More Than 50 Years of Successful Continuous Temperature Section Measurements by the Global Expendable Bathythermograph Network, Its Integrability, Societal Benefits, and Future

SYSTEMATIC REVIEW published: 24 July 2019 doi: 10.3389/ ma s.2019.00452 F on ie s in Ma ine Science | www. on ie sin.o g 1July 2019 | Volume 6 | A icle 452 Edi ed by: Amos Tie eyangn Kabo-Bah, Uni e si y o Ene gy and Na u al Resou ces, Ghana Re iewed by: Eleano F ajka-Williams, Na ional Oceanog aphy Cen e, Uni e si y o Sou hamp on, Uni ed Kingdom John Pa ick Ab aham, Uni e si y o S . Thomas, Uni ed S a es *Co espondence: Gus a o J. Goni [email p o ec ed] Special y sec ion: This a icle was submi ed o Ocean Obse a ion, a sec ion o he jou nal F on ie s in Ma ine Science Recei ed: 28 Oc obe 2018 Accep ed: 08 July 2019 Published: 24 July 2019 Ci a ion: Goni GJ, Sp in all J, B ingas F, Cheng L, Ci ano M, Dong S, Domingues R, Goes M, Lopez H, Mo ow R, Ri e o U, Rossby T, Todd RE, T inanes J, Zilbe man N, Ba inge M, Boye T, Cowley R, Domingues CM, Hu chinson K, K amp M, Ma a MM, Reseghe i F, Sun C, Bhaska TVS U and Volko D (2019) Mo e Than 50 Yea s o Success ul Con inuous Tempe a u e Sec ion Measu emen s by he Global Expendable Ba hy he mog aph Ne wo k, I s In eg abili y, Socie al Bene i s, and Fu u e. F on . Ma . Sci. 6:452. doi: 10.3389/ ma s.2019.00452 Mo e Than 50 Yea s o Success ul Con inuous Tempe a u e Sec ion Measu emen s by he Global Expendable Ba hy he mog aph Ne wo k, I s In eg abili y, Socie al Bene i s, and Fu u e Gus a o J. Goni1*, Jane Sp in all2, F ancis B ingas1, Lijing Cheng3, Mau o Ci ano4, Shen u Dong1, Rica do Domingues1,5, Ma los Goes1,5, Hosmay Lopez1,5, Rosema y Mo ow6, Ulises Ri e o1, Thomas Rossby7, Robe E. Todd8, Joaquin T inanes1,5,9, Na halie Zilbe man2, Molly Ba inge 1, Tim Boye 10, Rebecca Cowley11, Ca ia M. Domingues12,13, Ka he ine Hu chinson14,15, Ma in K amp16, Mau icio M. Ma a17, F anco Reseghe i18, Cha les Sun10, Udaya Bhaska TVS19 and Denis Volko 1,5 1A lan ic Oceanog aphic and Me eo ological Labo a o y, Na ional Oceanic and A mosphe ic Adminis a ion, Miami, FL, Uni ed S a es, 2Sc ipps Ins i u ion o Oceanog aphy, Uni e si y o Cali o nia, San Diego, La Jolla, CA, Uni ed S a es, 3In e na ional Cen e o Clima e and En i onmen Science, Ins i u e o A mosphe ic Physics, Chinese Academy o Sciences, Beijing, China, 4Depa men o Me eo ology, Ins i u e o Geosciences, Fede al Uni e si y o Rio de Janei o, Rio de Janei o, B azil, 5Coope a i e Ins i u e o Ma ine and A mosphe ic S udies, Uni e si y o Miami, Miami, FL, Uni ed S a es, 6Labo a o ie d’E udes en Geophisique e Oceanog aphie Spa iales, Toulouse, F ance, 7G adua e School o Oceanog aphy, Uni e si y o Rhode Island, Na aganse , RI, Uni ed S a es, 8Woods Hole Oceanog aphic Ins i u ion, Woods Hole, MA, Uni ed S a es, 9Technological Resea ch Ins i u e, Uni e si y o San iago de Compos ela, San iago de Compos ela, Spain, 10 Na ional Oceanic and A mosphe ic Adminis a ion, Na ional Cen e s o En i onmen al In o ma ion, Sil e Sp ing, MD, Uni ed S a es, 11 Commonweal h Scien i ic and Indus ial Resea ch O ganisa ion, Oceans and A mosphe e, Hoba , TAS, Aus alia, 12 ACE CRC, CLEX, Uni e si y o Tasmania, Hoba , TAS, Aus alia, 13 Ins i u e o Ma ine and An a c ic S udies, Uni e si y o Tasmania, Hoba , TAS, Aus alia, 14 Oceanog aphy Depa men , Uni e si y o Cape Town, Cape Town, Sou h A ica, 15 Labo a oi e LOCEAN/IPSL, So bonne Uni e si és (UPMC Uni e si és Pa is 06) CNRS-IRD-MNHN, Pa is, F ance, 16 The JCOMM in si u Obse a ions P og amme Suppo Cen e, Plouzané, F ance, 17 Ins i u e o Oceanog aphy, Fede al Uni e si y o Rio G ande (FURG), Rio G ande-RS, B azil, 18 ENEA, I alian Na ional Agency o New Technologies, Ene gy, and Sus ainable Economic De elopmen , San a Te esa Resea ch Cen e, Le ici, I aly, 19 Indian Na ional Cen e o Ocean In o ma ion Se ices, Minis y o Ea h Science, Hyde abad, India The i s eXpendable Ba hyThe mog aphs (XBTs) we e deployed in he 1960s in he No h A lan ic Ocean. In 1967 XBTs we e deployed in ope a ional mode o p o ide a con inuous eco d o empe a u e p o ile da a along epea ed ansec s, now known as he Global XBT Ne wo k. The cu en ne wo k is designed o moni o ocean ci cula ion and bounda y cu en a iabili y, basin-wide and ans-basin ocean hea anspo , and global and egional hea con en . The abili y o he XBT Ne wo k o sys ema ically map he uppe ocean he mal ield in mul iple basins wi h epea ed ans-basin sec ions a eddy- esol ing scales emains unma ched oday and canno be ep oduced a p esen by any o he obse ing pla o m. Some epea ed XBT ansec s ha e now been con inuously occupied o mo e han 30 yea s, p o iding an unp eceden ed long- e m clima e eco d Goni e al. Global XBT Ne wo k o empe a u e, and geos ophic eloci y p o iles ha a e used o unde s and a iabili y in ocean hea con en (OHC), sea le el change, and me idional ocean hea anspo . He e, we p esen key scien i ic ad ances in unde s anding he changing ocean and clima e sys em suppo ed by XBT obse a ions. Imp o emen in XBT da a quali y and i s impac on compu a ions, pa icula ly o OHC, a e p esen ed. Technology de elopmen o p obes, launche s, and ansmission echniques a e also discussed. Finally, we o e new pe spec i es o he u u e o he Global XBT Ne wo k. Keywo ds: expendable ba hy he mog aphs, su ace cu en s, subsu ace cu en s, me idional hea anspo , ocean hea con en , sea le el, ex eme wea he INTRODUCTION EXpendable Ba hyThe mog aphs (XBTs) a e ins umen s ha p o ide he simples and mos cos -e icien solu ion o equen ly ob aining empe a u e p o iles along ixed ansec s o he uppe housand me e s o he ocean. XBTs ha e been his o ically deployed by na ies, esea ch essels, and me chan ships. The i s XBT p obes we e es ed in 1959, and sys ema ic deploymen o XBTs began in he mid o la e 1960s. XBTs he ea e became he la ges sou ce o da a o he uppe ocean he mal eco d du ing he 1970s−1990s, wi h ∼89,000 XBTs deployed in 1990. XBTs hus p o ide one o he longes a ailable his o ical eco ds o uppe ocean empe a u e p o iles ( o ∼1,000 m dep h). Cu en ly, XBTs deployed along ixed ansec s a e g ouped in o wha cons i u es he Global XBT Ne wo k (Figu e 1, op panel). Du ing he pas 10 yea s, 15,000– 20,000 XBTs ha e been deployed annually. Mos o he XBTs being cu en ly deployed a e om he Deep Blue ype, which can each dep hs o 800 m (Cheng e al., 2014). Obse a ions om he Global XBT Ne wo k p o ide epea ed sec ions o empe a u e along ixed ansec s ha c oss egions ha a e c i ical o moni o ing, unde s anding, and assessing su ace and subsu ace dynamical p ocesses ha occu in he uppe ocean. Da a om he Global XBT Ne wo k ha e been used ex ensi ely o es ima e a iabili y and changes in nea - su ace ocean p ope ies (e.g., hea con en ) and dynamics (e.g., Le i us e al., 2012). XBT obse a ions in o med much o wha is known abou a iabili y and changes in global and egional uppe -ocean hea con en (OHC) be o e he nea - global A go p o iling loa a ay was implemen ed (Rise e al., 2016; Jayne e al., 2017). XBT obse a ions a e ex emely aluable in nea -coas al egions and in some a eas o he open ocean whe e hey a e he sole sou ce o epea ed hyd og aphic obse a ions ha esol e mesoscale ea u es o assessing anspo s. The cu en Global XBT Ne wo k collec s obse a ions a spa ial and empo al scales ha canno easibly be duplica ed by o he obse a ional pla o ms. While pla o ms such as p o iling loa s (Rise e al., 2016) and unde wa e glide s (Rudnick, 2016) now p o ide empe a u e p o iles, hey canno occupy epea ed, mesoscale- esol ing, ans-ocean basin ansec s ac oss majo cu en s on he ime scales ha a e egula ly sampled using XBTs om as -mo ing ships. Obse a ions om XBTs and om o he p o iling pla o ms should be seen as complemen a y. Fo example, XBTs p o ide a ge ed obse a ions in speci ic egions, while A go loa s p o ide backg ound in o ma ion needed o unde s and he p ocesses ha lead o he a iabili y obse ed by XBT obse a ions (Figu e 2). In addi ion, colloca ed obse a ions om XBTs and o he componen s o he Global Ocean Obse ing Sys em (GOOS) can be used o iden i y and assess po en ial e o s o biases wi hin he obse ing sys em. XBT obse a ions a e cu en ly mainly used o: 1) Moni o he s a e and spa ial and empo al a iabili y o key su ace and subsu ace ocean cu en s and bounda y cu en s, including hei anspo ; 2) Moni o he s a e and a iabili y o he Me idional Hea T anspo (MHT) and Me idional O e u ning Ci cula ion (MOC) ac oss ocean basins; 3) P o ide uppe ocean he mal obse a ions o es ima e global and egional OHC in a eas unde sampled by o he obse a ional pla o ms; 4) Ini ialize and alida e Ocean Fo ecas ing Sys ems; and 5) P o ide cons ain s h ough da a assimila ion o ocean eanalysis hindcas s. The wo spa ial modes o XBT deploymen cu en ly used in he Global XBT Ne wo k a e: 1. High Densi y o High Resolu ion (HD/HR): Usually ou o mo e epe i ions a e conduc ed annually along a ixed ansec wi h an a e age o one XBT deploymen abou e e y 10–50 km along he ship ack (35 XBT deploymen s pe day a a ship speed o 20 k s). This mode is aimed a ob aining high spa ial esolu ion in a single ealiza ion o esol e he spa ial s uc u e o mesoscale eddies, on s, and bounda y cu en s. These ansec s a e designed o esol e bounda y cu en s and o es ima e basin-scale geos ophic eloci y and mass and hea anspo s, including he MOC, and hea anspo . This is cu en ly he mos widely used deploymen mode. 2. F equen ly Repea ed (FR): Twel e o mo e epe i ions a e conduc ed annually along a ixed ansec , wi h six o mo e XBT deploymen s pe o med daily along he ansec e e y 100–150 km. This mode is aimed a ob aining epea su eys along hose ansec s whe e he e is high empo al a iabili y. This sampling mode is designed o p oduce well- esol ed mon hly ime se ies ha obse e speci ic ea u es o he he mal s uc u e (e.g., he mocline idges) o ha ob ain F on ie s in Ma ine Science | www. on ie sin.o g 2July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k FIGURE 1 | (Top) Loca ion o he HD/HR and FR XBT ansec s ecommended by he XBT Science Team du ing he 5 h Science Wo kshop held in Oc obe 2016. Table 1 p o ides a de ailed explana ion o each ansec . (Bo om) Annual pe o mance o he XBT ne wo k. Ta ge s a e se by he XBT Science Team and compa ed o ealiza ion numbe s o each ansec . The KPI de ined by he o al numbe o ealiza ion numbe s di ided by he numbe o all a ge ed ealiza ions is 82% o 2018. samples whe e in aseasonal a iabili y is s ong (e.g., he Indonesian Th ough low). The cu en ly ope a ed ansec s (Figu e 1, op panel) ollow ecommenda ions om he in e na ional e iew o he global uppe ocean he mal ne wo k (Smi h e al., 2001), OceanObs’99, OceanObs’09 (Goni e al., 2010), and ecen ecommenda ions om he XBT Science Team. P o iles om abou 90% o he XBT deploymen s a e ansmi ed in nea eal- ime in o he Global Telecommunica ion Sys em (GTS), making up ∼15% o he cu en eal- ime e ical empe a u e p o ile obse a ions (no including he con inuous empe a u e p o iles made by some moo ings). Some XBT ansec s ha e been in ope a ion o mo e han 30 yea s, he eby p o iding unique and aluable clima e eco ds. Fo example, AX10 (New Yo k o San Juan) has p o ided key in o ma ion abou he a iabili y in uppe ocean empe a u e wi hin he Gul S eam o mo e han 55 yea s (Molina i, 2004). PX06 (Auckland o Fiji) has been occupied since 1986 and was he i s ansec sampled in HD/HR mode; i has now been sampled mo e han 90 imes o e 30 yea s. In he Indian and Paci ic oceans, he FR ansec s IX01 (Wes e n Aus alia o Ja a) and PX02 (Da win, Aus alia o Indonesia) ha e been sampled o mo e han 35 yea s. Since he implemen a ion o he A go a ay in 1999 o sample he ocean in e io (Gould e al., 2004; Rise e al., 2016), he ocus o he XBT a ay has been o p ima ily moni o bounda y cu en s and ans- basin sec ions ha cap u e he me idional anspo o hea and mass. This e iew p esen s he cu en s a e o he Global XBT Ne wo k, majo scien i ic ad ances esul ing om he decades- long XBT eco d, and syne gy be ween he Global XBT Ne wo k and o he componen s o he obse ing sys em. Examples o how he XBT ne wo k con ibu es o bo h ope a ional oceanog aphy F on ie s in Ma ine Science | www. on ie sin.o g 3July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k FIGURE 2 | Loca ion o XBT ( ed) and A go loa (blue) measu emen s du ing he yea s 2007 (Top) and 2017 (Bo om) ha show he di e ence o spa ial sampling p oduced by hese obse ing pla o ms. The numbe s indica e he obse a ions made by each pla o m du ing hese 2 yea s. and moni o ing he s a e o he ocean, pa icula ly wi h espec o he MOC, OHC, and sea le el change, and ex eme wea he e en s, a e also highligh ed. XBT OPERATIONAL AND SCIENTIFIC OVERSIGHT XBT ope a ions a e coo dina ed on a global scale by he Ship O Oppo uni y P og amme Implemen a ion Panel (SOOPIP), a ne wo k o he Ship Obse a ions Team (SOT) which ope a es unde he amewo k o he Join Technical Commission o Oceanog aphy and Ma ine Me eo ology (JCOMM) o he Wo ld Me eo ological O ganiza ion (WMO) and UNESCO’s In e go e nmen al Oceanog aphic Commission (IOC). The JCOMM Obse a ions P og amme Suppo Cen e (JCOMMOPS) is asked wi h moni o ing he ope a ional e o s o he SOOPIP. I also implemen s Key Pe o mance Indica o s (KPI) and s a us maps o he cu en Global XBT ne wo k, e/de ined by he in e na ional communi y (Figu e 1, bo om panel). The Global XBT Ne wo k is a key componen o GOOS ha add esses hemes ela ed o clima e, ope a ional se ices, and ma ine ecosys em heal h. The ne wo k di ec ly o indi ec ly measu es Essen ial Ocean and Clima e Va iables, such as sea su ace empe a u e, subsu ace empe a u e, su ace and subsu ace cu en s, and ocean su ace hea lux. The scien i ic o e sigh and jus i ica ion is p o ided and assessed by he XBT Science Team. Scien i ic aspec s o XBT obse a ions a e discussed wi hin he XBT Science Team, which was c ea ed in 2011 and consis s o mo e han 30 expe s and scien is s om 19 ins i u ions and 10 coun ies. The ocus o his eam is o: •P o ide a oice in he scien i ic communi y o communica e XBT- ela ed esul s; •O ganize mee ings o he XBT communi y o discuss scien i ic ad ances in he use o XBT obse a ions; •Enhance in e na ional scien i ic collabo a ion; •Make ecommenda ions and p io i ize ansec s o he XBT ne wo k; •Make ecommenda ions on XBT da a managemen ; •Cul i a e links o ac i e and ecognized scien i ic and ope a ional panels o o he obse ing pla o ms. The XBT Science Team websi e1p o ides easy access o XBT da a, XBT-de i ed p oduc s and indica o s, and o he XBT- ela ed scien i ic and ope a ional in o ma ion. I also b ings scien is s oge he o highligh he uses o XBT da a, including uppe ocean he mal s uc u e and a iabili y, ocean cu en s, and hea anspo . 1www.aoml.noaa.go /phod/goos/xb science F on ie s in Ma ine Science | www. on ie sin.o g 4July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k In addi ion o he XBT Science Team, scien is s and ope a o s in ol ed in he XBT ne wo k pa icipa e in in e na ional panels ha add ess many aspec s o XBT ope a ions, da a managemen , and science. These panels p o ide a wide ange o ecommenda ions gea ed owa d in e disciplina y and complemen a y s udies, he con inuous epo ing o esea ch highligh s, and imp o emen o he XBT ne wo k in eg a ion wi h he GOOS. Some o hese panels a e: •SOOPIP: This panel coo dina es he ope a ional and da a managemen s anda ds o he implemen a ion and main enance o he Global XBT Ne wo k om olun ee ships. •IQuOD: The In e na ional Quali y con olled Ocean Da abase (IQuOD) p ojec ocuses on he c ea ion and dis ibu ion o a comple e, high quali y single ocean p o ile eposi o y, including me ada a, and assigned unce ain ies, mos ly o use in ocean clima e esea ch applica ions, da a assimila ion, and model e alua ion2. •GOSUD: The Global Ocean Su ace Unde way Da a (GOSUD) P ojec is an IOC p og am dedica ed o assembling and dis ibu ing quali y-con olled da a se s o unde way sea su ace empe a u e and salini y obse a ions collec ed by ca go ships and esea ch essels. •GTSPP: The Global Tempe a u e and Salini y P o ile P og am (GTSPP) p o ide essen ial subsu ace clima e a iables o empe a u e and salini y p o ile da a, as well as imely and comple e da a wi h documen ed quali y lags. I implemen s in e na ionally ag eed upon quali y con ol s anda ds and manages ocean da a in acco dance wi h he GOOS ac ion plan. KEY XBT SCIENTIFIC CONTRIBUTIONS Since he incep ion o he XBT ne wo k, XBT obse a ions ha e led o pionee ing esea ch ela ed o OHC, ocean cu en a iabili y, and wa e mass and hea anspo s. The con ibu ions o XBT obse a ions o scien i ic esea ch ha e been highligh ed in housands o publica ions and ha e also p o ided he basis o many s uden heses and disse a ions. On a e age, abou 100 pee - e iewed manusc ip s ha use XBT da a a e published annually. XBTs ha e p o ided some o he longes con inuous eco ds o ocean cu en s, wi h many o he exis ing ansec s su passing 30 yea s o unin e up ed obse a ions ac oss ocean basins on a leas a qua e ly basis. These include he su eillance o na ow bounda y cu en egions ha he global A go a ay wi h i s 3-deg ee spacing canno esol e. XBTs a e one o he ew obse a ional pla o ms capable o long- e m moni o ing o ocean cu en p ope ies a he su ace and a subsu ace dep hs and o measu ing ans-oceanic empe a u e sec ions a an eddy- esol ing esolu ion. The main enance o sus ained empe a u e p o ile obse a ions along hese ixed ansec s is c i ical o long- e m moni o ing o he p ope ies o key ocean cu en s and in eg a ed anspo ac oss basins. Scien is s om he XBT communi y ha e been success ul in de eloping and implemen ing no el me hodologies, including 2www.iquod.o g mul ipla o m and mul i a iable assessmen s, ha ha e become s anda d o moni o ing and analyzing he s a e and a iabili y o he ocean. In wha ollows, sec ion he complemen a i y o XBTs wi h o he obse ing pla o ms highligh s s udies ha discuss he syne gy o XBT ansec s wi h o he componen s o he global obse ing sys em. Sec ion Ocean cu en s, gy es, and ocean a iabili y shows examples o how XBT moni o ing has imp o ed unde s anding o ocean cu en s, gy es, and ocean a iabili y, while sec ions me idional hea anspo , global and egional ocean hea con en , and ope a ional oceanog aphy and ocean o ecas s highligh MHT, global/ egional OHC, and ope a ional oceanog aphy/ocean o ecas s, espec i ely. Sec ion Socie al bene i s o XBT obse a ions p o ides an o e iew o he socie al bene i s o XBT obse a ions, sec ion Da a managemen add esses XBT da a managemen , and sec ion Technological Imp o emen s discusses echnological imp o emen s. Finally, sec ion he u u e o he Global XBT Ne wo k p esen s he ision o he au ho s on he u u e o he Global XBT Ne wo k. The Complemen a i y o XBTs Wi h O he Obse ing Pla o ms Se e al s udies ha e combined XBT p o iles wi h colloca ed Conduc i i y, Tempe a u e, and Dep h (CTD), A go, and sa elli e al ime y obse a ions o es ablish, o example, a s a is ical dynamic heigh ela ionship. By linking dynamic heigh o cumula i e ba oclinic anspo ac oss an XBT sec ion, al ime ic dynamic heigh can be used o ex end he XBT sec ions in o a nea -con inuous long- e m ime se ies o ba oclinic anspo . The syne gy be ween XBT empe a u e p o iles and sea su ace heigh measu ed by sa elli es has been used ex ensi ely o moni o se e al cu en sys ems and egions, including he An a c ic Ci cumpola Cu en (ACC) sou h o Tasmania (Rin oul e al., 2002), he Agulhas e o lec ion and ACC on s sou h o A ica (Swa e al., 2008), he ACC on s in he D ake Passage (Sp in all, 2003), he Eas Aus alian Cu en (Zilbe man e al., 2018), ac oss he No h Paci ic gy e (Roemmich and Gilson, 2001), he B azil Cu en (Goni and Waine , 2001), he No h B azil Cu en (Fonseca e al., 2004), he Eas India Coas al Cu en s in he Bay o Bengal (She in e al., 2018), he Gul S eam (Molina i, 2011), and he Flo ida Cu en (Olson e al., 1983; Domingues e al., 2018). Sec ion Ocean cu en s, gy es, and ocean a iabili y shows examples o how XBT obse a ions a e in eg a ed wi h da a om o he obse ing pla o ms o assess he s a e and a iabili y o he ocean. The complemen a i y o XBT obse a ions o da a p o ided by o he obse ing pla o ms a e u he shown in his issue o MHT (F ajka-Williams e al., 2019) and bounda y cu en s (Todd e al., 2019). Ocean Cu en s, Gy es, and Ocean Va iabili y Gul S eam The Gul S eam, he Wes e n Bounda y Cu en (WBC) o he No h A lan ic, has been linked o changes in a ious wea he and clima e phenomena, including ex eme wea he e en s o e he No hwes A lan ic, he A lan ic Me idional O e u ning Ci cula ion (AMOC), and coas al sea le el ise (La i e al., 2000; F on ie s in Ma ine Science | www. on ie sin.o g 5July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k Hoskins and Hodges, 2002; Joyce e al., 2009; Kelly e al., 2010; Kwon e al., 2010). Fou XBT ansec s moni o he Gul S eam a di e en loca ions: AX08 (Cape Town o New Yo k), AX10 (New Yo k o Pue o Rico), AX32 (New Yo k o Be muda), and AXWBTS (Palm Beach, FL, o G and Bahama). The i s sus ained ime se ies o he posi ion o he Gul S eam, beginning in he ea ly 1950s, was ob ained by combining mechanical ba hy he mog aph measu emen s wi h XBT da a along AX10. These obse a ions showed ha me idional mig a ion o he Gul S eam is s ongly co ela ed wi h he No h A lan ic Oscilla ion (NAO) on decadal ime-scales and ha he me idional mig a ion is also simila o anomalies in Gul S eam uppe laye anspo and an eas -wes ex ension o he Gul S eam sou he n eci cula ion gy e (Molina i, 2004). The Gul S eam be ween he no heas e n Uni ed S a es and Be muda has been su eyed o nea ly 150 yea s. The H.M.S. Challenge collec ed he oldes documen ed empe a u e sec ion ac oss he Gul S eam in 1873 (Rossby e al., 2010). Be ween he la e 1960s and ea ly 1970s, he US Na al Oceanog aphic O ice made a la ge numbe o high esolu ion XBT sec ions om a ious passenge essels be ween 40◦N ( he ou e con inen al shel ) and 35◦N. These da a a e cu en ly being eassembled and will be a chi ed c uise-by-c uise. Since 1977 he Global XBT Ne wo k has included XBT deploymen s ac oss he shel ou o and some imes in o he Gul S eam on a mon hly basis along AX32 and in HD mode wi h ansec s AX10 and AX08. In la e 1992, a p og am o measu e uppe ocean cu en s along he New Je sey-Be muda sec ion was implemen ed using he M/V Oleande , a con aine essel. As pa o his e o , addi ional XBTs a e now being deployed ac oss he Gul S eam on a mon hly basis. A ecen analysis o he 20-yea ime se ies o AX10 HD da a (Figu e 3A) shows ha he Gul S eam expe iences s ong no h-sou h shi s, which can exceed wo deg ees o la i ude on seasonal ime scales. Howe e , he cu en i sel has no exhibi ed signi ican long- e m ends in loca ion (Figu e 3B) o in anspo (Figu e 3C). Ongoing esea ch indica es ha 20 yea s o measu emen s using AX10 da a show ha below he seasonal mixed laye he la ges empe a u e a iabili y in he Gul S eam occu s be ween 300 and 600 m dep h. This is impo an because sub opical mode wa e s a e ound wi hin his dep h ange. Geos ophic eloci y es ima ed om each AX10 sec ion using empe a u e measu emen s om XBTs and salini y in e ed om he his o ical T-S ela ionship (Goes e al., 2018) shows ha he empo al a ia ions in he XBT-de i ed geos ophic eloci y es ima es a e e ically cohe en . Combining AX10 obse a ions wi h sa elli e al ime y obse a ions has also esul ed in imp o ed unde s anding o Gul S eam changes o e a la ge egion (50◦-80◦W). Du ing 1993–2016, he Gul S eam was ound o expe ience a s ong sou hwa d shi eas o 65◦W a e passing he New England Seamoun chain (Figu e 4A). This sou hwa d shi was accompanied by a weakening o he Gul S eam (Figu es 4B,C). Wes o 70◦W, howe e , he obse ed FIGURE 3 | (A) Time-mean Gul S eam posi ion om sa elli e al ime y ( ed) and he loca ion o he XBT AX10 ansec (black). (B) Gul S eam posi ion om AX10 in deg ee la i ude. (C) Gul S eam anspo in he uppe 800 m wa e column om AX10 ( e e enced o 800 m); uni s a e S e d ups (106m3s−1). The backg ound colo shows he bo om dep h. FIGURE 4 | Zonally a e aged Gul S eam (A) posi ion, (B) speed, and (C) c oss- on sea su ace heigh (SSH) di e ence (p oxy o anspo ) de i ed om sa elli e al ime y. F on ie s in Ma ine Science | www. on ie sin.o g 6July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k FIGURE 5 | Dep h- ime diag am o he a e age empe a u e esiduals (seasonal cycle emo ed and one 1-yea low-pass il e applied) in he Flo ida S ai s using subsu ace empe a u e p o ile da a de i ed om 1,925 XBT p o iles (AX07 and AXWBTS) and 