S PACE W EATHER :C IVILIAN R ESEARCH TO O PERATIONS (R2O) AC OMMUNITY P ROPOSAL 1 OBJECTIVE The DoD and NOAA have the U.S. Government responsibility for providing operational space weather support for military and civilian cus‐ tomers. NASA has its own space weather re‐ quirements in support of human and robotic space exploration. In addition, the growing commercial sector also has a stake in space weather. Both the government and the com‐ mercial sector have an urgent need for vali‐ dated, physics‐based models that can give as much as 5 days of advanced warning of major space weather events with well quantified un‐ certainties. NOAA and commercial companies, however, are not funding basic research and thus have only marginal influence on the mod‐ els developed and used by the research com‐ munity. NASA and NSF are the primary federal agencies supporting basic research of space weather phenomena. They also support the develop‐ ment and community use of physics‐based “end‐to‐end” models at universities, non‐profit organizations, government laboratories and small businesses. In effect, they control the de‐ velopment and research applications of sophis‐ ticated, high‐end computational models that provide vastly improved capabilities to model complex space weather phenomena. There are other government agencies and/or laboratories that are also involved in civilian space weather research with somewhat limited mandate and resources. They include AFOSR and AFRL on the USAF side, ONR and NRL from the Navy and some other offices of the DoD. While there is remarkable cooperation among these agencies there are also very clear differ‐ ences in their mandates, cultures and philoso‐ phy. A smooth and efficient “pipeline” from the basic research community to civilian operations requires active and enthusiastic participation by everyone involved. In practice this means that there is an urgent need to establish a fair and efficient civilian research‐to‐operations (R2O) pipeline and governance structure that encourages participation by everyone involved in the space weather enterprise. In contrast with the civilian space weather R2O chain, the DoD has an established R2O transi‐ tion process. AFRL and NRL, and to a lesser ex‐ tent ARL, have the mandate to provide science and technology options and capability in this area to the operational DoD units. For space weather, it is the Air Force Space Command (AFSPC) that translates needed warfighter ca‐ pabilities into requirements for the acquisition community, which includes the Air Force Space and Missile System Center (SMC). In the case of AFSPC requirements translated into AFWA pro‐ curement needs, there are processes to develop research objectives within the AFRL and NRL. In theory, these requirements are coordinated with AFOSR and ONR to guide investments in the most basic research at the Technology Readiness Level (TRL) 1‐2 level. Products and capabilities that reach the level TRL 6 level within the labs can then be transitioned into operational capabilities through SMC. The goal of this document is to outline possible models for the civilian R2O process and gov‐ ernance. 2 PRESENT SITUATION In the last decade very significant progress has been achieved in our ability to understand and even forecast space weather. This progress was made possible by the significant investments made by NASA’s Heliophysics Division and in‐ creased funding opportunities offered by the DoD (four space weather MURIs) and NSF (an STC). The resulting capabilities include two “Sun‐to‐Mud” coupled model systems (the CISM model chain and the Space Weather Modeling Framework), a coupled model of the magneto‐
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
S P A C E W E A T H E R : C I V I L I A N R E S E A R C H T O O P E R A T I O N S ( R 2O ) A C OMMUN I T Y P R O P O S A L
1 OBJECTIVE
TheDoD andNOAAhave theU.S. Governmentresponsibility for providing operational spaceweather support for military and civilian cus‐tomers. NASA has its own space weather re‐quirements in support of human and roboticspace exploration. In addition, the growingcommercial sector also has a stake in spaceweather. Both the government and the com‐mercial sector have an urgent need for vali‐dated, physics‐based models that can give asmuch as 5 days of advancedwarning ofmajorspaceweather eventswithwell quantified un‐certainties. NOAA and commercial companies,however, are not funding basic research andthushaveonlymarginal influenceonthemod‐els developed and used by the research com‐munity.
NASAandNSFaretheprimaryfederalagenciessupporting basic research of space weatherphenomena. They also support the develop‐ment and community use of physics‐based“end‐to‐end”models at universities, non‐profitorganizations, government laboratories andsmallbusinesses. Ineffect, theycontrolthede‐velopmentandresearchapplicationsofsophis‐ticated, high‐end computational models thatprovide vastly improved capabilities to modelcomplexspaceweatherphenomena.
There are other government agencies and/orlaboratories that are also involved in civilianspaceweatherresearchwithsomewhatlimitedmandate and resources. They include AFOSRandAFRLontheUSAFside,ONRandNRLfromtheNavyandsomeotherofficesoftheDoD.
