EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/12

Acronyms and Abbreviations

A ASI Agenzia Spaziale Italiana, Italy AC Adjudication Committee ASIC Application Specific Integrated Circuit AC Alternate Current ASQC American Society of Quality Control ACCESS Advanced Composition ASW Application Software Experiment on Space Station ATIC Advanced Thin Ionization Calorimeter ACRO Pierre Auger Cosmic Ray Observatory ATLID Atmospheric LIDar (PAO) ATP Authority to Proceed AD Applicable Document ATP Authorization to Proceed AD Architectural Design AU Astronomical Unit ADC Analog to Digital Converter AVS AVionicS ADD Architectural Design Document AW AirWatch ADM The Atmospheric Dynamics Mission AWG American Wire Gauge AFEE Analog Front End Electronics AWG Astronomy Working Group (ESA) AGASA Akeno Giant Air Shower Array AWS AirWatch from Space AGN Active Galactic Nuclei B AI Action Item BABY Background BYpass AI Analog Input BAD Broadcast Ancillary Data AIRES AIR-shower Extended Simulations S/W BASIS Burst arc second Imaging and Spectroscopy AIS Automated Information system BATSE Burst and Transient Spectroscopy AIT Assembly Integration and Test Experiment AIV Assembly Integration and Verification BC Bus Controller AIV/T Assembly, Integration and Verification & BCD Binary Coded Decimal Test BD Block Diagram ALS Alenia Spazio BDR Baseline Design Review AM Atmosphere Monitoring BeppoSAX X-ray astronomy Satellite AME Analog Macrocell Electronics BF Bleeder Factor AMS Alpha Magnetic Spectrometer BFD Back Focal Distance AMSD Advanced Mirror Systems Development BG BackGround ANSI American National Standards Institute BINRAD Be-INduced RADiation AO Announcement of Opportunity BJ Blue Jet AO Atomic Oxygen BKG Background AOB Any Other Business BLAST Burst Location with and Arc Second APC AstroParticle and Cosmology Laboratory in Telescope Paris, France BM Business Manager APCU Auxiliary Power Control Unit BoL Beginning of Life APD Applicable Document / Approved Document BPS Bits Per Second APD Avalanche PhotoDiode BS Blue Starter APE Absolute Pointing Error BSP Board Support Package APID Application Process Identifier BSW Basic Software APM Attached Pressurized Module BTU British Thermal Units APS Active Pixel Sensor C AR Acceptance Review C&DH Command & Data Handling ARP Address Resolution Protocol C&SR Cost & Schedule Report AS Atmosphere Sounding C&W Caution and Warning ASAP Advanced Systems Analysis Program CA&CI Configuration Audit & Configuration ASAP As Soon As Possible Inspection ASCII American National Standard Code for CAD Computer Aided Design Information Interchange CADM Configuration And Data Management ASE Airborne Support Equipment CAE Computer Aided Engineering

1 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/12

CAL Calibration CO Communication & Outreach CALS Computer Aided Logistics Support COA Certificate of Acceptance CAO Center for Applied Optics COBE Cosmic Background Explorer CARSO Center for Advanced Research in Space COC Certificate of Compliance Optics, Trieste, Italy COF Columbus Orbital Facility CASA Chicago Air Shower Array CoG Center of Gravity CASA-MIA CASA MIchigan Array Co-I Co-Investigator CBR Cosmic Background Radiation COL Columbus CC Control Center Col-CC Columbus Control Centre CCB Change Control Board COMPTEL Compton Telescope CCC Columbus Control Center COQ Certificate of Qualification CCD Charge Coupled Device CORSIKA COsmic Ray SImulations for KAscade CCHP Constant Conductance Heat Pipes COTS Commercial Off-the-Shelf CCN Contract Change Notice COU Concept of Operations and Utilization CCRF Cannister Cleaning and Rotation Facility CPU Central Processing Unit CCS Command and Control Software CR Cosmic Ray, Cosmic Radiation CCSDS Consultative Committee for Space Data CR Change Request Systems CRC Cyclic Redundancy Code CCU Central Control Unit CRES Corrosion Resistant Steel CDD Configuration Data Document CRM Continuous Risk Management CDF (CdF) College de France, France CS Conducted Susceptibility CDH Control Data Handling CSCI Computer Software Configuration Item CDR Critical Design Review CSMA/CD Carrier Sense Multiple Access with Collision CE Conducted Emissions Detection CEPF Columbus External Payload Facility CSR Concept Study Report CER Center Export Representative CTC Command and Telemetry Computer CERN European Organization for Nuclear CTE Coefficient of Thermal expansion Research CTRs Conventional Volatile Condensable CERS Cost-Estimating Relationship Materials CES Control Electronics Subsystem CUC CCSDS Unsegmented Time Code CESR Centre d’Etudes Spatiales des CVS Concurrent Version System Rayonnements, France CW Continuous Wave CESR Cornell Electron Storage Ring CWPR Continuous Wave Pseudo Random CETDP Cross-Enterprise Technology Development CXB Cosmic X-ray Background Program CXO Chandra X-Ray Observatory CG Center of Gravity CYTOP Amorphous Peffluoro Alkenylvinylether CGMS Coordination Group for Meteorological D Satellites D/MSM Directorate of Manned Spaceflight and CGRO Compton Observatory Microgravity (ESA) CI Communications Interface D/SCI Directorate of Science (ESA) CI Configuration Inspection DAK Double Aluminized Kapton CIL Critical Item List DAQ Data Acquisition CLA Coupled Load Analysis DAQ Data Quality Assurance CM Center of Mass DaSS Data Service System CM Configure Management DB DataBase CMB Cosmic Microwave Background DBMS DataBase Management System CMD Command DC Direct Current CN Cosmic Neutrino DCA Digital to Analog Converter CNB Cosmic Neutrino Background DCE Digital Control Electronics CNES Centre National d’Etudes Spatiales, France DCL Declared Components List CNR Consiglio Nazionale delle Ricerche, Italy DD Design Document CNRS Centre Nationale de la Recherche DD Detailed Design Scientifique, France DD&VP Design, development & Verification Plan

2 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/12

DDP Delivery Data Package EIDP End Item Data Package DDR Detailed Design review EIRS EUSO Instrument Requirement DFEE Digital Front End Electronics Specification DIMES Diffuse Microwave Emission Survey EIS End Item Specification DM Dark Matter ELID ELectrolytic Inprocess Dressing DMS Data Management System ELS Emergency Landing Site D-MSM Directorate of Manned Spaceflight and EMC Electro Magnetic Compatibility Microgravity (ESA) EMI Electro Magnetic Interference DOC Document EP EXPRESS Pallet DoD Department of Defense EPD Entrance Pupil Diameter DoE Department of Energy EPDC Electrical Power Distribution & Conditioning DP Data Package EPF External Payload Facility (COF) DPA Destructive Parts Analysis EPO Education and Public Outreach DPM Data Processing and Monitoring EPOWG Education & Public Outreach Working DR Design Requirement Group DR Data Requirement EPP EXPRESS Pallet Payloads DRD Doument Requirement Description EPR Engineering Peer Reviews DRL Document Requirement List EPS Electrical Power Supply DRM Design Reference Mission ER Engineering Requirement DRR Document Release Record ESA European Space Agency DS Data Sheet ESAF EUSO Simulation and Analysis Framework D-SCI Directorate of Science (ESA) ESCS EUSO Standard Coordinate System DSP Digital Signal Processor ESD Electro Static Discharge DSRI Danish Space Research Institute, Denmark ESF Engineering Support Function DSU Data Storage Unit ESF European Science Foundation DV Direct Vision ESOC European Space Operations centre DWG Drawing ESR Executive Summary Report E ESRD EUSO Science Requirements Document E&W Emergency and Warning ESSC EUSO Science Steering Committee e.g. exempli gratia (for example) ESTEC European Space Technology Center (ESA) EAL Earth Atmospheric Lidar ETA Energetic Transient Array EAS Extended Air Shower ETSE Electrical Test Support Equipment ECIS Export Clearance Information Sheet EUSO Extreme Universe Space Observatory ECP Engineering Change Proposal EUT Equipment Under Test ECR Engineering Change Request EUTAS Enhanced Universal Trunnion Attachment ECSS European Co-operation for Space System Standardiz. EUV Extreme Ultra-Violet ECT Emulsion Chamber Technology EVA Extra Vehicle Activity EDAC Error Detection and Correction EVR ExtraVehicular Robotics EDBIF External DataBase InterFace EXIST Energetic X-ray Imaging Survey Telescope EE Extreme Energy EXP EXpress Pallet EECR Extreme Energy Cosmic Ray EXPA EXpress Pallet Adaptor EECR/ν Extreme Energy Cosmic Rays and EXPRESS Expedite the Processing of Experiments to Neutrinos Space Station EEE Electrical, Electronic, Electromechanical EXPS EXpress Pallet System EEPROM Electrically Erasable Programmable Read- F Only Memory FADC Fast Analog-to-Digital Converter EF Exposed Facility FAQ Frequently Asked Question EFL Effective Focal Length FAR Federal Acquisition Regulations EGRET Energetic Gamma-Ray Experiment FAR Final Acceptance Review Telescope FDI Failure Detection and Isolation EGSE Electrical Ground Support Equipment FDI Full Disk Image EI Engineering Integration FDIR Failure Detection, Isolation and Recovery

3 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/12

FE Front End GRB Gamma Ray Burst FEA Finite Element Analysis GRIS Gamma Ray Imaging spectrometer FEE Front End Electronics GS Ground Segment FEM Finite Element Model GSE Ground Support Equipment FF Free Flyer GSEA Ground Support Equipment & Activity FGSE Fluidic Ground Support Equipment GSFC Goddard Space Flight Center (NASA) FIRE Fluorescence Image Read-out Electronics GSOC German Space Operations Centre (ESA) FIRST Far Infrared-Submillimeter Space Telescope GTU Gate Time Unit FITS Flexible Image Transport System GUI Graphical User Interface FLAP Flight Automated Procedure GUT Grand Unified Theory FM Flight Model GZK Greisen - Zatsepin – Kuzmin FMEA Failure Modes and Effect Analysis H FMECA Failure Modes, Effects and Criticality H&S Healt and Status Analysis H/W (HW) Hardware FMS File Management System HDBK Handbook FNAL Fermi National Accelerator Laboratory HECR High Energy Cosmic Ray FOCUS Frontier Optical Coherent Ultrafast Center HETE High Energy Transient Explorer FOV (FoV) Field of View HFE Human Factors Engineering FPAG Fundamental Physics Advisory Group (ESA) HIMS Hot Interstellar Medium spectrometer FPGA Field Programmable Gate Array HiRes High Resolution Fly’s Eye Cosmic Ray FRAM Flight Releasable Attachment Mechanism Observatory FRC Facility Responsible Center (ESA) HK Housekeeping FRGF Flight Releasable Grapple Fixture HMI Human-Machine Interface FS Filter System HMU Harness Manufacturing Unit FS Flight Segment HOSC Huntsville Operations Support Center FS Focal Surface HOU Hands-On Universe FSA Focal Surface Assembly HPD Hybrid Photo Detector FSD Focal Surface Detector HPD Hybrid Photo Diode FTA Fault Tree Analysis HQ HeadQuarters FTP File Transfer Protocol HRDL High Rate Data Link FTSE Fluidic Test Support Equipment HRFM High Rate Frame Multiplexer FUSE Far Ultraviolet Spectroscopic Explorer HRV High Resolution Visible G HST GBM GLAST Burst Monitor HTML Hyper Text Markup Language GEVS General Environment and Verification HTTP Hyper Text Transport Protocol Specification HTTPS Secure Hyper Text Transport Protocol GF Geometrical Factor HTXS High Throughput X-ray Spectroscopy GF Grapple Fixture mission GIDEP Government-Industry Data Exchange HUBE Hopkins Ultraviolet Background Explorer Program HV High Voltage GIL Greisen - Il’ina – Linsley I GIS Geographic Information System I&T Integration and Test GLAS Geoscience Laser Altimeter System I/F Interface GLAST Gamma-Ray Large-Area Space Telescope I/O Input/Output GMM Geometrical Mathematical Models IAP Institut d’Astrophysique de Paris, France GMT Greenwich Mean Time IASF Istituto di Astrofisica Spaziale e Fisica GN NASA Ground Network Cosmica – CNR, Italy GN&C Guidance, Navigarìtion & Control IASF-Palermo IASF-CNR at Palermo, Italy (formerly IFCAI) GOES Geostationary Operational Environmental IC Instrument Controller Satellite ICA Interface Control Annex GP-B Gravity Probe-B ICC Integrated Cargo Carrier GPS Global Positioning System ICD Interface Control Document GRAM Global Reference Atmosphere Model ICF Instrument Configuration Files

4 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/12

ICMP Internet Control Message Protocol IS Interface System ICR Instrument Control and Readout ISAC Istituto di Scienze dell’Atmosfera e del Clima ICRC International Cosmic Rays Conference – CNR, Italy ICRR Institute for Cosmic Ray Research (Japan) ISAC-BO ISAC-CNR at Bologna, Italy ICS Integrated Cradle Structure ISAS Institute of Space and Astronautical Science, ID Identifier Japan IDD Instrument Definition Document ISCCP International Satellite Cloud Climatology IDD Interface Definition Document Project IDL Interactive Data Language ISM InterStellar medium IEC International Electro-Technical Commission ISN Institut des Sciences Nucléaires, France IEEE Institute of Electrical and Electronic ISNG Institut des Sciences Nucléaires de Engineers Grenoble, France IFCAI obsolete: use IASF-Palermo ISO International Standards Organization IFM In-Flight Maintenance ISPR International Standard Payload Rack IFMS Integrated financial Management System ISS International Space Station IFR Instrument Final Review IT Information Technology IFS Instrument Flight Segment IT Integration & Testing IGES Initial Graphics Exchange Specification ITAR International Traffic in Arms regulation IICD Instrument Interface Control Document ITS Integrated Truss Segment IIDD Instrument Interface Definition Document ITT Invitation To Tender IIRP Integrated Independent Review Plan IV&V Independent Verification and Validation IIRT Integrated Independent Review Team IVA Intra-Vehicular Activity IIT Image Intensifier IVP Instrument Verification Plan IM Instrument Manager IWG Instrument Working Group IN2P3 Institute National de Physique Nucléaire et J de Physique des Particlules, France JEM Japanese Experiment Module (on ISS) INFN Istituto Nazionale di Fisica Nucleare, Italy JPL Jet Propulsion Laboratories INIST Italian National Institute of Standards and JSC Johnson Space Center (NASA) Technology K INOA Istituto Nazionale di Ottica Applicata, KAL Keep Alive Line Firenze, Italy KEK High Energy Accelerator Research INS Inertial Sensor Organization, Japan INSU Institut National des Sciences de l’Univers, KHB Kennedy Handbook Paris, France KO Kick Off INTEGRAL International Gamma Ray Astrophysics KOM Kick Off Meeting Laboratory KSC Kennedy Space Center IP Internet Protocol KYA Keel Yoke Assembly IPC Industrial Policy Committee L IPM Input Protection Module LAN Local Area Network IPN Institut de Physique Nucléaire, Orsay, LAPP Laboratoire d'Annecy-le-Vieux de Physique France des Particules, France IPT Integrated Product Team LAPTH Laboratoire d'Annecy-le-Vieux de Physique IR Imaging Rate Théorique, France IR InfraRed LBL Lawrence Berkeley Laboratory, USA IR Instrument Requirement LCS Light Collection System IRB Instrument Reference Book LD Lens Diameter IRD Interface Requirements Document LEO Low Earth Orbit IRIG Inter-Range Instrumentation Group LET Linear Energy Transfer IRS Instrument Requirements Specification LHC Large Collider (at CERN) IS Image Space LI Lorentz Invariance IS Instrument Scientist LIDAR LIght Detection And Ranging IS Instrument Structure LIP Laboratório de Instrumentação e Física IS Instrument System Experimental de Partículas, Portugal

5 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/12

LISA Laser Interferometer Space Antenna MGSE Mechanical Ground Support Equipment LLL Least Light Loss MI Momentum of Inertia LLNL Lawrence Livermore National Labs MIDEX Medium-Class Explorer LNGS Laboratori Nazionali del Gran Sasso, Italy MIL Military LOS (LoS) Line of Sight MIME Multipurpose Internet Mail Extensions LOS (LoS) Loss of Signal MIP Mission Integration Plan LPNHE Laboratoire de Physique Nucléaire et de MIR Medium Infrared Hautes Energies, France MIS Miscellanea LPNHEP LPNHE in Paris, France MIUL Material Identification Usage List LRDL Low Rate data Link MLI MultiLayer Insulation LRT Low Rate Telemetry MM Mathematical Models LSA Logistics Support Analysis MM&OD Micro Meteoroids & Orbital Debris LSAR Logistics Support Analysis Record MMI Man Machine Interface LSb Less Signficant bit MMU Mass Memory Unit LSB Less Significant Byte MO Mission of Opportunity LSE Lead Systems Engineer MoA Memorandum of Agreement LSP Less Significant Part MOC Mission Operation Centre LSRD Launch Site Requirements Document MOCCA MonteCarlo Cascade LSSP Launch Site Support Plan MOM Minutes of Meeting LSW Less Significant Word MoS Margin of Safety LVLH Local Vertical Local Horizontal MoU Memorandum of Understanding LVPS Low Voltage Power Supply MP Multi-Phase LVSU Low Voltage Switching Unit MPA Max Planck Institut für Astrophysik, M Germany M&P Material & Processes MPD Marshall Policy Directive M/OD Meteoroid and Orbital Debris MPE Max Planck Institut für Extraterrestrische MADB Mission Archive and DataBase Physik, Germany MAMA MultiAnode Microchannel Array MPG Marshall Procedures and Guidelines photomultiplier MPG Multiple Point Ground MAP Microwave Anisotropy Probe MPIFR Max Planck Institut für Radioastronomie, MAP Mission Activity Planning Germany MAPMT Multi Anode Photo Multiplier Tube MPLM Multipurpose Pressurized Logistic Module MAPTIS Materials and Process Technology MRB Material Review Board Information System MRD Mission Requirements Document MARS MultiAnode Read-out System MRDL Medium Rate Data Link MASCS Monitoring, Alignment and Self-Calibration MRSA Material Readiness Support Activity System MS Manned Spaceflight MASS Maximum-energy Auger Air-Shower Satellite MS Military Specification MBS Mobile Base System MSb Most Significant bit MBS Mobile remote servicer Base System MSC Mobile Servicing Center MBS Most Significant Byte MSD Mass Storage Device MC MonteCarlo MSFC Marshall Space Flight Center (NASA) MCAS Mobile Common Attach System MSFC SD Marshall Space Flight Center Science MCC Mission Control Center Directorate (NASA) MCC-H Mission Control Center - Houston MSIS Mass Spectrometer Incoherent Scatter MCD Molecular Column Density MSL Mean sea Level MCI Mass, Centering and Inertia MSM Manned Spaceflight and Microgravity (ESA) MCP Micro Channel Plate MSP Most Significant Part MDD Mission Definition Document MSS MDM Multiplexer DeMultiplexer MSW Most Significant Word MDPS Meteoroids and Debris Protection System MT Mobile Transporter MDR Molecular Deposition Rate MTF Modulation Transfer Function MERAT Mean Effective RAdiation Temperature MTL Master Timeline

6 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/12

MTM Management and Technical Meeting OUST On-board Unit System Trigger MTR MidTerm Review OWL Orbiting Wide-Angle Light Collector MTSE Mechanical Test Support Equipment P N P/L Payload N/A Not Applicable P/N Part Number NAO National Astronomical Observatory, Japan P3 Port Truss Segment three NAS National Aerospace Standard PA Product Assurance NASA National Aeronautics and Space PA&S (PA/S) Product Assurance and Safety Administration, USA PAH Payload Accommodation Handbook NASDA National Space Development Agency, PALS Program Automated library System Japan PAO Pierre Auger Observatory (as ACRO) NATC NASA Threaded Coupling PAS Payload Attach System NDE Non-Destructive Evaluation PBMS Manned Space Programme Board NE Network Element PCB Printed Circuit Board NERSC National Energy Research Supercomputer PCC Laboratoire de Physique Corpuscolaire et Center, USA Cosmologie, CdF, Paris, France NFS Network File System PCS Portable Computer System NGST Next Generation Space Telescope PCU Power Control Unit NGXO Next Generation X-ray Observatory PD PhotoDetector NIR Near InfraRed PD Project Directive NIST National Institute of Standard Technology PDB Power Distribution Box NPD NASA Policy Directive PDGF Power Data Grapple Fixture NPG NASA Procedures and Guidelines PDL Payload Data Library NSF National Science Foundation, USA PDM Photo Detector Module NSSDC National Space Science Data Center, USA PDR Preliminary Design Review NSSTC National Space Science and Technology PDU Power Distribution Unit Center, USA PEP Payload Executive Processor NSTS National Space Transportation System (US PFM Pro-Flight Model ) PFR Payload Final Review NUV Near Ultra Violet PFR Portable Foot Restraint NVR Non-Volatile Residue PHS Process Health and Status O PHS&T Packaging, Handling Storage and OACT Osservatorio Astronomico di Catania, Italy Transportability OAFI Osservatorio Astronomico di Arcetri, PI Principal Investigator Firenze, Italy PIA Payload Integration Agreement OBC On Board Computer PICD Payload Interface Control Document OBDH On Board Data Handling PIP Payload Integration Plan OBT On Board Time PIRD Payload Interface Requirements Document OD Orbital Debris PIRN Preliminary Interface Revision Notice OdP Observatoire de Paris, France PLANCK Planck Cosmic Background Surveyor OF Optical Filter PLCU Payload Control Unit OLA Off Line Analysis PLD Payload OM Operations & Maintenance PM Programme Manager OM Optics Module PM Meeting OMS Orbiter Maneuvering System PM Project Manager OO Object Oriented PMC Project Management Council OPF Orbiter Processing Facility PMCS Project Management Control System ORU Orbit Replaceable Unit PMMA Poly Methyl Metha Acrylate OS Optical System PMR Payload Mid-term Review OSE On-orbit Support Equipment PMT Photo Multiplier Tube OSS Office of Space Science (NASA) PMT Project Management Team OTCM ORU/Tool Change-out Mechanism PMTR Payload Mid Term Review OTD Optical Transient Detector PO Project Office

7 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 8/12

POA Payload Orbital Replacement Unit RMS Remote Manipulator System Accommodation RMS Root Mean Squared POC Payload Operation Center RO&C Read-Out & Control POCC Payload Operation Control Center ROEU Remote Operated Umbilical POIF Payload Operations Integration Function ROM Rough Order of Magnitude PPAR Payload Preliminary Architecture Review ROOT An Object-Oriented Data Analysis PPL Preferred Parts List Framework developed at CERN PPRR Payload Preliminary Requirements Review ROS Real Object Space PRA Probabilistic Risk Assessment ROSAT Roentgen Satellite PRD Procedure Document RP Report PRF Pulse Repetition Frequency RPC Remote Power Controllers PRLA Payload Retention Latch Assy RPCM Remote Power Control Module PRN Problem Report Notification RQ Requirement PRN Pseudo-Random Noise RS Recommended Standard (e.g. RS-232) ProSEDS Propulsive Small Expandable Deployer RS Radiated Susceptibility System RSS Root Sum Squared PS Power System RT Remote Terminal PS Project Scientist RTD Resistant Temperature Device PSA Part Stress Analysis RTF Rich Text interchange Format PSB Power Supply Board RTLF Rack Test Level Facility PSD Power Spectral Density RTM Real-Time Monitoring PSF Point Spread Function RUL Remote User Location PSI Pounds per Square Inch RVC Requirements, Verification and Compliance PSRP Payload Safety Review Panel RVLF Random Vibration Load Factor PTI Post Transportation Inspection RX Receiver PUB Publication of papers (journal, conf. Proc.) RXTE Rossi X-ray Timing Explorer P-V peak to valley S Q S&MA Safety and Mission Assurance QA Quality Assurance S/C Spacecraft QCD Quantum Chromo Dynamics S/N Serial Number QL Quick Look S/N Signal-to-Noise QM Qualification Model S/S Sub System QMS Quality Management System S/W (SW) Software QR Qualification Review S3 Starboard Truss segment Three R SA Service d’Aéronomie, France R&D Research and Development SAA South Atlantic Anomaly RAM Random Access Memory SAC Scientific Analysis Component RCB Read-out & Control Board SAE Society of Automotive Engineers RD Requirements Document SAM Stratospheric Aerosol Measurement RDT Remote Data Terminal SAO Smithsonian Astrophysical Observatory RE Radiated Emissions SAS System Application Software REP Report SAX X-ray astronomy Satellite BeppoSAX REQ Requirement SBC Single Board Computer RF Radio Frequency SC Science RFA Request For Approval SC EUSO Science Consortium RFQ Request for Quotation SC Shipping Container RHIC Relativistic Heavy-Ion Collider SCC Stress Corrosion Cracking RI Rockwell International SCN Serial Configuration Number RIF Relative Illumination Fall out SCSI Small Computer System Interface RIKEN Institute of Physical and Chemical Research, SDA Scientific Data Analysis Japan SDC Science Data Center RM Redundancy Management SDE Software Development Environment RMA Restraints and Mobility Aids SDN Starboard Deck Nadir direction

8 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 9/12

SDR System Design Review / Report SRP Standard Repair Procedure SDX Starboard Deck X-axis SRR System Requirement Review SE Support Equipment SRU Shop Replaceable Unit SE System Electronics SS Study Scientist SE&I System Engineering & Integration SS Support Structure SEA Support Experimental Activity SS System Specification SEE Single Event Effects SSA Scientific Support Activity SEGR Single Event Gate Rupture SSAC Space Science Advisory Committee (ESA) SEL Single Event Latchup SSC Science Steering Committee SEU Single Event Upset SSCC Space Station Control Center SEU Structure and Evolution of the Universe SSFP Space Station Freedom Program (NASA Program) SSL Space Science Laboratory SIDD Standard Interface Definition Document SSP Space Station Program SIM Simulation SSPF Space Station Processing Facility SIRTF Space InfraRed Telescope Facility SSPPF Space Station Payload Processing Facility SLAC Stanford Linear Accelerator Center SSQ Space Station Qualified SLAST Shower initiated Light Attenuated to the SSRMS Space Station Remote Manipulator System Space Telescope SST Science Study Team (ESA) SM Standard Model ST System, structure and Thermal control SM Study Manager STA Structural Test Article SMD Standard Military Drawing STBD StairBoard SMP Super Massive Particle STC Structure & Thermal Control SMT Study Manager Team (ESA) STCS Structure & Thermal Control System SN Serial Number STD Standard SNR Signal-to-Noise Ratio STS Shuttle Transport System SNR SuperNovae Remnant STS Space Transportation System SO Science Objectives SWM SoftWare Maintenance SOC Science Operation Center SWOP S/W On-board Command SODC Science Operation and Data Center SWT Science Working Team SOFIA Stratospheric Observatory for Infrared SXI Solar X-ray Imager Astronomy T SOMTC Space Optics Manufacturing Technology T&OBDH Trigger and On Board Data Handling Center, USA T/R Transmit/Receive SOT Science and Outreach Team TA Telescope Array Project SoW Statement of Work TAB Trunnion Adapter Beam SOWG Science Objectives Working Group TAEM Terminal Area Energy Management SP Séminaire de Prospective TAL Trans Atlantic Landing SP Specification TAXI Transparent Asynchronous SPAP Safety and Product Assurance Plan Transmitter/Receiver Interface SPC Science Programme Committee (ESA) TBC To Be Confirmed SPDM Special Purpose Dexterous Manipulator TBD To Be Defined / Determined SPE Specification TBR To Be Resolved / Reviewed SPEC Specification TBS To Be Specified / Supplied SPL Sound Pressure Level TBV To Be Verified SPLC Standard PayLoad Computer TBW To Be Written SPOE Standard Payload Outfitting Equipment TC TeleCommand SR Scientific Requirement TCG TeleCommand Generation SR Software Requirements TCP Transmission Control Protocol SR System Requirement TCS Thermal Control System SRD Scientific Requirements Document TCU Thermal Control Unit SRD System Requirements Document TCU Trigger Control Unit SRG Spectrum-Roentgen-Gamma TD Topological Defect SRMS Shuttle Remote Manipulator System TDRS Tracking Data Relay Satellite

9 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 10/12

TDRSS Tracking & Data Relay Satellite System ULTRA UV Light Transmission and Reflection in TEA Torque Equilibrium Attitude Atmosphere TEB Tender Evaluation Board UMA Umbilical Mating Adapter TEO Trigger Electronics and On-board read-out UNISIM Shower Simulation Package TID Total Ionizing Dose Univ. University TIDE Thermal-Ion Detector Experiment UNM University of New Mexico, USA TIM Technical Interchange Meeting UR User Requirements TIS Total Integrated Scatter URD User Requirement Document TLC Telecommand URL Uniform Resource Locator TLM Telemetry USOS United States Orbital Segment TM Telemetry UT Universal Time TM Technical Memo UT University of Texas, USA TMH TeleMetry Handling UTA University of Texas in Austin, USA TML Total Mass Loss UTC Universal Time Constant TMM Thermal Mathematical Model UV Ultra Violet TMT Technical Management Team UV-PMMA-00 grade 000 UV transmitting PMMA TN Technical Note UVI Ultra Violet Imager ToC Table of Contents V TP Technical Paper V&V Verification and Validation TPS Thermal Protection System VAB Vehicle Assembly Building TPX Polymethyl-pentene VADR Vehicle Analysis Data Recording TR TRansfer to operation VC Virtual Channel TRASYS Thermal Radiation Analyzer System VCD Verification Control Document TREK Telescience REsources Kit VCHP Variable Conductance Heat Pipes TRL Technical Readiness Level VD Voltage Divider TRO Trigger and Read-Out VG Veiling Glare T-S Thermal Structure VG ViewGraph TSE Test Support Equipment VHE Very High Energy TT&C Telemetry, Tracking, and Commanding VLSI Very Large Scale Integration TVF Test & Validation Facility VMDB Vehicle Master Data Base TVIS Test and Verification Information System VME Versa Module Europa TVS Thermal Vacuum Stability VP Verification Plan TW Twisted VRCD Verification Requirements Compliance TWDS Twisted Double Shielded Document TWS Twisted Shielded VRSD Verification Requirements and Specification TX Transmitter Document U VU Vanderbilt University, USA UAD Unique ancillary Data W UAH University of Alabama in Huntsville, USA WBS Work Breakdown Structure UART Universal Asynchronous WCA Worst Case Analysis Receiver/Transmitter WFF Wallops Flight Facility UAS User Application Software WG Working Group UCB University of California in Berkeley, USA WIC Wideband Imaging Camera UCCAS Unpressurized Cargo Carrier Attach System WIRE Wide-field InfraRed Explorer UCLA University of California in Los Angeles, USA WP Work Package UCP Unpressurized Cargo Pallet WPD Work Package Description UDP User Datagram Protocol WR Wavelength Range UFP Uncertainty Flight Prediction wrt (w.r.t.) with respect to UHE Ultra High Energy WSGT White Sand Ground Terminal UHECR Ultra High Energy Cosmic Ray WWW World Wide Web UHEN Ultra High Energy Neutrino X UHF Ultra High Frequancy XEUS The X-ray Evolving Universe Spectroscopy UIF User InterFace Mission

10 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 11/12

XMM X-ray Multi-Mirror spectroscopy mission Y Z ZEONEX Amorphous Cycle-Olefin ZHR Zenith Hourly Rate

11 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: - Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 12/12

12 EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/35

1. THE EUSO SCIENTIFIC CASE ...... 3 1.1 INTRODUCTION ...... 4 1.2 THE CURRENT OBSERVATIONAL AND THEORETICAL SCENARIO ...... 6 1.2.1 Past Investigations ...... 6 1.2.1.1 The Spectrum of EECRs ...... 6 1.2.1.2 Sky Distribution...... 6 1.2.2 Present and Planned Ground-Based Observations...... 9 1.2.3 Future Space-Based Observations...... 9 1.2.4 The Current Theoretical scenario...... 9 1.3 ORIGIN ...... 11 1.3.1 Acceleration Process...... 11 1.3.1.1 Diffusion acceleration at Newtonian Shocks ...... 11 1.3.1.2 Unipolar induction...... 12 1.3.1.3 Non-linear particle-wave interaction...... 13 1.3.2 Astrophysical Sources...... 13 1.3.2.1 Active Galactic Nuclei and Dead Quasars...... 15 1.3.2.2 Neutron Stars ...... 15 1.3.2.3 Gamma Ray Bursts...... 16 1.3.3 Physics beyond the Standard Model...... 17 1.3.3.1 New Primary Particle and interactions ...... 17 1.3.3.2 Top-down Scenarios...... 17 1.3.3.3 Top-down Fluxes...... 18 1.4 PROPAGATION OF EECR ...... 19 1.4.1 The GZK Effect...... 19 1.4.2 The Propagation of Heavy Nuclei...... 21 1.4.3 Photon Propagation...... 21 1.4.4 Deflection in cosmic magnetic fields...... 22 1.5 EE NEUTRINO PHYSICS AND ASTROPHYSICS...... 24 1.5.1 Neutrinos Cross Sections ...... 24 1.5.2 Super GZK neutrinos ...... 26 1.5.2.1 Sources of SGZK neutrinos...... 27 1.5.2.2 Upper limits on diffuse neutrino fluxes ...... 28 1.5.3 Upper limits on diffuse neutrino fluxes...... 29 1.5.3.1 AGN and other accelerator sources...... 29 1.5.3.2 Reionization bright phase ...... 29 1.5.3.3 Necklaces ...... 30 1.5.3.4 Superheavy dark matter (SHDM)...... 30 1.5.3.5 Mirror matter ...... 31 1.5.3.6 GZK neutrinos ...... 31 1.5.3.7 τ neutrinos ...... 32 1.6 ATMOSPHERE SCIENCE...... 33 1.6.1 METEORS...... 33

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/35

1.6.2 Transient Luminous Phenomena...... 34 1.6.2.1 Events in Lower Atmosphere: Lightning ...... 34 1.6.2.2 Events in upper atmosphere : Elves, Sprites and Blue Jets...... 34 1.6.3 Atmosphere as a Physical System ...... 35

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/35

1. The EUSO Scientific Case The “Extreme Universe Space Observatory – EUSO” is an international, multi-agency mission, lead by ESA, aimed at investigating the nature and origin of Extreme Energy Cosmic Rays (EECRs, charged particles, photons, neutrinos, with E > 5x1019 eV, the conventional GZK energy) and at opening the channel of High Energy Neutrino Astronomy.

EUSO will pioneer measurements of EECR-induced Extensive Air Showers (EASs) from space, making accurate measurements of the primary energy, arrival direction and composition of EECRs, using a target volume far greater than is possible from the ground. Such data will shed light on the origin of EECRs, on the sources that are producing them, on the propagation environment from the source to the Earth, and, possibly, on the particle physics mechanisms at energies well beyond the ones achievable in man-made laboratories.

Two are the main goals of the proposed Mission: • Investigation of the highest energy processes present and accessible in the Universe through the detection and analysis of the Extreme Energy Component of the Cosmic Radiation (EECRs with E > 5x1019 eV, the conventional GZK energy). • Open the Channel of High Energy Neutrino Astronomy to probe the boundaries of the Extreme Universe and to investigate the nature and distribution of the EECR sources. A fall-out is represented by the systematic surveillance of Atmospheric Phenomena ( the Atmosphere as a Physical System, Electrical Discharges, Meteors , ... ).

The fundamental obstacle to understanding the nature and origin of EECRs is the lack of observational data. The observational goals of EUSO are therefore as follows: • Collect a large sample of EECRs; measure their energies, directions, and composition; • Make astronomical observations at extreme energies using charged particles, neutrinos, and/or gamma rays.

This will address the following fundamental observational questions: • How does the cosmic-ray spectrum continue beyond the existing data? Is there a maximum energy (Emax)? Are there particular point sources responsible for the spatially correlated event pairs already observed?2, 3 Is there anisotropy that indicates source regions? • Are the EECRs protons, nuclei, photons, neutrinos, or exotic particles? • What is the neutrino flux at extreme energies? • Are there point sources of neutrinos? Are active galactic nuclei (AGN) or gamma-ray bursts (GRBs) copious sources of neutrinos? Are there other sources? • What is the gamma-ray flux at extreme energies? Does it exhibit any predicted quantum gravity effects?

Interpreting the EUSO data will lead us to consider the following theoretical questions: • What processes and what astronomical objects can generate radiation at these extreme energies? • Are EECRs the result of high-energy neutrinos arriving from distant sources?

• Must we postulate topological defects4 and/or supermassive relic particles5 to explain the observations? • Is special relativity valid at extreme energies? • Are the EECRs a window to new physics at the TeV–PeV mass energy scale?

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/35

In this chapter we summarize the historical background and the present observational status of EECRs and EE neutrinos and we discuss scientific aspects related to origin and nature of EECRs and neutrinos, to their sources and propagation, to possible new physical mechanisms.

1.1 Introduction Historically cosmic rays have always been at the intersection of astrophysics with particle physics. This is still and especially true in current days where experimenters routinely observe atmospheric showers from particles whose energies reach macroscopic values up to about 50 Joules. This dwarfs energies achieved in the laboratory by about eight orders of magnitude in the detector frame and three orders of magnitude in the centre of mass. Their macroscopic energies likely links their origin to the most energetic processes in the Universe and possibly testify physics not yet discovered. Explanations range from conventional shock acceleration to particle physics beyond the Standard Model and processes taking place at the earliest moments of our Universe. While motivation for some of the more exotic scenarios may have diminished by newest data, conventional shock acceleration scenarios remain to be challenged by the apparent isotropy of cosmic ray arrival directions which may not be easy to reconcile with a highly structured and magnetized Universe. High energy cosmic ray (CR) particles are shielded by Earth's atmosphere and reveal their existence on the ground only by indirect effects such as ionization and showers of secondary charged particles covering areas up to many km2 for the highest energy particles. In fact, in 1912 Victor Hess discovered CRs by measuring ionization from a balloon (Hess, 1912) and in 1938 Pierre Auger proved the existence of extensive air showers (EAS) caused by primary particles with energies above 1015 eV by simultaneously observing the arrival of secondary particles in Geiger counters many meters apart (Auger et al., 1938, Auger & Maze, 1938).

Figure 1.1-1 Compilation of measurements of the Figure 1.1-2 UHECR flux as measured by the HiRes-I and differential energy spectrum of cosmic rays. Above HiRes-II detectors (Abu-Zayyad et al. 2002), and the AGASA 100 MeV the CR spectrum exhibits little structure experiment (Takeda et al., 2002}. Also shown is a fit to the data and is approximated by broken power laws ∝ E−γ: At of a superposition of a galactic and an extragalactic source the energy E ∼ 4 x 1015 eV called the ``knee'', the component. This figure is from the second reference in flux of particles per area, time, solid angle, and Ref.(Abu-Zayyad et al. 2002). energy steepens from a power law index γ ∼2.7 to one of index ∼3.0 (shown for comparison).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/35

After almost 90 years of CR research, their origin is still an open question, with a degree of uncertainty increasing with energy (Berezinsky et al., 1990, Geisser, 1998). Only below 100 MeV kinetic energy, where the wind shields protons coming from outside the solar system, the sun must give rise to the observed proton flux. Above that energy the CR spectrum exhibits little structure and is approximated by broken power laws ∝ E−γ: At the energy E∼ 4 x 1015 eV, called the ``knee'', the flux of particles per area, time, solid angle, and energy steepens from a power law index γ ∼2.7 to one of index ∼3.0. The bulk of the CRs up to at least that energy is believed to originate within the Milky Way Galaxy. Above the so called ``ankle'' at E = 5 x 1018 eV, the spectrum flattens again to a power law of index γ ∼2.7. This latter feature is often interpreted as a cross over from a steeper Galactic component, which above the ankle cannot be confined by the Galactic magnetic field, to a harder component of extragalactic origin. Over the last few years, several giant air showers have been detected both in ground detectors measuring the secondary shower particles directly in water tanks or scintillation counters (Takeda et al., 1998) and in fluorescence telescopes detecting the nitrogen emission induced by the shower (Bird et al., 1993, Abu Zayyad, 2002). This confirms the arrival of CRs with energies up to a few hundred EeV (1 EeV = 1018 eV), corresponding to about 50 Joules. The existence of such ultra-high energy cosmic rays (UHECRs) poses a serious challenge for conventional theories of CR origin based on acceleration of charged particles in powerful astrophysical objects. In addition, nucleons above ~70 EeV lose energy drastically due to photo-pion production on the cosmic microwave background (CMB) -- the Greisen-Zatsepin-Kuzmin (GZK, Greisen, 1966, Zatsepin & Kuzmin, 1966) effect which limits the distance to possible sources to less than ~100 Mpc (see section XXX). Heavy nuclei at these energies are photo-disintegrated in the CMB within a few Mpc (Stecker & Salomon, 1999 and references therein). Unless the sources are strongly clustered in our local cosmic environment, a cut-off in the spectrum above ~70 EeV is therefore expected (see, e.g., Blanton et al., 2001 and references therein). However currently there seems to be a disagreement between the AGASA ground array which detected about 8 events above 1020 eV, as opposed to about 2 expected from a cut-off, and the hires fluorescence detector whose measurements seems to be consistent with the presence of a cut-off. The resolution of this inconsistency may have to await next generation experiments such the Pierre Auger Observatory (Cronin, 1992) or EUSO. If the source spectrum continues beyond a few 1020 eV, the GZK cut-off is in fact not a cut-off in the strict sense because sources closer than the energy loss distance still contribute to the flux beyond the GZK "cut-off". Instead, a "recovery" of the spectrum is expected beyond the GZK threshold, at a flux level smaller than the flux right below the GZK threshold by a factor reflecting the fraction of source flux injected along the line of sight within the GZK distance. A measurement of this recovery therefore contains important information on distribution and spectra of the sources and thus should be a goal of any next generation experiment such as EUSO. Both source distributions that are roughly homogeneous or follow galaxy red-shift surveys predict a recovery at a flux level of 10-5km-2sr-1 yr-1, about a hundred times smaller than the flux at the GZK threshold. This flux is probably too small to be measurable even by the Pierre Auger project and thus necessitates an instrument with acceptance ∼106 km2sr, such as EUSO. Finally, apart from more speculative processes, the GZK effect itself gives rise to secondary neutrinos from pion decay (see section XXX) whose flux should be comparable to the primary UHECR flux around the GZK threshold within one or two orders of magnitude. Due to their small cross section the probability that such neutrinos produce showers observable in the atmosphere is ≤10-2. This disadvantage for the detection process is at the same time a blessing because it makes these elusive particles reach us unattenuated over cosmological distances and from very dense environments where all other particles (except gravitational waves) would be absorbed. Giving rise to showers typically starting deep within the atmosphere, they can also be distinguished from other primaries. Their discovery and subsequent measurement with reasonable accuracy provides additional motivation to build a detector with large acceptance area.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/35

1.2 The Current Observational and Theoretical Scenario

1.2.1 Past Investigations

1.2.1.1 The Spectrum of EECRs Six ground-based experiments have reported a total of XX events >1020eV (super-GZK events) during the last 40 years: 1 event from Volcano Ranch (Linsley, 1963), 4 from Haverah Park (Lawrence et al., 1991), 1 from Yakutsk (Afanasiev et al., 1996) , 1 from Fly’s Eye (bird et al., 1993) , 8 from the Akeno Giant Air Shower Array (AGASA), and 2 from the preliminary High-Resolution Fly’s Eye Experiment (HiRes-1). This corresponds to a flux of about one event per km2 per century. At present, the two highest energy CRs measured have the energy of ~3.2 X 1020eV (Fly’s Eye) and ~3.4 X 1020eV (AGASA). The origin of these events is mysterious since there are no visible source candidates within the GZK horizon except possibly M87, a radio-loud AGN about 20 Mpc away from us, and Cen-A (NGC5128), a radio galaxy at 3.4 Mpc. Neither of these is in the direction of any observed events. Large-scale isotropy of observed events suggests that many sources, rather than one or a few sources, are probably required. Although the existence of such Extreme Energy events it’s out of doubt, currently there seems to be a disagreement between the flux and spectrum of these events as measured by the AGASA ground array, which detected about 8 events above1020 eV, as opposed to about 2 expected from a cut-off, and therefore suggesting (at 4.2 σ) the existence of an extended Super-GZK spectrum, with respect to the HiRes fluorescence detector which seems consistent with a GZK cut-off, although inconsistent with the extended Super GZK spectrum at a level of 2.3 σ. The EECR energy spectra observed by AGASA and HiRes are shown in Figure 1.1-2..The exposures of AGASA (12 years observation with 100km2 area) and HiRes (4 years operation) are almost the same around GZK –3 energy. The overall E dependence has been removed from the figure to reveal the detail. The dashed line shows the expected spectrum based on the GZK hypothesis, and assuming that a homogeneous source population fills the universe. AGASA data clearly shows that the EECRs spectrum extends beyond the Greisen-Zatsepin-Kuzmin (GZK) cutoff. As has been shown by several authors (give ref.), the conflict could arise from a shift of unknown origin in the experimental energy scale of the two experiments. And indeed if systematics are taken into account the AGASA an HiRes measures agree within 2-3 σ. As a matter of fact, the measurement of the spectrum of EECRs is still uncertain and therefore no firm observational information is available in this challenging energy window. This murky experimental situation can be, only, resolved with a high statistics and high-energy resolution experiment, such EUSO.

1.2.1.2 Sky Distribution Cosmic rays (CRs) with lower energies of ~1018 eV show a small but significant anisotropy toward the galactic center (Hayashida N., et al.). If a similar fraction of higher energy events have a galactic origin, they should show a stronger anisotropy due to their increasing magnetic rigidity. However, AGASA (Takeda et al., 1999) and a world 19 19 data summary (Nagano & Watson, 2000) of CRs above 1×10 eV (fig. XXX) and above EGZK (~5×10 eV) (fig. XXX) show an isotropic distribution in the sky on large scales clearly suggesting an extragalactic origin. 19 Among the 58 events observed by AGASA in the In the EGZK (~5×10 eV) sky, small-scale clustering (i.e. events within the 2.5° AGASA angular resolution) has been observed. AGASA counted six pairs and one triplet of spatially correlated events with arrival times differing by <2 years (see Fig.). The world data shows nine pairs and two triplets. The chance coincidence probability for these clusters arising from an isotropic distribution is <0.04 percent (Takeda et al., 1999, Teshima, et al., 2001). The interacting galaxy VV141 is observed at 100 Mpc in the direction of the triplet.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/35

Figure 1.2.1.2-1 Galactic cosmic rays up to 1018 eV. A clear anisotropy toward the Galactic centre has been observed by AGASA (Hayashida et al., 1999).

Figure 1.2.1.2-2 Isotropic distribution of CR events with energy > 1019 eV observed (Takeda et al., 1999).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 8/35

Figure 1.2.1.2-3 AGASA Super GZK event distribution.

In Fig. XXX we report the distribution of events separation for E>4×1019 eV. Both data and expected distribution (smooth solid line) are shown. A clear excess below 3° is observed. This excess corresponds to 8 pairs (5+3) from 5 pairs and one triplet. If these multiplets are not the result of a statistical fluctuation, then the only way to explain their appearance is by assuming that compact sources are the origin of EE cosmic rays. From the multiplicity distribution of such doublets, and triplet, we can estimate the number of sources (~100) in the northern hemisphere. The inference to be drawn is that particles within each cluster may have the same source and travel with minimal magnetic deflection. There are currently two possible explanations of these experimental findings: the first one assumes negligible magnetic deflection. In this case most of the sources would have to be at cosmological distances which would explain the absence of nearby counterparts and the apparent isotropy would indicate that many sources contribute to the observed flux, where a subset of especially powerful sources would explain the small-scale clustering (Tinyakov & Tkachev, 2001). This scenario predicts the confirmation of a GZK cutoff.

Figure 1.2.1.2-4 Two points correlation function of the AGASA data for E>4×1019 eV. Both data and expected distribution (smooth solid line) are shown.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 9/35

The second scenario is most likely more realistic and takes into account the likely existence of large scale intervening magnetic fields with intensity B ∼ 0.1-1 µG, correlating with the large scale galaxy distribution. In this case magnetic deflection would be considerable even at the highest energies and the observed UHECR flux could be dominated by relatively few sources within about 100 Mpc. Here, large scale isotropy could be explained by considerable angular deflection leading to diffusion up to almost the highest energies and the small scale clustering could be due to magnetic lensing. As will be discussed in more details in section XXX the EUSO mission, with its expected high statistics, will open new avenues to determine the nature on EECR sources and possibly to measure their EE spectra.

1.2.2 Present and Planned Ground-Based Observations. The largest planned ground-based experimental facility is the Pierre Auger Observatory (Cronin, 1993) presently under construction in Argentina. The detection system combines the two major techniques: a fluorescence detector system to measure the longitudinal profile of the EAS and a surface array of detectors to sample its lateral distribution at the ground. It will consist of an array of 1,600 particle detectors, and 4 fluorescence light detectors similar to the ones used in the HiRes experiment. This hybrid detector will allow cross calibration and a check of the systematic uncertainties inherent in each of the techniques. Construction of the Pierre Auger Observatory is expected to be completed in 2006. It will have an aperture of 7,000 km2sr. Extrapolating the spectrum as a power law above 1020 eV, we expect about 70 events/yr if the data follow the AGASA spectrum or 30 events/yr if the data follow the HiRes spectrum. In about half a year, it will produce a number of events comparable to all previously observed events above 1020 eV. A second Auger observatory is proposed for the northern sky. A detailed comparison between The Pierre Auger and the EUSO mission is described in section…. If the HiRes spectrum is correct, Auger may still be too limited to follow the CR spectrum much higher in energy, or to obtain the detailed form of the spectrum with small statistical errors. It is possible that only a few events above 1X1021 eV (ZeV) will be recorded in 10 years of operation with both Auger observatories. Only 20 events are expected, even if the AGASA flux is right. At least an order of magnitude more statistics is desirable.

1.2.3 Future Space-Based Observations. John Linsley (Linsley, 1982) first suggested that the Earth’s atmosphere at night, viewed from space, constitutes a huge calorimeter for remotely observing EECRs. The collecting power of the night sky on the whole Earth, 4X108km2sr, is the ultimate limit for space observatories. By comparison, ground-based observatories are reaching a practical limit at ~104km2sr. The next generation of EECR detectors must be space based. The EUSO instrument will be the first of this new generation. A geometrical factor of 5×105 km2 sr, combined with an efficiency of 0.1, gives EUSO an effective geometry factor of ~5×104 km2 sr in its nadir viewing orientation. Table 1.1 In the Foreword section shows how EUSO compares with present and future ground experiments.

1.2.4 The Current Theoretical scenario. The question of how the EECR are produced is after 40 years from their discovery still open and represents one of the big challenge to be solved in the future. Acceleration scenarios are at the base of the so called “bottom-up” models. Alternatively, EECR can be generated from the top, injected as a result of the decay of very massive particles, relics of the Big Bang. The problems encountered in trying to explain UHECRs in terms of ``bottom-up'' acceleration mechanisms have been well-documented in a number of studies; see, e.g. (Hillas, 1984, Sigl et al., Norman et al., 1995). In summary, apart from energy draining interactions in the source the maximal UHECR energy is limited by the product of the accelerator size and the strength of the magnetic field containing the charged particles to be accelerated, similar to the situation in man-made accelerators such as at CERN. These criteria reveal that it is hard to accelerate protons and heavy nuclei up to the energies observed even in the most powerful astrophysical objects such as radio galaxies and active galactic nuclei. Adding to the problem, there are no obvious astronomical counterparts within ~100 Mpc of the Earth. At the same time, no significant large-scale anisotropy has been observed in UHECR arrival directions above ~1018 eV, whereas, as we said above, there are strong hints for small- scale clustering.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 10/35

More speculative ways to explain the UHECR observations consist of circumventing the distance restriction on the sources imposed by the GZK effect. The only way this can be done without invoking an as yet unknown new physics is that charged particles accelerated in sources at much larger distances give rise to a secondary neutrino beam which can propagate unattenuated. This neutrino beam has to be sufficiently strong to produce the observed UHECRs within 100 Mpc by electroweak interactions with the relic neutrino background (CNB), the neutrino analogue of the CMB (see Weiler, 1982).

Figure 1.2.4-1 Z0 burst from the annihilation with CNB relic neutrinos in

Virgo Cluster. The decay products of Z0 are gamma rays, nucleons and neutrinos, as firmly established by the CERN LEP experiments.

Neutrino interactions become especially significant if the relic neutrinos have masses mν in the eV range and thus constitute hot dark matter, because the Z boson resonance then occurs at an UHE neutrino energy 21 Eres=4×10 eV /mν eV. The decay products of this ``Z-burst'' would then be the source of UHECRs (Weiler, 1999; Fargion et al., 1999; Fodor et al., 2002). The big drawback of this scenario is the need of enormous primary neutrino fluxes that cannot be produced by known astrophysical acceleration sources. The distance restriction can also be circumvented by postulating either new particles or new interactions beyond the Standard Model of particle physics (section below) or a violation of the Lorentz symmetry. These possibilities do not, however, solve the problem of acceleration to and beyond the observed energies which has to be solved separately. In contrast, in the ``top-down'' scenarios, which will be discussed in section 5, the problem of energetics is trivially solved. Here, the UHECR particles are the decay products of some super-massive ``X'' 20 particles of mass mX >>10 eV, and have energies all the way up to mX. Thus, no acceleration mechanism is needed. The massive X particles could be metastable Super-Heavy (SH) relics of the early Universe with lifetimes of order of or above the current age of the Universe or could be released from topological defects (TDs) that were produced in the early Universe during symmetry-breaking phase transitions envisaged in Grand Unified Theories (GUTs). If the X particles themselves or their sources cluster similar to dark matter, the dominant observable UHECR contribution would come from the Galactic Halo and absorption would be negligible. Non-astrophysical solutions of the UHECR problem are of course in general quite model dependent. In addition, if the existence of the GZK cutoff will be confirmed, their motivation surely diminishes in general. On the other hand, even in this case UHECR can be used to test and constrain new physics beyond the Standard Model, such as new interactions beyond the reach of particle collider experiments, violation of symmetries (see section below) as well as Grand Unification and early Universe cosmology, such as the rate of TD and/or massive particle production in inflation at energies often inaccessible to accelerator experiments.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 11/35

The physics and astrophysics of UHECRs are intimately linked with the emerging field of neutrino astronomy (for reviews see Halzen & Hooper, 2002) as well as with the already established field of γ-ray astronomy (for reviews see, e.g., Ong, 1998; Catanese & Weeks, 1999). Indeed, all scenarios of UHECR origin, including the top- down models, are severely constrained by neutrino and γ-ray observations and limits. In turn, this linkage has important consequences for theoretical predictions of fluxes of extragalactic neutrinos above about a TeV whose detection is a major goal of next-generation neutrino telescopes: If these neutrinos are produced as secondaries of protons accelerated in astrophysical sources and if these protons are not absorbed in the sources, but rather contribute to the UHECR flux observed, then the energy content in the neutrino flux can not be higher than the one in UHECRs, leading to the so called Waxman-Bahcall bound for transparent sources with soft acceleration spectra (Waxman & Bahcall, 1999; Bahacall & Waxman, 2001; see also section xxx). If one of these assumptions does not apply, such as for acceleration sources with injection spectra harder than E-2 and/or opaque to nucleons, or in the top-down scenarios where X particle decays produce much fewer nucleons than γ-rays and neutrinos, the Waxman Bahcall bound does not apply, but the neutrino flux is still constrained by the observed diffuse γ-ray flux in the GeV range (Manneheim et al., 2001).

1.3 Origin The origin of EECRs is completely unknown. Existing data show an excess of super GZK particles. The present number of super-GZK events is too low to allow a quantitative investigation of their origin. Several explanations have been proposed. Some, bottom-up models, employ extreme values of the key parameters of known astrophysical objects. Other explanations, top-down models, invoke new physics and/or particular particle types that avoid the GZK cutoff. Inferences to be derived from the highest energy events in nature may span the fields of traditional astrophysics, particle and neutrino astrophysics, and possibly cosmology and fundamental physics. The following paragraphs present several proposed explanations for super-GZK events. Charged particles can be accelerated to very high energies through the conversion of the kinetic energy of plasma flows or plasma waves into the kinetic energy of a few particles in the surrounding space. These phenomena can occur in very different astrophysical sites, ranging from compact objects (neutron stars, black holes, white dwarfs) to extended objects (clusters of galaxies, cosmological accretion flows). The maximum energies that can be achieved in astrophysical accelerators, as well as in terrestrial accelerators, are always the result of a balance between the acceleration process and energy losses of the particles, or their escape from the accelerator. The details therefore depend upon the local conditions in the acceleration region. These criteria become extremely restrictive when the maximum energies that we are interested in fall in the range E>1020 eV, which is the domain of the so-called Extreme Energy Cosmic Rays (EECRs). Most astrophysical sources of EECRs, being distributed almost homogeneously in the universe, produce a spectrum of cosmic rays at the Earth that has the GZK imprint in it. If EUSO had to confirm the AGASA-like spectrum, then a new component might be required, probably of non astrophysical origin, in order to explain observations. Different astrophysical sources may be sites of particle acceleration, but the acceleration processes that may be invoked are often the same in a wide class of sources. In this section we discuss the basic physics underneath some of the better known acceleration processes, while a discussion of the sources may be found in the next section.

1.3.1 Acceleration Process Different astrophysical sources may be sites of particle acceleration, but the acceleration processes that may be invoked are often the same in a wide class of sources. In this section we discuss the basic physics underneath some of the better known acceleration processes.

1.3.1.1 Diffusion acceleration at Newtonian Shocks Diffusive shock acceleration is thought to be responsible for acceleration of cosmic rays in several astrophysical environments. Most of the observational evidence for this mechanism, also known as first order Fermi acceleration, has been provided by studies of heliospheric shocks, but there are indirect lines of evidence that acceleration occurs at other shocks as well. A particularly impressive example of particle acceleration is represented by the observation of gamma ray emission from the supernova remnant SN1006. These observations could be interpreted as inverse Compton emission of very high energy electrons, accelerated at the shock on the rim of

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 12/35

SN1006, though other radiation processes may contribute (Aharonian, & A.M. Atoyan). Shock acceleration has been studied carefully and a vast literature exists on the topic. Some recent excellent reviews have been written (see Jones & Ellison, 1991).

Figure 1.3.1.1-1 Scheme of diffusive shock acceleration. Charged particles scatter back and forth across a shock due to diffusion in the magnetic field present in the upstream and downstream fluids.

