702),
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/
A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo 1057
Figu e 11. A emo e-sensing e lec ance maximum band a io (as
de ined in ex ) ([443,490,510] /555 o [443,490,510] /560 i 555
no a ailable) as a unc ion o chlo ophyll aconcen a ion. All
chlo ophyll da a we e conside ed, bu o a gi en s a ion, HPLC
da a we e selec ed i a ailable. Da a wi hin 2 nm o he wa e-
leng hs we e used. Fo e e ence, he solid and do ed lines show
he NASA OC4 and OC4E 6 s anda d algo i hms, espec i ely
(h ps://oceancolo .gs c.nasa.go /a bd/chlo _a/, las access: 10 July
2019). The o al numbe o poin s was 3814, o which 79 % we e
om NOMAD.
AMT (164), ARCSSPP (189), BARENTSSEA (7188),
BATS (356), BIOCHEM (4592), BODC (895), CALCOFI
(4631), COASTCOLOUR (3322), CCELTER (254), CIMT
(204), ESTOC (100), GEPCO (56), PALMER (2865),
SEADATANET (5403) and TPSS (1000). The o al num-
be o chla_hplc obse a ions is 23 550, anging om
0.002 o 99.8 mg m−3(Fig. 7), wi h con ibu ions om
NOMAD (1309), SeaBASS (9478), MERMAID (707),
ICES (2994), HOT (193), GeP&CO (1536), BOUSSOLE
(397), AMT (902), AWI (750), BATS (334), BODC (735),
COASTCOLOUR (848), IMOS (103), MAREDAT (1024),
PALMER (1077), TPSS (1002) and TARA (161). The com-
bined chlo ophyll da ase (all chlo ophyll da a conside ed,
bu o a gi en s a ion, HPLC da a we e selec ed i a ailable)
has a o al o 79 731 obse a ions, wi h 10 %, 49 % and 41 %
espec i ely om oligo ophic (<0.1 mg m−3), meso ophic
(0.1–1 mg m−3) and eu ophic (>1 mg m−3) wa e s. When
compa ed wi h he p opo ions o he wo ld ocean in hese
ophic classes, 56 % oligo ophic, 42 % meso ophic and
2 % eu ophic (An oine e al., 1996), oligo ophic wa e s
a e unde ep esen ed ela i e o eu ophic wa e s in he
compila ion. The combined chlo ophyll da ase is une enly
dis ibu ed be ween each mon h o he yea in bo h he
No he n and Sou he n Hemisphe e, wi h highe co e age
in summe mon hs (Fig. 3). The e a e ewe da a in he
Sou he n Hemisphe e han in he No he n Hemisphe e
(Fig. 3). The spa ial dis ibu ion o he chlo ophyll alues
o he combined da ase (Fig. 8) shows a good ag eemen
wi h known biogeog aphical ea u es, such as lowe chlo o-
phyll alues in he sub opical gy es and highe alues in
empe a e, coas al and upwelling egions. Many egions
show a good spa ial co e age (e.g. A lan ic and Paci ic
Ocean), while o he s a e less well sampled (e.g. Sou he n
and Indian Ocean). O he con ibu ing da ase s, NOMAD
and SeaBASS p o ide a good spa ial co e age in many
egions (Fig. 9). O he da ase s also p o ide co e age om
se e al loca ions ac oss he globe (GEPCO, MAREDAT,
TARA). The ICES, MERMAID and BODC da a a e mainly
loca ed along he coas al egions o Eu ope. The AMT
and many AWI da a mainly co e he cen al pa o he
A lan ic Ocean, o he AWI da a co e he A lan ic sec o
and he Amundsen o Bellingshausen Sea o he Sou he n
Ocean and he wes e n sub opical and opical Paci ic. The
SEADATANET, ARCSSPP and BARENTSSEA p o ide
co e age o he A c ic egion and no he n seas o he No h
A lan ic. The obse a ions om BIOCHEM and TPSS a e
mos ly om he no hwes e n A lan ic, while CALCOFI,
CCELTER and CIMT p o ide da a o he wes e n coas o
No h Ame ica. The emaining da ase s p o ide obse a ions
o ixed loca ions: PALMER (wes e n An a c ic Peninsula),
COASTCOLOUR (17 coas al si es ac oss he wo ld), BATS
(Be muda, No h A lan ic), BOUSSOLE (Medi e anean),
HOT (Hawaii, No h Paci ic), IMOS (coas al si es a ound
Aus alia) and ESTOC (Cana ies, No h A lan ic). Figu e 9
shows all da a sou ces ha con ibu e wi h chlo ophyll
obse a ions, bu many o e lap each o he , especially a ound
Eu ope and No h Ame ica. Fo addi ional analysis and as
an example o he applica ions o he compiled da ase , he
combined chlo ophyll da a (chla_ luo and chla_hplc) we e
pa i ioned in o 5◦×5◦boxes, and o each box he numbe
o obse a ions, a e age alue and s anda d de ia ion we e
compu ed (Fig. 10a, b and c, espec i ely). The numbe
o obse a ions can be e y high (>1000) in some boxes
along he Eu opean and No h Ame ican coas lines and
ela i ely low (<20) in oceanic egions. Again, he e is
e idence in he a e age alue map (Fig. 10b) o well-known
biogeog aphical ea u es, such as he lowe chlo ophyll
in he sub opical gy es and highe alues in coas al and
upwelling a eas. The e is a close co espondence be ween
he spa ial pa e ns o he a e age and s anda d de ia ion
maps (Fig. 10b and c), which may be an indica o o he da a
quali y.