While there is remarkable cooperation amongthese agencies there are also very clear differ‐ences in theirmandates, cultures and philoso‐phy.Asmoothandefficient“pipeline”fromthebasicresearchcommunitytocivilianoperationsrequires active and enthusiastic participationby everyone involved. In practice this means
that there is anurgentneed to establish a fairand efficient civilian research‐to‐operations(R2O) pipeline and governance structure thatencouragesparticipationbyeveryone involvedinthespaceweatherenterprise.
IncontrastwiththecivilianspaceweatherR2Ochain, the DoD has an established R2O transi‐tionprocess.AFRLandNRL,andtoalesserex‐tentARL,have themandate toprovidescienceand technology options and capability in thisarea to the operational DoD units. For spaceweather, it is the Air Force Space Command(AFSPC) that translates needed warfighter ca‐pabilities intorequirements for theacquisitioncommunity,whichincludestheAirForceSpaceandMissileSystemCenter(SMC).InthecaseofAFSPCrequirementstranslatedintoAFWApro‐curementneeds,thereareprocessestodevelopresearch objectives within the AFRL and NRL.In theory, these requirements are coordinatedwith AFOSR and ONR to guide investments inthe most basic research at the TechnologyReadinessLevel (TRL)1‐2 level. Productsandcapabilities that reach the level TRL 6 levelwithin the labs can then be transitioned intooperationalcapabilitiesthroughSMC.
Thegoalofthisdocumentistooutlinepossiblemodels for the civilian R2O process and gov‐ernance.
2 PRESENT SITUATION
Inthelastdecadeverysignificantprogresshasbeenachievedinourabilitytounderstandandevenforecastspaceweather.Thisprogresswasmade possible by the significant investmentsmade by NASA’s Heliophysics Division and in‐creased funding opportunities offered by theDoD (four spaceweatherMURIs) andNSF (anSTC). The resulting capabilities include two“Sun‐to‐Mud”coupledmodelsystems(theCISMmodel chain and the SpaceWeather ModelingFramework), a coupledmodelof themagneto‐
R2OProposal page2of5 1/19/09
sphere – ionosphere – thermosphere(OpenGGCM), and the first assimilative iono‐sphere model (GAIM). It is important to notethat the DoDMURIs and theNSF STC injected$65M additional funding to the spaceweathermodel development community. By the natureof these awards most of the funding went touniversities.Aconsequenceofthisisthattodaythe model developer community is primarilybased at universities, a fact that greatly con‐tributed to the vitality and dynamism of thefield.
The last decade also witnessed the success ofthe Community Coordinated Modeling Center(CCMC).Createdbyaninteragencycooperation,CCMC not only became the place where mosthigh‐end global models are available for com‐munityuse,butitisalsoprovidingindependentmodelvalidationstudiesaswellasexperimen‐tal real‐time simulations in support of spaceweatheroperations.AtthecoreofCCMC’ssuc‐cess is the philosophy that it provides a com‐munity service by bridging the gap betweenmodeldevelopersandtheusercommunity.Asaconsequence, CCMC does not participate inmodeldevelopmentanddoesnotcompetewiththemodel developer community. This divisionof labor is critical for the trust between themodeldevelopersandCCMC.
A decade ago there was an attempt to createtwo rapidprototyping centers (RPC) to transi‐
tion physics‐based models from the CCMC tooperations.Unfortunately,theseeffortsdidnotbring fruit and they eventually failed. The ab‐senceofRPCshavecreatedasignificantholeinthecivilianR2Ochainandthisholesignificantlyhurts our national effort to achieve a smoothandefficienttransitionfromresearchtoopera‐tions,andultimatelyhurtsthenationalefforttocreate a robust and reliable space weatherforecastcapability.
3 THE MISSING LINK
Themainweakness of the present situation isthat the link between the operations centersandtheresearchcommunity isveryweak(seeFig.1). SinceneitherNOAAnorAFWA is fund‐ing external civilian space weather research,theoperationscentershavelittleornoleverageover the direction of the research community.Thereisverylittlecoordination(letalonetran‐sition)betweenCCMCand theoperations cen‐ters.ByitsnatureCCMCisprimarilyservingtheresearchcommunity,afactthatisadirectcon‐sequenceof the funding sources of CCMC: it islocated at GSFC and it is supported by the re‐searchfundingagencies,NASA,NSF,AFOSRandONR.
Thereisaveryclearnationalneedforstrongercenters, either funded by NOAA and AFWA orby revenues generated from the commercialsector.Suchcenterswillhavethetaskofready‐
Fig.1:Presentstatusofthecivilianspaceweatherresearchtooperationsprocess.ThereisagapbetweenCCMCandtheoperationscenter.
