u.s. integrated ocean observing system · regional observing systems, in support of goos. the...

U.S. DEPARTMENTOF COMMERCE

National Oceanicand AtmosphericAdministration

Vol. 14, No. 2 ● Spring 2004

EARTH SYSTEM MONITOR

A guide toNOAA's data and

informationservices

3News briefs

7Northern Hemisphereannular mode in the

ocean

11Central Californiacoastal upwelling

U.S. Integrated Ocean Observing System

DEP

ARTMENT OF COMMERC

E

★ ★

UN

ITEDSTATES OF AMER

ICA

Steve HankinChair, IOOS/DMAC Steering CommitteeNOAA Pacific Marine Environmental Laboratory

Thomas C. Malone, PhDDirector, OceanUS Office

Eric Lindstrom, PhDNASA Oceanography Program Scientist andNASA Liaison to OceanUS Office

Roz CohenOceanUS Office andNOAA National Oceanographic Data Center

Data and communications infrastructure

▲ Figure 1. The U.S. Integrated Ocean Observing System

— continued on page 2

INSIDE

Congress has directed the U.S. marine envi-ronmental science communities to come to-gether to plan, design, and implement asustained Integrated Ocean Observing System(IOOS). The IOOS is envisioned as a network ofobservational, data management, and analysessystems that rapidly and systematically acquiresand disseminates marine data and data productsdescribing the past present and future states ofthe oceans. Existing and planned observing sys-

tem elements will be integrated into a sustainedocean observing system that addresses both re-search and operational needs in the followingareas:• detecting and forecasting oceanic compo-

nents of climate variability;•␣ facilitating safe and efficient marine opera-

tions;• ensuring national security;• managing resources for sustainable use;• preserving and restoring healthy marine

ecosystems;• mitigating natural hazards;• ensuring public health.

The OceanUS Office, which is responsible forplanning and coordinating the development ofIOOS, was established in 2000 under the auspicesof the interagency National Oceanographic Part-nership Program (NOPP). The IOOS is the U.S.contribution to the Global Ocean Observing Sys-tem (GOOS). Planning for GOOS, an initiative ofthe Intergovernmental Oceanographic Commis-sion, began in the early 1990’s. GOOS is also a

2 Spring 2004EARTH SYSTEM MONITOR

EARTH SYSTEM MONITOR

The Earth System Monitor (ISSN 1068-2678) is published quarterly by the NOAANational Oceanographic Data Center.Past issues are available online at http://www.nodc.noaa.gov/General/NODCPubs/

Questions, comments, or suggestionsfor articles, as well as requests forsubscriptions and changes of address,should be directed to the editor.

The mailing address for the Earth SystemMonitor is:

National Oceanographic Data CenterNOAA/NESDIS E/OC1SSMC3, 4th Floor1315 East-West HighwaySilver Spring, MD 20910-3282

EDITORR. Torstenson

Telephone: 301-713-3281 ext.107Fax: 301-713-3302

E-mail: [email protected]

DISCLAIMERMention in the Earth System Monitor ofcommercial companies or commercialproducts does not constitute an endorse-ment or recommendation by the NationalOceanic and Atmospheric Administrationor the U.S. Department of Commerce.Use for publicity or advertising purposes ofinformation published in the Earth SystemMonitor concerning proprietary productsor the tests of such products is notauthorized.

U.S. DEPARTMENT OF COMMERCEDonald Evans, Secretary

National Oceanic andAtmospheric Administration

Conrad C. Lautenbacher, Jr.,Under Secretary and Administrator

NA

TIO

NA

LO

CEA

NICAND ATMOSPHERIC

ADMIN

IST

RA

TIO

N

U.S. DEPARTMENT OF COMMER

CE

U.S. IOOS, from page 1

Rosalind CohenNOAA/NESDISNational Oceanographic Data Center1315 East-West HighwaySSMC3, Rm 4822Silver Spring, MD 20910E-mail: [email protected] — continued on page 4

component of the Integrated GlobalObserving Strategy developed througha partnership of international agenciesconcerned with global environmentalissues.

IOOS is developing as two closelycoordinated components — global andcoastal — that encompass the broadrange of scales required to assess,detect, and predict the impacts ofglobal climate change, weather, andhuman activities on the marine envi-ronment (Figure 1). The global compo-nent consists of an internationalpartnership to improve forecasts andassessments of weather, climate, oceanstate, and boundary conditions forregional observing systems, in supportof GOOS. The coastal componentblends national observations in theExclusive Economic Zone (EEZ) withmeasurement networks that are man-aged regionally to improve assessmentsand predictions of the effects ofweather, climate, and human activitieson the state of the coastal ocean, itsecosystems and living resources, andthe nation’s economy. The coastal com-ponent encompasses the nation’s EEZ,Great Lakes, and estuaries. Detailedinformation on the planning andimplementation of the global andcoastal components of IOOS can befound in the proceedings of the 2002OceanUS Workshop (OceanUS, 2002);Parts I and II of the IOOS Implementa-tion Plan (OceanUS, 2003a; OceanUS,2003b); recent editions of the MTSJournal and Oceanography devoted toocean observing systems (MTS, 2003;Oceanography, 2003), and the proceed-ings of the 2003 Regional Summit(OceanUS, 2003c).