Charged particles scatter back and forth across a shock due to diffusion in the magnetic field present in the upstream and downstream fluids, as illustrated in Figure 1.3.1.1-1. At each shock crossing the energy of the particle increases by a small amount. The energy gain of a particle crossing the shock from upstream to downstream and back is proportional to the relative speed between the upstream and downstream flows, namely ∆E/E ∝V, with V ∼ c for non-relativistic shocks. In the reference frame of the upstream (downstream) fluid, the downstream (upstream) flow appears as moving in the direction of the shock, so that the acceleration process can be pictured as a reflection of the particle against a mirror always moving in the direction opposite to the motion of the particle. These multiple reflections gradually increase the energy of the particle, until either its energy losses become too fast or the particle escapes from the accelerator, because its diffusion length becomes larger than the size of the shock. The spectrum of the accelerated particles is completely determined by the properties of the shock and in particular by its compression factor. Under the assumption of the so-called test particle approximation, which is based on the assumption that the accelerated particles do not affect the shock structure, typical E-2 spectrum is obtained. There are now plenty of indications that this might not be the case, and the shock may turn non-linear. The non-linearity of the shock as particle accelerator is the consequence of two main physical effects, namely the back-reaction of the accelerated particles on the structure of the fluid velocity field, and the generation of the plasma waves that the particles scatter against, and that are responsible for the particle diffusion across the shock surface. Both effects become important when the pressure in accelerated particles becomes comparable to the dynamical pressure ρu2 of the plasma crossing the shock. There is the interesting possibility that the shock may behave as a self-regulating machine, in which the acceleration efficiency and the maximum energy of the particles adapt to each other through the back-reaction of the particles on the shock. (TBC)

1.3.1.2 Unipolar induction Shock acceleration is the typical example of a stochastic mechanism for particle energization. However, charged particles can also be accelerated in regular electric fields, typically generated due to unipolar induction. A rotating magnetic field B induces an electric E = 1/c v × B, as seen in the laboratory frame (Berezinsky et al. 1990), and references therein for a review). While the details of the acceleration process depend on the specific situation, the

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 13/35

basic principles are easily explained, in terms of the electromotive force (emf) between the inductor's surface and infinity. The order of magnitude of this emf is ΩR2 ∆V = B c where Omega is the angular speed and R is the size of the unipolar inductor. The magnetic field B also depends on the details of the astrophysical scenario that we apply the unipolar induction to. The apparent semplicity of this description hides a much more intricate scenario: the above equation only represents the difference in the potential between the rotator and infinity, but typically the magnetosphere of one of these objects is characterized by definition by E B = 0, so that there cannot be any acceleration along the magnetic field lines. This suggests that there must be regions where the condition E B = 0, is violated, and the particles can in fact be accelerated by the emf. These regions are called potential gaps, and their location in the magnetosphere of the unipolar inductor is hardly obtainable from first principles. Unfortunately the conditions in the gap region and around it are critical to estimate the maximum energy achievable by the accelerating particles: first, the electron-positron pairs that are often associated with the creation of the gap may easily short-cut the gap, and dramatically reduce the emf available for particle acceleration. Second, the photon background in the gap region can substantially affect the maximum energy through inelastic collisions with the accelerating particles. Third, the magnetic field configuration may be such that curvature radiation does hinder the possibility to reach high energies for the accelerated particles. The delicate balance among all these processes determines the possibility or the impossibility for the unipolar inductor to be an efficient accelerator.

1.3.1.3 Non-linear particle-wave interaction A non linear wave propagating in a plasma can induce the creation of an electrostatic field, due to a local charge separation. This electric field can in principle be responsible for the acceleration of particles to very high energies, depending upon the local conditions. An example of this acceleration mechanism was recently proposed in Chen et al. (2002). in this case a non-linear wave propagating through a plasma can develop a wake, that is the site for the development of a strong electric field due to charge separation. For a relativistic plasma moving with Lorentz factor Gamma_p, the maximum acceleration gradient can be written as π 2 eE 4 npe G = m c 2 ω e 2 Γ mc me c p

where E and ω are the electric field and frequency of the wave, ωp is the plasma frequency and np the plasma density. In general, in the astrophysical context there will be waves that generate a wake responsible for acceleration and others that decelerate particles, so that the acceleration process is stochastic in nature. In Chen -2 et al., (2002) the spectrum of accelerated particles was derived to be ∝ Ep , where Ep is the energy of the accelerated particles.

1.3.2 Astrophysical Sources In this section we consider in more detail the most plausible sources of EECRs and discuss their basic observational imprints and the acceleration processes that may be at work. Cosmological sources with a homogeneous spatial distribution is expected to generate a spectrum at the Earth which is characterized by the GZK feature. On the other hand, small scale anisotropies may also be expected as a result of the point-like nature of these sources, and represent an extremely important diagnostic tool to identify the nature of the sources. Galactic sources of EECRs seem to be disfavored by current experiments due to the lack of evidence for large scale anisotropies in the direction of the Galactic disk. As shown in recent simulations (see section xxx) the EUSO experiment has the potential to measure the shape of the GZK feature in the cosmic ray spectrum with high statistical significance, if it is there. If this feature will be in fact detected, this would represent the strongest evidence in favor of astrophysical sources distributed approximately homogeneously on cosmological scales.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 14/35

Figure 1.3.2-1 Different celestial sources as possible sites of EECRs acceleration. The size of the source is plotted vs. the magnetic field strength. Objects below the red dashed diagonal line can not 20 accelarate particle to 10 eV by shock acceleration. The solid line indicates the limit for ZeV protons, the dotted line is the limit for 1020 eV protons, both with extreme shock speeds of β=1.

The simplest modelling of Fermi acceleration by shock waves gives the maximum energy acquired by a particle of charge Ze: E max ≈ βc × Ze × B× L where L is the characteristic size of the acceleration region and βc is the shock velocity. Under certain configurations of the shocks and the magnetic field b can be replaced by a much larger factor of the order of 1-3.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 15/35

The above equation essentially states that the gyro-radius of the particle being accelerated must be contained within the acceleration region. Because it is reasonable to assume that energy losses take place in astrophysical accelerators it is more likely that the above value of Emax should be reduced by perhaps a factor of ten, depending on the details of the shock and environment. No known astrophysical object is able to easily accelerate particles to 20 10 eV. This is illustrated in Figure 1.3.2-1 [Hillas]; objects below the red dashed diagonal line cannot accelerate 20 particles to 10 eV by shock acceleration. The solid line indicates the limit for ZeV protons, the dotted line is the 20 limit for 1020 eV protons, both with extreme shock speeds of β=1. Even if 10 eV could be achieved, it is unclear how the accelerated particles can emerge from the dense radiation fields near the acceleration region without significant energy loss. In addition, most of the radio galaxies and AGN are at distances >100 Mpc from Earth. Acceleration, even to ZeV in these distant sources, is insufficient to explain the observed events.

1.3.2.1 Active Galactic Nuclei and Dead Quasars The term (AGN) is used here to identify a very broad class of objects, characterized by the presence of a massive black hole in the centre, and fuelled by an accretion flow that is responsible for the large brightness of the host Galaxy. The presence of jets is a characteristic of a subclass of the AGNs. When the fuel is almost exhausted, the AGN stops being active. Objects like these have been named dead quasars in (Boldt & Gosh, 1999). Three potential sites of production of EECRs exist : 1) the central regions of AGNs; 2) jets of some special class of AGNs; 3) accretion disks of dead quasars. Different acceleration mechanisms may potentially be at work in the three scenarios. In simplest form, the central region of an AGN can be modelled as an approximately spherical inflow, with a shock surface at 10-100 times the horizon radius of the black hole (Szabo & Protheroe, 1994). This shock may potentially be the site for acceleration of cosmic rays, through diffusive shock acceleration. Although the size of this shock might be large enough to reach 1020 eV, energy range, energy losses, due to photopion production and synchrotron-Compton scattering, could limit the maximum energy to much lower values, of the order of 106-107 GeV. Hot spots in F-R II radio galaxies appear to be more promising sites for the generation of EECRs, due to the lower photon density and lower values of the magnetic field in the acceleration region. The hot spot is terminated by a strong shock at which particles may be accelerated diffusively reaching a maximum energy that can be as high as 1021 eV for the most optimistic choice of the parameters. Recently (Boldt & Gosh, 1999) has been proposed that unipolar induction in the vicinity of the event horizon of a supermassive black hole hosted by what was in the past an AGN and that exhausted its accretion fuel, may energize charged particles to ultra high energies. The authors propose that the relatively under-dense material in the accretion flow and the low value of the magnetic field may reduce the effect of the losses compared to the case of AGNs, while keeping an electromotive force between the horizon and infinity which is large enough to accelerate EECRs, provided the mass of the black hole is large enough. Candidate sources that could fit this description were searched for in (Boldt, & M. Loewenstein, 2000) among the known massive under-luminous galaxies. The model has the important advantage that since the source is expected to have very low activity, it would not appear in the sky as a bright source easy to correlate to the direction of arrival of one of the observed EECR events. On the other hand, it has been proposed (A. Levinson, 2000) that the extremely high energy particles moving along the magnetic field lines may radiate part of their energy as curvature radiation, that would appear as gamma ray emission from the same sources of EECRs. At present no accurate calculation exists of the spectrum of the particles accelerated by dead quasars, and of the corresponding spectrum of particles at the Earth. Other astrophysical objects may well be suitable candidates as sources of EECRs, although a specific model may not have been worked out in the details. Recently, evidence has been found for a correlation of the arrival directions of cosmic rays with energy above 4×1019 eV with the spatial location of BL Lac objects (Tinyakov & Tkachev, 2001) ,many of which are at large redshifts. The authors estimate the probability that the coincidence may occur by chance in the amount of one part in 104. The correlation appears to be even more interesting after the recent claim (D. Gorbunov, et a., 2002) that a subsample of gamma ray loud BL Lac objects, probably seen even by EGRET, are in the direction of arrival of some ultra high energy cosmic rays.

1.3.2.2 Neutron Stars Rapidly rotating strongly magnetized neutron stars represent the most efficient unipolar inductors, with an electromotive force that may reach 1021 V (Venkatesan et al., 1997) between the polar region and the equator of the star. This potential is likely to be partially short-cut by the electron-positron pairs in the magnetosphere of the

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 16/35

neutron star, so that the effective emf may be appreciably less. The high conductivity of the plasma in the magnetosphere makes the condition E • B = 0 true almost everywhere, so that the emf may be present only in a gap, where this condition is violated. It is not clear so far what is the location of this gap, so that models have been formulated with gaps located at the polar cap or in alternative at the light cylinder, where the corotation speed of the magnetospheric plasma equals the speed of light (Chen et al., 1986). For a review of the physics of pulsar acceleration (Berezinsky et al., 1990) from which we deduce that the maximum achievable energy is typically around 1015 eV, mainly due to the limitation imposed by curvature energy losses during the motion of the particle along the open magnetic field lines. Similar results hold when the unipolar induction works in an accretion disk around either a neutron star or a stellar size black hole with possible microquasar activity. For the reasons outlined above, in general the presence of superstrong magnetic fields, such as those expected around magnetars and/or soft gamma ray repeaters, are unlikely to help to achieve higher energies, because although the emf gets larger, energy losses may also become more severe as well, unless quantum effects suppress the cross section for the radiative processes involved, as would be expected for very high energy particles in superstrong magnetic fields. In (Blasi et al., 2000) a phenomenological approach was taken to rule in favor of acceleration of heavy nuclei (as well as protons) to extremely high energies at the light cylinder of young neutron stars: it is well known that most energy lost by a spinning neutron star is not in the form of the radiation observed so far (Kennel & Coroniti, 1984) This suggests that this energy is actually converted into kinetic energy of a relativistic wind that is in fact required to explain the observed properties of plerion-like pulsars. The Lorentz factor of the wind must be of the order of 6 7 γW∼10 -10 for a Crab-like pulsar, although it is not clear so far what is the mechanism through which the energy radiated by the neutron star is converted into the kinetic energy of the wind. The approach of (Blasi et al., 2000) is to assume that at the light cylinder a small pollution in the form of Iron nuclei is present (protons may be present as well). In this case the nuclei may acquire the same Lorentz factor of the wind, that for young neutron stars may well 10 1 be in the useful range γW∼10 -10 . In order to escape freely from the plerion these nuclei should be able to cross the ejecta without suffering spallation, and they should also not be destroyed due to the inelastic interaction with the thermal photons from the surface of the neutron star. These conditions allow the authors to impose constraints on the magnetic field at the surface of the neutron star and on the age of the plerion when the acceleration and escape become possible. In case of Iron nuclei, the origin may be galactic, although this possibility does not appear to be favored by current data, due to the fact that an anisotropy is expected but not observed in the direction of the galactic disk. Protons should, for the same reason, be generated in neutron stars in other galaxies, as is the case also in other scenarios that may be applied to neutron stars.

1.3.2.3 Gamma Ray Bursts EECRs may be accelerated through the Fermi mechanism at the relativistic shock front created by the relativistic fireball of a Gamma Ray Burst (GRB) (Vietri, 1995, Waxman, 1995). When the burst explodes in the interstellar 15 medium, with magnetic field in the µG range, the maximum energy of the accelerated particles is Emax ≈10 eV Bµ, where Bµ is the magnetic field in µG. On the other hand, two of the most successful models for GRBs currently in the literature, namely the binary neutron star merger Narayan et al., 1992, and the SupraNova model (Vietri & Stella, 1998) predict that the GRB goes off in the relativistic wind of a neutron star, the so-called pulsar wind bubble (PWB). In this case the maximum energy can well be in the range of interest for the production of EECRs because the magnetic field is highly enhanced and has a radial dependence 1/r. The predicted spectrum of the accelerated particles is ∝E-γ with γ=2-2.5, and may possibility of acceleration of particles to energies well in excess of those required by observations. The detection extend to more than ∼1020 eV (Vietri et al., 2003). In (Chen et al., 2002) the wakefield acceleration mechanism has been invoked in the context of GRBs, to assess the of EECRs with energy in excess ≥5×1019 eV requires that if they are accelerated in GRBs, a magnetic field must be present in the intergalactic medium, in order to allow the burst event to appear diluted in time, so that more than one burst is observed at fixed time in EECRs. This is required by the near isotropy observed in the spatial distribution of the arrival directions. An upper limit to the intergalactic magnetic field is obtained from measurements of the Faraday rotation from distant quasars: for reasonable values of the parameters this upper limit is in the range 10-9-10-11 Gauss. In this field two phenomenological implications can be found: 1) the average deflection angle of EECRs from the same bursts is smaller than the angular resolution of the present experiments (several degrees); 2) particles generated at the same burst arrive at the detector in inverted temporal order (higher energy particles arrive first). Observations of small scale anisotropies in the AGASA detector do not seem to fulfill the second requirement, although the uncertainty in the energy determination are such that no definite conclusion can be inferred. In (Vietri et al., 2003) it has been pointed out that fluctuations may in fact be responsible for lower energy particles to arrive

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 17/35

earlier, although the probability to obtain the observed multiplets simply due to statistical fluctuations has not yet been evaluated. In (Stecker 2000, Scully & Stecker, 2002) it was argued that a mismatch exists between the energy available in gamma rays in GRBs and the energy in EECRs. The claim is based on the argument that if GRBs follow approximately the star-formation history, then the local rate of bursts should be low and unable to explain the observed flux of cosmic rays with energy above 1020 eV. The arguments in (Stecker 2000) were recently addressed in (Vietri et al., 2003): the calculated energy injection rate at energies above 1017 eV, for an injection spectrum E-2.2, is a factor ∼6 larger than that available in gamma rays from GRBs. The mismatch therefore remains, but it is somewhat reduced if compared with that claimed in previous work. In conclusion there seems still to be some problems in the energetics of GRBs as sources of EECRs, although may be somewhat mitigated by the uncertainties in the GRBs energy deposition in the gamma ray component.

1.3.3 Physics beyond the Standard Model

1.3.3.1 New Primary Particle and interactions A possible way around the problem of missing counterparts within acceleration scenarios is to propose primary particles whose range is not limited by interactions with the CMB. Within the Standard Model the only candidate is the neutrino, whereas in extensions of the Standard Model one could think of new neutrals such as axions or stable supersymmetric elementary particles. Such options are mostly ruled out by the tension between the necessity of a small EM coupling to avoid the GZK cutoff and a large hadronic coupling to ensure normal air showers (Gorbunov, et al., 2001). Also suggested have been new neutral hadronic bound states of light gluinos with quarks and gluons, so-called R- that are heavier than nucleons, and therefore have a higher GZK threshold. In both the neutrino and new neutral stable particle scenario the particle propagating over extragalactic distances would have to be produced as a secondary in interactions of a primary proton that is accelerated in a powerful active galactic nucleus which can, in contrast to the case of EAS induced by nucleons, nuclei, or γ-rays, be located at high redshift. Consequently, these scenarios predict a correlation between primary arrival directions and high redshift sources. In fact, possible evidence for a correlation of UHECR arrival directions with compact radio quasars and BL-Lac objects, some of them possibly too far away to be consistent with the GZK effect, was recently reported. The main challenge in these correlation studies is the choice of physically meaningful source selection criteria and the avoidance of a posteriori statistical effects. (Tinyakov and Tkachev, 2001). However, a moderate increase in the observed number of events will most likely confirm or rule out the correlation hypothesis. Note, however, that these scenarios require the primary proton to be accelerated up to at least 1021 eV, demanding a very powerful astrophysical accelerator.

1.3.3.2 Top-down Scenarios The difficulties of bottom-up acceleration scenarios discussed earlier motivated the proposal of the ``top-down'' scenarios, where UHECRs, instead of being accelerated, are the decay products of certain ``X'' particles of mass close to the GUT scale. Such particles can be produced in basically two ways: If they are very short lived, as usually expected in many GUTs, they have to be produced continuously. The only way this can be achieved is by emission from TDs left over from cosmological phase transitions that may have occurred in the early Universe at temperatures close to the GUT scale, possibly during reheating after inflation. TDs necessarily occur between regions that are causally disconnected, such that the orientation of the order parameter associated with the phase transition can not be communicated between these regions and consequently will adopt different values. Examples are cosmic strings, magnetic monopoles, and domain walls. The defect density is thus given by the particle horizon in the early Universe. The defects are topologically stable, but time dependent motion leads to the emission of particles of mass comparable to the temperature at which the phase transition took place. The associated phase transition can also occur during reheating after inflation. Alternatively, instead of being released from TDs, X particles may have been produced directly in the early Universe and, due to some unknown symmetries, have a very long lifetime comparable to the age of the Universe. In contrast to Weakly-Interacting Massive Particles (WIMPS) below a few hundred TeV which are the usual dark matter candidates motivated by, for example, supersymmetry and can be produced by thermal freeze out, such super-heavy X particles have to be produced non-thermally (see Kuzmin & Tkachev, 1999, for a review). In all these cases, such particles, also called ``WIMPZILLAs'', would contribute to the dark matter and their decays could

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 18/35

still contribute to UHECR fluxes today, with an anisotropy pattern that reflects the dark matter distribution in the halo of our Galaxy. However, these scenarios predict the absence of the GZK cut-off and thus will be ruled out if the existence of the GZK cutoff is confirmed. Furthermore, in order to reproduce the observed UHECR flux, they require strong fine tuning between the density in units of the critical density, ΩX, and lifetime, tX, of the X particle, -11 10 ΩX∼10 (tX/10 yr), making them appear somewhat artificial. It is interesting to note that one of the prime motivations of the inflationary paradigm was to dilute excessive production of ``dangerous relics'' such as TDs and super-heavy stable particles. However, such objects can be produced right after inflation during reheating in cosmologically interesting abundances, and with a mass scale roughly given by the inflationary scale which in turn is fixed by the CMB anisotropies to ∼1013 GeV. This mass scale is somewhat above the highest energies observed in CRs, which implies that the decay products of these primordial relics could well have something to do with UHECRs which therefore can probe such scenarios. The X particles could be gauge bosons, Higgs bosons, super-heavy fermions, etc. depending on the specific GUT. They would have a mass mX comparable to the symmetry breaking scale and would decay into leptons and/or quarks of roughly comparable energy. The quarks interact strongly and hadronize into nucleons (Ns) and pions, the latter decaying in turn into γ-rays, electrons, and neutrinos.

1.3.3.3 Top-down Fluxes Fig (neutrino chapter). shows results for the time averaged nucleon, γ-ray, and neutrino fluxes in a typical top-down scenario, along with constraints on diffuse γ-ray fluxes at GeV energies and neutrino flux sensitivities of future experiments. The spectrum was optimally normalized to allow for an explanation of the observed UHECR events, assuming their consistency with a nucleon or γ-ray primary. The flux below ≤ 2 ×1019 eV is presumably due to conventional acceleration in astrophysical sources. The photopion process on the CMB depletes the photon flux above 100 TeV, and the same process on the IR/O background causes depletion of the photon flux in the range 100 GeV--100 TeV, recycling the absorbed fluxes to energies below 100 GeV through EM cascading. The scenario in Fig (neutrino chapter) obviously obeys all observational constraints within the normalization ambiguities. Note that the diffuse γ-ray background measured by EGRET up to 10 GeV puts a strong constraint on these scenarios, especially if there is already a significant contribution to this background from conventional sources such as unresolved γ-ray blazars. However, this constraint is much weaker for TDs or decaying long lived X particles with a non-uniform clustered density. The energy loss and absorption lengths for UHE nucleons and photons are ≤ 100 Mpc. Thus, predicted UHE nucleon and photon fluxes are independent of cosmological evolution. The γ-ray flux below ≅ 1011eV, however, scales as the total X particle energy release integrated over all redshifts and increases with decreasing p. For mX=2 ×1016 GeV, scenarios with p<1 are therefore ruled out, whereas models with a co-movingly constant injection rate (p=2) are well within the limits. It is clear from the above discussions that the predicted particle fluxes in the top-down scenarios are currently uncertain to a large extent due to particle physics uncertainties (e.g., mass and decay modes of the X particles, the quark FF, the nucleon fraction, and so on) as well as astrophysical uncertainties (e.g., strengths of the radio and infrared backgrounds, extragalactic magnetic fields, etc.). We stress here that there are viable top-down scenarios which predict nucleon fluxes comparable to or even higher than the γ-ray flux at all energies, even though γ-rays dominate at injection. Some of these top-down scenarios would therefore remain viable even if UHECR induced EAS should be proven inconsistent with photon primaries. This is in contrast to scenarios with decaying massive dark matter in the Galactic halo which, due to the lack of interactions of injected particles, predict compositions directly given by the FFs, i.e. domination by γ-rays, and thus may be in conflict with observed compositions (Shinozaki et al., 2002) As discussed above, in top-down scenarios most of the energy is released in the form of EM particles and neutrinos. If the X particles decay into a quark and a lepton, the quark hadronizes mostly into pions and the ratio of energy release into the neutrino versus EM channel is r∼.3. The energy fluence in neutrinos and γ-rays is thus comparable. However, whereas the photons are recycled down to the GeV range where their flux is constrained by the EGRET measurement, the neutrino flux is practically not changed during propagation and thus reflects the injection spectrum. Its predicted level is consistent with all existing upper limits (see Fig neutrini 2) and should be detectable by several experiments under construction or in the proposal stage such EUSO. This would allow to directly see the quark fragmentation spectrum.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 19/35

1.4 Propagation of EECR

1.4.1 The GZK Effect Thirty eight years ago, Penzias and Wilson reported the discovery of the cosmic 3K thermal blackbody radiation which was produced very early on in the history of the universe and which led to the undisputed acceptance of the ``big bang'' theory of the origin of the universe. The perfect thermal character and smoothness of the CBR proved conclusively that this radiation is indeed cosmological and that, at the present time, it fills the entire universe with a 2.7K spectrum of radio to far-infrared photons with a density of ∼400cm-3. Shortly after the discovery of the CBR, Greisen and Zatsepin & Kuz'min independently predicted that pion-producing interactions of ultrahigh energy cosmic ray protons with CBR photons of target density ∼400cm-3 should produce a cutoff in their spectrum at energies greater than ∼5×1019 eV. This predicted effect has since become known as the GZK (Greisen-Zatsepin- Kuz'min) effect. For protons, this occurs when the pion production threshold is reached. The reaction p+γ→∆+→p+π° or n+π+ will quickly slow down the proton and lead to an effective attenuation length of 50 Mpc for a proton of 1020 eV. 50 Mpc is about the size of the Virgo cluster to which our galaxy belongs, and is just a small fraction of the size of the Universe.

Process Cutoff Energy Mean free path 14 20 20 γ + γ2.7K ≥10 eV (at 10 eV) 10 Mpc (at 10 eV) 19 p + γ2.7K → π° + X ≥ 5 10 eV 50 Mpc 22 ν + ν1.95K → (W/Z0) + X ≥ 4 10 eV 40 Gpc Table 1.4.1-1 Extreme energy processes that cutoff the energy spectrum of particles in Universe.

Following the GZK papers, Stecker and Berezinsky independently utilized data on the energy dependence of the photomeson production cross sections and inelasticities to calculate quantitatively the mean energy loss time for protons propagating through the CBR in intergalactic space as a function of energy. Based on calculated EECR proton propagation characteristics, Stecker suggested that the particles of energy above the GZK cutoff energy (hereafter referred to as trans-GZK particles) must be coming from within the ``Local Supercluster'' of which we are a part and which is centered on the Virgo Cluster of galaxies. Thus, the ``GZK cutoff'' is not a true cutoff, but a supression of the ultrahigh energy cosmic ray flux owing to a limitation of the propagation distance to a few tens of Mpc. The actual position of the GZK cutoff can differ from the ∼5×1019 eV predicted by Greisen. In fact, there could actually be an enhancement at or near this energy owing to a ``pileup'' of cosmic rays starting out at higher energies and crowding up in energy space at or below the predicted cutoff energy. The existence and intensity of this predicted pileup depends critically on the flatness and extent of the source spectrum, i.e., the number of cosmic rays starting out at higher energies. Scully & Stecker (2002) have determined the GZK energy, defined as the energy for a flux decrease of 1/e, as a function of redshift (see Fig.). At high redshifts, the target photon density increases by (1+z)3 and both the photon and initial cosmic ray energies increase by (1+z).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 20/35

Figure 1.4.1-1 The GZK cut-off energy versus redshift Figure 1.4.1-2 Predicted spectra for cosmic ray protons (Scully and Stecker, 2002) as compared with the data. The middle curve and lowest curve assume an E-2.75 source spectrum with a uniform source distribution and one tha follows the z distribution of the star formation rate respectively. The upper curve is for an E-2.35 source spectrum which requires an order of magnitude more energy input and exhibits a ``pileup effect'' (Scully and Stecker 2002).

Figure xxx, also from Scully & Stecker, shows the form of the cosmic ray spectrum to be expected from sources with a uniform redshift distribution and sources which follow the star formation rate such as would be expected if the source of EECRs were Gamma Ray Bursts (GRBs). The required normalization and spectral index determine the energy requirements of any cosmological sources which are invoked to explain the observations. Pileup effects and GZK cutoffs are evident in the theoretical curves in Figure 2. As can be seen in this figure, the present data from HiRes and AGASA appear to be statistically consistent with either the presence or absence of a pileup effect. Future data with much better statistics are required to determine such a spectral structure, but if its existence is confirmed in the future by more sensitive detectors such as EUSO, it would be evidence for the GZK effect.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 21/35

Figure 1.4.1-3 Mean energy loss times for protons Figure 1.4.1-4 The mean free path for ultrahigh energy γ- (Stecker 1968; Puget, Stecker and Bredekamp 1976) ray ttenuation vs. energy. The curve for electron-positron and nuclei originating as Fe (Stecker and Salamon pair production off the cosmic background radiation 1999). (CBR) is based on Gould & Schreder (1966). The two estimates for pair production off the extragalactic radio background are from Protheroe & Biermann (1996). The curve for double pair production is based on Brown et al., (1973). The physics of pair production by single photons in magnetic fields is discussed by Erber (1966). This process eliminates all photons above ∼ 1024 eV and produces a terrestrial anisotropy in the istribution of photon arrival directions above ∼1019eV.

1.4.2 The Propagation of Heavy Nuclei A conservative hypothesis for explaining the trans-GZK events is that they were produced by heavy nuclei. Stecker & Salamon (1999) have shown that the energy loss time for nuclei starting out as Fe is longer than that for protons for energies up to a total energy of ∼300 EeV (see Figure 3). An interesting new clue that we may indeed be seeing heavier nuclei above the proton-GZK cutoff comes from a very recent analysis of inclined air showers above 10 EeV energy by the Haverah Park group. These new results favor proton primaries below the p-GZK cutoff energy but they appear to favor a heavier composition above the p-GZK cutoff energy. This, of course, will be one of the scientific objectives of the next generation EECRs observatory such EUSO.

1.4.3 Photon Propagation Origin models such as Z-burst and ultraheavy halo dark matter (``wimpzilla'') decay or annihilation predict a high ratio of photons to protons in the EECRs. These ultrahigh energy photons can reach the Earth from anywhere in a dark matter galactic halo, because, as shown in Figure 4, there is a ``mini-window'' for the transmission of ultrahigh energy cosmic rays between ∼0.1 and ∼106 EeV. Photon-induced giant air showers have an evolution profile which is significantly different from nucleon-induced showers because of the Landau-Pomeranchuk-Migdal (LPM) effect. and because of cascading in the Earth's magnetic field By taking this into account, the AGASA group have placed upper limits on the photon composition of their UHECR showers. They find a photon content upper limit of 28% for events above 1019 eV and 67% for events above 3×1019 eV at a 95% confidence level with no indication of photonic showers above 1020 eV. A recent reanalysis of the ultrahigh energy events observed at Haverah Park indicates that less than half of the events (at 95% confidence level) observed above 10 and 40 EeV are γ-ray initiated. The profile of the highest energy Fly's Eye event (E=300 EeV) shows it not to be of photonic origin as determined by Halzen and Hooper (2002). In

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 22/35

addition, the AGASA group has found no indication of departures from isotropy as would be expected from photonic showers produced in the galactic halo, this admittedly with only 10 events in their sample. Again, much better statistics from detectors such as EUSO are indicated.

1.4.4 Deflection in cosmic magnetic fields Magnetic fields are omnipresent in the Universe, but their origin still lies in the dark (Grasso & Rubinstein, 2001). Best known are the magnetic fields in galaxies which have strengths of a few micro Gauss, but there are also some indications for fields correlated with larger structures such as galaxy clusters. Magnetic fields as strong as ~1 µG in sheets and filaments of the large scale galaxy distribution, such as in our Local Supercluster, are compatible with existing upper limits on Faraday rotation (Ryu et al., 1998, Blasi et al., 1999). It is also possible that fossil cocoons of former radio galaxies, so called radio ghosts, contribute to the isotropization of UHECR arrival directions (Medina-Tanco & Enßlin, 2001). Contrary to the case of electrons, for charged hadrons deflection is more important than synchrotron loss in the EGMF. To get an impression of typical deflection angles one can characterize the EGMF by its r.m.s. strength B and a coherence length lc. If we neglect energy loss processes for the moment, then the r.m.s. deflection angle over a distance Test of Fundamental Physics If we neglect energy loss processes for the moment, then the r.m.s. 1/2 deflection angle over a distance r≥ lc in such a field is θ(E,r)≅(2rlc/9) /rL, where the Larmor radius of a particle of charge Ze and energy E is rL\≅ E/(ZeB). In numbers this reads, for r≥lc :

−1 1 / 2 1 / 2  E   r   l   B  θ ( E , r ) ≅ 0.8° c  20       −9   10 eV   10 Mpc   1Mpc   10 G 

This expression makes it immediately obvious why a magnetized Local Supercluster with fields of fractions of micro Gauss prevents a direct assignment of sources in the arrival directions of observed UHECRs; the deflection expected is many tens of degrees even at the highest energies. This goes along with a time delay

τ ≅ θ 2 ≅ × 3 2 20 −2 2 −9 ( E,r ) r ( E,d ) / 4 1.5 10 Z ( E 10 ) ( r / 10Mpc ) ( lc / Mpc )( B / 10 G )yr

which can be millions of years. A source visible in UHECRs today could therefore be optically invisible since many models involving, for example, active galaxies as UHECR accelerators, predict variability on shorter time scales. On the other hand, an EGMF of this size could explain the observed large scale UHECR isotropy by diffusion and the small-scale clustering by magnetic lensing, even if most of the sources are relatively nearby. In fact, numerical simulations of nucleons in a simplified scenario where the UHECR source density in the Local Supercluster is idealized as a Gaussian sheet of a few Mpc thickness and about 20 Mpc length containing a magnetic field with a Kolmogorov spectrum and an r.m.s. strength B proportional to the same profile have lead to the following result: About 10 sources in the Local Supercluster and a maximal field strength of ~.3µG lead to arrival direction multi-pole moments and autocorrelation functions consistent with the AGASA data (Isola & Sigl, 2002). In particular one finds that as long as the observer is surrounded by magnetic fields above about 0.1 µG, which is possible but not obvious for our environment, roughly 10 or more sources lead to multi-poles and autocorrelations marginally consistent with present data which are limited to the Northern hemisphere, although consistency of large scale multi-poles is somewhat worse than for more extended EGMFs. An example for 100 sources is shown in Figs.~\ref{fig6.2.1} .A proton injection spectrum roughly ∝E-2.4 extending up to ∼1021eV can reproduce the sub-GZK spectrum and tends to predict a spectrum somewhere between the AGASA and HiRes observations above GZK energies, see Fig.~\ref{fig6.2.3}. In contrast, if the observer is in a region of EGMF field strength much smaller than ∼0.1µG, the UHECR sky distribution reflects the highly structured large scale galaxy distribution and is thus inconsistent with the observed isotropy.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 23/35

Figure 1.4.4-1 Trajectory for a proton with energy of Figure 1.4.4-2 Deflection of protons with energies 1019 eV in our galaxy. In the calculation antiproton is from 2x1019-1020 eV using models for the galactic ejected from the Earth. magnetic field as in Stanev (1997). The particel is injected at the Earth.

Figure 1.4.4-3 The angular power spectrum C(l) as a Figure 1.4.4-4 Predicted spectrum observed by AGASA function of multi-pole l, obtained for the AGASA exposure for the scenario discussed in (Isola & Sigl, 2002). function, for N=57 events observed above 40 EeV, sampled from 12 simulated configurations of 100 sources, emitting a proton spectrum ∝E-2.4 up to 1021eV. The magnetic field strength at the observer is ∼0.13\mu\,$G. The diamonds indicate the average over 12 realizations, and the left and right error bars represent the statistical and total (including cosmic variance due to different realizations) error, respectively, see text for explanations. The histogram represents the AGASA data. The overall likelihood significance is ∼.15.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 24/35

One also finds that sources outside our Local Supercluster do not contribute significantly to the observable flux if the observer is immersed in magnetic fields above about 0.1 µG and if sources reside in magnetized clusters and super-clusters: For particles above the GZK cutoff this is because sources outside the Local Supercluster are beyond the GZK distance. Sub-GZK particles are mainly contained in the magnetized environment of their sources and thus exhibit a much higher local over-density than their sources. The confidence levels that can be obtained for specific models of our local magnetic and UHECR source neighborhood will greatly increase with the increase of data anticipated from future experiments. Full sky coverage alone will play an important role in this context as many scenarios predict large dipoles for the UHECR distribution. Whereas current northern hemisphere data are consistent with the simulated scenario at the ∼1.5 sigma level, a comparable or larger exposure in the southern hemisphere would be sufficient in this case to find a dipole at several sigma confidence level. Modeling our cosmic neighborhood and simulating UHECR propagation in this environment will therefore become more and more important in the coming years. This will also have to include the effects of the Galactic magnetic field and an extension to a possible heavy component of nuclei. For first steps in this direction see, e.g. (Alvarez-Muniz et al., 2001) and (Bertone et al. 2002) respectively.

1.5 EE neutrino Physics and Astrophysics

1.5.1 Neutrinos Cross Sections Astronomy at the highest energies must ultimately be performed by neutrinos because the Universe is transparent to no other known radiation. Detection of astrophysical neutrinos demands an extraordinarily large volume. EUSO will significantly increase the target volume compared with ground-based detectors, enabling exploration of the neutrino Universe. Neutrino primaries have the advantage of being well established particles. However, within the Standard Model their interaction cross section with nucleons, whose charged current part can be parametrized by (Gandhi et al., 1998)

σ SM ≅ × −32 19 0.363 2 νN ( E ) 2.36 10 ( E 10 eV ) cm

for 1016 eV≤ E ≤\la1021 eV, falls short by about five orders of magnitude to produce air showers as they are observed. This extrapolated cross-section is uncertain by a factor of a few. However, we expect it to be large enough for cosmic neutrinos to produce observable numbers of atmospheric showers in the 1013 tons of atmosphere observed by EUSO. Moreover it has been suggested that the neutrino-nucleon cross section, σνN can be enhanced by new physics beyond the electroweak scale in the CM frame, or above about a PeV in the nucleon rest frame. Neutrino induced air showers may therefore rather directly probe new physics beyond the electroweak scale. One possibility consists of a large increase in the number of degrees of freedom above the electroweak scale (Domokos & Kovesi-Domokos, 1999). One of the largest contributions to the neutrino-nucleon cross section turns out to be the production on the brane representing our world of microscopic black holes which extend into the extra dimensions. The production of compact branes, completely wrapped around the extra dimensions, may provide even larger contributions (Ahn et al., 2003). The cross sections can be larger than in the Standard Model by up to a factor ∼100 for reasonable parameters (Feng & Shapere, 2002). 19 -27 2 The UHECR data can be used to put constraints on cross sections satisfying σνN(E ≥10 eV) ≤10 cm . Particles with such cross sections would give rise to horizontal air showers which have not yet been observed. Resulting upper limits on their fluxes assuming the Standard Model cross section are shown in Fig-pc.~\ref{fig5.1}.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 25/35

Figure 1.5.1-1 Standard Model Neutrino Cross Sections

Figure 1.5.1-2 Theoretical neutrino flux upper limits based on the γ-ray bound (red thick line) derived from the EGRET diffuse γ- ray flux (shown on left part between 30 MeV and 100 GeV), the Mannheim-Protheroe-Rachen limit for optically thin sources(blue dotted line), and the Waxman-Bahcall limit (green dashed line) for redshift evolution typical for active galactic nuclei are also shown. For comparison shown are the atmospheric neutrino flux,, as well as existing upper limits on the diffuse neutrino fluxes from MACRO, AMANDA, BAIKAL, AGASA, the Fly's Eye and RICE experiments, and the limit obtained with the Goldstone radio telescope (GLUE), as indicated.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 26/35

Figure 1.5.1-3 Predictions for the differential fluxes of γ-rays (dashed line), nucleons (dotted line), and neutrinos per flavor (solid line, assuming maximal mixing among all flavors)in a top-down model characterized 14 by p=1, mX = 2 ×10 GeV, and the decay mode X→ q+q, assuming the FF in MLLA without supersymmetry, with a fraction of 10% nucleons. The calculation used the code described in Ref (Lee, 1998); and assumed the minimal URB version consistent with observation and an EGMF of 10-12 G. Measured CR, neutrino, and γ- ray fluxes are as in previous Figure where also a fit to the CR flux below 1017 eV is shown. Also shown are expected sensitivities of the currently being constructed Auger project to electron/muon- (upper dotted) and tau-neutrinos (lower dotted) and the planned projects telescope array (TA) the fluorescencevCherenkov detector MOUNT, and the space based OWL, the water-based NT200+ and ANTARES, and the ice-based ICECUBE, as indicated.

1.5.2 Super GZK neutrinos The abbreviation `SuperGZK neutrinos' implies neutrinos with energies above the Greisen-Zatsepin-Kuzmin 19 cutoff EGZK ∼5 × 10 eV. Soon after theoretical discovery of the GZK cutoff it was noticed that this phenomenon is accompanied by a flux of UHE neutrinos, that in some models can be very large (Berezinsky & Zatsepin, 1969). In 80s it was realized that topological defects can produce unstable superheavy particles with masses up to the GUT scale (Witten, 1985) and neutrinos with tremendous energies can emerge in this process (Hill et al., 1987). It has been proposed that SuperGZK neutrinos can be detected with help of horizontal Extensive Air Showers (EAS) (Berezinsky & Smirnov, 1975).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 27/35

The exciting prospects for detection of SuperGZK neutrinos have appeared with the ideas of space detection, e.g. in the projects EUSO. The threshold of neutrino detection in these experiments is about 1×1020 eV, and therefore they are ideal instruments for detection of SuperGZK neutrinos.

1.5.2.1 Sources of SGZK neutrinos The sources can be of the accelerator and non-accelerator origin. In Fig XXX some predicted fluxes are summarized. Accelerator sources- Low energy photons inside the source or outside it (e.g. CMB photons) are usually more efficient for neutrino production than gas. The problem, as discussed in parXXX is acceleration to E >>1020eV. For 21 shock acceleration the maximum energy can reach optimistically Emax ∼10 eV for protons. There are some other, less developed accelerator mechanisms (e.g. unipolar induction, acceleration in strong e-m waves), which hopefully can provide larger Emax. The large neutrino fluxes at the superGZK energies is therefore a problem for accelerator sources. The non- acceleration sources, topological defects and superheavy relic particles, involve physics beyond the Standard Model. Very high neutrino energies, in excess of the GUT scale, are possible for these sources. The fluxes are normally limited by the cascade bound. The largest flux (above the standard cascade bound) is predicted by the mirror matter model. Non-accelerator (top-down) sources. -Topological Defects and Superheavy Dark Matter (SHDM) can easily produce required superGZK energies. In both cases the superGZK neutrinos are produced in the decays of very heavy particles, unstable gauge and Higgs particles in case of TDs, and quasistable particles in case of SHDM. 16 Neutrinos are born mostly in pion decays and have Emax∼ 0.1 mX. A natural upper limit for mass is mX ≤mrGUT∼10 GeV. TDs differ substantially by mechanisms of SH particle production. They include emission of X-particles through cusps in superconducting cosmic strings. In this case the energy of X-particle is boosted by the Lorentz-factor of the cusp which can reach γ ∼ 103 - 106. In case of network of monopoles connected by strings, monopoles move with large acceleration a and can thus radiate the gauge bosons such as gluons, W±,Z0. Neutrinos appear in their decays, in case of gluons through production and decays of . The typical energy of radiated quanta is E ∼a ΓM, where ΓM is the Lorentz-factor of the monopole. In necklaces, where each monopole (antimonopole) is attached to two strings, all monololes and antimonopoles inevitably annihilate in the end of evolution, producing neutrinos in the this process (see Subsection \ref{subsec:neckl}). Vortons can decay due to quantum tunnelling, producing neutrinos. In some of the cases listed above the neutrino energies can be larger (due to Lorentz factor) than the GUT scale. SHDM is very efficiently produced at inflation due to gravitational radiation and is accumulated in galactic halos with overdensity ∼105. Neutrinos are produced at the decay of these particles mostly in extragalactic space.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 28/35

Figure 1.5.1.2-1 Cosmic neutrino fluxes from Bottom-up and Top-down models: AGN(A) and AGN(B) from Mannheim; GRB from Waxman & Bahcall (1999) and Vietri (1995) extension; GZK neutrinos from Protheroe & Johnson; TD(A) from Protheroe & Stanev (1997); TD(B) form Sigl et al. (1997), TD(C) from Sigl et al.,SM relic from Kalashev et al (2001)., Hatched region from atmospheric neutrino background

1.5.2.2 Upper limits on diffuse neutrino fluxes There are different upper bounds for the diffuse neutrino flux. Cascade upper limit Production of HE neutrinos is + - accompanied by photons and electrons which start e-m cascade colliding with target photons (γ+γtar→e +e ,e+γtar →e'+γ'). The cascade photons get into range observed by EGRET, and the energy density of these photons, ωcas, 2 ≤ π ω put the upper limit on diffuse HE neutrino flux, E Iν ( E ) ( c 4 ) cas under assumptions that neutrinos are produced in the chain of decays of particles (e.g. pions, W and Z0), where electrons or photons appear too, and that the sources are not opaque for the cascade photons. In fact cascade can develop inside a source, provided that produced cascade photons are not absorbed by gas in the source. This bound has a great generality. It is valid for photon and proton target and for any spectrum index (in this case the bound becomes up to factor 2 stronger. Waxman-Bahcall (WB) upper limit If (Waxman & Bahcall, 1999) UHE protons escape freely from a source, their flux should be less than the observed one. It gives the WB bound on neutrino flux produced by escaping protons. It is obtained for a specific shape (1/E2) of the production spectrum. This bound is much stronger than the cascade limit, but it is not valid for many sources and models (e.g. for TDs and SHDM, where production of protons is negligible, for acceleration sources at high redshift, from which UHE protons do not reach us, for clusters of galaxies, since time of CR exit from there is larger than age of universe, for some specific AGN models, e.g. the Stecker model etc). However, it is valid for diffuse flux from some interesting sources like GRBs and some models of AGN.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 29/35

Mannheim-Protheroe-Rachen (MPR) (Mannheim et al., 2001) upper bound- As compared to the WB bound, the MPR limit is valid for sources with various optical depths and for different maximal acceleration energies. It is weaker than WB bound, and for high and low energies differ not too much from the cascade limit. Since UHE protons is subdominant component for most sources of superGZK neutrinos (see next Section), the WB and MPR bounds are not valid for them, and the cascade bound becomes most appropriate.

1.5.3 Upper limits on diffuse neutrino fluxes

1.5.3.1 AGN and other accelerator sources The fluxes of UHE neutrinos produced in pγ collisions with CMB photons by UHE protons from AGN (quasars and 21 Seyfert galaxies) were calculated in (Berezinsky & Smirnov, 1975) up to Emax∼10 eV. The assumed cosmological evolution of the sources up to redshift z=14 with the evolutionary index m=4 increases strongly the flux. The cascade upper bound has been taken into account. Recently the detailed calculations of diffuse neutrino fluxes from evolutionary and non-evolutionary sources have been performed in (Kalashev et al., 1999) The acceleration 22 mechanisms are not specified, but it is assumed that generation spectra are flat and Emax can reach 3× 10 eV. Neutrinos are produced in collisions with CMB photons. The calculated fluxes agree by order of magnitude with those from (Berezinsky & Smirnov, 1975).

Figure 1.5.3.1-1 Diffuse νµ neutrino flux from the Figure 1.5.3.1-2 Diffuse neutrino flux from necklaces. The reionization bright phase at redshift z0=20 and z0=5. The curve p shows the UHE proton flux, compared with the cascade upper bound for νµ neutrinos is shown by curve AGASA data, and two curves γ show UHE photon flux for `cascade limit’. two cases of absorption.

1.5.3.2 Reionization bright phase The WMAP detection of reionization implies an early formation of stars at large redshifts up to z ∼20, able to reionize the universe. A plausible process is formation of very massive, M> 100 M , Population III stars with ~ subsequent fragmentation to SN. Two-burst scenario of reionization is plausible: at z∼15 and z ∼6 The fraction of baryonic matter in the form of compact objects is ∼0.01 at z∼10. Fragmentation of massive Pop III stars into pre- supernovae and black holes results in CR acceleration by various mechanisms (shocks, jets in miniquasars and 50 hypernovae etc). Not specifying them we shall assume that the energy release in the form of CR is Wcr∼5×10 erg/1M ,, generation spectrum is ∼E-2.1 and E =1×1013GeV. Neutrinos are produced in collisions with CMB ~ max photons. In Fig.XXX we present the calculated neutrino spectrum for two values of z (20 and 5), together with cascade upper limit for one neutrino flavour Berezinsky & Gazizov). This model is very similar to the galactic bright phase (for a review see Berezinsky et al., 1990).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 30/35

1.5.3.3 Necklaces Necklaces are hybrid TDs (monopoles connected by strings) formed in a sequence of symmetry breaking phase transition G→ H × U(1) → H × Z2 In the first phase transition the monopoles are produced, and at the second each monopole gets attached to two strings.

Figure 1.5.3.3-1 Diffuse neutrino flux from mirror necklaces Figure 1.5.3.3-2 Diffuse neutrino flx from SHDM \protect\cite{BKV,BK-MC}. The curve γ shows UHE photon flux from the halo compared with the AGASA data, and the curve N shows the UHE nucleon flux from the halo.

The symmetry breaking scales of these two phase transitions, ηm and ηs, are the main parameters of the 2 necklaces. They give the monopole mass, m∼4π ηm /e, and tension of the string, µ∼2π\ηs . The evolution of necklaces is governed by ratio r=m/µd, where d is a length of a string between two monopoles. During the evolution this length diminishes due to gravitational radiation, and in the end all monopoles and antimonopoles annihilate, producing high energy neutrinos as the dominant radiation. The diffuse neutrino flux from necklaces is given (Aloisio et al., 2003) in Fig. \XXX

1.5.3.4 Superheavy dark matter (SHDM) Production of SHDM particles naturally occurs in time varying gravitational field of the expanding universe at post- inflationary stage (Chung et al., 1999) This mechanism of production does not depend on the coupling of X- particles with other fields, and occurs even in case of sterile X-particles. The relic density of these particles is mainly determined (at fixed reheating temperature and inflaton mass) by mass mX. SHDM can constitute all CDM observed in the universe, or be its subdominant component. The range of practical interest is given by masses (3 - 10)1013 GeV, at larger masses the SHDM is subdominant. SHDM is accumulated in the Galactic halo with 5 10 overdensity ∼ 10 . In many elementary-particle models SH particles can be quasi-stable with lifetime τX >> 10 yr. Such decaying particles produce UHECR with photons from the halo being the dominant component. The measured flux of these photons with corresponding signatures (anisotropy in the direction of Galactic Center and the muon-poor EAS) determines τX experimentally. The energy spectrum of HE particles is now reliably calculated in the SUSY-QCD framework as ∼dE/E2 there is an agreement between the calculations of different groups (Berezinsky & Kachelriess, 2001; Sarkar & Toldra, 2002; Barbot & Drees, 2002) Neutrino flux from SHDM is given in Fig.\XXX according to (Berezinsky & Kachelriess, 2001; Sarkar & Toldra, 2002; Barbot & Drees, 2002). It is produced by decays of X-particles in extragalactic space, while UHECR signal is produced mainly by photons from X-particle decays in the Galactic halo.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 31/35

1.5.3.5 Mirror matter Mirror matter can be most powerful source of UHE neutrinos not limited by the usual cascade limit (Berezinskty & Vilenkin, 2000) Existence of mirror matter is an interesting theoretical idea which was introduced by Lee and Yang (1956), Landau (1957) and most notably by Kobzarev, Okun and Pomeranchuk (1966) to have a space reflection operator Is commuting with Hamiltonian [Is,H]=0. The mirror particle space is generated by operator Is which transfers the left states of ordinary particles into right states of the mirror particles and vice versa. The mirror particles have interactions identical to the ordinary particles, but these two sectors interact with each other only gravitationally. Gravitational interaction mixes the visible and mirror neutrino states, and thus causes the oscillation between them. A cosmological scenario must provide the suppression of the mirror matter and in particular the density of mirror neutrinos at the epoch of nucleosynthesis. It can be obtained in the two-inflaton model (Berezinsky & Vilenkin, 2000). The rolling of two inflaton to minimum of the potential is not syncronised, and when mirror inflaton reaches minimum, the ordinary inflaton continues its rolling, inflating thus the mirror matter produced by the mirror inflaton. While mirror matter is suppressed, the mirror topological defects can strongly dominate (Berezinsky & Vilenkin, 2000). Mirror TDs copiously produce mirror neutrinos with extremely high energies typical for TDs, and they are not accompanied by any visible particles. Therefore, the upper limits on HE mirror neutrinos in our world do not exist. All HE mirror particles produced by mirror TDs are sterile for us, interacting with ordinary matter only gravitationally, and only mirror neutrinos can be efficiently converted into ordinary ones due to oscillations. The only (weak) upper limit comes from the resonant interaction of converted neutrinos with DM neutrinos: ν +ν → 0 DM Z .I n (Berezinsky & Vilenkin, 2000) UHE neutrino flux from mirror TDs is presented according to calculations R.Aloisio, et al., 2003. This model provides the largest flux of superGZK neutrinos being not restricted by the standard cascade bound.

1.5.3.6 GZK neutrinos The most conventional process to create cosmic EE neutrinos is the GZK cut-off itself through the ∆+→π++…→ν+…decay sequence. In Fig… the GZK neutrinos spectra under different assumptions for EECR source is shown (Yoshida & Teshima, 1993). The flux of neutrinos above 1019 eV strongly depends on the maximum EECR spectrum. The GZK neutrinos flux could provide useful information on the turn-on time, evolution parameter and the maximum energy of primary EECR.

Figure 1.5.3.6-1 Eenergy spectra of EHE neutrinos produced by the GZK mechanism for energies of cosmic rays of 1021 and 1022 eV (from Yoshida & Teshima, 1993).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 32/35

1.5.3.7 τ neutrinos EUSO has the possibility to detect τ neutrinos (ντ) at even much lower energies by measuring the light from the upward showers they produce. Tau neutrinos interact near the Earth’s surface after penetrating the whole Earth and produce τs that exit the Earth and decay in the atmosphere. While electron neutrinos (νe) and mu neutrinos (νµ) are fully absorbed by the Earth at energies >1014 eV, a ντ will regenerate through the Earth. Regeneration is an inevitable consequence of the repeated charged current processes ντ+N→τ+X and τ→ντ+X. The end result is to produce an emerging upward shower of energy, 1015–1018 eV. Above the energy threshold of ~1∞1015 eV, the EUSO instrument can detect collimated beams of Cherenkov light emitted in a narrow cone by these upward showers. Earth-skimming neutrinos are another class of neutrino-initiated showers that have been recently discussed (Weiler T., 1982; Weiler T., 1999; Fargion et al., 1999, Bottai & Giurgola, 2002) These neutrinos graze the Earth and travel through a small column density of Earth in which they interact. The resulting shower emerges into the atmosphere. The rate of such Earth-skimming events actually grows with a decreasing cross section, as 1/σνN. If EUSO, viewing the horizon, can detect Earth-skimming neutrinos then EUSO can measure the neutrino-nucleon cross section from the angular dependence of the Earth-skimming rate. Such a measurement would establish the value of this cross section at high energies.

Figure 1.5.3.7-1 Schematic drawing of atmospheric showers induced by neutrinos. Upward ντ . could be detected as upward going showers induced by τ decay.

Figure 1.5.3.7-2 Schematic drawing of two different observational techniques forupward neutrinos events: fluorescence based and direct Cherenkov detection.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 33/35

1.6 Atmosphere Science By using the atmosphere as a natural detector for the fluorescence induced by ECCRs, EUSO on the ISS will systematically survey a large portion of space providing a complete dataset of UV measurement in the 300-400 nm band for at least 3-years. Besides the Cosmic Ray events, this dataset will contain recorded information about all atmospheric UV-phenomena that occurred during the observation time (i.e. meteors, lightning, elves, sprites, blue jets …).

1.6.1 METEORS The term “Meteors” is generally referred to describe the whole phenomena associated with the entry into the Earth’s atmosphere of a solid particle from the interplanetary space. Aim of this note is to give basic general information as a support to investigate the potentiality of EUSO to detect meteors by looking at the streak of UV signal resulting from the interaction of a solid particle with air when it enters into the Earth’s atmosphere. The inflowing particles are both of natural origin (meteoroids from primordial interplanetary matter or deriving from fragmentation of asteroids and comets, Grun et al., 1985) and man made origin (space debris). Meteoroids have mass ranging from 10-21 kg to 1015 kg and enter the top of the Earth’s atmosphere with velocity between 11.2 and 72 km/s with a flux-mass relation that cannot be expressed by a simple power law ( see figure).

Figure 1.6.1-1 Flux (number of object /year over the whole Earth surface) as a function of mass (Ceplecha: Influx of Interplanetary bodies onto Earth. Astronomy and Astrophysics, 263; 361 – 366 – 1992.)

Meteoroids that retain the orbit of their parent body ( process of fragmentation ) can create periods of high flux and are called “ streams”(periodic meteor showers, Jenniskens P., 1994¸ Cour-Palais B.G., 1969), they are highly directional and are limited to specific time of the year. Random fluxes of solid particles with no apparent pattern are called “ sporadic “ component of the meteoroids influx (sporadic meteors) and constitute a major part of the whole complex of the interplanetary bodies. Due to their origin, space debris particles have the following properties: the flight direction is nearly parallel to the Earth’s surface, the flux depends on the altitude (EUSO ~400 km) and the debris environment evolves with time. Most debris has circular orbits. In order to estimate the flux of particle that EUSO on the ISS is expected to detect, implemented models for both meteoroids and space debris can be used (Divine N., 1993; Staubach P., et al. 1996, Anderson, 1991; Kessler, et al. 1996), keeping in mind that mostly all of them contain several approximations and uncertainties mainly resulting from uncertainties in knowledge of particles densities and masses. Therefore the necessity to collect more consistent sets of experimental data about meteoroid mass/velocity distribution which in turn is one of the expected outcome of EUSO observing meteors.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 34/35

1.6.2 Transient Luminous Phenomena. Electro-optical phenomena often occur in the lower atmosphere such as storms and their associated lightning and in upper atmosphere such as elves, sprites and blue jets. Due to their transient aspect, their luminosity and their locations they can appear as fake Cosmic Ray EAS events. They can have a non negligible influence on the Euso duty cycle and on the trigger efficiency.

1.6.2.1 Events in Lower Atmosphere: Lightning Among luminous phenomena, lightning are by far the most abundant. They propagate over several kilometers at the speed of light and can mimic a cosmic shower as far as Euso is sensitive to the fast component of the emitted flash light. Lightning usually occur between the bottom of clouds and the ground, from space observation they should be attenuated and diffused inside the cloud and should appear as faint diffuse flashes. However around every ten negative lightning, one positive lightning occurs between the top of the cloud and ground. This positive lightning are much more intense due to accumulation of charges at the top of the cloud, and the emitted light is much less attenuated by the cloud itself. The global distribution of lightning on Earth was studied by the Optical Transient Detector (OTD) satellite and is shown for the year 1999 on the next figure. One can observe that lightnings are by far much more abundant above land than above oceans . The observed frequency is estimated at 3.2 Lightning/km2/year. Due to its large FOV, this number can be converted in 1600 lightnings per day in EUSO ! These lightning will be mainly concentrate above the equatorial landscape, where several tens of flashes can appear simultaneously inside the field of view. This number has to be modulated by the day/night variation of lightning . Taking into account the Euso duty cycle at 12%, the number of expected observation will be ~200 lightnings per day in the FOV.

Figure 1.6.2.1-1 Lightning frequency from satellite OTD.

1.6.2.2 Events in upper atmosphere : Elves, Sprites and Blue Jets. Transient luminous phenomena , which develop in the upper atmosphere up to 100 km in altitude , were observed recently (early 90’s ). These phenomena are extremely elusive, lasting from few micro-seconds to hundred of milliseconds, difficult for visual observation. They are characteristic of a considerable amount of energy transfer

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 35/35

from the low atmosphere to the ionosphere and beyond, where up to now the coupling was considered to be substantially weak. These phenomena appear in various shape which are classified in three main categories: Elves, Sprites and Blue Jets; associated phenomena (beads , fingers…etc) can be added. Other phenomena such as Terrestrial Gamma Ray Bursts observed by the satellite GRO/BATSE looking at Nadir above clouds, can have a common origin. Elves, historically discovered first, are somehow the optical counterpart of the extreme low frequency (ELF) perturbations in the radio-wave propagations around earth and known since a long time. Elves, observed at an altitude of 100 km, look like a circular ring expanding to hundreds of kilometres with the speed of light. These are extremely luminous with a time scale of the order of 1 millisecond. It looks like a brief illumination of the ‘Airglow’. Blue Jets are slightly conic beams developing at a speed of 100 km/second starting from the top of clouds upward to 30~40 km. Sprites develop from the top of clouds up to 100 km with an optical energy of 10-50 kJoules.