R2OProposal page3of5 1/19/09
ing physics‐based models for operations. ThisincludestakingTRL6modelsthatarewellvali‐dated and have been demonstrated to standaloneontheirownmerit,movingthemthroughTRL 7 (test linkage of models to input opera‐tionaldatastreamsandoutputprocessorswithsuccessfulsystemleveltests),andthensuccess‐fully demonstrating them at TRL 8 (prototypedemonstration of coupled operational inputdata and model operations with vali‐dated/verified output and after a reasonable“burn‐in” full operations testingperiod). Tran‐sition to TRL 9 (turn‐key actual operations) isthe goal to this transition activity but it is theresponsibility of SWPC, AFWA, or commercialoperators to implement systems at this level.TRLs 7‐8 focus on “hardening” the codes foroperational use, ensuring input redundancy,writingtheinterfacecodebetweenmodelsanddata, applying rigorous software engineeringpractices, generating detailed documentations,extensivetestinginoperationallyrelevantenvi‐ronments, and providing support services foroperational use. While many of these tasksmust involve the model developers (whoshouldbe fundedtoprovidethissupport),oneoptionistodevelopaSpaceWeatherPredictionTestbed(SWPT)thatcouldbeatagovernmentlaboratory, a nonprofit institution or run byindustry, including small businesses, whichcould carryoutmost of the tasks.Anotherop‐tion is the broadening of CCMC and co‐locate
SWPTwithCCMC.
4 SWPT
There is a clear need to transition physics‐basedmodels to theoperationalagencies fromthe CCMC. Either the SWPT or a broadenedCCMC could provide a facility where spaceweathermodels can be tested and “hardened”in a development setting that closely matchesthose within the operational agencies. Thesemodels will be supported through a code re‐pository, documentation, tutorials and helpdesks.TheSWPTandtheCCMC,inconjunctionwith industry and small business partners,should take an aggressive education campaignabout the need for space weather services topotentialusersandhelp themdevelop tools tointerfacewithspaceweatherinformation.
In its mature form the SWPT could be a rela‐tivelysmall(~10FTE)andindependentopera‐tion.Itshouldhaveverylimiteddirectassocia‐tionwithmodeldevelopersoranyorganizationthat is participating in space weather science.Thisindependenceiscriticalforachievingtrustand credibility, both from theoperational cen‐ters(AFWAandSWPC)andfromthemodelde‐velopers. One solution would be a small busi‐nessthatissolelydedicatedtoSWPT.Thelive‐lihoodof this smallbusinesswoulddependonthe successof SWPTand the satisfactionof itsfundingagencies(AFWAandSWPC).
Fig.2:TheproposedcivilianR2Ochain
R2OProposal page4of5 1/19/09
5 SWPT GOVERNANCE
SWPTwould be a non‐scientific support orga‐nization, focusing on software engineering tomake the link between space weather systemoutputsand specialized customerneeds,docu‐mentation, standards and best practices inte‐gration, code repositories and software sup‐port.Theywouldnotmakescientificoropera‐tional decisions and would work with the op‐erationalagencies,CCMCandthedevelopersofthecodesthataretransitioned.
In all the SWPToptions, a SteeringCommitteeco‐chaired by representatives of AFWA andSWPCwouldmakescientificandpoliticaldeci‐sions.The co‐chairsof theSteeringCommitteewouldselectthemodelstobetransitionedwithadvise from themembers.A competitiveproc‐ess for model selection within the bounds ofclear requirements and guidelines is a provenmethod for allowing the best capabilities tocome forward. This is necessary, since AFWAandSWPChavetheresponsibility forU.S.Gov‐ernment operational forecasting of spaceweather. The Steering Committee would alsoevaluatetheworkofSWPTandwouldhavetherighttoimplementchangesinSWPTifneeded.TheSteeringCommitteememberswouldcomefromAFRL,CCMC,NASA,andNSF.Itwouldalsohave representatives from the university com‐munity, nonprofits/small businesses, and fromgovernmentlaboratories.
6 PROPOSED R2O CHAIN
Fig.2 shows the proposed R2O chain. TheCommitteeonSpaceWeatherof theNSWPco‐ordinatestheentirespaceweatherenterprise.
Attheresearchend,thecoordinationismainlyamong the research funding agencies: AFOSR,NASA, NSF and ONR. It is hoped that the re‐searchcommunitywillbeabletowinlargeandprestigiousawards,likeMURIandSTC.