Coinciden obse a ions o chlo ophyll aconcen a-
ion and emo e-sensing e lec ance a e a ailable a 3814
s a ions. These obse a ions a e mos ly om NOMAD
(79 %), MERMAID (9 %), COASTCOLOUR (6%), and
SeaBASS (5 %). The maximum o h ee selec ed band
a ios o emo e-sensing e lec ance is plo ed agains
chlo ophyll aconcen a ion (Fig. 11). The chla alues used
a e he combined HPLC and luo ome ic chlo ophyll a,
and o he s, he closes spec al obse a ion wi hin 2 nm
was used. The maximum band a ios we e calcula ed as he
maximum o [ s(443) / s(555), s(490) / s(555),
www.ea h-sys -sci-da a.ne /11/1037/2019/ Ea h Sys . Sci. Da a, 11, 1037–1068, 2019

1058 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
Figu e 12. The dis ibu ion o (a) aph a 44Xnm, (b) aph a 55Xnm, (c) adg a 44Xnm, (d) adg a 55Xnm, (e) bbp a 44Xnm, ( ) bbp a
55Xnm, (g) kd a 44Xnm, and (h) kd a 55Xnm. Da a we e i s sea ched o a 445 and 555 nm and hen wi h a sea ch window up o 8 nm
o include da a a 547 nm. The g aphical con en ion is iden ical o Fig. 2.
s(510) / s(555)] o [ s(443) / s(560),
s(490) / s(560), s(510) / s(560)] i s(555) was no
a ailable. The ela ionship be ween maximum band a io and
chlo ophyll is close o he NASA OC4 and OC4E 6 s anda d
algo i hm (h ps://oceancolo .gs c.nasa.go /a bd/chlo _a/)
simila ly based on maximum band a ios, p o iding
con idence in he quali y o he compiled da a.
The inhe en op ical p ope ies (aph, adg and bbp) a e
a ailable a 550 unique wa eleng hs be ween 300 and
850 nm. The e is a o al o 3293, 1654 and 792 obse a-
ions, o aph, adg and bbp, espec i ely. Fo aph he o-
al numbe o obse a ions is dis ibu ed among NOMAD
(1190), TPSS (966), COASTCOLOUR (593), AWI (458),
SeaBASS (14) and MERMAID (72). Fo adg he con i-
bu ions a e as ollows: NOMAD (1079), COASTCOLOUR
(531), SeaBASS (11) and MERMAID (33). The bbp ob-
se a ions come om NOMAD (371), COASTCOLOUR
(154), SeaBASS (32) and MERMAID (235). The da a dis-
ibu ion o aph, adg and bbp a 44Xnm and 55Xnm
o each da ase is p o ided in Fig. 12a– . Median al-
ues o aph, adg and bbp a 44Xand 55Xnm o each
da ase a e summa ized in Table 3. Fo addi ional analy-
sis, he ollowing band a ios o he abso p ion coe icien s
we e calcula ed: aph(490) /aph(443), aph(412) /aph(443),
adg(443) /adg(490) and adg(412) /adg(443). Da a wi hin
2 nm o he wa eleng hs we e used o maximize he numbe
o poin s. The dis ibu ion o he a ios is shown in Fig. 13.