1.6.3 Atmosphere as a Physical System . Continuous operation of the LIDAR, mainly due to requirements of thermal stability for the laser source, gives the opportunity to acquire climatological data useful for the statistics of the effective frequency of occurrence of the observed events, and it could be also utilized to retrieve additional information of local atmospheric conditions from successive passages.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/8

2. THE EUSO OBSERVATIONAL APPROACH...... 2 2.1 SHOWER GENERATION, FLUORESCENCE AND ČERENKOV PHOTONS ...... 2 2.2 A SCHEMATIC VIEW OF THE EUSO DETECTOR...... 4 2.3 THE EUSO EFFICIENCY ...... 5 2.4 THE ACCEPTANCE OF EUSO WITH AND WITHOUT CLOUDS ...... 6 2.5 THE DUTY CYCLE AND THE EUSO COUNTING RATES ...... 7 2.6 THE EUSO OBSERVATIONAL METHOD AND THE NEED FOR AN ATMOSPHERIC SOUNDING...... 8

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/8

2. The EUSO Observational Approach The use of the fluorescence (plus Čerenkov) production as a measure of the characteristics of an UHECR has already been used and will be used by future ground-based projects. The observational approach of EUSO, which looks at the atmosphere from a space-based telescope, placed on the International Space Station ISS, contains some peculiarities with respect to ground-based approaches. They deserve to be outlined, to stress the different problems and opportunities which arise from them. Since the ISS covers the whole Earth surface in the latitude range ±51° approximately, and moves at a speed of ~ 7 km/s, the variability of the scene seen by EUSO is much higher than that observed by a ground-based experiment. The optics aperture, moreover, implies that a portion as large as ∼200 000 km2 is observed at once. In such a large field of view, the atmosphere status and hence the expected detection acceptance can have a certain degree of non-homogeneity, whose effect has to be considered and evaluated. This is by far the largest complication a space-based detection approach has to face, with respect to the ground-based experiments. On the other hand, the use of a space-based detector, has a number of very significant advantages: • The non-proximity of the EUSO detector to the shower considerably diminishes the severe problems associated to the determination of the solid angle and to the differential attenuation of the atmosphere traversal suffered by the UV light, also within the same shower.

• The near-constant fluorescence emission rate at any height below the stratosphere allows to make simple and justified assumptions on the relationship between the energy and the fluorescence yield at the shower maximum as well as regards the relationship between the time width of the shower and the altitude at which it is produced.

• The observation of the fluorescence from above allows the method to be much less sensitive to the presence of most aerosols that are limited to altitudes below the atmosphere boundary layer. These effects will be discussed in more details in Chapter 6, in connection with the actual baseline for the EUSO detector. The present chapter will present an overview of the observational approach used by EUSO with the aim to show that, with the above advantages as well as others, the EUSO telescope will provide a powerful tool to tackle the demanding and exciting physics of UHECR.

2.1 Shower Generation, Fluorescence and Čerenkov Photons The observational approach of EUSO is based on the measurement of fluorescence and Čerenkov photons produced by the UHECR shower as it progresses through the atmosphere. A hadronic UHECR (radiation length ≈ 40 g cm-2 at E = 1020 eV) penetrating the earth’s atmosphere will interact promptly and generate an Extensive Air Shower (EAS). After a first interaction, a shower of secondary particles will be produced. These secondary particles, whose number (>1011) is proportional to the energy of the primary UHECR, will deposit their energy in the atmosphere. Figure 2.1-1 shows the expected shower profile, as a function of the traversed depth in atmosphere. A fraction of the total energy carried by the charged particles (≈ 0.5%) will be converted into fluorescence photons through the excitation of N2 molecules. A highly beamed Čerenkov radiation is generated as well by the ultrarelativistic particles in the shower. Figure 2.1-2 shows the expected amount of UV light accompanying an EAS induced by a 1020 eV primary proton. The fluorescence light is emitted isotropically by the N2 molecules along the shower’s path. The fluorescence yield depends on the local pressure and temperature, but appears rather constant with the altitude, below 15 km, as it will more diffusely discussed in Section 6.1.2. The EUSO observational approach mainly relies on the fact that, thanks to the huge amount of emitted light, a fraction (~ 10-11) of these photons will reach a light-collecting device of few squared meters, placed outside the atmosphere, at the ISS height of ∼400 km, looking downward to the Earth at night, as schematised in Figure 2.1-3. Typically, for a 1020 eV shower, a few thousand photons will reach the EUSO detector. As the EUSO telescope has a Fresnel lens system associated to a fast counting, pixelized focal surface, EUSO will detect not only the number of photons arriving but also their direction and time of arrival. It is the observation of this specific space-time correlation that identifies, very precisely, the presence of an UHECR shower.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/8

Figure 2.1-1 The profile of a vertical shower, as a Figure 2.1-2 The expected UV light accompanying an function of the slant depth, as generated by GIL (full EAS generated by a 1020eV primary proton, 45° zenith thick line, and by CORSIKA (10 events superimposed- inclined: a) the UV Čerenkov photons; b) the light full lines). fluorescence UV photons emitted in 25 m steps along the shower path.

Figure 2.1-3 Artistic view of the EUSO Figure 2.1-4 Time profile of photons. The green curve observational approach: “doing astronomy shows the fluorescence photons and the blue line the looking downward the night Earth atmosphere”. diffused Čerenkov photons. The peak at 1540 µs is the reflection off the Čerenkov photons by the sea albedo. The dotted black curve shows the fluorescence profile if atmosphere transmission effects are neglected.

Figure 2.1-4 shows the time profile of the photons reaching the EUSO optical system for a typical shower (E 20 shower =10 eV, θshower = 60° at nadir from EUSO). It displays the typical features of such showers: a (fluorescence) bell-like curve showing a maximum at an altitude at which the particle production of the shower is maximum. This

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/8

figure reveals one of the major advantages of the fluorescence technique: its ability to describe the longitudinal development of the shower. Ground-based array detectors only observe those particles that reach their location. It is important to notice that, apart from the incident energy, the integrated number of photons and the time duration of the shower will also depend on the shower angle. A vertical shower will develop rapidly (∆t ∼ 100 µs) and, in some case, the maximum of the shower development (Smax ) and hence of the fluorescence yield will not be reached before ground. Conversely, a quasi-horizontal shower, at high altitude, will develop over a long track and its width in time will be much larger, possibly reaching values ∆t ≥ 300 µs. Because the velocities of the secondary shower particles are higher than the local velocity of light, Čerenkov emission is also present. This emission is focused (within a cone of ~ 1.3° radius) along the shower axis and will be visible by EUSO through two effects. The first effect is the possible diffusion of the Čerenkov photons by the atmosphere’s molecular and aerosol content, i.e. the Rayleigh and Mie large angle scattering processes. The second effect is due to the “albedo” of the ground surface, viewed as a rough discontinuous surface, as far as the refraction index is concerned. A similar effect can also be due to the presence of clouds which will act as an efficient reflective layer. Depending on the optical depth of the clouds, the effective albedo can reach up to 80%. In the case of a water surface, the albedo is typically of the order of 5 – 8 %. In Figure 2.1-4, the narrow peak to the right of the broad fluorescence peak corresponds to the Čerenkov photons reflected from the sea. The Čerenkov process corresponds to a large band in wavelength (≈ 200 - 500 nm) and has a λ−2 wavelength dependence.

2.2 A Schematic View of the EUSO Detector In order to observe these photons, the EUSO detector is constructed as large field-of-view, UV sensitive, ultra-fast camera, as it will be widely described in Section 5.

Figure 2.2-1 Schematic view of the EUSO main detector, Figure 2.2-2 Time profile of photo-electrons (p.e.). the EECR/ν telescope. This spectrum is calculated from the one of Fig.2.1-4 by taking into account the photon → p.e. conversion of the detector.

Figure 2.2-1 gives a schematic description of the main EUSO detector, the EECR/ν telescope: a light weight double Fresnel lens optical system which focuses the photons upon a focal surface covered by Multi-Anode-Photo- Multipliers (MAPMT). The focal surface is thus covered by approximately 200 000 pixels. Each pixel images, at ground level, an area equivalent to ~ 0.8×0.8 km2. Each pixel is associated to a fast counting electronics. Each MAPMT is also associated to an analog electronic chain that extends the dynamical range of the electronics, necessary in the case, for example, of the detection of large Čerenkov yields. The time sequence of the electronics is defined by the Gate Time Unit, GTU. The GTU defines the time unit during which the fast front-end electronics will count the arriving photoelectrons (Np.e), and represents the resolution

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/8

time for the fluorescence signal. As discussed in Section 5, the typical EUSO GTU is of the order of few µs (2.5 µs in the current baseline design). With such a value, the detection of a shower will typically consist of ~ 50 GTU. The acquisition system is controlled by a trigger system that takes advantage of the signal profile and its space-time correlation. Two main parameters control the trigger scheme. The parameter Nthre sets the level at which a given pixel will be considered “hit”: if Np.e. ≥ Nthre in one GTU, the corresponding pixel is considered to be activated. If, for Npers successive GTU, one or more pixels are activated, a valid shower will be identified by the trigger system. Figure 2.2-2 shows the p.e. time spectrum obtained from the photon spectrum given in Figure 2.1-4; one can read Nthre along the vertical axis and Npers as the number of consecutive bins above Nthre. The EUSO trigger scheme takes therefore advantage of the very special space-time correlation that is produced by the shower. The random photon noise background, produced by natural (starlight, airglow of moonlight) or anthropomorphic sources do not possess such specific space-time correlations and will therefore be eliminated, with a certain efficiency, by this trigger scheme, as it will be discussed in Chapter 6.

2.3 The EUSO efficiency The efficiency of EUSO will depend on the energy of the showers. Figure 2.3-1 shows a typical EUSO efficiency curve as a function of the shower energy in clear sky as well as in cloudy conditions (trigger setting: Nthre = 4, Npers= 4; see below for the definition of the cloudy condition). Three energy ranges can be identified. At low energy (E ≈ 1019 eV) showers are not identified because the number of produced photons times the photon to p.e. conversion (≈ 0.06) is such that the shower signal is dominated by the background noise. As energy increases the signal overcomes the background and more and more showers are identified by the trigger scheme. At the highest energies, the signal-to-noise ratio poses no problems to the trigger and all showers are observed. The residual inefficiency is due to those vertical showers whose the fluorescence maximum is not reached before ground. This type of efficiency curve can be generated for various trigger parameters and experimental conditions. This will be studied in more details in Section 6.1.4.

Figure 2.3-1 EUSO efficiency in clear sky and cloudy conditions. The trigger conditions correspond to a threshold of 4 and a persistency of 4.

It is important to notice that the shape of such efficiency curves will also depend on the shower angle. Figure 2.3-1 has been generated by a statistical ensemble of showers. If the ensemble is restricted to showers of a specific angular bin ( θ > 60°, for example), the efficiency curve will be shifted to lower energy. Thus one observes,

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/8

once again, that EUSO characteristics such as efficiencies and energy threshold, for example, are dependent not only on the detector and atmosphere parameters but also on the shower properties.

2.4 The Acceptance of EUSO with and without Clouds One of the most important parameters that will define the EUSO efficiency is the atmosphere situation: cloud and aerosol content. As the EUSO detector will cover a wide range of longitudes and latitudes, a study of the statistical effect of the atmosphere conditions can only be based on a precise statistical knowledge of the atmosphere conditions. This can be done by making use of a variety of atmosphere (cloud) databases. Such a database is given by the ISCCP (International Satellite Cloud Climatology Project). This database provides, since 1983, and in 3 hours steps, the cloud properties as measured by a set of satellite-based observations, covering the entire Earth with a segmentation in 280×280 km2 cells, obtained dividing the globe in 2.5° large latitude steps and variable longitudinal steps. Typical cloud properties available are: local fractional cloud amount, height, optical depths,… Figure 2.4-1 shows the acceptance of EUSO, as a function of shower angles, in the case of clear sky conditions. The dotted red curve gives the maximum geometrical acceptance. The black histogram gives the acceptance for showers that trigger EUSO and for which the fluorescence maximum (Smax) and the Čerenkov peak is observed with a signal-to-noise ratio S/N ≈ 3 and 2, respectively (so called golden showers). The red histogram shows those showers for which the Smax is not observed and the green histogram those showers for which the Čerenkov peak is not seen. The acceptance for golden showers is seen, for a shower energy of 1020 eV, to be about 80 % of the total acceptance (duty cycle × 608 000 km2 sr). Such a plot is constructed using approximately 60 000 randomly generated showers.

Figure 2.4-1 EUSO acceptance in clear sky conditions Figure 2.4-2 Same as figure 2.4-1 but for the ISCCP for a proton shower of 1020 eV. The black curve shows cloud conditions. the acceptance (a.u.) of well detected showers. The dotted red curve gives the maximum geometrical acceptance. The red histogram shows events when only the Čerenkov is well detected. The green histogram shows events when only the fluorescence is well detected.

When ISCCP cloud conditions are randomly introduced, the acceptance is reduced as shown by Figure 2.4-2. This reduction is most important for vertical showers because the presence of clouds will, in most case, hide Smax from observation. For inclined showers (θshower > 60°) the effect is small because these showers produce their maximum at altitudes above the average cloud top height. The resulting overall acceptance, for golden showers, is

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/8

reduced, at this energy, to 50 % of the geometrical acceptance. Note that the red curve in Figure 2.3-1 shows the energy evolution of the EUSO efficiency in cloudy conditions. The effect of clouds does not vary so much with the shower energy because the Smax altitude depends logarithmically on the primary energy. To first order, cloud effects can be considered as a shower energy independent effect. A more detailed discussion of the effect is given in Section 6.2.2.

2.5 The Duty Cycle and the EUSO counting rates In order to predict the shower counting rate, the acceptance must take into account the duty cycle, i.e. the fraction of time during which shower detection will be possible. Apart from operational limitations (not considered in this section), the duty cycle is essentially determined by the amount of background photons. This background is determined by starlight, airglow and moonlight (see Section 6.1.3). During moonless nights the background is estimated (measured) to be ∼ 500 photons m-2 ns-1 sr-1, corresponding to ∼ 0.3 p.e./µs per pixel at the focal surface level. The duty cycle corresponding to moonless nights (moon below the horizon) is 12.6 %. Figure 2.5-1 shows the duty cycle as a function of the number of added background due to moonlight. One observes that if a 20 % increase in the background is accepted (~ 100 photons m-2 ns-1 sr-1 from the moon), the duty cycle reaches 25 %. In the following calculations, this value of 25 % will be used as a realistic estimate of the duty cycle. It should be noted that the presence of clouds will influence the background level. This effect will be estimated in Section 6.1.3. Using the acceptance obtained from the ISCCP cloud effect and the above duty cycle, estimates of the EUSO counting rates can be given. For these calculations, two hypotheses have been used. The first one assumes that a Super-GZK spectrum is observed corresponding to a power law exponent of –2.7 and normalized to a value of 3.6×10-33 (eV-1 m-2 s-1 sr-1) for E = 1019 eV. The second assumption is one with a GZK spectrum. The energy profile of this GZK is taken from the work of V. Berezinsky et al., 1990.

Figure 2.5.1 EUSO duty cycle, as a function of the Figure 2.5.2 “GZK” (blue) and “Super-GZK” (red) moon acceptable light differential counting rate spectrum.

Figure 2.5-2 shows, in red, the differential spectrum observed in the Super-GZK case for a 3 year counting period. The number of showers observed above 1020 eV is >4000. Between 3×1019 eV and 1020 eV, more than 9000 showers are expected to be observed with a mean efficiency of 50 %, not including the cloud efficiency effects. The same figure shows (in blue) the differential counting rates observed for the GZK case. In this situation, one observes that the GZK decrease can be precisely measured as well as the GZK recovery. The number of events observed in this situation for energies above 1020 eV is >400 ( ∼7000 at E < 1020 eV). One can therefore conclude that, whatever the final answer given to the AGASA-Hires controversy, EUSO will be able to bring valuable and abundant data to the study of this physics.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 8/8

2.6 The EUSO Observational Method and the need for an Atmospheric sounding. The general EUSO observational approach, described in the previous sections, depends on the altitude at which the shower develops, on the atmospheric parameters and, in a first approximation, on the possible presence of clouds. This has led the EUSO collaboration to study a number of atmosphere sounding devices: a stand-alone LIDAR system (baseline), a laser coupled to the EUSO detector, a wide angle Infra-red camera (options). Aim of these devices would therefore be to monitor the cloud situation at the moment and location of the shower detection; this would allow to correct the EUSO acceptance due to cloud presence and then to help in the evaluation of the altitude of the shower development. The atmosphere sounding devices, listed above in order of decreasing complexity, have an impact of the resources (power, mass and financial) allocated to the shower detection device (the EUSO EECR/ν telescope). For these reasons we also raised and investigated the answer to the following question: to which extent can EUSO be qualified as a self-sufficient atmospheric detector, able do make an auto-diagnosis of the atmospheric parameters, down to the accuracy needed for its own scientific objectives? The baseline option, i.e. a stand-alone LIDAR, will be extensively discussed in Chapter 3 from the observational point of view and showing its expected performances. The technical issue of equipping EUSO with a stand-alone LIDAR will be exploited in Chapter 5, where the Instrument is described. The potentiality of EUSO as a self-sufficient atmosphere detector capable of monitoring the cloud status and of retrieving the correct shower parameters will be discussed in Section 6.4, in connection with the experimental performances that the actual design of the EUSO detector will allow. It is however worth to mention here the features that give to the observation of UHECR from space, as it will be done by EUSO, several advantages connected to the impact of the atmospheric variability, and mainly to the presence of clouds: i. For most of the observed showers, the shower development takes place in the higher portion (> 3-5 km) of the atmosphere where cloud and aerosol perturbation (Mie scattering) are smallest. This is NOT the case of ground based fluorescence observations of UHECR which are very sensitive to aerosol below the atmosphere boundary layer.

ii. The transparency, at these altitudes, is high and does not depend critically on a precise knowledge of the altitude. For example, the Rayleigh transmission factor is of the order of 0.5 and varies by less than 10 % per km (altitude).

iii. The presence of an optically thick cloud (a large majority of the clouds present in the lower atmosphere) signals itself by a precise signal: a clear increase (3-10 times) of the reflected Čerenkov signal, embedded in the fluorescence peak, and a strong suppression of the photon yield at lower altitudes.

iv. Most of the effects to be expected from the presence of clouds or aerosol would translate into an underestimation of the UHECR energy and NOT an overestimation. The case of high altitude sub-visible clouds (optical depth ≤ 0.1) will, for example, decrease the atmosphere transmission by ≈ 30 % and will not lead to an over-prediction of the energy.

v. The main physics goal of EUSO is the study of UHECR with energies above 1020 eV where photon statistics are important and the shape of the signal well defined. Lower energy showers will mostly be used for comparison with Auger results. In this case, a statistical correction for cloud and aerosol effects may be sufficient.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/13

3. THE ATMOSPHERE SOUNDING: SCIENTIFIC RATIONALE TO HAVE A LIDAR COUPLED TO THE EUSO DETECTOR ...... 2 3.1 SCIENTIFIC RATIONALE...... 2 3.1.1 The signal source: the fluorescence yield and the Čerenkov light of an EAS ...... 2 3.1.2 The clear atmosphere assumption and its ideal transmission properties ...... 3 3.1.3 The real atmosphere: single event feature distortion and the LIDAR rôle ...... 4 3.2 EUSO-AS OBSERVATIONAL APPROACH...... 6 3.2.1 LIDAR with one laser wavelength ...... 7 3.2.2 LIDAR with more than one laser wavelength ...... 7 3.3 FLUORESCENCE PROFILE RETRIEVAL...... 9 3.3.1 Retrieval in the presence of high cirrus clouds ...... 9 3.3.2 Retrieval in the presence of mid-low aerosol/cloud layers...... 10 3.4 CONCLUSIONS...... 12

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/13

3. The Atmosphere Sounding: Scientific Rationale to have a LIDAR coupled to the EUSO Detector The EAS induced by EECR/ν produces an intense Čerenkov beam, collimated around the shower axis, whose reflection/diffusion can be detected superimposed to the fluorescence signal. The transport of both the fluorescence and of the Čerenkov signal is conditioned by the atmospheric conditions. This section (described in unabridged form in the supporting document EUSO-AS-REP-002-1) addresses the issue of directly monitoring the atmosphere acting as the active target of EUSO, in order to retrieve systematically the atmospheric transmission/scattering properties along the path of observation from the EAS trajectory to the EUSO receiver. This information is needed for the correct interpretation of the fluorescence and scattered Čerenkov signal profiles in the retrieval of the identity and energy of the primary particle. The presence of cloud systems and aerosol, as a function of their vertical distribution and optical depth, can affect the acceptance of the space-time intensity profile of the signal reaching EUSO. The EAS development and fluorescence yield are expected to be characterized by intrinsic fluctuations that limits the retrieval of the energy of the primary particle to an accuracy of the order of 15 %. The degree of precision with which the atmosphere condition has to be known, at the level of event by event in space and time for the EUSO observations, has to be driven by the final goal of retrieving the correction factor due to the atmospheric perturbation with an accuracy of the same order or better. The intrinsic scientific interest, for the Atmospheric Physics community, of a detailed monitoring of the Earth atmosphere viewed by the ISS should be addressed and duly considered even if it is beyond the goal of this study.

3.1 Scientific Rationale The knowledge of the statistical aerosol/cloud coverage and of their vertical distribution and optical depth, over the entire mission duration, will allow the evaluation of the effective statistics of the EECR/ν events observed by EUSO, in terms of frequency of occurrence of the observed events with respect to the whole effective observation time of the mission: this capability will be limited not only by the daylight and moonlight conditions, but also by the occurrence of real atmospheric conditions that may interfere with the detection of the event. The distortion of single event measurements in terms of energy value reconstruction, primary particle identification (height of shower maximum), as a consequence of the distortion induced on the observed fluorescence and Čerenkov signal profiles by the real atmospheric conditions (transmission/diffusion variability, due to aerosol/clouds distribution and optical depth), can introduce significant errors in the shower parameters determination . “Local” knowledge of the scattering and light absorption properties of the atmosphere where the EAS occurred can be provided by means of a multi wavelength LIDAR system.

3.1.1 The signal source: the fluorescence yield and the Čerenkov light of an EAS In Figure 3.1.1-1 the simulated vertical intensity profiles of Čerenkov (left) and fluorescence (right) signals generated by UHECR of different energy and trajectory are reported. These profiles have been calculated from a model and express the number of photons, generated by EAS within the spectral band of acceptance of EUSO (300 - 400 nm) as a function of the height of penetration of the shower, considering a standard atmosphere without extinction.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/13

Figure 3.1.1-1 Simulated vertical intensity profiles of Čerenkov (left) and fluorescence (right) signals generated by UHECR of 1020 eV and 1021 eV energy of primary particle and for 0° and 60° incidence zenith angle of the shower.

3.1.2 The clear atmosphere assumption and its ideal transmission properties Several simulation have been carried out to evaluate the signal strengths that should be acquired by EUSO and by a LIDAR, following an event generated by the passage of an UHECR particle through the atmosphere. As a first step the profiles of the fluorescence, scattered Čerenkov and back-scattered laser as they should be observed by EUSO, have been calculated considering standard atmospheric profiles of pressure and temperature. According to the EUSO main instrumental parameters, assuming to utilize EUSO also as a UV LIDAR receiver with a 10 % overall detection efficiency and considering a laser source of 100 mJ of energy per pulse for the LIDAR, the fluorescence (green curves), scattered Čerenkov (violet curves) and scattered laser (blue curves) observed, expressed as the number of photoelectrons detected as a function of height above the sea level (a.s.l.), are reported in Figure 3.1.2-1. In the same plots the profiles of the collective UV signal of fluorescence plus scattered Čerenkov (orange curves) are reported, as they will be simultaneously acquired by EUSO. Figure 3.1.2-1 The expected number of photoelectrons in Comparing the orange curves (total acquired 20 signal), to the corresponding green curves EUSO for an UHECR of energy 10 eV (proton), (fluorescence only), the shift in the position and respectively vertical (left) and inclined 60° (right) and for a intensity of the maximum, due to the scattered UV laser source of 100 mJ. Čerenkov contribution, is clearly evident. So the retrieval of Čerenkov contribution along the observed signal profile, and not only the Čerenkov landmark, should be taken into account for the determination of the position and intensity of the maximum of fluorescence. Also note the intensity of the back-scattered laser signal: even if based on a single shot acquisition, it is about two orders of magnitude higher than the scattered Čerenkov signal at sea level and several order of magnitude higher than the scattered Čerenkov at upper altitudes. While the scattered Čerenkov signal can be observed only from the lower altitudes up to a few kilometres, the backscattered laser signal is strong enough to be detected from an altitude in excess of 20 km. This demonstrates the advantage resulting from the implementation of a LIDAR system in EUSO to retrieve event-by-event atmospheric transmittance profiles at least up to the tropopause level, where aerosol/cloud structures of relevant optical thickness could be present, (but not revealed by the Čerenkov echo ). The fluorescence to scattered Čerenkov ratio is as large as 2 at about 2 km a.s.l. and about 1 at sea level, for vertical showers.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/13

The effect of the increasing influence of the scattered Čerenkov is more evident for slant trajectories, where the same values of the fluorescence on scattered Čerenkov ratio are encountered at about 4 km a.s.l. and 3 km a.s.l. respectively for a slant 60° trajectory of the shower. In clear sky conditions it should be possible to separate the scattered Čerenkov contribution to the observed signal from the fluorescence using standard atmospheric models as reference, (this means that it should be possible to unambiguously determine the reality of such conditions). Instead, if aerosol/cloud structures are present it could be very difficult, if not impossible, to extract the fluorescence profile from the total observed signal, without an independent measurement of the atmospheric conditions that allow to retrieve the vertical distribution and optical properties of such structures, in fact the final amount of the observed signal is determined by the extinction/scattering properties of the medium as a whole (molecules plus aerosol/clouds). A LIDAR system is indeed the ideal instrument to do it. When starting to analyze the profiles acquired by EUSO, the primary particle is unknown (e. g. the source signal intensity as generated by the shower is unknown), and the fluorescence and scattered Čerenkov components are mixed together; if also the transmission/scattering properties of the atmospheric region interested by the shower development are unknown, any approach to retrieval, even if based on a statistical assumption about the atmospheric conditions, could be affected by a large amount of uncertainty, due to the possibility of an extreme discrepancy of the statistical assumption from the real occurred condition at the time of the observation. In the next paragraphs several simulation outputs will be shown to demonstrate how the possible presence of aerosol/cloud structures, if not revealed and duly considered, could introduce a large systematic error in the identification and energy retrieval of the primary particle.

3.1.3 The real atmosphere: single event feature distortion and the LIDAR rôle Based on available meteorological statistics, the annual average clear sky frequency (report on visual clouds only) is 18 % over land and 3 % over seas. The total cloud cover amounts on the average to 65 % over ocean areas and to 52 % over land. Similar results have been obtained from ISCCP (68 % and 51 %, respectively). Moreover there is a high probability of simultaneous occurrence of aerosol layers and different cloud types at different heights in the atmosphere. If we start to consider real atmospheric conditions, instead of standard clear sky models, all the effects of distortion (mentioned in the introduction) on the observed fluorescence and scattered Čerenkov signals, in particular on the height location and intensity of the maximum of fluorescence, are still present and are generally enhanced. A representative set of simulations have been carried out to estimate the influence of the real atmosphere, instead of a standard one, in the shape and intensity of both fluorescence and scattered Čerenkov signals, as they will be acquired by EUSO. This has been done introducing real aerosol/molecular backscattering ratio profiles acquired both by airborne (and ground-based LIDAR systems) as input parameters in the simulations. The following figures show results that can be directly compared with the analogue results for clear sky condition shown in Section 3.1.2. As an example, Figure 3.1.3-1 depicts profiles similar to those shown in Figure 3.1.2-1, but obtained introducing in the calculations the real atmospheric conditions retrieved from observation made from a ground- based LIDAR station at Lampedusa, Italy. They have been obtained in presence of a Saharan dust layer of optical thickness 1, extending from the ground to about 6 km height a.s.l. The distortion due to the combined effects of the additional scattered Čerenkov signal and of the aerosol extinction on the observed fluorescence signal intensity modifies the observed profile from the green curve to the orange curve. Comparing for example the profiles obtained in the real case for vertical incidence with the one corresponding to clear sky conditions, there is a shift of ~2 km in the height location of the observed maximum of fluorescence, that corresponds to more than 100 g cm-2 of atmospheric depth; furthermore there is a variation of about 40 % in the intensity of the observed maximum of fluorescence. All these effects will produce a significant systematic error in the retrieval of the identity and the energy of the primary particle. Furthermore the situation depicted in Fig. 3.1.3-1 shows a typical case in which the retrieval of the aerosol extinction from a LIDAR measurement may be affected by large uncertainty if the backscattered laser signal is acquired at only one wavelength, since these data will not make it possible to retrieve the aerosol size distribution and consequently their extinction/backscattering properties (see later). The knowledge of these parameters is necessary both to evaluate the cumulative aerosol plus molecular extinction effect and to extract the scattered Čerenkov contribution from the observed signal and reconstruct the fluorescence signal profile. Generally the presence of aerosol/clouds results in a larger amount of scattered radiation when compared to the purely molecular

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/13

situation, so the distortion induced on the fluorescence profile by the scattered Čerenkov radiation in presence of aerosol/clouds is more important. This demonstrates the importance of knowing the scattering properties of the atmosphere for each observation, in order to reconstruct the unperturbed fluorescence profile, avoiding wrong assumptions on the atmospheric conditions from statistics or standard models.

Figure 3.1.3-1 The expected number of photoelectrons in EUSO, generated by an UHECR of energy 1020 eV (proton), respectively vertical (left) and inclined 60° (right). UV photons are transported from the source to the detector in a possible real atmospheric situation, with a Saharan dust layer of optical depth 1 extending from ground up to about 6 km.

We will now also consider the influence of clouds. As for the case of aerosol, real atmospheric profiles have been introduced in the simulations: these are based on observations over oceanic areas carried out by the avionic LIDAR ABLE during the flight from Recife, Brasil, to Isla do Sal on October 1999. Same as for the aerosol case of Fig. 3.1.3-1, Figure 3.1.3-2 and Figure 3.1.3-3 show what should be the effects of the cloud vertical distribution on the fluorescence and scattered Čerenkov signals acquired by EUSO for two different atmospheric conditions observed. Again, the overall effect is to introduce a systematic distortion to the acquired signal resulting in a significant shift of the height location and in a relevant intensity reduction of the maximum of fluorescence. As for the clear sky conditions, the profiles have been calculated for a 1020 eV energy of the primary particle; a 10 % overall detection efficiency has been assumed for EUSO and a 100 mJ energy per pulse has been assumed for the laser source. Considering vertical trajectories, the shift in the height location of the signal observed varies from about 2 to 3 km, depending on the vertical cloud structure considered. Also note in Figure 3.1.3-2 that the landmark of the scattered Čerenkov (i.e. the Čerenkov echo from the ground surface) for the 60° inclined shower is completely extinguished by the upper cloud layers. In this case the signal scattered by the lower cloud may be confused with the landmark, introducing an additional systematic error of about 2 km in the determination of the height location of the maximum of fluorescence. In the same figure it is shown that the laser landmark is still observable in the same atmospheric condition. Figure 3.1.3-3 gives an other significant demonstration of the potential misinterpretation of the fluorescence profile that could arise in absence of a systematic sounding of the atmospheric conditions. Comparing the profiles of 60° inclined showers of Figure 3.1.2-1, which refers to clear sky condition, with the corresponding one of Figure 3.1.3-3, where one cloud stratus of optical depth 1 extending from about 11 km to about 13.5 km is present, the intensity of the observed maximum of fluorescence varies of more than 150 %. Note that in this case there is no evidence of the presence of the cloud from the scattered Čerenkov signal. However the structure and location of the cloud is clearly revealed by the back-scattered laser signal. The possibility to retrieve also the optical depth of the cloud will depend on the signal-to-noise ratio obtainable by the LIDAR measurement, in order to separate unambiguously the molecular echoes from the signal scattered by the cloud and, whenever it shouldn’t be possible, on the possibility to acquire the back-scattered laser signals at more than one wavelength, in order to retrieve the extinction/scattering properties of the cloud.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/13

Figure 3.1.3-2 The expected number of photoelectrons in EUSO, generated by an UHECR of energy 1020 eV (proton), for a shower in a real atmospheric condition observed by the ABLE airborne LIDAR during the flight from Recife to Isla do Sal in October 1999. Two cloud layers of optical depth 1.3 and 0.4 around 2 and 4 km respectively are present.

Figure 3.1.3-3 Same of Fig.3.1.3-2 but for a different observed atmospheric situation: one cloud stratus of optical depth 1 extending from about 11 km to about 13.5 km is present.

3.2 EUSO-AS observational approach An “event-by-event“ approach, which asks for an “in-situ” real time sounding of the atmospheric condition, is strongly advised in order to make a right interpretation of the signals acquired by EUSO taking into account the possible relevant distortion introduced by the extreme variable conditions of the real atmosphere (transmission/scattering). This approach represents the only way to apply a systematic correction that may be relevant to the observed signal profiles and that cannot be extrapolated from statistical knowledge of the atmospheric condition. The laser beam through its atmospheric propagation suffers collisions with molecules and aerosol which provide a backscatter signal as well as an attenuation factor. The interaction of the photons with molecules at the wavelengths of interest can be considered elastic if the presence of absorbing gases is neglected, and the mechanism of scatter is simply represented by a cross section and a phase function. The number density of the scattering molecules can be calculated from the hydrostatic equation if the temperature is known. When aerosols are present the mechanism is rendered somewhat more complex since aerosols may exhibit absorbing properties, a size distribution, complicated angular scattering patterns, significant de-polarization. In LIDAR usage the transport of radiation along the beam through optically thin layers is generally assumed to be affected only by single scattering, so that the entire process may be described at one wavelength by two parameters, namely the back- scattering cross section per unit volume and the attenuation to backscatter ratio. Supposing now that the atmosphere be composed of molecules only it is evident that, if the medium is of sufficiently low optical density and the profile of backscatter is known, the attenuation is simply obtained. If vice

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/13

versa the atmosphere contains molecules and aerosols, the retrieval of attenuation can be considerably more difficult. These considerations particularly apply to the short UV wavelengths where the extinction and scattering due to molecules is large.

3.2.1 LIDAR with one laser wavelength Simulations have been carried out that indicate the difficulty of retrieving atmospheric extinction from an atmospheric LIDAR backscatter experiment if only one wavelength is used, and the advantages resulting from the utilization of other available laser harmonics, also considering that the EUSO spectral range of acceptance is in the ultraviolet and that the main candidate to be utilized as the laser source of the LIDAR is a diode pumped NdYAG laser that emits at the fundamental wavelength of 1064 nm (near infrared) and shall be doubled and tripled to emit also the second and third harmonic at 532 nm (green visible) and 355 nm (ultraviolet) respectively. There are a few ways to retrieve atmospheric extinction from a LIDAR if only one laser wavelength is radiated. The simplest and standard way is to obtain and compare echoes from ranges where clouds and/or aerosol are not expected to be present. This is generally the case in the upper troposphere and lower stratosphere regions where echoes are clearly attributable to the presence of molecule; this is particularly true at the first and second laser harmonic, because the molecular echoes are generally weak at these wavelengths, therefore the occasional presence of aerosol and thin clouds can be clearly detected and the attenuation of the beam in crossing the cloud clearly determined. On the contrary, the higher molecular echoes at the third harmonic (the only laser wavelength detectable by EUSO), reduces considerably the contrast between molecular and aerosol echoes, resulting in a lower capability of detection and in a poorer accuracy in locating the aerosol/cloud structures. The upper and lower edge of a thin cloud are a function of temperature, are generally sharp, and clean air conditions are known to exist above and below it. This method starts of not being applicable in the lower troposphere where the presence of aerosol layers may be continuous and extend to heights from the ground up to 7-8 km. Furthermore, since the profile is not obtained from a single shot but from an integration, the presence of broken clouds may be leading to a wrong result. Other ways to obtain attenuation are related to the retrieval of Raman scattering echoes due to N2 or O2: but the echoes are weak and not likely to be observable from the Space Station. A third way would be to analyse the spectrum by way of a Fabry Perot interferometer, since the Doppler broadening differently affects aerosol and molecular echoes; the technique implies some level of sophistication and requires a stable and narrow band source, and is the one proposed for the Aeolus/Aladin project for a wind retrieving LIDAR.

3.2.2 LIDAR with more than one laser wavelength The level of uncertainty can be considerably reduced by using several wavelengths. A Nd-YAG laser coupled to an harmonic generator can usefully provide, in addition to the fundamental at 1064 nm, the second and third harmonic, at 532 nm 355 nm respectively. The outputs available are not dissimilar in intensity. To enhance the capability of obtaining the aerosol characteristics, more wavelengths could be added. Higher order harmonics would not be as useful since the atmospheric extinction (absorption plus scattering) grows rapidly, in the UV, in particular due to photo-dissociation. To extend the range towards the IR and fill gaps in the visible would require additional sources. Simulations based on using the three wavelengths, easily available as the first, second and third harmonic from a NdYAG laser, have been carried out are presented here. It will be shown, as a first example, that the presence of aerosol may provide LIDAR profiles where the aerosol backscatter exactly compensates the attenuation induced on the molecular echo at one wavelength and that therefore would not be distinguishable from the profile obtained from a purely molecular atmosphere if additional wavelengths are not used. The profiles reported in Figure 3.2.2-1 display the backscattered laser signals calculated, integrated over 100 shots, that should be acquired by a LIDAR system of typical characteristics installed on the Space Station, having a 500 mm aperture diameter telescope and 150 mJ, 120 mJ and 100 mJ pulse energy at the 1st , 2nd and 3rd harmonic respectively. For each wavelength the pure molecular echoes and the molecular plus aerosol echoes have been computed. The change in backscatter as a function of wavelength follows the Rayleigh law: β(λ) ∝ 1/λ4, where β is the backscattering coefficient and λ is the wavelength, for the molecular component; while the change in backscatter as a function of wavelength for the aerosol was assumed to follow the simple law: β(λ) ∝ 1/λ , which is generally adopted unless detailed values are available. The aerosol extinction to backscatter ratio can be expressed as a function of wavelength, particle index of refraction and size distribution. For the test case reported in Figure 3.2.2-1,

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 8/13

the extinction to backscatter ratio has been computed at the three wavelengths for dust particles with a typical log- normal size distribution of 0.1 micron modal radius. The aerosol to molecule backscatter ratio for the third harmonic was assumed to exactly compensate the effects of backscatter and extinction. In fact while the two curves at 355 nm exactly coincide except near ground, where a surface reflectivity has been included; the 2nd and 1st harmonic LIDAR profiles are very different and clearly show the presence of aerosol from ground level up to about 8 km.

Figure 3.2.2-1 LIDAR backscatter profiles that would be obtained at different harmonics for an atmosphere with and without aerosol where, when present, the content of aerosol, expressed as the backscattering cross section per unit volume, and the extinction to backscatter ratio were assumed to exactly compensate the effects of backscatter and extinction at the UV 355 third harmonic.

An other important parameter that could be extracted from multiple wavelength measurements is the colour ratio, that is the ratio between the signals acquired at different wavelengths from the same scattering volume. The colour ratios relative to the real atmospheric conditions reported in Figure 3.1.3-1 have been computed and plotted in Figure 3.2.2-2 for three different size distributions. As can be see from the figure the colour ratios have a different behaviour in the three cases, demonstrating how this type of information could be helpful in order to retrieve the size distribution of the observed aerosol structures: consequently their extinction/scattering properties can be obtained, that will determine the amount of perturbation of the observed fluorescence and scattered Čerenkov signals.

Figure 3.2.2-2 Simulated colour ratios obtainable from LIDAR observations at multi wavelength of the real atmospheric condition reported in Fig. 3.1.3-1. Different size distributions have been assumed for the aerosol: dust type with modal radius of 0.1, 0.3 and 3 microns respectively from left to right.

Figure 3.2.2-3 compares the one way and two way transmittance profile at the third harmonic of the laser respectively for tropical standard clear sky condition, in the presence of an extended aerosol layer having a vertical distribution and optical properties that will produce the LIDAR echoes depicted in Figure 3.2.2-1. If only one wavelength is utilized to acquire the LIDAR signal, in particular the third harmonic of the NdYAG laser (355 nm) that is the only one included in the EUSO spectral range of acceptance, the presence of the aerosol would not be

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 9/13

revealed. Therefore a clear sky transmittance profile (red curve in Figure 3.2.2-3), will be wrongly assumed for the fluorescence profile correction for atmospheric extinction; while the real transmittance profile to be considered should be the violet one reported in Figure 3.2.2-3, that shows a considerable higher extinction starting from the height location of the top of the aerosol layer down to the ground. Of course this test case is very particular, but it gives a good idea of the level of uncertainty that could be associated to a single wavelength atmospheric sounding instead of a multiple wavelength approach. The conclusion is that the use of a stand alone LIDAR system, operating at two or three wavelengths, could be useful if not essential in retrieving the one-way attenuation at 355 nm. In reality the retrieval of the correct attenuation implies solving an inverse problem and will be largely helped by additional information obtained from climatology.

Figure 3.2.2-3 – Comparison of atmospheric transmittance profiles between clear sky conditions and when aerosol is present having a vertical distribution and optical properties that will produce the LIDAR echoes depicted in Fig. 3.2.2-1. The one-way and two-way transmissions from ground to 20 km along the vertical at the third harmonic of the laser are shown. The one way transmission affects the fluorescence signal, while the two way transmission affects the scattered Čerenkov and laser signals.

3.3 Fluorescence profile retrieval Following the general consideration reported at the beginning of this chapter, in order to estimate the effectiveness of the additional information coming from LIDAR measurements in the EUSO data analysis, several cases have been investigated.

3.3.1 Retrieval in the presence of high cirrus clouds The test case reported in Figure 3.1.3-3 has been further investigated to compare the results obtainable in the retrieval of the unperturbed fluorescence signal as generated by the EAS development, if the LIDAR information is exploitable or not, in terms of reduction of the systematic error on the intensity of the maximum determination. Starting from the ideal source profile of fluorescence for a 10e20 eV primary particle and a 60° slant trajectory reported in Figure 3.1.1-1, the expected profile of total fluorescence plus scattered Čerenkov as they will be detected by EUSO have been calculated, adding also the expected quantum noise. To retrieve the unperturbed fluorescence from this observed profile, the correction for atmospheric extinction of the signal should be applied. As a first approximation we neglect the correction of the scattered Čerenkov component which in this case amounts to less than 10 % of the total signal around the location of the maximum (see Figure 3.1.3-3). In absence of the LIDAR information, the presence of the cloud is not revealed by EUSO, since the scattered Čerenkov signal from the cloud located above 11 km is too weak to be detected. In this case a standard extinction profile would be applied to correct the observed signals. Otherwise from the LIDAR measurements it is possible to detect unambiguously the presence of the cloud and to extract its optical depth, from which the effective real atmospheric extinction profile can be derived with an accuracy that will depend both on the cloud optical depth value itself and on the signal-to-noise ratio of the acquired LIDAR signals. In this test case in the presence of a single cloud layer, clearly detectable even at the third harmonic of the laser, the unperturbed fluorescence profile reconstruction using only the backscattered UV laser signal acquired by the EUSO receiver will be considered.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 10/13

In Figure 3.3.1-1 the outputs of the simulation done are reported. The orange cross shaped points are the total noisy signal acquired by EUSO and the red solid curve is the corresponding fit simply obtained by smoothing. The light blue “x” shaped points are the unperturbed signals retrieved if a standard molecular atmospheric profile is utilized to correct the data and the blue curve is the corresponding fit. The “o” shaped points represent the unperturbed signal retrieved using the extinction atmospheric profile (considering also the error associated to this profile) as derived from the LIDAR measurement and the green curve is the corresponding fit. Finally the solid black curve is the real unperturbed fluorescence signal (without any atmospheric attenuation) that should be recovered at the end of the analysis.

Figure 3.3.1-1 Simulation outputs of the signal observed by EUSO for a primary particle of 1020 eV and a 60° slant trajectory in the atmospheric condition of Fig.3.1.3-3, and the corresponding signals retrieved considering the extinction profile of a standard clear atmosphere or the extinction profile retrieved by LIDAR measurements.

A large difference is evident between the intensity of the maximum retrieved under the wrong assumption of clear sky conditions and the expected value, which leads to underestimating the energy of the primary particle to a value that is only about 55 % of the real one. The difference between the retrieved value and the expected one is significantly lower if the extinction profiles obtained from the LIDAR are utilized. In this case the residual discrepancy leads to overestimate the maximum of fluorescence to a value that is about 105 % of the real one (e.g. only 5 % discrepancy). It could be also explained as a result of the neglected correction of the scattered Cerenkov signal; in fact looking again at the Figure 3.3.1-1, the retrieved fluorescence profile has values higher than expected only at the lower heights where the scattered Čerenkov signal is more significant and it should be subtracted.

3.3.2 Retrieval in the presence of mid-low aerosol/cloud layers When more complex aerosol/clouds structures occur, use of a single wavelength LIDAR measurement would imply a large uncertainty in the retrieval of the atmospheric extinction profile, that could impair the effective advantage of the LIDAR implementation. In these cases a significant advantage would arise from the use of a three-wavelength stand alone LIDAR measurements in the unperturbed fluorescence profile reconstruction. In our simulations the LIDAR specifications have been assumed to satisfy a signal-to-noise ratio better than 2 for the acquired backscattered laser profile from ground up to 20 km. As an example a test case with one thick cloud located around 2 km and an aerosol layer extending from the cloud up to about 6 km is reported. Figure 3.3.2-1 displays the signal observed by EUSO for an UHECR of 10e20 eV and of 30° incident zenith angle as computed for the assumed atmospheric conditions. A single wavelength approach will leave in this case a large uncertainty in the retrieval of the atmospheric transmittance from 6 km to the ground along the path of observation, (the expected height location of the maximum of fluorescence). In fact from the laser echo profile observed by EUSO (blue curve in Fig. 3.3.2-1) it would only be

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 11/13

possible to estimate the minimum amount of extinction from the aerosol layer as it reduces the intensity of the observed signal from the molecular echo profile represented by the light blue curve to the attenuated profile corresponding to the brown curve. Figure 3.3.2-2 shows the signals acquired by the stand-alone LIDAR at the three harmonics of the laser for the same atmospheric conditions. Using the multi-wavelength information it is possible to make some assumption about the size distribution of the observed aerosol and consequently to retrieve their extinction profile. A comparative analysis has been carried out to demonstrate the advantage that will arise from this approach. The outputs of the analysis are plotted in Figure 3.3.2-3.

Figure 3.3.2-1 Simulated signals (fluorescence plus Figure 3.3.2-2 Signals acquired by the stand-alone scattered Čerenkov) observed by EUSO for a slant 30° LIDAR in the same atmospheric conditions of EAS generated by a 1020 primary particle and for a 100 mJ Fig.3.3.2-1 corresponding to the three harmonic laser pulse energy at 355 nm, if a total extinguish cloud is echoes of the laser. located at 2 km and an aerosol layer extending up to 6.5 km is present. The associated ideal fluorescence signal which should be retrieved from the observed profiles is also reported.

In the single wavelength analysis a standard molecular extinction profile plus the minimum extinction of the aerosol layer as it could be valued from the UV laser echo observed by EUSO have been applied, while in the multi-wavelengths case an atmospheric extinction profile retrieved has been applied. The correction for the scattered Čerenkov components, assuming a 1020 eV primary particle has also applied considering standard or retrieved scattering properties for the one wavelength and multi wavelengths analysis respectively. The comparative averaged results of several simulation runs are depicted by the solid blue curve and solid red curve for the one wavelength and multi wavelengths analysis respectively. Comparing these curves with the expected profile (black curve) of the fluorescence the higher accuracy achievable by the multi wavelengths analysis is clearly evident. In fact, while the expected intensity and height location of the maximum of fluorescence are close to the values obtained from the multi wavelengths analysis, the fluorescence profile retrieved from the single wavelength analysis, using only the information acquired by EUSO from the third harmonic, leads to a displacement of the maximum of fluorescence of about 1 km and to an underestimate of about 12 % of its intensity. The root mean square difference between the source fluorescence curve and the fluorescence curve retrieved taking into account the modelled atmospheric extinction in the plotted region from the height of the cloud to 10 km, is 0.75 for the one wavelength retrieval approach and 0.13 for the multi wavelengths approach respectively.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 12/13

Figure 3.3.2-3 Outputs of the unperturbed fluorescence profile retrieval analysis. All the profiles are reported as the number of photoelectrons detected as a function of the height of the atmospheric volume of observation, thus they all extinguish after the height location of the cloud. The solid black curve is the same in both the plots and shows the expected fluorescence profile that should be retrieved from the analysis. Red boxes correspond to the calculated fluorescence plus Čerenkov signals observed by EUSO and again they are the same in the left and right plot; the red slim curves in both the plots are the corresponding fit. On the plot the results of the one wavelength retrieval analysis are reported: the violet crosses are the retrieved fluorescence plus Čerenkov signals and the violet slim curve is the corresponding fit ;the green circles are the retrieved fluorescence only and the solid green line is the corresponding fit; the solid blue curve is the retrieved fit of the fluorescence profile averaged over several simulation runs Instead on the right plot the results of the multiple wavelength retrieval analysis are reported : the yellow triangle are the fluorescence plus Čerenkov signals and the yellow slim curve is the corresponding fit; the brown x are the retrieved fluorescence only and the solid brown curve is the corresponding fit. The solid red curve is the retrieved fit of the fluorescence profiles averaged over several simulation runs.

3.4 Conclusions

a) The signal acquired by EUSO is the sum of scattered Čerenkov and fluorescence components. The relative importance of the scattered Čerenkov over the fluorescence increases at low altitudes, even in clear sky conditions, reaching a factor of about 0.5 at 2 km and about 1 at sea level, for vertical showers

b) From the existing data it can be demonstrated that the ideal clear sky condition (absence of clouds and aerosol), represented by conventional standard models, is present in nature only in limited cases. Atmospheric conditions are highly variable in space and time and only a systematic sounding of the atmosphere could allow a realistic evaluation of the distortion induced by the real atmospheric condition to the observed fluorescence profiles at the time of occurrence of the event.. In the absence of an atmospheric sounding device it could be very difficult and sometimes impossible to distinguish clear sky conditions from “polluted” conditions, due to the presence or aerosol/clouds, especially when mid and high level clouds are present

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 13/13

c) The aerosol content and the variability of the cloud coverage of the real atmosphere can introduce relevant distortions to the signal profile expected by EUSO in clear sky condition, that will result in a considerable shift in the height of the observed maximum. For example a desert dust layer, typically extending from ground to a few kilometres, produces a vertical shift of about 2 km in the position of the observed maximum of fluorescence due to vertical incidence EECR (value corresponding to about 100 g cm-2 atmospheric depth).

d) The presence of aerosol and/or clouds, even if they are located well above the height of the maximum of fluorescence, can produce a considerable extinction of the fluorescence signal. It results in a significant reduction of the intensity of the observed maximum, thus introducing a systematic error in the retrieval of the energy of the primary particle. For example a cirrus of optical depth 1 above 11 km could reduce the intensity of the observed maximum to more than a factor of two. These effects could be systematically corrected if the optical depths of aerosol/clouds layers should be known. They can be retrieved from the LIDAR measurements.

e) In the case of high altitude clouds ( > 10 km), the opacity of these clouds reduce significantly the signal reaching EUSO and yet the presence of the cloud would not be revealed by the Čerenkov signal echo; while the LIDAR atmosphere sounding device would clearly notice it.

f) In the presence of low altitude clouds of high optical depths, the landmark of Čerenkov radiation could be completely extinguished by the cloud and may be confused with the signal scattered by the cloud itself, determining a systematic error in the height location of the maximum of fluorescence. In such cases the landmark signal, especially from the first harmonic, may be still observed by using a LIDAR.

g) Slant trajectories and the presence of aerosols and/or clouds increases the relative contribution of scattered Čerenkov with respect to the fluorescence. To separate the scattered Čerenkov component from the total observed signal the scattering properties of the atmosphere along the observation path should be known. They can be retrieved by a multiple wavelength LIDAR system.

h) A single wavelength LIDAR, especially if operating in the ultraviolet spectral region, allows to retrieve the atmospheric transmittance profile in limited cases: when aerosol/clouds layer of relevant optical depth are clearly confined between atmospheric region of purely molecular content.

i) In most of the cases, the use of a stand alone LIDAR system, operating at two or three wavelengths, could be useful if not essential in retrieving the one-way attenuation at 355 nm. In reality the retrieval of the correct attenuation implies solving an inverse problem and should take advantage of available climatological knowledge, and, sometimes, of educated guess.

j) Continuous operation of the LIDAR, mainly due to requirements of thermal stability for the laser source, gives the opportunity to acquire climatological data useful for the statistics of the effective frequency of occurrence of the observed events, and it could be also utilized to retrieve additional information of local atmospheric conditions from successive passages.

To complement the information, for the choice of the LIDAR technique, the following Document is appended: • “Inputs for EUSO Phase A study: Comparison between lidar techniques for AS-EUSO application by numerical performance simulation”, ON/EUSO/1.00/2003/18 July 2003, Observatoire de Neuchâtel, Neuchâtel, Switzerland.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/7

4. THE REQUIREMENTS FOR EUSO ...... 2 4.1 SCIENTIFIC REQUIREMENTS ...... 2 4.2 OBSERVATIONAL REQUIREMENTS ...... 2 4.3 INSTRUMENT REQUIREMENTS...... 3 4.3.1 Discussion of the Instrument requirements (Science to Instrument flow-down ???)...... 5 4.3.1.1 Instrument triggering optical efficacy...... 5 4.3.1.2 Effective aperture ...... 5 4.3.1.3 Pixel size and angular resolution...... 5 4.3.1.3.1 Angular resolution perpendicular to the line of sight...... 6 4.3.1.3.2 Angular resolution parallel to the line of sight...... 6 4.3.1.4 Time resolution...... 7

4.3.1.5 XMAX 7 4.3.1.6 Noise and Background ...... 7

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/7

4. The Requirements for EUSO In this document, the term requirement indicates a figure or condition that must be met in order to not compromise the scientific case of the EUSO mission as described in the section [1.xx]. A goal, in contrast, indicates a figure or condition which seems technically feasible with moderate extra effort and which enhances the capabilities of the instrument significantly in a way to strengthen the scientific outcome of the mission.

4.1 Scientific Requirements The EUSO Scientific Objectives, described in section [1.xx], convert into Scientific Requirements that represent, together with the Observational Requirements, the guideline for the formulation of the Instrument Requirements. Where is the derivation from Scientific Objectives to Scientific Requirements described ? • Deep survey of EECR with E>1020 eV: − 3% statistical error for E≥1020 eV, with the observations of about 1000 events, based on the extrapolation from the AGASA spectrum; − 7% statistical error for E≥1020 eV, with about 100-200 events in presence of a dominant GZK effect. • Spectral Study in the GZK Region: − ∆E/E < 30% (goal ∆E/E < 20% with suitable Atmospheric Sounding). Possibly we should add here at which energy we require such a resolution: E≥1020 eV ? • Point Source/clusters identification: − all Sky coverage; − reconstructed angular resolution at E ≥ 1020 eV: ∆α ≤ 2° (goal ∆α ≤ 1°), in order to ………………………………………………………………………………….. Possibly we should add here at which energy we require such a resolution: E≥1020 eV ? • Primary identification: 2 2 − Longitudinal shower profile with Xmax determination: ∆Xmax ≤ 35 g/cm (goal ∆Xmax ≤ 20 g/cm ), in order to have a resoloution better (three times) than the typical proton-iron Xmax separation. • Neutrino Observation: 2 − Downward Super GZK events, observation of EAS with Xmax > 2000 g/cm . − Upward tau neutrinos (energy range above 1015 eV): observation of EAS exploiting the direct Cherenkov emission and the longitudinal EAS profile. Do we really see them ? • Observation of “slow” phenomena in the atmosphere as Meteors, Blue Jets and Sprites: − “Slow” phenomena by analogy with “fast” phenomena (EECR) are detectable events but with duration time from few ms up to seconds.

4.2 Observational Requirements The Observational Requirements are defined by combination of the not alterable ISS attitude (including orbit inclination, altitude, and pointing capability) and the Observational Approach consequences.

• Instrument must look toward the Earth with its optical axis parallel to the Nadir direction within ± 5°. • Neither parts nor components of the ISS must obstruct the FoV of the Instrument within the maximum required angular value of ± 30° around Nadir.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/7

• Instrument is required to operate with its dedicated voltages converted (internally to the Instrument) from the main 120 V ISS power Bus. • Instrument data I/O requires ISS 1553 standard bus for download and upload link. • Three years on-orbit nominal operation plus two years of extended mission are required for the EUSO Instrument in order to fulfil the scientific objectives.

4.3 Instrument Requirements The Instrument Requirements are a compilation of high-level requirements on the performance, design and operation of the EUSO Instrument on the ISS. These requirements are direct consequences of the Scientific Requirements and Observational Requirements, as defined in section 4.1 and 4.2, and of the specific constraints imposed by the accommodation of the Instrument on the Columbus External Payload Facility on the ISS. In this section the requirements are simply summarized. A quantitative and simplified discussion of the derivation of the main Instrument Requirements is given in section 0. Full Monte Carlo simulations provide the final justification of the Instrument Requirements, as discussed in section [6]. Definition of some of the quantities is given in Appendix A: EUSO baseline parameters. Some of the points will be discussed in more detail in the relevant sections of chapter [5]. • The Instrument shall be sensitive in the wavelength range WR:=(330nm÷400nm). This bandwidth provides the best signal to noise ratio for detecting both the air fluorescence signal and the Cherenkov signal. Below 330nm the signal is strongly suppressed by atmospheric ozone absorption. The bandwidth is the result of a trade-off among signal, background and the system design (the optics in particular), becoming more and more complex as the bandwidth is expanded. [TBCheked: with the airglow spectrum !!!!!]

• The Instrument must collect as many photons as possible, in WR, in order to be able to detect the faint signal from the less energetic EAS and discriminate the EAS from the background. As a consequence a large area collecting optics module is required, with a triggering optical efficacy greater than 1.5 m2 in order to collect enough photons to guarantee the required energy resolution.

• An optics with a large FoV (half-angle γMAX τ 25° with a goal of γMAX ∼ 30°) is required, in order to observe a mass of atmosphere as large as possible, achieving an instantaneous geometrical factor (km2·sr) suitable to fulfil the Scientific Requirements.

• An effective focal length of the main optics as short as possible is required, in order to have a FS (and therefore a Photo-Detector) as compact as possible thus minimizing mass and volume.

• The observational duty cycle, ηO, must be large enough to guarantee a sufficient exposure: ηO τ 0.20 (goal: ηO ∼ 0.25). The Instrument duty cycle, ηI, defined as the fraction of time the Instrument is fully ON, shall be: ∼ (ηO+0.05) to take into account switch on/off and reset/initialization times, self-diagnostics as well as, possibly, some required Calibration and/or Monitoring activity.

• The Instrument shall have an angular granularity of the order of ∆α δ 0.1°, in order to ensure an angular resolution on the EAS direction of the order of a few degrees.

• A Photo-Detector (including its electronics) operating at the single-photon counting level and exhibiting

good photon detection efficiency (εPD τ 0.12) is required. It must be fast enough to be able to follow the space-time development of the EAS (sampling time of the order of µs) and reconstruct the EAS kinematical parameters from one single observation point.

Note: (sampling time well below ≈ 0.1 µs) was REMOVED; we actually have a sampling time of one GTU ! do we need less ?

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/7

• A low noise and good signal to noise ratio are required to detect the faint signal produced by the less energetic EAS and discriminate the EAS from the background. Small cross-talk and after-pulse rate are also required to avoid degradation of the energy and angular resolution. Therefore the intrinsic Instrument noise (all sources) is required to be less than three GHz on the whole Photo-Detector. [cross-check with light-tightness]

• The following statement is very delicate; it would require a more solid justification….