541 CTD cas s sampled du ing 1995–2016. ends du ing 1993–2006 we e e y weak. This ype o s udy is impo an because he sea su ace empe a u e (SST) g adien associa ed wi h he Gul S eam con ibu es signi ican ly o he g ow h o midla i ude s o m ac i i y, s o m acks, and in ensi y (Chang e al., 2002; Kushni e al., 2002; Nakamu a e al., 2004). Flo ida Cu en The Flo ida Cu en is he WBC ha eeds in o he Gul S eam and ca ies bo h he e u n low om he sub opical wind-d i en gy e and he uppe b anch o he AMOC. The Flo ida Cu en is ou inely moni o ed by wo XBT ansec s: AX07 (Miami o Gib al a ) and AXWBTS. While he AMOC has long been ecognized as an impo an componen o he clima e sys em, changes in he in ensi y o he Flo ida Cu en anspo and hea ca ied by he cu en ha e also been ecen ly acknowledged as key d i e s o egional sea le el changes along he US Eas Coas (Eze , 2013; Domingues e al., 2016). An analysis o XBT p o iles om hese ansec s e eals subs an ial yea - o- yea changes in he Flo ida Cu en empe a u e, which can exceed ±1◦C o e he ull ime eco d (Figu e 5). The ime se ies e eals ha empe a u e anomalies a e mos ly cohe en h oughou he en i e wa e column (e.g., la e 2015), al hough ∼30% o he ime he anomalies abo e and below 100 m ha e opposi e signs (e.g., ea ly 1997). XBT da a also e ealed an unp eceden ed wa ming o he Flo ida Cu en du ing 2014– 2015, which ollowed a ela i ely cold pe iod in 2010–2013. Du ing he 2014–2015 e en , he en i e wa e column in he Flo ida S ai s was ∼0.5◦C wa me han a e age condi ions. As discussed in sec ion egional sea le el changes, hese changes a e key d i e s o coas al sea le el anomalies in he egion. Tempe a u e changes in he Flo ida Cu en a e also ound o be unco ela ed wi h changes in he in ensi y o i s low (Domingues e al., 2018). These phases o wa ming and cooling o he Flo ida Cu en ha e impo an impac s on egional sea le el changes along he US Sou heas Coas . B azil Cu en The B azil Cu en (BC) is he WBC o he Sou h A lan ic sub opical gy e. The e a e wo XBT ansec s ha c oss he BC: AX18 (Buenos Ai es o Cape Town) a 34◦S and AX97 (Rio de Janei o o Ilha da T indade) a 22◦S. S a ed in 2002 and 2004, espec i ely, AX18 and AX97 a e he longes con inuous e o s o assess he s uc u e and a iabili y o he BC. The BC is o key impo ance in closing he mass budge in he Sou h A lan ic, since i is he WBC ha closes he sub opical gy e, anspo ing wa e s om subpola egions, hus cons i u ing an in eg al pa o he AMOC. Un il he implemen a ion o hese wo XBT ansec s, mos o he BC obse a ions elied on spa se c uise da a, sho pe iod moo ing deploymen s, o models. A ecen s udy (Lima e al., 2016) used geos ophic eloci y ields cons uc ed om AX97 da a o show ha models gene ally mis ep esen he s uc u e o he a iabili y o his cu en , simula ing i as oo deep, and oo wide. AX97 ansec s ha e esol ed he high mesoscale a iabili y associa ed wi h he BC ha can mani es in insho e o o sho e s a es, depending on ansien eddies and he semi-pe manen Cape o São Tomé eddy (Mill e al., 2015). Du ing he summe o 2009–2010, an ex eme wa m SST e en (>3◦C) was iden i ied nea 22◦S o he coas o B azil, which was associa ed wi h a mosphe ic eleconnec ions om a Cen al Niño e en in he Paci ic (Majumde e al., 2019). Du ing he wa m SST e en , he XBT-de i ed geos ophic BC anspo (12 S ) was h ee imes la ge han a e age. This anomalous anspo was physically linked o inc eased coas al upwelling and ba oclinici y in he egion (Goes e al., 2019). These p ocesses enhance he SST g adien ac oss he BC o Cabo F io, B azil, which gene a es wind con e gence/cu l and hickens he a mosphe ic bounda y laye , impac ing local wea he and p ecipi a ion (e.g., Pezzi e al., 2016). Fu u e wo k will include assessing he sub opical gy e a iabili y and BC on al changes o egional wea he pa e ns. Eas India Coas al Cu en in he Bay o Bengal The uppe laye ci cula ion o he Bay o Bengal (BoB) is known o ha e s ong seasonal a iabili y (Eigenhee and Quad asel, 2000). Du ing he no heas monsoon, he Eas India Coas al Cu en (EICC) is he WBC o he BoB and lows equa o wa d along he eas coas o India o S i Lanka. She in e al. (2018) used 27 yea s o epea ed XBT sec ions ha c oss he wes e n (Chennai o Po Blai ) and no hwes e n (Kolka a o Po Blai ) egions o he BoB o s udy he EICC and i s in e annual a iabili y. The EICC was ound o be seasonally e e sing, lowing polewa d om Feb ua y o July wi h a anspo o 5 S and hen lowing equa o wa d om Oc obe o Decembe wi h a anspo o 3 S . In Ma ch, 7 S in he EICC lows no heas wa d in he no hwes e n BoB. Weak no hwes wa d low (2 S a mos ) occu s du ing he emainde o he calenda yea . The Indian Ocean Dipole (IOD) is ound o ha e a signi ican in luence on EICC a iabili y. Remo e wind o cing om he equa o ial Indian Ocean associa ed wi h he EICC gene a es a no hwa d (sou hwa d) anomalous anspo o 5 S (7 S ) du ing win e o posi i e (nega i e) IOD e en s (She in e al., 2018). F on ie s in Ma ine Science | www. on ie sin.o g 7July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k FIGURE 6 | Simul aneous co ela ion o mon hly anomalies o opical A lan ic SST (con ou ) and pseudo wind s ess ( ec o s) wi h he anspo o (A) No h Equa o ial Unde cu en (NEUC) and (B) No h Equa o ial Coun e cu en (NECC) a in e annual imescales. Anomalies a e calcula ed ela i e o he mon hly means. The boxes ep esen he egions o maximum co ela ion. T opical A lan ic Cu en Sys em The AX08 ansec moni o s and assesses he opical A lan ic sys em o su ace and subsu ace cu en s and coun e cu en s a ∼23◦W. AX08 ansec da a and sa elli e-de i ed sea heigh ields e ealed ha al ime y da a alone could no be used o iden i y and moni o all cu en s in he opical A lan ic, pa icula ly he unde cu en s (Goni and Ba inge , 2002). In a mo e ecen s udy, Goes e al. (2013a) combined XBT da a wi h his o ical empe a u e-salini y ela ionships, al ime ic sea le el anomalies, and A go-based s e ic heigh da a o es ima e densi y and eloci y p ope ies o he opical A lan ic eas wa d cu en s o he en i e al ime ic pe iod (1992-p esen ). Goes e al. (2013a) associa ed he a iabili y o he No h Equa o ial Unde cu en (NEUC) and No h Equa o ial Coun e cu en (NECC) wi h he main modes o in e annual a iabili y in he opical A lan ic (Figu e 6), pa icula ly he A lan ic Me idional Mode (AMM) and associa ed excu sions o he In e opical Con e gence Zone. The NECC and NEUC anspo s we e ound o be ou -o - phase; he NECC (NEUC) is associa ed wi h posi i e (nega i e) AMM and led by he s eng hening (weakening) o he ade winds. Al hough sa elli e al ime y measu emen s ha e su icien empo al and spa ial esolu ion o esol e mos o he highly a iable su ace p ocesses nea he equa o , he XBT da a we e c i ical in sampling he e ical and me idional s uc u e o he subsu ace cu en s, which a e gene ally be ween 200 and 300 m deep and 100–150 km wide (Goes e al., 2013a). The Eas Aus alian Cu en , he Eas Auckland Cu en , and he Tasman Sea XBT ansec s PX30 (B isbane o Fiji) and PX34 (Sydney o Welling on) c oss he Eas Aus alian Cu en (EAC), he WBC o he Sou h Paci ic gy e. XBT ansec PX06 (Auckland o Fiji) c osses he Eas Auckland Cu en (EAuC). These ansec s a e among he longes unning HR lines in he Global XBT Ne wo k and ha e now been sampling along nea - epea ansec s o o e 30 yea s (Table 1). Geos ophic eloci y es ima es ob ained by combining XBT and sa elli e al ime y da a ha e shown ha he eas wa d low om he sepa a ed EAC occu s in dis inc pe manen ilamen s (Ridgway and Dunn, 2003; Ridgway e al., 2008), demons a ing he banded na u e o he mean eloci y ield. Hill e al. (2011) showed ha sou hwa d anspo in he Tasman Sea is s ongly an i-co ela ed wi h he eas wa d anspo o he Tasman F on (PX06) no h o New Zealand. Mo eo e , a mul i-decadal sou hwa d shi in he Sou he n Hemisphe e wes e ly winds has esul ed in less eas wa d anspo in he Tasman F on and g ea e sou hwa d anspo in he EAC Ex ension. This wo k, ollowing a p e ious analysis by Roemmich e al. (2005), sheds ligh on no only long- e m empe a u e and salini y ends in he Tasman Sea bu also he ecosys em impac s o clima e change in he EAC sys em. These XBT da a ha e signi ican ly con ibu ed o ou unde s anding o he mass and hea budge s in he Tasman egion and he o ma ion, sp eading, cha ac e is ics, and a iabili y o Sou h Paci ic Sub opical Mode Wa e (Roemmich and Co nuelle, 1992; Roemmich e al., 2005; Tsubouchi e al., 2007; Holb ook and Maha aj, 2008). Conside able e o o e he pas 10 yea s has ocused on expanding ou knowledge o he empo al a iabili y o he EAC and EAuC anspo s a in e annual o decadal ime scales, al hough unce ain ies emain. The XBT-de i ed anspo ime- se ies show in e annual a iabili y wi h a pe iod o abou 4 yea s and a decadal end owa d lowe eas wa d anspo (Hill e al., 2008). This end is consis en wi h changes in he wind s ess cu l ha a e belie ed o ha e caused he EAC o ex end a he sou h o e he pas decade (Cai e al., 2005; Roemmich e al., 2007; Hill e al., 2011). In e es ingly, in con as o he EAC, he e has been no signi ican end in he EAuC anspo o e he pas 30 yea s, and he e is li le co ela ion in a iabili y wi h he la ge-scale o local wind o cing (Fe nandez e al., 2018). Imp o ed es ima es o he oceanic ad ec ion o hea in he EAC egion would ha e a bene icial impac on wea he o ecas s, modeling o ma ine ecosys ems, and ishe ies managemen (Su he s e al., 2011). T anspo es ima es ac oss PX30 show ime-mean and low- equency a iabili y o he EAC anspo ha a e consis en wi h o e lapping and nea ly colloca ed moo ed obse a ions by Sloyan e al. (2016) (Figu e 7). S udies F on ie s in Ma ine Science | www. on ie sin.o g 8July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k TABLE 1 | Lis o all cu en ly ope a ional XBT ansec s, wi h yea o implemen a ion, mode o ope a ion, and main ocean p ope ies hey obse e. T ansec S a yea Cu en sampling mode Main objec i es AX_WBTS 1995 HD/HR and FR S a e and a iabili y o he Flo ida Cu en AX01 2000 HD/HR No h A lan ic subpola gy e. Va iabili y o MHT in he no he n limb o he he mohaline ci cula ion o he No h A lan ic. AX02 2008 HD/HR Lab ado Sea egion, pa hways and o e lows o wa e s. AX07 1994 HD/HR MHT in he No h A lan ic along ∼30◦N, assessmen o decadal a iabili y in he No h A lan ic Ocean. Va iabili y o he Flo ida Cu en . AX08 2000 HD/HR Main zonal cu en s, coun e cu en s, and unde cu en s in he opical A lan ic Ocean. Gul S eam. A lan ic sub opical gy es. AX10 1996 HD/HR Va iabili y o loca ion and anspo o he Gul S eam, hei link o he NAO, sea le el, and wea he e en s. AX18 2002 HD/HR Me idional mass and hea anspo in he Sou h A lan ic and B azil Cu en . Some imes a somewha no he n ansec ha uns om Rio de Janei o o Cape Town, e e ed o as AX17, is ca ied ou . AX22 1996 HD/HR In e ocean exchanges be ween Sou h A lan ic and Paci ic oceans, and An a c ica, An a c ic Ci cumpola Cu en . AX25 2005 HD/HR In e ocean exchanges be ween Indian Ocean and A lan ic Ocean wa e s, An a c ic Ci cumpola Cu en . AX32 2000 HD/HR Moni o ing o he Gul S eam. AX90 2013 HD/HR Moni o s he su ace- o-bo om empe a u e o all wa e be ween Sco land and Iceland. AX97 2004 HD/HR Moni o s he zonally in eg a ed ba oclinic anspo o he B azil Cu en and i s associa ed mesoscale a iabili y. IX01 1983 FR Indonesian Th ough low moni o ing. IX21 1994 HD/HR Agulhas Cu en . IX28 1992 HD/HR T anspo s ac oss he Sou he n Ocean in conjunc ion wi h AX25 and AX22. MX04 2011 HD/HR Va iabili y o ci cula ion o Ty henian Sea PX02 1983 FR Indonesian Seas and he Indonesian Th ough low moni o ing. PX05 2009 HD/HR Eas Aus alian Cu en , he low la i ude bounda y cu en in he Solomon Sea, and Ku oshio Cu en . PX06/PX09/PX31 1986 HD/HR Pa o he Tasman Box (PX30,PX34,PX06). Sampling he Eas Auckland Cu en and he zonal opical Paci ic cu en sys em (PX06, PX09, and PX31). PX11/IX22 1986 FR Indonesian Th ough low, in egions o e y shallow wa e and high cu en s. PX30 1991 HD/HR EAC bounda y cu en egions. Pa o he Tasman Box (PX30,PX34,PX06). PX34 1991 HD/HR Pa o he Tasman Box (PX30,PX34,PX06). PX37/PX37S 1991 HD/HR Cali o nia Cu en Sys em. PX38 1993 HD/HR Sub opical/subpola Paci ic gy e. PX40 1998 HD/HR Ku oshio Cu en and in e io sub opical gy e ha combine syne gis ic measu emen s o HD XBT da a wi h al ime y and A go obse a ions a e conduci