IOOS is composed of three sub-systems: the Observing Subsystem(collection of environmental measure-ments and their transmission from plat-forms); the Data Management andCommunications Subsystem (DMAC -discovery and delivery of data within

IOOS and interoperability with otherrelevant observing systems); and theModeling and Analysis Subsystem(evaluation and prediction of the stateof the marine environment). A coher-ent strategy for integrating marine datastreams across disciplines, organiza-tions, time scales, and geographic loca-tions is central to the success of IOOS.The DMAC Subsystem is the primaryintegrating element of IOOS. It pro-vides the linkages among IOOS compo-nents and partner organizations, as wellas to systems in other disciplines (i.e.,terrestrial and atmospheric), and tointernational marine data managementsystems.

Data management & communicationsIn the spring of 2002, the OceanUS

Office appointed the Data Managementand Communications Steering Com-mittee (DMAC-SC) to develop a de-tailed, phased implementation plan forthis IOOS subsystem. The DMAC Plan(Hankin and DMAC-SC, 2003) consistsof three main parts: Part I provides anoverview of the system requirements,the unique challenges it presents, andthe plan for addressing them; Part IIpresents the detailed implementationplan leading to an Initial OperationalCapability for the DMAC subsystemwithin five years; and Part III, the Ap-pendices, provides in-depth discussionof key technical topics.

IOOS Data FlowIOOS observations originate at in-

ternational, national, and regionalobserving systems (Figure 2, page 5).The raw measurements from these sys-tems are transferred by various mecha-nisms (e.g., mail, telephone, radio andsatellite transmission, microwave, andthe Internet) to centers where primarydata assembly and (typically) qualitycontrol occurs. Generally some form ofprimary data assembly is requiredbefore IOOS data can be utilized. Thisprocessing may involve, for example,hand entry of data from log sheets, orreal-time conversion of instrumentvoltages to physical units.

The DMAC subsystem “begins”with the output from the primary dataassembly centers. At the entry points to

3Spring 2004 EARTH SYSTEM MONITOR

News briefsNew International GeophysicalYear, IGY-2

The 50th anniversary of the Interna-tional Geophysical Year (IGY) and of theWorld Data Centers (WDC) will occur in2007. A resolution supporting IGY-2 wasapproved by the U.S. House and the Sen-ate. The resolution states that the Presi-dent should submit to Congress, byMarch 15, 2004, a report detailing thesteps taken in carrying out the possibleactivities and organizational structures forIGY-2 in 2007-2008. The Directors of theNational Academy of Sciences and theNational Science Foundation and theAdministrator of the National Aeronauticsand Space Administration are tasked withpreparing the report. The National Geo-physical Data Center and the WDCs playan important role in IGY-2, as it is definedby the IGY-2 task team.

For NESDIS, the most significantelement of IGY-2 is the electronic Geo-physical Year and the creation of "virtualobservatories." The electronic GeophysicalYear (eGY) is intended to promote im-proved access to all geophysical datawhether they reside in data centers, inlaboratories, or on a scientist's web site."Virtual Observatories" will improve theuse of geophysical data by 1) organizingall available sources of data for a specificregion and 2) merging these comprehen-sive data sources with data assimilationmodels and physical models to computethe best description of the environmentfor a given location and time. "VirtualObservatories" are needed by satelliteoperators to determine and mitigate thecauses of operational anomalies and bythe transportation industry to determineprecise locations using GPS systems.

GOES space environment monitorsupports Mars Mission

The Odyssey Spacecraft in orbitaround Mars carries instrumentation toevaluate the radiation environment in thevicinity of Mars, to plan the design offuture manned spacecraft and missions tothe planet. The Martian Radiation Envi-ronment Experiment (MARIE) is a particlespectrometer that measures radiation (20-450 MeV) from solar energetic particleevents and galactic cosmic rays. TheANSER Corporation is using the GOESenergetic particle archives to provide asecond point of reference for the MARIEdata and to validate the results.

NOAA progresses toward parallelCORS-West facility

Representatives from NOAA Satellitesand Information, NOAA Oceans, andNOAA Research, including the SpaceEnvironment Center (SEC) and ForecastSystems Lab (FSL) met with an audienceof over 50 state and local government,industry, and university constituents onFebruary 25 to celebrate progress towardthe establishment of a ContinuouslyOperating Reference Station (CORS)-Global Positioning System (GPS) paralleldata facility in Boulder, Colorado. Talksfocused on the CORS-West facility andthe geo-location, meteorological, andspace weather products created fromCORS-GPS data.