Se e al obse a ions we e ound o be ou side he h esholds
used in he IOCCG epo 5 (2006) o quali y con ol (see
do ed e ical black lines in Fig. 13). These poin s a e high-
ligh ed he e o in o ma ion bu e ained in he da abase, as
hese we e mos ly om NOMAD and he e was an in e -
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/
A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo 1059
Figu e 13. The dis ibu ion o abso p ion coe icien s band a ios:
adg(443) /adg(490), adg(412) /adg(443), aph(490) /aph(443) and
aph(412) /aph(443). Da a wi hin 2 nm o he wa eleng hs we e
used. The g aphical con en ion is iden ical o Fig. 2. The e ical
dashed lines show he lowe and uppe h esholds used o qual-
i y con ol in he IOCCG epo 5. The o al numbe o poin s o
adg a ios a e di ided be ween NOMAD (89 %), COASTCOLOUR
(7 %), MERMAID (3 %) and SeaBASS (1 %). The o al numbe o
poin s o aph a ios a e di ided be ween NOMAD (36 %), TPSS
(29 %), COASTCOLOUR (18 %), AWI (14 %), MERMAID (2 %)
and SeaBASS (1 %).
es o p ese e his da ase as a whole. Also, no disca ding
hese da a allows u he manipula ion wi h di e en qual-
i y con ol c i e ia. On he annual scale, he obse a ions o
he inhe en op ical p ope ies a e s ongly unde ep esen ed
in he Sou he n Hemisphe e whe e he e is a comple e ab-
sence o da a in se e al mon hs o he yea (Fig. 3). O e -
all, he geog aphic co e age o obse a ions o aph, adg and
bbp (Fig. 14) is poo , wi h mos open ocean egions no be-
ing sampled, excep o he A lan ic Ocean. Small clus e s o
da a a e loca ed in pa icula coas al egions.
Finally, o he di use a enua ion coe icien o down-
wa d i adiance (kd) he e a e 25 unique wa eleng hs be-
ween 405 and 709 nm. The e is a o al o 2454 obse -
a ions om NOMAD (2266), SeaBASS (118) and MER-
MAID (70). Da a dis ibu ion o kd a 44Xand 55Xnm o
each da ase is shown in Fig. 12g and h. No kd da a a hese
wa eleng hs we e a ailable o he SeaBASS da ase (only
a 490 nm). Median alues o kd a 44Xnm span be ween
0.08 m−1(NOMAD) and 0.1 m−1(MERMAID), whe eas
a 55Xnm he kd alues a e app oxima ely 0.1 m−1(NO-
MAD and MERMAID). NOMAD p o ides he bes geo-
g aphical co e age (Fig. 15), wi h a highe co e age in he
A lan ic, compa ed wi h o he oceans. Wi h he excep ion o
he coas al egions o No h Ame ica and he Sea o Japan,
mos coas al egions a e no sampled. In he No he n Hemi-
sphe e, kd is dis ibu ed oughly e enly ac oss all mon hs o
he yea , bu in he Sou he n Hemisphe e he e a e ew da a
poin s du ing he aus al win e and none a all in Sep em-
be (Fig. 3). Fo o al suspended ma e ( sm) he e is a o-
al o 1546 obse a ions di ided be ween COASTCOLOUR
(1199) and MERMAID (347). The obse a ions o sm a e
a ailable in a g ea e numbe in he No he n Hemisphe e
(Fig. 3) and a e dis ibu ed ac oss se e al coas al egions
a ound Eu ope, he Medi e anean Sea, he Sou h China Sea,
Indonesia and Aus alia (Fig. 15).
Al hough mos o he s a ions wi h concu en a iables
a e om he NOMAD da ase , o comple eness, an exami-
na ion o bio-op ical ela ionships is p o ided (Fig. 16). The
ela ion be ween aph a 443 nm and chlo ophyll a(Fig. 16
a) ag ees wi h B icaud e al. (2004). A o al o 2953 poin s
exis wi h hese wo a iables a ailable (34 % om NO-
MAD, 32 % om TPSS, 11 % om AWI, 11% om COAST-
COLOUR, and he emaining 12 % om MERMAID and
SeaBASS). The ela ion be ween he sum o aph and adg a
443 nm and s a 443 nm (Fig. 16 b) shows a simila dispe -
sion, wi h he excep ion o some sca e ed poin s, o an equi -
alen analysis on he IOCCG epo 5 (see hei Fig. 2.3).