e o unde s anding he along-cu en a iabili y o he EAC, esol ing bo h he majo je s and he EAC eci cula ion, and imp o ing es ima es o he basin-scale anspo s o mass, hea , and eshwa e in he shallow Sou h Paci ic MOC. A pilo p ojec ha will me ge da a om he XBT ne wo k wi h mul idisciplina y da a om A go loa s, sa elli es, glide s, and ocean moo ings is p esen ly unde way o connec ocean dynamics and p oduc i i y in he EAC and o e he con inen al shel . Ku oshio Cu en The Ku oshio Cu en , he WBC o he No h Paci ic gy e, is sampled by XBT ansec PX40 (Honolulu o Yokohama) ha began in 1998. This ansec is o en combined wi h XBT ansec s PX37 (San F ancisco o Honolulu) and PX10 (Honolulu o Guam) o es ima e he comple e ans-basin mean hea and eshwa e anspo s in he No h Paci ic (Ueha a e al., 2008; Douglass e al., 2009, 2010; Auad e al., 2011; Nagano e al., 2012, 2016). An analysis o he o al hea budge o he No h Paci ic Ocean, including hea s o age, ai -sea lux, and hea anspo by he ocean ci cula ion, was ca ied ou using HR XBT da a and an ocean da a assimila ion model (Douglass e al., 2009, 2010). The mean o se be ween he no hwa d hea anspo om XBT da a and ha es ima ed om he model is due o he low model esolu ion nea he WBC and o a me idional o se in he simula ed posi ion o he No h Equa o ial Cu en . Model- based and obse a ional analyses show good ag eemen in hei empo al a iabili y, demons a ing la ge in e annual a iabili y in he ocean hea anspo . The hea anspo and hea s o age componen s la gely balance one ano he , wi h less a iabili y in he ai -sea exchange componen . Nagano e al. (2012, 2016) used he PX37/40 ansec da a o quan i y he a iabili y in he in e io . Thei in eg a ed analysis o XBT, p o iling loa , and sa elli e al ime y da a showed ha F on ie s in Ma ine Science | www. on ie sin.o g 9July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k No h o Cape Ha e as, mo e han 10,000 XBT obse a ions om ansec s AX10, AX08, and AX32 e ealed ha he obse ed sea le el decline along he coas coincided wi h a cooling o he wa e column o e he shel (no shown). Sea le el decline in his a ea was la gely accoun ed o by an inc ease in a mosphe ic p essu e combined wi h a small con ibu ion om cooling o he wa e column o e he con inen al shel (Domingues e al., 2018). Sus ained XBT obse a ions allowed o he iden i ica ion o key changes in hese bounda y cu en s ha con ibu ed o coas al looding e en s a ec ing highly popula ed u ban a eas. DATA MANAGEMENT Da a T ansmission XBT p o iles a e gene ally ansmi ed om ship o sho e using sa elli e communica ions ne wo ks (e.g., I idium, A gos, Inma sa ). When nea eal- ime ansmission is no possible, he p o iles a e sen o ansec ope a o s once he ship a i es in po . Each p o ile unde goes a quali y con ol (QC) p ocess in which a se ies o es s assesses he o e all quali y o he measu emen s. Some da a cen e s apply ini ial au oma ic p ocedu es; p o iles ha ail hese es s mo e o a isual QC (VQC) s age. O he da a cen e s p oceed di ec ly o he VQC s age. In VQC, he p o iles a e isually inspec ed and quali y lags a e applied. The QC es s check o he p esence o spikes, cons an alue p o iles, ex eme dep h, and empe a u e alues, impossible da es and loca ions, e ical g adien s and in e sions, wi e b eaks, sea loo con ac , e c. (Bailey e al., 1994; Thada hil e al., 2001). Once he p o ile QC phase is comple e, all p o iles app o ed du ing his p ocess a e encoded in o FM 63-XI Ex . BATHY ( he adi ional alphanume ic code o epo ing empe a u e p o iles) and/o BUFR (Bina y Uni e sal Fo m o he Rep esen a ion o me eo ological da a) bulle ins and submi ed o he GTS o wo ldwide dis ibu ion in nea eal- ime. The GTS is a co e componen o WMO’s Wo ld Wea he Wa ch P og amme and con ibu es o he apid collec ion and dis ibu ion o sa elli e, in si u, and o he p ocessed da ase s (WMO, 2015b). The collec ion and dis ibu ion o XBT da a is ou inely pe o med h ough GTS cen e s in he Uni ed S a es, Aus alia, Japan, F ance, Canada, and B azil. The cen e s in he i s ou coun ies also dissemina e he p o ile da a and associa ed me ada a in BUFR o ma . BATHY encoded GTS dis ibu ions a e g adually being discon inued wi hin he XBT communi y, in acco dance wi h he WMO manda e o ully mig a e o BUFR. The easons behind his decision a e based on he de elopmen o new and dynamic equi emen s, a highe olume and complexi y o da a and me ada a, a p omo ion o au oma ion, and he limi a ions o he adi ional ixed alphanume ic codes such as BATHY, which es ic he numbe o me ada a ields and do no include QC lags. In BUFR, XBT p o iles a e encoded in o he ope a ional common sequence 315004, which inco po a es all o he common me ada a ields, as well as ull esolu ion da a (WMO, 2015a). Da a acking ac i i ies include he collec ion o XBT BATHY and BUFR epo s a i ing om he GTS. Moni o ing he di e en s ages o he da a managemen p ocess se es o FIGURE 11 | Time-se ies o he a e age empe a u e esiduals (seasonal cycle emo ed) o he uppe 300m o he wa e column in he Flo ida S ai s ( ed, T300), o he mos e ic anomalies de i ed om he empe a u e da a obse ed in he Flo ida S ai s (magen a), and o he Flo ida Cu en (FC) olume anspo ( ed and blue illed cu e) measu ed in he Flo ida S ai s using elephone cable ol age di e ences, and complemen ed using sa elli e al ime y da a (g ay). All ime se ies a e displayed a e applying a 1-yea low pass il e . gene a e epo s, de ec anomalies and da a gaps, and analyze he pe o mance and la ency o he da a collec ion and dis ibu ion sys em. The da a o igina o s e ain he o iginal and delayed- mode QC p o iles and in e media e p oduc s. XBT da a pos ed o he GTS in nea eal- ime a e collec ed by he Ma ine En i onmen al Da a Sec ion (MEDS) o he Oceans Science B anch, Fishe ies and Oceans in Canada, along wi h o he ocean empe a u e p o ile da a, and elayed as a package e e y 3 days o he Na ional Oceanic and A mosphe ic Adminis a ion (NOAA) Na ional Cen e s o En i onmen al In o ma ion (NCEI) in he US. NCEI hos s he long- e m a chi e cen e o GTSPP and p ese es XBT da a in he GTSPP Con inuously Main ained Da abase (CMD). The delayed-mode p o iles ( hose ha a e ei he no pos ed o he GTS o a e ull esolu ion o a e ully quali y con olled eplacemen s o he GTS e sions) a e sen o NOAA/NCEI o inclusion in he GTSPP, om whe e hey a e dissemina ed and en e o he global da a se s such as he NOAA/Wo ld Ocean Da abase, hus becoming pa o he da a low o he JCOMM Ma ine Clima e Da a Sys em. E e y o he yea , he da a cen e s and XBT p o ide s ga he unde he auspices o GTSPP o discuss po en ial imp o emen s o quali y con ol and global dissemina ion o XBT da a. Da a Quali y Decades o e o ha e been made by he XBT communi y o imp o e XBT da a quali y (e.g., Hanawa e al., 1995), and signi ican p og ess has been achie ed in da a quali y imp o emen s since OceanObs’09. Mo e ools and me hods a e now being used o be e unde s and he accu acy o F on ie s in Ma ine Science | www. on ie sin.o g 16 July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k FIGURE 12 | Global OHC ime se ies in he uppe 2000 m a e applying h ee XBT da a imp o emen schemes: Le i us e al. (2009)-L09 (blue); Gou e ski and Reseghe i (2010)-GR10 (g een); and Cheng e al. (2014)-CH14 ( ed) based on he mapping me hod p oposed by he Chinese Ins i u e o A mosphe ic Physics (IAP) in Cheng and Zhu (2016). XBT all a es and empe a u e measu emen s. These ools include adi ional side-by-side XBT and CTD compa isons (e.g., Thada hil e al., 2002; Hamon e al., 2012; Cowley e al., 2013; Cheng e al., 2018), es s in swimming pools and wa e anks (e.g., B ingas and Goni, 2015), Geophysical Fluid Dynamics models (Ab aham e al., 2012, 2014; Go man e al., 2014; Shepa d e al., 2014), compa ison wi h A go and sa elli e al ime y da a (DiNezio and Goni, 2010), and empo al changes o biases (DiNezio and Goni, 2011; Good, 2011; Gou e ski, 2012). A ca e ul analysis o he di e en indi idual p obe ypes is in p og ess (Reseghe i e al., 2018), as di e en p obe ypes ha e di e en cha ac e is ics in p obe design ha may impac da a quali y. An o e iew o he p og ess made in imp o ing da a quali y can be ound in Cheng e al. (2016b). In 2016, he XBT science communi y ecommended he use o an XBT da a imp o emen scheme based on he bias co ec ions (Cheng e al., 2014, 2016b). The new XBT scheme allows o imp o ed XBT obse a ions in he es ima es o global OHC. Using he h ee XBT da a pe o mance me hods (Le i us e al., 2009; Gou e ski and Reseghe i, 2010; Cheng e al., 2014) esul s in nea -iden ical global OHC changes in he uppe 2000 m since 1966 based on he mapping me hod p oposed in Cheng and Zhu (2016) (Figu e 12). In addi ion, empo al, and spa ial a iabili y o loca ions and anspo s o ocean cu en s, es ima es o MHT and MOC, and he de e mina ion o mixed laye dep hs a e obus o any XBT da a imp o emen scheme (Goes e al., 2015b; Houpe e al., 2015). TECHNOLOGICAL IMPROVEMENTS XBT P obes Based on heo e ical and obse a ional expe imen s, imp o emen s ha e been p oposed o he accu acy o bo h he XBT dep h es ima e and he measu ed empe a u e. To imp o e es ima es o p obe dep h, he addi ion o p essu e swi ches has been p oposed. P essu e swi ches a e small esis o s ha a e ac i a ed a ce ain dep hs du ing he p obe descen , ma king hose dep hs in he p o ile wi h spikes. These spikes a e il e ed du ing pos p ocessing, and hei dep hs a e eco ded and used o co ec he de i ed-dep h es ima es o he ull p o ile. In a heo e ical s udy, Goes e al. (2013b) showed ha one p essu e swi ch can limi dep h e o s om 2% o dep h o ∼3.5 m. The implemen a ion o p essu e swi ches may inc ease he cos o XBT p obes, an issue ha will be join ly assessed by he manu ac u e and he scien i ic and ope a ional communi ies. The p obe- o-p obe a iabili y o he linea dep h bias migh also be educed by using a igh e weigh ole ance o he p obes. A p esen , he s a ed weigh ole ance o Deep Blue p obes, he mos widely used p obes, is ±2.5 g (±1 g o he me al head and ±1.5 g o he wi e). Howe e , educing he ole ance o ±1.1 g in a sea ial did no p oduce signi ican imp o emen s (Goes e al., 2017). Addi ional es s a e needed o assess he impo ance o igh e weigh ole ance on p obe linea biases o con i m he esul s o heo e ical assessmen s (G een, 1984; Ab aham e al., 2012). The empe a u e accu acy o XBTs s a ed by he manu ac u e (Lockheed Ma in Sippican, Inc.) is 0.2◦C. Changes in p obe speci ica ions and acquisi ion sys ems can impac his accu acy. Goes e al. (2017) ound ha he mis o calib a ion, pe o med in a s ic ly con olled empe a u e ba h, can imp o e XBT accu acy o 0.03◦C a p ac ically no addi ional cos . XBT Launche Sys ems Many ad ances ha e been made o e he yea s in collec ing and dis ibu ing XBT da a mo e e ec i ely. Ini ially, XBT F on ie s in Ma ine Science | www. on ie sin.o g 17 July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k p obes we e deployed by a ained ope a o using a hand- launching sys em. On many p ojec s, i is necessa y o deploy XBTs on a 24-h-a-day schedule as he ship s eams along i s cou se. To educe he wo kload and pe sonnel, an XBT p obe au olaunche was de eloped ha allowed his wo k o be pe o med by one pe son. New au olaunche s can be p eloaded wi h a numbe o p obes (6–12) ha a e hen deployed a p ede e mined launch imes o posi ions. Au olaunche s ha e been de eloped by se e al ins i u ions, including NOAA, Sc ipps Ins i u ion o Oceanog aphy, CSIRO, and he Uni e si y o Rhode Island. Fo example, a ecen ly de eloped Au oma ed eXpendable Ins umen Sys em (AXIS; F a an oni e al., 2017) in 2012, enabled XBT sampling ac oss an en i e sec ion om he con inen al shel o Be muda wi hou he need o an obse e on boa d. Au olaunche s a e moun ed o he s e n o he ship and cabled o a oom, whe e hey in e ace wi h he da a acquisi ion compu e . The leng h o he cable- un can a y om ship o ship bu is on a e age >75 m. Laying he cable alongside he ship can be di icul and ime-consuming because o he limi ed amoun o deck space a ailable o ins alla ion. To alle ia e he ins alla ion and b eak-down o he cu en se up, a powe independen , wi eless au olaunche using a s anda d wi eless access poin , a ba e y, a sola panel, and o he o - he-shel equipmen and so wa e ools, has been de eloped as a “cable eplacemen ” o he s anda d XBT au olaunche sys em (F a an oni e al., 2017). The imp o ed se up will consis o Wi i echnology coupled wi h a emo e desk op clien ha in heo y can be ope a ed using only a able compu e om wi hin he essel o ope a ed om a land-based s a ion ia he I idium sa elli e ne wo k. Da a Acquisi ion and T ansmission Sys ems The da a acquisi ion eco de is he backbone o collec ing accu a e XBT da a. A new XBT p o o ype da a eco de is cu en ly being es ed o imp o e he numbe o da a d opou s in he ansmissions and o educe he cos o se icing and upg ading exis ing da a acquisi ion sys ems. These measu es may educe ha dwa e cos s by 85%. His o ically, he eal- ime ansmission o XBT da a had been mos ly ca ied ou using he Inma sa -C sa elli e sys em. Wi h he de elopmen o a mo e cos -e ec i e, I idium- based ansmission sys em, he a e age ansmission cos pe XBT p o ile was educed by 95% pe p o ile du ing he las 10 yea s since Ocean Obs‘09. Al hough o iginally de eloped o be used o XBT obse a ions, hese ansmission sys ems ha e also been expanded o ansmi o he ypes o da a, such as he mosalinog aph (TSG), pCO2, and ma ine wea he obse a ions. THE FUTURE OF THE GLOBAL XBT NETWORK Twen y yea s a e OceanObs’99, he Global XBT Ne wo k con inues o inc ease in alue, no only h ough he g owing leng h o he decadal ime-se ies along indi idual ansec s, bu also due o in eg a i e ela ionships wi h o he elemen s o he ocean obse ing sys em. Uniquely, he Global XBT Ne wo k p o ides spa ial and empo al sampling ha canno as ye be ep oduced by o he exis ing pla o ms. One o he key s eng hs o he ne wo k is ha XBTs ha e low ope a ional cos s and can be eadily deployed on a epea basis wi h a ying spa ial esolu ion. I is expec ed ha he Global XBT Ne wo k will emain ac i e and be enhanced o e he nex 10 yea s. We conclude wi h a lis o key aspec s ha he scien i ic communi y has de e mined o be impo an o u u e s udies in ol ing XBT obse a ions. •Sampling s a egies. One unique quali y o XBT obse a ions is hei abili y o sample along ixed ans-basin ansec s and ac oss bounda y cu en s in a sus ained ashion, which p esen ly canno be ep oduced by any o he pla o m. O he componen s o he ocean obse ing sys em (e.g., p o iling loa s, glide s, moo ings, e c.) p o ide complemen a y p o iles o ocean empe a u e and o he p ope ies in hese egions; howe e , none can eplica e he apidly-occupied ansec s in nea ly epea ed loca ions ha ha e been ob ained by XBTs o decades. •Main enance o long clima e eco d. Se e al o he ime se ies ini ia ed and s ill main ained by XBTs ha e been in place o 30 yea s o longe . Du ing he nex decade, XBTs a e likely o emain an in eg al pa o he coo dina ed obse ing e o ha con inues collec ing key oceanic empe a u e measu emen s o moni o ing bounda y cu en s (sec ion Ocean cu en s, gy es, and ocean a iabili y), MHT es ima es ac oss ocean basins (sec ion Me idional Hea T anspo ), and global OHC assessmen s (sec ion Global and egional ocean hea con en ). •Imp o emen o da a quali y. As wi h o he obse ing pla o ms, expe imen s and s udies will con inue o be ca ied ou o imp o e he quali y o XBT obse a ions. This will be add essed by con inuing o educe e o s in each subg oup o XBT da a (i.e., da a o he same p obe ype, da a om he same yea , e c.) (sec ion Technological Imp o emen s) and by imp o ing p obe design o inc ease he p ecision o each indi idual measu emen (sec ion The u u e o he Global XBT Ne wo k). The con inuous imp o emen o XBT da a quali y jus i ies he me ging o XBT da a wi h da a om o he pla o ms (i.e., A go, CTD), allowing o be e moni o ing and analysis o clima e change and a iabili y (i.e., sec ion Global and egional ocean hea con en ). •Me idional hea anspo . The Global XBT Ne wo k con inues o p o ide key assessmen s o oceanic empe a u e p o iles a di e en la i udes, pa icula ly in he No h Paci ic and Sou h A lan ic oceans o moni o he cu en s a e o he MOC and associa ed MHT. These da a will con ibu e o s udies ha link ans-basin hea anspo s wi h a mosphe ic ci cula ion ha may in luence egional and global clima e and ex eme wea he , aiding in he de elopmen o o ecas s and ou looks o high-impac ex eme wea he e en s. •Simul aneous me eo ological and oceanog aphic obse a ions. Me eo ological senso s can be easily in eg a ed in o exis ing XBT ansec s o p o ide key me eo ological da a collec ed simul aneously wi h uppe ocean he mal obse a ions o calcula e su ace hea and mois u e luxes, which a e c i ical o wea he and clima e esea ch. O he F on ie s in Ma ine Science | www. on ie sin.o g 18 July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k ins umen a ion ha can be ins alled on ships o oppo uni y include pCO2 sys ems, con inuous plank on eco de s, acous ic cu en Dopple p o ile s, e c. •Sea le el change. S udies o sea le el change a ibu ions, such as ha being pe o med o he US Eas Coas (Domingues e al., 2018), se e as examples o simila s udies ha may be conduc ed ou side coas al a eas whe e XBT obse a ions con inue o p o ide long- ime se ies o a iabili y o ocean cu en s, such as he B azil Cu en . Ocean obse a ions, including hose om XBTs in hese coas al a eas, a e c i ical o he con inuous unde s anding and moni o ing o key d i e s o dis up i e, and o en imes des uc i e, looding e en s due o ele a ed sea le els. •Submesoscale Ocean Dynamics. O cu en in e es in oceanog aphy is he moni o ing o submesoscale ea u es and p ocesses (<10 km) ac oss s ong bounda y cu en s, mesoscale eddies, and meande s. The XBT ne wo k can con ibu e o his e o in coo dina ion wi h semi-Lag angian obse ing pla o ms, such as unde wa e glide s and d i e s. Fo example, he challenges ha glide s may encoun e while measu ing ac oss s ong cu en s could be a oided by inc easing he spa ial sampling along selec ed po ions o XBT ansec s. •In e nal ides. As we mo e o ine - esolu ion al ime ic obse a ions, wi h along ack Syn he ic Ape u e Rada (SAR) missions (e.g., Sen inel-3) and he u u e Su ace Wa e and Ocean Topog aphy (SWOT) 2D missions, hese long ime se ies o XBT obse a ions a e being eassessed. High- equency in e nal ide a iabili y was his o ically il e ed ou o XBT da a o concen a e on he la ge -scale eddies and ci cula ion. Now ha al ime y is capable o obse ing he sea le el a ia ions o hese signals, he e a e oppo uni ies o da a mining o he olde XBT da a o help alida e he al ime ic in e nal ide obse a ions, as well as ocean models including in e nal ides. Fu u e XBT o glide obse a ions along SAR-al ime y o SWOT acks will p o ide in aluable e ical s uc u e o help in e p e hese dynamical p ocesses. •High no he n la i ude obse a ions. The exis ing XBT ansec s AX01 (G eenland o Denma k) and AX90 (Iceland o Fa oe Island o She land Islands) in he subpola No h A lan ic ha e p o ided aluable in o ma ion on me idional olume and hea anspo s (e.g., Rossby e al., 2018). The e is u u e po en ial o signi ican ly enhance he p esen -day obse ing sys em in he high la i udes by es ablishing a new XBT ansec be ween con inen al No way and S alba d. Possible ins umen a ion o a No wegian supply essel wi h a shipboa d ADCP and an XBT launche would p o ide accu a e measu emen s o ocean cu en s and empe a u e luxes ac oss his mos impo an A c ic ga eway a high spa ial and empo al esolu ion. This will esul in imp o ed moni o ing o oceanic luxes in o he A c ic Ocean, a egion expe iencing d ama ic clima e change. •Obse ing sys em expe imen s and obse ing sys em simula ion expe imen s. Bo h OSEs and Obse ing Sys em Simula ion Expe imen s (OSSEs) a e needed o ca y ou quan i a i e e alua ions o he impac o ocean obse a ions, including XBTs. OSEs se e o assess he impac o ac ual obse a ions on ocean o ecas s o eanalyses, while OSSEs p o ide a igo ous app oach o e alua e he po en ial impac o new obse ing sys ems o o imp o e he sampling o cu en obse a ions. Wi h he implemen a ion o new obse ing pla o ms i is necessa y o quan i a i ely assess he complemen a y alue o a sui e o empe a u e p o iles a di e en spa ial and empo al scales o a ange o s udies. •A pla o m o deploy o he obse ing ins umen s. Vessels in ol ed in he wo k o he SOT, and pa icula ly in he XBT ne wo k, o en also suppo o he ne wo ks, e.g., h ough he deploymen o au onomous ins umen s (d i e s, loa s) o ins alla ion o unde way sys ems (e.g., TSGs). Coo dina ion and moni o ing o ship con ibu ions ac oss all obse ing ne wo ks is o g owing impo ance, no only o a be e exploi a ion o syne gies (e.g., main enance and logis ics), bu also o no o e bu dening ships wi h oo many asks o a a ie y o pu poses. JCOMMOPS5has de eloped online ools ha will allow o a cen alized and ha monized egis a ion o c uises, ins umen s, and deploymen plans, all e e ing o a commonly used ship e e ence lis wi h unique iden i ie s. •Hu icane applica ions. A po en ial applica ion o XBTs is o imp o e seasonal hu icane ou looks. In he Paci ic Ocean, PX09 (Honolulu o Su a)/PX31 (Los Angeles o Su a) and PX40 da a a e used o de i e OHC es ima es o imp o e opical cyclone in ensi y o ecas s (Shay and B ews e , 2010; McCaskill e al., 2016). The AX08 ansec c osses he de elopmen egion o A lan ic hu icanes, a egion whe e coupled models gene ally p esen a cold bias and whe e cyclone de elopmen is a ec ed by eddy, in e annual, and decadal uppe OHC a iabili y ia u bulen hea luxes. The use o AX08 da a o assess and imp o e ocean models has he po en ial o also imp o e seasonal ou looks and/o in ensi ica ion o ecas s o A lan ic hu icanes (Domingues e al., 2019). •Redundancy o obse a ions. Finally, i is impo an o ecognize ha some edundancy in he obse ing sys em is needed, especially o assis au oma ic quali y con ol p ocedu es. Fo ins ance, ha ing XBT da a in he icini y o p o iling loa s can help de ec e o s in one o he o he ins umen . AUTHOR CONTRIBUTIONS GG led he w i ing and o ganiza ion o he manusc ip , as well as he esea ch whose esul s a e pos ed in se e al sec ions. JS led he esea ch pos ed in se e al sec ions and con ibu ed wi h w i ing and commen s. FB, LC, MC, SD, RD, MG, HL, RM, UR, TR, RT, JT, NZ, MB, TB, RC, CD, KH, MK, MM, FR, CS, UB, and DV con ibu ed o he w i ing o sec ions, pos ed commen s, and p o ided igu es. 5www.jcommops.o g F on ie s in Ma ine Science | www. on ie sin.o g 19 July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k FUNDING GG, FB, SD, UR, MB, RD, and DV we e suppo ed by a g an om he NOAA/Ocean Obse ing and Moni o ing Di ision (OOMD) and by NOAA’s A lan ic Oceanog aphic and Me eo ological Labo a o y (AOML). The pa icipa ion o JS and NZ in his s udy was suppo ed by NOAA’s Global Ocean Moni o ing and Obse ing P og am h ough Awa d NA15OAR4320071 and NSF Awa d 1542902. CD was unded by he Aus alian Resea ch Council (FT130101532 and DP160103130); he Scien i ic Commi ee on Oceanic Resea ch (SCOR) Wo king G oup 148, unded by na ional SCOR commi ees and a g an o SCOR om he U.S. Na ional Science Founda ion (G an OCE-1546580); and he In e go e nmen al Oceanog aphic Commission o UNESCO/In e na ional Oceanog aphic Da a and In o ma ion Exchange (IOC/IODE) IQuOD S ee ing G oup. LC was suppo ed by 2016YFC1401800. ACKNOWLEDGMENTS The au ho s acknowledge he many agencies and ins i u ions ha suppo he implemen a ion and main enance o he Global XBT Ne wo k, and da a managemen and esea ch ac i i ies, including bu no limi ed o: NOAA (Uni ed S a es), Na ional Science Founda ion (Uni ed S a es), Commonweal h Scien i ic and Indus ial Resea ch O ganiza ion (CSIRO), Bu eau o Me eo ology (BOM), In eg a ed Ma ine Obse ing Sys em (IMOS), a na ional collabo a i e esea ch in as uc u e suppo ed by he Aus alian go e nmen (Aus alia), Na ional Agency o New Technologies, Ene gy and Sus ainable Economic De elopmen (ENEA) (I aly), Tohoku Uni e si y (Japan), Uni e si y o Miami (Uni ed S a es), Uni e si y o Tasmania (Aus alia), Sc ipps Ins i u ion o Oceanog aphy (Uni ed S a es), Na ional Ins i u e o Wa e and A mosphe ic Resea ch (New Zealand), Labo a oi e d’E udes en Géophysique e Océanog aphie Spa iales (LEGOS) (F ance), Ins i u e o A mosphe ic Physics and Chinese Academy o Sciences (China), Uni e si y o Rhode Island (Uni ed S a es), BIO (Be muda), Uni e si y o Cape Town (Sou h A ica), Fede al Uni e si y o Rio G ande do Sul (B azil), Fede al Uni e si y o Rio de Janei o (B azil), Uni e si y o Pa is (F ance), Na ional Ins i u e o Oceanog aphy (India), MEDS (Canada), Se icio de Hid og a ía Na al (A gen ina), Se icio de Hid og a ía Na al (B azil), S a e Uni e si y o New Yo k a S ony B ook (Uni ed S a es), Indian Na ional Cen e o Ocean In o ma ion Se ices (India), and he Woods Hole Oceanog aphic Ins i u ion (Uni ed S a es). The au ho s acknowledge he olun ee con ibu ion o con aine shipping lines and hei c ews o con inuously p o iding suppo o ope a ions and logis ics o deploy XBT p obes. REFERENCES Ab aham, J. P., Ba inge , M., Bindo , N. L., Boye , T., Cheng, L. J., Chu ch, J. A., e al. (2013). A e iew o global ocean empe a u e obse a ions: implica ions o ocean hea con en es ima es and clima e change. Re . Geophys. 