The National Geodetic Survey (NOS/NGS) and National Geophysical DataCenter (NESDIS/NGDC) are working to-gether to establish a parallel facility that isgeographically separate from the primaryCORS facility in Silver Spring, MD. CORS-West would actively collect, process, dis-tribute, and archive CORS data, providinguninterrupted services to users and ensur-ing the collection and access to theseimportant data.

The value of CORS data to users wasestimated (May 2002) at $72 million peryear. This annual value has more thandoubled every year since the inception ofthe CORS program in 1994. The parallelfacility in Boulder, CO, ensures continuedcollection and access to these data in theevent the primary facility in Silver Springis down.

With the expanded use of CORS datain near real-time weather and spaceweather products, the Boulder facilityprovides faster access, a level of redun-dancy, and improved services - both toNOAA Research and to the surveying andgeodetic community. This effort involvesthree NOAA line offices, plus the SpaceEnvironment Center which will soon be-come part of the National Weather Ser-vice, a fourth line office.

NOAA Multibeam III meetingThe Third NOAA Multibeam Sonar /

Ocean Mapping Workshop was held atthe National Geophysical Data Center inMarch 2004. Significant exchanges ofdata content requirements, processingstreams, and product requirements weremade between the various NOAA com-munities using multibeam systems: CoastSurvey, Sanctuaries, Fisheries, Research,and Ocean Exploration. Of particular em-phasis for this conference was an ex-tended discussion of Fisheries' need forvolumetric mapping of the water columnfor fish inventory analysis, a three-dimen-sional (rather than the traditional two-dimensional) use of multibeam mappingsystems. Agreement was reached on theinitial steps toward coordinating the gen-eration of metadata and the cooperativeplanning of collection efforts. The criticalconcept developed was that ocean map-ping needs to be managed NOAA-wide,as it represents a primary activity for allthose offices involved with the "O" inNOAA. The three-day meeting included15 presentations, two breakout sessions, atour of the Forecast Systems Laboratories,"Science on a Sphere," and a tour ofNGDC.

Climatography of the United StatesNo. 20, 1971-2000 period

The Climatography of the UnitedStates No. 20 (CLIM20), Monthly StationClimate Summaries for the 1971-2000period of record have been completedand will be available online in March2004 (http://www.nndc.noaa.gov/cgi-bin/climatenormals/climatenormals.pl?directive=prod_select).The new CLIM 20 updates and expandsthe previous version, last published in1985 and covering the period 1951-1980. The types of statistics includemeans, medians, extremes, mean numberof days exceeding threshold values, andprobabilities from monthly precipitationand freeze data. There is also a table foreach station with heating, cooling, andgrowing degree days for various tempera-ture bases.

The summaries can be ordered byindividual station, state, or subscriptionthrough the National Oceanic and Atmo-spheric Administration's National ClimaticData Center Online Store. A CLIM20 CD-ROM product will be available by earlysummer 2004.


the DMAC Subsystem, data share acommon Data Communications Infra-structure - standards and protocolswhich support:• IOOS-wide descriptions of data sets

(via metadata);• the ability to search for and find

data sets of interest (data discov-ery);

• the ability to access data in aninteroperable manner from clientapplications (data transport);

• the ability to review the datathrough common Web browsers(online browse).

The DMAC Subsystem also includesArchive Centers which, in the matureIOOS, will utilize the Data Communica-tions Infrastructure for the receipt anddistribution of data. The DMAC Sub-system “ends” with the delivery of dataand data products in usable form tousers who wish to have access to thesedata and data products, or to organiza-tions and applications that producevalue-added information products forend users. These information productsinclude text forecasts, maps of oceanvariables and coastal features, materialsfor educational curricula, and timeseries showing trends useful forresource managers.

DMAC subsystem - data communicationsinfrastructure

The Data Discovery capability ofthe DMAC Subsystem depends on theexistence of accurate and timely meta-data accompanying all IOOS data sets.The DMAC Plan specifies that all IOOSmetadata records must be compliantwith the Federal Geographic Data Com-mittee (FGDC) metadata standards.Metadata creation is typically the re-sponsibility of the data providers.TheDMAC Plan calls for tools and trainingto be made available to assist dataproviders with this task. The establish-ment of a community-based core meta-data working group is recommended todevelop an FGDC profile that addressesthose elements that are shared in com-mon by all DMAC metadata. The Planalso calls for additional working groupsto address more specialized, discipline-specific needs. The working groups will

decide which metadata standards toadopt, and how they need be extendedto address conflicts in terminology.Tools and techniques for translationamong different metadata standards arealso discussed in the Plan.

A fundamental concept for DMACis the “Web Services”, which supportthe ability to connect independent datamanagement systems. In the DMACPlan, the term Web Services is used torefer to reusable software componentsbased on the eXtensible Markup Lan-guage (XML) that provide a standard-ized way for computer systems torequest data and data processing fromone another. These services useHypertext Transfer Protocol (HTTP),the ubiquitous communications proto-col of the World Wide Web (Web).They are accessed through the familiarURIs (Universal Resource Identifiers)that begin with “http://”. DMAC willdefine uniform means to access diversedata and metadata management sys-tems through the use of Web Services.