• Analog signal taken by last dynode (or grouping anodes in a convenient way) is required for the detection of diffusely reflected Cherenkov and dynamic range extension.

• Suggested replacement:

• A dynamic range sufficient to detect the less energetic as well as the most energetic EAS is required. Moreover the dynamic range should allow the detection of the diffusely reflected Cherenkov flash.

• An efficient, selective and reliable trigger system, capable of a good background rejection, is required in order to cope with the limited data storage, data transfer and computing capabilities available on-board.

The overall Instrument dead-time, τDEAD, (including the effects of trigger and read-out) shall be: τDEAD ≤ 3%. Triggering efficiency shall be greater than 0.95 for hadronic EAS with energy E ≥ 1020 eV. [check]

• Modularity of the photo-detector and the associate electronics is required. Macro-Cell modules (not necessarily corresponding to the Photo Detector Modules) constituting the photo-detector must work independently in order to ensure, in case of failure of one of more modules, a sufficient and safe operation level.

• Atmospheric Sounding is required for an enhanced characterization of the atmosphere profile.

• All the constraints and requirements related to the Space mission have to be accounted for. Mandatory characteristics are therefore: a compact and robust system with low mass, volume and power consumption, suitable structural properties, radiation hardness, low sensitivity to magnetic fields of the order of the gauss, high reliability and operating stability over a period of a few years.

• EUSO shall be designed in compliance with all applicable requirements for payloads on the STS, the ISS and the CEPF.

• The Instrument must be protected from possibly dangerous environmental factors, including intense light. The required light-tightness in WR (this WR depends on the location of the filters) is less than one GHz total photon rate on the whole Focal Surface.

• The Instrument lifetime design shall be: three full years on-orbit plus two years of extended mission plus two years of pre-launch storage on ground.

• Mission success definition (at end of mission: three plus two years on orbit) shall be: end of mission with at least 0.8 of the Instrument fully working (TBC). The reliability figure for the whole Experiment/Mission shall be: 0.95 mission success probability (TBC). Note that all the Instrument design is basically driven by the optics. Appendix A: “EUSO baseline parameters”, summarizes the baseline parameters and goals in a detailed and exhaustive way.

TO be completed/revised.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/7

4.3.1 Discussion of the Instrument requirements (Science to Instrument flow-down ???) Refer to the Appendix A, “EUSO baseline parameters”, for detailed explanations about symbols and values.

In this section a number of simplified arguments, driving the Instrument parameters choice, are presented. All the following, naïve, considerations apply to an hadron-induced EAS, with a zenith angle less than 75°, observed by EUSO in clear sky conditions. Whenever possible the following arguments are based on analytical, even though possibly approximate, calculations. When this is not easily feasible the input from full simulations is used. All the Instrument requirements are consolidated by the results from full Monte Carlo simulations. The full simulation consolidates these considerations and make them more precise, see section [6.xxxx]. The following assumptions are used. 3 1. Exponential density profile in the atmosphere, with: ρ0 = 1.29 kg/m and L=8.3 km. 20 2 2. Gaisser-Hillas proton induced EAS longitudinal profile at E = 10 eV with: λ = 65 g/cm , XMAX = 880 2 2 g/cm , X0 = 35 g/cm . 3. Most of the EAS properties, as observed by EUSO, depend on zenith angle, azimuth angle and location inside the FoV of the EAS. When the dependence is not explicitly mentioned typical cases are considered. Note that typical EAS induced by a primary proton with E = 1020 eV, with a zenith angle less than 75°, are seen as tracks with an angular span not larger than about 5°; therefore they can be considered as short track inside the EUSO FoV and changes of the Instrument properties with field angle can be neglected. 4. The range of visibility of the EAS is arbitrarily defined as the space-time region where the rate of detected photon for the EAS is larger than the rate from random background.

4.3.1.1 Instrument triggering optical efficacy The triggering optical efficacy is the basic parameter driving the photon detection capabilities of the EUSO Instrument. In fact, among all the parameters affecting the photon detection capabilities, it is the only one that can be, in principle, dimensioned to within the external constraints. The required energy resolution implies a relative error due to the statistics of the detected photons not larger than σN/N ∼ (0.10 ÷ 0.15), that is a detected number of photons of the order of one hundred. The irradiance at EUSO, produced in clear sky conditions by a typical EAS induced by a primary proton with E = 1020 eV at 45° zenith angle and falling at half of full FoV, is 5.5×102 photons per squared meters, as it can be easily estimated analytically. As the typical overall Photo-Detector detection efficiency is εPD ∼ 0.12 the required typical triggering optical efficacy is 1.5 m2. At an energy E = 5×1019eV the relative error due to the statistics of the detected photons is thus expected to be ∼0.15.

4.3.1.2 Effective aperture The effective high energy (asymptotic) aperture (i.e. the effective aperture at very high energy, when the triggering efficiency equals one) is approximately given by the relation: eff =η η −τ π 2 2 2 γ A O C (1 DEAD ) H tan M in terms of the ISS height, H, (H ∼ 430 km) the Instrument half FoV angle, γM, (γM τ 25°) the observational duty cycle, ηO, and the cloud coverage efficiency, ηC, quantifying the effect of real cloud coverage on the EAS detection efficiency (ηC ∼ 0.5, from simulation ISCCP database). Cross check with simulations. The comparison with the Pierre Auger Observatory, when one requires for EUSO an aperture one order of magnitude larger implies the requirement: η −τ 2 γ ≥ O (1 DEAD ) tan M 0.07

Here if we ask for ten times PAuger, taking into accout that γM might go down to 25°, we might not be strictly compliant !!!

4.3.1.3 Pixel size and angular resolution The EAS direction (EAS velocity vector), as seen by EUSO, can be decomposed in two components: one parallel to the line of sight from EUSO to the EAS instantaneous position (EUSO-EAS line of sight) and the other perpendicular to it. Any EAS will be seen as a point moving inside the EUSO FoV with a direction and an angular

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/7

velocity depending on the EAS direction. As the speed of the EAS is known (it is equal to the speed of light) the former depends on the observed angular velocity of the EAS inside the FoV, while the latter is given by the observed EAS direction on the FS. The pixel size is driven by the Scientific Requirements and constrained by available resources. It affects: the trigger efficiency and contamination (roughly speaking the signal to background ratio on the pixel scales as 1/δ, δ := pixel size); the angular resolution; XMAX resolution (i.e. particle identification capabilities). The pixel size has to be not larger than the optics Point Spread Function in order not to spoil the optics resolution.

_ The number of pixels has a strong impact on the Instrument budgets and complexity. A trade-off on the pixel size is therefore of outmost importance. An approximate and simplified analysis, leading to determine the required pixel size, is presented below. Note that the following elementary analysis ignores the effect of the background (making the angular resolution worse) and assumes to have an unbiased estimator of the EAS arrival direction. Therefore the requirements derived must be considered as necessary requirements. On the other hand the use of the diffusely reflected Cherenkov flash might improve the angular resolution.

4.3.1.3.1 Angular resolution perpendicular to the line of sight

The expected angular resolution on the EAS direction perpendicularly to the line of sight, ∆βperp, is readily estimated by assuming to perform a linear fit. In this case one has for the error on the angle: δ 1 1 ∆β ≈ , perp Σ 12 L N

where N is the number of detected photons, ΣL is the standard deviation of the observed EAS image (easily determined from simulations) and δ is the error on the position on the FS which can be roughly taken as the pixel size because the pixel size will be chosen to match, approximately, the optics point spread function. Using again a typical primary proton induced EAS with E = 1020 eV at 45° zenith angle one has L ∼ 1° and N ∼ 102. Exploiting the Gaussian-like shape of the EAS longitudinal profile (neglecting the fact that the Gaussian may be truncated) the relation between the observed EAS length, LEAS, (i.e. the range of the sampled values) and ΣL turns out to be: L ≈ 5 ΣL (from the relation between range and standard deviation of Gaussian distributions). Therefore one finds that, in order to aim to get an angular resolution of the order of ∆βperp∼ 1°, one needs δ ∼ 0.1°.

4.3.1.3.2 Angular resolution parallel to the line of sight The relation between the observed angular velocity and the angle between the EAS velocity vector and the line of sight, β, is the well known relation: − β β ω = c 1 sin  = c     tan  D  cos β  D  2  where c is the speed of light and D is the distance of the moving object (in the present case one assumes that the EAS develops in the lower few km of atmosphere and therefore the EAS distance, D, is known). Due the non- linear relations between ω and β the estimation of the best fit error is more complex in this case. Therefore the error on β is estimated by assuming a simple measurement of the angular velocity of the EAS. One obtains: δ δ ∆β = ∆ω ω =  + t  /   .  LEAS TEAS  Using again a typical primary proton induced EAS with E = 1020 eV at 45° zenith angle one has L ∼ 1° and N ∼ 2 10 , TEAS ≈ 100 µs. By assuming an EAS sampling time (GTU) no larger than δt ∼ 2.5 µs the second term is smaller than the first one. One finds ∆β ∼ 0.15 rad by assuming again δ ∼ 0.1°. One might assume that, with a best fit, this results will scale as N-1/2 obtaining the desired ∆β ∼ 0.015 rad ∼ 0.1°. Therefore one finds that, in order to aim to get an angular resolution of the order of ∆β ∼ 1°, one needs δ ∼ 0.1° and δt ∼ 2.5 µs. Note that, as long as the GTU is smaller than the pixel transit time the error is dominated by the pixel size and not by timing.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/7

4.3.1.4 Time resolution While the time resolution required to follow the space time development of the EAS is in the range of a few µs, a better time resolution is preferred in order to aim to improve the EAS reconstruction by suitable fitting procedure. Moreover, due to the low light level, the Photo-Detector shall be better operated at the single photon counting level, with an electronics processing based on binary electronics. In fact a binary electronics allows a considerable power saving with a large number of channels and it is matches with the low light flux received. However, in order to guarantee a sufficient dynamic range, the time resolution of the Photo-Detector shall be of the order of 10 ns as a goal, sufficient to detect the EAS maximum of a with E = 1021 eV at 45° zenith angle and falling at half of full FoV, without a significant photon pile-up. In order to increase the dynamic range an analogue electronics is foreseen to handle higher signal from diffusely/ reflected Cherenkov and, possibly, primary EHECR with higher energies.

4.3.1.5 XMAX 2 The requirement ∆Xmax ≤ 35 g/cm is satisfied following the previous Instrument Requirements. In fact the relation between slant depth, X, and the coordinate along the EAS, w, is: dX cosθ = X . dw L Therefore for a typical primary proton induced EAS with E = 1020 eV, at zenith angle larger than 45° falling at 2 2 half FoV, one has that the ∆w corresponding to ∆Xmax ∼ 35 g/cm (Xmax ∼ 880 g/cm ) is larger than 0.47 km. Therefore, neglecting the geometrical and kinematical details of the EAS development, it is larger than half the pixel size projected at the Earth. The required Xmax resolution is therefore compatible with the pixel size. The goal 2 resolution ∆Xmax ∼ 20 g/cm might be reached by a suitable fitting procedure.

4.3.1.6 Noise and Background Due to the low expected signal level the noise level has to be kept well below the physical background level. The random background detected by the whole Photo-Detector is expected to be in the range (0.3÷1.0)·1011 Hz for photons in the operating wavelength range WR. Therefore the intrinsic Instrument noise (all sources) is required to be less than three GHz (whole Photo- Detector) in order to be negligible with respect to the natural background. Note that this requirement also include the effect of the stray-light.

TO be completed/revised.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/68

5. THE EUSO INSTRUMENT...... 4 5.1 EUSO INSTRUMENT OVERVIEW ...... 4 5.1.1 EUSO Instrument operating principle...... 4 5.1.2 EUSO Instrument Architecture overview ...... 5

5.2 THE EHECR/ν TELESCOPE BREAKDOWN...... 5 5.2.1 Optics Module...... 7 5.2.1.1 Optics module key design parameters ...... 7 5.2.1.2 Optical Design ...... 7 5.2.1.2.1 Phase A Baseline Design ...... 8 5.2.1.2.2 Performance, Trades Studies, and Tolerance...... 8 5.2.1.3 Optics Manufacturing...... 9 5.2.1.3.1 phase A Prototypes...... 9 5.2.1.3.2 Manufacturing Plan ...... 9 5.2.1.3.3 Fixturing...... 10 5.2.1.3.4 Material...... 10 5.2.1.4 Optics Module Structure ...... 10 5.2.1.4.1 Structure Description...... 10 5.2.1.4.2 Assembly ...... 11 5.2.1.4.3 Metrology ...... 11 5.2.1.4.4 Testing...... 11 5.2.1.5 Optical Filter...... 12 5.2.1.6 The optical adapter or Light Collection System (LCS) ...... 13 5.2.1.7 Error Budget ...... 14 5.2.1.8 Analysis and Notes on EUSO Baseline Optics Throughput Efficiency Budget...... 15 5.2.1.9 Advanced Designs ...... 18 5.2.2 The Focal Surface Photo-Detector...... 20 5.2.2.1 The FS architecture...... 20 5.2.2.2 The MAPMT ...... 22 5.2.2.2.1 The MAPMT voltage divider ...... 23 5.2.2.3 Elementary-Cell...... 23 5.2.2.4 Photo-Detector Modules (PDM) and the Mechanical Supporting Structure (MSS) ...... 24 5.2.2.5 Summary of the main FS PD parameters ...... 26 5.2.2.6 Focal Surface Systematic Errors...... 26 5.2.2.7 Procurement and assembly Procedure ...... 27 5.2.2.8 Further Options ...... 27 5.2.2.9 The Optical Adapter (Light Collector System, LCS)...... 27 5.2.3 The Electronics...... 29 5.2.3.1 Architecture of the EUSO electronics ...... 29 5.2.3.2 The EUSO electronics design and requirements...... 29 5.2.3.3 Front-end, read-out and trigger electronics design description ...... 30 5.2.3.3.1 The principle of measurement and triggering ...... 30 5.2.3.3.2 The Front-End electronics...... 31 ASIC design, development and prototyping ...... 32

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/68

5.2.3.3.3 RO&C Functional and architectural description...... 33 5.2.3.3.4 The EUSO trigger...... 35 5.2.3.4 The system electronics ...... 35 5.2.3.4.1 The control electronics ...... 35 5.2.4 The Atmosphere Sounding (AS) Instrumentation ...... 36 5.2.4.1 The LIDAR...... 36 5.2.4.1.1 EUSO Requirements ...... 36 5.2.4.1.2 AS Scientific Requirements ...... 36 5.2.4.1.3 AS Performance Requirements...... 37 5.2.4.1.4 Sub-system Definition ...... 37 5.2.4.1.5 Sub-system Description ...... 37 5.2.4.1.6 Sub-system location...... 38 5.2.4.1.7 Constituting Elements ...... 38 5.2.4.1.8 Functional, Performance and Physical Requirements ...... 38 Functional Requirements ...... 39 Physical Requirements ...... 39 Structure and Mechanism Requirements ...... 40 Electrical Requirements ...... 40 Communication Requirements ...... 40 Thermal Requirements ...... 40 5.2.4.1.9 Instrument Configuration & Performances...... 41 5.2.4.1.10 Environmental Interfaces ...... 42 Contamination...... 42 Ground ...... 42 On-Orbit ...... 42 On Ground Cleanliness...... 42 5.2.4.1.11 Preliminary design...... 42 5.2.4.1.12 Critical issues and open points ...... 43 Possible Solutions ...... 43 5.2.4.2 Other Devices...... 44 5.2.5 The Power Supply System ...... 45 5.2.5.1 The buffer batteries ...... 45 5.2.5.2 The FS PS system ...... 45 5.2.5.2.1 Distributed PS system for the FS (totally modular)...... 47 5.2.5.2.2 Sextant Centralized PS system for the FS...... 47 5.2.5.2.3 Discussion...... 49 5.2.6 The Thermal control System...... 49 5.2.7 Instrument System Engineering ...... 50 5.2.7.1 Instrument System Configuration...... 51 5.2.7.2 Interfaces 52 5.2.7.3 Components ...... 52 5.2.7.3.1 The lid...... 52 5.2.7.3.2 The protection layers: meteoroids and orbiting debris protection...... 52 5.2.7.3.3 The protection layers: light tightness ...... 52 5.2.7.4 Budgets 52 5.2.7.5 Envelope and Main Coordinate System ...... 53 5.2.7.6 Sub-systems Mechanical Environment and Interfaces...... 54

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/68

5.2.7.7 Sub-systems Thermal Environment and Interfaces...... 54 5.2.7.8 Instrument Fields of View ...... 54 5.2.7.9 Safety and reliability aspects ...... 54 5.2.8 Instrument Functional Parameters versus Instrument requirements ...... 54 5.3 INSTRUMENT MODULE ASSEMBLY, INTEGRATION AND VERIFICATION...... 55 5.3.1 Model Philosophy and Testing ...... 55 5.3.2 Verification approach...... 55 5.3.2.1 Verification Method ...... 55 5.4 INSTRUMENT CALIBRATION...... 57 5.4.1 Pre-flight calibration...... 57 5.4.1.1 Level 1 Pre-flight calibration (MAPMT calibration) ...... 58 5.4.1.2 Level 2 Pre-flight calibration (Macro-Cell calibration)...... 58 5.4.1.3 Level 3 Pre-flight calibration (EUSO detector calibration)...... 58 5.4.2 On-board calibration...... 58 5.4.2.1 Optical design...... 59 5.4.2.2 Light source ...... 60 5.4.2.3 Sensors and pre-amplifiers gain calibration by threshold scanning ...... 61 5.4.3 EUSO field monitor (UV-Optical CCD camera, and IR CCD camera)...... 63 5.4.4 Ground Light Source ...... 63 5.4.4.1 Ground Light Source (Flasher)...... 64 5.4.4.2 Ground light source(LIDAR) ...... 65 5.4.5 Instrument Calibration: operations...... 67 5.4.5.1 Technical calibration ...... 67 5.4.5.2 Scientific calibration...... 67

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/68

5. The EUSO Instrument

5.1 EUSO Instrument overview

5.1.1 EUSO Instrument operating principle

The air fluorescence light from the UHECR induced EAS appears as a thin luminous disk, whose intensity is proportional to the number of charged particles in the EAS, with a variable radius of the order of 0.1 km, and an EAS front thickness of the order a few meters. It moves through the atmosphere approximately at the speed of light. The fluorescence light at the maximum of the EAS development and the total fluorescence light are proportional to the primary particle energy. The additional observation of the impact point, on land, sea or cloud, and the measurement of the timing of the diffusely reflected Čerenkov light provides additional information, useful to improve the EAS reconstruction. The apparent EAS angular size can vary from a few up to tens of degrees, depending on many parameters, including the nature of the primary particle, its energy and arrival direction. A proton induced EAS of energy 1020 eV will be typically seen by the EUSO Instrument as a few detected photons per pixel per µs, over several aligned pixels during tens to hundreds of µs. This pattern will be in many cases followed, with a time delay up to hundreds of µs by the Čerenkov light spot, peaked within a few pixels and within a small time window of a few µs. A detector with a sufficiently fast time response allows one to determine the direction of the primary EHECR by means of one single observation point. In fact any EAS will be seen as a point moving inside the FoV with a direction and an angular velocity depending on the EAS direction. The EAS direction (EAS velocity vector) can be decomposed into its parallel and perpendicular components with respect to the line joining the EUSO observation point to any suitable point of the EAS (EUSO-EAS line of sight). As the speed of the EAS is known (equal to the speed of light) the former depends on the observed angular velocity of the EAS inside the FoV, while the latter is given by the observed EAS direction on the FS. The peculiar characteristics of the EAS, in particular the kinematical ones, allow one to distinguish them from the various backgrounds, due to a typically different space- time development. Since the EAS travels at the speed of light, the photons reaching EUSO from any point on the EAS lags behind the passing EAS front by a time: + ∆ = AB BC = AC t ϑ c c tan 2 as can be deduced from Figure 5.1.1-1.

Figure 5.1.1-1 Kinematics of the EAS

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/68

Add some more geometry/kinematics formulas

5.1.2 EUSO Instrument Architecture overview

The EUSO Instrument is conceptually very simple. It consists of an EECR/ν Telescope and of an Atmospheric Sounding device. The Telescope is basically a fast, high-granularity, large-aperture and large Field-of-View digital camera working in the near-UV wavelength range (330nm ÷ 400nm) at the with single photon counting capability. It is made of a collecting Optics focusing the EAS image onto a Photo-Detector assembly located on the Focal Surface of the Optics. The EECR signal is separated by the background and processed on-board, before raw data are sent down to ground-stations. Atmospheric Sounding Instrumentation provides useful information on the atmosphere profile, helping the EECR events reconstruction. Though an Instrument aiming to watch from Space the EAS produced in the atmosphere by EECR is conceptually simple, its design is a challenging task, mainly because the EECR flux reaching the Earth is very small, the observable signal is very faint and the apparatus has to operate in Space within the constraints and resource limitations imposed by a Space Mission. EUSO operates as a Time Projection Chamber to measure the properties of UHECR induced EAS. A two dimensional projection of an EAS is reconstructed from the distribution of the light signals imaged onto the FS. The third projection, the distance along the line of sight, is measured from time delays. The absolute altitude of the EAS can be further evaluated from the diffusely reflected Čerenkov signal. EUSO measures both the fluorescence and the Čerenkov photons generated by the EAS. Ionization of the nitrogen of the atmosphere by the charged particles in the EAS leads to an isotropic emission of fluorescence light along the EAS trajectory. Fast charged particles emit Čerenkov light within a small cone along the EAS direction. The Čerenkov light can be detected directly for EAS pointing to the Telescope or by reflection from ground, sea or clouds for EAS pointing to the Earth. An UHECR will appear as a luminous point propagating almost at the speed of light along a segment of few to tens km long. Its image will appear sequentially in the FS, starting with a faint signal and gradually increasing to a maximum before fading gradually away. The signal will depend on the energy, the altitude and the inclination of the EAS. For a typical EAS of 1020 eV energy, there will be typically few photoelectrons per pixel per µs on several aligned pixels during ~ (10 ÷100) µs. This pattern will be in most cases followed with a time delay up to ~ 100 µs by a spot of the Čerenkov light, peaked within one or two pixels and within a small time window of about a fraction of µs to few µs.

5.2 The EHECR/ν Telescope breakdown

The EUSO Instrument baseline is made of the following parts, from a functional and logical point of view.

1) The main (refractive) optics, the Optical Module (OM), based on Fresnel lenses, to collect the incoming light and focus the EAS track image onto the Photo-Detector on the Focal Surface. 2) The Photo-Detector (PD) on the Focal Surface (FS) of the main optics, detecting the near-UV light focused by the main optics and providing both position and arrival times. 3) The System Electronics (SE), controlling the Instrument at system level, performing the Trigger and Data Handling at system level and interfacing to the EUSO payload. 4) The Power System (PS). 5) The Thermal Control System (TCS). 6) The Instrument Structure (IS), including the external and internal structures, the body and envelope, the lid and the baffle as well as the meteoroids, orbiting debris and thermal protections and blankets. 7) The Atmospheric Sounding (AS) Instrumentation. 8) The Monitoring, Alignment and Self-Calibration system (MASCS). 9) The Ground Support Equipment (GSE).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/68

To reference the summary table of the baseline Instrument parameters: see the separate appendix file. It will be either inserted here or a reference to the appendix will be made: see Appendix A

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/68

5.2.1 Optics Module

The Optics Module (OM) is made of: the two Fresnel lenses (the external and the internal one); the aperture stop; the filters which, in the current baseline, will be located on the FS; the required harness, local and I/F harness; the interfaces among the different components of the OM and those between the OM and the rest of the Instrument; the supporting structure; all the ancillary components required for the proper operation, monitoring and control of the OM, including the Thermal Control, Monitoring, Alignment and Self-Calibration systems.

5.2.1.1 Optics module key design parameters

The flow down from the scientific objectives of the EUSO mission yields the baseline requirements for the Optics Module (OM), including the imaging characteristics of the light-collecting Telescope. The key design parameters for the OM are as follows. • A triggering efficacy, in square meters, τ 1.6 - 0.28 γ2 - 2.1 γ4 (where γ is the angle between the optical axis and the direction to the EAS in radians) is required for the collection and detection of the faint fluorescence from EAS with energies above 3×1019 eV and full Instrument sensitivity above 1×1020 eV. • A full angle FoV ≥ 50°. This large FoV allows EUSO to view a sufficiently large volume of atmosphere to assure a sufficient statistical sample of EAS events at energies above 1020 eV. • A 6 arcmin (=0.1°) angular resolution. This resolution corresponds approximately to (0.75 ÷ 0.87) km on the Earth. • A 330nm ÷ 400nm optical bandwidth. This bandwidth provides the best signal to noise ratio for detecting both the air fluorescence signal and the Čerenkov signal. • Fast optics with f/#=1.00 to f/#=1.25 that are limited to be ≤2.5 meters in diameter. The deployment of EUSO by the Space Shuttle and its attachment to the ISS place severe limits on the mass and size. The EUSO mission has an unusual set of requirements to be met by an optical system. The large aperture, the wide FoV, and the low f/# requirements, coupled with the mass and size constraints are major challenges. On the contrary the resolution of the EUSO Telescope is three orders of magnitude less stringent than a diffraction- limited astronomical Telescope. In order to optimise the design performance, we have developed a set of new optimisation parameters that allow the optical design to adjust and compensate for the scientific requirements. We employ a method we call least light loss (LLL), that adjusts and weights the wide field angle more heavily than the on-axis field. Since the number of visible EAS increases as the square of the field angle, we need systems whose response is as flat as possible. The LLL optimisation of the system has the power to compensate for the inherent cosine cubed roll-off in the triggering efficacy.

5.2.1.2 Optical Design The EUSO optical design evolved from the Optical Wide-angle Lens (OWL) design studies conducted at UAH since 1995 under NASA’s Cross-Enterprise-Technology-Development Program (CETDP). Early on, the investigation found purely reflective systems unsatisfactory. Reflective systems, using either a single reflective or a traditional two-mirror system, are inherently limited to a narrow FoV at low f/#’s. After studying reflective, catadioptric and refractive systems, we found that only refractive systems could satisfy the full set of constraints imposed by the EUSO requirements. This result has been placed on a firm theoretical basis. For a refractive system, there are three deficiencies that must be overcome: (1) mass, (2) material absorption and (3) chromatic aberrations. Fresnel lenses solve the first and minimize the second. A Fresnel lens essentially uses the surface profiles of a standard lens, divides it into zones, and collapses it onto a thin plate that may have some base curvature. By using Fresnel lenses in place of standard lenses, we can achieve a large-aperture, wide-FoV, and low-f/# system with drastically reduced mass and UV absorption. Finally, the optical design addresses the issue of

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 8/68

chromatic aberration. First, there is the choice of material, for which we have identified several plastics with both low absorption and low dispersion in the UV bandwidth. Fully polychromatic performance from in the 330nm ÷ 400nm wavelength range is then achieved by adding a micro-faceted structure on the first and last surface, as shown in Figure 5.2.1.2-1. This microstructure leads to an additional bending of the light rays due to diffraction. The dispersive nature of the diffraction is opposite to that of the material, resulting in chromatic error compensation.

Figure 5.2.1.2-1 - EUSO OM Design

5.2.1.2.1 Phase A Baseline Design In January 2003 at the conclusion of NASA’s phase A study as reported in the Concept Study Report (CSR), the baseline optical design consisted of two spherically-base-curved, double-sided Fresnel lens with a central stop. The baseline design presented in the CSR and the ESA Mid-Term Review was a monochromatic system optimised for the 357 nm wavelength. The lenses were 2.5-m diameter and 15-mm thick. The final design is a polychromatic system where the front surface of the first lens and back surface of the second lens have micro-grating structures on them for chromatic aberration correction. Our choice of material is presently special grade UV transmitting polymethyl-methacrylate (UV-PMMA-000), which has ~50% smaller dispersion than commercial UV-PMMA over the 70-nm UV bandwidth. With this new grade 000 material, we have opened the EPD to 2.3-m while lowering the f/# to f/1.00. Increasing the EPD from 1.9m to 2.3m, increases the collection efficiency by 46%. CodeV results show that this new system maintains well controlled, but slightly bigger, spots over the full-FoV, however, the off- axis vignetting roll off is greater. Further improvements were made by implementing a least-light-loss requirement in the design [Takahashi, Lamb and Hillman, 1999], making the vignetting roll off nearly flat. Since completing the phase A CSR in the USA, we have continued studies to advance the optical design, both to improve optical performance and to lower manufacturing risks. A major advance builds upon the work initiated by A. Zuccaro [Zuccaro, 2003]. By radially segmenting each Fresnel lens into three zones, each with a different spherical base curvature, A. Zuccaro created an effective aspheric base curvature and showed better than a 20% reduction in the encircled energy spots. The further reduction of spot sizes is expected using the ZEMAX program to design lenses with aspheric bases curvatures. With aspheric base curvatures, it may be possible to design lenses that are sufficiently achromatic without the use of surface micro-gratings.

5.2.1.2.2 Performance, Trades Studies, and Tolerance With the phase A baseline design, we used CodeV to analyse manufacturing and alignment errors, including tilt, decentring, thickness, wedge, and index variations. In addition, we have used the optical system analysis program ASAP (Breault Research Organization) to evaluate the systems throughput. With ASAP we are able to create a full physical model of the system, including true facet shapes, back-cuts, surface reflections, and material absorption.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 9/68

This work has been summarized and incorporated into an error budget for the Optics Module. Finally, a new subroutine that incorporates the exact ray tracing of the optical system has been produced and introduced into the ESAF simulation framework so that end-to-end simulations of EUSO can be performed.

5.2.1.3 Optics Manufacturing

5.2.1.3.1 phase A Prototypes During phase A the following investigations were conducted with regard to the manufacturing of the EUSO OS: 1) Petal moulding technology development 2) Material selections and trade studies 3) Tool tip radius 4) Facet width and depth 5) Facet curvature 6) Surface smoothness measurements 7) Fixturing

Figure 5.2.1.3-1 Moulded Petal Segment Figure 5.2.1.3-2 Diamond Turning Meniscus lens

The segmented phase-A design was tested for manufacturability. A petal moulding technique was developed during phase-A, as shown by Figure 5.2.1.2.-1. Two 1-m size test articles of the centre piece, a plano-convex lens with facets on one side and a thin doubled sided meniscus were manufactured during the phase A study. Testing of these lenses allowed characterization of the Moore M40 diamond turning machine, shown in Figure 5.2.1.3-2 cutting the 1-m meniscus, to determine what machine upgrades will be required to manufacture the EUSO optics and the definition of the fixturing required to support a large meniscus element on the vacuum chuck.

5.2.1.3.2 Manufacturing Plan The concept presented at the Midterm Review was for a segmented lens consisting of a turned centre and moulded petals. Since this review, the confidence we gained with test manufacturing the one-meter meniscus and the design of a support structure that can carry a monolithic 2.5-meter lens have convinced us that a one-piece lens can be used. Due to the increase in throughput and the simplicity of integration, the option to manufacture each lens in one piece by direct diamond turning is our new baseline. Since manufacturing quality has not been

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 10/68

confirmed for neither direct diamond turning nor moulding, a parallel investigation of moulding lens segments will continue into phase B. A complete prototype of the OM will be manufactured during phase B using the best manufacturing method to verify the OM design and manufacturing plan. The EUSO OM baseline is for the lenses to be diamond turned directly into PMMA plastic. For the EUSO system, 1-mm constant width segmentation would produce 2,500 facets across the Fresnel lens. This would lengthen the manufacturing time required and increase the amount of scattered light off the facet back cuts. For these reasons, a segmentation using facets that are of constant depth of 1 mm and varying widths has been chosen.

5.2.1.3.3 Fixturing An approach to fixturing the lenses has been developed and demonstrated that is scalable from the 1-meter tests to 2.5 meters without subjecting the lenses to undue stresses or strains. This approach produces symmetric grooves on both surfaces without damage that have proved to be sufficiently coaxial. This fixturing concept, developed in phase A, will be used to produce the 2.5 m prototype lenses for the EUSO project during phase-B.

5.2.1.3.4 Material All the candidate lens materials selected have been shown to be relatively free from susceptibility to radiation damage, solarization, and UV photochemistry. Results from long-term vacuum tests and simulated 3-years radiation tests show less than 3% degradation. Manufacturing data obtained recently from Sudbury Neutrino Observatory and Mitsubishi has shown that ultra pure UV-PMMA can be obtained with dispersion in the 310nm to 400nm range that is ~50% what has been used in previous optical designs.

5.2.1.4 Optics Module Structure

5.2.1.4.1 Structure Description The OM and component hardware are shown in Figure 5.2.1.4.1-2. The OM consists of the two monolithic Fresnel lenses, an aperture plate, and lens support frames necessary to support the lenses so they will survive launch loads and to maintain optical alignment during operational phases. The lens material is PMMA, which has a substantially greater coefficient of thermal expansion (CTE) than that of most structural materials, including composites. Therefore, it was decided that Aluminium 7075 would be used for the lens frames because it has a larger CTE and a high strength to weight ratio. A flexural mounting system will be utilized to athermalize the optics and mitigate differences in CTE. Flexures are designed to allow the lens to expand radially, while maintaining structural strength axially and tangentially. The sensitivity of a Fresnel Telescope to errors caused by deflections is much less than that of a conventional Telescope—deflection sensitivity is on the order of mm for Fresnel alignment. The current concept for attachment to the EUSO structure is via each the outer ring of each lens frame, utilizing three quasi-determinant high-load flexures or trunnions. The outer ring of each lens frame serves as the primary load path from the EUSO structure. Twelve radial struts connect the lens outer ring to an inner ring of the frame for each lens. There are 24 flexures located on the outer ring, and 24 flexures located on the inner ring. Trade studies were performed to determine the optimum size and location of the frames for the monolithic lenses. The lens frame was constrained circumferentially and out-of-plane at the three mounting locations as shown in figure 5.2.1.4.1-1. Analyses were performed using preliminary design load factors and random vibrations specified in EUSO-SP-AI-0001, “EUSO P/L Instrument Interface Requirements Document” (IIRD). Random vibration loads were calculated using the criteria in the IIRD and the methodology defined in SSP 52005B, “Payload Flight Equipment Requirements and Guidelines for Safety-Critical Structures.” Acoustic loads were not included due to the unknown acoustic environment to which the lenses will be exposed when mounted inside the EUSO. Four thermal load cases also were run, representing the operating temperature range of 10 + 10 °C, and the survival temperature range of 25 + 50 °C. Safety factors specified in SSP 52005B were applied to the stress results to calculate yield and ultimate margins of safety. Safety factors for composite structures were used on the PMMA. Analysis results showed that the lens and the lens structure components all had positive margins of safety. Although the flexures had negative margins of safety for some load cases, this was expected due to the immaturity of flexure design at this time. There is assurance that this will be corrected once the detailed design is completed. The first natural frequency of the lens and lens structure is 16 Hz.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 11/68

Figure 5.2.1.4.1-1 Finite Element Model (front and rear) Figure 5.2.1.4.1-2 Lens 1 with its frame

5.2.1.4.2 Assembly The lens frame structure will be made from aluminium alloy 7075, using standard aluminium manufacturing technology. The aperture ring will be made from aluminium plate and will be attached to the Lens 1 outer ring. The lens frames will be assembled using a ground-support equipment (GSE) alignment stand, and the GSE will have methods of calibrating each frame to ensure that alignment is within tolerances during assembly. Flexures will then be mounted on each frame, and the lens will be mounted onto flexures. Each flexure will also be adjustable to allow fine alignment adjustments. The lens assembly will be tested to verify alignment is acceptable. System testing will be performed to ensure that the OS meets alignment criteria.

5.2.1.4.3 Metrology Techniques for evaluating the performance of the large OS proposed here have not been addressed in industry but were developed during the CETDP research and development at UAH. After each lens is produced, the mechanical properties and optical performance will be evaluated for consistency with design specifications. Current industry standard Fresnel lens characterization is a visual inspection of groove geometry using a high- powered microscope followed by an optical performance test. Planned measurements are on the essential specifications of the physical properties include the following: i. Surface profilometry. ii. Facet angle measurements. iii. Lens thickness micrometry. iv. Lens curvature measurement. Planned measurements on the essential specifications of the optical properties (imaging accuracy and transmission) are as follows: i. Two-pinhole images and the rms separation. ii. Photometric fluence distribution. iii. Integral photometry. iv. Veiling glare and stray-light measurements. Optical testing will be conducted at the 2.5 m collimator at Wright-Patterson Air Force Base.

5.2.1.4.4 Testing Structural testing will be performed to certify the OM for flight. This will include structural load testing, modal survey testing, random vibration testing, and acoustic testing of the components and sub-system, as appropriate. Several non-flight hardware end items are anticipated, including transportation container(s), an OM metrology and

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 12/68

integration structure, alignment tooling, and lifting/handling devices. A ground version of a metering structure between the two lenses will also be fabricated to allow the two lenses to be tested as a lens system prior to delivery. Optical testing will be conducted at the 100-inch collimator at Wright-Patterson Air Force Base.

5.2.1.5 Optical Filter The Schott BG3 absorption filter has been selected as the baseline filter for EUSO. This filter will have a multi-layer coating that cuts off transmission sharply above 400 nm and is anti-reflective [Gambicorti, Mazzinghi and Pace, 2003]. Figure 5.2.1.5-1 shows the transmission characteristics of the filter at the extreme incident angles of the light. For light incident at the largest angles, the blue and red curves show the effect of S and P polarized light respectively. This filter transmits >90% of the light in the region 330nm ÷ 400nm for most incidence angles with less than 10% throughput above 440nm and less than 1% at 460nm. The anti-reflective coating virtually eliminates reflection losses.

Figure 5.2.1.5-1 This figure shows the transmission of MgF2/HfO2 + AR coating applied to a BG3 filter. The addition of

the MgF2/HfO2 multilayer coating provides a shape cut-off above 400 nm with little loss of transmission below 400 nm. The two black curves show the transmission of light at normal incidence and at the largest angle of incidence. For the latter case, the blue and red curves show the transmission for S and P polarized light.

The filter can be mounted on the glass window of the MAPMT. The bonding material, identified as space qualified and UV transmitting (96% for 330nm ÷ 400nm), is EPO-TEK 301-2 [EPOTEK]. The Schott BG3 filter was tested for compatibility in the LEO environment. The filter was exposed to ionizing radiation levels present in the ISS orbit. The transmission of the BG3 filter was measured before and after the exposure to detect changes in the filter throughput. Transmission in the 300nm ÷ 400nm band decreased less than 3% after receiving the equivalent of a 3-year exposure in space. This performance is acceptable for the EUSO filter. Samples were also included in phase-A vacuum testing for a 10-month period and showed no detectable change in their properties. The concept to bond the filters directly to the PMT glass window has been studied by the EUSO-Japan Team. In the tests a BG3 filter was glued to a PMT window using the epoxy EPO-TEK 301-2. A series of vibration tests were then performed on this assembly. This mounting technique was demonstrated successfully. The baseline filter design calls for a separate filter for each of the FS sensors. The physical size of the baseline filter is 2 mm thick. The US-EUSO team is to procure the filters and deliver the final product to the Japan-EUSO team. The filter supplier will cut the filters with the specified shape and dimensions and apply the multiplayer

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 13/68

coating to the BG3 glass. The filters will undergo thermal-vacuum testing and transmission measurements. The Japan-EUSO team will mount the filters on the MAPMTs and complete the testing for the EUSO Photo-Detector.

5.2.1.6 The optical adapter or Light Collection System (LCS) The baseline design calls for the sides of the filters tapered to improve the effective geometrical collection efficiency of the FS detectors to 95%. The base of the filter has a cross section that matches the input window of the baseline EUSO photo-detector (Hamamatsu MAPMT R8900-M36). An optional improvement suggested by [Gambicorti, Mazzinghi and Pace, 2003] is to add a thin plastic lens beneath the BG3 filter and a truncated pyramid with mirrored sides below the lens. Simulations show that this lens- pyramid combination can improve the effective geometrical collection efficiency to 99% and still meet the mass requirements. This would provide better overall throughput for the EUSO Instrument.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 14/68

5.2.1.7 Error Budget

Parameter Value Transmittance Source and Comments Field Angle (degrees) -30 -20 -10 0 10 20 30

Fresnel Facet Backcuts See Note 1. ASAP ray trace Blockage and Losses Design 0.68 0.88 0.91 0.92 0.91 0.88 0.68 Hillman

PMMA Material Absorption PMMA data 0.95 PMMA table: Adams, Takahashi Anti-reflection on two with 1.5% Surface Reflections 0.90 residual loss Surface Roughness TIS Loss 20 nm RMS 0.89 See Note 2. TIS analysis: Hillman

50-100 nm Diffractive Facet Cut Depth RMS See Note 3. Diffraction Efficiency Irregularity (uncoated) 0.95 Analysis: Hillman

Root & Peak Tool Error Tool Radius See Note 4: Geometric Analysis diffractive losses 0.5 micron 0.90 with Corrections:Zissa, Hillman

Support Structure Mechanical Obscuration Design 0.88 Ferguson

Total Throughput Efficiency 0.39 0.50 0.52 0.53 0.52 0.50 0.39

See Note 5 on LLL optimization and Note 6: Alignment Tolerance Analysis, Takahashi, .Hillman, Fraction of photons in the 5mm diameter Pitalo. Non-symmetrical due to trigger bucket bucket used 0.74 0.61 0.54 0.74 0.57 0.68 0.82 Lens 1 to Lens 2 tilt On Orbit Trigger Throughput Efficiency 0.29 0.31 0.28 0.39 0.30 0.34 0.32

Absorption change due to radiation 0.015 0.99 NASA Test Report 2000 Losses due to contamination of the surfaces 0.01 0.99 Malone End of Mission (EOM) Trigger Throughput Efficiency 0.28 0.30 0.27 0.38 0.29 0.33 0.31

EOM Trigger Efficacy (m2) 1.17 1.25 1.14 1.58 1.20 1.39 1.29 Trigger Efficacy Req. (m2) 1.35 1.52 1.57 1.58 1.57 1.52 1.35

Table 5.2.1.7-1 Baseline Design EUSO Optics Throughput Trigger Efficiency Based on 2.3 m EPD

This optical design is within 13% (averaged over the FoV) of meeting the Optical Triggering Efficacy requirement at the end of the mission. Advanced Optical design concepts are being explored that promise to improve the throughput enough to exceed the Optical Trigger Efficacy requirement, see section 4.1.1.8. ???

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 15/68

5.2.1.8 Analysis and Notes on EUSO Baseline Optics Throughput Efficiency Budget This budget is based on earlier work done with a design having a 1.9 m EPD.

Note 1: Fresnel Facet Back-cuts Blockage and Losses based on ASAP ray trace analysis. Source: L. W. Hillman. Entries in the table correspond to ASAP ray tracing analysis. This analysis modelled the complete Fresnel lens structure with 1mm constant depth facets. A uniform source illuminated the full front lens at the four angles 0°, 10°, 20° and 30°. The perpendicular irradiance was held fixed. The number of rays collected within a 10mm×10mm area at the FS compared to the number of rays collected by the entrance pupil area when the source is at 0°. This ratio yields the change in the efficiency with observation direction. The Fresnel facets result in a throughput efficiency of 92% at nadir and 68% at 30°, which includes the approximately cosine obliquity factor of the entrance pupil. Because Code V is unable to account for facet loss, but properly accounts for lens vignetting losses, lens-vignetting losses are taken out of this line. They are included in line labelled “Fraction of photons in the trigger bucket”.

Note 2. Surface Roughness TIS Loss. Source: L.W. Hillman. Surface roughness will degrade the transmission by scattering the light. This loss is characterized Total Integrated Scatter (TIS) and depends on the surface rms roughness and the wavelength, λ. When the roughness rms is less than ≈λ/3, then the transmittance efficiency accounting for TIS is given by the classic formula, 2   2π   T = exp − RMS * ∆n TIS   λ       where ∆n is the change refractive index across the boundary, rms is the root-mean-square surface height variation, and λ is the wavelength. This function is plotted in the Figure 5.2.1.8-1 for ∆n = 0.5. Based on an rms ~12nm, sets TIS transmittance to ~95%. It should also be noted that this loss can be further minimized by the application of a near index matching smoothing layer.

Note 3. Diffractive Facet Cut Depth Irregularity. Source L.W. Hillman. The diffraction efficiency of a first- order blazed grating optimized for wavelength λ0 is given by λ−∆ π η∆ λ =2 0 z − ∆ =sin(x ) (znx , ) sinc 1 where sinc( ) . λπx In this expression, ∆z is the error in the grating depth, λ is the operating wavelength, and ∆n is the index change. Assuming that the cutting of the depth of the grating has an rms accuracy of σ and is Gaussian, the average efficiency becomes 2 1  z  TT(,λσ ) = η (,)exp z λ − dz . σπ∫  σ    This function is plotted in Figure 5.2.1.8-2. Setting the rms accuracy to ~ 50nm÷75nm yields a combined diffraction efficiency of 95% for gratings. The effects of this error can be reduced by applying a smoothing layer overcoat with a material having different refractive index (∆n ~0.15).

Note 4. Root & Peak Tool Error Diffractive Losses. Source: Zissa and Hillman. A straight geometrical shadow model of was conducted by David Zissa. He determined the geometric area coverage by distortion of the facet due to tool size (radius = 0.5 µm) at root and distortion (0.2 µm) at crest to be approximately 14%. Because the size of the crest is less than wavelength and the root radius 2/3 larger, the effective area that scatters is roughly 70% the geometrically calculated number. The estimate for the scattering loss is 10% with a effective transmittance of 0.9.

Note 5. Advanced Design Gain Using LLL Optimization. Source Takahashi. Given the optical throughput requirements, it is evident that the F/1.25 1.9m EPD baseline design presented at the Mid-Term Review falls short of these requirements at the high FoV angles. We have addressed this critical issue by increasing the EPD to 2.3m and exploiting a design technique we call LLL Optimization. The UAH optical design team first explored this

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 16/68

method in 1999. The scientific baseline set for the efficacy of the optical sub-system (i.e. the conversion of a photon irradiance at EUSO into the number of photons that lie within in a pixel bucket on the focal plane) has been based upon Monte Carlo simulation. The critical parameter is that the number of photons in one gate-time unit incident on a photo-multiplier pixel (i.e. the bucket) is larger than a triggering threshold. Once triggered, photon counts from this and surrounding pixels are all rapidly captured. Hence the statistics and probability of capturing and recording the EAS from UHECR (i.e. number of events per energy per year) depends on the efficacy integrated over the full FoV. The LLL optimization weighs the throughput and spot size for the high-field angles higher than the near-axis angles. Therefore the spot size is smaller off-axis than on-axis. This compensates for the roll-off due to vignetting and facet losses. Hence as the aperture is opened up and more photons are collected by the optical system, the additional off-axis photons are more concentrated into their bucket than the on axis photon. In fact, in exploring the baseline change going from 1.9m to 2.3m, we only increased the on-axis 0° triggering efficacy by 8%, but increased the 30° by 34% (a simple scale of the pupils results in a 46%--a result that is impossible to achieve since the rms spot sizes do grow in each case.). This advanced LLL 2.3m EPD f/#=1.00 design therefore comes close meeting the baseline science requirements. It is presently being optimized and analyzed for EUSO.

Note 6. Alignment Tolerance Analysis. Source: Takahashi, Zuccaro, Pitalo. The alignment tolerances listed under “Other Manufacturing and Integrations Errors” are the errors that the OM manufacturer is responsible for, but that are not accounted for with the facet and diffractive budget items. This includes:

i. Lens 1 Surface 1 To Surface 2 Centration. ii. Lens 2 Surface 1 To Surface 2 Centration. iii. Lens 1 Surface 1 To Surface 2 Tilt iv. Lens 2 Surface 1 To Surface 2 Tilt v. Lens 1 Thickness Change vi. Lens 2 Thickness Change

The alignment tolerances listed under “System Alignment Tolerances” include the alignment errors that are the responsibility of the EUSO System Integrator. These include:

i. Lens 1 To Lens 2 Tilt ii. Lens 1 To Lens 2 Centration iii. Focal Plane Axial Displacement iv. Focal Plane Lateral Displacement v. Focal Plane Tilt

Note: The defocus effect due to dn/dt is not accounted for at this time. Additional analysis with a detailed thermal model is required. For the purposes of this report, we assume that the optics will be kept in focus though the operating temperature range.

rms Spot Size: To achieve the desired angular resolution, an rms spot size of ≤ 5 mm is desired. The average spot size, weighted by the observed area at each angle is 5.2 mm. The spot size as a function of angle between the direction to the EAS front and the optical axis is given in the table below. Both the weighted mean spot size and the spot size at all angles slightly exceed the requirement. Work is ongoing on an optics design that uses aspheric base curvatures for the lenses. This design promises to provide smaller spots.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 17/68

Angle (degrees) rms Spot Size in mm

0.00 5.07 5.00 5.61 10.00 6.59 15.00 6.72 20.00 5.70 25.00 4.77 30.00 4.06

Conclusions on Optical Design. At the conclusion of this phase A study, a firm set of optical performance requirements has been established for the optical sub-system. This is expressed as field dependence of the efficacy of the optical system, which relates the photon irradiance at the EUSO Telescope to the number of photons per GTU that a 5mm diameter pixel bucket receives. To attempt to meet these requirements, we have advanced the NASA phase A baseline of a 1.9 EPD f/#=1.25 design presented at the Mid-Term Review by applying LLL optimization, opening the EPD to 2.3m and decreasing f/#=1.25 to f/#=1.00. This design is achromatic across the spectral range 330nm ÷ 400nm. This optical design is within 13% of meeting the specified performance. Optical design concepts are being explored that will, if successful, exceed the throughput requirements see section 4.1.1.8 ???.

Figure 5.2.1.8-1 Transmission efficiency change with surface roughness. Solid blue, green and red lines correspond to single surface at wavelengths 337, 357, & 391 respectively. Dotted line is average accumulation of the four surfaces.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 18/68

Figure 5.1.2.8-2 Diffraction efficiency into the first-order for the three wavelengths as a function of the rms error in the grove depth.

5.2.1.9 Advanced Designs The US phase A ended with a baseline design for the optical system with EPD of 1.9m and f/#=1.25, which was reported in the ESA phase A Mid-Term Report. In the Mid-Term Report design, the rms spot size is minimized across the FS. In the interim US phase A/B period, efforts to improve this design have continued. Four areas for improvement are being investigated. These investigations have lead to advanced designs that are reported in this ESA phase A Final Report. First, new data has been obtained for less dispersive PMMA. Second, optical systems with larger EPD and lower f/# have been explored. Third, a new optimization based an increased performance at the large field angles has been employed. We call least light loss (LLL) optimization. Finally new designs that use on aspherical base curvatures for the Fresnel lenses are being investigated. Our concentration continues to focus on improving the throughput of the system, improving the angular resolution and reducing the manufacturing risks. The baseline design reported here has incorporated several of these advanced approaches. • A better grade PMMA with a dispersion of 1.83% from 310nm to 40nm. All our designs until now have used a dispersion of 4.25% in the same range. • The use of anti-reflection coatings that increase the total throughput by 5%. • Increased the aperture stop. A gain in throughput can be obtained by increasing the optical aperture. This increase in aperture however must be coupled with our new optimization routine LLL. The increased aperture designs achieve spots where all there rays at the highest FoV lie within the 5mm diameter spot. The stop position is optimized for minimizing the energy vignetting (i.e. rays that pass though the entrance pupil but miss one of the lens). We have demonstrated designs that achieve around 30% increase in the triggering throughput at full FoV based on a 2.3m EPD compared to the 1.9m EPD Mid-Term baseline. Several of the advanced design concepts have not yet been fully investigated. These include the following: • Aspheric base curvature instead of a spherical base curvature. Preliminary analysis shows that this can result in a 35% decrease in spot size measured as encircled energy. With smaller spots, chromatic

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 19/68

blurring may be small enough that one or both of the diffractive micro-gratings can be eliminated. This alone would improve throughput by 10%-15%. • Current designs have not optimized the facet back-cut angles. Analysis will be conducted to determine how much the throughput of the system can be increased if the facet back-cuts are made parallel with the light rays passing through the facet. Throughput can possibly be increased 10%.

Development of Least Light Loss (LLL) optimization: The firm definition of the triggering throughput requirements in terms of the OS triggering efficacy [i.e for a given incident photon radiance (photons per square meter) at EUSO, the required number of photons within a 5mm diameter circle surrounding the focal spot on the focal surface] leads to a firm basis for optimizing the optical design. The Science Requirement for triggering throughput emphasizes high throughput at large field angles to insure satisfactory photon statistics for 1020 eV events. Opening the aperture from 1.9 to 2.3 meter naturally results in the collection of 46% more photons; however, in our previous analyses, these extra photons were either blurred into a larger spot on the focal surface. Some even missed the second lens completely, the so-called vignetting loss. The net gain in triggering throughput was minimal. However, by employing our new LLL optimization, which weights the off-axis throughput and minimization of off-axis vignetting loss very heavily, we can achieve optical designs with substantial increase in the number of photons in a 5 mm diameter spot. This optimization technique has proven so powerful as to yield nearly field-angle-independent triggering efficacy and which compensates for the part of cos3(θ) losses at filed angles, θ. We anticipate that further work on these advanced design concepts will yield optical systems with even greater triggering sensitivity.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 20/68

5.2.2 The Focal Surface Photo-Detector The design of the Photo-Detector (PD) for EUSO is a challenging task due to the many requirements and constraints. In particular the accommodation on the ISS has to be compatible with the (limited) resources available, including mass, volume, power and telemetry. Possibly unconventional solutions might be required. The Focal Surface (FS) Photo-Detector (PD) must detect near ultraviolet photons in the 330nm ÷ 400nm wavelength range with single photon counting capability. The PD must be capable of acquiring time-resolved images of EAS with the frame rate of ∼2.5 µs (the GTU: Gate Time Unit). It should resolve single photon signals with a double-hit resolution of about 10 ns. It should be also reliably and stably operate in the space environment within the limited power budget and for required EUSO Instrument lifetime. The PD, is made of autonomous and independent modules, the Photo-Detector Modules, (PDM) installed on a common supporting structure. Each PDM is made of: • the sensors (MAPMT); • the optical adapters or light collection systems (LCS), whenever required; • the Front-End electronics; • the sections of the read-out, trigger, control and data-handling electronics which is decentralized and performed at the PDM level; • the components required for safety/emergency operations (i.e. light sensors, temperature sensors, dust/contamination sensors; • the required harness, local and I/F harness; • the interfaces among the different components of the PD and those between the PD and the rest of the Instrument; • the housing for all the above-mentioned components and the focal surface PD supporting structure; • all the ancillary components required for the proper operation, monitoring and control of the PD, including the parts pertinent to the PD of the Power, Thermal Control and Monitoring, Alignment and Self-Calibration systems. The baseline design of the PD on the FS is a closely packed mosaic of multi-anode photo-multipliers (MAPMT), approximating the curved focal surface, with a bialkali photo-cathode deposited on the UV-glass entrance window. The MAPMT features single-photon counting capability with fast response and high gain. Photo- Multipliers are a well-established and mature technology featuring reliability in space application and availability at the commercial level with reasonably low cost [RDFS1,2,3,4]. The maximization of the effective photon detection efficiency of the FS is a strong requirement both in the choice of MAPMT and the FS design. We define the FS efficiency as:

εPD = εFILL εSAR εDQE

where εFILL is the fraction of the geometrical occupancy of MAPMT units on the focal surface, εSAR is the ratio of the sensitive area to the area of the MAPMT unit (sensitive area ratio) and εDQE is the overall photon detection efficiency of the MAPMT, including the photo-cathode quantum efficiency and the photoelectron collection efficiency.

5.2.2.1 The FS architecture One of the main driving points of the design is to combine different functions as much as possible, in order to save mass and volume. The packing of the devices has to be optimized to reduce losses in the geometrical acceptance, due to dead regions between the close packed devices, and defocusing effects, originating from a positioning of the sensor at some distance from the ideal FS. A modular structure is preferred. The overall structure shall consist of small autonomous functional units (elementary-cells, EC) assembled in larger modules. The EC consists of a limited number of MAPMT sharing some common resources. The EC can be a thick multi-layered Printed Circuit Boards (PCB). Several EC form a PDM. PDM are independent structures tied to each other by a common support structure and having a shape determined by the layout of the FS. The EC concept is useful because it implements the desired modular structure, it allows the sharing of many resources among four MAPMT, such as the supporting PCB, the HV/LV lines, the MAPMT voltage divider, cables, connectors and electronic chips. This sharing improves the economy and

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 21/68

makes design, production and testing easier. It will be assumed that a module is made of an array of close- packed MAPMT with a suitable shape, possibly surrounded by a border of variable thickness running all around the MAPMT, to leave a sufficient clearance. Of course the detailed design of the module geometry will require the final fine-tuning of the geometry which needs the final design of the main EUSO optics, defining the FS. The PD is assumed to be arranged in a modular structure based on the following tree. • Single sensors (MAPMT). • An EC is both a physical and logical grouping of four sensors. It is an autonomous system with resource sharing; it is the largest item which cannot be reworked. • A PD-Module (PDM) is a physical grouping of a suitable number of EC. The number of EC per PDM might be different for different PDM, according to the geometry. A macro-cell is a logical grouping of a suitable number of EC, for triggering and electronics purposes. The PDM and the macro-cell concepts are a priori distinct. • The FS is divided into a suitable number of identical sectors, each one made of a suitable number of PDM and macro-cells. Each elementary-cell contains four units of MAPMT R8900-03-M36 [RDFS6], additional optical components to increase εSAR (referred to as optical adapters), HV/LV power connectors, an HV voltage divider for any of the MAPMT units, common heat dissipation facilities, common miscellaneous services and monitoring devices, all of the Front-End electronics plus as much as possible of the trigger/read-out electronics integrated in the EC and connectors for signal transmission and control lines. All these components are assembled onto a printed circuit board housing. A suitable number of elementary-cells are assembled as a Photo-Detector Module (PDM) and the PDM will be assembled on the Mechanical Supporting Structure (MSS) to form the Focal Surface Assembly (FSA). The block-scheme of the FS, from the functional point of view, is shown in Figure 5.2.2.1-1.

Figure 5.2.2.1-1 Block-scheme of the PD and its interfaces. Figure quality to be improved

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 22/68

5.2.2.2 The MAPMT The current baseline choice of the MAPMT is the Hamamatsu R8900-03-M36, featuring weak electrostatic focusing between the photo-cathode and the first dynode. R8900 is a modified version of R7600, which was the baseline choice at the beginning of phase-A. The fabrication process is common to both of these types and their reliability in the space environment is at the same level [RDFS5]. The R8900 has a bialkali photocathode deposited on a UV-glass of 0.8mm thick. The photoelectron emitted from the photocathode is weakly focused into the acceptance of a 12-stage stack of metal channel dynodes and amplified by the factor of ∼106 when the high voltage of –900V is applied with a standard tapered voltage divider. The uniformity of the quantum efficiency is at the level of 2%. The quantum efficiency is dependent on the temperature and is ∼20% at room temperature. The photoelectron collection efficiency is ∼70%, independent of the temperature. The εDQE is defined as the ratio of the number of anode signal pulses to the number of incident photons on the MAPMT input window. The number of signal pulses in this document is defined as the number of pulses exceeding the 1/3 of single photoelectron peak in the pulse height distribution. The value of εDQE is measured as: -1 dεDQE / dT = a εDQE, with a = -0.0037°C and εDQE (T=30°C)=0.14 The sensitivity is distributed in visible and near-UV wavelength regions and the quantum efficiency is peaking at the wavelength of 420nm. The near-UV photons in the wavelength region of 330nm ÷ 400nm will be selectively delivered by employing optical filters on the top of MAPMT units.