51, 450–483. doi: 10.1002/ og.20022 Ab aham, J. P., Go man, J., Reseghe i, F., Spa ow, E., S a k, J., and Shepa d,. T. (2014). Modeling and nume ical simula ion o he o ces ac ing on a sphe e du ing ea ly-wa e en y. Ocean Eng. 76, 1–9. doi: 10.1016/j.oceaneng.2013.11.015 Ab aham, J. P., Go man, J. M., Reseghe i, F., Spa ow, E. M., and Minkowycz, W. J. (2012). D ag coe icien s o o a ing expendable ba hy he mog aphs and he impac o launch pa ame e s on dep h p edic ions. Num. Hea T ans e P . 62, 25–43. doi: 10.1080/10407782.2012.672898 Auad, G., Roemmich, D., and Gilson, J. (2011). The cali o nia cu en sys em in ela ion o he no heas paci ic ocean ci cula ion. P og. Oceanog . 91, 576–592. doi: 10.1016/j.pocean.2011.09.004 Bailey, R., G onell, A. M., Phillips, H., Tanne , E., and Meye s, G. (1994). Quali y Con ol Cookbook o XBT Da a. CSIRO Ma ine Labo a o ies Repo s 221. Boye , T., Domingues, C. M., Good, S. A., Johnson, G. C., Lyman, J. M., Ishii, M., e al. (2016). Sensi i i y o global ocean hea con en es ima es o mapping me hods, XBT bias co ec ions, and baseline clima ologies. J. Clim. 29, 4817–4842. doi: 10.1175/JCLI-D-15-0801.1 Boye , T. P., An ono , J. I., Ba ano a, O. K., Coleman, C., Ga cia, H. E., G odsky (2013). Wo ld Ocean Da abase 2013. NOAA A las NESDIS 72. B annigan, L., Lenn, Y.-D., Rippe h, T. P., McDonagh, E., Che eskin, T. K., and Sp in all, J. (2013). Shea a he base o he oceanic mixed laye gene a ed by wind shea alignmen . J. Phys. Oceanog . 43, 1798–1810. doi: 10.1175/JPO-D-12-0104.1 B ingas, F., and Goni, G. (2015). Ea ly dynamics o deep blue XBT p obes. J. A mos. Oceanic Tech. 32, 2253–2263. doi: 10.1175/JTECH-D-15-0048.1 Cai, W., Shi, G., Cowan, T., Bi, D., and Ribbe, J. (2005). The esponse o he Sou he n Annula Mode, he Eas Aus alian Cu en , and he sou he n mid- la i ude ocean ci cula ion o global wa ming. Geophys. Res. Le . 32:L23706. doi: 10.1029/2005GL024701 Chang, E. K. M., Lee, S., and Swanson, K. L. (2002). S o m ack dynamics. J. Clim. 15, 2163–2183. doi: 10.1175/1520-0442(2002)015<02163:STD>2.0.CO;2 Cheng, L., Ab aham, J., Goni, G., Boye , T., Wij els, S., Cowley, R., e al. (2016b). XBT science: assessmen o ins umen al biases and e o s. Bull. Ame . Me . Soc. 97, 924–933. doi: 10.1175/BAMS-D-15-00031.1 Cheng, L., Luo, H., Boye , T., Cowley, R., Ab aham, J., Gou e ski, V., e al. (2018). How well can we co ec sys ema ic e o s in his o ical XBT da a? J. A mos. Oceanic Tech. 35, 1103–1125. doi: 10.1175/JTECH-D-17-0122.1 Cheng, L., T enbe h, K. E., Fasullo, J., Boye , T., Ab aham, J., and Zhu, J. (2017). Imp o ed es ima es o ocean hea con en om 1960-2015. Sci. Ad . 3:e1601545. doi: 10.1126/sciad .1601545 Cheng, L., T enbe h, K. E., Palme , M. D., Zhu, J., and Ab aham, J. P. (2016a). Obse ed and simula ed ull-dep h ocean hea -con en changes o 1970–2005. Ocean Sci. 12, 925–935. doi: 10.5194/os-12-925-2016 Cheng, L., and Zhu, J. (2016). Bene i s o CMIP5 mul imodel ensemble in econs uc ing his o ical ocean subsu ace empe a u e a ia ions. J. Clim. 29, 5393–5416. doi: 10.1175/JCLI-D-15-0730.1 Cheng, L., Zhu, J., Cowley, R., Boye , T., and Wij els, S. (2014). Time, p obe ype and empe a u e a iable bias co ec ions o his o ical expendable ba hy he mog aph obse a ions. J. A mos. Oceanic Tech. 31, 1793–1825. doi: 10.1175/JTECH-D-13-00197.1 Ciu a di, T., Napoli ano, E., Iacono, R., Reseghe i, F., Rai e i, G., and Bo done, A. (2016). Analysis o su ace ci cula ion s uc u es along a equen ly epea ed XBT ansec c ossing he Ligu ian and Ty henian Seas. Ocean Dyn. 66, 767–783. doi: 10.1007/s10236-016-0954-y Cowley, R., Wij els, S., Cheng, L., Boye , T., and Kizu, S. (2013). Biases in expendable ba hy he mog aph da a: a new iew based on his o ical side-by-side compa isons. J. A mos. Oceanic Tech. 30, 1195–1225. doi: 10.1175/JTECH-D-12-00127.1 DiNezio, P. N., and Goni, G. (2010). Iden i ying and es ima ing biases be ween XBT and A go obse a ions using sa elli e al ime y. J. A mos. Oceanic Tech. 27, 226–240. doi: 10.1175/2009JTECHO711.1 DiNezio, P. N., and Goni, G. (2011). Di ec e idence o a changing all- a e bias in XBT manu ac u ed du ing 1986-2008. J. A mos. Oceanic Tech. 28, 1569–1578. doi: 10.1175/JTECH-D-11-00017.1 F on ie s in Ma ine Science | www. on ie sin.o g 20 July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k Domingues, C. M., Chu ch, J. A., Whi e, N. J., Gleckle , P. J., Wij els, S. E., Ba ke , P. M., e al. (2008). Imp o ed es ima es o uppe -ocean wa ming and mul i-decadal sea-le el ise. Na u e 453, 1090–1093. doi: 10.1038/na u e 07080 Domingues, R., Ba inge , M., and Goni, G. (2016). Remo e sou ces o yea - o-yea changes in he seasonali y o he Flo ida cu en anspo . J.Geophys. Res. 121, 7547–7559. doi: 10.1002/2016JC012070 Domingues, R., Goni, G., Ba inge , M., and Volko , D. (2018). Wha caused he accele a ed sea le el changes along he U.S. Eas Coas du ing 2010-2015? Geophys. Res. Le . 45, 13,367–13,376. doi: 10.1029/2018GL081183 Domingues, R., Goni, G., Swa , S., and Dong, S. (2014). Wind o ced a iabili y o he an a c ic ci cumpola cu en sou h o A ica be ween 1993 and 2010. J. Geophys. Res. 119, 1123–1145. doi: 10.1002/2013JC008908 Domingues, R., Kuwano-Yoshida, A., Cha don-Maldonado, P., Todd, R. E., Halliwell, G. R., Kim, H. S., e al. (2019). Ocean obse ing sys ems in suppo o s udies and o ecas s o opical and ex a opical cyclones. F on . Ma ine Sci. 6:446. doi: 10.3389/ ma s.2019.00446 Dong, S., Ba inge , M., Goni, G., and Ga zoli, S. (2011). Impo ance o he assimila ion o A go loa measu emen s on he me idional o e u ning ci cula ion in he Sou h A lan ic. Geophys. Res. Le . 38:L18603. doi: 10.1029/2011GL048982 Dong, S., Ba inge , M. O., Goni, G. J., Meinen, C. S., and Ga zoli, S. L. (2014). Seasonal a ia ions in he Sou h A lan ic me idional o e u ning ci cula ion om obse a ions and nume ical models. Geophys. Res. Le . 41, 4611–4618. doi: 10.1002/2014GL060428 Dong, S., Ga zoli, S. L., Ba inge , M. O., Meinen, C. S., and Goni, G. J. (2009). The a lan ic me idional o e u ning ci cula ion and i s no hwa d hea anspo in he Sou h A lan ic. Geophys. Res. Le . 36:L20606. doi: 10.1029/2009GL039356 Dong, S., Gille, S. T., and Sp in all, J. (2007a). An assessmen o he Sou he n ocean mixed laye hea budge . J. Clim. 20, 4425–4442. doi: 10.1175/JCLI4259.1 Dong, S., Gille, S. T., Sp in all, J., and Fe ze , E. J. (2010). Assessing he po en ial o he A mosphe ic In a ed Sounde (AIRS) su ace empe a u e and speci ic humidi y in u bulen hea lux es ima es in he sou he n ocean. J. Geophys. Res. 115:C05013. doi: 10.1029/2009JC005542 Dong, S., Gille, S. T., Sp in all, J., and Gen emann, C. (2006a). Valida ion o he ad anced mic owa e scanning adiome e o he ea h obse ing Sys em (AMSR-E) sea su ace empe a u e in he Sou he n Ocean. J. Geophys. Res. 111:C04002. doi: 10.1029/2005JC002934 Dong, S., Goni, G., and B ingas, F. (2015). Tempo al a iabili y o he Sou h A lan ic me idional o e u ning ci cula ion be ween 20◦S and 35◦S. Geophys. Res. Le . 42, 7655–7662. doi: 10.1002/2015GL065603 Dong, S., Hau ala, S. L., and Kelly, K. A. (2007b) In e annual a ia ions in uppe -ocean hea con en and hea anspo con e gence in he wes e n No h A lan ic. J. Phys. Oceanog . 37, 2682–2697. doi: 10.1175/2007JPO 3645.1 Dong, S., Sp in all, J., and Gille, S. T. (2006b). Loca ion o he An a c ic pola on om AMSR-E sa elli e sea su ace empe a u e measu emen s. J. Phys. Oceanog . 36, 2075–2089. doi: 10.1175/JPO2973.1 Douglass, E., Roemmich, D., and S amme , D. (2009). Da a sensi i i y o he ECCO s a e es ima e in a egional se ing. J. A mos. Oceanic Tech. 26, 2420–2443. doi: 10.1175/2009j echo641.1 Douglass, E., Roemmich, D., and S amme , D. (2010). In e annual a iabili y in No h Paci ic hea and eshwa e budge s. Deep Sea Res. 57, 1127–1140. doi: 10.1016/j.ds 2.2010.01.001 Du ack, P. J., Glecke , P. J., Lande e , F. W., and Taylo , K. E. (2014). Quan i ying unde es ima es o long- e m uppe -ocean wa ming. Na . Clim. Change 4, 999–1005. doi: 10.1038/nclima e2389 Eigenhee , A., and Quad asel, D. (2000). Seasonal a iabili y o he Bay o Bengal ci cula ion in e ed om TOPEX/Poseidon al ime y. J. Geophys. Res. 105, 3243–3252. doi: 10.1029/1999JC900291 England, M. H., McG ego , S., Spence, P., Meehl, G. A., Timme mann, A., Cai, W., e al. (2014). Recen in ensi ica ion o wind-d i en ci cula ion in he Paci ic and he ongoing wa ming hia us. Na . Clim. Change 4, 222–227. doi: 10.1038/nclima e2106 Eze , T. (2013). Sea le el ise, spa ially une en and empo ally uns eady: why he US Eas Coas , he global ide gauge eco d, and he global al ime e da a show di e en ends. Geophys. Res. Le . 40, 5439–5444. doi: 10.1002/2013GL0 57952 Fe nandez, D., Bowen, M., and Su on, P. (2018). Va iabili y, cohe ence and o cing mechanisms in he New Zealand ocean bounda y cu en s. P og. Oceanog . 165, 168–188. doi: 10.1016/j.pocean.2018.06.002 Fi ing, Y. L., Che eskin, T. K., and Mazlo , M. R. (2011). Ve ical s uc u e and anspo o he an a c ic ci cumpola cu en in d ake passage om di ec eloci y obse a ions. J. Geophys. Res. 116:C08015. doi: 10.1029/2011JC006999 Fonseca, C., Goni, G. J., Johns, W. E., and Campos, E. J. D. (2004). In es iga ion o he No h B azil cu en e o lec ion and no h equa o ial coun e cu en a iabili y. Geophys. Res. Le . 31:L21304. doi: 10.1029/2004GL020054 F ajka-Williams, E., Dong, S., Meinen, C., Pe ez, R., Goni, G., Volko , D., e al. (2019). A lan ic Me idional o e u ning ci cula ion: obse ed anspo s and mechanisms. F on . Ma . Sci. 6:260. doi: 10.3389/ ma s.2019.00260 F a an oni, D. M., O’B ien, J. K., Flagg, C., and Rossby, T. (2017). AXIS– An au onomous expendable ins umen sys em. J. A mos. Oceanic Tech. 34, 2673–2682. doi: 10.1175/JTECH-D-17-0054.1 Gleckle , P. J., Du ack, P. J., S ou e , R. J., Johnson, G. C., and Fo es , C. E. (2016). Indus ial-e a global ocean hea up ake doubles in ecen decades. Na . Clim. Change 6, 394–398. doi: 10.1038/nclima e2915 Goes, M., Babcock, E., B ingas, F., O ne , P., and Goni, G. (2017). The impac o imp o ed he mis o calib a ion on he expendable ba hy he mog aph p o ile da a. J. A mos. Oceanic Tech. 34, 1947–1961. doi: 10.1175/JTECH-D-17-0024.1 Goes, M., Ba inge , M., and Goni, G. (2015b). The impac o his o ical biases on he XBT-de i ed me idional o e u ning ci cula ion es ima es a 34◦S. Geophys. Res. Le . 42, 1848–1855. doi: 10.1002/2014GL061802 Goes, M., Ch is ophe sen, J., Dong, S., Goni, G., and Ba inge , M. O. (2018). An upda ed es ima e o salini y o he a lan ic ocean sec o using empe a u e–salini y ela ionships. J. A mos. Oceanic Tech. 35, 1771–1784. doi: 10.1175/JTECH-D-18-0029.1 Goes, M., Ci ano, M., Ma a, M. M., and Majumde , S. (2019). Long- e m moni o ing o he B azil Cu en anspo a 22S om XBT and al ime y da a: Seasonal, in e annual and ex eme a iabili y. J. Geophys. Res. 124, 3645–3663. doi: 10.1029/2018JC014809 Goes, M., Goni, G., and Dong, S. (2015a). An op imal XBT-based moni o ing sys em o he Sou h A lan ic me idional o e u ning ci cula ion a 34◦S. J. Geophys. Res. 120, 161–181. doi: 10.1002/2014JC010202 Goes, M., Goni, G., Ho mann, V., and Pe ez, R. C. (2013a). Va iabili y o he A lan ic o -equa o ial eas wa d cu en s du ing 1993–2010 using a syn he ic me hod. J. Geophys. Res. 118, 3026–3045. doi: 10.1002/jg c.20186 Goes, M., Goni, G., and Kelle , K. (2013b). Reducing biases in XBT measu emen s by including disc e e in o ma ion om p essu e swi ches. J. A mos. Oceanic Tech. 30, 810–824. doi: 10.1175/JTECH-D-12-00126.1 Goni, G., and Ba inge , M. (2002). Su ace cu en s in he opical A lan ic ac oss high densi y XBT line AX08. Geophys. Res. Le . 