Data Discovery will be achievedthrough searches of IOOS FGDC meta-data records using both commercialWeb search engines such as Google andYahoo and more specialized geospatialsearches. DMAC metadata searcheswill allow users to formulate queriesthrough Web portals (Web sites), andwill also support machine-to-machinesearches using Web Services interfaces.More advanced data discovery methodssuch as the Semantic Web (http://www.w3.org/2001/sw) will be evaluatedfor use by DMAC and incorporated asthey mature.

As described earlier, IOOS data willbe accessible through DMAC DataTransport standards and protocols, sub-sequent to primary data assembly(Figure 1). The OPeNDAP [Open sourceProject for a Network Data Access Pro-tocol; this is a non-profit corporationformed to develop and maintain themiddleware formerly known as theDistributed Ocean Data System (DODS);see http://www.unidata.ucar.edu/packages/dods] data access protocol isthe most widely tested and acceptedsystem for distributed data accesswithin the marine sciences. OPeNDAP,which has been in use since 1995, un-derlies the National Virtual Ocean Data

System (NVODS); NVODS was createdwith support from a Broad AgencyAnnouncement (BAA) issued by theNational Oceanographic PartnershipProgram in 2000. OPeNDAP employs adiscipline neutral approach to theencapsulation of data for transport(data structures or vocabulary are nottied to a single field of study). Thisfeature of OPeNDAP is very importantfor IOOS because of the wide variety ofdata types and users it encompasses.The DMAC-SC has designatedOPeNDAP as an initial operationalcomponent for transport of griddeddata, and a pilot component for deliv-ery of non-gridded data.

"Operational" is the designationused in IOOS for activities and systemsin the fourth, most mature stage ofdevelopment; earlier stages are research,pilot, and pre-operational; see the IOOSImplementation Plan Part I. "Pilot" isthe designation used in IOOS for activi-ties and systems in the second stage ofdevelopment leading to the operationallevel.

All data in the mature DMACsubsystem must be delivered with con-sistent syntactic and semantic datamodels (or families of data models) toachieve the desired level of interoper-ability. Syntactic refers to the atomicdata types in a data set (ie., ASCII,binary, real, etc.), the data structures,dimensions of the data arrays, etc;semantic refers to the meaning of vari-ables (ie., T represents ocean temperature), special value flags (ie., -1 meansmissing data, 0 means land data, etc),units, etc. The DMAC Plan calls forcommunity-based working groups to beconvened to develop comprehensivedata models. The semantic data modelmust harmonize with existing modelsin other data communities, especiallythose models used by geographic infor-mation systems (GIS) and climate mod-eling. It must standardize controlledvocabularies, including the encoding ofocean biological and taxonomic datasuch as those used by the Ocean Bio-geographic Information System (OBIS);OBIS is a globally distributed networkof systematic, ecological, and environ-mental data systems (http://iobis.org). Itmust also describe a diverse range of

U.S. IOOS, from page 2



▲ Figure 2. Data flow in the U.S. Integrated Ocean Observing System (IOOS). Elements that are outlined in darker gray are classifiedas components of DMAC.


data structures, including spectral andfinite element models, arbitrary curvi-linear coordinate systems, and multi-level hierarchies of in situ measure-ments. Users of the mature DMAC Sub-system will find that software applica-tions which they use for scientificanalyses and product generation areadapted to work directly with DMACWeb Services. Access to distributedIOOS data will be transparent.

A basic browsing and visualizationcapability that extends across all datain the DMAC Subsystem will berequired for effective management.Standard web browsers will provideuniform access to geo- and time-refer-enced graphics. Users will reach theOn-line Browse functionality throughDMAC Data Discovery, which can thenuse DMAC Data Transport protocols toaccess specific subsets of data. The prin-ciple users of the Online Browse willinitially be marine data specialists whohave responsibilities for managingelements of IOOS, however, the capa-bilities will doubtless be useful to manyother user groups. An NVODS render-ing and visualization tool, the LiveAccess Server (LAS), has proven to beeffective in the delivery of OnlineBrowse capabilities across a broad rangeof marine data types; http://www.ferret.noaa.gov/LAS. The LAS has been desig-nated as an initial pre-operationalcomponent of Online Browse by theDMAC-SC; "pre-operational" is the des-ignation used in IOOS for activities andsystems in the third stage of develop-ment leading to the operational level.Other emerging technologies such asOpen GIS may be integrated into theOnline Browse and Data Transport so-lutions as they mature.

