Unsolicited, Improving Elevation Change Time Series from ICESat-2 Observations

WE PROPOSE TO PERFORM RESEARCH ON TWO TOPICS RELATED TO ICESAT-2. THE FIRST TOPIC IS A CONTINUATION OF THE PROJECT ENTITLED "IMPROVING ELEVATION CHANGE TIME SERIES FROM ICESAT-2 OBSERVATIONS" AND THE SECOND TOPIC IS RELATED TO THE CALVAL EFFORTS OF ICESAT-2. 1 CONTINUATION OF PROJECT FROM OCTOBER 2016 THE MAIN GOAL OF THE PROJECT ENTITLED "IMPROVING ELEVATION CHANGE TIME SERIES FROM ICESAT-2 OBSERVATIONS" IS TO VALIDATE ICESAT-2 LAND ICE ELEVATION PRODUCTS USING SERAC (SURFACE ELEVATION RECONSTRUCTION AND CHANGE DETECTION). THIS ENTAILS THE COMPUTATION OF TIME SERIES WITH SIMULATED ICESAT-2 DATA BASED ON THE ATL06 DATA PRODUCT FOLLOWED BY THE COMPARISON WITH TIME SERIES DETERMINED IN ATLLL BY BEN SMITH. WE HAVE RECENTLY EXPLORED THE POSSIBILITY OF COMPUTING SERAC TIME SERIES DIRECTLY FROM THE ATL03 DATA PRODUCT AND WE PROPOSE TO EXPAND THESE EFFORTS INTO A FORMAL STUDY. SUCH A STUDY WOULD ENTAIL A RIGOROUS ASSESSMENT OF SERAC TIME SERIES WITH THOSE DETERMINED BY BEN SMITH MADE AVAILABLE AS DATA PRODUCT ATLLL. AS SUCH THE RESEARCH WOULD PROVIDE A VALIDATION/ATION OF THE ATLLL PRODUCT. FURTHERMORE IT WOULD RESULT IN A LIST OF PROS AND CONS OF BOTH APPROACHES (SERAC AND ATLLL) IN TERMS OF SURFACE PATCH SIZE SMOOTH ICE SHEETS AND RUGGED SURFACES OF CREVASSED OUTLET GLACIERS. THE PROPOSED RESEARCH WILL ALSO INCLUDE A STUDY OF THE PHOTON POINT CLOUD AND THE DETERMINATION OF SIGNAL PHOTONS OVER MODERATELY TO HEAVILY CREVASSED SURFACES. WHAT EXACTLY ARE THE SIGNAL PHOTONS REPRESENTING OF THE TRUE SURFACE (ESTABLISHED E.G. BY OIB UNDER FLIGHTS WITH ATM)? HOW SUITABLE IS THE CHARACTERIZATION OF THE CREVASSED SURFACE BY THE SIGNAL PHOTONS FOR CHANGE DETECTION? 2 CONTRIBUTION TO CALVAL OF ICESAT-2 WE PROPOSE TO ANALYZE THE ACCURACY AND PRECISION OF ICESAT2 DATA BY COMPUTING COMBINED TIME SERIES WITH EXISTING ICESAT/LCESS/LVIS DATA AND NEW ICESAT2 DATA. WE CAN PERFORM THE STUDY AS SOON AS ICESAT- 2 DATA WILL BE RELEASED OR MADE AVAILABLE TO US. THE LOCATIONS OF TIME SERIES TO BE USED IN THE STUDY CAN BE DETERMINED AS THE ICESAT-2 GROUND-TRACKS HAVE ALREADY BEEN ESTABLISHED. THIS MEANS WE ARE ABLE TO SELECT SUITABLE LOCATIONS BEFORE THE LAUNCH OF ICESAT-2. SUITABLE EXISTING TIME SERIES ARE THOSE (I) WITH TIME EPOCHS CLOSE TO THE DATE OF THE ICESAT2 DATA (II) HAVING LONG ENOUGH TIME SPANS AND (III) BE ON FAIRLY STABLE GROUND BE WITH VERY LOW SNOW ACCUMULATION (E.G. AT HIGHER ELEVATION ON THE EAST ANTARCTIC ICE SHEET). TIME SERIES WITH THESE PROPERTIES CAN BE CALCULATED PRIOR TO THE LAUNCH OF THE ICESAT-2. THUS WITHIN A FEW WEEKS AFTER THE FIRST RELEASE OF ICESAT-2 DATA WE ARE READY TO DELIVER THE RESULTS OF THE ACCURACY/PRECISION ANALYSES BECAUSE WE CAN INCORPORATE ATL03/ATL06 DATA PRODUCTS INTO SERAC ALREADY NOW. WE WILL PERFORM THE TIME SERIES ANALYSIS BY FITTING A CURVE (EITHER POLYNOMIAL OR PSPLINES) TO THE LASER POINTS EXTRAPOLATE THE FITTED CURVE TO THE TIME OF LCESAT-2 TIME EPOCHS AND DETERMINE THE DIFFERENCE OF THE EXTRAPOLATED POINT TO THE REAL ICESAT-2 MEASUREMENT. THE ANALYSIS WILL INCLUDE A RIGOROUS ERROR ASSESSMENT OF THESE DIFFERENCES. WHILE OUR PREVIOUS PROJECT ENTITLED IMPROVING ELEVATION CHANGE TIME SERIES FROM ICESAT-2 OBSERVATIONS FOCUSED ON USING ATL06 DATA PRODUCTS THIS PROPOSAL WILL ALSO DRAW ON GLOBAL GEOLOCATED PHOTONS OF ATL03. OUR PROPOSED WORK BEGINS WITH THE PHOTON EVENTS THAT HAVE BEEN CLASSIFIED AS LIKELY BACKGROUND PHOTON EVENTS AND LIKELY SIGNAL PHOTON EVENTS WITH 3 CONFIDENCE LEVELS. WHEN EXAMINING THE POINT CLOUD OF SIGNAL PHOTON EVENTS IT BECOMES OBVIOUS THAT THE PHOTON EVENTS ARE LOCATED ABOVE AND BELOW THE TRUE SURFACE. MOREOVER THE DENSITY OF THE POINT CLOUD TENDS TO BE HIGHEST AROUND THE SURFACE WITH A CONSIDERABLE DROP-OFF AWAY FROM IT. WE WILL MODIFY ALGORITHMS FROM GENERAL POINT CLOUD PROCESSING MAINLY FROM COMPUTER VISION TO FINE TUNE THE PROCESS OF FINDING THE SURFACE. WE WILL ALSO APPLY CORRECTIONS SUCH AS THE CORRECTION TO ACCOUNT FOR THE FIRST PHOTON BIAS TO OBTAIN AN UNBIA