29, 71.1–71.4. doi: 10.1029/2002GL015873 Goni, G., Roemmich, D., Molina i, R., Meye s, G., Sun, C., Boye , T., e al. (2010). “The ship o oppo uni y p og am,” in P oceedings o OceanObs’09: Sus ained Ocean Obse a ions and In o ma ion o Socie y, eds J. Hall, D. E. Ha ison, and D. S amme (Auckland: ESA Publica ions). Goni, G. J., and Waine , I. (2001). In es iga ion o he B azil Cu en on a iabili y om al ime e da a. J. Geophys. Res. 106, 31,117–31,128. doi: 10.1029/2000JC000396 Good, S. A. (2011). Dep h biases in XBT da a diagnosed using ba hyme y da a. J. A mos. Oceanic Tech. 28, 287–300. doi: 10.1175/2010JTECHO773.1 Go man, J. M., Ab aham, J. P., Schwalbach, D. B., Shepa d, T. S., S a k, J. R., and Reseghe i, F. (2014). Expe imen al e i ica ion o d ag o ces on sphe ical objec s en e ing wa e . J. Ma . Ocean Biol. Oceanog . 3:2. doi: 10.4172/2324-8661.1000126 Gould, J., Roemmich, D., Wij els, S., F eeland, H., Ignaszewsky, M., Jianping, X., e al. (2004). A go p o iling loa s b ing new e a o in si u ocean obse a ions. Eos T ans. Ame . Geophys. 179, 190–191. doi: 10.1029/2004EO190002 Gou e ski, V. (2012). Using GEBCO digi al ba hyme y o in e dep h biases in he XBT da a. Deep Sea Res. 62, 40–52. doi: 10.1016/j.ds .2011.12.012 Gou e ski, V., and Reseghe i, F. (2010). On dep h and empe a u e biases in ba hy he mog aph da a: De elopmen o a new co ec ion scheme based on analysis o a global ocean da abase. Deep Sea Res. 57, 812–833. doi: 10.1016/j.ds .2010.03.011 G een, A. W. (1984). Bulk dynamics o he expendable ba hy he mog aph (XBT). Deep-Sea Res. 31, 415–426. doi: 10.1016/0198-0149(84)90093-1 F on ie s in Ma ine Science | www. on ie sin.o g 21 July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k Hamon, M., Re e din, G., and Le T aon, P. Y. (2012). Empi ical co ec ion o XBT da a. J. A mos. Oceanic Tech. 29, 960–973. doi: 10.1175/JTECH-D-11-00129.1 Hanawa, K., Rual, P., Bailey, R., Sy, A., and Szabados, M. (1995). A new dep h ime equa ion o Sippican o TSK T-7, T-6 and T-4 expendable ba hy he mog aphs (XBT). Deep Sea Res. 42, 1423–1451. doi: 10.1016/0967-0637(95)97154-Z Hill, K. L., Rin oul, S. R., Coleman, R., and Ridgway, K. R. (2008). Wind o ced low equency a iabili y o he Eas Aus alian cu en . Geophys. Res. Le . 35:L08602. doi: 10.1029/2007GL032912 Hill, K. L., Rin oul, S. R., Ridgway, K. R., and Oke, P. R. (2011). Decadal changes in he Sou h Paci ic wes e n bounda y cu en sys em e ealed in obse a ions and ocean s a e es ima es. J. Geophys. Res. 116:C01009. doi: 10.1029/2009JC005926 Holb ook, N. J., and Maha aj, A. M. (2008). Sou hwes Paci ic sub opical mode wa e : a clima ology. P og. Oceanog . 77, 298–315. doi: 10.1016/j.pocean.2007.01.015 Hoskins, B. J., and Hodges, K. I. (2002). New pe spec i es on he No he n Hemisphe e win e s o m acks. J. A mos. Sci. 59, 1041–1061. doi: 10.1175/1520-0469(2002)059<1041:NPOTNH>2.0.CO;2 Houpe , L., Tes o , P., Du ieu de Mad on, X., Somo , S., D’O enzio, F., Es ou nel, C., e al. (2015). Seasonal cycle o he mixed laye , he seasonal he mocline and he uppe -ocean hea s o age a e in he Medi e anean Sea de i ed om obse a ions. P og. Oceanog . 132, 333–352. doi: 10.1016/j.pocean.2014.11.004 Ishii, M., Fukuda, Y., Hi aha a, S., Yasui, S., Suzuki, T., and Sa o, K. (2017). Accu acy o global uppe ocean hea con en es ima ion expec ed om p esen obse a ional da a se s. SOLA 13, 163–167. doi: 10.2151/sola.2017-030 Jayne, S. R., Roemmich, D., Zilbe man, N., Rise , S. C., Johnson, K. S., Johnson, G. C., e al. (2017). The a go p og am: p esen and u u e. Oceanog aphy 30, 18–28. doi: 10.5670/oceanog.2017.213 Jiang, C., Gille, S. T., Sp in all, J., and Sweeney, C. (2014). D ake Passage oceanic pCO2: e alua ing CMIP5 coupled ca bon–clima e models using in-si u obse a ions. J. Clim. 27, 76–100, doi: 10.1175/JCLI-D-12-00571.1 Joyce, T. M., Kwon, Y.-O., and Yu, L. (2009). On he ela ionship be ween synop ic win e ime a mosphe ic a iabili y and pa h shi s in he Gul S eam and Ku oshio ex ension. J. Clim. 22, 3177–3192. doi: 10.1175/2008JCLI2690.1 Kelly, K. A., Small, R. J., Samelson, R. M., Qiu, B., Joyce, T. M., Kwon, Y.-O., e al. (2010). Wes e n bounda y cu en s and on al ai –sea in e ac ion: Gul S eam and Ku oshio Ex ension. J. Clim. 23, 5644–5667. doi: 10.1175/2010JCLI3346.1 Kushni , Y., Robinson, W. A., Blade, I., Hall, N. M. J., Peng, S., and Su on, R. (2002). A mosphe ic GCM esponse o ex a opical SST anomalies: syn hesis and e alua ion. J. Clim. 15, 2233–2256. doi: 10.1175/1520- 0442(2002)015<2233:AGRTES>2.0.CO;2 Kwon, Y.-O., Alexande , M. A., Bond, N. A., F ankignoul, C., Nakamu a, H., Qiu, B., e al. (2010). Role o he Gul S eam and Ku oshio–Oyashio sys ems in la ge-scale a mosphe e–ocean in e ac ion: a e iew. J. Clim. 23, 3249–3281. doi: 10.1175/2010JCLI3343.1 La i , M., A pe, K., and Roeckne , E. (2000). Oceanic con ol o decadal No h a lan ic sea le el p essu e a iabili y in win e . Geophys. Res. Le . 27, 727–730. doi: 10.1029/1999GL002370 Lee, S.-K., Pa k, W., Ba inge , M. O., Go don, A. L., Hube , B., and Liu, Y. (2015). Paci ic o igin o he ab up inc ease in Indian Ocean hea con en du ing he wa ming hia us. Na . Geosci. 8, 445–449. doi: 10.1038/ngeo2438 Legle , D. M., F eeland, H. J., Lumpkin, R., Ball, G., McPhaden, M. J., No h, S., e al. (2015). The cu en s a us o he eal- ime in si u global ocean obse ing sys em o ope a ional oceanog aphy. J. Ope . Oceanog . 8, s189– s200. doi: 10.1080/1755876X.2015.1049883 Lenn, Y.-D., and Che eskin, T. K. (2009). Obse a ions o Ekman cu en s in he Sou he n Ocean. J. Phys. Oceanog . 39, 768–779. doi: 10.1175/2008JPO3943.1 Lenn, Y.-D., Che eskin, T. K., and Sp in all, J. (2008). Imp o ing es ima es o he an a c ic ci cumpola cu en s eamlines in d ake passage. J. Phys. Oceanog . 38, 1000–1010, doi: 10.1175/2007JPO3834.1 Lenn, Y.-D., Che eskin, T. K., Sp in all, J., and Fi ing, E. (2007). Mean je s, mesoscale a iabili y and eddy momen um luxes in he su ace laye o he an a c ic ci cumpola cu en in d ake passage. J. Ma . Res. 65, 27–58. doi: 10.1357/002224007780388694 Lenn, Y.-D., Che eskin, T. K., Sp in all, J., and McClean, J. L. (2011). Nea - su ace eddy hea and momen um luxes in he an a c ic ci cumpola cu en in d ake passage. J. Phys. Oceanog . 41, 1385–1407. doi: 10.1175/JPO-D-10- 05017.1 Le i us, S., An ono , J. I., Boye , T. P., Ba ano a, O. K., Ga cia, H. E., Loca nini, R. A., e al. (2012). Wo ld ocean hea con en and he mos e ic sea le el change (0- 2000 m), 1955-2010. Geophys. Res. Le . 39:L10603. doi: 10.1029/2012GL051106 Le i us, S., An ono , J. I., Boye , T. P., Loca nini, R. A., Ga cia, H. E., and Mishono , A. V. (2009). Global ocean hea con en 1955-2008 in ligh o ecen ly e ealed ins umen a ion p oblems. Geophys. Res. Le . 36:L07608. doi: 10.1029/2008GL037155 Li, Y., Han, W., Hu, A., Meehl, G. A., and Wang, F. (2018). Mul idecadal changes o he uppe Indian Ocean hea con en du ing 1965–2016. J. Clim. 31, 7863–7884. doi: 10.1175/JCLI-D-18-0116.1 Lima, M. O., Ci ano, M., Ma a, M., Goes, M., Goni, G., and Ba inge , M. O. (2016). An assessmen o he B azil Cu en ba oclinic s uc u e and a iabili y nea 22◦S in dis inc ocean o ecas ing and analysis sys ems. Ocean Dyn. 66, 893–916. doi: 10.1007/s10236-016-0959-6 Liu, Q.-Y., Feng, M., Wang, D., and Wij els, S. (2015). In e annual a iabili y o he Indonesian h ough low anspo : a e isi based on 30 yea expendable ba hy he mog aph da a. J. Geophys. Res. 120, 8270–8282. doi: 10.1002/2015JC011351 Lopez, H., Dong, S., Lee, S.-K., and Goni, G. (2016). Decadal modula ions o in e hemisphe ic global a mosphe ic ci cula ions and monsoons by he Sou h A lan ic me idional o e u ning ci cula ion. J. Clim. 29, 1831–1851. doi: 10.1175/JCLI-D-15-0491.1 Lopez, H., Goni, G., and Dong, S. (2017). A econs uc ed Sou h A lan ic me idional o e u ning ci cula ion ime se ies since 1870. Geophys. Res. Le . 44, 3309–3318. doi: 10.1002/2017GL073227 Lyman, J. M., Good, S. A., Gou e ski, V. V., Ishii, M., Johnson, G. C., Palme , M. D., e al. (2010). Robus wa ming o he global uppe ocean. Na u e 465, 334–337. doi: 10.1038/na u e09043 Lyman, J. M., and Johnson, G. C. (2008). Es ima ing annual global uppe -ocean hea con en anomalies despi e i egula in si u ocean sampling. J. Clim. 21, 5629–5641. doi: 10.1175/2008JCLI2259.1 Majumde , S., Goes, M., Poli o, P. S., Lumpkin, R., Schmid, C., and Lopez, H. (2019). P opaga ing modes o a iabili y and hei impac on he wes e n bounda y cu en in he Sou h A lan ic. J. Geophys. Res. 124, 3168–3185. doi: 10.1029/2018JC014812 Ma in, M. J., Balmaseda, M., Be ino, L., B asseu , P., B assing on, G., Cummings, e al. (2015). S a us and u u e o da a assimila ion in ope a ional oceanog aphy. J. Ope . Oceanog . 8, s28–s48. doi: 10.1080/1755876X.2015.1022055 McCaskill, C., Shay, L. K., B ews e , J. K., and Meye s, P. C. (2016). De elopmen and assessmen o he sys ema ically me ged Paci ic Ocean egional empe a u e and salini y (SPORTS) clima ology o ocean hea con en es ima ions. J. A mos. Oceanic Tech. 33, 2259–2272. doi: 10.1175/JTECH-D-15-0168.1 Mill, G. N., da Cos a, V. S., Lima, N. D., Gabioux, M., Gue a, L. A. A., and Pai a, A. M. (2015). No hwa d mig a ion o Cape São Tomé ings, B azil. Con . Shel Res. 106, 27–37. doi: 10.1016/j.cs .2015.06.010 Molina i, R. L. (2004). Annual and decadal a iabili y in he wes e n sub opical No h A lan ic: signal cha ac e is ics and sampling me hodologies. P og. Oceanog . 62, 33–66. doi: 10.1016/j.pocean.2004. 07.002 Molina i, R. L. (2011). In o ma ion om low-densi y expendable ba hy he mog aph ansec s: No h A lan ic mean empe a u e s uc u e and quasi-decadal a iabili y. P og. Oceanog . 88, 131–149. doi: 10.1016/j.pocean.2010.12.013 Mo ow, R., Donguy, J.-R., Chaigneau, A., and Rin oul, S. R. (2004). Cold co e anomalies a he suban a c ic on , sou h o Tasmania. Deep Sea Res. 51, 1417–1440. doi: 10.1016/j.ds .2004.07.005 Mo ow, R., and Kes ena e, E. (2014). Nine een-yea changes in su ace salini y in he Sou he n Ocean sou h o Aus alia. J. Ma . Sys. 129, 472–483. doi: 10.1016/j.jma sys.2013.09.011 Mo ow, R., Valladeau, G., and Sallée, J. (2008). Obse ed subsu ace signa u e o Sou he n Ocean decadal sea le el ise. P og. Oceanog . 77, 351–366. doi: 10.1016/j.pocean.2007.03.002 Mun o, D. R., Lo enduski, N. S., S ephens, B. B., Newbe ge , T., A igo, K. R., Takahashi, T., e al. (2015). Es ima es o ne communi y p oduc ion in he Sou he n Ocean de e mined om ime se ies obse a ions (2002–2011) o nu ien s, dissol ed ino ganic ca bon, and su ace ocean pCO2 in d ake passage. Deep Sea Res. 114, 49–63. doi: 10.1016/j.ds 2.2014.12.014 F on ie s in Ma ine Science | www. on ie sin.o g 22 July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k Nagano, A., Ichikawa, H., Yoshikawa, Y., Kizu, S., and Hanawa, K. (2012). Va ia ion o he sou hwa d in e io low o he No h Paci ic sub opical gy e, as e ealed by a epea hyd og aphic su ey. J. Oceanog . 68, 361–368. doi: 10.1007/s10872-012-0102-3 Nagano, A., Kizu, S., Hanawa, K., and Roemmich, D. (2016). Hea anspo a ia ion due o change o No h Paci ic sub opical gy e in e io low du ing 1993–2012. Ocean Dyn. 66, 1637–1649. doi: 10.1007/s10236-016-1007-2 Nakamu a, H., Sampe, T., Tanimo o, Y., and Shimpo, A. (2004). “Obse ed associa ions among s o m acks, je s eams and midla i ude oceanic on s,” in Ea h’s Clima e: The Ocean–A mosphe e In e ac ion, Geophysical Monog aph, Vol. 147, Ame ican Geophysical Union (Washing on, DC), 329–346. doi: 10.1029/147GM18 Napoli ano, E., Iacono, R., Ciu a di, T., Reseghe i, F., Poulain, P. M., and No a s e ano, G. (2018). The Ty henian In e media e Wa e (TIW): cha ac e iza ion and o ma ion mechanisms. P og. Oceanog . 170, 53–68. doi: 10.1016/j.pocean.2018.10.017 Nie es, V., Willis, J. K., and Pa ze , W. C. (2015). Recen hia us caused by decadal shi in Indo-Paci ic hea ing. Science 349, 532–535. doi: 10.1126/science.aaa4521 Oke, P. R., La nicol, G., Fujii, V., Smi h, G. C., Lea, D. J., Guinehu , S., e al. (2015a). Assessing he impac o obse a ions on ocean o ecas s and eanalyses: Pa 1, Global s udies. J. Ope . Oceanog . 