Again, he sca e ed da a we e e ained in he inal able o
p ese e he NOMAD da ase . A o al o 1112 poin s exis
o which hese h ee a iables a e a ailable (97 % om NO-
MAD). The ela ion be ween he a io s(490) / s(555) and
kd(490) (Fig. 16c) shows a good ag eemen wi h he NASA
KD2S s anda d algo i hm (h ps://oceancolo .gs c.nasa.go /
a bd/kd_490/). A o al o 2280 poin s exis o which hese
h ee a iables a e a ailable (93 % om NOMAD). The ela-
ion be ween he a io s(490) / s(555) and bbp a 555 nm
(Fig. 16 c) shows a good ag eemen wi h he ela ion sug-
ges ed by Tiwa i and Shanmugam (2013). A o al o 365
poin s exis o which hese h ee a iables a e a ailable
(89 % om NOMAD).
4 Da a a ailabili y
In o ma ion abou he da a a ailabili y can be ound in Ap-
pendix B.
5 Conclusions
In his wo k, a compila ion o bio-op ical in si u da a is p e-
sen ed, esul ing om he acquisi ion, homogeniza ion and
in eg a ion o se e al se s o da a ob ained om di e en
sou ces. The compiled da a ha e a global co e age and span
he pe iod om 1997 o 2018. Minimal changes we e made
o he o iginal da a, o he han he ones occu ing om con-
e sion o s anda d o ma and quali y con ol. In si u mea-
su emen s o he ollowing a iables we e compiled: emo e-
sensing e lec ance, chlo ophyll aconcen a ion, algal pig-
men abso p ion coe icien , de i al and colou ed dissol ed
o ganic ma e abso p ion coe icien , pa icle backsca e ing
coe icien , di use a enua ion coe icien o downwa d i a-
diance and o al suspended ma e .
The inal se o da a consis s o a subs an ial numbe o
in si u obse a ions, a ailable in a simple ex able and p o-
cessed in a way ha could be used di ec ly o he e alua ion
www.ea h-sys -sci-da a.ne /11/1037/2019/ Ea h Sys . Sci. Da a, 11, 1037–1068, 2019
1060 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
Table 3. Summa y o median alues o aph, adg and bbp a 44Xand 55Xnm o each da ase (as shown in Fig. 12a– ). Da a we e i s
sea ched o a 445 and 555 nm and hen wi h a sea ch window up o 8 nm o include da a a 547 nm.
Median aph Median adg Median bbp
44x nm 55x nm 44x nm 55x nm 44x nm 55x nm
SeaBASS 0.0549 0.0074 0.0711 0.0222 0.0035 0.0025
MERMAID 0.0282 0.0052 0.1149 0.0286 0.0080 0.0052
NOMAD 0.0353 0.0046 0.0515 0.0112 0.0030 0.0022
COASTCOLOUR 0.0665 0.0096 0.1259 0.0175 0.0047 0.0037
AWI 0.0208 0.0032 – – – –
TPSS 0.0454 0.0071 – – – –
Figu e 14. Global dis ibu ion o obse a ions o inhe en op ical p ope ies (algal pigmen abso p ion coe icien aph, de i al plus CDOM
abso p ion coe icien adg, and pa icle backsca e ing coe icien bbp) in he inal able.
Figu e 15. Global dis ibu ion o di use a enua ion coe icien o downwa d i adiance (kd) and o al suspended ma e ( sm) pe da ase
in he inal able. The sm and kd poin s om MERMAID o e lap each o he in he wes e n Black Sea (∼40◦N, 30◦E) and he A c ic
(∼70◦N, 120◦W).