DMAC Subsystem - data archive andaccess

A DMAC Archive working groupwill be convened early in the imple-mentation of the DMAC Subsystem.This group will establish guidelineswhich must be adhered to by all IOOSArchive Centers. The working groupwill inventory existing archives andanticipated data streams to ensure that

U.S. IOOS, from page 5 all data deemed appropriate for archivalhave designated Archive Centers. Theywill also participate in the developmentof DMAC metadata standards to ensurethat elements essential for data steward-ship are included. Key features ofthe Archive Centers are as follows:• Data will be accessible online to the

greatest extent possible, at no costto users. Data from offline sourceswill similarly be available at notmore than the cost of providingservice.

• Data and metadata will be madeavailable using DMAC Data Transport protocols, metadata standards,and Data Discovery interfaces.

• Data and metadata migration planswill be developed to accommodateevolution of media and systems,and ensure long-term preservationof irreplaceable data. All relevantIOOS data policies and Federalregulations will be followed.

Next stepsPlanning and concept development

for the IOOS are well underway. Imple-mentation plans are being formulatedand budget initiatives are movingforward. The ocean community isworking with the OceanUS Office toestablish regional associations whichwill oversee coastal observing systemdevelopment on a regional scale. ANational Federation of RegionalAssociations is being established toenhance cooperation among regionsand provide a unified national interfacefor these efforts. The global componentof IOOS is more advanced than thecoastal component, but furtherexpansion is needed in the global arenaas well to meet IOOS goals.

The DMAC Plan provides aroadmap for implementation of a datacommunications infrastructure that willmeet the needs of the marine environ-mental observing community. The in-formation technology required to meetmost of the DMAC requirements, whilechallenging, can be developed fromexisting capabilities through relativelystraightforward software engineering.The major challenge facing DMAC is

one of coordination and cooperationamong the IOOS partners and usercommunities. DMAC will only succeedif barriers to participation are mini-mized, and participants realize thatadopting DMAC data and metadatastandards, communication protocols,software, and policies is to their advan-tage. A strong outreach program is criti-cal to foster continued coordination,cooperation, and support for IOOSwithin the regional, national, andinternational marine environmentalobserving communities.

ReferencesHankin, S. and the DMAC Steering Committee,

2003. The U.S. Integrated OceanObserving System (IOOS) Plan for DataManagement and Communications(DMAC), Parts I-III, OceanUS, Arlington,VA (Sept. 2003 Draft). [http://dmac.ocean.us/dacsc/imp_plan.jsp]

Hankin, S., L. Bernard, P. Cornillon, F. Grassle,D. Legler, J. Lever, and S. Worley, 2003.Designing the Data Management andCommunications Infrastructure for theU.S. Integrated Ocean Observing System.MTS Journal, 37(3): 51-54.

MTS, 2003. Ocean Observing Systems. MarineTechnology Society Journal, 37(3) 159pp.

Oceanography, 2003. Ocean Observatories /Ocean Eddies in the 1539 Carta Marina /Ice Keel Scour Marks on Mars / Impact ofClimate Variability on Right Whales. TheOceanography Society, 16(4).

OceanUS, 2002. Building Consensus: Towardan Integrated and Sustained OceanObserving System (IOOS), OceanUS,Arlington, VA, 175pp. [http://www.ocean.us/documents/docs/Core_lores .pdf]

OceanUS, 2003a. Implementation of the InitialU.S. Integrated Ocean Observing System.Part I. Structure and Governance. OceanUS, Arlington, VA. 36pp. [http://www.ocean.us/documents/docs/ioos_plan_ 6.11.03.pdf]

OceanUS, 2003b. Implementation of theInitial U.S. Integrated Ocean ObservingSystem. Part II. Building the Initial IOOS.OceanUS, Arlington, VA. 42pp.

OceanUS, 2003c. Regional Ocean ObservingSystems: An Ocean.US Summit.Washington, DC. OceanUS, Arlington,VA. 32pp. [http://www.ocean.us/documents/componentsIOOS.jsp] ■


The Northern Hemisphere annular modein the ocean

Boundary currents and heat transportation

Kathryn A. Kelly and Shenfu Dong*Applied Physics LaboratorySchool of OceanographyUniversity of Washington*presently at the Scripps Institution ofOceanography

Strong western boundary currentsin the Northern Hemisphere oceanstransport heat from the warm tropicalregions to the midlatitudes, wheremuch of the heat is fluxed to the atmo-sphere as the warm currents encountercooler air masses. Some of this heatcontinues on into the subpolar gyre towarm the high-latitude regions andsome heat is stored in a region of recir-

culating currents with deep wintertimemixed layers.

Observations of sea surface height(SSH) anomalies from the TOPEX/POSEIDON radar altimeter since 1992suggest that there are large interannual-to-decadal variations in the structure ofthese current systems (Qiu, 2000; Vivieret al., 2002; Dong and Kelly, 2004). InFigure 1 the SSH anomaly has beencombined with an estimate of the meanSSH to illustrate the nature of the circu-lation changes. Large-scale anomaliesoccur just to the south of the currentcore (the closely spaced SSH contours);maps of SSH show similar changes inthe Atlantic and in the Pacific, with an

expanded region of high SSH in 1993(not shown) and 1999, and a con-tracted region of high SSH in 1996.Changes in SSH correspond to changesin heat content, because a warmingwater column expands, increasing SSH.

