OCO-2 HAS PROVIDED THREE YEARS OF GLOBAL OBSERVATIONS OF THE VERTICALLY INTEGRATED ATMOSPHERIC CARBON DIOXIDE (CO2) MOLE FRACTION XCO2. VARIATIONS IN XCO2 IN BOTH TIME AND SPACE PROVIDE INFORMATION THAT CAN BE USED TO INFER SURFACE FLUXES OF CO2 TO THEATMOSPHERE BUT MUST BE INTERPRETED IN LIGHT OF OTHER FACTORS (INCLUDING FINE-SCALE ATMOSPHERIC TRANSPORT AND RETRIEVAL ERRORS) THAT ALSOAFFECT OBSERVED VARIATIONS. DURING THE FIRST PHASE OF OCO-2 RESEARCH (2015-PRESENT) OUR PROPOSAL TEAM USED GEOSTATISTICAL ANALYSIS TOQUANTIFY THE SYNOPTIC/MESOSCALE VARIABILITY AND ERROR CHARACTERISTICS OF OCO-2 OBSERVATIONS. HERE WE PROPOSE TO USE THIS FRAMEWORKTO EXPLORE THE TERRESTRIAL OCEANIC AND HUMAN MECHANISMS DRIVING THE SUBANNUAL TO INTERANNUAL VARIABILITY IN ATMOSPHERIC XCO2OBSERVATIONS. OCO-2 PROVIDES A UNIQUE OPPORTUNITY TO ADDRESS THIS QUESTION SINCE MOST SURFACE IN SITU CO2 OBSERVATIONS SUCH AS FROMTHE NOAA GREENHOUSE GAS NETWORK ARE MOSTLY SITED IN COASTAL AND ISLAND LOCATIONS; THUS WHILE THERE IS A GOOD RECORD OF BACKGROUNDGLOBAL CO2 THERE ARE VERY FEW DIRECT OBSERVATIONS OVER THE OTHER CONTINENTS ESPECIALLY IN THE TROPICS AND FOR LARGE SWATHS OF THEOCEAN. WE WILL LEVERAGE THE GLOBAL COVERAGE OF OCO-2 OBSERVATIONS TO PARTITION INTERANNUAL VARIABILITY (IAV) IN CARBON FLUXES AMONGLAND AND OCEAN PROCESSES WITH A PARTICULAR FOCUS ON UNDERSTANDING THE RESPONSE TO THE 2015/2016 EL NINO. OUR RESEARCH WILL BEORGANIZED AROUND THE FOLLOWING OBJECTIVES:OBJECTIVE 1: CHARACTERIZE THE MAGNITUDE AND SPATIAL PATTERN OF XCO2 VARIABILITY AT 10X10 RESOLUTION USING A GEOSTATISTICAL ANALYSISFRAMEWORK.OBJECTIVE 2: ATTRIBUTE THE XCO2 ANOMALIES IDENTIFIED UNDER OBJECTIVE 1 TO TERRESTRIAL OCEAN AND HUMAN-MEDIATED SURFACE CO2 FLUXANOMALIES USING A SIMPLE FINGERPRINTING-INVERSE MODEL APPROACH TAKING INTO ACCOUNT THE RIGOROUS UNCERTAINTY ESTIMATES FROM OURGEOSTATISTICAL FRAMEWORK. WE EXPECT THAT THIS ANALYSIS WILL YIELD INSIGHTS INTO THE MECHANISTIC LINKS BETWEEN REGIONAL XCO2VARIATIONSAND VARIATIONS IN LAND AND OCEAN CLIMATE VIA SUCH DRIVERS AS TEMPERATURE PRECIPITATION AND CHANGES IN OCEAN CIRCULATION.OBJECTIVE 3: COMPARE XCO2 ANOMALIES AND SURFACE CO2 FLUX ANOMALIES FROM THE EMPIRICAL FINGERPRINTING MODEL AGAINST ANCILLARYOBSERVATIONS AND MODELS FOR TERRESTRIAL AND OCEAN SYSTEMS INCLUDING THE OCO-2 LEVEL 4 FLUX PRODUCT.THE APPROACH WE TAKE IS TO BALANCE THE (A) INSIGHTS THAT CAN BE GAINED FROM INCREASED SPATIAL AND TEMPORAL CHARACTERISTICS OF THE DATAWITH (B) SPATIAL SCALES THAT PERMIT HYPOTHESIS TESTING AND MODEL FALSIFICATION AS OPPOSED TO GENERATING A SPATIALLY FULLY RESOLVED CARBONBUDGET (E.G. A LEVEL 4 FLUX PRODUCT). IN PARTICULAR WE WILL FOCUS ON UNDERSTANDING THE FACTORS THAT DRIVE VARIATIONS IN XCO2 FROMYEAR-TO-YEAR. THESE METHODS WILL BE APPLIED TOWARD ANALYZING CARBON CYCLE PROPERTIES BEFORE DURING AND AFTER A LARGE EL NIO EVENT ADVANCING BOTH KEY SCIENCE QUESTIONS FOR THE OCO-2 MISSION AND FOR CARBON CYCLE SCIENCE. OUR PROPOSAL WILL ADVANCE KEY OBJECTIVESOF THE OCO-2 MISSION BY LINKING OCO-2 OBSERVATIONS TO OTHER SATELLITE DATASETS (INCLUDING MODIS SMAP GRACE AND AVHRR) INFORMULATING THE FINGERPRINTING MODEL.