Figure 5.2.2.1-2 Mechanical drawing of the Hamamatsu R8900-03-M36

The device cross section, in its top view, is 25.7 mm × 25.7 mm (with 26.2 mm maximum side, from manufacturing tolerances) and its length is 27.2 mm as shown in Figure 5.2.2.1-2. About 83% of the device cross section is the sensitive area. The small deficit of sensitivity due to the spacing between the close packed devices can be partly recovered by employing an optical adaptor as described in section 5.2.16. The value of εSAR is ∼0.95 assuming the optical adaptor is a tapered filter glass (BG3) and its top has the same dimension as the device cross section. We note that the εSAR may effectively exceed 1.0 if we extend the top of the optical adaptor in order to recover part of the gap between neighbouring MAPMT, required for engineering reasons. The choice of the pixel size is a trade-off between many different requirements. A smaller pixel size of the FS is preferred for a better determination accuracy of the primary particle arrival direction. However a trade-off among pixel size, number of channels (impacting on the budgets and Instrument complexity) and Science minimum requirements has to be done. Available anode segmentation of R8900 series are 4×4, 5×5, 6×6. Taking the 1 mm spacing between units in configuring the array, the optimal choice is currently considered to be the 6×6 anode format, which gives the pixelization of ∼(4.5mm × 4.5mm) on the FS. However the possibility to use a ∼(3.3mm × 3.3mm) pixel size on the FS, as it was in the original design, is still kept as an option. The pulse rise-time is ∼1.5ns with the transit time spread of ∼0.3ns, which is sufficiently fast to resolve single photon signals arriving in the time difference of 10ns [RDFS6]. The anode capacitance to ground is nominally 2.8pF

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 23/68

[RDFS6]. The gain non-uniformity among anodes in a MAPMT unit is 1:3, and the average gain non-uniformity over different MAPMT units is 1:2. This non-uniformity is within the range of the Front-End programmable threshold and of the high-voltage power supplies (see section 5.2.3). The anode cross-talk between neighbouring anodes in a MAPMT unit is 7% [RDFS6] or, which introduces 0.4 degree uncertainty in determination of arrival direction of primary cosmic rays of 1020eV with the zenith angle of 60 degrees. [AP, OC: so big ???]. The overall anode dark current after 30 min storage in darkness is of the order of 1nA, which is smaller than the night background level by two orders of magnitude. Other relevant specifications of these MAPMT are listed below. • One unit of R8900-03-M64 has 27.3g mass. • Both switch-on and allowed operating temperature ranges are –30°C ≤ T ≤ +40°C. The storage temperature range is –30°C ≤ T ≤ +50°C. • Maximum acceleration is 12.7 G(rms) to ensure the gain degradation of less than 10% and measurable increase of dark current, with the break-point exceeding 20G(rms). The vibration in the frequency range over 1 kHz has a drawback because it increases the dark current level. • Magnetic field influence is less than 0.1 in the relative gain variation up to 0.2mT, which is sufficiently large compared with the variation of geomagnetic field on the ISS orbit. • The anode sensitivity decreases to about 78% of its initial sensitivity after 3+2 years on-orbit at 50% lid opening ratio. • The ageing for 3+2 years on-orbit is 0.985, ensuring the gain degradation of less than 25%, assuming 4.6 µA average anode current, corresponding to the expected random background and 50% lid opening ratio. Even considering an average anode one thousands times larger and 100% lid opening ratio, the reliability is 0.936. • The influence of the radiation is dominated by the degradation of the entrance window transparency. The transparency may be degraded to 90% after receiving the dose beyond 1.4x105R, which is far smaller than the expected dose of 103R during 5 year operation on the ISS orbit.

5.2.2.2.1 The MAPMT voltage divider The baseline MAPMT sensors (Hamamatsu R8900 series) are 12-dynode MAPMT. The manufacturer’s suggested divider ratio is: ( PK (3,2,2,1,1,1,1,1,1,1,1,2,5) A ). This is assumed to be the optimal one, albeit further studies will be performed in order to confirm this choice for EUSO. In the present baseline each MAPMT has its own Voltage Divider (VD) which is implemented on the EC base-board PCB. A bias scheme with grounded anodes is assumed, in order to simplify the connection of the PMT anode to the Front-End electronics, so that the cathode must be supplied a negative high voltage. The MAPMT will be operated at the lowest possible voltage in order to minimize their aging taking into account that the real amount of light that the MAPMT will receive on-orbit is difficult to 5 quantify. The typical operating value is therefore assumed to be V0 = –850 V (nominal gain 5.3⋅10 , with a range of values from V0 = -800V to V0 = -1000 V. The bleeder current and decoupling capacitors at the last dynodes were chosen in order to minimize power consumption while handling the signals properly. Currently a triple HV power supply is identified as the baseline, while a trade-off with a double power supply is on-going. In the current design (triple voltage supply) the voltage divider is dimensioned in order to have a bleeder current as large as 50 times the average anode current due to the background, while the short signal pulses are handled by the decoupling capacitors in the last stages. The bleeder dimensioning takes into account that for a E = 5·1020 eV EHECR the maximum number of photo-electrons per pixel per µs is about 20, while a one thousand photon pulse is easily handled by a few tens of pF decoupling capacitors, with a relative gain change at the percent level. The power consumption of each MAPMT will be 9.6 mW with an expected worst-case relative gain change of 0.05 when operating at V0 = -1000 V. The details of the calculation can be found in [SSFS].

5.2.2.3 Elementary-Cell

Each elementary-cell contains 4 units of MAPMT R8900-03-M36, optical adaptors, HV/LV power connector, HV dividers, Front-End ASIC chips, connectors for signal transmission and control lines assembled on a printed circuit board.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 24/68

The four MAPMT units are located on a base-board together with the passive electronics of the voltage dividers. The MAPMT can be inserted in the upper part of the socket. The lower part of the socket contains a ball grid array that is soldered to the elementary-cell board. A photograph of a prototype of the elementary-cell base- board is shown in Figure 5.2.2.3-1.

Figure 5.2.2.3-1 A prototype of elementary-cell base-board equipped with ball grid array sockets.

The base-board is fixed to the PDM supporting back-planes, by means of screws and dowel pins. This allows a precise relative positioning of the different EC onto the PDM and redundant support. The base-board is fixed on the back-plane supporting structure of the PDM by means of dowel pins and screwed by means of steel screws. Screws and dowel pins fit into metallized holes drilled into the PCB. They also act as thermal bridges to convey the heat generated on the base-board away from it, to the PDM. As any EC is totally autonomous and it is individually fixed to the PDM supporting back-planes, we may freely choose the shape of the PDM, possibly including curved shapes. The signals from the MAPMT anodes are carried on the back-side of the thick base-board where to a Front-End chip housed on the back-side of the base-board. The MAPMT side of the base-board also houses the components of the voltage dividers to power the dynode chain. Up to one voltage divider per MAPMT (i.e. four in total) can be housed onto the base-board. The thick base-board includes a copper layer, 0.5 mm thick, to help to drain away from the base-board the heat dissipated on the base-board by the voltage dividers and Front-End electronics. This can be accomplished by conduction, through screws, pins and, possibly, dedicated heat bridges. Tests on the prototypes were also carried on, confirming the expectations. By using a surface-mount housing the thickness of the baseboard can be kept relatively small, compatibly with the mechanical requirements, thus saving mass. Assuming to avoid a direct soldering of the MAPMT one needs a suitable socket. A prototype socket, for surface mounting, was produced. To ensure that the mechanical resistance and electrical contacts are kept all life long, and to ensure electrical insulation, good thermal conduction and good protection, both the volume between the MAPMT and socket and the volume between the socket and the base-board PCB shall be potted. Moreover potting between MAPMT and base-board and all around the four MAPMT will ensure electrical and mechanical insulation, mechanical damping, structural strength, components containment and, possibly, light tightness and thermal conduction. The impossibility to perform a complete visual inspection can be overcome, for instance, by X- ray inspection techniques and/or by defining a suitable alternative testing functional procedure, both for the MAPMT connections and for the ASIC. X-ray inspection has been already performed on the prototypes.

5.2.2.4 Photo-Detector Modules (PDM) and the Mechanical Supporting Structure (MSS) Elementary-cells are assembled in PDM. PDM are assembled on the MSS into an FSA. Following a suitable layout of the MAPMT on the FS, the EC are installed onto the PDM supporting back-plane, by means of screws and dowel pins. Cables are routed to and from the EC via holes left on the PDM back-plane. Each PDM back-plane is fixed onto the FS supporting structure via a suitable number of screws, depending on the shape and dimensions of the PDM itself. Screws also help to drain the heat away from the EC. Tests on the prototypes were also carried on, confirming the expectations. PDM design depends on the design of FSA. Two types of FS layouts have been developed at a comparable level. They are referred to as X-Y layout and r-ϕ layout as schematically shown in Figures 5.2.2.4-1,2,3 for the XY layout and in Figures 5.2.2.4-4,5,6 for the r-ϕ layout. Both of them are capable of

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 25/68

approximating the curved focal surface within the required tolerance. Note that in both cases the central part will not be used because it will be housing components of the Calibration sub-system (see section 5.4).

Figure 5.2.2.4-1 Design of the FS Figure 5.2.2.4-2 Backside view of Figure 5.2.2.4-3 Design of the supporting structure for the XY layout. the FSA in the XY layout. PDM for XY layout.

Figure 5.2.2.4-4 Design of the FS supporting Figure 5.2.2.4-5 A possible Figure 5.2.2.4-6 Design of the structure for the r-φ layout. arrangement of PDM in the PDM for r-φ layout. r-φ layout (six identical sextants).

The FSA will be attached to the main EUSO structure with 12 attachment points, which are equally spaced on the perimeter ring of the MSS. The MSS is designed to have the first modal frequency higher than 50Hz. The back- side of the MSS will house the harness as well as the TCU. It will also provided the required light tightness. The temperature uniformity is expected to be about ±2°C during the observation period assuming uniform thermal dissipation over the FSA. The 2°C temperature non-uniformity causes a thermo-mechanical deformation of the FS supporting structure by 0.032 mm in the centre along the Telescope axis. The allowable temperature range is ±60°C which corresponds to the 1 mm tolerance of the distance between the OM and the FSA. The 93% of the total FS corresponding to the ±30 degree FoV is covered by PDM (εFILL=0.93) using 5544 MAPMT units.

Some more details on the mechanical design of the supporting structure and a few words on the thermal aspects would be helpful….

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 26/68

5.2.2.5 Summary of the main FS PD parameters A summary of the main parameters, for a typical layout choice, is shown in Table 5.2.2.5-1. The resulting physical scheme of the FS is shown in Figure 5.2.2.5-1.

LAYOUT r-ϕ X-Y Number of MAPMTs 5544 6272 Number of EC 1386 1568 Number of PDM 6×10 128 Number of different PDM shapes 10 9 Average number of EC per PDM 23.1 (= 92.4 MAPMT) 12.25 (= 49.0 MAPMT)

Table 5.2.2.5-1 Summary of the main parameters for typical layout choices.

Figure 5.2.2.5-1 Physical scheme of the FS.

5.2.2.6 Focal Surface Systematic Errors

We obtain εFS = 0.12 at the temperature of 30°C in the baseline design described above. The influence of unexpected changes or misidentification of the FS performance to physics quantities can be evaluated from the systematic errors arising from the uncertainty in the value of εFS. The systematic error of εFS results in the uncertainty of the scale in determination of primary energy of cosmic rays.

The requirement of |∆εFS / εFS| ≤0.05 imposes additional requirements in the ground-base characterization, the adjustment accuracies of Front-End electronics and the accuracy of the on-board calibration.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 27/68

• The uncertainty in the MAPMT efficiency must be characterized and monitored within |∆ε3 / ε3| ≤0.05. • The threshold level of the Front-End comparator must be adjusted with the accuracy of 0.25 relative to the threshold level assuming nominal threshold is set at the 1/3 of the single photoelectron peak [RDFS6]. • The high voltage must be adjusted or controlled with the accuracy of ±24V or better when the high voltage is –900V. • The temperature of the photocathode must be monitored with the accuracy of ∆T ≤ ± 13.5°C. • Anode cross talk must be characterized with the accuracy better than ±5%. • The gain uncertainty of dynodes and pre-amplifiers must be characterized with the accuracy better than ±5%.

5.2.2.7 Procurement and assembly Procedure The MAPMT will be procured after the screening of the gain non-uniformity of 1:3 in each MAPMT unit and the gain non-uniformity of 1:2 over different MAPMT units. Delivery is expected to be two years after the contract. The MAPMT will be characterized by measuring the efficiency and gain maps as a function of the incident light angle, temperature and high voltage for about 1% of whole units. The detailed characteristics on mechanical robustness, aging level, after pulse effects will be studied for them. These detailed characterizations will be simplified for the remaining units if the unit-to-unit variation of characteristics can be reasonably estimated from limited number of characteristics such as the anode uniformity and gain for normal incident photons. Otherwise, all the units will be characterized in detail. The characterization data will be used to select MAPMT units to be assembled into elementary-cells and PDM so that HV and Front-End electronics can be adjusted within the adjustment requirement described in section 5.2.3. It will also be used for the correction of the possible change of FS performance on-orbit using the environmental monitoring and calibration. The MAPMT will be assembled into elementary-cells and PDM after the selection. The photo-detection performance will be checked by illuminating each PDM placed on the focal surface position through the prototype optics to be fabricated during phase-B. The tested PDM will be assembled into an FSA and calibrated after assembly as a complete set of EUSO Telescope. The detection efficiency will be characterized with the accuracy better than 3% before launch, and will be monitored with the accuracy better than 5%.

5.2.2.8 Further Options The Flat Panel PMT (Hamamatsu R8400) might be available in a short time. The device cross section is 52mm x 52mm and the sensitive area amounts 89% of the device cross section. This is more suitable for maximizing the value of ε2 since we can decrease the insensitive device spacing. In addition, it has more flexibility in anode segmentation ranging 8x8 through 16x16, corresponding to the pixel size of 6.6mm through 3.3mm according to the improved metal channel dynode with the finer channel pitch. This option requires minor engineering changes, since four units of R8900 can be approximately substituted by one unit of R8400 in an elementary-cell.

5.2.2.9 The Optical Adapter (Light Collector System, LCS) The ∼83% of the PMT R8900 cross section is the sensitive area. The small deficit of sensitivity due to the spacing between the close packed devices can be partly recovered by employing an optical adaptor. The optical adapter consists of the simple tapered glass, which can be also used as an optical filter. The optical adapter is glued to the

MAPMT window using the candidate epoxy EPOTEK 301-2. The value of εSAR is ∼0.95 assuming the optical adaptor is a tapered filter glass (BG3) and its top has the same dimension as the device cross section. We note that the εSAR may effectively exceed 1.0 if we extend the top of the optical adaptor in order to recover part of the gap between neighbouring MAPMT, required for engineering reasons.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 28/68

Figure 5.2.2.9-1 Preliminary concept study of the optical adapter (purple part).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 29/68

5.2.3 The Electronics

5.2.3.1 Architecture of the EUSO electronics The EUSO electronics is designed to perform normal data taking operation, monitoring and control of the EUSO Instrument and it can be divided into the following main parts. a) The System Electronics (SE) sub-system: this includes the system level trigger and all the electronics needed for the Instrument operation, monitoring and control. This sub-system is housed in a standard electronics box, the TCU (Trigger and Control Unit). b) The FS electronics: including all the electronics needed to bias and read out the MAPMT (described in section 5.2.2.2.1), the Front-End electronics, the first levels of triggering logic, the Macro-Cell electronics and the electronics for the FS monitoring. All this electronics is conceived as being physically part, of the Focal Surface Assembly1. c) The Atmosphere Sounding unit electronics (see section 5.2.4). d) The electronics of the Calibration System (see section 5.4). e) The electronics needed at system level, including the Thermal Control System electronics and the lid control electronics (see section 5.2.7). This section will describe the first two items, as they are the ones directly related to the EAS detection and they need to be treated in a logically unified way (Front-End, read-out and trigger electronics). The other items will be described in the relative sections.

5.2.3.2 The EUSO electronics design and requirements The design of the EUSO electronics is a challenging task due to the following main requirements and constraints [RDEL1,RDEL2]. • Volume limitations, imposed by the close packing of a large number of sensors with a large number of channels on the FS, and power limitations, imposed by the limited available power budget per channel, both suggest the design of a custom ASIC for the Front-End electronics and the first levels of the triggering logic. The power consumption largely depends on the level of complexity and richness of functions that will be implemented in the final devices. The preliminary design and the measurements done on prototypes have shown that the minimum power can be as low as 1.2 mW per channel, at the Front-End level, compatible with the available power budget. Depending on the complexity of the Front-End electronics, and in particular the size of the analogue section, the power consumption might be significantly higher. • The electronics must be radiation tolerant at the level required by the space environment for the mission time. It must be either intrinsically latch-up free or a suitable anti latch-up system must be foreseen in the EC circuitry. All digital and control sections must be protected against SEU (Single Event Upset) in order to guarantee data integrity. The ASIC technology must be chosen in order to fulfil the above mentioned requirements. Already certified technologies should be preferred. If that is not feasible an appropriate qualification program will be carried on. • The use of a binary Front-End electronics is preferred in order to save power, while achieving the single photon counting capability required by the Science and simplify the design. To avoid spoiling the detection capabilities this choice dictates a fast Front-End electronics, in order to reduce the effect of pile-up.

1 The Front-End Electronics (FEE) and the first levels of triggering logic will be housed in a single custom ASIC. This ASIC, called MARS (Multi-Anode Readout System), will include both digital and analogue functions. During phase A, two separate Working Packages called DFEE and AFEE (Digital and Analogue Front-End Electronics) were developed fro the preliminary study. In this document the two work packages will be treated as a single project as they will be from phase B.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 30/68

Relation to single photon counting ? The FEE must be able to count single photoelectrons with minimum inefficiency. As a consequence the preamplifier stage, the noise figure and the discriminator must be good enough to allow setting a threshold well below the amplitude of a single photoelectron signal given by the MAPMT. A challenging goal is to be able to set the threshold to ¼ of a photoelectron signal with 10% precision. The FEE must be able to count photons at very high speed. In case of a typical E = 1021 eV proton induced EAS at 45° zenith angle falling at half of full FoV the detected photon rate can be as high

as RMAX ≈ 50/µs. The double hit resolution (i.e. the minimum time above which two signals are resolved) should be compatible with this figure and therefore a goal of T2HIT = 10 ns double hit resolution is set. It is very important to have good linearity so as to provide a correct energy measurement up to these values. The single photon counting technique is still efficient for triggering at this level but it may have severe linearity problems because the counting efficiency depends on the photon rate. The counting rate can be further increased when there is reflection from clouds with a large albedo. The dynamic range will be thus extended by integrating the electric charges using analogue electronics. • The energy measurement, especially in the low energy range, depends critically on the photon counting efficiency, i.e. it is very important to take under control the PMT gain and the threshold level. The Analogue Front-End Electronics (AFEE) will allow the calibration of the gain of MAPMT and thresholds of discriminators, it will provide the corrections to the binary electronics counting losses and it will allow gain stability control. An automatic calibration procedure will be defined. Storage of the pedestal values and gain values will be needed. • The electronics must be able to withstand the background rate, keeping the capability to trigger on useful signals with a limited dead time induced by fake triggers processing.

5.2.3.3 Front-end, read-out and trigger electronics design description The EUSO Photo-Detector will have to distinguish the well defined space-time contiguity structure produced by the EAS from the random background and other space-time correlated background of different origin (see section on background [???]). The random background expected at the PD level is ~(0.3 ÷ 0.5) detected photons per µs per pixel in moonless conditions. In addition to that luminous phenomena in the atmosphere (airglow, blue jets, ...) and human activities might produce a background few order of magnitudes higher at some localized areas and times.

5.2.3.3.1 The principle of measurement and triggering The trigger system has to be able to provide a fast trigger with hundreds of thousands of channels. It has to be selective in order to tag the EAS produced signal while rejecting the background in an efficient way. The trigger system [REF] is a critical point of EUSO. The envisaged system is modular, based on the logical concept of the Macro-Cell, which is an assembly of sensors, logically working as a single entity as far as triggering purposing are concerned. The trigger module has been studied to provide different kind of triggers including a normal mode for EAS detection, a slow mode for atmospheric phenomena and a special mode for detector calibration. The read-out electronics has been designed to obtain an effective reduction of channels and data to read-out, developing a method that reduces the number of the channels without penalizing the performance of the detection system. Rows wired-OR and columns wired-OR connections have been adopted inside every single Macro-Cell for diminishing the number of channels to read-out while keeping the triggering capabilities. The basic method of measurement and triggering exploits the single photon counting technique. A charge integration approach is also implemented to handle signals at high energy and from strong Čerenkov light. By using both analogue and digital electronics, the detection is possible with a threshold as low as a fraction of the mean amplitude of the single photo-electron signal up to about four thousands photo-electrons. As described in Section 5.2.2, MAPMT are physically grouped into Elementary-Cells made of four MAPMT. A suitable number of Elementary-Cells form a Photo-Detector Module (PDM). The triggering logic is based on almost independent modules called Macro-Cells, each with its own electronics. Rows wire-ored and columns wire-ored routing connections have been adopted inside every single Macro-Cell. Each pixel can be identified by its position in X and Y in the Macro-Cell, diminishing the number of information that needs to be read-out. A free running method has been adopted to store temporarily the information in cyclic memories and recover the relevant data at the time that a good trigger signal occurs. To reject the background, the EUSO electronics operates with several trigger levels.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 31/68

• Each time the signal coming from a MAPMT pixel exceeds an analogue threshold, a fast discriminator recognizes the arrival of a single photo-electron event. The pulse is counted by a pixel-level 10-bit counter. • When a pre-set counter value is reached within a given GTU (Gate Time Unit, the programmable basic time unit), a pixel-level trigger is issued, the (X,Y) lines of the pixel are marked into the (X,Y) memory and the pulse counting output is enabled during the remaining GTU time. • When a pre-set counter value at the Macro-Cell level is reached within a given GTU, due to the pulses coming from the enabled pixels, a Macro-Cell level trigger is then issued. A hard wired (X,Y) proximity device for adjacent GTU at the Macro-Cell level further rejects fake triggers in presence of large photon background. • The software reconfigurable System Trigger will continuously monitor the Macro-Cell level trigger activity searching for time-persistency patterns. When the given software criteria for a valuable pattern are met, the System Trigger will issue a system–level “alert” trigger that will activate the Macro-Cell proximity device which will start a very fast “track finding” (TF) process using the information of X and Y lines. At the end of the TF process, through a command line, the result will be issued to the System Trigger enable it to start the read-out sequence after a preset exposure time. Three trigger modes are considered in the present base line: • Normal mode with a GTU of 2.5 µs for routine data taking. • Calibration mode, with a GTU of_____,(FAST or SLOW ???) for calibration runs. • Slow mode with a programmable GTU up to a few ms, for the study of meteorites and other atmospheric luminous phenomena. An auto-level-trigger function within the system trigger is also under consideration. This would allow the Instrument to be set at the optimum trigger levels in case of varying background conditions due to slowly transient phenomena (moon phases, varying cloud coverage, large urban areas, …).

5.2.3.3.2 The Front-End electronics The Front-End electronics must amplify the small signals from the sensors with a limited available power and without spoiling the fast sensor response. A programmable threshold is required to discriminate the signals. It is required to mask noisy or bad channels, in order not to compromise the trigger operation. Detected photons must be counted for each channel during an externally driven time interval (the Gate Time Unit, GTU) in order to fire the first level trigger when the number is above a programmable digital threshold. The required logical signals for the trigger should be generated. The number of photons detected per channel must be stored for a sufficient time interval waiting for read-out in case of trigger. Required features are a very compact design with minimal distance between the sensors and the Front-End electronics, a completely modular system with minimal cabling and self- triggering capabilities. The EUSO Front-End electronics will be implemented in the form of an Application Specific Integrated Circuit (ASIC).The Front-End ASIC will provide both the MAPMT signal interface and the pixel-level trigger. A simplified block-diagram of the baseline ASIC implementation, referring to a 36 anodes MAPMT, is shown in Fig. 5.2.3.3.2-1. The main functions of the EUSO ASIC are the following:

• Collect the anodic signals via DC coupling. • Discriminate the amplified anodic signals above a programmable analogue threshold. • Provide a photon counting capability with about 10 ns double-hit resolution. • Compare for every pixel the counted events within the GTU with a programmable digital value. • Enable the next incoming pulses from that channel to be routed out along a fast digital OR channel and activate the associated (X,Y) address lines. • Store the number of photoelectrons counted in each GTU into internal ring memories and allow readout of these memories when a trigger occurs; the depth of these memories will be large enough to allow waiting for the trigger. Current baseline is 128 GTU depth.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 32/68

• Integrate dynode signals and store in analogue memories. When a trigger occurs, the charge stored into the analogue memories will be transferred to an external ADC (housed in the EC very near the ASIC) and converted to digital data that is sent to the PDM through the serial line. • Accept commands, parameters and settings from a serial line and read-out through a serial line. • To better fulfil the science requirements a set of anode analogue channels working with a principle similar to the dynode one, might be included if this happens to be compatible within the power and volume budgets.

Figure 5.2.3.3.2-1 EUSO ASIC baseline block diagram.

ASIC design, development and prototyping The design of the ASIC is critical mostly because of the strong constraints imposed by the very limited power budget. During the phase A the INFN Genova group and the ISN Grenoble group have performed a detailed study of both the digital and analogue section of this device, building a few small prototypes in order to check simulation results. The critical sections of this project, particularly the front end pre-amplifier, the channel discriminator, the threshold DAC, the charge integrator and the analogue memories, have been designed, simulated, built and tested. The results of these studies are described in detail in [RDEL2]. These studies have proved that it is possible to implement the required Front-End functions, digital functions (programmable single pixel threshold, digital counters, X and Y trigger logic, internal memories 128 GTU deep) and dynode analogue functions while keeping the total power requirements below about 1.2 mW per channel. Simulations and prototype measurements have shown that double hit resolution of the order of 10 ns is possible

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 33/68

within these constraints. All prototypes have been built using 0.35 µm, 3.3 V technology, suitable for space qualification.

5.2.3.3.3 RO&C Functional and architectural description The preliminary architectural block diagram of the Read-out & Control Board is sketched in Figure 5.2.3.3.3-1 for a Macro-Cell made of nine Elementary-Cells.

Figure 5.2.3.3.3-1 Preliminary architectural block diagram of the Read-Out & Control Board.

The basic structure of the RO&C electronics contains the following items. • The Elementary-Cell Array interface, mainly the (X,Y) and photon counting ring memories and the analogue Front-End electronics ring memories, some strobe lines and the communication bus. • The auxiliary trigger logic, track-finding device. • The analogue Front-End electronics interface. • The Power Supply System interface, mainly the power lines and some command and housekeeping lines. • The TCU interface, mainly the proper set of serial-parallel converters as well as driver and receivers. • A number of 8-bit counters (one for each trigger segment) and a digital threshold register used to generate the Macro-Cell trigger. • Digital multiplexers connected to a parallel-serial register used during data download. • Command decoder and router logic. • A service ADC (with an analogue multiplexer) used to collect the PDM data (voltages, temperatures, ...).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 34/68

In the present baseline, all the interfaces between Macro-Cells and TCU are digital balanced differential lines based on a RS422-like model (possibly implemented as LVDS). An option under study is to implement an optical (IR) link between the PDM and the TCU to reduce the harness mass. The System Trigger logic is mainly implemented within an FPGA to allow flexibility. The FPGA itself has to be SW-reconfigurable by proper writing of its configuration registers. Most of the read-out and control blocks will be implemented into a dedicated FPGA hereafter named Control Logic FPGA. The Control Logic FPGA will have the following functions: • Generate the ASIC setting parameters. • Generate the clock signal to reset the internal counters of all the ASIC and Macro-Cells at the beginning of each GTU time window and to load the (X,Y) lines and photon counting counters into Macro-Cell ring memories at the end of each GTU window. • Manage the writing of the memories and the address generation for the ring memory operation. • Manage the reading of the memories and the address generation during data download. • Drive the multiplexer address lines when the data are streamed from the memories into a parallel-in-serial- out register. • Send the telemetry data to the TCU by means of the serial interface.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 35/68

5.2.3.3.4 The EUSO trigger

The triggering logic was first proposed in [RDEL1]; it is fully described in [DREL] and [xxxxx]

5.2.3.4 The system electronics The system electronics (SE) [REF], is made of: • the Central Control Unit, controlling the Instrument at System level; • the Trigger Control Unit performing the final level of triggering and Data Handling; • the Power System control; • the P/L interface Unit; • the required harness, local and I/F harness; • the interfaces among the different components of the SE and those between the SE and the rest of the Instrument; • the boxes housing all the above mentioned components; • any other ancillary component required for the proper operation, monitoring and control of the EUSO Instrument, including the parts pertinent to the SE of the Power, Thermal Control and Monitoring, Alignment and Self-Calibration systems.

5.2.3.4.1 The control electronics The control electronics [REF] is in charge of managing the operations of the Instrument. The EUSO Instrument baseline includes a nominal and a redundant (cold) TCU which also manages the interface with the CEPF. The TCU will include a suitable micro-processor with dedicated SW running on it for both common services and for science data processing. A survey of all its envisaged functions is given in the following. • System Trigger as well as Trigger and Control Handling. • Interface with the different Instrument components. • reception, validation and distribution of telecommands. • Collection and pre-processing of the scientific (consisting of the position and arrival time of detected photons) and HK data, to check the correct configuration and operation of the Instrument. • Preparation, storage and transmission of the Scientific and HK telemetry packets. • Management of the power distribution and the configuration to the PDM. • Management of the lid mechanism (motor drivers and position sensors), of the Thermal Control System (heaters drivers and sensors). • The control of the Instrument operative modes during observations, diagnostic and calibration intervals (these modes also include the autonomous maintenance of the detector in safe conditions). • Management of the Atmospheric Sounding Instrumentation. • Management of emergency situations. • Provision of data patches and dump capability for on-board software programming.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 36/68

5.2.4 The Atmosphere Sounding (AS) Instrumentation The AS sub-system will provide information about the clouds location and the atmospheric transmission and attenuation of ultraviolet fluorescence and Čerenkov signals generated by the EECR in order to provide correcting factors to the EUSO Telescope measurements. The AS sub-system, as a fall out, will, also, afford global atmospheric sounding from the ISS platform, giving the opportunity to acquire data of interest for atmospheric physics studies, planetary radiative balance and climatologic studies. In particular, to accomplish the objectives of EUSO experiment, the AS sub-system will be based on a LIDAR system and it will include a scanning device (coelostat) which will allow to point the LIDAR toward those atmospheric regions where an EECR event has been previously detected.

5.2.4.1 The LIDAR The proposed implementation of the LIDAR for EUSO is described in [ASDR].

5.2.4.1.1 EUSO Requirements The EUSO mission requirements are strongly related to the definition of the AS sub-system requirements for the following reasons. The precision achievable of the EUSO measurements will be influenced by the absorption, scattering and reflection processes of the light (fluorescence and Čerenkov) on its flight path from the production point to the EUSO Instrument. These effects will depend on the presence of relatively variable atmospheric constituents, namely gases, aerosol, clouds. The following atmospheric factors are of importance. • The attenuation of the fluorescence light in its path to EUSO, caused by absorption and scattering by the above mentioned air constituents, will result in a biased measurement of the amount of total light created by the EAS and hence of the initial energy of the primary particle inducing the EAS. • For the detection of the Čerenkov light the presence of a well-defined scattering layer, such as the ground, the sea or other water surface or a sharply defined cloud top is critical. An incorrect evaluation of the altitude at which the Čerenkov light is reflected will bias the identification of the identity of the primary particle and its energy. • The amount of Čerenkov light detected by EUSO can also be used as an independent measure of the energy of the primary particle. The precision with which the reflection properties (albedo) of the reflective medium as well as the extinction properties of the intermediate medium will be known, shall determine the quality of the measurement. [AP: The angular dependence af the reflection coefficient is normally strong….We should explain how this can be done.] Several models have been implemented to evaluate the effect of real atmosphere perturbation of both the fluorescence and Čerenkov signals generated by an EAS [RD6 and RD13] and the outputs from that models clearly show the strong distortion that could be introduced by different typical atmospheric condition, especially due to the presence of clouds and aerosol layer whose distribution and optical depth are extremely variable. The available time (about 30 s on average) should allow for a reasonable scanning of the EAS trajectory in order to retrieve the atmospheric condition, both of the atmospheric region interested by the EAS and along the optical path from the EAS to the EUSO receiver, that affect the intensity of the observed fluorescence and Čerenkov signals. Systematic knowledge of the atmospheric condition where an EAS has been occurred is necessary to fulfill the EUSO Scientific Requirements.

5.2.4.1.2 AS Scientific Requirements From the above considerations the following scientific requirements have been identified for the AS sub-system: AS-SR1: Detection of the presence of clouds and/or aerosol layers along the direction of observation of the detected EAS. AS-SR2: Detection of the landmark. AS-SR3: Determination of the thick clouds top level with at least 100 m vertical resolution. AS-SR4: Retrieval of the optical depth of semi-transparent clouds with at least 30% of accuracy.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 37/68

AS-SR5: Retrieval of thick clouds reflectance. AS-SR6: Retrieval of clouds and aerosol extinction/scattering properties

5.2.4.1.3 AS Performance Requirements In order to fulfill the above mentioned scientific requirements related to the atmospheric sounding operations and the EUSO Instrument Requirements, the AS sub-system should have the following performance that will be referred as Atmospheric Sounding Instrument Requirements (AS-IR). AS-IR1: Sensitivity that allows to measure the molecular backscattering signals from up to 20 km a.s.l. with a signal to noise ratio better than 2 within 1 s. This is needed to attempt a reasonable analysis of the LIDAR data in order to retrieve the atmospheric properties of interest. AS-IR2: Capability to acquire all the dynamic range of the backscattering signal that typically extends for several decades from single photon counting rate, because this is the possible intrinsic intensity variation of the acquired backscattering signal, for a single laser pulse echo, when both the weak molecular echo from the higher atmospheric level and the strong echoes from the ground and/or from the optically thick low level clouds are present (RD6 and R13). AS-IR3: Sampling rate of the backscattering signals better than 1.5 MHz in order to achieve a range resolution better than 100 m, considering the two way travel of the LIDAR signal to and from the backscattering volume. AS-IR4: Simultaneous acquisition of backscattering signals at multiple wavelengths and different polarization. This is very important, if not essential, because it represent the only possibility to unambiguously identify the presence of aerosol/clouds and to retrieve some information about the size distribution of the observed aerosol/clouds to which the aerosol extinction properties and consequently the amount of the perturbation introduced on the observed fluorescence signals are strongly dependent. AS-IR5: Pointing capability within few seconds of time from the detection of an EAS with a pointing accuracy better than 1 mrad, in order to scan the EAS trajectory with at least one pixel resolution for a reasonable time until the region of interest is within the FoV of EUSO.

5.2.4.1.4 Sub-system Definition We define as Atmosphere Sounding sub-system (AS) the measuring device for atmospheric sounding incorporated in the non-standard EUSO P/L to be accommodated on the lower balcony, SDX location, of the ESA Columbus External Payload Facility (CEPF) on the International Space Station (ISS). The AS will provide a wide-angle (about 60° FoV) observation of the Earth atmosphere by means of the combined utilization of a stand alone LIDAR associated to a scanning device and the EECR/ν Telescope, making possible during the night-time ISS orbit the detection of the backscattered echo of the light beam transmitted by the laser source of the LIDAR. The stand-alone LIDAR baseline configuration has been derived from consolidated experience and its development will be based on existing technology, while the utilization of the EECR/ν Telescope as an UV LIDAR receiver coupled with the stand-alone LIDAR is under investigation and will be developed specifically for this project.

5.2.4.1.5 Sub-system Description The real time knowledge of the scattering and light absorption properties of the local atmosphere where the EAS occurred is provided by the LIDAR. The main parts of a LIDAR system are the transmitter, which generate laser pulses, and the receiver, which collects and detect the radiation scattered back from the atmospheric regions crossed by the laser pulse. The EUSO-AS sub-system will be based on an hybrid configuration. The transmitter will be a space qualified diode pumped NdYAG laser, and there will be a dedicated stand-alone LIDAR receiver for the entire harmonics complex, plus the EECR/ν Telescope as a receiver for the 3rd harmonic of the laser. This hybrid configuration should allow performances similar to the ones obtainable using a large aperture LIDAR, but with a significantly smaller aperture of the stand-alone LIDAR receiver. The receiver of the stand-alone LIDAR will contain several optoelectronic channels for the acquisition of the backscattered signals at the three harmonics of the laser source, and possibly for different polarization state of the radiation. The large aperture EECR/ν Telescope will acquire only the echoes from the 3rd harmonic of the laser that is in the central portion of the UV spectral range of

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 38/68

EUSO. The simultaneous acquisition of echoes at several wavelengths and polarizations will allow to retrieve accurate profiles of the atmosphere, sounding the presence and nature of clouds and aerosol and evaluating the transparency and scattering properties of the atmosphere, in order to correct and calibrate the UV fluorescence and Cherenkov signals acquired by EUSO. A LIDAR is intrinsically a very narrow FoV device, so the stand-alone LIDAR will be also composed of a scanning device in order to scan both the laser beam and the stand-alone receiver within the about 60° total FoV of the EECR/ν Telescope.

5.2.4.1.6 Sub-system location The stand-alone LIDAR sub-system (laser and multi wavelengths receiver) will be allocated outside the external envelope of the EECR/ν Telescope and attached to the same supporting structure.

5.2.4.1.7 Constituting Elements Laser source. A space qualified NdYAG diode pumped laser will generate laser pulses at nominal 10 Hz (TBC) pulse repetition frequency (PRF) at the fundamental emission wavelength of 1064 nm, plus the second and third harmonics (532 nm and 355nm). The same laser should operate up to 100 Hz PRF for brief bursts in order to scan the region of interest after the detection of an EECR event. The laser head will contain multiple cavities for redundancy in order to assure proper operation all the mission time long. Note that the choice of using also the first and second harmonics, while improving significantly the LIDAR performance, doesn’t affect the input power needs for the laser, due to the necessity of to generate and combine the first and second harmonic to produce the third harmonic, which is the most important for EUSO because it is the only one that can provide direct measurement of the atmospheric properties within the spectral region of acceptance of EUSO. EECR/ν Telescope. It will be used to collect and acquire the UV backscattered laser signal. From a preliminary analysis has emerged that the utilization of the EECR/ν Telescope also as an UV LIDAR receiver will not impact significantly on EUSO technical budgets, (only some additional memory and CPU power consumption are requested). The central portion of the FS (that is not utilized for EAS detection) could contain a dedicated detector unit to acquire also the laser backscattered signal of the second harmonic. LIDAR Telescope. For the stand-alone LIDAR it will be a Cassegrain or Dahl-Khirkham type. This Telescope will be used to collect the backscattered echoes at all the three harmonics of the laser. Detector box. For the stand-alone LIDAR it will contain optics, detectors and electronics for up to six (TBC) channels of detection (three wavelengths, two polarizations). The number of channels doesn’t impact significantly on the mass and power budgets of the stand-alone LIDAR (less than 1 kg of mass and less than 5 W of input power are estimated for each detection channel). Scanning device. It is a motorized scanning mirror attached to the stand-alone LIDAR Telescope that allows to point simultaneously both the laser beam and the Telescope in the same direction, in order to track the atmospheric region where an EAS is occurred. System and Sub-System Engineering, Power sub-System and Thermal Control sub-System. The EUSO Instrument structure holds together all sub-systems, including the LIDAR, providing the necessary stiffness for the relative alignment among sub-systems and the strength to withstand all mechanical environments (on ground and on orbit), it protects the Instrument from debris/meteoroids and provide the connection with the Payload (on ground and on orbit). The stand-alone LIDAR supporting structure shall be attached on a suitable external location of the EUSO Instrument structure. Light-tightness in daylight orbit phases will be managed autonomously by the stand-alone LIDAR sub-system by means of filters, diaphragms and/or shutters inside the optical receiver. The AS Power System branch will manage the power needs of the AS device. The AS Thermal Control System branch will manage the thermal needs of the AS device.

5.2.4.1.8 Functional, Performance and Physical Requirements All the following requirements are referred to the stand-alone LIDAR. No additional requirements are foreseen for the EECR/ν Telescope to be implemented as an UV LIDAR receiver for the AS sub-system except for limited additional mass, volume and power (CPU and memories) and increased complexity of the electronics.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 39/68

Functional Requirements

• Normally the axes of the EECR/ν Telescope, the laser beam and the stand-alone LIDAR receiver shall point to the nadir direction. • The laser beam shall be maintained aligned with the optical receiver. A fine alignment of better than one mrad will be required between the laser and the stand-alone LIDAR Telescope, while a coarse alignment of better than 0.1° is requested between the stand-alone LIDAR and the EECR/ν Telescope. • The information about the absolute pointing direction of the Instrument with at least 0.1° (TBC) of angular resolution shall be available. [This means a dedicated star-tracker ???] • Due to the Space Station orbital flight velocity the available time during which the same region of interest remains within the FoV of EUSO will be typically no more than about 30 s. A few seconds pointing operation is required to the scanning device to allow a reasonable scanning of the EAS trajectory. • The scanning device shall allow to point the LIDAR within the entire FoV of the EECR/ν Telescope. • The laser shall be kept running continuously for reasons of thermal stability. This condition offers the advantage of having a continuous coverage of the atmospheric conditions along the ISS track. • Two PRF will be used: a nominal lower PRF for statistical observation of the atmosphere and a higher PRF to maximize the signal to noise during the few seconds allowable to scan the region where an EAS has occurred. • Very narrowband filtering shall be applied if also daytime observation will be carried out to improve statistical knowledge of atmospheric condition during the entire mission and consequently to improve statistics on EAS occurrence from the knowledge of the effective possibility of observation. • Multi wavelength sounding is required to unambiguously detect and characterize the presence of aerosols and clouds. • The acquisition of the echoes at multiple wavelengths must be simultaneous and triggered by the laser pulse emission. • The exposed optical surfaces of the Telescope and the coelostat should be protected from contamination during docking operations and periodically re-boost phases of the ISS.

Physical Requirements Power – The stand-alone LIDAR sub-system needs about 300 W of continuous input power to manage both the nominal 10 Hz (TBC) PRF operation and the burst mode 100 Hz (TBC) PRF operation, during which peak power of about one kW for tens of seconds will be required, after the detection of an EECR event. This input power will be always requested to the EUSO Payload, while the peak power will be managed by the LIDAR sub-system itself . Limited additional power resources are envisaged to implement on the EECR/ν Telescope the capability to be utilized also as a LIDAR receiver. Mass – The total mass of the stand-alone LIDAR has been evaluated to be around 140 kg. The achievement of this goal for the mass will be object of the trade-off study in progress in order to fulfill the overall mass constraint for all the EUSO Instrument. Limited additional mass is envisaged to implement on the EECR/ν Telescope the capability to be utilized also as an UV LIDAR receiver. Stiffness – The supporting structure of the stand-alone LIDAR sub-system shall avoid deformation that will introduce more than 0.1 mrad divergences between the transmitter and receiver optical axes. Envelopes – The entire Instrument surface shall be covered with protective and thermal insulation layers. The laser unit and the detector box shall be partially contained into pressurized envelopes due to the presence of high voltage and to guarantee safety operational conditions for optics and electronics. Thermal control –The AS sub-system will be integrated on the EUSO Instrument and therefore the thermal analysis of the AS sub-system will be enclosed in the main Instrument thermal analysis. The main Instrument will contain in its budgets the eventually needed resources for the AS sub-system thermal control. Fine thermal

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 40/68

stabilization will be required for several components inside the laser (pumping diodes) and the Detector Box (filters, detectors) within few degrees around their operative temperature.

Structure and Mechanism Requirements

• The common structure that will support together the EECR/ν Telescope and the stand-alone LIDAR shall guarantee the alignment between the two systems within less than 0.1°. • The several parts of the stand-alone LIDAR shall be contained in an assembly compatible with the constraints imposed to the payload dimensions by the available space into the Shuttle cargo bay. • All the moveable parts and motorized mechanisms will be space qualified and shall guarantee proper operation all the mission time long. • The scanning device shall assure a two axis tilting orientation of the scanning mirror of ± 30° around the primary axis (coaxial to the Telescope optical axis) and of ± 15° around the secondary axis (normal to the Telescope axis) within a few seconds. • A lid for the Telescope shall prevent from contamination during docking operations and periodically re- boost phases of the ISS.

Electrical Requirements

• A total input power of 300 W shall be provided continuously to the stand-alone LIDAR by the EUSO P/L. • The AS sub-system will include a dedicated power unit to manage the appropriate power needs for the Instrumentation. The voltage conversion and electrical power storage and distribution toward the several components (laser, detectors, motors and electronics) of the sub-system will be managed by the sub- system itself. • The power line of the stand-alone LIDAR receiver shall be separated and decoupled from the one dedicated to the laser transmitter in order to avoid severe electrical noise disturbance on the high sensitivity detectors and acquisition electronics inside the detector box.

Communication Requirements Telemetry – A 2.5 Kbps (TBC) flux of data during normal mode operation (10 Hz of PRF) and 25 Kbps (TBC) during burst mode operation (100 Hz of PRF) are expected from the stand-alone LIDAR sub-system assuming to integrate the signals every 100 shots for each stored profile. Additional 10 kbps (TBC) telemetry will be required for the data acquired by the EECR/ν receiver when used as UV LIDAR receiver. These telemetry values has been evaluated also considering that opportune pre-processing (integration on multiple shots) will reduce the amount of data to be transmitted. Tele Commands – Communication and data transfer between the EECR/ν receiver and the stand-alone LIDAR receiver will be managed by EUSO internal resources. The stand-alone LIDAR will need to receive the information about the location of the observed EAS in order to point and track the corresponding atmospheric region. The stand-alone LIDAR shall provide a trigger to the EECR/ν receiver in order to synchronize the acquisition with the laser pulse emission at the end of pointing operations. A large mass memory of about 100 Mb per orbit (~1.7 Tb total) will hold the telemetry data (scientific, housekeeping) to be downloaded to the ground station (rate and time: TBD).

Thermal Requirements

• Thermal gradients shall be contained in order to avoid structural deformations that will introduce more than 0.1 mrad (TBC) divergences. • An average of about 300 W thermal power shall be dissipated from the laser. • About 100 W thermal power shall be dissipated from the detector box.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 41/68

• Thermal stabilization within a few degrees around 20°C (TBC) operational temperature will be required for several optical and electro-optical devices inside the laser and the detector box in order to assure their stabilization and lifetime. In particular a fine thermal stabilization will be required for the laser pumping diodes and the optical filters and detectors inside the Detector Box.

5.2.4.1.9 Instrument Configuration & Performances

Fig. 5.2.4.1.9-1 AS sub-system configuration

• The stand-alone LIDAR optical receiver reference axis shall be parallel to the EECR/ν Telescope reference axis, both pointing nadir. • The alignment between the laser and the stand-alone LIDAR telescope shall be of about 0.1 mrad. • The alignment between the s. a. LIDAR and the EECR/ν telescope reference axis shall be better than 0.1°. • The laser beam and the telescope of the LIDAR subsystem shall be simultaneously pointed in few seconds in the direction of occurrence of an event detected by the EECR/ν Telescope with an accuracy of 1 mrad (TBC). • The atmospheric region interested by an EECR event will be tracked until it exits from the Instrument total FoV (no more than about 60 seconds) during which the laser beam will be pulsed up to 100 Hz. In normal mode, the AS subsystem will point to nadir and the pulse repetition rate of the laser will be of 10 Hz.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 42/68

• The laser source will emit simultaneously pulses at three wavelengths corresponding to the 1st, 2nd and 3rd harmonics of the NdYAG laser respectively at 1064 nm (near infrared); 532 nm (green visible) and 355 nm (ultraviolet). • The stand-alone LIDAR receiver will collect simultaneously the back scattered echoes at all the three harmonics of the laser and eventually the polarized and depolarized components of the backscattered signals. • The EECR/ν Telescope shall be utilized to acquire the total 3rd harmonic backscattered echo. • The central portion of the EECR/ν Telescope FS could be dedicated to acquire also the second harmonic. • Using this configuration and adopting the technical specification reported in the following paragraph, it should be possible to achieve the instrument performance required.

5.2.4.1.10 Environmental Interfaces Impact of operative environment versus performance is as follows: • the AS subsystem is sensitive to dust and gases/vapors with possible interaction concerning corrosion and condensation on optical/UV class mirror surfaces and dielectric coatings; • the AS subsystem is sensitive to light sources (over the 350nm ÷ 1100nm band for the stand alone LIDAR) originating from the ISS or the service vehicle docking to the ISS; • the AS subsystem is sensitive to thermal changes outside the range defined by normal operation conditions; • the AS subsystem is sensitive to electric discharges.

Contamination Reflecting surfaces of the externally exposed mirrors of the Coelostat and of the telescope are the most sensitive parts to contamination from LEO environment. A limited level of degradation of the performance of such reflecting surfaces is envisaged during the mission period.

Ground No particular protection is required for the stand-alone LIDAR if it is exposed to standard atmospheric conditions except for a closed lid on the receiver aperture to avoid dust contamination.

On-Orbit Mirrors shall be protected from debris and plasma by means of appropriate shielding maintaining unobstructed the operative FoV. A lid for the telescope should prevent from contamination during docking operations and periodically re-boost phases of the ISS.

On Ground Cleanliness Standard atmospheric condition could be considered acceptable cleanliness level only if the receiver aperture is conveniently closed by a lid; a clean room class 10000 environment (TBC) is advised to avoid contamination of exposed optical coatings.

5.2.4.1.11 Preliminary design A preliminary design of the stand-alone LIDAR has been based on the design of ABLE, an airborne LIDAR that has been entirely developed at the University of Rome “La Sapienza” (see e.g. Fiocco et al. 1999). This LIDAR has been installed on the M55 Geophysica Russian stratospheric aircraft, and successfully used in several campaigns since 1997. From an operational point of view, starting from ABLE’s design, the main modifications are related to the introduction of a more sophisticated scanning head and the implementation of a dedicated thermal control system. While, from a technological point of view, the main innovations are related to the space qualifications of all

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 43/68

the components, in order to guarantee safety and reliability of the system for the entire period of the mission in orbital environment condition. Many technological aspects of LIDAR in space have been just investigated in several research and industrial projects, also under ESA financial support (ATLID; Aladin). Existing knowledge acquired in the development of such projects will be highly useful for the completion of this preliminary studies. Figure 5.2.4.11-1 illustrates a potential layout of the stand-alone LIDAR.

Figure 5.2.4.11-1 Stand-alone LIDAR concept design.

In order to make the Coelostat more compact, and utilize a single scanning mirror of minimal dimensions, the stand-alone LIDAR should be accommodated with the primary axis of the telescope tilted in the plane defined by the nadir direction and the flight axis at about 165° respect to the flight direction, (that produces a mass reduction of about 20% for the Coelostat). The total envelope of the stand-alone LIDAR will be approximately of 0.7×1×1.9 m3. This envelope will be compatible with the dynamic envelope limitations inside the NSTS cargo bay.

5.2.4.1.12 Critical issues and open points

• Mass: to be optimized with the goal of achieving a total mass of 140 Kg including contingences. • Power: to be minimized with respect to present configuration. The utilization of accumulators/batteries is foreseen to provide the peak input power to the laser during burst mode operation (100 Hz PRF) after an EAS detection. • Thermal Control possible impacts on Instrument configuration (e.g. radiators). • Pointing: instantaneous orientation of the EUSO Instrument, including the AS subsystem, respect to the local nadir direction shall be known in order to retrieve the absolute pointing direction with an accuracy of better than 0.1°. • Telemetry: trigger signals and data transfer between the EECR/ν receiver and the stand alone LIDAR shall be implemented in order to attempt synchronous acquisition and storage of the atmospheric profiles relative to the same observed region of interest.

Possible Solutions

• The mass of the stand-alone LIDAR can be reduced, if a single axis scanning head with a smaller mirror will be implemented. In this case the Space Station motion will perform the scan in the other direction to cover the whole FoV of the EECR/ν Telescope. This solution introduces a percentage delay and reduction in the allowable time for the observation of the region of interest, so that science objectives and expected results will be approximately reduced of the same percentage.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 44/68

• The laser input power reduction (it should be kept above 250 W in any case for reasons of thermal stability) will result in a significant degradation of the EUSO-AS subsystem performance. Instead a more restrictive limitation in the duty cycle for the burst mode operation (actually 25%) and the optimization of the thermal design are envisaged to reduce the input power demand. • Thermal: common thermal analysis with Instrument will allow design consolidation. • Possible implementation of gyroscopes and/or a star tracker might be considered.

5.2.4.2 Other Devices See EUSO-SDA-REP-012 note [other reports on IR cameras ?]

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 45/68

5.2.5 The Power Supply System The Power Supply system will receive the primary power from the ISS and it will convert and distribute it to the different components as required. 1) OM: power might be required for heaters (TBC); 2) FS: power is required to power the MAPMT (HV) and to operate the electronics; 3) SE: power is required to operate the TCU; the TCU will drive the PS system; 4) System Engineering: power is required to operated heaters (TBC), the lid, actuators and emergency operation. 5) AS: power is required to operate the laser and electronics.

5.2.5.1 The buffer batteries The use of on-board batteries is foreseen [RDPS1], exploiting the short observation duty cycle, to store some energy during day-time when the Front-End , read-out an trigger electronics, consuming a large fraction of the total power, can be in principle switched-off. The details of the power consumption, conversion efficiencies and contingencies assumed for the various sub-systems is discussed in [IIDD]. In addition to that the batteries can act as a buffer in case of a primary power cut. To make the current EUSO Instrument design compliant with the total power allocation the baseline configuration adopted consists in the use of Li-Ion batteries. The maximum required power is only necessary during the data-taking phase of the Instrument, for an observational duty cycle estimated as ηOBS = 0.25. An Instrument duty-cycle of ηINSTR = 0.30 will be assumed for battery sizing purposes, to take into account switch on/off and initialization times, calibration runs (TBC) and contingency. Li-ion batteries are preferred, with respect to other technologies, for the following reasons: high specific energy, high efficiency, low self-discharge rate, no memory effect (see Figure 5.2.5.1-1). The main drawback is of use of Li- Ion batteries is safety.

Figure 5.2.5.1-1 Comparison between different types of battery technologies.

5.2.5.2 The FS PS system The FS PD is by far the most complex sub-system of the PS system. Only this part will be described in some detail in this section, as it is the only one with significant implications at system level. For the other parts a menton will be

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 46/68

made in the relevant sections. Two, schemes, basically, might be considered (with some possible variants) [RDPS2]:

1) a distributed DC/DC architecture; 2) a centralised DC/DC architecture.

The distributed architecture foresees a dedicated DC/DC converter for each PDM, providing the power needed by the PDM either directly from the 120 VDC primary bus or via some additional intermediate DC/DC converter and a suitable intermediate bus, decoupling the PDM from the primary bus. The centralised architecture foresees a set of high-power DC/DC converters, each one providing a given set of LV or HV lines, which are common to many PDM. The envisaged solution foresees a double power conversion, a first conversion from 120 VDC to a more common avionic bus, (28 VDC, TBC) followed by a second conversion to the required secondary voltages. This design would allow partial re-use of some commercial and space-qualified DC/DC design. The final choice will take care of many requirements and implications at system-level, including thr main design drivers listed below: 1) overall mass, power and volume budgets; 2) power conversion efficiency; 3) redundancy versus reliability; 4) harness simplification: mass and layout; 5) HW items reduction; 6) some pertinent requirements, such as: a. fault isolation management: it shall be possible to exclude a faulty PDM from the system; b. ON/OFF commanding: it shall be possible to switch ON/OFF and to independently regulate any individual PDM; it is required to provide a way to locally adjust the HV bias within each PDM, in order to allow better gain equalisation, also to compensate MAMPT ageing; c. it is required that the loss of anything at the elementary-cell level (including MAPMT and ASIC) shall result in the loss of the involved micro-cell only, without loosing the full PDM; d. it is required that the connection to and from the PDM are made through connectors, to simplify the integration and testing; 7) EMC/EMI. The trade-off between the different options is going on.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 47/68

5.2.5.2.1 Distributed PS system for the FS (totally modular) One PS board per PDM is assumed, providing all the voltages required to the PDM by direct conversion from the ISS 120 V power bus. This approach is shown in Figure 5.2.5.2.1-1.

Fig. 5.2.5.2.1-1 Distributed PS system for the FS (150 PDM are assumed).

5.2.5.2.2 Sextant Centralized PS system for the FS Each sextant of the FS is powered by a suitable number of PS boards distributing the required voltages to the PDM and converting the power received by an Input Protection Module which converts the ISS 120 VDC to a suitable intermediate voltage (28 VDC, TBC). This approach is shown in Figure 5.2.5.2.2-1 and Figure 5.2.5.2.2-2.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 48/68

Figure 5.2.5.2.2-1 Scheme of the distributed PS system for the FS.

Figure 5.2.5.2.2-2 Scheme of the distributed PS system for the FS at the PDM level.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 49/68

5.2.5.2.3 Discussion The approach centralized per sextant looks harder from the engineering point of view but it is based on existing, commercial and tested devices. It is clear that the modular approach has advantages from the engineering point of view, as it makes integration, test and replacements easier because everything is included inside the PDM. The conversion efficiency seems to be the most critical parameter in order to adopt a decentralized approach. In fact the low power required by each PDM (no more than a few W per PDM) and the different voltages required (3 LV plus two or three HV) makes hard to obtain a high conversion efficiency which is the main critical parameter with this approach. Current guess-estimates for this parameter range from less than 0.5 (based on extrapolation from existing devices) up to a more optimistic 0.7, for a completely new design. Also the mass budget and redundancy should be more carefully checked. No PS board with all the required characteristics and good enough efficiency is known to exist as of today for a distributed approach. During phase B the Collaboration will carry on a dedicated R&D study to evaluate the design of a dedicated PS board for the PDM with a better efficiency figure than available today. For the reasons mentioned above the current baseline is the Sextant Centralized one, as it is based on existing, commercial and tested devices, even if it does not look optimal from the engineering point of view.

5.2.6 The Thermal control System

It includes: • thermal control hardware (thermal blankets, heaters, thermostats, heat pipes and radiators) at Instrument level, including the main radiator;

• the thermal control hardware required by the different Instrument parts the thermal control of the different Instrument parts (including OM, FS, AS);

• the interfaces between the thermal control system at Instrument level and the thermal control of the different Instrument parts.