Dominant modes of heat contentBecause the SSH time series (10

years) is too short to give a reliable esti-mate of the dominant modes of heatcontent variability, we compute theempirical orthogonal functions (EOFs)of heat content from the Joint Environ-mental Data Analysis Center (JEDAC)for 1955-2001. The fraction of variance

▲ Figure 1. Sea surface height maps from the TOPEX/POSEIDON altimeter. At the left is the Gulf Stream region in the North Atlantic,and at right is the Kuroshio Extension region in the North Pacific, for years (top) 1996 and (bottom) 1999. Units are in meters.More positive SSH indicates more heat stored in the ocean.



Kathryn A. Kelly, Principal OceanographerAir-sea Interaction and Remote SensingDepartmentApplied Physics LaboratoryBox 355640University of WashingtonSeattle, Washington 98195Email: [email protected]

Northern hemisphere, from page 7in the first two modes is 26% and 16%respectively, and both are significant atthe 95% level (Kelly and Dong, inpress). The first mode of heat content(Figure 2a) shows maxima in the exten-sion regions of the western boundarycurrents, with coherent variations be-tween the North Atlantic and the NorthPacific. Variability in the second mode(not shown) is primarily in the PacificOcean. The heat content time seriesreveal similarities with climate indices:the first mode of heat content (Figure2b) resembles the Arctic Oscillation(AO) Index (Thompson and Wallace,1998), whereas the second mode ofheat content resembles the PacificDecadal Oscillation (PDO, Hare andMantua, 2000).

An EOF analysis of SSH from thealtimeter of 1992-2002 gives modes(not shown) that resemble those of theJEDAC heat content, but with the firstand second mode order reversed. Thereversal of the mode order suggests thatthe Pacific mode variance was largerduring the 1990s than was typical forthe longer period. The SSH mode timeseries also resemble these same climateindices, with the coherent mode (EOF2) resembling the AO.

Northern Annular Mode in the oceanThe Northern hemisphere Annular

Mode (NAM, Wallace and Thompson,2000), describes zonally coherent fluc-tuations in sea level pressure (SLP). TheAO Index is the principal component(time series) of the SLP mode. A posi-tive AO corresponds to anomalouslystrong westerlies throughout the north-ern hemisphere.

Coherent variations of heat con-tent are correlated (r = 0.49, significantat 95%), with AO leading the heat con-tent by 13 months, which suggests thatthe heat content anomalies are causedby changes in the strength of the

midlatitude westerly winds. The sim-plest explanation, that stronger windscause larger air-sea fluxes and thereforea loss of ocean heat, can be ruled out bythe sign of the correlation (Figure 2).Strong westerlies correspond to highheat content.

The upper ocean heat budgetTo understand the causes of ob-

served fluctuations in ocean heat con-tent and SSH, parallel studies of theupper ocean heat budget were con-ducted for the western North Pacificand North Atlantic (Vivier et al., 2002;Dong and Kelly, 2004). A simple upperocean layer model was forced by surfaceheat fluxes, winds, and currents to pre-dict temperature for the regions shownin Figure 1. Geostrophic velocity wasspecified from altimeter data andEkman velocity was estimated fromwind stress. The North Atlantic and theNorth Pacific heat budgets both showedthat advection is the dominant termresponsible for the interannual changesin upper ocean heat content; net sur-face heat fluxes are nearly as large asthe advection term, but not consis-tently correlated with the tendency ofheat content. The contribution of geo-strophic currents to advection is sub-stantially larger than that by the Ekmancomponent (Qiu, 2000; Dong andKelly, 2004).

Based on these heat budget studies,we argue that the observed changes inheat content reflect changes in heattransport caused by changes in oceancirculation, not by air-sea fluxes.Therefore, the correlation with strongwesterlies likely reflects a strengtheningof ocean circulation that in turn causesan increase in advection.

The heat content anomalies southof the western boundary currents arecaused by changes in the upper oceanthermal structure, in waters that arepart of the upper ocean mixed layeronly in winter. Positive anomalies ofheat content (and SSH) are negativelycorrelated with the thickness of the 18-degree layer (subtropical mode water)in the Gulf Stream region, based on a50-yr record (Kwon, 2003; Dong, 2004).A similar relationship between themode water layer thickness and fluctua-tions in the Kuroshio Extension has

been found by Yasuda and Hanawa(1997). The negative correlation meansthat subtropical mode water contrib-utes a heat content deficit to the upperocean.

