IT IS A STATED AIM OF NASA TO I) ANSWER THE QUESTION HOW DO MAGMATIC SYSTEMS EVOLVE UNDER WHAT CONDITIONS DO VOLCANOES ERUPT AND HOW DO ERUPTIONS AND VOLCANO HAZARDS DEVELOP? AND II) TO IMPROVE OUR ABILITY TO FORECAST THE BEHAVIOR OF EARTH'S VOLCANOES USING REMOTE SENSING DATA SUCH THAT WE MIGHT MITIGATE THEIR EFFECTS ON SOCIETY. THE PROPOSED WORK WILL SUPPORT THESE GOALS. 1. WE WILL PRODUCE A 30 YEAR GLOBAL MEASUREMENT RECORD OF VOLCANIC THERMAL UNREST BY INTEGRATING LOW AND HIGH SPATIAL RESOLUTION MEASUREMENTS. WE WILL IMPROVE THE SENSITIVITY OF THE CURRENT MODVOLC ALGORITHM TO ALLOW LOWER INTENSITY THERMAL EMISSION THE KIND OF UNREST THAT PRECEDES ERUPTION TO BE QUANTIFIED FOR EARTHS VOLCANOES. WE WILL IMPLEMENT THE NEW ALGORITHM ON THE NPP VIIRS DATA STREAM TO INCREASE THE TEMPORAL FREQUENCY WITH WHICH THIS THERMAL UNREST CAN BE MEASURED AND TO ENSURE CONTINUITY OF GLOBAL VOLCANO MONITORING FROM THE EOS INTO THE JPSS ERAS. WE WILL EXPLOIT THE HISTORICAL GLOBAL AVHRR DATA ARCHIVE EXTENDING OUR GLOBAL THERMAL EMISSION TIMES-SERIES BACK TO 1986. WE WILL QUANTIFY THERMAL EMISSION FROM ALL EARTHS ACTIVE VOLCANOES USING HIGH RESOLUTION LANDSAT (TM ETM AND TIRS) AND TERRA ASTER DATA BACK TO 1984. THIS HIGH RESOLUTION RECORD WILL ALLOW US TO QUANTIFY LOW INTENSITY (E.G. HYDROTHERMAL) VOLCANIC ACTIVITY THAT FALLS BELOW THE DETECTION LIMIT OF MODIS-VIIRS-AVHRR THUS PROVIDING A MORE COMPLETE MEASUREMENT RECORD OF THERMAL UNREST DURING PERIODS OF REPOSE. THIS WILL ALSO ALLOW US TO CALIBRATE THE LOW SPATIAL RESOLUTION (BUT HIGH TEMPORAL RESOLUTION) TIME-SERIES. 2. ALL OF THESE PRODUCTS WILL BE MADE AVAILABLE TO THE COMMUNITY AT HTTP://MODIS.HIGP.HAWAII.EDU 3. PREDICTING THE EVOLUTION OF THE INTENSITY OF AN ERUPTION ONCE IT HAS BEGUN HAS PROVEN HARDER THAN DETECTING UNREST ASSOCIATED WITH ERUPTION ONSET. WE WILL USE DEEP LEARNING TECHNIQUES TO ANALYZE THE 30 YEAR ARCHIVE WE WILL DEVELOP TO DETERMINE WHETHER ESCALATIONS DE-ESCALATIONS AND CESSATIONS OF THERMAL EMISSION AND UNREST CAN BE PREDICTED AND WITH WHAT UNCERTAINTY. 4. KEY PARAMETERS IN DYNAMIC FORECAST MODELS INCLUDE THE LOCATION COMPOSITION AND VOLATILE CONTENT OF THE MAGMA AS WELL AS MASS FLUXES OF MAGMA AND GASES. THE CHANGE IN LAVA EFFUSION RATE DURING AN ERUPTION (DECAY CONSTANT) IS RELATED TO PRESSURE RE-EQUILIBRATION IN THE CHAMBER AND THE NET RATE OF OUTFLOW AND RECHARGE. WE WILL ANALYZE A COMPREHENSIVE GLOBAL MULTI-DECADAL DATA SET OF I) LAVA EFFUSION RATES FOR ALL EFFUSIVE ERUPTIONS IN THE 30 YEAR TIME-SERIES AND II) THE DECAY CONSTANTS ASSOCIATED WITH THESE LAVA MASS FLUXES TO DETERMINE HOW INFORMATION EXTRACTED FROM SATELLITE-DERIVED EFFUSION RATE MEASUREMENTS CAN BE RELATED TO MAGMA SUPPLY CONDITIONS.