To be written

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 50/68

5.2.7 Instrument System Engineering The System Engineering (main structure, body and envelope) of the EUSO Instrument deals with: • the main supporting structure, supporting all the EUSO components and internal structures, and interfacing with the EUSO payload; • the payload mechanical items (I/F), actuators and mechanisms; • a system for protecting the interior of the Instrument from undesired light; • a system for protecting the interior of the Instrument from meteoroids and orbiting debris; • a system for protecting the interior of the Instrument from plasma (if necessary); • the motorized shutter (lid), to protect the Instrument against undesired light; • the baffle: a minimum section to protect the protruding optics from debris/plasma/ATOX plus an optical extension for off-field light protection during observation; • anything required to keep the interior of the Instrument safe; • the emergency operation systems, provisions and procedures; • the required harness, local and I/F harness; • the interfaces among the different above-mentioned components and those between the different above- mentioned components and the rest of the Instrument; • any other ancillary component required for the proper operation, monitoring and control of the EUSO Instrument, including the relevant parts of the Power, Thermal Control and Monitoring, Alignment and Self- Calibration systems. The EUSO Instrument is composed of a Flight Segment and a Ground Segment. The Flight Segment shall be composed of all hardware and software provisions necessary to successfully meet the mission objectives. The main items are summarized below. • Space environment compatibility • Science observation tasks • Electrical power switching and distribution to sub-system and components • Autonomous processing capability • Failure detection, safing and reporting • Commands reception, processing and distribution • Housekeeping data processing • Data exchange capability with P/L • Telemetry generation and packing • Thermal control • Calibration The main Ground Segment elements are the following. • Breadboards and Prototypes • Structural and Thermal Models • Ground Support Equipment (GSE) • Mission support facilities

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 51/68

5.2.7.1 Instrument System Configuration Figure 5.2.7.1-1 shows the Instrument system scheme and the links with the EUSO P/L and the ISS, while Figure 5.2.7.1-2 depicts the system configuration (TBC).

Figure 5.2.7.2-1 EUSO Instrument System scheme.

Figure 5.2.7.2-2 EUSO Instrument System configuration (TBC).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 52/68

5.2.7.2 Interfaces

• Optical Module. Each lens shall interface with the main structure by means of three interfaces. Heaters shall be mounted on the lens supporting structure as needed. • Focal Surface Photo-Detector. The focal surface structure shall interface with the main structure by means of twelve interfaces and shall support the SE electronic boxes and the FS-related thermal control provisions. • System Electronics. The electronic box shall be mounted on the FS structure. It will interface with all the other sub-systems and with the payload. • Atmosphere Sounding Device. The AS device shall be mounted on the main structure, externally, and it shall be provided with a dedicated radiator and protection system. • Structure. The main structure shall hold together all sub-systems, providing the necessary stiffness for the relative alignment among sub-systems and the strength to withstand all mechanical environments (on ground and on orbit), protects the Instrument from debris/meteoroids, assures light tightness in daylight orbit phases and provides the interfaces with the Payload (on ground and on orbit). • Thermal Control system. The Thermal Control system shall maintain all Instrument parts within their operating temperature ranges, in all mission conditions, by means of distributed thermal provisions. • Power System. The power system receive the power from the payload and distributes it to all the EUSO components requiring power. It is controlled by the TCU.

5.2.7.3 Components

5.2.7.3.1 The lid An essential part of the main structure is the lid, having a high electro-mechanical reliability (10000 actuations a year), protecting the optics from impinging atomic oxygen, plasma and debris. An integrated baffle shall also protect front lens from debris and shall limit stray-light.

5.2.7.3.2 The protection layers: meteoroids and orbiting debris protection to be written

5.2.7.3.3 The protection layers: light tightness to be written

5.2.7.4 Budgets See [IIDD].

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 53/68

5.2.7.5 Envelope and Main Coordinate System

Figure 5.2.7.5-1 EUSO Instrument main dimensions and main coordinate system location.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 54/68

5.2.7.6 Sub-systems Mechanical Environment and Interfaces TBD: Updated Mechanical/Structural analyses in progress

5.2.7.7 Sub-systems Thermal Environment and Interfaces TBD: Updated Thermal analysis in progress

5.2.7.8 Instrument Fields of View The EUSO Telescope unobstructed FoV is +/- 30 degrees (TBC; pointing nadir). The AS Device unobstructed FoV is +/- 30 degrees (TBC; pointing nadir).

5.2.7.9 Safety and reliability aspects Measures to solve/minimize the following safety and reliability issues shall be identified as the design evolves: high voltage lines (e.g. isolation), laser power (e.g. radiation reaching the ground), batteries (e.g. leakage), mechanisms reliability/safety (e.g. lid, moving parts), materials safety (e.g. glass/plastic fragmentation), LIDAR laser head cooling fluid, Probability of Perforation (Debris, meteoroids), contamination/cleanliness (i.e. photo-polymerization of lenses material)

5.2.8 Instrument Functional Parameters versus Instrument requirements A few key issues, related to the EUSO Instrument design, are summarized in this section in order to clarify some peculiar aspects of the EUSO Instrument design. • The required angular resolution of the optics is forgiving with respect to the usual optics requirements. • All the alignment and alignment stability requirements are mitigated by the large (4.5 mm) pixel size (equivalently 0.11° angular granularity). • The small optics f/# requires a good fit of the Photo-Detector surface to the optics focal surface: only a discrepancy of a few mm at most can be tolerated to avoid too large defocusing effects. In fact, ∆w, the average displacement perpendicular to the average direction of the incident photons (approximately parallel to the plane tangent to the FS) is roughly given by: ∆w 1 ≈ ≈ tanϑ ∆z 2 f /# MAX

where ∆z is the displacement from the FS along the average direction of the incident photons, and θMAX is the maximum angle of incidence of the photons with respect to the average direction of the incident photons. In order to keep the defocusing error small with respect to the pixel size ∆z δ 1 mm is required. • The focusing and alignment capabilities of the optics and Photo-Detector must be kept, all the mission long, with respect to mechanical and thermal aspects. This is made more complex by the large dimensions. • The temperature of the lenses and FS Photo-Detector must be kept stable to within (TBD) to avoid worsening of the image quality. • The HV must be kept stable to within ±24V in order to keep the MAPMT gain stable. • The LV must be kept stable to within (TBD) in order to guarantee the stability of the DAC and discriminator thresholds. • The light tightness requirement will limit the aging due to stray-light and it will limit the background during data-taking. • A baffle (TBD) long is required to limit the off-field light reaching the FS during data taking to be one order of magnitude less than the in-field random background light.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 55/68

5.3 Instrument Module Assembly, Integration and Verification

5.3.1 Model Philosophy and Testing The development logic of the Instrument is based on a “proto-flight” approach. Figure 5.3.2.1-1 shows the EUSO Development logic, where two models (at sub-system and system level) are identified: the “Structural Thermal Model” (STM) and the “Proto-Flight Model “ (PFM). STM shall be the model utilised for mechanical and thermal qualifications of the Instrument. More critical technologies and equipment, where prototypes development are necessary, shall be also foreseen. EUSO Instrument development is strictly interconnected with the EUSO P/L items development.

5.3.2 Verification approach The EUSO Instrument verification control activities, data and reports shall be supported by a verification data base. Each requirement shall be traced and the higher level requirement from which it is derived shall be indicated. The program verification phases are defined as follows. • Development: Verification activities (i.e. analysis and development test) oriented to support and justify the design choices. They do not require, in general, flight representative articles and, in principle, are not used to achieve the formal verification conclusion. • Qualification: Verification activities (i.e. analysis, review of design, inspection and qualification test) oriented to formally and contractually demonstrate that the design, implementation and manufacturing methods have resulted in hardware and software which conforms to specification requirements. • Acceptance: Verification activities (i.e. inspection and acceptance test) oriented to formally and contractually demonstrate that the flight product, manufactured in agreement with the qualified design and free from workmanship defect.

5.3.2.1 Verification Method Possible verification methods are test, analysis, review of design, inspection or a combination of them. • Test: Verification method in which technical means, such as the use of special equipment, Instrumentation, simulation techniques and the application of established principles and procedures are used to determine compliance with requirements. The analysis of the data derived from tests is an integral part of the test program. The test method is preferred when it is feasible, unless one of the other methods can establish a sufficient level of confidence whilst being cost and schedule effective. • Analysis: Analysis is used for verification in lieu of, in combination with, or in support to test to verify requirements. The selected techniques may typically include system engineering analyses, statistics, qualitative analyses, modelling and computer simulations. Since similarity needs first an engineering analysis, verification by similarity is included in this verification method. It is used if it can be shown that the article is similar to another article that has already been qualified to equivalent or more stringent criteria • Review of Design: Verification method based on the validation of previous records, certifications, etc. or by evidence of validated design documents, approved design reports, technical descriptions, engineering drawings, etc., showing unambiguously that the requirements are satisfied. • Inspection: Verification method to determine conformance to requirements without the use of special laboratory equipment, procedures, test support items or services. Inspection uses standard quality control methods to verify compliance with requirements of construction features, workmanship standards and physical condition.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 56/68

Figure 5.3.2.1-1 EUSO Development Logic (TBC)

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 57/68

5.4 Instrument Calibration The Instrument calibration is one of the most important and critical issues of the EUSO mission. For example, the reliable discussion on GZK cut-off really depends on the absolute energy scale. We can classify the EUSO Instrument calibration into four categories: • pre-flight calibration; • on-board calibration; • additional calibrations by ground based supporting facilities; • atmospheric calibration. The issue concerning the atmospheric calibration will be discussed mainly in section [xxx] The EUSO Telescope can be characterized by the following parameters: the throughput efficiency of the two

Fresnel lens system, τl, the transmittance of the filter τf, the collection efficiency of the sensor,η, and the quantum efficiency ε of the detector and the Front-End electronics and trigger efficiency, τe. Therefore, the photon yield ∆Q at distance r, producing a number ∆S of photo-electrons in each pixel, can be expressed as

εκηT TT Tα A ∆S = f l e ∆Q 4πr 2

where τα is the transmittance of the atmosphere, A is the EUSO Telescope Entrance Pupil Area and κ is a factor included to account for some fraction of the flux that may not fall on the given pixel owing to the optics point spread function. Note that most of the factors in the formula above depend on other parameters such as the field- angle and wavelength. The role of an on-board calibration is to determine the contribution from the Instrumental terms: ε κ η τf τe τl. In order to accomplish this goal the following items must be calibrated: • the Fresnel lenses; • the filters and optical adapters; • the FS Photo-Detector; • the Front-End and trigger electronics.

Note that, due to the relatively low number of detected photons per pixel, the effect of systematic errors, as long as bias is not present, can be minimized provided they are kept below the error due to Poisson photon statistics. Since the on-board resources are very limited, the detailed absolute calibration should be done on ground, before launch, as much as possible. The requirement for the on-board calibrations should be essentially relative calibrations, the flat fielding of the focal surface devices, monitoring the time variation of gains and efficiency of the sensors and the aging effects of the Fresnel lenses, the filter transmittance and sensors. The time variation of each MAPMT gain may be corrected periodically by adjusting the HV or discrimination level during the mission period. It should be noted that EUSO focal surface detector is working in photon counting regime, hence, the discrimination level directly affects the photo-electron collection efficiency and also noise rate.

5.4.1 Pre-flight calibration We need to consider several items on Pre-flight calibration, the transmittance of Fresnel lenses, the transmittance of optical filter, the quantum efficiency and the dynode gain of each pixels, the electric gains and so on. Each calibration will be discussed in each element chapters. Here, we discuss the most important pre-flight calibration system, a light pulser system for the PMT calibration (focal surface calibration) in macro cell unit. The information collected by EUSO is made of anode and dynode pulses from the many pixelised photomultipliers. The counting unit of time is 10 ns, and the pulse width at the base is very close to this value. In the actual simulations, the photomultipliers pulses are taken as perfectly square with infinite rise and decay times, and exactly 10 ns wide. In order to test in real conditions (pulses with about 3 ns rise and decay times, with some variations due to Poisson statistics around 1 photo-electron, or not perfect functioning, like after-pulses) how EUSO

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 58/68

will react to any imaginable EAS in the presence of any imaginable noise, we build a light pulser, with one light source for each pixel. This is also a very valuable tool to test trigger algorithms in real situations. The largest EASs to be detected are contained in one macro-cell (a square of 64 by 64 (4096) pixels). Hence we will have a matrix of 64 by 64 UV (370 nm) LED’s. In order to give a stable output, they will at each pulse provide about 10000 photons, and attenuate by exactly the same factor the light (filters plus optical fibres). Each pixel will then receive exactly one photo electron in average. The time interval between the arrival of the first and last EAS photon is slightly less than 1 ms, that is 105 times 10 ns. Hence, each LED will be addressed by a fast (100 MHz) memory with 105 values, that is 15 bits. The addressing system will be a point to point addressing, similar to the addressing of pixels in a computer monitor, where the “millions of colours” will be replaced by 105 times 10 ns. The LED drivers will be very simple and give a pulse so that one gets about 1000 photo-electrons if no filter, and one with filter. The light will go to the pixels through optical fibres. At the fibre exit, a small Teflon sheet will randomize the light to illuminate uniformly the pixel. Each pixel will be light shielded from its neighbours. One will send the X-Y-T information of the simulated EAS to the system, adding the required noise. The pulser being loaded, it is ready to send its pulses at the right places and the right time. The matrix will be able to move in order to be on any part of the focal surface.

5.4.1.1 Level 1 Pre-flight calibration (MAPMT calibration) The quantum efficiencies and dynode gains of MAPMT pixels are calibrated one by one after gluing BG-3 filters using multi-colour light sources LED. Among the measured MAPMT, we choose MAPMT which satisfy suitable requirements (uniformity, gains, Q.E.) as EUSO focal surface detector elements. Then, MAPMTs are sorted into several groups by the parameter of the gain. MAPMTs with similar gain are grouped together. The absolute calibration of the detection efficiency of each pixel will be done with an accuracy of 3% or better.

5.4.1.2 Level 2 Pre-flight calibration (Macro-Cell calibration) After the selection and grouping of MAPMTs, we assemble them as Macro-Cells. In this stage we define the location of each Macro-Cell on the focal surface. The light beam received by MAPMT / Macro-Cell distributes in a cone, the shape and the directional distribution of which depend on the location on the focal surface. In order to estimate the correct response of MAPMT / Macro-Cells, we need to simulate this angular distribution for the incidence photons. For this purpose we use the EUSO optics itself (for instance the first prototype lenses built during phase B) and a collimated light source which illuminates the EUSO optics with parallel uniform light. Then we can characterize the quantum efficiency, the gain and the uniformity of MAPMT / Macro-Cells under the realistic condition. To test the Macro-Cells with read-out system, we prepare the LED array light source which emulates the EAS images (a moving light source). It will be located at the focus of the collimator parabolic mirror or directly attached to Macro-Cells by fibre-optics cables after the attenuation.

5.4.1.3 Level 3 Pre-flight calibration (EUSO detector calibration) This is the final pre-flight calibration. We assemble the flight model EUSO detector Instrument in the laboratory, and then calibrate it using the collimated light source. In this stage, we calibrate the EUSO Instrument from end to end. The effective transmittance of lenses and QE of MAPMTs, dynode gain of the MAPMTs, electronics gains and cross talks between channels are calibrated, and then final pre-flight calibration tables are produced.

5.4.2 On-board calibration The resources in flight are very limited. Hence the calibration system on-board must be very simple and stable. However, we need to get enough information of EUSO detector for data analysis and adjustment of gain, HV and discrimination level. The major features for the on-board calibration systems (light sources) are listed as follows. • Spectro-radiometry in the range 330 nm ÷ 400 nm. • Photon counting (support DC mode and Pulse mode) • Time resolution (pulse mode): 20 ns. • Background: <104 photons/ (pixel s) • Uncertainty of absolute light intensity < 10%.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 59/68

• Light intensity range (DC mode): 0.01p.e. ~ 0.1p.e. • Light intensity range (Pulse mode) : 0.1p.e. ~ 100p.e. • Very low impact on the mass budget. • Low power consumption. • No movable components. • Monitoring function for the light source intensity. • Fitting the EUSO structure. The major concern for the calibration system is to produce a stable illumination on the focal surface. This stability must be continuously monitored during the Instrument lifetime. The completely uniform illumination on the focal surface may be ideal; however, small geometrical non-uniformity within 20% level on the focal surface is acceptable. But this non-uniform light distribution must be well calibrated on the ground before the flight.

5.4.2.1 Optical design The possible configuration of the light source in the EUSO optics is shown in Figure 5.4.2.1-1. The light source is located at the centre of the focal surface. The light passes through the lenses and is reflected back inside the EUSO Telescope by a diffuse surface reflector arranged on the rear side of the shutter(lid). Seven small PMT are embedded inside this reflector and measure the intensity distribution of light after passing through the two Fresnel lenses. The light intensity distribution on the focal surface is shown in Figure 5.4.2.1-2.

Figure 5.4.2.1-1 Ray tracing of the EUSO Telescope with a point-like Lambertian source placed at the centre of the focal surface and a diffuse reflector at the rear side of the lid. After the diffuse reflection, the light enters the EUSO Telescope and illuminates the focal surface. The light intensity and uniformity at the diffuse reflector is monitored by seven PMTs embedded in the reflector and lid.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 60/68

Figure 5.4.2.1-2 Intensity distribution of the radiation arriving on the focal surface using a radiation source arranged on the focal surface.

The required photon flux of 5×109 ph/s on the Focal Surface can be achieved using a source emitting 1015 ph/s (this light intensity reduction comes mainly from the geometrical factor in the light source, as it will be explained later). This number of photons is enough to calibrate all the pixels on Focal Surface with about one second exposure. There is one possible problem in this configuration. When the lid is open in observation mode, this diffuse surface may reflect background radiation into the EUSO Telescope. However, we can control this effect by manufacturing the reflector surface with many small-dimple structures. This structure may reduce the reflectance at large angles significantly.

5.4.2.2 Light source The requirements for a point-like UV light source are a uniform emission profile, and multi-wavelength emission to enable spectro-radiometric capabilities of the calibration system. Among the available UV sources, UV LEDs based on nitride technology are the most suitable. GaN, InGaN, and AlGaN LED emits in the spectral region ranging from 350 nm up to 450 nm. They are commercially available by some companies. Figure 5.4.2.2-1 shows a possible arrangement to produce a point like Lambertian source, whose emission flux can be easily adjusted and monitored. An UV LED with angular aperture of 15˚ illuminates a 0.2 mm pinhole that is placed on a 0.5 mm thick Teflon film. The pinhole diameter and the LED-pinhole distance can be adjusted to define the emission solid angle and the photon flux. The Teflon film is a very good diffuser for UV radiation and the selected thickness allows the transmittance of 0.2 ÷ 0.3 in the spectral region of interest. After the Teflon film, a diaphragm can be placed at a certain distance. Again, the diaphragm diameter and the distance can be selected in order to define the photon flux emitted from this source. Diaphragm constrains the aperture of the outgoing beam light to 30° half-angle. The stability of the emitted radiation can be monitored by the solid-state photodiode arranged in the gap between the diffuser and the diaphragm. Spectro-radiometry can be achieved using different colour LED. Figure 5.4.2.2-2 illustrates a possible arrangement of a light source with 3 LED. The LED are alternatively switched on, and the emitted radiation can be collected by an optical fibre, whose aperture angle fits perfectly with the LED emission angle. The optical fibres are connected to the pinhole. The silicon-based components of the source (the electronics and the photodiode) pose a problem of radiation hardness; therefore, the box must be built using a 4 mm thick Al layer. This has a direct impact on the mass budget and thus a careful design will have to minimize radiation shielding.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 61/68

Figure 5.4.2.2-1 Sketch of the source arrangement to Figure 5.4.2.2-2 Schematic drawing of the source get a point-like Lambertian UV source. arrangement for a multi-wavelength UV radiation source. The accommodation of 3 LEDs emitting at different wavelengths is shown in the front view on the right.

The impacts of on-board calibration system on mass budget and power budget are summarized in Table 5.4.2.2-1 and Table 5.4.2.2-2.

MASS BUDGET Box 850 g LEDs 60 g Photodiode 20 g Teflon, fibers, mirror 50 g Electronics 200 g Reflector, detectors and readout electronics 1000 g TOTAL 2180 g Table 5.4.2.2-1 Mass budget of one source and detectors on the FS.

POWER BUDGET LEDs 0.36 W Photodiode < 0.1 W Electronics 1 W Reflector, detectors and readout electronics 1 W TOTAL 2.36 W Table 5.4.2.2-1 Mass budget of one source and detectors on the FS.

5.4.2.3 Sensors and pre-amplifiers gain calibration by threshold scanning The precise knowledge of the photon detection efficiency is crucial for EUSO. The energy measurement, indeed, depends directly on that. It is therefore very important to provide the Instrument with the capability to measure, even on flight, the combined gain of the photomultiplier and of the Front-End preamplifier, in order to be able to set the digital threshold of DFEE to the right value, and to check that this value is correct. In the absence of an independent

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 62/68

analog measurement of the photoelectron signal, the best way to compute this combined gain is to perform a threshold scan, either using the night glow background or using a controlled light source (on-board light source, e.g. an LED). During a threshold scan, the DFEE threshold is varied in small steps from a very low value (e.g. 20% of the nominal single photoelectron peak amplitude) up to a high value (e.g. 2 times the single photoelectron peak amplitude) and the average number of counts detected in a GTU of µs. Using the DFEE internal memory buffers, for each value of the threshold, it is possible to count the detected photons for 128 consecutive GTU. The GTU length in this case is more conveniently set to 10 µs instead of the usual 2.5 µs. The result is a plot similar to the one shown in Figure 5.4.2.3-1, where indeed a real threshold scan performed on a real DFEE prototype, built and tested during phase A, is shown.

Figure 5.4.2.3-1 Example of threshold scan.

If the shape of the single photoelectron signal around the peak is approximately Gaussian, the result of the threshold scan is a sampling of the error function that is the integral of this Gaussian. Using simple fitting methods, it is possible to measure the position of the peak and also the width with high precision. This method takes automatically into account the non-linearity of the DAC that control the threshold and also its non-uniformities among channels because the scan is done individually on each channel. The complete scan of a Macro-Cell, including contingency and assuming not to use more than 10% of the TCU computing power, requires about 4 min and yields 54 Mbit of data. These 54 Mbit will be best analyzed on ground and the computed thresholds sent back to EUSO. This operation could be done for example twice a day, in order to perform a complete calibration scan of the detector roughly every two months. As stated above the threshold scan can be done either using the natural night glow background or a controlled LED source placed on board. The first method has the advantage that it can be used during normal data taking but it relies on an unknown source of light. The LED system has the advantage that can be used at any time when the lid is closed (EUSO daytime). After the thresholds are computed, we need a system to check that the result is correct and that the counting efficiency of the focal surface is the one you expect. For this reason, we think that a LED system is essentially needed anyhow, because to measure the efficiency you need a known light source and for this purpose the night glow is useless. Of course, to check the efficiency we will not do any scan. We will just set the threshold to the working values and then measure the efficiency by pulsing the focal surface with a well know pulse rate and intensity.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 63/68

5.4.3 EUSO field monitor (UV-Optical CCD camera, and IR CCD camera)

The FoV (FoV) of EUSO will cover the vast area on the ground (a circle of about 500 km diameter). In order to identify / reject spurious background events, the FoV monitoring system in UV-optical and IR is helpful. Furthermore, for the reconstruction of the energy spectrum from observed data, we need not only the absolute energy scale, but also the exact collection aperture for cosmic rays. IR camera gives us the map image of the water vapour distribution in the atmosphere. We consider two possible EUSO FoV monitoring system, UV-Optical camera (0.3~0.6µm) and IR camera (~10µm). These cameras capture images inside the EUSO FoV immediately after the EAS trigger with an exposure time programmable between 1 ms and 100 ms, because we do not know what kind of phenomena may occur in the FoV and the purpose of these camera is to reject unexpected background events. If the lightening or other unknown background light triggers the Instrument, we may get some hints in the captured image. This system helps significantly the understanding of EUSO signals / triggers and establishing the identification of EAS and other background. The IR camera of ~10 µm is sensitive to the cloud (water vapour in the atmosphere), and it is usually used in weather satellites. IR camera takes pictures randomly / periodically as frequently as possible, until the data transmission rate is allowed. If we employ jpeg compression, one picture (256 x 256 x 8bits) can be compressed down to ∼100 kbits. If we limit 10 Mbits/orbit, we can capture 100 pictures in one orbit. A summary of the characteristics is shown in table Table 5.4.2.3-1

CAMERA Pixels Wavelength Mass Power UV/optical TV quality 300 nm ÷ 600 nm 3.0 kg 20 W IR 256×256 (8bits) 8 µm ÷14 µm 1.8 kg 12 W

Table 5.4.2.3-1 Specifications of UV-optical, and IR camera

5.4.4 Ground Light Source

Please compare and cross-check with section 5.4.5

Proper analysis of the atmospheric fluorescence signals produced by EHECR and acquired by EUSO requires both calibration of the EUSO Telescope and knowledge of the transmission properties of the intervening atmosphere. To achieve the latter, a separate Atmospheric Measuring (AM) system will be used during the EUSO mission. The AM system will provide data from which one can infer the attenuation of the UV fluorescence signals reaching EUSO. To validate these results, a second system consisting of a set of multiple Ground Light Sources (GLS) will be employed. The GLS will provide a direct and independent verification of the accuracy of the reconstructed luminosity of EHECR. There are two possible solutions for GLS, one is the intense flasher and the other is ground based LIDAR system. The use of hybrid GLS system may be more redundant and effective.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 64/68

5.4.4.1 Ground Light Source (Flasher) We consider a total of 11 flasher lamps, 1 airborne unit and 10 ground-based units, deployed in different geographical locations to sample a variety of atmospheric conditions and terrestrial backgrounds. On average, the ISS will pass over one of the GLS units each night for a total of 1,000 measurements during the EUSO mission. The airborne unit will be installed in an upward viewing portal of a P-3B research aircraft stationed at NASA Wallops Flight Facility (WFF). The portal is enclosed by a glass bubble to protect the GLS unit from the slipstream of the aircraft. During the EUSO mission time, we are planning 30 under-flights on the P-3B that will be performed over land and over open water. Over-flights at sea will be planned that exclude the Eastern Seaboard of the US from EUSO FoV. Each under-flight by the P-3B will target a specific overpass by ISS and a specific altitude between 2 and 6 km. The central element of the GLS is a xenon flash lamp with high-intensity output in short flashes lasting a few 2 µs. The intensity of the Hamamatsu flash lamp L6604 is 0.3 µJ/cm at a distance of 0.5 meters between 300nm ÷400nm and is used in our baseline concept. The signal detected at EUSO for this lamp is estimated to be >500 photoelectrons per flash, assuming the EUSO baseline design and 50% absorption in the atmosphere. The flash-to flash variation for this model is 3 percent when operated below 30 Hz. In an over-flight lasting 30s, a total of 900 flashes would be detected by EUSO. The integrated signal from a single overpass would then have a statistical variation < 1 percent. Data shows that the spatial non-uniformity of this flash lamp is less than 5 percent over a 60° FoV. A thin UV diffuser will be used to ensure this variation remains <1 percent over any 5° FoV. The life cycle of these flash lamps is > 107 flashes without significant degradation in performance, which greatly exceeds our requirements.

Figure 5.4.4.1-1 The expected signals recorded by EUSO in each shot by the GLS(flasher) as a function of the nadir angle. In the calculation, Rayleigh scattering is taken into account. Mie scattering may degrade the intensity by 20~50% from this curve under the good weather condition.

The flash lamp is operated by a high-voltage source and trigger circuitry. Four flash lamps are grouped together to form a single GLS unit that is housed in a weatherized canister. Three of the tubes have narrow band- pass filters centred on the principal N2 fluorescence lines at 337, 357, and 391nm. The fourth flash lamp uses a

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 65/68

wide band-pass filter (330nm ÷ 400 nm) like that used on the EUSO focal surface. The weatherized canister has an upward looking window made of a UV glass that is covered by a mechanical shutter. These GLS are controlled remotely through the internet from the data centre. When GLS are in the FoV of EUSO, GLS flashes with ~30Hz repetition synchronous with UT-seconds. The image of GLS flash on the focal surface of EUSO is a spot and this light spot moves with the angular velocity of ~0.03°/0.03s. Hence, if the GLS passes through near the centre of FoV, we will see in total ~2000 flashes across the focal surface with 0.03 degree steps. We can estimate the integrated transmittance from the ground to EUSO by comparing the intrinsic flash powers monitored at GLS and the received photons observed by EUSO. Simultaneously, the on-board LIDAR system shots the laser toward the GLS and measure the differential transmission coefficient in the each altitude in the atmosphere. Then we can estimate the accuracy or systematic error of LIDAR measurements by comparing LIDAR and GLS data.

5.4.4.2 Ground light source(LIDAR) The LIDAR system on ground may be very effective to the EUSO diagnostics and calibration, because we can simulate EAS tracks using the laser beam. The third harmonic of NdYAG laser give us the 355nm UV light. In order to simulate the EAS of 3x1020eV, we need at least the power of ~50mJ, under the assumption of horizontal shot (~20degrees elevation angle). This kind of laser is available in commercial base, from companies, for example, Big Sky Laser, and Quantel. The elevation angle of laser shot and the laser power should be carefully optimized by the Monte Carlo simulation with the consideration of the attenuation of laser beam in the atmosphere and the image quality obtained by EUSO. If we determine the laser power and the elevation angle of the laser beam (~20 degrees), we can estimate the necessary receiver mirror diameter (~1m diameter size). Then, we can get backscattered photons in the receiver, with the level of ~800p.e./2.5µs and ~20p.e./2.5µs in each shot from 30km and 60km distances, respectively. For example, after 100 shots integration, we can get more than 1000p.e. in each time bin, then we can characterize the atmosphere precisely. In the figure 5.5.4.2 the signal strengths detected by EUSO and by ground LIDAR receiver are shown. The advantage of LIDAR is that we can measure the beam intensity profile by itself along the laser beam with an accuracy of 5-10%. This gives us a good opportunity to calibrate EUSO and atmosphere independently. As a ground light source for EUSO the shot in the elevation angle of 20~30 degrees is optimum and probably the fixed directional LIDAR system is robust and minimize the maintenance of mechanical parts of the system. The laser shot frequency of NdYAG Laser can be tuneable up to 10Hz~30Hz. In these horizontal shot, the laser beam comes in the pure molecular scattering region (Rayleigh scattering region) in the top atmosphere after travelling ~30km (corresponds to actual height of 10km). Then the laser beam travels another ~30km in pure molecular region. Once, the laser beam is in the pure molecular region, we can get the boundary condition for the LIDAR equation, because we know the relation of backscattering light intensity and the beam intensity in pure molecular scattering. Then we can solve LIDAR equation from the top of the atmosphere toward the ground and finally we can derive the laser beam intensity profile and the attenuation coefficient along the laser beam pass in the atmosphere differentially. Then finally we can also derive the transmittance of the light in the atmosphere in each height differentially. The laser with the elevation angle of 20 degrees can be seen as a long track by EUSO (30~50km length). Since the profile of the laser beam intensity is derived by the ground LIDAR itself as mentioned above, we can estimate the number of photons which will arrive to the EUSO optics entrance after the scattering, if the scattering is dominated by Rayleigh process. The scattering angle of photons which will be received by EUSO is always larger than 40 degrees and in such large angle scattering, Rayleigh process usually dominate under the good weather condition. Mie differential cross section is peaked at the forward direction and smaller contribution in the larger angles. Probably for EUSO calibration, we use the laser beam above 3km from ground because we can describe the scattering just by Rayleigh. The simultaneous operation of on-board LIDAR system and ground LIDAR system gives us more redundant measurement, and we can characterize the atmosphere much better than one system and reduce the systematic error in the measurement significantly. The angular resolution and the energy determination systematic error of EUSO detector are experimentally derived by reconstructing the laser beam. With 10~20 ground-based LIDAR systems, we have chance to calibrate the EUSO ~one time a day. These systems must be controlled from the data centre through the internet. The ground LIDAR will be turned on 30mins before the ISS passage to get stability of the laser and to get enough data to characterize the local atmosphere and will be operated continuously until the end of the ISS passage.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 66/68

Figure 5.4.4.2-1 The signals from Ground-based LIDAR shot (50mJ shot, 355nm NdYAG 3rd harmonic) in the elevation angle of 20 degrees. Number of p.e. is calculated in the integration time of GTU(2.5µsec). The Mirror size and Q.E. of LIDAR receiver is assumed as 1m diameter and 0.1. The signal strength detected by EUSO between 10km and 60km is ranging in the reasonable region, from several p.e. to 100p.e.. The signal strength received by LIDAR is enough to characterize the atmosphere after ~1000 shots. We can solve the atmospheric transmittance with high accuracy and reduce the systematic errors using ground-based LIDAR and on-board LIDAR.

Please explain how to deal with the atmospheric absorption from the photon emission point to EUSO.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 67/68

5.4.5 Instrument Calibration: operations Instrument calibration is crucial for the achievement of the scientific goals of the mission. Calibration activities scheduling might represent an important part of the mission planning activities and calibration data might represent a sizeable fraction of the telemetry. Calibration activities are categorized in two different types: technical calibration, which monitor the health, status and performance of the Instrument components, and scientific calibration, which will emulate the response of the system to scientific events and allow to calibrate the physical quantities determination procedure.

5.4.5.1 Technical calibration Technical calibration activities monitor the health, status and performance parameters of the optical system, of the Photo-Detector on the FS and of the Atmospheric Sounding Instrumentation. The required absolute calibration operation will be performed on ground and then a calibration matrix will be obtained (pre-flight calibration). Calibration runs on-orbit will allow to monitor the health, status and performance variation of the different components (on-board calibration). Scheduling of the on-board calibration activities will be performed from ground. According to on-ground (time- tagged) commanding, the on-board system will go into technical calibration mode and activate the Instrument calibration system. Technical calibration will be normally scheduled during day-time periods, when no scientific observation with EUSO is possible. Since power saving configurations will be adopted during day-time, calibration activities will take place either at the beginning or at the end of the data-taking period, when the Instrument in still/already in normal observation configuration, and have a foreseen duration of the order of a few minutes. Calibration data volumes are a concern, because of the high segmentation of the EUSO FS. The calibration of the FS with not all at once but per Macro-cell or Macro-cell group in different orbits could be a favourable solution from the point of view of the use of the available resources: power, telemetry and observation time. Data compression of calibration data may be required and is envisaged (see section [???] for a brief discussion of telemetry budgets). As a reference, the scenario of the FS calibration (one Macro-cell calibration in each orbit) is considered in section [???] under the assumption with a very low hit pixel occupancy (of the order of 10-1, in order to allow for single photoelectron peak calibration) and for a large number of GTU (of the order of 4K, for a 5% resolution). This design requires a reasonable telemetry budget. Calibration results may lead to periodic updates of the system configuration files (e.g. pixel masking). These will be performed from ground with a periodicity which is TBD.

5.4.5.2 Scientific calibration Scientific calibration involves the test of portions of the focal surface with events that mimic the scientific events and allow calibrating the “light to energy conversion”. This can be done on-board using external light sources (see section 5.4.4), strategically placed on top of mountain and other remote areas for this purpose. Ground Light Sources (Flasher) uses xenon flash lamps (four tubes per GLS unit) with high-intensity output in short µs flashes, with intrinsic luminosity known to 1% level. This will test how well the luminosity of the EAS is reconstructed. If we deploy 10 GLS station, on average, EUSO will fly over one of the GLS sites per night, for typically up to about one min. The EUSO passes need to be scheduled (from the ISS ephemeris). Scientific calibration activities will be planned on ground and included in the up-linked activity schedule files. A configuration file for the GLS should exist on-board and be loaded/executed automatically when EUSO is over a GLS. Atmospheric transmission monitoring with on-board LIDAR will be required during GLS calibration. As GLS operations must be planned with the overall EUSO operations, the EUSO observation schedules will be made available to the GLS operation team. Conversely, HK data from the GLS will be made available to the SODC for integration with the remaining EUSO data. Also envisaged is the use of a ground based LIDAR for scientific calibration purposes. In this case EUSO would see not the direct light from the ground source but rather a light trace propagating and interacting in the atmosphere, like in EHECR events. Moreover, the detection by the ground based LIDAR of the back-scattered echoes would allow the characterisation of the light intensity at each point. From the point of view of operation, the above considerations for GLS can be applied also to ground LIDAR calibration: calibration would be scheduled from ground, adequate configuration files would have to be loaded on-board, the activity schedule would have to be

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 68/68

made available to the ground calibration team, and the corresponding HK data would have to be delivered to the SODC.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/43

6. EXPECTED PERFORMANCE...... 2 6.1 END-TO-END SIMULATION ...... 2 6.1.1 The Physics Process Simulation ...... 3 6.1.2 The Atmospheric Model...... 4 6.1.2.1 The Atmosphere as a light emission medium ...... 4 6.1.2.2 The atmosphere as a transmission medium...... 6 6.1.3 The Expected Background ...... 11 6.1.3.1 Natural night sky diffuse and slowly varying sources...... 11 6.1.3.1.1 Diffuse night brightness...... 11 6.1.3.1.2 Airglow...... 12 6.1.3.1.3 Total Moonless expected background...... 12 6.1.3.1.4 Moon Phases...... 13 6.1.3.2 Artificial light of the Night sky...... 13 6.1.4 The Detector Simulation...... 13 6.2 FLUX AND SENSITIVITY ...... 17 6.2.1 Acceptance and Aperture ...... 17 6.2.2 Flux and Sensitivity - Extreme Energy Cosmic Rays ...... 19 6.2.3 Flux and sensitivity - Extreme Energy Neutrino...... 19 6.3 THE EXPERIMENTAL RESOLUTION ...... 23 6.3.1 Direction ...... 23 6.3.2 Energy Resolution ...... 26 6.3.3 Elemental Composition ...... 28 6.4 EUSO AS A SELF-SUFFICIENT ATMOSPHERE DETECTOR ...... 31 6.4.1 Estimation of the Acceptance of EUSO due to cloud presence ...... 31 6.4.2 Shower altitude estimate and shower energy measurement...... 32 6.5 METEORS OBSERVATION BY EUSO...... 34 6.5.1 The Meteor basic physical processes: modelling...... 34 6.5.2 The Earth Atmosphere...... 35 6.5.3 Motion of a spherical massive body in the Atmosphere...... 36 6.5.4 The Ablation process...... 37 6.5.5 Meteors as source of UV radiation ...... 38 6.5.6 Meteor simulation ...... 39 6.5.7 The case of micrometeorites ...... 41 6.5.8 Data retrieval and approach to the analysis...... 42

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/43

6. Expected Performance In this Chapter we will work out the expected performances as a consequence of the Observational approach of EUSO, folded to the actual detector design and features. We will therefore discuss the End-to-end simulation which brings from the physics process to the expected signal we expect to observe in EUSO in sect. 6.1, The acceptance and the expected sensitivity in Sect. 6.2, and the resolution in retrieving the original event from the experimentally observed signal in Sect. 6.3. We will finally discuss the possibility of using the detected photon signal as an atmospheric monitoring device in Sect. 6.4 and the capability and sensitivity for the slow fluorescence- emitting phenomena like meteors to be detected by EUSO in Sect. 6.5.

6.1 End-to-End Simulation The flow-chart in Figure 6.1-1 summarizes the most important items that the EUSO Simulation and Data Reduction system has to face, and identifies its building blocks. The innovative approach of EUSO (as it is the first experiment which will detect EASs from space and in particular through their fluorescence emission), implies a series of challenges also as far as the simulation and event reconstruction are concerned. The goal is that of obtaining an end-to-end simulation chain finalised to study the detector response function. The expected signal has to be worked out, as an output of the simulation algorithms, as it will be experimentally recorded for real events, with as much as possible the same conditions, to the extent of our knowledge of the physical processes and instrumental efficiencies. The reconstruction algorithms have to be developed in such a way of deconvoluting the effect of the experimental measurements, with the aim of tracing back the features of the original events, from the experimental observables. The aim of the Phase A study has been that of identifying the problems, develop the algorithm to treat them and derive the expected detector performances. All these tasks have been successfully performed, leaving to a second stage of development the goal of building a completely coherent software frame and the analysis of all the refinements, as the prediction of model dependent interaction mechanisms and, to a large extent, the elemental composition discriminating power.

Figure 6.1-1 The EUSO simulation and event reconstruction flow-chart.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/43

6.1.1 The Physics Process Simulation The trade-off of EAS generation is between the desirable detailed description given by the most common and sophisticated Montecarlo programs [CORSIKA: Heck et al., 1998; AIRES: Sciutto, 1999], but paid with a very high computing time, even in the “thinning” version, and a less detailed simulation, focused on the correct treatment of the longitudinal shower profile, the fluorescence yield and Čerenkov light emission, with less emphasis on the tracking of all the secondary particles produced in the shower. The decision taken at this stage of the study is to use a simulation based on the hybrid approach UNISIM [Bottai, 2001] and on shower parameterisation [GIL: Linsley, 2001a.b; Catalano et al., 2001; EUSO-SIM-REP-010; SLAST: EUSO-SDA-REP-015] approaches. In UNISIM the shower is simulated following all the particles produced in the showering process, according to the SIBYLL [Engel et al.,1992; Fletcher et al., 1994; Engel et al., 1999] minijet model for high energy hadronic interactions, down to a threshold energy (E=1017eV, where E is the energy of the primary particle, has been selected as a trade-off value between the accuracy of the fluctuation reproduction and the execution speed). Below this threshold a parametric description of the shower, based on a library of pre-simulated events is used, to work out the longitudinal profile. A global NKG fit is then used to generate the lateral particle distribution, relevant to simulate the shape of the shower front, where the fluorescence and Čerenkov light starts from. The generation of Čerenkov light during each step of shower propagation is calculated using the energy distribution of electrons given in Hillas, 1982. The angular distribution of the emitted photons is simulated using the parameterisation of electrons angular distribution [Baltrusaitis et al., 1987] convoluted with the Čerenkov cone emission. The UNISIM package includes also the LPM effect for electromagnetic interactions and it is able (unique in the panorama of the existing shower generators on the market) to simulate neutrino interactions inside the target mass seen by the detector. As additional unique feature, UNISIM is able to simulate showers which develops perfectly horizontal without any interaction point with ground. Neutrinos charged current (CC) and neutral current (NC) differential cross sections used in the simulation for neutrino interactions have been calculated in the framework of QCD improved parton model. The effect of LPM interaction in neutrino induced showers is discussed in [Bottai, 2001]. Both the shower maximum and its profile are affected (sometimes two clear peaks are observed in the profile due to hadronic and electromagnetic shower superposition, the latter being shifted by the LPM effect).

Figure 6.1.1-1 Longitudinal profiles for a sample of vertical proton showers generated by UNISIM with E=1020eV

The results of these generators have been compared to the prediction of the CORSIKA Montecarlo, as far as the longitudinal profile is concerned. Figure 6.1.1-1 shows the shower profiles for protons generated by UNISIM. Also shown are the shower parameters obtained using the CORSIKA Montecarlo, with the same hadronic

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/43

interaction generator (SIBYLL 1.7). The same comparison, between the GIL analytical parameterisation and a sample of CORSIKA/QGSJET events, has already been shown in Figure 2.1-1.

The output of the generator is the EAS shower profile N(Ep,X;AM,γ,ν), the number of charged particles as a function of:

• the energy of the primary particle Ep; • the slant depth X in the atmosphere;

• the nature of the primary, hadron of mass number AM, photon γ, or neutrino ν. This is our starting point to build-up the expected signal for EUSO.

6.1.2 The Atmospheric Model The primary particles collide with air nuclei after entering the Earth’s atmosphere, then secondary particles are produced and the shower development is mainly determined by the amount of traversed air. Adopting the same primary parameters, showers develop quite similar to the grammage traversed. However, differences occur in the geometrical development because of the relationship between atmospheric depth and altitude. The Atmosphere acts thus as a target for EUSO. We will detect the UV light emitted by the EASs, both in the fluorescent de- excitation of mainly the N2 molecules of the air, as a consequence of the punch through of the secondary particles produced in the shower, and in the Čerenkov radiation beam which accompanies the propagation of the relativistic particles in the shower. With concern to both this phenomena the atmosphere plays twofold a major role. It is the light emission medium. Atmosphere is also the transmission medium where the light propagates (and attenuates) from the source to the detection site: the EUSO telescope placed at an altitude of ∼400 km, far beyond the height of the atmosphere. As a target, as a light emission medium and as a transmission medium, the knowledge of the features of the atmosphere represents a key factor of success of the EUSO experiment.

6.1.2.1 The Atmosphere as a light emission medium The fluorescence yield. As mentioned the fluorescence light is produced in the interaction of charged particles with air. The fluorescence spectrum of air consists almost entirely of nitrogen contribution. Light is emitted isotropically and is dominant in the wavelength region from 300 up to 450 nm. The light comes from the de- excitation of the two electronic states: the 2P band and the 1N band. From Figure 6.1.2.1-1 below ones sees that the main contributing transitions lines are the 2P at 337.1, 357.7 and 315.9 nm and the 1N at 391.4 and 427.8 nm. Due to the fact that the EUSO bandwidth covers the range 300÷400 nm and since the atmospheric transmission is highly suppressed in the region below 330 nm by the ozone layer absorption, the simulation study concentrated, in the present Phase A study, on the fluorescence yield in the wavelength range 330÷400 nm. The simulation of the fluorescence yield, in our present work is based on the results of F. Kakimoto [Kakimoto et al.,1995] where the fluorescence produced by 1.4 MeV electrons, have been measured in the 337, 357 and 391nm wavelength bands. The dependence of the fluorescence yield with density ρ and temperature T is given by the gi(ρ,T) functions respectively for the 1N (391 nm) and 2P(337 and 357 nm) bands.

ρ A ρ A (6.1) g ()ρ , T = 1 g ()ρ , T = 2 1 ()+ ρ 2 ⋅ ()+ ρ F 1 1 B 1 T 2 ,760 F 1 1 B 2 T

3 -5 Where ρ is in units of g/cm and T in Kelvin. F1, A1, A2, B1, B2 are constants and are 1.044 10 , 0,574 and 0,927 in units of cm2/g and 6500 and 1850 in units of cm3/g/K-1/2 respectively. The relative weights of the 337 and 357 wavelengths have been extracted from the A.N. Bunner Ph.D. Thesis [Bunner, 1964]. The dependence on ionization loss is also clearly established and the yield of fluorescence photons emitted by an electron of energy E is then parameterised by the following formula, in photons/(m×particle).

(g (ρ,T) + g (ρ,T)) ()()ρ = ⋅ 1 2 Nγ E; ,T dE / dx E ()dE / dx = E E0

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/43

where ρ is the local density and T the temperature, E0 is a reference energy (E0 = 1.4 MeV in Baltrusaitis,1987). The dE/dx dependence given in Kakimoto et al., 1995, has been approximated by weighting the energy spectrum of the e.m. particles in the shower and assuming the mean value, Emean = 32 MeV, and (dE/dx)E= Emean. Thus, the overall yield is normalized to achieve 4.8 photons per meter at normal conditions of pressure and temperature. A recent work by M. Nagano [Nagano et al., 2003], updated the results given in Kakimoto et al., 1995. They confirm the pressure and temperature dependence given in (6.1), as a good approximation of the contribution given by six different emission bands. The overall yield is however higher by ∼10 %. A more detailed simulation algorithm is under development within our collaboration. It is worth to be said that, since the subject is of wide interest within the UHECR research community, a worldwide effort [FIWAF, 2002] is in progress by various collaborations (EUSO, Auger, HiRes) to achieve a precision of < 5 %. The energy dependence and the humidity dependence of the fluorescence yield are the main concerns, where a definite experimental effort is still required. Figure 6.1.2.1-2 shows the behaviour of the total (right axis) and relative (left axis) yield of the three lines as a function of the altitude in a standard US atmosphere. An interesting feature is the rough independence of the yield from the altitudes below 15 km: this is due to a compensation effect between the decrease of the number of molecules on one side (a smaller number molecules are excited by the secondary particles), and the increasing weight of radiative versus collisional processes on the other side (more excited molecules de-excite through fluorescence emission at low density). A detailed work on the fluorescence yield dependence upon the atmospheric condition and its impact on the EUSO observation is contained in [Lebrun, 2002].

Figure 6.1.2.1-1 The different fluorescence Figure 6.1.2.1-2 On the left axis (black curve) the transitions in nitrogen. fluorescence yield, photons/(m×ch. part.) as a function of altitude. On the right (coloured curves) the relative abundance of the three main fluorescence lines used in the calculations.

The Čerenkov light. Čerenkov light is produced as the charged particles in the shower travel in the atmosphere faster than the speed of light in air. Čerenkov light is highly beamed around the shower axis, in a cone whose aperture is given by θmax ≅ 1.3°, in the atmosphere. > = 2 − The energy threshold for Čerenkov light production is given by the formula: E Eth mc 2(n 1) where n is the wavelength-dependent index of refraction at the current atmospheric depth. The number of Čerenkov photons emitted by a single particle of energy E and velocity β=pc/E, per unit path length, in the wavelength range λ1÷λ2 is described by the well known equation:

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/43

λ () 2 dN p E 2   = 2παz 1 − 1 dλ (6.2) ∫  []β ()λ 2  dx λ n 1   with α being the fine structure constant, z being the charge number. The total amount of light produced by the shower per meter, at the depth X, has thus to take into account both the refraction index dependence on wavelength and depth, as well as the charged particle energy distribution in the shower f(E,X):

∞ dN()X dN p (E) = f (E, X ) dE (6.3) dx ∫ dx Eth

Since n(λ) changes by less than 5% in the range 300÷400 nm, a wavelength independence of the refraction index has been assumed in this work. The change of the refraction index as a function of the air density has instead been taken into account. The generation of Čerenkov light during each step of shower propagation is calculated in UNISIM using the energy distribution of electrons given in [Hillas, 1982]. The angular distribution of emitted photons is simulated using the parameterisation of electrons angular distribution [Baltrusaitis, 1987] convoluted with the Čerenkov cone emission.

6.1.2.2 The atmosphere as a transmission medium. The transmission of the photons in the atmosphere is affected both by scattering (Rayleigh, Mie) and absorption effect. The Rayleigh scattering is produced by the air molecules, and is wavelength dependent. When the diameter of the targets is of the order of the radiation wavelength, the Mie scattering takes place. Aerosols (dust, smoke, ...) in atmosphere and droplets (clouds) are responsible for Mie scattering of light. In the UV range of interest, absorption is mainly due to the ozone.

Figure 6.1.2.2-1 Transmission as a function of the Figure 6.1.2.2-2 The atmosphere transparency as a altitude computed with LOWTRAN7, for two function of the radiation wavelength in a U.S. Standard wavelengths, in case of the 1976 US Standard atmospheric environment. model. Contributions of Rayleigh and Mie (“Navy maritime” aerosols) scatterings are shown.

The impact of Rayleigh scattering, Mie scattering and Ozone absorption for different wavelengths and different atmospheric profiles has been studied with the LOWTRAN7 [Kneyzis et al., 1996] program which allows to compute the light transmission, for different atmospheric models. One of the most well known model is the “1976 US Standard” [USSA-76] which has been used for most of the computations. It is important to notice that scattered

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/43

light can reach the detector after one or several scattering processes. The transmitted one is the minimum amount of light reaching the detector. In Figure 6.1.2.2-1, the transmission is plotted as a function of the altitude, given in kilometres, of the light emission point, for two different wavelengths. To get the transmission curves shown, the US Standard Model was used. It can be seen that the main light attenuation is due to Rayleigh scattering, in particular for low wavelength values, and that it is the dominant attenuation process in upper atmosphere. For λ = 337 nm, 40% of the light emitted from ground is transmitted to EUSO, and 83% of the total attenuation is due to Rayleigh scattering. Figure 6.1.2.2-2 shows the expected wavelength dependence of the attenuation for light emitted at ground level. While rather transparent for large wavelengths, ozone presents a sharp cut-off below 330 nm. Therefore, the influence of ozone in EUSO is strongly dependent on fluorescence light spectrum and on the bandwidth of the telescope. In Figure 6.1.2.2-3 the effect of ozone on the light transmission as a function of the wavelength is plotted. To estimate the influence of the ozone thickness, the US standard ozone profile has been scaled by factors 1, 1.5 and 2. The effect is non negligible below 330 nm, and then will have a larger impact on the Čerenkov light transmission than on the fluorescence one. The effect (reflection, diffusion) of a cloudy sky on the fluorescence and Čerenkov signals is treated in the next section. Using other atmospheric models has a tiny influence on the above results, provided that the aerosol contents are not modified, and even more, to the extent that clear, ideal atmosphere is a good representation of the real situation we will get on an event by event basis. Figure 6.1.2.2-4 shows the attenuation of light when changing the aerosol models (the atmospheric model is fixed). The aerosols are mainly present in the boundary layer, and then the attenuation could be important for light emitted or transmitted at low altitudes.

Figure 6.1.2.2-3 Light transmission as a function of the Figure 6.1.2.2-4 Transmission as a function of the wavelength in case of ozone absorption. The curves altitude computed with LOWTRAN7, for different corresponds to three ozone profiles (US standard one types of aerosol contents. scaled by factors 1, 1.5 and 2).

The observational approach of EUSO, which looks at the atmosphere from a space-based telescope, placed on the International Space Station ISS, contains some peculiarities, as far as the impact of the atmosphere is concerned, as it has been outlined in Chapter 2, which turn essentially into a complication and several advantages: • The atmosphere variability, due to the ISS speed, and its non-homogeneity, mainly due to the wideness of the atmosphere portion contained in the field-of-view is the largest complication to be handled. The first step in the study of atmosphere variability and its effect on the radiative transfer is to test different atmospheric models. The US Standard Atmosphere 1976 models are commonly used in the atmosphere community. The U.S. Standard Atmosphere Supplements, includes tables of atmospheric parameters for five northern latitudes (15, 30, 45, 60, 75), for summer and winter conditions. The grid in latitude, longitude and time may be not precise enough for our purpose, for simulation and even more in view of the retrieval of the real event pattern. Using US standard values as inputs whatever the space-time location of the shower can lead to large systematic errors, in the UHECR energy determination or particle identification. As far as we are concerned by the global properties and profiles such as pressure, temperature and the number densities of the main constituents it seems better to use the empirical models recommended by

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 8/43

the Committee for Space Research (COSPAR). These models are based on 40 years of data of various types and are continuously updated. They can be used in various fields of EUSO simulation from fluorescence yield variation through nitrogen number density, to airglow through atomic oxygen profile, even on the mechanical side evaluating the impinging of atomic species at 400 km on various part of the telescope. Neutral densities from ground to thermosphere are given within a latitude, longitude and date grid, which can be easily accommodated to the EUSO trajectory purpose. The MSISE model doesn’t contain secondary constituents at low concentration at a given altitude such as water vapor, ozone or aerosols, which are of importance for light production and transmission. This lack of information can be recovered by the use of US-Standard Atmosphere 1976 profiles or by the data of a dedicated measurement. Comparisons between US-Standard and NRLMISE-00 Model 2001 [Picone et al., 2001] density profiles from ground to 20 km have been performed, in order to check that the profiles of the US standard models can be reproduced with the data from NRLMSISE; the results show that profiles are consistent within ±2.5% at all altitudes. The use of MSISE model, although not implemented in this phase, will be added, in the full simulation software, with a twofold goal: modelling, in a detailed way, the real atmospheric environment, and retrieving, at the real event analysis stage, the actual atmosphere seen during the data taking.

• The fact that EUSO is looking to the Earth at its Nadir (±30°, the width of the FoV), from a large distance (>~400km), together with the thinness of the atmosphere layer where the showers develop (<~40 km) turn out into several advantages: a) the proximity effect is rather small (all the different segments of the shower lie approximately at the same distance, ±5%, from the detection optics). Even between different showers both the distance (the subtended solid angle) and the amount of traversed material, seen as an attenuation medium, varies at most by a factor cos(30°); b) the line of sight from the shower to the detector is aligned with the atmospheric density gradient (±30°). In particular, most of the fluorescence light doesn’t experience the Mie scattering on the Aerosol particles, mainly present in the boundary layer (h ≤ 2500 m a.s.l). Ground based experiments, whose line-of-sight, mainly for distant showers, almost entirely lies in this range, maximise the light attenuation and the correlated calibration problems due to this effect. This remains, nevertheless, also for EUSO the most troublesome effect, since it affects the light coming from the region where the shower maximum is 2 20 expected, mainly for vertical showers (Xmax≥ 800 g/cm at E>10 eV, corresponding to ∼2000 m a.s.l. for vertical showers). The two sources of signal in EUSO, fluorescence and Čerenkov light, suffer a different attenuation as a consequence of the atmosphere traversal. Fluorescence light in fact, isotropically emitted by the EAS, directly reaches EUSO from its source, eventually suffering multiple scattering effect. Čerenkov light on the contrary, which is highly beamed around the shower axis, travels with the EAS “front” down to its “beam dump” surface, and diffuses up again from it to the detection site. This means a double path: from the source point to the Earth (the clouds), and then back to EUSO, for that part that, according to the reflecting surface albedo, diffuses upward in the atmosphere. Figure 6.1.2.2-5 shows, for the same shower already shown in Figure 2.1-4, a typical one of the EUSO “zoo”, the effect of the atmosphere as an attenuation medium. The light curve, as a function of the arrival time to the detector, is shown comparing the amount of photons expected in a “completely transparent atmosphere” (no attenuation) to the same profile in a “clear sky atmosphere” (U.S. Standard), both showing the contribution of the fluorescence light, the Čerenkov flash on the reflecting surface and the Čerenkov light due to the light “backscattered from the Čerenkov beam”. The photons in the Čerenkov beam in fact, whose intensity is very high (it grows up to 1015÷1016 photons in the wavelength region 300÷400nm at the shower maximum), have a non negligible probability of being “back-scattered” in the EUSO “line of sight”, and contribute, starting from the shower maximum position, to the total amount of light collected by the EUSO optical system We will call pb(θ) this backscattering probability at an angle θ. This effect is discussed in EUSO-AS-REP-001.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 9/43

Figure 6.1.2.2-5 Time distribution of the photons arriving at the EUSO optics for a proton shower of E=1020 eV with an incidence angle of 60°. The fluorescence and Čerenkov contributions are distinguished. The black histogram corresponds to the photons flux if the atmosphere was totally transparent.

The comparison of “a standard, clear sky atmosphere” with what we expect as the “real atmosphere” seen by EUSO, has been dealt with in two different ways: • the atmosphere cloudiness modifies the detector acceptance and the expected duty cycle. The quantitative aspects of this modification will be discussed in detail in sect. 6.2.1 and 6.2.2 in a statistical way, on the basis of the study given in EUSO-SIM-REP-010. We have used the ISCCP International Satellite Cloud Climatology Project database [ISCCP, 1999] which provides extensive data on precise cloud situations around the globe as observed with satellite based atmosphere sounding devices. The ISCCP has divided the globe in equal 280x280 km2 areas and gives, on a 3 hour period, a number of cloud parameters for each of these cells. Such data have been collected from 1983 onwards. Among the many accessible parameters for each cell, we use:

◊ the clear sky fraction, ◊ the average cloud optical depth, ◊ the average cloud top altitude, ◊ the latitude, longitude and the nature (land, coast or sea). Notice that, due to the type of atmospheric sounding devices used on these satellites, the data used in these calculations correspond to daytime measurements. The extension to night-time condition is in progress. Special attention should be given to the notion of the local clear sky fraction. In fact, each cell is itself divided into a variable number of sub-pixels. Each of these pixels is observed for cloud presence and the clear sky fraction is the fraction of the clear pixel. This definition does not therefore take into account the topology of the clear or cloudy pixels. Due to the extension of an UHECR shower, this may lead to an overestimation of clear sky conditions. The effect of such over-estimation will be discussed below.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 10/43

As an example, Figure 6.1.2.2-6 gives, on the left, a representation of the total cloud amount (1-the clear sky fraction) over the globe, averaged over the 1983-2001 period. The graph on the right shows this quantity as a function of latitude. This data, combined with reference to the seasonal variation, shows that, although could situations vary significantly over short periods of time, the average in time is quite stable. This allows us to have some confidence on the average values and hence the statistical significance that will be extracted from the results presented below. Another relevant effect of the clouds on the light signal collected by EUSO is connected to the effect of the cloud albedo on the Čerenkov beam. While both Fluorescence and Čerenkov backscattered light reach EUSO travelling directly from the source to the detector suffering the attenuation due to their path and the spread due to multiple scattering, the reflected Čerenkov signal intensity is strongly related to the albedo of the surface where the EAS impinges. Since cloud albedo can be much more effective (as large as 80%) than Earth surface (<10%), this effect has a large impact on the shape of the observed signal when clouds are present, as it will be discussed in the next sections. An experimental measurement of the different surface albedo is being performed by the EUSO Collaboration (ULTRA experiment, see Annex B).