Heat flux anomalies forced by heatcontent

During periods of high heat con-tent in the western boundary currents,fluxes of heat from the ocean to theatmosphere are also anomalously large.To estimate the effect of the coherentheat content anomalies (Figure 2a) onair-sea fluxes, we regress the NCEP/NCAR net surface fluxes onto the firstmode of heat content (Figure 2b). Theregression coefficients (Figure 3) arenegative with maximum amplitudes inthe western boundary current exten-sion regions, consistent with regions ofhigh heat content forcing anomalousfluxes of heat from the ocean. The mag-nitude of these flux anomalies is aslarge as 15-25 watts per square meter,approxmately 20% of the annual meanoceanic heat loss in these regions.

The association of larger oceanicheat losses with increased ocean heatcon itent, caused by advection, suggeststhat the ocean may be forcing changesin the atmosphere in the vicinity of thewestern boundary currents oninterannual-to-decadal time scales. Theidea that the western boundary cur-rents force an atmospheric response hasbeen raised by numerous authors, par-ticularly in regard to changing the win-tertime “storm tracks” (Rodwell et al.,1999; Nakamura et al., 1997; Joyce etal., 2000).

ConclusionsA surprising fraction (26%) of the

variations in midlatitude heat contentare coherent between the North Atlan-tic and the North Pacific and correlatedwith the Northern hemisphere AnnularMode (NAM) or Arctic Oscillation (AO).These coherent variations have theirlargest amplitudes in the extensionregions of the northern hemispheremidlatitude western boundary currents,the Gulf Stream and the Kuroshio Ex-tension. Although it is tempting toattribute the ocean heat content varia-tions to changes in the surface fluxes



▲ Figures 2a,b. Northern Hemisphere Annular Mode. At top (a) are the first EOF's of heat content (colors) and sea level pressure(contours), and below (b) are the time series of SLP (Arctic Oscillation Index, solid) and heat content (dashed). Heat content and SLPmodes are negatively correlated with the AO leading heat content anomalies by 13 months.


associated with changes in wind speed,the sign and phase of the correlations isinconsistent with a passive response ofthe ocean to air-sea heat fluxes.

Studies of the upper ocean heatbudgets near western boundary currentextensions have shown that the heatcontent variations are caused by oceanadvection, rather than by surfacefluxes. Much of the anomalous heatfrom advection is stored south of theboundary currents, in the subtropicalmode waters. The source of the changesin ocean advection, which in turncause most of the interannual changesin heat content, appears to be forcingby midlatitude winds.

Changes in ocean heat contenthave implications for the interannual-to-decadal scale climate variationsthrough their contributions to air-seafluxes: ocean heat content anomaliesforce fluxes, rather than the other wayaround, as in the ocean interior. Theair-sea flux anomalies associated withchanges in ocean heat content are ofthe order of 20% of the annual meanvalue. Although the geographical ex-tent of these regions is small, they con-tribute most of the midlatitude air-seafluxes.

ReferencesDong, S., 2004: Interannual variations in up-per ocean heat content, heat transport, andconvergence in the Western North Atlantic.Ph.D. thesis, School of Oceanography, Univer-sity of Washington.Dong, S. and K. A. Kelly, 2004: The heat bud-get in the Gulf Stream region: The importanceof heat storage and advection. J. Phys.Oceanogr., in press.Hare, S. R., and N. J. Mantua, 2000: Empiricalevidence for North Pacific regime shifts in 1977and 1989. Progress in Oceanography, 47,103-145.Joyce, T. M., C. Deser, and M. A. Spall, 2000:Relation between decadal variability of Sub-tropical Mode Water and the North AtlanticOscillation. J. Clim., 13, 2550-2569.Kelly, K. A., and S. Dong, 2004: The relation-ship of western boundary current heat trans-port and storage to midlatitude ocean-atmosphere interaction. “Ocean-AtmosphereInteraction and Climate Variability,” editors: C.Wang, S.-P. Xie, and J. A. Carton, AGU mono-graph, in press.Kwon, Y.-O., 2003: Observations of generalcirculation and water mass variability in theNorth Atlantic subtropical mode water region.Ph. D. thesis, School of Oceanography, Univer-sity of Washington.

Nakamura, H., G. Lin, and T. Yamagata, 1997:Decadal climate variability in the North Pacificduring recent decades. Bull Am. Meteor. Soc.,78, 2215-2225.Qiu, B., 2000: Interannual variability of theKuroshio Extension system and its impact onthe wintertime SST field. J. Phys. Oceanogr.,30, 1486-1502.Rodwell, M. J., D. P. Rowell, and C. K. Folland,1999: Oceanic forcing of the wintertimeNorth Atlantic Oscillation and European cli-mate. Nature, 398, 320-323.Thompson, D. W. J, and J. M. Wallace: 1998.The Arctic Oscillation signature in the winter-time geopotential height and temperaturefields. Geophys. Res. Lett., 25, 1297-1300.(http://www.jisao.washington.edu/ao)Vivier, F., K. A. Kelly, and L. Thompson: 2002.Heat budget of the Kuroshio Extension region:1993-1999. J. Phys. Oceanogr., 32, 3436-3454.Wallace, J. M. and D. W. J. Thompson: 2002.The Pacific center of action of the NorthernHemisphere annular mode: real or artifact?J. Clim., 15, 1987-1991.Yasuda, T. and K. Hanawa, 1997: Decadalchanges in the mode waters in the midlatitudeNorth Pacific. J. Phys. Oceanogr., 27, 858-870. ■

Northern Hemisphere, from page 8

▲ Figure 3. Relationship of net surface flux to heat content. Regression coefficients of net surface flux onto the first EOF of heatcontent (Figure 2b). Negative values indicate flux of heat from the ocean to the atmosphere associated with positive heat contentanomalies.