IfaCCMC‐directedoption is selected, it ispro‐posed that the governance structure of CCMCbe modified, independent of which option isimplemented. CCMC is amature and very suc‐cessful enterprise that has earned the trust ofthemodelingandusercommunities.Itdoesits
jobwell.ThepresentCCMCSteeringCommitteeis exclusively composed of representatives ofgovernmentagenciesandlaboratories.ThenewCCMCSCwould replace the present governingstructure. Representatives of NSF and NASAwould co‐chair the SC. The members of theCCMC SC would include representatives ofAFOSR, SWPC and AFWA, 3 representativesfrom universities, 1 representative from non‐profit or small business research enterprises,and 2 representatives of other governmentlaboratories(oneof thesemightbeONR).Thiscomposition would ensure broad representa‐tion of the research community and participa‐tion by the operations community (AFWA andSWPC).ThespecificsoftheCCMCSC(lengthofterm,selectionprocess,etc.)needtobeworkedoutinaseparatedocument.
As itwasdiscussedabove, theSWPTSCwouldbe co‐chairedby representativesofAFWAandSWPC. The specifics of the proposed govern‐ance model is described in section 5 of thisdocument.
7 THE FAIRNESS ISSUE
Whoever implements theSWPT, andwhateverformittakes,thecommunitywillbeconcernedthat selection of a particularmodel for opera‐tionalstatuswouldeffectivelyanointitas“spe‐cial status”, giving it an advantage in futurefunding competitions and possibly undermin‐ing development of other models. This wouldstifle competitive innovation in the modelingcommunity and damage the ongoing advance‐mentsthatareneededbytheoperationalcom‐munity.
Onewaytohelpmitigatethisproblemmightbeto establish mechanisms to compare the per‐formance of operational models with othermodelsinthecommunity,withtheexplicitgoalofupgradingoperationalcapabilitybybringinginmoreordifferentmodels.Thisistheconceptbehind ensemble model operational forecast‐ing.Onecouldimagine,forexample,thatCCMC,AFRL,orotherscouldruncomparisonsagainstexisting forecast metrics, or against metricsthat represent desired new forecast capabili‐ties, using a wide range of models from the
R2OProposal page5of5 1/19/09
community.Thiscouldhaveseveralbenefits:
1) Stimulate competitive improvements inexistingoperationalforecastproducts,andprovide a way to assess different modelsfornext‐generationforecastgoals.
2) Provideabasistointroducemultiplemod‐els intooperations forensemble‐like fore‐castingorforforecastingparticularthingsatwhichdifferentmodelsexcel.
3) Reduce the finality of “anointing” a par‐ticular model and provide a transparentway to recover from selection decisionsthatmaybeimperfect.
4) Ensemblemodeling is a process that cap‐tures the best of models, helps to definethe uncertainty, and stimulates develop‐ment. CCMC and SWPT would certainlywanttocollaborateonbringingmodels toanensemblecapabilityfortransitioningtooperations.
It is recommended that the operational agen‐ciesandSWPT implementmechanisms tosup‐portongoingreassessmentsofselectedmodelsas comparedwith others. Thiswould facilitatecomparisons and development, and should re‐ducethetimeandeffortneededtobringinnewmodels.
Another widely shared concern is that SWPTshouldnotbepartofany institutionthat is in‐volved in space weather model development.Suchamodelwould inherentlycreatean insti‐tutional conflict of interest and severely un‐derminethe integrityandcredibilityof theen‐tireR2Ochain.
Finally,objectivestandardsthatareclearlyde‐fined in written documents are absolutely es‐sential for the credibility and fairness of theentireprocess. Ifwedonotdo this, themodelselectionprocesswillnotbetransparentandalarge fraction of themodel development com‐munitywillbealienated.Itisveryimportanttonote that model developers get “fame andglory” by having their models transitioned tooperations, but very little funding. Developingan appropriate incentive system is, therefore,veryimportant.
8 SUMMARY
Thisdocumentoutlines twopossibleorganiza‐tions of the space weather research to opera‐tionschain.Whiletheseoptionsarenotuniqueandofferonlyahighleveldescriptionofasolu‐tion, it is strongly felt that these solutionswould maintain a vibrant and healthy modeldevelopment community, make use of theuniquerolethatsmallbusinesscanplayintak‐ing space weather to commercial customers,while ensuring efficient transition of first‐principles based models to operations. It alsosatisfies the requirements of the operationscommunityasunderstoodbytheauthorsofthisdocument.
9 AUTHORS
Thisproposalwaswrittenby a small groupofspaceweathermodeldevelopers:
TamasGombosiRollinM.GerstackerProfessorEngineeringUniversityofMichigan
JoachimRaederAssociateProfessorofPhysicsUniversityofNewHampshire
RobertSchunkProfessorofPhysicsUtahStateUniversity
KentTobiskaPresidentandChiefScientistSpaceEnvironmentTechnologies
Related Documents