• the atmospheric situation in the EUSO FoV and its variability will however impact also on the signal shape on an event-by event basis. The clouds coverage, the cloud topology, all the thermodynamical parameters in the area where an EAS develops and the trigger comes from, has to be known and corrected for to have a correct reconstruction of the parent longitudinal distribution generated by the primary Cosmic Ray. To do this, a dedicated atmosphere sounding method is being developed within the EUSO proposal. A LIDAR system will be coupled to the main detection Instrument, with two acquisition modes: a free-running mode, when LIDAR will scan the FoV at low sampling rate (10 Hz) and supplement the existing meteorological and atmospheric satellites data; a triggered mode, with the possibility of pointing the region where a trigger has been detected and obtaining, with a fine sampling (100 Hz), the real atmospheric parameters, in the region where the event has been detected. The details of the LIDAR and the precision of its response function as far as the reconstruction of the EAS profile is concerned, are discussed in Chapter 3.

Figure 6.1.2.2-6 Example of ISSCP data. The mean annual total cloud amount. The clear sky fraction is defined as 1-total cloud mount. On the right, seasonal variations as a function of latitude.

As a concluding remark, we can summarise the effect of atmospheric transmission, from the point of view of the “end-to-end simulation”, introducing the transmission efficiency τm(r1,r2;λ) (for an atmospheric model “m”), from point r1 to point r2, of a photon of wavelength λ. For our purposes we can distinguish

• τm(r0,rE;λ) as the transmission efficiency for fluorescence photons and Čerenkov backscattered photons originating from point r0 and reaching the EUSO position rE;

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 11/43

• τ′m(r0,rE;λ)= τm(r0,rG;λ)×αG× τm(rG,rE;λ) as the transmission efficiency for Čerenkov photons originating from point r0 and reaching the EUSO position rE after a reflection on ground/clouds in position rG, characterised by a reflectivity αG (assumed as isotropic in the upper hemisphere [EUSO-SEA-REP-001]. This rather crude approximation will be improved also as a consequence of the results expected from the ULTRA experiment, see Annex B).

6.1.3 The Expected Background The background to the fluorescence signal detected by EUSO is being extensively studied: here again the innovative approach of the EUSO experiment plays a pioneering role. Looking at nadir, it will in fact be sensitive to the light coming from space and upward reflected through Earth albedo and to any source of light located in upper atmosphere below 430 km [Maccarone et al., 2001]. The major sources of background have already been discussed in Sect. 2.5. We will focus our attention here to the noise that this background can cause to the observed signal. The background sources we have been taking into account in the Phase A study are [EUSO-SIM-REP-006, Berat et al, 2003]: • natural night sky diffuse and slowly varying sources; • man made sources like city lights; • transient luminous phenomena in lower and upper atmosphere.

6.1.3.1 Natural night sky diffuse and slowly varying sources. The first category includes the whole sky stars and planets brightness, Moon phases and diffuse night sky brightness, as light coming from the outer space, to which EUSO will be sensitive through Earth albedo. The so- called Airglow is the main component of the light produced in upper atmosphere and contributes both as light directly detected by EUSO and as reflected light.

6.1.3.1.1 Diffuse night brightness. Leaving apart the contribution due to the Moon phases, which deserves a separate treatment deeply connected with the Experimental Duty Cycle, we evaluated the diffuse night sky brightness taking the data from the “1997 reference of the night sky brightness” work [Leinert et al., 1998] performed under the International Astronomical Union recommendations. For the band EUSO is concerned with, three main components contribute: the Zodiacal light for 100 photons/m2/sr/ns and the Diffuse Star Light (DSL or Faint stars) for 80 photons/m2/sr/ns, and the 2 Airglow (O2) which amounts to 300 photons/m /sr/ns, as a reference value [250÷600 as a range]. A lower contribution has been estimated from the distinct star and planets. Normalizing to the Sun flux in U band the total integrated flux from distinct stars incoming on the top of atmosphere will contribute to 190 photons/m2/ns. The sum of the three main components of the diffuse night sky light, as mentioned above, implemented with the distinct star flux averaged over 2π (30 photons/m2/sr/ns), amounts to a total flux (reference value) incoming above atmosphere (no moon, no planets): Φ =510 photons/m2/sr/ns corresponding to M = 22.0 /arcsec2 Let us take two different approaches for calculating the reflection on Earth surface and hence the expected level of the Background measured by the EUSO telescope, distinguishing situation where clouds are present or absent. The reflectivity of the atmosphere and the reflectivity of the earth surface are the dominant parameters which govern the intensity of the photon background when observed at Nadir. The optical depth of a standard clear sky atmosphere is around 0.9 in the considered wavelength range; this corresponds to a transmission of ~0.4 for a downward vertical flux through a vertical column of air to the ground. Excepted for ozone, there is no absorption. The fraction of the reflected light by atmosphere and the remaining fraction reaching the ground are roughly estimated in the following table, for various atmospheric transmission coefficient and an isotropic downward flux.

Transmission from top to ground 0.2 0.4 0.5 0.6 0.7 0.9 % Fraction of reflected light 42 35 31 26 22 10 % Fraction of light reaching ground 58 65 69 74 78 90 Table 6.1.3.1.1-1 Reflected and Transmitted light in atmosphere

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 12/43

For a total atmospheric transmission of T=0.4 above sea, whose reflectivity lies around 8%, 35% of the light is reflected by atmosphere and 5.2% by the sea. The total albedo is 40.2%. Above cloud whose top height is 3 km with a reflectivity of 40%, the transmission from top atmosphere to 3 km is now T~0.6, then 26% of the light is reflected by atmosphere above the cloud and the remaining is reflected by the cloud to 29.6%. In this case the total albedo is then 55.6%. The presence of clouds (with the above characteristics) should therefore increase the reflectivity by a factor 1.38. The above numbers are given as a guide for our estimate, but should vary according to the specific conditions. The reference value of Φ = 510 ph/m2/ns/sr turns then into a mean number of expected detected photons: 2 ΦD ≅200 ph/m /ns/sr in “clear sky condition” on EUSO from diffused night brightness

2 ΦDc ≅280 ph/m /ns/sr/ in “cloudy sky condition” on EUSO from diffused night brightness An important experimental cross check of the validity of our assumptions is given by the direct measurement done during the BABY balloon flights [Giarrusso et al., 2001; Catalano et al., 2002], expressly devoted to the measurement of the background for EUSO, and performed in 1998 and 2002 by a component of our Collaboration. These flight, performed in moonless nights over the Mediterranean area, where the pollution due to human induced background can’t be completely avoided, gave a mean value of: 2 ΦD ≅300 ph/m /ns/sr at 40 km height

which is fairly consistent with the extrapolation given in this Section.

6.1.3.1.2 Airglow. The night Airglow is possibly the most important source of background which EUSO will be sensitive to, for moonless and clear sky with no pollution light: it is an isotropic source of light located in a thin layer of upper atmosphere centred near an altitude of 100 km. It will be in direct view of EUSO looking at nadir from an altitude of 400 km, and it will also contribute through its reflected light from atmosphere (already taken into account for in the diffuse night sky light) back. The BABY measurements, performed at an altitude of ∼40 km, are therefore blind to its direct component. The production zone comes from the molecular oxygen dissociation by solar radiation, the maximum production rate occurs in a thin Chapman layer located near 100 km with a few kilometres width. Once produced, the atoms encounters collisions in the mono-atomic oxygen residual gas leading to the formation of molecular oxygen witch decays via the Hertzberg transition band located mainly in the 300-400 nm band. The intensity of the emitted light depends on the solar radiation exposure, then depends on the inclination on ecliptic (season and latitude variation) and on the initial solar flux (solar cycle variation). The relaxation time of atomic collision induces a local time dependence and a longitude variation. As already mentioned, taking into account the latitude range spanned by the ISS, we estimate the mean airglow intensity will vary between 250 and 600 ph/m2/ns/sr.

6.1.3.1.3 Total Moonless expected background. Table 6.1.3.1.3-1 summarises the total flux of photons expected to be seen by EUSO, as a result of the sum of atmospheric reflected light and of the direct view of the Airglow, that we can refer to as the “steady or slowly varying sources of background”. We will therefore discuss the EUSO performances (Sect. 6.2, 6.3) assuming 500 ph/m2/ns/sr as the baseline for the expected noise level, and also discussing the expected improvement in case of low level background, (300 ph/m2/ns/sr), as suggested by some experimental measurement [Levedinski et al, 1965].

Incident flux Reflected Light Airglow EUSO Background ph/m2/ns/sr Above Sea (Cloud) Direct View ph/m2/ns/sr Minimum 460 190 (260) 250 440 (510) Reference 510 200 (280) 300 500 (580) Maximum 810 320 (450) 600 920 (1050) Experiment BABY 300 250-600 550-900 Table 6.1.3.1.3-1 EUSO background

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 13/43

6.1.3.1.4 Moon Phases. An estimate of the Moon Phase that EUSO can tolerate impacts in a severe way with the expected duty cycle. In Sect. 2.5 we suggested that a tolerable level of moon-light induced background is 100 ph/m2/ns/sr. This value corresponds in fact to an increase of the expected background level of ∼20% with respect to the mean moonless value, which is of the same order of the incertitude given in Table 6.1.3.1.3-1. On the other hand, as it has already been discussed in Sect. 2.5 and in EUSO-SIM-REP-009, this value implies an increase of the expected duty cycle (total fraction of detection time) up to 25%.

6.1.3.2 Artificial light of the Night sky.

We expect that city lights and gas oil extraction flares are the most important sources of artificial light of the night sky. A detailed study is in progress, based on the first world Atlas of Zenith artificial night sky brightness at sea level was published recently [Cinzano et al., 2001]. Data were taken from nadir observation by satellites in the astronomical photometric V band. Flux data are mapped with an intensity relative to the diffuse night sky light flux, 2 which in V band is Bv= 860 ph/m /ns/sr. The intensity ranges from 0.1 Bv (above oceans) up to values greater than 27 Bv (above the largest cities). Even if the extrapolation from V-band to U-band is not straightforward, making the assumption that the same ratio holds in the U-map, with reference to the previously discussed reference value of 510 ph/m2/ns/sr, as it does in the V-band, we can guess the pollution level induced by the presence of the cities within our FoV. The guideline of the analysis is that a city will blind single pixels for a time corresponding to the persistency of its image within the relevant portion of the field of view. Since the ISS velocity is 7 km/sec and the pixel size is ∼0.7 Km, the expected persistence of a “steady human source” is of the order of 100 msec.

6.1.3.3 Transient Luminous Phenomena.

Electro-optical phenomena often occur in the lower atmosphere such as storms and their associated lightnings and in upper atmosphere such as elves, sprites and blue jets. Due to their transient aspect, luminosity and locations they can appear as fake events. They will have a non negligible influence on the EUSO duty cycle and on trigger efficiency, as far as the primary scientific objective is concerned. Their study can however be a scientific significant by-product of the mission. Their features are discussed in [Berat et al, 2003; Christian, 1999] and will be the subject of a dedicated simulation study in the fore-coming phases of the project, in order to evaluate the expected trigger rate. We summarise here, in Table 6.1.3.3-1, the number of events expected to occur within the EUSO FoV, assuming a duty cycle of 25%.

Lightning 400/day Sprites, Blue Jets ∼2/day Elves 20÷40/day Table 6.1.3.3-1 Expected Transient Atmospheric Phenomena

6.1.4 The Detector Simulation The height of the ISS orbit and hence the position of the Detection telescope is a crucial feature of the EUSO observational approach: the highest the height, the largest is the expected statistics of event falling inside the EUSO field-of-view; unfortunately, the highest the height, the smallest is the expected amount of collected light. Since in fact fluorescence light is expected to be emitted isotropically, a solid angle reduction factor:

2 εΩ= A⋅cos γ/4π(r0-rE)

has to be paid when light is travelling from the source position r0 to the detection position rE (A is here the optical system effective area, γ is the angle of the (r0 - rE) vector with respect to the telescope axis, with a common point in rE). The same solid angle factor, for the Čerenkov light reflected on ground will be given by:

2 ε′Ω= A⋅cos γ/2π(rG-rE)

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 14/43

where the normalisation factor takes into account that the radiation is assumed to be reflected isotropically only in the upper atmosphere, from point rG. Čerenkov backscattered light is collected with an efficiency εΩb, which 2 depends on the scattering angle θs under which the EAS trajectory is seen by the EUSO telescope (∝ 1+cos θs). With reference to the number of particles N(Ep,X(r0);AM) generated by a shower with a primary of Energy Ep, atomic number AM, at the point r0, after a depth X of traversed atmospheric matter, the photon signal intensity dNp produced at the telescope optical system entrance window, by a shower length dl is:

dNp (r0) = N(Ep,X(r0);AM)⋅{εΩ ⋅ [Σλεf⋅τ(r0,rE;λ)] + pb⋅εΩb⋅ [∫ (dNc/dl) ⋅ τ(r0,rE;λ)dλ ] + ε′Ω⋅ [∫ (dNc/dl) ⋅τ′(r0,rE;λ)dλ]} dl

where the integration is taken on the wavelength region 300nm≤λ<400nm and the efficiencies are, as defined in the previous sections:

• εf ≡ the wavelength-dependent fluorescence yield per unit length;

• dNc/dl ≡ the Čerenkov photon yield per unit length;

• pb(θ ) ≡ the Čerenkov backscattering probability at an angle θ, pointing to EUSO and measured from the shower direction;

• τ(r0,rE;λ), τ′(r0,rE;λ) ≡ the atmospheric transmission coefficient for direct and reflected radiation;

• εΩ,,εΩb, ε′Ω ≡ the geometrical solid angle factors for fluorescence, backscattered and reflected Čerenkov light. Figure 6.1.2.2-5 shows the arrival time distribution of the photons at the pupil entrance (with reference to the time of the first interaction). During Phase A study, the detector simulation has been done by a detailed parameterisation of the various efficiency terms entering the signal processing during the detection process: a) the amount of collected light, directly connected to the active diameter of the optical system; b) the FoV opening angle, directly connected to the effective aperture; c) the duty cycle, whose value is connected to the Detector design to the extent that: ◊ the photodetection system is able to filter the noise due to steady light sources; ◊ the atmospheric cloudiness can be recovered to the live time. d) the Point Spread Function of the Optics System here defined as the diameter of the area where 63% of the light intensity is contained (Sect....) e) the detector efficiency, including the Q.E. of the MAPMT’s, the dead area in the focal surface, the Front End elctronics efficiency, f) the photosensitive device (MAPMT) dimension, defining the pixel size on the focal surface and, convoluted with the PSF of the optical system, the (angular) pixel size in the FoV. Table 6.1.4-1 summarizes the baseline that we have assumed, for the simulation studies, during phase A, as far as the detector simulation is concerned. The optimisation of the single parameters will be discussed in the text.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 15/43

Parameter Value EUSO altitude 430 km EUSO active diameter (EPD) Ø 1.9 m EUSO Field of View (FoV) 30° ISS latitude (min,max) ± 51° EUSO observational duty cycle 18% - 25% (energy dependent) Optics triggering throughput Efficiency [1/(π⋅ (EPD/2)2)]{1,581-0,258θ2-2,131θ4} (see Optics Requirement Table) where θ is the angle of incidence(rad) on the EUSO optics (0°≤θ≤30°) Photodetector efficiency 0.12 ∗ Optics PSF(dRMS) 5 mm GTU 2.5µsec

Trigger parameters Threshold: Persistency= (Nthre:Npers)=4:4 Background noise level 500 ph/m2/nsec/sr ≈ 0.7 p.e. /pixel/GTU at nadir, ≈ 0.,5p.e. /pixel/GTU at FoV edge Focal surface pixel size 0.8x0.8 km2 at nadir for M36 MAPMT Cloud description ISCCP: cloud fraction, cloud top height, cloud optical depth Table 6.1.4-1 EUSO Detector parameters baseline assumed during Phase A Simulation studies

It should be emphasised that the baseline for the simulation studies comes out from an f# 1.25 for the optical system. A recent improvement (Chapter 5) switched to a more fast system with f# 1.0. This obviously implies >40% improvement in the entrance pupil area and hence of the collected light from EAS, (and a different optical throughput) whose effect is under study as far as the overall detector performances is concerned (mainly the reduction of the energy threshold for the primary Cosmic Rays). The final end-to-end- simulation will be based on a software architecture, called ESAF, EUSO Simulation and Analysis Framework [EUSO-SDA-REP-008], based on the world-wide accepted ROOT [Brun et al., 2001] and GEANT4 [Agostinelli et al. 2003] packages, whose design has been completed during Phase A, and whose implementation did already start with the development of a detailed Detector description. The Detector Simulation contained in ESAF, whose 1st version has already been released to the Collaboration and is actually being tested will contain a direct tracing of the single photons and the detailed geometry of the single photosensitive elements. The actual detector simulation software has been used to work out the trigger efficiency and to optimise it. Figure 6.1.4-1 shows, for the same event shown in Figure 6.1.2.2-5, the expected photo-statistics. The number of photons registered per Gate Time Unit is shown, together with the expected noise level. As it has already been outlined in Chapter 2, where the EUSO observational approach has been described, the main ingredients for the trigger logic are:

• the Gate Time Unit (GTU), ∆tGTU. It is the basic time unit for counting the number of photoelectrons Npe in a pixel, with a double pulse resolution time of 10 ns;

• the photoelectron threshold Nthr (i.e. the minimum number of photoelectrons Npe piling-up in a pixel in a GTU, such that for Npe>Nthr that pixel will be set as “hit” ;

• the persistency level Npers (i.e. the minimum number of consecutive GTUs in which an EAS is expected to develop, giving in EUSO a detectable signal, Npe>Nthr).

The first level trigger occurs when the “hit” condition Npe>Nthr is detected with a persistence larger than Npers in a given “space trigger unit”, the macrocell, obtained OR-ing the “hits” of n×n pixels. This trigger scheme takes therefore advantage of the very special space-time correlation that is produced by the shower. The random photon noise background, produced by natural (starlight, Airglow of moonlight) or anthropomorphic sources do not possess such specific space-time correlations and will therefore be eliminated,

∗ dRMS is defined as the “spot size”, i.e. the RMS diameter of a δ-focused light beam on the fpcal surface.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 16/43

with a certain efficiency, by this trigger scheme. A second trigger level, consisting in a fast “space proximity test” of “hits” in consecutive GTUs allows to further reject the fake trigger due to the random noise. Thus, the setting of the trigger parameters is determined by a trade-off between the desired extension of the energy range downward toward faint showers (keep Nthr as low as possible) and a) the need of a tolerable rejection power against the fake trigger contamination of the sample , b) a track bright and long enough to be reconstructed.

The ∆tGTU=2.5 µs corresponds to the minimum time needed by the particles of a (horizonal) shower to pass through the FoV of a single pixel.

Figure 6.1.4-1 Same as figure 6.1.2.2-5 but for a realisation of detected photoelectrons.

Figure 6.1.4-2 shows the expected trigger efficiency, as a function of the primary energy Ep, for different choices of the triggering condition (Nthr:Npers) and for “contained events”, whose hit point with the ground surface is inside the EUSO FoV. The fake trigger rate depends on the expected background level and the dimension of the macrocell: with a background level of 500 ph/m2/ns/sr and a macrocell grouping of 6×6 MAPMTs, each containing 6×6 pixels, i.e. n×n=1296 pixels, the rate of first level fake trigger rate can be handled by a fast second level “track finding algorithm” when (Nthr=4:Npers=4). Figure 6.1.4-2a thus represents the expected “energy threshold behaviour”. Figures 6.1.4-2b,c show how the expected scenario is changing when a more stringent trigger condition would be required (as a consequence of a higher background environment, for instance).

(a) (b) (c)

Fig. 6.1.4-2 The EUSO detection efficiency as a function of energy for three different trigger conditions: (a) Nthr=4, Npers=4; (b)

Nthr=4, Npers=5; (c) Nthr=5 Npers=5. Baseline (blue curves) and Requirements (red curves) Optics Performance are reported.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 17/43

From the above consideration, it follows that the setting of the trigger parameters will depend on the background level as well as by the amount of UHECR photon production, itself dependent on the shower energy

(E shower) and on the shower angle (θshower).

6.2 Flux and Sensitivity

6.2.1 Acceptance and Aperture Because the energy and the angle of incidence of the primary UHECR will vary, actual photon spectra will vary in shape and intensity. For large incident angles, the spectrum will spread over a much larger time span. Beside the ideal shower, shown in fig.6.1.2.2-5, there will be a large “zoo” of possible situations and space-time shapes of the signal that will trigger EUSO. Figure 6.2.1-1a shows the case of a 85° shower. One observes that because of its inclination, the Čerenkov beam is exhausted on its way down to earth by the diffusion atmosphere processes. The resulting Čerenkov reflected peak (t=3100 µsec) is very small and will probably not be observed by EUSO. In this case the number of photons (p.e.) detected would be 20 000 (1140). The case of a perpendicular shower gives a different picture. Figure 6.2.1-1b shows the case of a vertical shower encountering a dense cloud whose cloud top is located at 3 km. In this case, the fluorescence bell-like shape is truncated and the Čerenkov peak is considerably enhanced by the large albedo (40%) of such clouds. It is therefore quite clear that, in view of the varying atmosphere situations and the varying angle of incidence, the quality of the detection of UHECR showers will vary greatly.

(a) (b) Figure 6.2.1-1 Time distribution of the photons arriving at the EUSO optics as for figure 6.1.2.2-5 but for two different shower situations: (a) a quasi-horizontal (θ=85°) proton originated shower. Notice the small reflected Čerenkov peak at t=3100 µsec; (b) a vertical shower (θ=0°) proton shower in the presence of a dense cloud at 3km altitude. The fluorescence max is hidden, the Čerenkov peak is instead considerably enhanced.

In order to quantify the “quality” of shower detection by EUSO we define two detection criteria: Fmax and Cp, The Fmax condition corresponds to the fact that the showers has “triggered” the EUSO detector (see below) and that the maximum of the fluorescence peak has been observed at a sufficient p.e. (photoelectron) level. The Cp condition corresponds to the observation of the Čerenkov peak at a sufficient p.e. level. We then group the detected showers in four different categories: 1) showers where both Fmax and Cp conditions are satisfied, 2) showers where only Fmax is observed, 3) showers where only Cp is satisfied, 4) neither Fmax or Cp is satisfied.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 18/43

Showers belonging to group1 are called “golden showers” as they should provide the best conditions for energy and angle of incidence measurements. Group 2 showers should normally provide a correct energy measurement but in reality, as atmosphere corrections need altitude information, the absence of the Čerenkov peak will degrade this information (except for quasi-horizontal showers where the altitude is can be derived from the time duration of the fluorescence signal). Group 3 showers will be difficult to analyse in most cases. If the fluorescence peak is large enough to be detected (even if its maximum is not seen) some directional and energy information may be extracted. The important Čerenkov peak that usually accompanies such showers (most of these showers come about because of the presence of a dense cloud top) may give further information. Group 4 showers are lost for EUSO. Using the detector baseline as it has been described in Sect. 6.1.4 and the trigger condition discussed therein, the detector acceptance has been studied, for proton induced showers, generating a large number of (proton) UHECR showers. The showers are generated in such a way as to represent, statistically, an isotropic flux of UHECR. The energy of the shower is varied from 1019 eV to 1021 eV, which spans the energy range that can possibly be covered by EUSO. The generated showers represent all possible geometrical configurations: • Type 1 showers: downward going showers that hit the earth surface within the EUSO field of view; • Type 2 showers: downward going showers whose impact point falls outside the EUSO FoV, but which intercept it; • Type 3 showers: upward going showers that intercept the EUSO FoV. In Figure 2.4-1, repeated herewith as Figure 6.2.1-2, we showed the “acceptance” of EUSO in the case of 20 perfect clear sky conditions and for 10 eV protons showers as a function of the shower angle of incidence (θshower). The dashed red curve in that figure shows the geometrical limit of the acceptance. The black curve gives the acceptance for golden showers. The red histogram shows the acceptance of showers when only the Čerenkov peak is identified (condition Cp). The green histogram shows the exclusive detection acceptance for Fmax showers. In this (optimistic) situation, the acceptance for golden showers represents 80% of the maximum geometrical acceptance. This geometrical aperture is equal to 608 000 km2 sr.

Figure 6.2.1-2 Acceptance of EUSO as a function of the shower angle, for protons of 1020 eV in the case of a clear sky. The acceptance for golden showers is represented by the black histogram. The red

histogram show the showers with Cp condition, the green one those for which only Fmax is observed.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 19/43

Using the statistical data obtained from the ISCCP, as discussed in Sect. 6.1.2.2, the acceptance of EUSO can be calculated in the same manner for cloudy scenarios. Figures 6.2.1-3 show this acceptance, as a function of 19 20 21 θshower for proton energies of 5×10 ,10 and 5×10 eV. For the case of 1020 eV proton, one observes that showers are correctly detected in 50% of the cases (black histogram). Most of the incorrectly detected showers are showers where the Fmax condition has not been satisfied, whereas the Čerenkov peak (Cp) is seen. This is due to high dense clouds that mask the fluorescence maximum and reflect (with a high albedo) the Čerenkov photons produced prior to encountering the cloud top. Because of the Persistency threshold parameter (Ntpers), most of these showers are rejected: the red histograms show however the statistical significance of those showers that, even if the shower maximum is not detected, do satisfy the trigger condition. A special reconstruction algorithm should be developed to analyse them, as many information as possible from them (direction of shower, energy of shower). They are however not taken into account, in this report, to work out the expected flux sensitivity (Sect. 2.5). A technical detail is worth to be mentioned, as far as the ISCCP statistical data are treated. In ISCCP data, the clear sky fraction represents 35% of the cases. Special attention has been given to the notion of the local clear sky fraction. In fact, each cell is itself divided into a variable number of sub-pixels in the ISCCP archive. Each of these pixels is observed for cloud presence and the clear sky fraction is the fraction of the clear pixel. This definition does not therefore take into account the topology of the clear or cloudy pixels. Due to the extension of an UHECR shower, this may lead to an overestimation of clear sky conditions. The effect of such over-estimation has been roughly estimated as follows: we have biased the ISCCP data by assuming that, in each cell, clouds are always present and that their characteristics are given by those of the cloudy pixels of this cell. Note that in 2% of the cases, the cell is totally clear and that, therefore, cloud data are not available.

19 20 20 (a) Ep = 5×10 eV (b) Ep = 10 eV (c) Ep = 5×10 eV Figure 6.2.1-3 The EUSO acceptance as a function of the shower zenith angle, for different primary proton energy,in case of ISCCP cloudy sky. The acceptance for golden showers is represented by the black histogram.

Doing so, we can derive a lower limit to the acceptance in a “full cloudy sky”: the expected fraction of 50% of Golden events, is reduced to ~45% at 1020 eV for the primary proton energy.

6.2.2 Flux and Sensitivity - Extreme Energy Cosmic Rays The expected flux of the EUSO experiment has been discussed in Chapter 2.5, as far as the UHECR are concerned. We will therefore concentrate here our attention to the EUSO sensitivity to UHE neutrino events and on the other physics items.

6.2.3 Flux and sensitivity - Extreme Energy Neutrino EUSO, with a field of view FoV=60o from 430 km altitude, will be able to monitor from the space an atmospheric mass M ≈ 2×1018 g with an energy threshold close to 5×1019 eV. Such enormous amount of target mass will offer the unique opportunity to detect cosmic neutrino events at extreme high energy.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 20/43

Electron, muon and tau neutrinos interact through deep inelastic scattering with the nucleons of the atmosphere, generating an electromagnetic plus an hadronic detectable cascade in case of νe charged current (CC) interactions and only a hadronic detectable shower in all the other cases (the probability to detect muon bursts and tau double bang is small and is not considered for the moment). Hence the tau and muon neutrinos charged current interactions will be seen like a neutral current one (NC). The produced showers will be detected and reconstructed using signals coming from the fluorescence light emitted isotropically along the shower, from the Čerenkov light backscattered in the atmosphere and from Čerenkov light diffusively reflected from the ground. Since the expected shower statistics, collected in three years of data taking, is of the order of 103, neutrino showers must then be recognized and selected from showers induced by protons or nuclei, with a rejection power of 10-4. The simulation of shower development, fluorescence and Čerenkov light generation and transmission, and detector response has been performed here using UNISIM simulation package. For the relevant effect of LPM on neutrino showers see also Bottai, 2001. Due to the weakness of neutrino cross sections the neutrino flux is not significantly attenuated while traversing the atmosphere (even at the extreme high energy and for horizontal directions). As a result the probability distribution for the altitude of neutrino interactions inside the atmosphere follows the density profile of the atmosphere itself (Figure 6.2.3-1) and the main signature which identifies neutrino showers from hadron/photon showers is the slant depth where the showers takes place. Figure 6.2.3-2 shows a sample of simulated almost horizontal showers initiated by protons and neutrinos CC interactions. A selection based on the slant depth of the shower maximum appears at the cleanest signature to reach the desired rejection power, as shown in Figure 6.2.3-2, upper plot.

Figure 6.2.3-1 Altitude (in m a.s.l.) of shower Figure 6.2.3-2 The longitudinal profiles for νe (CC) maxima. Red points are for protons (3 years of data (green) and proton (red) showers, at θ=80o and taking with 10% duty cycle) while green points are E=1×1020 eV. Upper plot: as a function of slant depth. neutrinos with arbitrary large statistics and E-1. Lower plot: The width around the shower maximum.

Neutrino showers do in fact exhibit several different features with respect to hadron induced showers, due to the different superposition of hadronic and electromagnetic showers. However, such features have large fluctuations and can hardly be used to give a clear signature, identifying primary neutrinos from protons The main distinctive feature of shower profiles in space is that low altitude neutrino showers will be shorter than high altitude proton horizontal showers due to the different atmospheric density in which they develop (Fig. 6.2.3-2, lower graph) In Figure 6.2.3-3 the position of the shower maximum (Xmax) is directly plot, both for neutrino and hadron initiated showers. The Xmax selection criterion, which appears very effective, is however based on the accuracy of Xmax reconstruction (Sect. 6.3.3). Once the shower direction has been reconstructed the arrival time of the

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 21/43

Čerenkov reflected light will give a very precise signature of the depth of the shower. A neutrino selection of 2 Xmax>1400 g/cm can be in principle used, while a more realistic value for selection will be established only after a sophisticated and detailed study of possible ghosts events and long migration in reconstructed values which survive to quality cuts. For almost horizontal showers, however, the probability of detecting the Čerenkov light reflected on ground surfaces drops in a significant way (Figure 6.2.3-4), and hence the Xmax reconstruction accuracy. These showers, very inclined, are characterised for being also low altitude tracks in the neutrino primary case. Thus, they can be discriminated from protons looking at their space development. The time duration of the registered signal in the detector appears a clear signature to discriminate proton from neutrino showers for the cases where no Čerenkov light is detected (Figure 6.2.3-5) . Better results are expected from more sophisticated shape analysis which are in progress. It is useful to characterize the detector performance in terms of target mass for neutrinos. For such purpose we have simulated neutrino interactions (CC + NC) using an isotropic angular distribution over 0o < θ < 90o inside the 18 ν EUSO FOV (2×10 g of atmosphere). Then the effective neutrino acceptance Aeff (g sr) for an isotropic flux can be calculated as: ν 18 Aeff (g sr) = 2π 2×10 (triggered and selected events)/(generated events)

and the event rate Rν e for an isotropic neutrino dφ/dE is: ν Rν = ∫ dφ/dE Aeff NA σν η dE

where σν is the neutrino total cross section cross section on nucleon iso-scalar target (CC or NC), NA the Avogadro number, and η is the duty cycle.

Figure 6.2.3-3 Slant depth of shower maximum Figure 6.2.3-4 The angular distribution of neutrino triggered (Xmax) for protons (red) and neutrino (green) showers (blue) The same when Čerenkov flash detection showers. Same event sample as for fig 6.2.3-1. is required (red) (albedo of ground=0.08). Same event sample as for fig 6.2.3-1

ν The results for Aeff are shown in Figure 6.2.3-6 for different selection option and in case of clear sky and clouds distribution (see previous paragraphs). The detection efficiency ε =(triggered and selected events)/(generated events), averaged over the isotropic angular distribution, is always well below because a significant part of neutrinos, apart from the almost horizontals where such ratio approach unity at higher energies, interact so close to the ground that the shower has not enough space to properly develop inside the atmosphere.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 22/43

As a consequence of the EUSO ability of detecting UHE neutrino-induced showers, the flux sensitivity limit of ev/(year×energy decade) is shown in Figure 6.2.3-7, compared to different theoretically expected fluxes of astrophysical neutrinos.

Figure 6.2.3-5 The time duration of the neutrino showers with no detectable Čerenkov flash in Fig 6.2.3-4 sample.

Figure 6.2.3-6 Neutrino acceptance with clear sky Figure 6.2.3-7 Neutrino fluxes per flavour (full mixing (blue lines) and with ISCCP expected distribution of neutrino oscillation scenario): AGN(A) and AGN(B) from clouds frequency (red lines). Continuos lines are for Mannehim 1995; GZK(A) from Protheroe and Johnson, CC interactions while dotted lines are for NC 1995; GZK(B) from Kalasev et al., 2002; TD(A) from interactions. Left panel use events selected to fulfill Protheroe and Stanev, 1996; TD(B) from Sigl et al., 1997; trigger condition and Xmax> 1400g/cm2, in the right TD(C) from Sigl, 2000; SM Relic particle from Kalashev panel we add also the selection for the visibility of et al., 1999; ZBURST from Kalasev et al., 2002. shower maximum (Xmax above clouds or ground). The red thick line represents the neutrino flux per flavour which produce one events/year per energy decade in EUSO in case of clouds and shower maximum visible (lowest curve for CC and NC effective acceptance in fig. 6.2.3.6), with duty cycle 25%.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 23/43

6.3 The Experimental Resolution As described elsewhere in this document, the EUSO detection system is based on the imaging of the shower through the single photon counting in any pixel (~0.8×0.8 km2 on ground at nadir), within a very short time window (the experiment Gate Time Unit, GTU). The shower will appear as a “single track” event whose duration, position and intensity are related to the arrival direction, energy and nature of the primary UHECR particle. The tracks are observed in presence of the intrinsic background produced by any event giving light in the same UV band of interest for EUSO.

6.3.1 Direction The space-time image of the shower triggering the telescope will be mainly given in terms of X-T and Y-T projections of the collected photoelectrons, X and Y being the coordinates inside the field-of-view; the time coordinate T measures the shower development in depth, thus providing info about the shower length in the third direction (the height in the atmosphere). The reconstruction of the two projections X-T and Y-T will therefore give, in the ideal case, an unambiguous estimate of the shower direction, to the extent that the reconstruction algorithm is able to recognize the sequence of correlated space-time pixels out of the uncorrelated background. Different methods, based on linear fitting or on pattern recognition procedures, have been used to determine the arrival direction of the shower, and all methods gave us results in strict agreement among them.

Figure 6.3.1-1 Schematic view of a track along the Figure 6.3.1-2 Average length (and RMS) in GTU units two projection view (as registered by the read-out (2.5 µs) of the significant fluorescence tracks as extracted electronics). Shower simulated at energy ~5×1020 by the clustering procedure (see text). eV and arrival direction zenith angle ~ 60°.

Just as an example of reconstruction algorithm, if we only take into account the position of the “fired pixels” along the two X-T and Y-T projection views, we can apply binary pattern recognition to extract the significant tracks, followed by fitting techniques to determine their shape parameters, and then 3D geometry to finally combine the information and to derive the arrival direction of the EAS shower (a detailed description of this procedure, named CARMF, Cluster Analysis and Robust Median Fit, is given in the note EUSO-SDA-REP-002). Figure 6.3.1-1 shows an example of such data seen as binary images. From each projection view, the significant track is extracted by using a two-dimensional single-link clustering algorithm: the data are seen as a random graph where the weight is the Euclidean distance between the nodes. Given a suitable threshold, defined as Euclidean distance in the space of each projection plane, a cluster track will be identified by all those pixels whose inter-distances are less than the threshold; the track will be considered significant if it contains more than a minimum number of points. Clustering threshold and significant number of points are automatically evaluated from the statistics of the data set. Fig. 6.3.1-2 shows the dependence of the cluster track length on energy and zenith angle of the simulated showers. The CARMF procedure has been applied on a large set of data (at least 1000 events for each sub-set) simulated at energy ranging from 3×1020 to 1021 eV, and for different zenith angles of the shower arrival direction. The sample

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 24/43

dataset was simulated under the following main conditions and parameters: proton as primary particle, no clouds present in the atmosphere, no information about the reflected Čerenkov light, GTU equal to 2.5 µs, background level equal to 300 photons m-2 ns-1 sr-1, trigger level ≥ 5 pe/GTU/pixel, and persistency level equal to 3.

Named (θ0,φ0) the zenith and azimuth angles of the simulated shower, and (θ,φ) the reconstructed ones, respectively, the “angular resolution”, AngRes, is defined as the angular distance in the sky between reconstructed and simulated arrival direction:

-1 AngRes = cos (sin θ ·sin θ0 ·cos φ ·cos φ0 + sin θ ·sin θ0 ·sin φ ·sin φ0 + cos θ ·cos θ0)

Let us define “Encircled Power” the percentage number of our samples enclosed within a given angular distance. As an example, Figure 6.3.1.3 (left panel) shows the behaviour of the Encircled Power for the sample data set simulated at 3×1020 eV shower energy. The dependence of this estimator from the energy is detailed in the right panel of Figure 6.3.1-3 for the sample data-set simulated at 30° and 75° zenith angle (with respect to the EUSO axis) of the arrival direction.

Figure 6.3.1-3 Encircled Power vs Angular Resolution of the reconstructed data sets (CARMF procedure). Left panel: zenith angle dependence for showers simulated at energy 3×1020 eV. Right panel: energy dependence for showers simulated at 30° and 75° zenith angle with respect the EUSO nadir axis.

Such a kind of representation allows us to read the expected Angular Resolution at any level of Encircled Power we consider. Figure 6.3.1-4 shows the value of AngRes as obtained at 68% level for the entire sample data- set, while the schematic representation given in Figure 6.3.1-5 refers to the AngRes average value (the radius of the circles). Notice that the AngRes being a positive definite quantity, its average occurs at a level less than 68% of the Encircled Power. The results show that the arrival direction can be reconstructed with an accuracy ranging from tenths of degrees (0.2°) to a few degrees, depending on several variables and parameters as the EAS incident zenith angle and energy, the pixel size on the focal surface, the photosensitive device (MAPMT) dimension. The angular resolution is also strictly related to the length of the track, and this length depends on the energy as well as on the inclination (zenith angle) of the arrival direction of the shower (see Fig.6.3.1-2); the results confirm our expectation: at higher energies the angular resolution improves as well as it does so for the most inclined showers. It has to be noted that showers with only fluorescence information were analysed through the CARMF procedure, so these results are expected to improve when the Čerenkov landing-point flash data is added, independently from the reconstruction method used. Figure 6.3.1-6 shows the behaviour of the Angular Resolution, determined via a linear fitting method, when the Čerenkov foot-print information is used (golden showers). In the figure, the dependence on different pixelisation of the focal surface detector (MAPMT36 and MAPMT64) is also shown as well as the effect of the aberration in the arrival direction reconstruction. By weighting these results with the EUSO acceptance for “golden” showers in the case of cloudy sky (as previously shown in Figure 6.2.1-3) the average value of the Angular Resolution is of the order of 2.7° for showers simulated at 1020 eV.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 25/43

Figure 6.3.1-4 Angular Resolution at 68% level of Figure 6.3.1-5 Average values (circle’s radius) of the Encircled Power for the entire simulated sample data Angular Resolution vs shower energy and zenith set (CARMF procedure, fluorescence events only). angle of the arrival direction (CARMF procedure, fluorescence events only).

Figure 6.3.1-6 Angular Resolution for showers simulated at energy 1×1020 eV and for different pixelisation of the focal surface (MAPMT36-baseline and MAPMT64). The effect of aberration in the arrival direction reconstruction is also presented (linear fit procedure on “golden showers”).

In conclusion, even in this preliminary analysis, the results prove the EUSO capability of addressing astronomical investigations. The directional resolution accuracy in fact shows to be of the same order of the expected bending of protons in an extragalactic magnetic field.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 26/43

6.3.2 Energy Resolution As we already discussed when estimating the EUSO acceptance, we can distinguish three categories of events triggered by EUSO, which will have different precision in their reconstructed features, and have to be treated in different ways, from the point of view of the reconstruction:

• Events where the maximum of the fluorescent signal is detected (Fmax condition described in Sect. 6.2 is satisfied, and we call them “Fluorescent events”),

• Events where the Čerenkov footprint signal is detected (Cp condition described in Sect. 6.2 is satisfied, and we call them “Čerenkov events”),

• Both fluorescent and Čerenkov signals (Fmax.AND.Cp conditions are satisfied, and we already defined them as “Golden events” in Section 6.2). Here we describe in short our reconstruction algorithms for Golden events and for Fluorescent events, mainly as far as the energy reconstruction is concerned and, in the following subsection 6.3.3, as far as the estimate of the position of the shower maximum is concerned. The detailed description of the energy reconstruction algorithms are summarized in EUSO-SDA-REP-016 note. Notice that Golden events do participate also to the Fluorescent events category. Its worth to notice the complementarity of the two categories: the events where the Čerenkov footprint position information is not present, “Fmax-only” events (see Figs. 6.2.1-2,3) are dominating in the large zenith angle acceptance region, where “Golden” are rather severely suppressed. The two reconstruction methods, when applied to “Golden” and “Fmax-only” events, cover essentially the full domain of zenith angles. When we talk of “Fluorescent events” we mean treating all the events, also the Golden ones, as the Čerenkov footprint position were not present, thus comparing, in the angular region θ ≤ ∼70°, the two reconstruction procedures. The primary energy estimators most commonly used in fluorescence experiments, where the longitudinal profile of the shower is measured, are the track integral and the number of particles at the shower maximum. This means, in the EUSO case, that the energy of the primary particle can be reconstructed using either using the integrated number of p.e. of the shower or the number of p.e. in the shower maximum, or both of these quantities. The small number of p.e. detected per one GTU makes the second method statistically less accurate. We will therefore concentrate on the estimate of the track integral. As it is shown in EUSO-SDA-REP-016, the total number of photons produced by the shower is inversely proportional to the air density along the track. Therefore we can use, as an energy estimator, the information on the density at the position of the shower maximum; we find that: -1 E ∝ ρ(hmax) Ntot <εdεfτatm> where Ntot is the total number of p.e. detected, <εdεfτatm> is the average product of the fluorescent yield εf , of the detector efficiency εd and the atmosphere attenuation τatm, and ρ(hmax) is the air density at the shower maximum. We reconstruct the energy using this kind of relation assuming a Poisson fluctuations of p.e. and the following relative errors to account for the uncertainties of the atmosphere attenuation (± 7.5%) and fluorescent yield (± 5%). Figure 6.3.2-1a,b display the relative error in the energy determination for the complementary sets of “Golden” and “Fmax-only” events, i.e. the events in which one can not take advantage of the Čerenkov footprint signal. One can conclude that a resolution ∆E/E∼25% can be achieved in both cases. For the sake of completeness, Fig. 6.3.2- 1c shows the same histogram, worked out for the full sample of Fluorescent events. From their comparison it shows up how the knowledge of the Čerenkov position reduces the systematics in the energy estimate of the shower. As it will be shown in Sect. 6.3.3 this effect is mainly due to a systematic underestimate of the reconstructed height of the shower maximum in this case (see Figure 6.3.3-8).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 27/43

(a) (b) (c)

Figure 6.3.2-1 Reconstructed/Simulated Energy distribution for Golden events (a), Fmax-only events (b), fluorescent events (c).

Fig.6.3.2-2,3,4 show the ±∆E/E (RMS) of (Reconstructed-Simulated)/Simulated energy distribution for Golden (red) and Fluorescent (blue) events as a function of UHECR energy, the zenith angle and the shower position within the FoV, expressed as the angular distance from the Nadir. One can conclude that, at 1020 eV EUSO will be able to measure the UHECR energy with a resolution ∆E/E=±25% for Golden events (±30% when only fluorescence maximum is detected) which is reduced to ±20% for more energetic showers. An examination of Fig.6.3.2-4 shows that energy reconstruction moderately depends on the UHECR zenith angle however being more efficient for near horizontal showers. The energy reconstruction precision appears rather independent from the shower maximum position in the FoV of EUSO telescope.

Figure 6.3.2-2 RMS of (Rec.-Sim.)/Sim. Figure 6.3.2-3 As for fig. 6.3.2-2, Figure 6.3.2-4 As for fig. 6.3.2-2, as a energy distribution for Golden (red) and as a function of UHECR zenith function of the position of the shower Fluorescent (blue) events as a function angle. maximum within the FoV (in deg from of UHECR energy. the nadir direction).

This kind of analysis represents a first attempt to work-out, from the collected data, the estimate for the energy of the incoming primary. It is however worth to mention that the goal for the next phases of the project development is to improve the reconstruction algorithm and the retrieval of the triggered shower, by exploiting all the capabilities of the experimental technique. Table 6.3.2-1 summarizes the intrinsic errors we expect to have on the quantities which affect the retrieval of the longitudinal profile of the shower from the detected p.e. space-time shape. Just propagating these errors we can set the intrinsic energy resolution to ∆E/E = ±14%, as a goal for the further development of the reconstruction algorithms. The largest contribution is still connected to the statistical error due to the photoelectron collection efficiency. A careful control of the “in-flight” performances of the detector, including the features of our active target, is a mandatory requirement to reach this level of accuracy. Notice that no systematic error has been included, at this level, to take into account for the mass incertitude of the primary cosmic ray.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 28/43

Parameter Error on each parameter (est.) Error observed on Energy ±FWHM/2 Missing Energy ± 2.5% Energy Æ e- ± 2.5% 1/2 p.e. statistics (~ ± 1.2N p.e. ) ± 12% (mean value for 1020eV showers) e- Æ fluorescence ± 10% Æ aiming ± 5% ± 5% Atmosphere Transmission ± 7.5% Æ aiming ± 5% ± 5% Optics+Detector ± 5% ± 5% Temperature+Pressure ± 1% < ± 1%

All ± 14% Table 6.3.2-1 List of the error estimates for different parameters and their impact on the energy measurement.

6.3.3 Elemental Composition The position of the shower maximum is the most powerful estimator of the mass of the primary hadron. Under this respect, the so-called ”Golden events” give a more precise signature because of the existence of the absolute measurement of the Čerenkov foot-print on Earth or clouds. For these events the reconstruction of the altitude of the maximum of a shower having both fluorescent and Čerenkov signals relies on the arrival time delay between the light originating from the shower maximum and that coming from the Čerenkov spot on a reflective surface (Earth, cloud,…). In Figures 6.3.3-1,2 we display both absolute (in km) and relative errors in the Hmax determination for Golden events. The spread of the precision at large zenith angles is related to the absolute uncertainty in the height of the maximum, which occurs at larger values for inclined showers. One can conclude a resolution better than 10-15% for Hmax determination.

Figure 6.3.3-1 Absolute error (in km) in the Hmax Figure 6.3.3-2 Relative error in the Hmax reconstruction reconstruction for proton initiated showers. for proton initiated showers.

Figure 6.3.3-3 show, for the same golden events the Hmax correlation plot, between generated (mc) and reconstructed values, in case of proton and neutrino showers. Figure 6.3.3-4 shows, for proton showers, the 2 correlation plot in the variable Xmax (slant depth in g/cm ), which is the most powerful indicator of the primary particle 2 nature. A precision of ∆Xmax< ±50 g/cm is achieved, also in this preliminary stage of the analysis, compatible with the possibility of discriminating heavy from light primaries. Figure 6.3.3-5 shows the expected Xmax behaviour for proton, oxygen and iron primaries.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 29/43

Figure 6.3.3-3 Correlation between simulated and reconstructed values of Hmax for proton initiated showers (a), and for neutrino initiated showers (b).

2 Figure 6.3.3-4 Correlation plot, in Xmax (g/cm ), for Figure 6.3.3-5 The expected Xmax behaviour for proton, proton initiated Golden showers. iron and oxigen showers.

Reconstruction of the shower maximum parameters for Fluorescent events is less evident since there is no reference point like a reflected Čerenkov spot. However one can use the intrinsic properties of the fluorescent light production in a shower to reconstruct a characteristics altitude of its development. As it has been discussed in Sect. 6.2, we expect to observe showers with no Čerenkov peak at large zenith angle (θ >70°). The identification of the shower altitude is in this case connected to the hadron.vs.neutrino shower rejection. As it has been discussed in Sect. 6.2.3 one expects the time duration of the shower to be a clear signature of high altitude hadron showers.vs. low altitude neutrino showers. Figure 6.3.3-6 shows a temporal development of the fluorescent signal for two horizontal showers originated at 5 km altitude (dashed line) and 20 km altitude (solid line). One can note that the second shower is significantly longer in time with an order of magnitude larger total number of photons than the first one. This is easy to understand noting that the total number of fluorescent photons is proportional to the fluorescent yield per unit length which is slowly varying function in range from 0 to 20 km and to the length of the shower which is inversely proportional to the air density. Thus qualitatively one can conclude that the ratio of the number of photons in the shower maximum to the total number of photons is sensitive to the air density and thus to the altitude of the shower development. We generalize this idea for the case of arbitrarily inclined showers what makes possible to reconstruct altitude of the shower maximum relying of fluorescent information only. Figure 6.3.3-7 shows the correlation plot between simulated and reconstructed Hmax for proton initiated showers. A typical error in the Hmax determination is about 0.4-0.5 km with 0.3-0.4 km gaussian sigma spread. In Figure 6.3.3-8 we display a relative resolution of the determination of Hmax of proton initiated showers. One can clearly see that this method gives 20-40% error in the Hmax determination for near vertical showers and becomes as precise as 15-20% at zenith angles larger than 60o.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 30/43

Figure 6.3.3-9, finally, shows the correlation plot in the Xmax variable for very inclined fluorescent events (θ>60°). With reference to the discussion of Sect. 6.2.3, where we expect those event to be hadronic ones or neutrino-induced, relying on the slant depth at which they develop, the signature of neutrino showers as the 2 detected events with Xmax > 1700 g/cm appears quite clear.

Figure 6.3.3-6 Two horizontal proton initiated Figure 6.3.3-7 Relative error in the Hmax reconstruction showers: at 5 km altitude (dashed line) and 20 for proton initiated. km altitude (solid line).

2 Figure 6.3.3-8 Relative error in the Hmax reconstruction Figure 6.3.3-9 Correlation plot, in the Xmax (g/cm ) for proton initiated showers. reconstruction, for proton initiated showers(θ >60°).

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 31/43

6.4 EUSO as a self-sufficient atmosphere detector As it has been mentioned several times throughout this document, the EUSO observational approach contains several features which turn out into opportunities, as far as the impact of the atmosphere, seen as an active target but also a transmission and attenuation medium for the signal to be observed is concerned. Let us recall them as they have been listed in Sect. 2.6, in order to discuss them in connection with the detector performances expected for EUSO: 5) For most of the observed showers, the shower development takes place in the higher portion (> 3-5 km) of the atmosphere where cloud and aerosol perturbation (Mie scattering) are smallest. This is NOT the case of ground based fluorescence observations of UHECR which are very sensitive to aerosol below the atmosphere boundary layer. 6) The transparency, at these altitudes, is high and does not depend critically on a precise knowledge of the altitude. For example, the Rayleigh transmission factor is of the order of 0.5 and varies by less than 10 % per km (altitude). 7) The presence of an optically thick cloud (a large majority of the clouds present in the lower atmosphere) signals itself by a precise signal: a clear increase (3-10 times) of the reflected Čerenkov signal, embedded in the fluorescence peak, and a strong suppression of the photon yield at lower altitudes (see Fig.6.4.2-1). 8) Most of the effects to be expected from the presence of clouds or aerosol would translate into an underestimation of the UHECR energy and NOT an overestimation. The case of high altitude sub-visible clouds (optical depth ≤ 0.1) will, for example, decrease the atmosphere transmission by ≈ 30 % and will not lead to an over-prediction of the energy. 9) The main physics goal of EUSO is the study of UHECR with energies above 1020 eV where photon statistics are important and the shape of the signal well defined. Lower energy showers will mostly be used for comparison with Auger results. In this case, a statistical correction for cloud and aerosol effects may be sufficient. These features gives EUSO the opportunity of being a self sufficient atmosphere detector, to the extent that it can accomplish its main scientific objectives, as it will be discussed in the forthcoming sections. To complement the information contained hereafter, the following Document is appended, about the possibility of retrieving the data on UHECRs, in absence of LIDAR assistance, by exploiting the reflected Čerenkov light by the cloud tops: “Assessment of the atmospheric monitor for the EUSO Science Specifications”, Appendix A by Y. Takahashi to the Report EUSO-AS-REP-001-1.1, 10 October 2002.

6.4.1 Estimation of the Acceptance of EUSO due to cloud presence The presence of optically thick, low altitude (< 5 km) cloud should lead to visible effects on the fluorescence time profile: a significant increase (with respect to earth-sea) of the Čerenkov peak, embedded in the fluorescence signal and a strong suppression of photon arrival for later times. This effect has to be compared to the natural Poisson fluctuations of the p.e. probability. Figure 6.4.1-1 shows examples of this fluctuation effect for 1020 and 3×1020eV showers. A predicted increase in the efficiency (2 times) of the detector should further diminish the fluctuations. An inspection of the p.e. time spectrum of individual showers should therefore allow to classify them into 3 categories: i. Showers that do not show any presence of cloud effects ii. Showers that show complicated structures revealing a complex cloud structure before and beyond the location of the fluorescence maximum (Smax). These showers would be rejected and the acceptance correspondingly reduced. iii. Showers showing cloud presence at altitudes below Smax. In this case the energy of the shower will be measurable both from the value of p.e. yield at Smax and from the integral of p.e. up to Smax (see below). The inclusion of a cloud atmosphere-sounding device should further confirm these diagnostics. Furthermore, the statistical evaluation of the reduction of the acceptance, due to the cloud presence, can be precisely obtained from existing database (ISCCP, … ). This value will be compared to the value obtained from the above method.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 32/43

Figure 6.4.1-1 Shower time spectrum (θshower = 60°) of photon-electrons with the presence of a cloud at 4 km altitude. Left panel: E = 1020 eV; right panel: E = 3×1020 eV.

6.4.2 Shower altitude estimate and shower energy measurement … yet to be revised …

Because of the particular behaviour of UHECR shower development in the atmosphere, it is well established that the number of charged particles at Smax is proportional to the shower energy:

20 10 Nmax ≈ k × (E/10 eV) with k = 7.0×10

Furthermore, the fluorescence yield per charged particle, depends, for altitudes less than 15 km, only on the track length, namely

E ~ 4-5 photons/m/particle

The combination of these results shows that the number of photons (or p.e.), at the maximum of the shower, will be a measure of the energy:

Fmax= Nmax exp [ λ/(4πR) ]

where λ is the track length seen by a pixel during a GTU and R the altitude of the ISS. λ is dependent on the angle of the shower, the measurement of which is obtained independently from the space–time correlation of the incoming photons (see section XX). Note that because the value of R is large with respect to the average altitude at Smax, the uncertainty in this quantity is very small (<1%). The above considerations are confirmed by simulations. Figure 6.4.2-1 shows the RMS relative value of the reconstructed value of the Smax altitude (labelled hmax) as a function of the shower angle.

hmax(rec) - hmax (exact))/ hmax (exact)

On observes that this uncertainty rarely exceeds 25%, corresponding to ≈ 1 km at the altitude of Smax. This induces an uncertainty in the transmission coefficient of the order of 10%. This uncertainty is even smaller for large shower angles (>50°) which constitute most of the well measured showers (see Figure 2.4-2). In conclusion, the observation of UHECR air showers from space is a well understood and controlled method that contains self correcting features. The impact of cloud presence will be signalled by visible signal whose clarity will increase with shower energy. The reduction of the acceptance due to cloud presence can be estimated by the inspection of the shower signal and compared to statistical evaluations. The measurement of the energy, which is

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 33/43

proportional to the number of p.e. at the maximum of fluorescence yield, can be determined with sufficient precision (≈10%). The addition of an atmosphere sounding device, i.e. an IR camera covering the field of view with a 0.1° resolution (≈200 000 pixels, equivalent to the pixel resolution of the EUSO telescope focal plane) would allow to confirm the presence of clouds as well as their local altitude.

Figure 2.6.2-1 Hmax resolution as a function of the shower angle

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 34/43

6.5 Meteors Observation by EUSO “Meteors” represent for EUSO an observable as a “slow“ phenomenon with seconds to minutes characteristic duration, to be compared to the “fast“ phenomenon typical of the Extensive Air Showers induced by the energetic Cosmic Radiation, ranging from microseconds to milliseconds. The investigation reported regards the evaluation of the light curve in the (300–400) nm UV band produced by the meteoroids and space debris interacting with the Earth atmosphere; the aim is the assessment of the visibility of the meteor phenomenon by EUSO and the estimate of the capability of the Mission for measuring the solid body flux impinging on Earth. The approach considers the deceleration and ablation of a meteoroid as a consequence of frictional forces due to the resistance offered by air molecules to the solid body. The target goal is the characterization of the “meteor UV-light curve”, namely, the variation of its luminosity along the trail. Computations are made for different values of the initial size/mass of the meteoroid, entry velocity and angle of impact entering the atmosphere. The results are presented by using, as a reference scale, units in Magnitudo to facilitate direct comparison with the common literature in the field. The simulations have been conceived on the basis of schematic general processes without any a priori reference to observational data: a calibration with real data will be made by using observation of characterized streams (for more details and general assumptions see EUSO-SIM-REP-008 and EUSO-SDA-REP-013 notes).

