THE HUBBLE DEEP AND ULTRA DEEP FIELDS ARE ICONIC AND ALMOST CERTAINLY THE MOST STUDIED HST OBSERVATIONS. WFIRST WILL ALSO HAVE THE OPPORTUNITY TO REACH GALAXIES AT THE SAME EXTREMELY FAINT MAGNITUDES BUT WITH A FOV 140 TIMES THAT OF HST WF3. THIS WILL OPEN UP THE STUDY OF THE HIGHEST REDSHIFT GALAXIES TO THE TYPE OF STATISTICAL STUDIES THAT HAVE MADE MODERN COSMOLOGY A PRECISION SCIENCE. JWST WILL ALSO DRAMATICALLY INCREASE OUR ABILITY TO PROBE FAINT GALAXIES AND HIGHER REDSHIFTS. IN OUR PROPOSAL WE AIM TO MAKE THEORETICAL PREDICTIONS FOR WHAT WILL BE SEEN IN THE WFIRST AND JWST DEEP FIELDS PROPOSE AND TEST ANALYSIS METHODS AND INVESTIGATE SOURCES OF SYSTEMATIC ERROR. OUR MAIN TOOL WILL BE STATE OF THE ART COSMOLOGICAL HYDRODYNAMIC SIMULATIONS THAT ARE WELL MATCHED TO THE LARGE VOLUMES INVOLVED. THESE SIMULATIONS INCLUDE OUR FLAGSHIP RUN BLUETIDES A 0.5 GPC ON A SIDE 0.7 TRILLION PARTICLE RUN THAT IS BY FAR THE LARGEST HYDRO SIMULATION YET. WE HAVE PUBLISHED MANY EARLY RESULTS FROM PHASE I OF BLUETIDES WHICH EVOLVED TO REDSHIFT Z=8 AND ARE CURRENTLY USING A LARGE ALLOCATION ON THE NCSA BLUEWATERS SUPERCOMPUTER TO CARRY OUT PHASE II OF BLUETIDES WHICH IS EVOLVING TO LOWER REDSHIFTS (Z~5) ENABLING US TO MAKE CONTACT WITH MUCH LARGER DATASETS AND SIMULATE THE POPULATION OF GALAXIES BRIGHT AND FAINT THAT WILL BE SEEN BY WFIRST JWST AND EUCLID. BLUETIDES REPRESENTS A SIGNIFICANT ADVANCE ON OUR MASSIVEBLACK SIMULATION WHICH PREVIOUSLY REACHED THESE Z~5 REDSHIFTS. IMPROVEMENTS INCLUDE A MODEL FOR INHOMOGENEOUS REIONIZATION A PRESSURE-ENTROPY FORMULATION OF SPH AND A MODEL FOR STAR FORMATION FROM MOLECULAR HYDROGEN. MANY SIGNIFICANT CODE OPTIMIZATIONS NOW ALLOW IT TO RUN AT THE TRILLION PLUS PARTICLE LEVEL SO THAT MASS RESOLUTION IS ~20 TIMES BETTER THAN MASSIVEBLACK. THE VOLUMES ARE WELL MATCHED: THE WFIRST FOV IS 0.75 DEGREES WIDE AND BLUETIDES SPANS 3.89 DEGREES AT REDSHIFT Z=6. IN OUR PROPOSED WORK FULL POPULATION SYNTHESIS MODELLING APPLIED TO TENS OF BILLIONS OF STAR PARTICLES MEANS THAT FULLY SPATIALLY RESOLVED SPECTRA WILL BE AVAILABLE FOR EVERY GALAXY. WE WILL MAKE SIMULATED WFIRST AND JWST DEEP FIELDS AND CROSS-REFERENCE SOURCES DETECTED IN THEM WITH GALAXY PROPERTIES IN THE FULLY 3 DIMENSIONAL SIMULATION OUTPUTS. THIS SHOULD BE AN INVALUABLE RESOURCE TO TEST THE RELIABILITY OF GALAXY PROPERTIES AT REDSHIFTS Z>5 INFERRED FROM OBSERVATIONS. UNLIKE BLUETIDES PHASE I PHASE II IS ALLOWING THE SIMULATION TO ENTER INTO THE LOWER REDSHIFT REGIME WHICH HOSTS THE CURRENTLY FARTHEST KNOWN QUASARS. WE WILL BE ABLE TO VALIDATE THE SIMULATION MODELING OF BLACK HOLE ACCRETION AND GROWTH AND MAKE PREDICTIONS FOR A WHOLE NEW POPULATION OF FAINT AGN TO BE OBSERVED BY WFIRST. OUR INCLUSION OF REIONIZATION PHYSICS WILL MEAN THAT WE WILL BE ABLE TO SEARCH FOR THE SIGNATURES IN WFIRST OBSERVATIONS OF THE EFFECTS OF PATCHY REIONIZATION ON GALAXY FORMATION. FOR EXAMPLE IT HAS BEEN PREDICTED THAT UV RADIATION WILL SUPPRESS COOLING AND CAUSE LARGE SCALE FLUCTUATIONS IN GALAXY PROPERTIES (COLORS STAR FORMATION RATES). PERHAPS MOST EXCITINGLY FOR COSMOLOGY OUR INITIAL STUDIES HAVE SHOW THAT BARYON ACOUSTIC OSCILLATIONS IN CLUSTERING MAY BE DETECTABLE BY WFIRST AT REDSHIFTS Z~6-8. THIS IS DEPENDENT ON THE WFIRST HIGH LATITUDE SURVEY OBSERVING STRATEGY AND ALSO THE FINAL DESIGN AND SPECTRAL COVERAGE OF THE WFIRST GRISM BOTH OF WHICH OUR PROPOSED MODELING COULD CONCEIVABLY INFLUENCE. OUR PROPOSED THEORETICAL STUDY OF THE WIDEST PLANNED DEEP FIELDS LINKS DIRECTLY TO THE NASA OBJECTIVE OF UNDERSTANDING OF THE STRUCTURE AND EVOLUTION OF THE UNIVERSE. THE PREDICTIONS FOR WHAT WFIRST AND JWST WILL SEE IN THESE INTEGRATIONS FROM LARGE SCALE CLUSTERING DOWN TO INDIVIDUAL GALAXY PROPERTIES WILL BE BOTH STATE OF THE ART AND START FROM FIRST PRINCIPLES. THIS WILL LEAD TO THE OPPORTUNITY FOR SURPRISING NEW THEORETICAL RESULTS AND FOR RELIABLE TESTS OF OBSERVATIONAL ANALYSIS TECHNIQUES.