8, s49–s62. doi: 10.1080/1755876X.2015.1022067 Oke, P. R., La nicol, G., Jones, E. M., Kou a alou, V., Spe e ik, A. K., Ca se, F., e al. (2015b). Assessing he impac o obse a ions on ocean o ecas s and eanalyses: Pa 2, Regional applica ions. J. Ope . Oceanog . 8, s63–s79. doi: 10.1080/1755876X.2015.1022080 Olson, D. B., B own, O. B., and Emme son, S. R. (1983). Gul S eam on al s a is ics om Flo ida S ai s o Cape Ha e as de i ed om sa elli e and his o ical da a. J. Geophys. Res. 88, 4569–4577. doi: 10.1029/JC088iC08p04569 Pezzi, L. P., Souza, R. B., Fa ias, P. C., Ace edo, O., and Mille , A. J. (2016). Ai -sea in e ac ion a he Sou he n B azilian Con inen al shel : in si u obse a ions. J. Geophys. Res. 121, 6671–6695. doi: 10.1002/2016JC011774 Pilo, G. S., Oke, P. R., Ryko a, T., Coleman, R., and Ridgway, K. (2015). Do Eas Aus alian cu en an icyclonic eddies lea e he Tasman Sea? J. Geophys. Res. 120, 8099–8114. doi: 10.1002/2015JC011026 Pol on, J. A., Lenn, Y. D., Elipo , S., Che eskin, T. K., and Sp in all, J. (2013). Can d ake passage obse a ions ma ch Ekman’s classic heo y? J. Phys. Oceanog . 43, 1733–1740. doi: 10.1175/JPO-D-13-034.1 Reseghe i, F., Cheng, L., Bo ghini, M., Yashayae , I. M., Rai e i, G., and Zhu, J. (2018). Assessmen o quali y and eliabili y o measu emen s wi h XBT Sippican T5 and T5/20. J. A mos. Oceanic Tech. 35, 1935–1960. doi: 10.1175/JTECH-D-18-0043.1 Rhein, M., Rin oul, S. R., Aoki, S., Campos, E., Chambe s, D., Feely, R. A., e al. (2013). “Obse a ions: ocean,” in Clima e Change 2013: The Physical Science Basis. Con ibu ion o Wo king G oup I o he Fi h Assessmen Repo o he In e go e nmen al Panel on Clima e Change, eds T. F. S ocke , D. Qin, G.-K. Pla ne , M. Tigno , S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P. M. Midgley (Camb idge, UK; New Yo k, NY: Camb idge Uni e si y P ess), 255–315. Ribo i, A., So gen e, R., Oli a, A., O ilia, F., Bo ghini, M., and Reseghe i, F. (2016). Indica ion o ecen wa ming p ocess a he in e media e le el in he Ty henian Sea om SOOP XBT measu emen s. Medi e . Ma . Sci. 17, 467–475. doi: 10.12681/mms.1196 Ridgway, K. R., Coleman, R. C., Bailey, R. J., and Su on, P. (2008). Decadal a iabili y o Eas Aus alian cu en anspo in e ed om epea ed high- densi y XBT ansec s, a CTD su ey and sa elli e al ime y. J. Geophys. Res. 113:C08039. doi: 10.1029/2007JC004664 Ridgway, K. R., and Dunn, J. R. (2003). Mesoscale s uc u e o he mean Eas Aus alian Cu en Sys em and i s ela ionship wi h opog aphy. P og. Oceanog . 56, 189–222. doi: 10.1016/S0079-6611(03)00004-1 Rin oul, S. R., Sokolo , S., and Chu ch, J. (2002). A 6 yea eco d o ba oclinic anspo a iabili y o he An a c ic ci cumpola cu en a 140◦E om expendable ba hy he mog aph and al ime y measu emen s. J. Geophys. Res. 107:19-1-19-22. doi: 10.1029/2001JC000787 Rise , S. C., F eeland, H. J., Roemmich, D., Wij els, S., T oisi, A., Belbéoch, M., e al. (2016). Fi een yea s o ocean obse a ions wi h he global A go a ay. Na . Clim. Change 6, 145–153. doi: 10.1038/nclima e2872 Roemmich, D., and Co nuelle, B. (1992). The sub opical mode wa e s o he Sou h Paci ic Ocean. J. Phys. Oceanog . 22, 1178–1187. doi: 10.1175/1520- 0485(1992)022<1178:TSMWOT>2.0.CO;2 Roemmich, D., and Gilson, J. (2001). Eddy anspo o hea and he mocline wa e s in he No h Paci ic: a key o in e annual/decadal clima e a iabili y? J. Phys. Oceanog . 31, 675–688. doi: 10.1175/1520- 0485(2001)031<0675:ETOHAT>2.0.CO;2 Roemmich, D., Gilson, J., Da is, R., Su on, P., Wij els, S., and Rise , S. (2007). Decadal spin up o he deep sub opical gy e in he Sou h Paci ic. J. Phys. Oceanog . 37, 162–173. doi: 10.1175/JPO3004.1 Roemmich, D., Gilson, J., Willis, J., Su on, P., and Ridgway, K. (2005). Closing he ime- a ying mass and hea budge s o la ge ocean a eas: he Tasman box. J. Clim. 18, 2330–2343. doi: 10.1175/JCLI3409.1 Rossby, T., Flagg, C., Cha ik, L., Ha den, B., and Søiland, H. (2018). A di ec es ima e o olume, hea , and eshwa e exchange ac oss he G eenland-Iceland-Fa oe-Sco land Ridge. J. Geophys. Res. 123, 7139–7153. doi: 10.1029/2018JC014250 Rossby, T., Flagg, C., and Donohue, K. (2010). On he a iabili y o Gul S eam anspo om seasonal o decadal imescales. J. Ma . Res. 68, 503–522. doi: 10.1357/002224010794657128 Rudnick, D. L. (2016). Ocean esea ch enabled by unde wa e glide s. Ann. Re . Ma . Sci. 8, 519–541. doi: 10.1146/annu e -ma ine-122414-033913 Sallée, J. B., Spee , K., and Mo ow, R. (2008). Response o he An a c ic ci cumpola cu en o a mosphe ic a iabili y. J. Clim. 21, 3020–3039. doi: 10.1175/2007JCLI1702.1 Sch oede , K., Chiggia o, J., Josey, S. A., Bo ghini, M., A ac i, S., and Spa nocchia, S. (2017). Rapid esponse o clima e change in a ma ginal sea. Scien. Rep. 7:4065. doi: 10.1038/s41598-017-04455-5 Shay, L. K., and B ews e , J. K. (2010). Oceanic hea con en a iabili y in he eas e n Paci ic Ocean o hu icane in ensi y o ecas ing. Mon. Wea. Re . 138, 2110–2131. doi: 10.1175/2010MWR3189.1 Shepa d, T., Ab aham, J. P., Schwalbach, D. S., Kane, S., Sigling, D., and Ha ing on, T. (2014). Veloci y and densi y e ec on impac o ce du ing wa e en y o sphe es. J. Geophys. Remo e Sens. 3:129. doi: 10.4172/2169-0049.1000129 She in, V. R., Du and, F., Gopalk ishna, V. V., Anu inda, S., Chai anya, A. V. S., Bou dalle-Badie, R., e al. (2018). Signa u e o indian Ocean dipole on he wes e n bounda y cu en o he Bay o Bengal. Deep Sea Res. 136, 91–106. doi: 10.1016/j.ds .2018.04.002 Shi, J.-R., Xie, S.-P., and Talley, L. D. (2018). E ol ing ela i e impo ance o he Sou he n Ocean and No h A lan ic in an h opogenic ocean hea up ake. J. Clim. 31, 7459–7479. doi: 10.1175/JCLI-D-18-0170.1 Sloyan, B. M., Ridgway, K. R., and Cowley, R. (2016). The eas aus alian cu en and p ope y anspo a 27◦S om 2012 o 2013. J. Phys. Oceanog . 46, 993–1008. doi: 10.1175/JPO-D-15-0052.1 Smi h, N., Ha ison, D., Bailey, R., Al es, O., Delc oix, T., Roemmich, K., e al. (2001). “The uppe ocean he mal ne wo k,” in F om: Obse ing he Oceans in he 21s Cen u y, eds C. Koblinsky and N. Smi h (Melbou ne, VIC: Bu eau o Me eo ology), 259–284. Sp in all, J. (2003). Subsu ace s uc u e o in e annual empe a u e anomalies in he Aus alian sec o o he Sou he n Ocean. J. Geophys. Res. 108:3285. doi: 10.1029/2002JC001494 Sp in all, J. (2008). Long- e m ends and in e annual a iabili y o empe a u e in D ake Passage. P og. Oceanog . 77, 316–330. doi: 10.1016/j.pocean.2006. 06.004 Sp in all, J., Che eskin, T. K., and Sweeney, C. (2012). High- esolu ion unde way uppe ocean and su ace a mosphe ic obse a ions in D ake Passage: syne gis ic measu emen s o clima e science. Oceanog aphy 25, 70–81. doi: 10.5670/oceanog.2012.77 Sp in all, J., Go don, A. L., Wij els, S. E., Feng, M., Hu, S., Koch-La ouy, A., e al. (2019). De ec ing change in he Indonesian seas. F on . Ma . Sci. 6:257. doi: 10.3389/ ma s.2019.00257 S ephenson, G. R., Gille, S. T., and Sp in all, J. (2012). Seasonal a iabili y o uppe ocean hea con en in D ake Passage. J. Geophys. Res. 117:C04019. doi: 10.1029/2011JC007772 S ephenson, G. R., Gille, S. T., and Sp in all, J. (2013). P ocesses con olling uppe -ocean hea con en in D ake Passage. J. Geophys. Res. 118, 4409–4423. doi: 10.1002/jg c.20315 F on ie s in Ma ine Science | www. on ie sin.o g 23 July 2019 | Volume 6 | A icle 452 Goni e al. Global XBT Ne wo k Su he s, I. M., Young, J. W., Bai d, M. E., Roughan, M., E e e , J. E., B assing on, G. B., e al. (2011). The s eng hening Eas Aus alian Cu en , i s eddies and biological e ec s - an in oduc ion and o e iew. Deep Sea Res. 58, 538–546. doi: 10.1016/j.ds 2.2010.09.029 Swa , N. C., Gille, S. T., Fy e, J. C., and Gille , N. P. (2018). Recen Sou he n Ocean wa ming and eshening d i en by g eenhouse gas emissions and ozone deple ion. Na . Geosci. 11, 836–841. doi: 10.1038/s41561-018-0226-1 Swa , S., Speich, S., Anso ge, I. J., Goni, G. J., Gladyshe , S., and Lu jeha ms, J. R. E. (2008). T anspo and a iabili y o he An a c ic Ci cumpola Cu en sou h o A ica. J. Geophys. Res. 113:C09014. doi: 10.1029/2007JC004223 Thada hil, P., Ghosh, A. K., Sa up ia, J. S., and Gopalak ishna, V. V. (2001). An in e ac i e g aphical sys em o XBT da a quali y con ol and isualiza ion. Comp. Geosci. 27, 867–876. doi: 10.1016/S0098-3004(00)00172-2 Thada hil, P., Sa an, A. K., Gopalak ishna, V. V., Ve hamony, P., and A aligidad, N. (2002). XBT all a e in wa e s o ex eme empe a u e: a case s udy in he An a c ic Ocean. J. A mos. Oceanic Tech. 19, 391–397. doi: 10.1175/1520-0426-19.3.391 Thompson, A. F., Gille, S. T., MacKinnon, J. A., and Sp in all, J. (2007). Spa ial and empo al pa e ns o small-scale mixing in D ake Passage. J. Phys. Oceanog . 37, 572–592. doi: 10.1175/JPO3021.1 Todd, R., Goes, M., Ba inge , M., Dong, S., Goni, G., Volko , D., e al. (2019). Global Pe spec i es on obse ing ocean bounda y cu en sys ems. F on . Ma ine Sci. 6:423. doi: 10.3389/ ma s.2019.00423 Tonani, M., Balmaseda, M., Be ino, L., Blockley, E., B assing on, G., Da idson, F., e al. (2015). S a us and u u e o global and egional ocean p edic ion sys ems. J. Ope . Oceanog . 8, s201–s220. doi: 10.1080/1755876X.2015.1049892 Tsubouchi, T., Suga, T., and Hanawa, K. (2007). Th ee ypes o Sou h Paci ic sub opical mode wa e s: Thei ela ion o he la ge-scale ci cula ion o he Sou h Paci ic Sub opical Gy e and hei empo al a iabili y. J. Phys. Oceanog . 37, 2478–2490. doi: 10.1175/JPO3132.1 Ueha a, H., Kizu, S., Hanawa, K., Yoshikawa, Y., and Roemmich, D. (2008). Es ima ion o hea and eshwa e anspo s in he No h Paci ic using high- esolu ion expendable ba hy he mog aph da a. J. Geophys. Res. 113:C02014. doi: 10.1029/2007JC004165 Ve dy, A., Co nuelle, B., Mazlo , M. R., and Rudnick, D. L. (2017). Es ima ion o he opical paci ic ocean s a e 2010-13. J. A mos. Oceanic Tech. 34, 1501–1517. doi: 10.1175/JTECH-D-16-0223.1 Vignudelli, S., Cipollini, P., Reseghe i, F., Fusco, G., Gaspa ini, G. P., and Manzella, G. M. R. (2003). Compa ison be ween XBT da a and TOPEX/Poseidon sa elli e al ime y in he ligu ian- y henian a ea. Ann. Geophys. A mos. Hyd osphe es Space Sci. 21, 123–135. doi: 10.5194/angeo-21-123-2003 Von Schuckmann, K., Le T aon, P. Y., Aaboe, S., Fanjul, E. A., Au e , E., and Axell, L. (2018). Cope nicus ma ine se ice ocean s a e epo . J. Ope . Oceanog . 11, S1–S142. doi: 10.1080/1755876X.2018.14 89208 Wang, B., Liu, J., Kim, H.-J., Webs e , P. J., and Yim, S.-Y. (2012). Recen change o he global monsoon p ecipi a ion (1979–2008). Clim. Dyn. 39, 1123–1135. doi: 10.1007/s00382-011-1266-z Wij els, S., and Meye s, G. (2004). An in e sec ion o oceanic wa eguides: a iabili y in he indonesian h ough low egion. J. Phys. Oceanog . 34, 1232–1253. doi: 10.1175/1520-0485(2004)034<1232:AIOOWV>2.0.CO;2 Wij els, S. E., Willis, J., Domingues, C. M., Ba ke , P., Whi e, N. J., G onell, A., e al. (2008). Changing expendable ba hy he mog aph all a es and hei impac on es ima es o he mos e ic sea le el ise. J. Clim. 21, 5657–5672. doi: 10.1175/2008JCLI2290.1 WMO (2015a). Manual on Codes. Vol. I.2. WMO-306. WMO, Gene a. ISBN:978- 92-63-10306-2. WMO (2015b). Manual on he Global Telecommunica ion Sys em. WMO-386. WMO, Gene a. ISBN: 978-92-63-10386-4. Zilbe man, N. V., Roemmich, D. H., Gille, S. T., and Gilson, J. (2018). Es ima ing he eloci y and anspo o wes e n bounda y cu en sys ems: a case s udy o he Eas Aus alian Cu en nea B isbane. J. A mos. Oceanic Tech. 35, 1313–1329. doi: 10.1175/JTECH-D-17-0153.1 Con lic o In e es S a emen : The au ho s decla e ha he submi ed wo k was no ca ied ou in he p esence o any pe sonal, p o essional, o inancial ela ionships ha could po en ially be cons ued as a con lic o in e es . Copy igh © 2019 Goni, Sp in all, B ingas, Cheng, Ci ano, Dong, Domingues, Goes, Lopez, Mo ow, Ri e o, Rossby, Todd, T inanes, Zilbe man, Ba inge , Boye , Cowley, Domingues, Hu chinson, K amp, Ma a, Reseghe i, Sun, Bhaska TVS and Volko . This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License (CC BY). The use, dis ibu ion o ep oduc ion in o he o ums is pe mi ed, p o ided he o iginal au ho (s) and he copy igh owne (s) a e c edi ed and ha he o iginal publica ion in his jou nal is ci ed, in acco dance wi h accep ed academic p ac ice. No use, dis ibu ion o ep oduc ion is pe mi ed which does no comply wi h hese e ms. F on ie s in Ma ine Science | www. on ie sin.o g 24 July 2019 | Volume 6 | A icle 452