▲ Figure 1. Krill biomass increase during the typical upwelling period of 2001. Image: MBARI.

Upwelling on the central California coastAn introduction to the Sanctuary Integrated Monitoring Network

The Monterey Bay (California)National Marine Sanctuary is an inter-nationally recognized location for ma-rine research, resource management,and policy. Over forty institutions andorganizations in the greater MontereyBay area are currently conducting re-search, which includes long-term moni-toring programs that are essential tothe further understanding and healthof the marine ecosystem.

The Sanctuary Integrated Monitor-ing Network (SIMoN) serves as a focalpoint to integrate the existing monitor-ing programs and to identify gaps ininformation. SIMoN makes the moni-toring data available to managers, deci-sion makers, the research community,and the general public. More informa-tion on SIMoN can be found at thewebsite: http://www.mbnms-simon.org.

Josh Pederson, Sarah Smith, andJenne HolmgrenMonterey Bay National Marine Sanctuary

Jenne HolmgrenSIMoN OutreachMonterey Bay National Marine SanctuaryMonterey, California 93940Email: [email protected]

One of the major studies in theSanctuary that SIMoN serves is that ofthe central coastal upwelling, whichtypically begins in March and endsaround July. The effects of this seasonaloccurrence can not only be seen inmarine organisms such as phytoplank-ton, zooplankton (Figure 1), and thefish and marine mammals that feedupon them, but in the weather as well.The traditional foggy summers on theCentral California coast are attributedto the physical interaction between thewarm summer air and the newly up-welled cold ocean surface waters. Asthese waters cool the summer air, mois-ture collects and forms the coastal fogcommon to the Sanctuary.

Scientists have developed instru-ments and technology to detect up-welling at the level of the primaryproducers. One instrument, the fluo-rometer, detects pigments, primarilychlorophyll, that phytoplankton use to

perform photosynthesis. Chlorophyllmeasurements are used by oceanogra-phers to determine the level of biologi-cal productivity throughout the foodchain. Satellite imagery is used to detectpigmentation of the phytoplanktonwhich reveals the quantity and distri-bution of phytoplankton over a givenarea of the ocean.

In the Monterey Bay NationalMarine Sanctuary (Figure 2, page 12) avariety of research groups are dedicatedto collecting and analyzing informationabout the upwelling process. The UCSanta Cruz Center for Integrated Ma-rine Technologies (CIMT) is collectingdata on wind, primary productivity,krill, ocean temperature, seabirds,whales, and other components of theupwelling system. This data is analyzedand integrated to produce an enhancedunderstanding of the relationship be-tween primary productivity and highertrophic level organisms such as fishes,marine mammals, and seabirds.


12 Spring 2004EARTH SYSTEM MONITORU

.S. DEP

AR

TM

ENT

OF C

OM

MER

CE

Natio

nal O

ceanic an

d A

tmo

sph

eric Ad

min

istration

Publication Distribution Facility

1315 East-West H

ighway

Silver Spring, M

D 20910-3282

ATTN

: Earth System M

onitor

Address C

orrection Requested

OFFIC

IAL BU

SINESS

Penalty for Private Use $300

Upwelling, from page 11

▲ Figure 2. Map of the Monterey Bay National Marine Sanctuary region; imagefrom the SIMoN home page at http://www.mbnms-simon.org.

The Monterey Bay Aquarium Re-search Institute (MBARI) also studiescoastal upwelling in the Sanctuary. TheSimulations of Coastal Ocean Physicsand Ecosystems (SCOPE) project atMBARI assimilates data from satellitesand ocean sensors to model the inter-connected physical, chemical, and bio-logical processes associated withupwelling. These models will helpunderstand the distribution of marinemammals, seabirds, and fish during aperiod of upwelling, and will helpresource management as well as direct

future research projects and observa-tional efforts.

With so many marine organismsutilizing the upwelling process for foodand reproduction, it is important thatMonterey Bay National Marine Sanctu-ary staff understand the dynamics ofthis annual process in order to effec-tively manage these interconnectedresources during this time of the year.Fortunately, there are programs such asCIMT and SCOPE that are dedicated toenhancing the Sanctuary's knowledgeof this incredible oceanic phenomenon.■

u.s. integrated ocean observing system · regional observing systems, in support of goos. the...

Documents