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/
A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo 1061
Figu e 16. Examples o bio-op ical ela ionships in he inal me ged able: (a) aph(443) e sus chlo ophyll a. The o al numbe o poin s
(2953) is di ided be ween AWI (334), COASTCOLOUR (335), MERMAID (214), NOMAD (991), SeaBASS (124) and TPSS (955). Fo
e e ence he solid line shows he eg ession om B icaud e al. (2004). (b) [aph(443) +adg(443)] e sus s(443). The o al numbe o
poin s (1112) is di ided be ween MERMAID (33) and NOMAD (1079). (c) [ s(490) / s(555)] e sus kd(490). The o al numbe o poin s
(2280) is di ided be ween MERMAID (62), NOMAD (2117) and SeaBASS (101). Fo e e ence he solid line shows he NASA KD2S
s anda d algo i hm (h ps://oceancolo .gs c.nasa.go /a bd/kd_490/, las access: 10 July 2019). (d) [ s(490) / s(555)] e sus bbp(555). The
o al numbe o poin s (365) is di ided be ween MERMAID (33), NOMAD (324), and COASTCOLOUR+SeaBASS (4). Fo e e ence he
solid line shows he ela ion p oposed by Tiwa i and Shanmugam (2013). A sea ch window o 2 nm was used o panels (a) and (b), and a
sea ch window o 5 nm was used o panels (c) and (d) o include da a a 560 nm when no a ailable a 555 nm.
o sa elli e-de i ed ocean-colou da a. The majo ad an ages
o his compila ion a e ha i me ges six commonly used
da a sou ces in ocean-colou alida ion (MOBY, BOUS-
SOLE, AERONET-OC, SeaBASS, NOMAD and MER-
MAID), ou da a sou ces de eloped o ocean-colou ap-
plica ions (AWI, COASTCOLOUR, TPSS and TARA) and
17 addi ional se s o chlo ophyll aconcen a ion da a (AMT,
ICES, HOT,GeP&CO, ARCSSPP, BARENTSSEA, BATS,
BIOCHEM, BODC, CALCOFI, CCELTER, CIMT, ESTOC,
IMOS, MAREDAT, PALMER and SEADATANET) in o a
simple ex able ee o duplica ed obse a ions. This com-
pila ion was ini ially c ea ed wi h he in en ion o e alua -
ing he quali y o he sa elli e ocean-colou p oduc s om
he ESA OC-CCI p ojec , bu i can also be used o o he
www.ea h-sys -sci-da a.ne /11/1037/2019/ Ea h Sys . Sci. Da a, 11, 1037–1068, 2019
1068 A. Valen e e al.: A compila ion o global bio-op ical in si u da a – e sion wo
Taylo , B. B., To ecilla, E., Be nha d , A., Taylo , M. H., Peeken, I.,
Rö ge s, R., Pie a, J., and B ache , A.: Bio-op ical p o inces in
he eas e n A lan ic Ocean and hei biogeog aphical ele ance,
Biogeosciences, 8, 3609–3629, h ps://doi.o g/10.5194/bg-8-
3609-2011, 2011a.
Taylo , B. B., To ecilla, E., Be nha d , A., Taylo , M. H., Peeken, I.,
Rö ge s, R., Pie a, J., and B ache , A.: Phy oplank on pigmen s,
composi ion, hype spec al ligh ield da a and bioop ical p op-
e ies du ing POLARSTERN c uise ANT-XXV/1. PANGAEA,
h ps://doi.o g/10.1594/PANGAEA.819099, 2011b.
Thuillie , G., He sé, M., Labs, D., Foujols, T., Pee e mans, W.,
Gillo ay, D., Simon, P. C., and Mandel, H.: The sola spec al
i adiance om 200 nnm o 2400 nm as measu ed by he SOL-
SPEC spec ome e om he ATLAS 1-2-3 and EURECA mis-
sions, Sol. Phys., 214, 1–22, 2003.
Tiwa i, S. P. and Shanmugam, P.: An op ical model o de i ing he
spec al pa icula e backsca e ing coe icien s in oceanic wa e s,
Ocean Sci., 9, 987–1001, h ps://doi.o g/10.5194/os-9-987-2013,
2013.
T ees, C. C., Kennicu II, M. C., and B ooks, J. M.: E o s associ-
a ed wi h he s anda d luo ime ic de e mina ion o chlo ophylls
and phaeopigmen s, Ma . Chem., 17, 1–12, 1985.
Valen e, A., Sa hyend ana h, S., B o as, V., G oom, S., G an , M.,
Tabe ne , M., An oine, D., A none, R., Balch, W. M., Ba ke , K.,
Ba low, R., Bélange , S., Be hon, J.-F., Be¸sik epe, ¸S., B ando,
V., Canu i, E., Cha ez, F., Claus e, H., C ou , R., F ouin, R.,
Ga cía-So o, C., Gibb, S. W., Gould, R., Hooke , S., Kah u,
M., Klein, H., K a ze , S., Loisel, H., McKee, D., Mi chell, B.
