REDUCED-SCALE MODELING AND DATA ASSIMILATION FOR DECADAL-SCALE FORECASTS OF THE GEOMAGNETIC FIELD THIS PROPOSAL ADDRESSES SCIENCE CHALLENGES FROM THE NASA EARTH SURFACE AND INTERIOR (ESI) WORKSHOP "WHAT ARE THE DYNAMICS OF THE EARTH'S MAGNETIC FIELD AND ITS INTERACTIONS WITH THE REST OF THE EARTH'S SYSTEM?" THE EARTH'S MAGNETIC FIELD IS GENERATED BY CONVECTIVE MOTION WITHIN THE LIQUID OUTER CORE AND CHANGES ON TIME SCALES THAT VARY FROM DECADES TO MILLIONS OF YEARS. THESE TIME SCALES ARE KNOWN FROM THE PALEOMAGNETIC RECORD AND MORE RECENT OBSERVATIONS FROM SAILING SHIP LOGS AND GEOMAGNETIC OBSERVATORIES. DURING THE PAST SEVERAL DECADES NASA HAS BEEN INVOLVED IN THE STUDY OF THE TIME VARYING GEOMAGNETIC FIELD MOST RECENTLY WITH SATELLITE MISSIONS IN PARTNERSHIP WITH OTHER NATIONS E.G. ORSTED OR SWARM. THESE DATA HAVE BEEN COMBINED WITH NUMERICAL MODELS DEVELOPED AT NASA GSFC TO SIMULATE THE FLUID MOTION IN THE OUTER CORE THE (MAIN) SOURCE OF THE GEOMAGNETIC FIELD. THE PROCESS BY WHICH DATA ARE COMBINED WITH NUMERICAL MODELS IS CALLED "DATA ASSIMILATION". DATA ASSIMILATION GENERATES ACCURATE ESTIMATES AND FORECASTS OF CHANGES IN THE EARTH'S MAGNETIC FIELD. SUCH FORECASTS HAVE IMPORTANT SOCIETAL APPLICATIONS PARTICULARLY IN NAVIGATION. INSIGHTS IN GEOMAGNETIC VARIABILITY ALSO HELP US TO UNDERSTAND THE EARTH'S ROTATIONAL VARIATIONS A KEY RESEARCH OBJECTIVE OF NASA'S EARTH SURFACE AND INTERIOR PROGRAM BECAUSE ROTATIONAL VARIATIONS DEFINE THE TERRESTRIAL REFERENCE FRAME WHICH IS USED FOR ALL EARTH SCIENCE AND APPLICATION MISSIONS. THE PROPOSAL'S SCIENTIFIC VALUE LIES PRIMARILY IN NEW INFORMATION ABOUT THE FLUID MOTION IN THE OUTER CORE AS WELL AS THE EXTENSIVE USE OF THE INTERNATIONAL GEOMAGNETIC REFERENCE FIELD PRODUCED EVERY 5 YEARS. DATA ASSIMILATION IN GEOMAGNETIC MODELING AND FORECASTING WAS PIONEERED BY NASA SCIENTISTS AND SINCE THEN HAS ATTRACTED THE ATTENTION OF MANY GEOMAGNETIC FIELD MODELERS. MAKING THE BEST FORECASTS OF THE GEOMAGNETIC FIELD ON DECADAL TIME SCALES REQUIRES THAT ALL AVAILABLE OBSERVATIONS (SATELLITE DATA DATA FROM THE EXTENSIVE OBSERVATORY NETWORK THE LONGER PALEOMAGNETIC RECORD) ARE EFFECTIVELY ASSIMILATED INTO GEOMAGNETIC FIELD MODELS. HOWEVER THERE ARE SEVERAL FUNDAMENTAL QUESTIONS THAT REQUIRE CAREFUL ANSWERS BEFORE DATA ASSIMILATION CAN BE TRULY SUCCESSFUL IN GEOMAGNETIC FORECASTING. MY PROPOSED RESEARCH FOCUSES ON ANSWERING ONE OF THESE QUESTIONS: WHAT IS THE MOST EFFECTIVE DATA ASSIMILATION METHOD THAT LEADS TO THE LARGEST GAINS IN FORECAST SKILL OF NASA'S DECADAL-SCALE GEOMAGNETIC FORECASTING SYSTEM? MY PROPOSED WORK WILL MAKE USE OF NASA'S GEOMAGNETIC DATA ASSIMILATION SYSTEM (MOSST-DAS UNIQUE IN THE US) WHICH COMBINES A NUMERICAL MODEL OF THE EARTH'S GEOMAGNETIC FIELD WITH OBSERVATIONS COLLECTED OVER THE PAST FEW DECADES DURING SATELLITE MISSIONS BY NASA OR WITH NASA PARTICIPATION. A SPECIFIC PRODUCT THAT I WILL CREATE IS AN EXTENSIVE NUMERICAL AND COMPUTATIONAL TEST-SUITE THAT CAN BE USED BY NASA SCIENTISTS FOR RAPID-PROTOTYPING AND TESTING OF NEW DATA ASSIMILATION ALGORITHMS FOR NASA'S OPERATIONAL GEOMAGNETIC FORECASTING SYSTEM.