G., Moisan, T., Mulle -Ka ge , F., O’Dowd, L., Ond usek, M.,
Poul on, A. J., Repecaud, M., Smy h, T., Sosik, H. M., Twa -
dowski, M., Voss, K., We dell, J., We nand, M., and Zibo di,
G.: A compila ion o global bio-op ical in si u da a o ocean-
colou sa elli e applica ions, Ea h Sys . Sci. Da a, 8, 235–252,
h ps://doi.o g/10.5194/essd-8-235-2016, 2016.
Valen e, A., Sa hyend ana h, S., B o as, V., G oom, S., G an , M.,
Tabe ne , M., An oine, D., A none, R., Balch, W. M., Ba ke ,
K., Ba low, R. G., Bélange , S., Be hon, J.-F., Besik epe, S.,
Bo sheim, Y., B ache , A., B ando, V., Canu i, E., Cha ez, F.,
Cianca, A., Claus e, H., Clemen son, L., C ou , R., F ouin,
R., Ga cía-So o, C., Gibb, S. W., Gould, R., Hooke , S. B.,
Kah u, M., Kampel, M., Klein, H., K a ze , S., Kudela, R.,
Ledesma, J., Loisel, H., Ma ai, P., McKee, D., Mi chell, B.
G., Moisan, T., Mulle -Ka ge , F., O’Dowd, L., Ond usek, M.,
Pla , T., Poul on, A., Repecaud, M., Sch oede , T., Smy h, T.,
Smy he-W igh , D., Sosik, H. M., Twa dowski, M., Vellucci,
V., Voss, K., We dell, J., We nand, M., W igh , S., and Zi-
bo di, G.: A compila ion o global bio-op ical in si u da a o
ocean-colou sa elli e applica ions – e sion wo, PANGAEA,
h ps://doi.o g/10.1594/PANGAEA.898188, 2019.
We dell, P. J. and Bailey, S. W.: An imp o ed bio-op ical da a se
o ocean colo algo i hm de elopmen and sa elli e da a p oduc
alida ion, Remo e Sens. En i on., 98, 122–140, 2005.
We dell, P. J., Bailey, S., Fa gion, G., Pie as, C., Knobelspiesse,
K., Feldman, G., and McClain, C.: Unique da a eposi o y acil-
i a es ocean colo sa elli e alida ion, EOS T ansac ions AGU,
84, 377–392, h ps://doi.o g/10.1029/2003EO380001, 2003.
Zhang, X., Hu, L., and He, M.-X.: Sca e ing by pu e seawa e : E -
ec o Salini y, Op . Exp ess, 17, 5698–5710, 2009.
Zibo di, G., Holben, B. N., Hooke , S. B., Mélin, F., Be hon, J.-F.,
Slu ske , I., Giles, D., Vandema k, D., Feng, H., Ru ledge, K.,
Schus e , G., and Al Mandoos, A.: A ne wo k o s anda dized
ocean colo alida ion measu emen s, EOS T ans. Am. Geophys.
Union, 87, 293–297, h ps://doi.o g/10.1029/2006EO300001,
2006.
Zibo di, G., Holben, B. N., Slu ske , I., Giles, D., D’Alimon e, D.,
Mélin, F., Be hon, J.-F., Vandema k, D., Feng, H., Schus e , G.,
Fabb i, B. E., Kai ala, S., and Seppälä, J.: AERONET-OC: A ne -
wo k o he alida ion o ocean colo p ima y adiome ic p od-
uc s, J. A mos. Ocean. Tech., 26, 1634–1651, 2009.
Zindle , C., B ache , A., Ma andino, C. A., Taylo , B.
B., To ecilla, E., Kock, A., and Bange, H. W.: Sul-
phu compounds, me hane, and phy oplank on du ing
SONNE c uise SO202/2 (T ansb om Sonne). PANGAEA,
h ps://doi.o g/10.1594/PANGAEA.820607, 2013.
Ea h Sys . Sci. Da a, 11, 1037–1068, 2019 www.ea h-sys -sci-da a.ne /11/1037/2019/