6.5.1 The Meteor basic physical processes: modelling The velocity required to a meteoroid travelling in the interplanetary space to enter into the Earth Atmosphere must be at least equal to the escape velocity of 11.2 km/s for an object close to the Earth surface (corresponding to a kinetic energy of about 6x104 joule/g). As the meteoroid is slowed down by friction with atmospheric gas molecules, a complex series of processes of energy conversion occurs as exemplified in Figure 6.5.1-1 [Bronshten, 1983]. The kinetic energy is partially converted into heat which in turn can vaporize (ablation process) and ionize the surface material of the solid body and dissociate and ionize the surrounding atmospheric gas. The electronic transitions resulting from the excitation of atoms involved (both atmospheric and of the meteoroid) produce a luminous source which travels with the meteoroid. Taking into account the formation of shock waves produced in the air ahead of the meteoroid, it is estimated that a fraction about 0.1 to 1 percent of the kinetic energy of the meteoroid is transformed into radiation in the optical-UV range. The “meteor visible flight” lasts few seconds when observed by standard Cameras (limit to 0 Mag), Super-Schmidt Cameras (limit to 3 Mag) or Low Light Level Television (LLLTV) with sensitivity extending to a Magnitudo about 9 [Ceplecha et al., 1998]; the missing “dark flight” of the meteoroid can be substantially longer in duration and may be traced by EUSO observing the UV emission at a much higher level of sensitivity monitoring the phenomenon from space (limiting Magnitudo up to ≈18 when normalized to observation from ground in clear atmospheric conditions). The basic assumption for an “EUSO oriented” numerical simulation of the meteor light curve can be summarized in the following steps: • Adoption of a numerical model for the Earth Atmosphere extending to the height limit which encloses the section of physical interest to the meteor phenomenon, i.e. up to about 900 km altitude above the sea level. • Assumption for the geometry of the meteoroid/debris of a sphere entering the Atmosphere and progressively slowed down by the drag due to the friction with the surrounding air molecules; the trajectory and speed of the solid body is computed as a function of time. • No fragmentation is considered. • A model of a simple ablation process is adopted, which takes into account the progressive mass reduction of the body during its flight through the Atmosphere as a result of the outer surface vaporization resulting by the energy dissipation by heat production. • The simulation considers the sphere as a moving source of radiation and computes the UV light curve as a function of time/trajectory converting and balancing the loss of kinetic energy into luminosity.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 35/43

Figure 6.5.1-1 Energy conversion in meteor phenomena

6.5.2 The Earth Atmosphere A profile of the Earth Atmosphere is required for the meteor simulation to provide a comprehensive set of values of temperature, pressure, density, absolute and cinematic viscosity, sound speed and molecular weight of the air as a function of altitude. For the Homosphere (i.e. the Earth Atmosphere below 105 km altitude, where complete vertical mixing yields a near-homogeneous composition of about 78.1% N2, 20.9% O2, 0.9%Ar and 0.1% CO2 and trace constituents) the US Standard Atmosphere 1976 Model has been adopted as reference [USSA-76]. The Model gives the temperature as a function of altitude calculated for piecewise linear temperature profile segments. Air density and pressure have been then calculated by assuming the homosphere as a perfect gas of constant molecular weight M = 28.964 in hydrostatic equilibrium conditions function of altitude. For the Heterosphere, from about 105 km of altitude up to the Exosphere, (region where it can be assumed that the constituents are in diffusive equilibrium with no vertical mixing and the concentration profiles develop for each constituent independently under the influence of gravity and thermal diffusion), the Mass Spectrometer Incoherent Scatter MSISE-90 Atmosphere Model output data (as obtained for mean solar activity (F10.7=(F10.7)avg=140) and daily mean geomagnetic index Ap=300) have been adopted as reference [Hedin, 1991]. MSISE-90 determines temperature, density and number concentration of the major constituents up to exospheric altitudes as a function of the atmospheric state parameters (i.e. geodetic latitude, local solar time, universal time/geographic longitude, day of the year, daily solar flux index, mean solar flux index, daily mean geomagnetic index). The upper limit of the heterosphere (i.e. max Earth Atmosphere altitude) has been set at 900 km above the Earth surface. The MSISE-90 output data set has been divided in different altitude regions and data points fitted in each interval with high order polynomial in order to obtain a continuous function for air temperature, pressure, density and molecular weight parameters as a function of altitude. For aerodynamic calculation in the rarefied atmosphere the mean free path length, the speed of sound and the dynamic viscosity have been approximated using the following relations:

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 36/43

γ ⋅ Tp22 L= ; a= k⋅ ; µ= ⋅ L ⋅⋅⋅ρ a; ⋅⋅⋅2 ⋅⋅ p2π davg γρ 3 πγ

L = Mean Free Path of a molecule (m); -1 a = Speed of Sound (m⋅ s ); -1 -1 µ = Dynamic Viscosity (kg⋅⋅ s m ); γ = Ratio of Specific Heats (= 1.44, in an Nitrogen dominated environment); ⋅ -10 davg = Mean Collision Diameter for Nitrogen (= 3.62 10 m); -23o -1 k = Boltzmann Constant (= 1.3807⋅⋅ 10 JK ); o T = Temperature ( K); -2 p = Pressure (N⋅ m ).

6.5.3 Motion of a spherical massive body in the Atmosphere The simulation program adopted calculates position and velocity of the meteoroid in the atmosphere as a function of time in a bi-dimensional space, by using a Cartesian reference system with origin set at the Earth surface, in correspondence of the point of impact of the object in the atmosphere assumed at a given altitude (up to 900 km) and with Y-axis directed towards Nadir; the meteoroid angle of entry is referred to the horizontal, i.e. to a direction tangent to the atmospheric surface at the point of impact of the object into the atmosphere (see Figure 6.5.3-1). The deceleration equation is based on the assumption that the momentum m·dv lost by a meteoroid is proportional to the momentum of the oncoming air flow; the mass impinging upon a mid-sectional area S at velocity v in time dt is S·ρ·v·dt. We obtain the equation: dv 2 m=-⋅⋅⋅⋅Γ S ρ v dt where Γ is the drag coefficient, expressing the portion of the momentum of the oncoming flow that is converted into deceleration of the body. In aerodynamics, the drag coefficient is usually denoted by cx or CD, where cx = 2Γ. The solid body is represented by a geometrical sphere (shape factor A = 1.21), with perfectly smooth surface and of homogeneous mass (i.e. no cracks or fracture weaken the solid body structure); chemical and mineralogical composition of the meteoroid are not taken into account and a constant density is assumed. To simplify the model neither fragmentation processes due to stress by high speed impact with air molecules are considered nor the deformation of the meteoroid during the flight in the atmosphere is taken into account and the body is considered self-similar during the meteor phenomenon (shape variation parameter µ constant and equal to ⅔). As far as the motion is concerned, the following constrains were adopted in the simulation of the atmospheric entry: • The meteoroid is travelling in a uniform gravitational field. • No correction is introduced for Earth Surface curvature, considered flat when comparing the Earth radius with the atmospheric thickness, nor for atmosphere curvature at low entry angle values. • The spherical meteoroid is supposed to be a non-rotating object; such an assumption makes it possible to avoid a continuous correction in the direction of flight due to the Magnus effect (deviation in the direction of flight due to rotation on one of its axis). • For the aerodynamic force, no lift component is considered in the calculation (lift coefficient = 0.0). • Shock wave formation, as a first approximation, is not taken into account. Therefore a punctual determination of the local atmospheric parameters (air pressure, density and temperature) cannot be made during the meteor appearance; on the other hand, being mostly focused on the ablation phenomena and the emission of UV radiation, we can globally take into account shock wave effects by setting proper values for the heat transfer and the luminosity coefficients, which give an estimate on the sharing of the kinetic energy between thermal and luminous component. • For aerodynamic calculations, air molecules flow regime definition/transition during the meteoroid flight in the atmosphere is taken into account by a simple appropriate calculation of the drag coefficient (Cd). It is assumed a strong Reynold’s number (Re) dependence (i.e. air viscosity and density) of Cd at lower meteoroid velocity, while a Mach number (M) dependence is advocated for higher speed. Different

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 37/43

experimental datasets are used based on the different range of object velocity/fluid viscosity ratios. A high order polynomial fitting procedure is then used to obtain a continuous function Cd=f(Re,M) and a numerical computation of Cd is made. This approximation avoids the complication arising from to the fact that the transitional flow regime is not well understood and it is therefore difficult to cover the conditions from free-molecular flow to continuum flow (subdivided in incompressible, compressible transonic supersonic and hypersonic flow based on the Mach number) in a process anyway dominated by a large number of parameters.

Figure 6.5.3-1 The reference system adopted in meteor simulation

6.5.4 The Ablation process The mass-loss equation follows from the assumption that in the reference frame of the solid body a certain portion ∆ of the rate of energy loss ½Sρv3 of the oncoming stream of molecules is expended on the ablation (vaporization or fusion and spraying) of mass dm in an interval time dt. If ζ is the latent heat of vaporization or fusion of the meteoroid material in units of energy (including the energy that must be delivered to a mass dm in order to heat it from its initial temperature T0 to its evaporation or melting temperature), then the mass-loss equation takes the form 3 dm S⋅⋅ρ v =-∆ ⋅ dt 2⋅ζ Some assumptions are made in order to simplify the ablation process modelling: • The percentage of the total meteoroid kinetic energy (which transformation is distributed according to a complex function between heating/ablation, radiation, ionization, fragmentation, shock waves etc...) converted into thermal energy is considered constant during the meteoroid flight through the atmosphere. Namely, the kinetic energy converted into heat is assumed for each simulation run not to be a function of the meteoroid speed and mass and the heat transfer coefficient is maintained constant over the entire calculation. • In the ablation model considered, the heat generated by friction is initially deposited on the external surface of the solid and propagates uniformly over the entire surface independently from the direction of flight. The

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 38/43

heat is supposed to conduct inward the body with low efficiency (low thermal conductivity of the meteoroid) so that only the outer husk melts. The resulting vaporized material is then removed by effect of the air molecules flowing around the solid, exposing newer surface to the ablation process with a progressive reduction of the particle mass and size. • No change in the mineralogical composition of the meteoroid is assumed as effect of the large amount of heat involved in the ablation process so that the density remain unchanged in the meteor phenomenon.

6.5.5 Meteors as source of UV radiation Starting point for the calculations is the assumption that a fraction of the meteoroid kinetic energy is converted into electromagnetic radiation (luminous flux), according to the following relation [Bronshten 1983]: dm v2 L = τ⋅⋅- dt 2 where: L is the optical luminosity (i.e. luminous energy); τ is the dimensionless coefficient of radiative efficiency (luminosity coefficient); dm/dt is the instantaneous ablation rate. The coefficient τ represents the part of the kinetic energy of the vaporizing meteoroid material which is converted into radiation. In order to simplify the process, according to Sect. 6.5.4, we have assumed τ not to be a function of the speed and mass of the meteoroid and it is kept constant over each simulation run; this is equivalent to say that the fraction of the total meteoroid kinetic energy converted into luminous energy is constant during the meteoroid flight in the atmosphere. We have carried out numerical simulations for the meteor magnitudo M; implementation for observations from space platforms is in progress. The following steps have been considered in the calculation of the UV radiation parameters:

• The program calculates first the instantaneous optical luminosity (L), then the optical luminous flux (F) as seen from an observer at ground in correspondence of the origin of the reference axes, and finally derives the value of instantaneous meteor visual absolute magnitude Mvisual by using the expression: ⋅ F Mvisual = - 2.5 log10  F0 -9 where F0 = 3.38 x 10 according to the calibration of Allen [Allen, 1973; see also Hills & Goda, 1993].

• To take into account the wavelength dependence of the meteors light flux, assumption on a particular meteor spectrum has to be made by setting a proper colour index (C.I.) parameter and, assuming Mvisual equal to Moptical, the absolute UV magnitude (Muv) in the (300-400) nm range can be calculated by using the relation:

MM(300-400)nm= optical - C.I.

where M(300-400nm) is the UV absolute magnitude, Moptical is the optical absolute magnitude and C.I. is the colour index. By adopting the C.I. “correction” parameter, even if in a very simplified way, the program can simulate emission of radiation from meteors with different composition, overcoming the fact that mineralogical/chemical meteoroid composition has not been taken into account for ablation.

-26 2 • In order to calculate the equivalent flux density Sν in Jansky (Jy) (1 Jansky=10 watts/m /hz) from the

values of magnitudes at the effective wavelength λeff, the following relation has been used [Allen, 1973]:

(-0.4 × Mυ ) Sυυ = A× 10

where Sν, and Mν are the flux density (in Jansky (Jy)) and the meteor magnitude while Aν is a constant. For meteor simulation values adopted are Auv=1910 and Aoptical=3830, corresponding to the U and V filter respectively.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 39/43

6.5.6 Meteor simulation The combined effect of forces of drag and gravity (with the assumption of absence of lift, as stated above) in the presence of ablation results in a set of differential equations that fully describe the trajectory of a meteoroid entering the Earth atmosphere. The equations are: dv ρ⋅⋅Av2 dm ∆⋅ρ ⋅Av ⋅ 3 =-C ⋅ and = - dtD m dt 2⋅ζ where: t is the time; m, v, and A are, respectively, the meteoroid mass, velocity and cross-sectional area; ρ is the air density; CD is the drag coefficient; ζ is the heat of ablation of the meteoroid material; ∆ is the heat transfer coefficient. The numerical method adopted in the simulation is the “explicit forward finite difference approximation to the differential equations”. For input parameters the following numerical values have been assumed: • Initial meteoroid mass. Meteoroid mass ranges covers several order of magnitudes, from 10-21 kg for space-dust particles to micrometeorites (≈10-9 kg) and up to 1015 kg in the case of very large object [Ceplecha, 1992]. • Meteoroid density. The density range of meteoroid has been determined by studying the objects which, during the flight in the atmosphere, partially survived to the ablation and were collected as meteorites at the Earth surface. In a broad sense, meteorites can be generally classified as stone-meteorite, with reference density ρ=3.4, and iron-meteorite, with density reference ρ=7.8. [Grimmer, 1948]. • Entry velocity. The meteoroid entry velocity is determined by the combination of the solar escape velocity of an object at the Earth’s orbit (42 km/s) and the orbital speed of the Earth (30 km/s). The meteoroid entry velocity is therefore set in the (12–72) km/s range. • Entry angle. The range is between 90o (impact perpendicular to the atmospheric surface) and 0o (impact tangent to the atmospheric surface). • Entry altitude. The impact of the meteoroid with the atmosphere has been set at 900 km above the Earth surface, being this altitude considered the limit above which the atmosphere density can be considered negligible in a friction process. • Heat of ablation. The heat of ablation is a function of the material type and the specific process of ablation. An average value of 5x106 J/kg is commonly adopted [Passey & Melosh, 1980; Baldwin & Shaeffer, 1971]. • Heat transfer coefficient. According to literature a typical value of 0.01 is adopted [O’Keefe & Ahrens, 1982; Allen et al., 1963; Liu, 1978]. −− • Luminosity coefficient. According to existing determinations: 310⋅≤≤⋅42τ 210 [Bronshten , 1983]: • Color Index. To be chosen according to the spectral emission of meteors. By default Vega has been assumed as reference (C.I. =0.0). • EUSO UV threshold. To be set according to EUSO technical specification: a numerical value of 18 has been adopted at present [B. Sacco, private communication]. • Time calculation step. As stated above, the accuracy of the explicit forward finite difference method is improved by adopting smaller time increments. A control is made during the run so that the time step needs to be reduced if the meteoroid change in mass or in velocity/position is greater than a predetermined value (e.g.10%) during each interval. Figure 6.5.6-1 shows, as an example, typical results obtained with the simulations with respect to observables like the magnitudo track profile, the track length and duration as a function of the threshold imposed on the magnitudo and the penetration depth in the atmosphere. Meteoroid mass, velocity and angle of entry are the parameters under investigation and are parameterised for the different run sets while all the other parameters (CD, ζ, ∆, τ, C.I. …) are kept unchanged in all the simulations.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 40/43

Figure 6.5.6-1 Typical Light Curve for meteors (Magnitudo vs Trajectory Parameters) and Mass variation vs time and altitude in the atmosphere (Entry Velocity = 12 km/s; Entry Angle =30°; Entry Mass = 0.001 kg).

(v=72 km/s) (theta=30o) (Mass =1 kg) (v=12 km/s) (theta=30o) (Mass =1 kg)

UVthreshold (M) UV Duration (s) UVthreshold (M) UV Duration (s) 18.0 20 18.0 120 5.0 20 5.0 44 0.0 13 0.0 19 -5.0 6 -5.0 4 -7.0 4 -6.0 0 -9.0 2 -10 0 (v=72 km/s) (theta=30o) (Mass =0.001 kg) (v=12 km/s) (theta=30o) (Mass =0.001 kg)

UVthreshold (M) UV Duration (s) UVthreshold (M) UV Duration (s) 18.0 17 18.0 104 5.0 10 5.0 14 0.0 3 3.0 5 -1.0 0 2.0 0 Table 6.5.6-1 Duration UV emission above threshold for various values of entry mass, angle and velocity

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 41/43

Figure 6.5.6-2 Simulated duration of meteors as a function of mass for different values of entry velocity and angle

(v=12 km/s) (theta=30o) (UV Threshold =18)

Mass (kg) UVmax (M) UVmax Altitude (km) UV End Altitude (km) UV Duration (s) 0.001 2.4 155 136 104 0.010 -0.8 135 122 118 0.100 -3.9 122 113 119 1.000 -7.0 114 107 120 (v=72 km/s) (theta=30o) (UV Threshold =18)

Mass (kg) UVmax (M) UVmax Altitude (km) UV End Altitude (km) UV Duration (s) 0.001 -0.9 314 269 17 0.010 -3.8 273 233 18 0.100 -6.7 236 200 19 1.000 -9.7 200 172 20 (v=12 km/s) (theta=60o) (UV Threshold =18)

Mass (kg) UVmax (M) UVmax Altitude (km) UV End Altitude (km) UV Duration (s) 0.001 0.3 143 127 72 0.010 -2.8 127 117 73 0.100 -6.0 117 109 74 1.000 -9.0 110 103 74 (v=72 km/s) (theta=60o) (UV Threshold =18)

Mass (kg) UVmax (M) UVmax Altitude (km) UV End Altitude (km) UV Duration (s) 0.001 -3.0 285 243 11 0.010 -5.9 246 209 11 0.100 -8.7 200 180 12 1.000 -11.8 179 155 12

Table 6.5.6-2 Mass dependence of the UVMax, UVMax altitude, UV emission duration at various entry angles and velocities of the meteoroids

6.5.7 The case of micrometeorites According to Whipple, a “micrometeorite” is defined as an extraterrestrial body that is sufficiently small to enter the Earth's atmosphere without being destroyed by the encounter with the atmosphere [Whipple, 1950]. More

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 42/43

specifically, for meteoroids with mass of the order of 10-9 kg, the surface/mass ratio of the solid body consents an instant surface heat dissipation so that, as its motion is retarded by atmospheric resistance, the micrometeorite radiates energy rapidly enough that its temperature remains below its melting point and therefore ablation process does not occur. The maximum temperature Tm to which the micrometeorite can be heated without appreciable vaporization is just below the melting point of the least refractory material in the meteoroid, approximately 1200 oK to 1700 oK for typical stones. Iron, iron oxides and silica also fall within this range. =− ⋅ () -9 2 Table 6.5.7-1 gives the calculated Magnitudo ( Mvisual 2.5 log10FF / 0 where F0 = 3.38 x 10 J/s/m according to the calibration of Allen [Allen, 1973] for a micrometeorite of mass= 10-9 kg (density ρ=3.5 g/cm3; Temperature =1700 oK; emissivity ε=0.5) at different ground distance.

Ground Distance (km) Optical luminosity Optical Flux (F) Optical Magnitudo Mvisual 50 0.05 1.58 E-12 8.3 100 0.05 3.95 E-13 9.8 200 0.05 9.88 E-14 11.3 300 0.05 4.39 E-14 12.2 Table 6.5.7-1 Calculated magnitudo for micrometeorites

Micrometeorites cannot be observed as “meteors” by optical (DV, cameras or LLLTV) methods, since they do not produce a significant luminosity above the sensitivity threshold. It may be assumed that the upper end of their mass range corresponds approximately to the lowest mass of a particle detectable by the most sensitive radars. “Micrometeoroids” are recorded mainly by detectors on spacecraft, meteoric-particle collision on rockets and balloons and collections on sea-beds and glacial deposits. As yet none of these methods enables the determination of micrometeoroids orbits; their mass distribution can be determined fairly reliably but in calculations of micrometeoroids streams the different methods show an error spread over four order of magnitude. EUSO, with its high sensitivity (Mag up to ~18) and signal/noise ratio is expected to be able to provide a significant direct “micrometeor” observation.

6.5.8 Data retrieval and approach to the analysis Meteor observational sensitivity for EUSO and other common standard technique are shown in Figure 6.5.8-1.

Figure 6.5.8-1 Meteor observational sensitivity for EUSO

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 43/43

Figure 6.5.8-2 The path of a meteor in the EUSO field-of-view

EUSO will observe the portion of the meteor below the altitude of 400 km (see Figure 6.5.8-2). The trajectory of a solid body traveling in the atmosphere and intersecting the cone of observation of EUSO will determine a UV profile projected as a “track” on the instrument focal surface. The observation is photometric (no spectroscopy) with high frequency sampling in time and position. The data retrieved can be exploited in the context of the so called “Integral Light” approach of investigation. They will be the result of meteor UV emission whatever the physical process responsible. The observation of the meteor “integral UV-light”, besides tracking the meteoroid trajectory and dynamics will allow reconstructing the solid body initial entry mass and its mass variation along the trajectory. The focal surface track is a complex function of several parameters, both instrumental and solid body dependent. Figure 6.5.8-3 shows an example of a meteor track as seen on the focal surface. The vector velocity of the meteoroid can be inferred by the shape of the track by subtracting the constant ISS vector velocity (~7 km/s). Track progression velocity, the geometry and intensity of the UV-signal will provide a powerful tool for a detailed reconstruction of the meteor UV light curve.

Figure 6.5.8-3 EUSO meteor observation. (A) - Different frames are taken at constant time steps ∆t and are represented by sections of the “cone” with different colors (first frame is pink, last is green). The sections move along the EUSO travel direction and are equally spaced because of the constant velocity (~7 km/s). The sections increase their diameter with time because they refer to progressively lower “cone”-meteor intersection altitude. Meteors velocities at each time are represented by colored vectors whose module decrease with time (altitude). The meteor “spots” moves with curved trajectories on the “observation cone” projection shown in the (B) panel.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/11

8. MISSION OVERVIEW...... 2 8.1 ORBITAL PARAMETERS...... 2 8.1.1 Operational and Life Limits ...... 2 8.2 INSTRUMENT TO PAYLOAD INTERFACES...... 3 8.2.1 Interface planes definition ...... 4 8.2.2 Physical interface at launch ...... 5 8.2.3 On orbit physical interface...... 6 8.2.4 Electrical interface ...... 7 8.2.5 Command and Data Interface ...... 8 8.2.5.1 Data rate capability...... 8 8.3 MISSION REQUIREMENTS ...... 8 8.3.1 Mission success definition...... 8 8.3.2 Mission/Experiment Reliability ...... 8 8.3.3 Instrument Duty Cycle...... 8 8.4 MISSION OPERATIONS...... 8 8.4.1 Operations ...... 8 8.4.1.1 Ground Operations ...... 8 8.4.1.2 On-Orbit Operations...... 8 8.4.1.2.1 EUSO Instrument On orbit Installation (P/L operation, For Info Only)...... 8 8.4.1.2.2 Operative Modes ...... 10 8.4.1.2.3 Non-Operative Modes...... 10 8.4.1.2.4 Contingency Modes ...... 11 8.5 COMMISSIONING...... 11

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/11

8. Mission Overview The International Space Station (ISS) will provide the opportunity to accommodate the EUSO Instrument on an observing location outside the Columbus module. The EUSO Instrument shall be transported and operated in orbit by means of the NSTS (Shuttle carrier) and the EUSO Payload. Figure 8-1 shows the location of the Instrument in the frame of the International Space Station.

EUSO Instrument

Figure 8-1 EUSO Instrument and the International Space Station

8.1 Orbital Parameters The EUSO Instrument design shall be compatible with the nominal orbit of the ISS:

Altitude : 351 ÷ 460 km Inclination : 51.6 degrees Period : 90 minutes

8.1.1 Operational and Life Limits The EUSO Instrument shall be designed for three years on-orbit nominal operations, plus two years of extended mission, plus two years of storage on ground, based on a system reliability figure higher than 0.95. System malfunctions recovery with consequent Instrument Hardware reconfiguration and/or flight software updates shall be managed by Ground segment as contingency operations. The EUSO Instrument shall nominally operate during all its life cycle except for the following cases: • Contingency operations and relevant troubleshooting • Boarded Science Instruments contingency operations and relevant troubleshooting (if not specifically required to support the Instrument investigation) • CEPF contingency operations and relevant troubleshooting

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/11

• ISS/Columbus contingency operations • ISS orbital re-boost • STS or other spacecraft docking • ISS/Columbus power nominal switch-off • ISS EVA activities (as necessary) • ISS Remote Manipulator operations (as necessary)

8.2 Instrument to Payload Interfaces Figure 8.2-1 shows the various parts of the Instrument and Payload.

Figure 8.2-1 EUSO Instrument and EUSO Payload (exploded view) components

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/11

8.2.1 Interface planes definition Figure 8.2.1-1 shows the ideal interface planes between the Instrument and the P/L. At launch two P/L elements shall be directly attached to the Instrument, that are the Power Data Grapple Fixture (interface for on orbit handling) and the Passive MCAS (interface to Columbus adapters, as per Figure 8.2.3-1).

Figure 8.2.1-1 EUSO Instrument interface planes scheme

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/11

8.2.2 Physical interface at launch The physical interfaces between Instrument and P/L in launch configuration are shown in Figure 8.2.2-1.

Passive MCAS PDGF (P/L part) (P/L part)

INSTRUMENT

Cradle (P/L part) Instrument trunnions

Active MCAS (P/L)+ Adapter struct (P/L)+ Active FRAM (P/L) (attached to Cradle)

Figure 8.2.2-1 Physical interfaces in launch configuration

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/11

8.2.3 On orbit physical interface Figure 8.2.3-1 shows the on orbit interface adapters for the Instrument. In particular, the Passive MCAS shall mate its Active MCAS counterpart.

Figure 8.2.3-1 On orbit physical interface

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/11

8.2.4 Electrical interface Figure 8.2.4-1 shows the electrical scheme interface between EUSO Instrument and EUSO P/L. The Instrument will include a dedicated power converter to transform the 120 V supply from the P/L to appropriate voltages for the Instrumentation. All data I/O will be managed over an ISS 1553 bus standard. A large mass memory will hold the telemetry data (scientific, housekeeping) to be downloaded to the ground station.

EUSO EUSO INSTRUMENT P/L

Figure 8.2.4-1 Electrical interface between Instrument and P/L

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 8/11

8.2.5 Command and Data Interface The EUSO Instrument shall be supported, when the EUSO P/L is installed on CEPF through the Adapter/FRAM, with the following data links and discrete signal I/Fs: • One redundant IEE 802.3-10BaseT link • One redundant MIL-STD-1553 bus • One contact status monitor (TBC) • Three (TBC) Active Driver monitor • Two (TBC) Analogue monitor • Two (TBC) 4-20 mA Current Loop monitor • Two (TBC) 5V Pulse cmd • One (TBC) 28V Pulse cmd.

8.2.5.1 Data rate capability

The EUSO Instrument maximum data rate shall not exceed 100 kbps.

8.3 Mission Requirements

8.3.1 Mission success definition After three years plus two years extended mission: at least 0.8 of the channels fully working.

8.3.2 Mission/Experiment Reliability Mission success probability: ≥ 0.95

8.3.3 Instrument Duty Cycle

The Instrument duty cycle (the fraction of time the Instrument is ON and ready to take data), shall be: ηI = ηO + 0.05

8.4 Mission Operations

8.4.1 Operations

8.4.1.1 Ground Operations No special operations are envisaged, apart from those relevant to optical calibration activities.

8.4.1.2 On-Orbit Operations

8.4.1.2.1 EUSO Instrument On orbit Installation (P/L operation, For Info Only) During this phase the EUSO Instrument is extracted from the Cargo Bay and then installed and activated on Columbus EPF, following these basic steps: • The P/L adapters assembly (Active MCAS+Adapt. Struct+Active FRAM, Figure 8.2.2-1) is robotically taken from the P/L cradle and mounted the CEPF as per Figure 8.2.3-1. • The Instrument is unlocked from the cradle and robotically installed on the adapter assy

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 9/11

The sequence shown in Figure 8.4.1.2.1-1 depicts the complete EUSO deployment from the Cargo Bay to the final location on Columbus nadir EPF (Figure 8.4.1.2.1-2).

Figure 8.4.1.2.1-1 EUSO Instrument On Orbit Installation

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 10/11

Figure 8.4.1.2.1-2 EUSO Instrument on CEPF

8.4.1.2.2 Operative Modes Operations with both the EHECR/ν Telescope and the AS Device have to be made continuously, except for cases of paragraphs 8.4.1.2.3 and 8.4.1.2.4.

8.4.1.2.3 Non-Operative Modes Non-operative modes occurs in Sunlit periods and non-quiescent periods. Sunlit periods, occurring every orbit. These periods are defined as any when reflected or scattered solar radiation may enter the EUSO acceptance angle. In such a condition, a lid must close the EHECR/ν Telescope entrance. A suitable set of light-sensors (TBD) oriented at an angle somewhat larger than the Instrument half-FoV

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 11/11

will sense the increasing and decreasing change in light intensity caused by the day/night orbit cycle allowing to trigger a command to the mechanism of the lid before/after the terminator line. The time of occurrences of the terminators will be predictable, so that timed commands may also be loaded in advance, with the light sensor providing a back-up of closure/aperture command, if required. Non-quiescent periods, during docking/undocking of any vehicle, periods when the Shuttle Orbiter is docked, and re-boost operations. Under such conditions, also to avoid the increases in the contamination environment, the lid must close the EHECR/ν Telescope optics entrance, and the AS Device must be protected against possible contamination.

8.4.1.2.4 Contingency Modes Periods of reduced power availability or contingency transfers on other ISS locations (TBC). In case of a general shortage of electrical power and contingency episodes, both the EHECR/ν Telescope and the AS Device cannot be operated at the required power levels. These periods are not always predictable, and their duration is also unknown. Operations concept

8.5 Commissioning

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/7

9. EUSO OPERATIONS ...... 2 9.1 INSTRUMENT OPERATIONS ...... 2 9.1.1 System Functionalities ...... 2 9.1.2 System Modes of Operation ...... 3 9.1.3 Calibration...... 4 9.1.4 Science Operation Modes ...... 4 9.1.5 Science Ground Segment...... 5 9.1.6 Scientific Data Products ...... 7 9.2 INTERFACES WITH THE COMMUNITY...... 7 9.3 MANAGEMENT OF OPERATIONS...... 7

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/7

9. EUSO Operations EUSO operations will ensure that the scientific objectives of the mission can be achieved, optimising the use of all the available resources. The operations system must keep the correct functioning and communication of the apparatus, and make sure that the mission data is safely collected, processed and stored on ground. The EUSO operations system is the integrated system of hardware, software, people and procedures that must cooperate to accomplish these objectives. The operations system involves both the flight and ground segments of the mission. The communication between them plays an important role. On-board, scientific data and housekeeping parameters are collected and the corresponding telemetry is sent to ground. The control and monitoring of the different instrument subsystems is performed, based on on-board autonomous procedures and on telecommands sent from ground, making sure EUSO is always in a safe state. On ground, telemetry is received, processed, monitored and archived. Telecommands are prepared for uplink, according to a defined mission activity planning. The mission operations phase will bring to three years the in orbit nominal operations period, plus two years of extended mission.

9.1 Instrument Operations The definition of the level of autonomy of the instrument operations is crucial for the dimensioning of the EUSO flight system, in particular of the Trigger Control Unit . It also impacts on the definition of the ground system, since a larger in-flight autonomy implies more care with ground based test and validation procedures, while a lower level of autonomy means heavier communications and mission planning activities. In the case of EUSO, basic constraints (no permanent contact is provided and the uplink/downlink rates are limited) impose a relatively large autonomy.

9.1.1 System Functionalities The basic functions of the instrument operations system can be summarised as: • Collection and pre-processing of scientific data;

• Collection and verification of Housekeeping data;

• Preparation, storage and transmission of Telemetry packets;

• Telecommands handling;

• Control of the PhotoDetector Module trigger and readout configuration;

• Management of the power distribution;

• Management of the Thermal Control System;

• Management of the lid;

• Management of the calibration devices;

• Management of contingency situations;

• Interface to the Atmosphere Sounding device. On board, the different subsystems are controlled and monitored by an Instrument Control and Readout system (ICR), which handles science and housekeeping data collection and controls the different instrument subsystems. It also provides the interface to the atmospheric sounding system. Relevant items for monitoring in the instrument are: temperatures, power status, luminosity level, lid status, magnetic field, pressure, time, data handling status, AS system status. The ICR contains the on-board SW, which manages the different ongoing processes and the command sequences received from ground. Different configuration files are available on-board and loaded in

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/7

agreement to the selected operation modes and conditions. The ICR functionalities will be implemented by the TCU.

9.1.2 System Modes of Operation A preliminary view of the different operational states of the instrument is shown in Figure 9.1.2-1, together with the possible transitions between them and the telemetry packets (housekeeping, HK, and science, SC) generated in each of them. The transitions between modes are controlled by the on-board SW and can be triggered by time- tagged commands contained in the schedule files up-linked from ground, by the receptions of ISS ephemeris parameters and by the on-board subsystem monitoring processes. Mode changes caused by exception states may be triggered by both the ISS and the on-board monitoring system.

Figure 9.1.2-1 Schematic view of the different EUSO operation states.

The non-observation system modes are: Power off, Initialisation, Standby, Setting, Contingency, Engineering and Technical calibration. In the Power off mode the whole instrument is powered off, except for the stay-alive power. When power is applied to the TCU the instrument goes autonomously to the Initialisation mode. In the Initialisation mode communications are initialised, the TCU and the on-board SW are initialised and verified. The system is ready to receive telecommands and produces the HK packets related to communications and to the status of the TCU. On the reception of a Start telecommand from ground the system is powered on step by step and goes to Standby mode. The full set of instrument periodic HK parameters is generated and down-linked. The full data handling SW is initialised. Information dump from a given subsystem (patch/dump capabilities) may be performed at this stage. At the execution of a Get ready command (up-linked from ground), the high voltage to the PDMs is switched on, while the lid is kept closed. The system is in Engineering mode and the High Voltage is enabled. Configuration table and SW updates may be performed in Setting mode, accessible both from the Standby and Engineering modes. Technical calibration may be initiated on the execution of a specific telecommand. In this case specific data packets will be generated. All the transitions mentioned above are ground- commanded. In case of exceptional events the system will automatically reconfigure itself into Contingency mode. As a baseline, ground commanding is required for recovery. There must be a “watchdog” to reboot the computer if needed, and the SW and parameter files should be stored in a memory that is supplied by the stay-alive line. The scientific system modes are labelled Observation modes in Figure 9.1.2-1. The system can autonomously go from the Engineering mode to EECR Observation mode if a day/night transition is signalled (by ISS time or orbit

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/7

information) and from Observation to Engineering mode in the reverse situation. The HV will be in standby during orbital day. For each scientific operation mode there are different configuration files that can be loaded. They are selected according to TC parameters or autonomously. On the scientific modes, the instrument generates the full set of HK telemetry plus scientific packets, which format and contents depend on the specific mode. In the current baseline, within each operational state, the on-board system performs all routine operations in an autonomous way: scientific data collection pre-processing and storage, HK data collection and monitoring of the instrument configuration and status, lid actuation, management of the power and thermal control systems. Summarizing, the way in which transition between operational states are commanded largely defines the level of autonomy of the system. All routine operations (i.e. operations taking place in any standard orbit) will be autonomous, regulated by the on-board system. Up-linked command sequence files will contain only “non-routine” operations: calibrations, slow mode operation, specific changes in configuration files.

9.1.3 Calibration Instrument calibration is crucial for the achievement of the scientific goals of the mission. Calibration activities scheduling might represent an important part of the mission planning activities and calibration data might represent a sizeable fraction of the telemetry. Calibration activities are of two very different types: technical calibration, which monitor the health, status and performance of the instrument components, and scientific calibration, which will mimic the response of the system to scientific events and allow to calibrate the physical quantities determination procedure. Technical calibration activities, described in section 5.4.2, are performed with the lid closed and use dedicated on-board equipment. Specific trigger and readout configurations are foreseen. Technical calibration activities include several different calibration procedures. They monitor the health, status and performance of the optical system, of the photo-detectors throughout the entire FS and of the AS instrumentation. The required absolute calibration operations will be first performed on ground and a calibration matrix will be available. Calibration runs on-orbit will allow the monitoring of the health, status and performance variation of the different components. Scheduling of the calibration activities will be performed from ground. According to ground (time-tagged) commanding, the on-board system will go into technical calibration mode and activate the instrument calibration system. Technical calibration will be normally scheduled during day-time periods, when no scientific observation with EUSO is possible. Since power saving configurations will be adopted during day-time, calibration activities will take place either in the beginning or at the end of the data-taking period, when the Instrument is still/already operational, and have a foreseen duration of the order of a few minutes. Calibration data volumes are a concern, especially due to the high segmentation of the EUSO FS. The calibration of the FS not all at once but per PDM or PDM group in different orbits is a favourable solution from the point of view of the use of the available resources. The complete focal surface calibration will be performed roughly once every two months. Calibration results may lead to updates of the system configuration files (e.g. pixel masking). Scientific calibration activities, described in section 5.4.4, are performed with the instrument in normal (UHECR) operation conditions: lid open, UHECR trigger and readout configuration (TBC). Ground light sources (flasher or LIDAR) will be used. On average, EUSO fly over one of the flasher GLS sites per night, for typically 20- 60 seconds. A similar pass rate is envisaged for LIDAR GLS. The EUSO passes need to be scheduled (from the ISS ephemeris). Scientific calibration activities will be planned on ground and included in the up-linked activity schedule files. A configuration file for the GLS should exist on-board and be loaded/executed automatically when EUSO is over a GLS, provided a scientific calibration is scheduled in the plan up-linked from ground. Atmospheric transmission monitoring will be required during GLS calibration. As GLS operations must be planned with the overall EUSO operations, the EUSO observation schedules will be made available to the GLS operation teams. Conversely, HK data from the GLS will be made available to the Scientific Operations and Data Centre (SODC, see next sections) for integration with the remaining EUSO data.

9.1.4 Science Operation Modes EUSO will operate in different modes, collecting several types of scientific data, in order to accomplish the scientific mission goals: • EECR mode – observation with standard (UHECR) trigger conditions;

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/7

• Slow mode – observation mode suited for events of the order of seconds (e.g. meteors);

• Calibration – both scientific and technical instrument calibration (see previous section). Since EUSO can only operate at night, routine operations are firstly determined by day/night cycles. As a baseline, the system will be in UHECR observation mode whenever possible. Switching to slow mode might be scheduled on the basis of predicted atmospheric ephemeris (e.g. meteorites, aurora, storms) or automatically activated, but it must not affect the achievement of the primary scientific goal of the mission. During normal operations, most activities are performed autonomously. All mission operations will be conducted according to the mission timelines, and to the control and contingency recovery procedures. Calibration activities and any maintenance activities will be scheduled from ground and executed as needed. It is presently expected that scientific calibrations will occur routinely on a daily basis. Technical calibrations will take place during the day (lid closed). Normal observation will also be interrupted in non-quiescent periods or by specific ISS ephemeris (e.g. docking of the shuttle) as well as any event which might threaten the safety of the instrument or of the ISS crew.

9.1.5 Science Ground Segment The ground segment is the complete ground infrastructure needed to operate EUSO. It is composed by the ISS ground stations, the Columbus Control Centre (Col-CC) and the dedicated EUSO scientific ground segment, the Science Operations and Data Centre (SODC). The SODC will carry on, in parallel, mission planning, TC preparation, verification and uplink, telemetry reception, monitoring and pipeline processing and on-board SW maintenance activities. It will also maintain the mission archive. The baseline scenario assumes that mission operations are controlled by the Col-CC, while the SODC handles the scientific operations and data processing. The SODC receives telemetry from the instrument via the Col-CC and sends telecommands to the Col-CC for uplink. During the operations phase the SODC has a 24 hours duty cycle, mainly to handle instrument monitoring activities. However, heavier activities should as much as possible be performed during the day. This is the case of mission planning and calibration activities. The foreseen large on-board autonomy requires particular care with test and validation activities. During operations the SODC hosts an average of 20 people, including permanent staff and visiting scientists. During commissioning, it is expected that Instrument and PI team members are hosted at the SODC. The basic SODC functionalities can be organized in three units:

1. TC Generation (TCG), Mission Activity Planning (MAP) and Test and Validation Facility /SW Maintenance (TVF/SWM) This functional unit of the SODC contains different modules related to system commanding and uplink. The mission activity planning role of the SODC consists on the scheduling of calibration activities, system configuration updating and other maintenance activities, and on the definition of the time-sharing between the different operation modes. Schedule files will have to be prepared. They might be short (1 orbit) or longer term activities. Whenever possible, long term planning and in advance scheduling should be the rule. Calibration procedures might be an important part of mission planning activities. The activity schedule generation receives as input the mission requirements (observation times, ephemeris, calibration needs, …) and constrains (orbit, day/night cycles, ISS ephemeris, …). Activity schedule preparation includes the validation of the activity schedule files (in view of the constraints) before TC sequence generation. On ground, there must be an on-board SW validation, maintenance and test facility. As a minimum, all the key (interface and communication) subsystems must be present in this environment. However, it is desirable to have the emulation of at least one full module of the system (e.g. one TCU and one full macro-cell, excluding the PMTs). This allows SW debugging and test, both for SW updates and in case of failure on-board. As a result of the mission activity planning and test, validation and maintenance activities, TC or TC sequences will have to be generated, verified and prepared for uplink. This functional block of the SODC is sketched in Figure 9.1.5-1.

2. TM Handling (TMH), Data Processing and Monitoring (DPM) Science and HK telemetry received from EUSO are pre-processed (unpacked, de-compressed, formatted) and streamed according to their type and to the instrument operation mode. Scientific data are pipeline processed within the SODC. They are pre-selected, the raw signals are calibrated, pattern reconstruction algorithms and event calibration parameters are applied. They are finally validated, archived and distributed. The monitoring of scientific data (quick-look analysis) is performed at the raw data level and at different stages of the processing. HK

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/7

parameters are used for instrument monitoring (instrumental checks). The HK telemetry includes relevant ISS ancillary data. This allows the monitoring of the instrument health and status, the control of the quality of the scientific data collected and of the pipeline processing procedures, as well as the following of the instrument evolution and the trend analysis of its performance. A quick response to exceptional scientific events could be envisaged. This functional block of the SODC is sketched in Figure 9.1.5-2..

MAP Activity SW schedule development preparation

TC SW sequence Test and generation validation

TCG TC SW patch Config file validation TC generation update TC

TVF/SWM

TC Uplink

Figure 9.1.5-1 Science Operations and Data Center: functional block of the TCG, MAP and TVF/SWM unit.

EUSO telemetry

TMH pre-processing & streamming atm data Science+HK calib data science data data DPM

atm monitoring processing calibration processing calib science calib atm monitor data params archive logs+data params params archive manager calib + atm + HK params manager

Figure 9.1.5-2 Science Operations and Data Center: functional block of the TMH and DPM unit.

3. Archive, database and user support The archiving system and in particular its user interface should take into account the different phases of the mission lifecycle and the different types of users, keeping in mind that all the relevant mission data will eventually be of public domain. Thus, both the archive contents and the interface design should be kept flexible, extensible and modular. The interface to the data archive should be on-line, user friendly and providing enough flexibility for the expert users. Parts of the archive will be password protected during the period of data property.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/7

The EUSO scientific archive should include, organise and give access to different sets of information: scientific data, atmospheric data, calibration matrices and energy/position calibration files, analysis SW, quick-look analysis tools, monitoring logbooks, housekeeping parameters, instrument users manuals, and, in general all relevant documentation and SW tools. Access to external databases (e.g. atmospheric, geographic, ISS) has to be provided in a transparent way to the user.

9.1.6 Scientific Data Products

The main objective of the EUSO mission is to obtain a detailed description of the Ultra High Energy Cosmic Rays spectrum together with a map of the arrival directions. Additionally, EUSO will detect data from phenomena intrinsic to the atmosphere or induced by the flux of meteoroids incoming from space. Measurements of the properties of the atmosphere (clouds, transparency, …) are required to correctly “certify” all such events. EUSO data will be collected, processed into standard formats (TBD), distributed to the science teams and made available to the scientific community in agreement with the ESA rules and directives. As described in the previous paragraphs, the EUSO Archive will be implemented for storing all the mission data, including both telemetry (science/housekeeping) and ancillary data (attitude, ephemeris and time-conversion data, ...). To allow both the re-analysis of the reduced data from telemetry and the scientific analysis and interpretation, three levels of data products are foreseen: • Level Zero (raw data) – essentially including telemetry data

• Level One (reduced/calibrated/reformatted data: from telemetry to reduction) – consisting of event records in which the events are identified and the data for each event is calibrated. This data product will also contain correlated information for each event such as, for example, the global positioning system time, the ISS ephemeris, local apparent time, and knowledge of the instantaneous pointing of the EUSO instrument. Data from the atmosphere sounding device and from the GLS units system will be present too. The format of the Level One data product is under study.

• Level Two (integrated information: from the analysis to the interpretation) – this should contain integrated information from the analysis of Level One certified events, as shower longitudinal profile, arrival direction, energy, nature of the primary,…

A proper data analysis system will facilitate community involvement.

9.2 Interfaces with the Community The user community will evolve with the different phases of the mission exploitation: from the EUSO instrument teams and expert users in the beginning of the operations, to the whole EUSO collaboration during the mission operation phase, and to the scientific community at large one year (TBC) after the end of operations. While most of the data quality control and processing is expected to be performed during the operations phase or shortly after, both mission archive and user support functionalities will have to be provided for several years during the post- operations phase. During operations, the SODC will assure the access of all the EUSO scientific consortium to the mission data.

9.3 Management of Operations The baseline scenario assumes that mission operations are controlled by the Col-CC, while the SODC handles the scientific operations and data processing. Scientific operations are conducted by the SODC under the authority of the EUSO PI.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/4

10. COMMUNICATION AND OUTREACH ...... 2 10.1 EUSO-EPO: THE EDUCATION AND PUBLIC OUTREACH PROGRAM...... 2 10.2 COMMUNICATION WITHIN THE CONSORTIUM ...... 4 10.3 PLANS AND NEEDS ...... 4

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/4

10. Communication and Outreach We live in a society which takes advantage of the technologies offered by the scientific progress. In contrast to the tremendous gains offered by science in the last fifty years, students' and the public's interest in working and learning science has sharply diminished. In western countries the decrease of students majoring in physics amounts to a factor of two over the past ten years. There is a deficit of scientific communication, education, popularisation, and the positive perception of physics. To communicate the message of science has now become an obligation. In particular it is essential to reach the young public and their teachers. The EUSO experiment, the appear and allure of the International Space Station, and more generally astrophysics and the study of cosmic radiation offer an ideal tool for the purpose of motivating the next generation of scientists. Many studies have shown that young people are engaged and motivated by space science and astrophysics. EUSO will give students hands-on use of real data, real experiences in science, and enable them to study and share in the discovery of the most energetic particles in the Universe. During the Phase A study, a detailed design of the Education and Outreach Program for EUSO has been prepared (see document EUSO-EPO-DD-001); here we present in brief its fundamentals together with a note about the communication within the consortium, future plans and needs.

10.1 EUSO-EPO: The Education and Public Outreach Program An Outreach Program will aim at explaining to the largest possible public the challenges of to-day physics research at the extreme borders of the Universe, hoping to attract young people into the field. In the case of EUSO, the aim of the Outreach program is to make the experiment known in circles as large as possible and through EUSO to promote an example of “Research in development”. There are two levels of action: • information, to develop interest in the general public about science and technology, and • education, to reach the young people within the educational system. Several initiatives can be thought of depending on the public to be addressed.

Society at large The use of Internet is now becoming widespread, and a good Web page with a section aiming at the general public is being developed. It can be viewed at http://www.euso-mission.org The EUSO collaboration being widespread in the world, it is important that “national” pages be developed with links to a central version. Such pages should include a general presentation of the experiment giving the translation of an “official” page, but could be complemented by regional news.

Colleagues Cosmic rays physics touches the domains of both particle physics and astrophysics. The usual ways of disseminating information about EUSO will be developed: presentations at conferences and articles in specialised journals. To this aim, it is good to maintain as soon as possible a set of quality documents, which could be deposited on the Web pages. One can also communicate through magazines of the community. A first article in the widely distributed CERN Courier has been published in March 2002. Further articles have appeared in the ESA Bulletin in May 2002, and in the ESA On Station Newsletter in December 2002. A citation on EUSO is present in an article about cosmic ray in Pour la Science in February 2002 and in the March 2003 edition of Sky and Telescope.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/4

Journalists Journalists constitute an important interface between the physicists and the public at large. It is often at a personal level that contacts can be made. It is also possible to contact them as a body. The list of scientific journalists can be easily found in a given country. A dossier should be prepared with a presentation of the physics aimed at by EUSO shown in an enticing way. This dossier would then be sent with the name of a physicist available for further inquiries as a local contact. This initiative should start with journalists working at scientific magazines. Articles in newspapers touch a larger audience but they can only be envisaged in the rare case of « an event » for example at the time of the launch of the telescope.

The general public Several initiatives can be thought of : • A model of the experiment available for display in scientific museums. It could constitute the focus of an exhibition on cosmic rays among other examples of detectors like cloud chambers, spark chambers, scintillation counters, ... A model of the EUSO observational approach has been already prepared by a group of students belonging to the “Lycée Blaise Pascal” of Longuenesse in France. It represents a rotating earth observed by the Space Station. Cosmic rays are materialized by a bunch of optical fibres which light up when reaching the atmosphere controlled by a button panel.

• A display of large photographs in public areas such as airports, commercial centres, public gardens… For example one can think of a set of 100 large photos exploring various aspects of research from the infinitely small to the infinitely large. These photos can easily be found at CERN, ESA, ESO, …

• A series of conferences for the public at large.

• A film tracing the progress of the EUSO experiment since its beginning. This should start right at the end of Phase A Study.

• A book on “EUSO and the cosmic rays”. Several new collections of cheap scientific books have sprung recently. Editors are willing to publish such a book at author’s expenses. In particular contacts have started with Editions Le Pommier which is willing to prepare multilingual 60 pages booklet.

Teachers Pamphlets and books with simple explanations and photographs should be prepared to be sent to teachers. In particular this material could be given to Scientific Clubs which exist in some secondary schools. It is also important to offer courses or activities to teachers in their educational program. This can be organised in “summer sessions” around activities discussed in the next chapters. Data analysis tools, as those based on the Hands-On Universe software package, should be spread and taught throughout all EUSO nations. There are strong Global Hands-On Universe collaborations in France, Germany, Japan and the US. An emerging HOU system is growing in Italy. This is a natural and proven method to train teachers, develop, and disseminate materials so that EUSO data can be useful in classrooms.

Students At the highest level, the groups forming the collaboration are encouraged to take students to prepare them for PhD theses. The extremes of the Universe as well as the ISS fascinates young people and theses proposed on the subject should attract good candidates. Cosmic rays shower everywhere on earth, and they are easy detectable for the most common of them. To reach the young public, it is proposed to set-up arrays of detectors to measure the flux and some properties of cosmic rays in schools. This allows to discuss several aspects of physics both theoretical and experimental. In such an endeavour, students must feel as actors of research. Several initiatives in the US, and more recently in Germany, France, Poland, Portugal have been started. In France, a prototype experiment is being set-up in the “Lycée Pierre Forest” of Maubeuge, starting with two borrowed scintillation counters and their electronics. In Portugal a similar program involves ten secondary schools,

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/4

with LIP as a central reference point. The aim of the American project (CROP project in the State of Nebraska, Chicanos in Los Angeles, …) is to equip up to several hundred schools for a total budget of 1.3 million US dollars granted from NSF. EUSO scientists are collaborators or in communication with these projects.

10.2 Communication within the Consortium Besides the necessary electronic mail tools, the communication within the EUSO Consortium is guaranteed by its official web site “EUSOweb” at http://www.euso-mission.org where, apart from the public pages, a restricted area reserved to the Collaboration allows to exchange specific information: in this area, and for the time being, EUSO Teams mainly found a simple database system devoted to the management of meetings and news of interest for EUSO. Moreover, most of the relevant EUSO Consortium documentation (as, for example, scientific and technical reports) can be retrieved from here. A set of figures, graphics, images, photo and movies related to EUSO is under preparation too. A service of News/Bulletin automatically sent towards the Collaboration is foreseen starting from Phase B. The main depository of all EUSO Consortium documentation (based on the Livelink database management system) has been settled at ESA-ESTEC (http://astro.esa.int/livelink/ ) where EUSO Teams can search, read and copy out documents. The technical management of the site is guaranteed by ESA; the content is under the control of two “gatekeepers”, one from ESA, the other one from the EUSO Consortium. When a document is included by a gatekeeper in the Livelink archive, a message is automatically sent to the distribution list specifically settled (it mainly includes the Responsible and Deputy of the various EUSO groups, as defined in the “Management” Chapter of this Phase A Study Report) .

10.3 Plans and Needs EUSO is about research in physics and the main aim of the experiment is to measure the spectrum of cosmic rays in the domain of the highest possible energies never looked at. At the same time it is a technical and sociological adventure. For the public it constitutes a good example of Research in the Making. With a dedicated effort, an Outreach program can develop the taste for physics among the public and in particular among young people. The proposal to set up a cosmic rays search through an array of schools is a way to directly communicate with students at an early age. In Paris contacts have been made with a major scientific museum, the “Palais de la Découverte”. The museum is very eager to house one of the stations thus linking itself to the network of schools nearby. To be successful, such a program requires a physicist ready to be the contact person with a school. Ideally it would be good to associate one physicist with one class. In Italy, contacts with “La Città della Scienza” in Neaples have been started and will be further continued and improved next Autumn. A minimum of scientific culture is necessary to understand the world of to-day. The aim of and Outreach program is precisely to fill the present gap between the scientists and the public. Such a program of Outreach is sketched. Tests are starting and contacts with schools have been made.

Money is necessary in order to go ahead with the proposals presented here. The EUSO-EPO programme would be supported by the European Space Agency for the discovery and dissemination of new knowledge from space, and find its place within the general goals of the ESA communication strategy.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/6

11. INTERNATIONAL COLLABORATION AND MANAGEMENT...... 2 11.1 GENERAL STRUCTURE ...... 2 11.2 THE ESA SCIENCE STUDY TEAM (SST)...... 2 11.3 THE EUSO CONSORTIUM ...... 3 11.3.1 Participant Nations and Institutes/Institutions ...... 4 11.3.2 The EUSO Science Steering Committee (ESSC)...... 5 11.3.3 The EUSO Working Groups ...... 5 11.3.3.1 The Instrument Working Group (IWG)...... 5 11.3.3.2 The Science Objectives Working Group (SOWG) ...... 6 11.3.3.3 The Education and Public Outreach Working Group (EPOWG)...... 6 11.4 PROCUREMENT PHILOSOPHY ...... 6 11.5 INDUSTRIAL MANAGEMENT ...... 6 11.6 DEVELOPMENT PHILOSOPHY AND SCHEDULE ...... 6 11.7 EVALUATION OF COST TO THE EUSO CONSORTIUM FOR PHASE B TO MISSION COMPLETION ...... 6

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/6

11. International Collaboration and Management Introduction …

11.1 General Structure EUSO is a collaborative effort of research groups in Europe, Japan and US, forming the so-called EUSO Consortium which is responsible for the EUSO Instrument. The Phase A study of EUSO consisted of parts conducted by both ESA and the EUSO Consortium. ESA supported the Phase A Study of EUSO through its Science Study Team (ESA-SST). The coordination between ESA and EUSO Consortium activities has been guaranteed by the key-people (ESA Study Scientist, ESA Study Manager, EUSO Principal Investigator, EUSO Instrument Manager) as schematically shown in Figure 10.1.1,and enhanced by a number of different types of EUSO meetings.

ESA EUSO Consortium Payload level activities Instrument level activities

Scientific Aspects: ⇔ Scientific Program: Study Scientist Principal Investigator (EUSO-PI) A. Parmar (D-SCI) L. Scarsi

Technical & Managerial Aspects: ⇔ Technical Program: Study Manager Instrument Manager (EUSO-IM) G. Gianfiglio (D-MSM) O. Catalano Figure 11.1-1 Schematic view of the EUSO general structure during Phase A study.

11.2 The ESA Science Study Team (SST) ESA supported the Phase A study of EUSO through its Science Study Team (ESA-SST) chaired by the ESA Study Scientist assisted by a Deputy; its membership comprised the ESA Study Manager, the EUSO Principal Investigator, the EUSO Instrument Manager, and four external scientific experts. For the time being, the ESA responsibilities have been shared between the two directorates of Science (D-SCI) and Space Station (D-MSM). The Science Study Team responsibilities have been mainly devoted to: • support ESA in the Phase A study of EUSO,

• provide scientific oversight of the mission,

• consolidate the technical design to meet the scientific requirements within the constraints of the ISS attached payloads programme,

• assist in top-level system trade-offs,

• recommend on readiness for Phase B,

• produce the Phase A final report under the coordination of the Study Scientist.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/6

11.3 The EUSO Consortium The EUSO Consortium, responsible for the EUSO Instrument, is formed by more than 120 scientists from several institutions in Europe, Japan, US. The EUSO Consortium is led by the Principal Investigator (EUSO-PI) assisted by a Deputy-PI. The Technical Program is led by the Instrument Manager (EUSO-IM) assisted by a Deputy-IM, too. The Consortium conducted the EUSO Instrument Phase A study together with specific supporting studies and technology developments. During the Phase A study, the main EUSO Consortium responsibilities included: • provide ways and means to support the EUSO-PI in refining and updating the scientific requirements of the EUSO mission, as well as in detailing the EUSO Instrument design on the basis of the scientific requirements and mission constraints that will apply;

• provide the interface to the industrial activities on the Instrument;

• provide plans for science, routine, emergency and calibration on-board operations;

• implement the nationally funded part of the mission;

• define the Instrument schedule and suitable major milestones leading from the definition phase through to the operational and archival phases and final mission products.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/6

11.3.1 Participant Nations and Institutes/Institutions The EUSO mission is a truly international effort …controllare

Nations and Scientific Institutions participating to EUSO

Italy IASF CNR - Istituto di Astrofisica Spaziale e Fisica Cosmica, Palermo ISAC CNR - Istituto di Scienze dell'Atmosfera e del Clima, Bologna INFN - Istituto Nazionale di Fisica Nucleare - Catania, Firenze, Genova, Torino, Trieste INOA - Istituto Nazionale di Ottica Applicata, Firenze CARSO - Center for Advanced Research in Space Optics, Trieste OACT - Astrophysical Observatory, Catania OAFI - Arcetri Astrophysical Observatory, Firenze Scuola Normale Superiore, Pisa University of Genova, DIFI - Department of Physics, Genova University of Palermo, DIFTER - Dept. of Physics and Related Technologies, Palermo University of Rome La Sapienza, Department of Physics, Roma University of Firenze, Department of Physics and Department of Astronomy, Firenze University of Torino, Department of Physics, Torino University of Trieste, Department of Physics, Trieste France APC - Astroparticle and Cosmology Laboratory, Paris CdF - College de France, Paris CESR - Centre d'Etudes Spatiales des Rayonnements, Toulouse IAP - Institut d'Astrophysique, Paris LAPP - Laboratoire d'Annecy-le-Vieux de Physique de Particules, Annecy LAPTH - Laboratoire d'Annecy-le-Vieux de Physique Theorique, Annecy LPSC - Laboratoire de Physique Subatomique et de Cosmologie, Grenoble LPTHE - Laboratoire de Physique Theorique et des Hautes Energies, Paris OdP - Observatoire de Paris, Paris SA - Service d'Aeronomie, Grenoble Germany MPIfP - Max Planck Institute for Physics, Munich MPIfRA - Max Planck Institute for RadioAstronomy, Bonn MPIHLL - Max Planck Institute, Halbleiterlabor, Munich University of Wuertzburg,Institute of Theoretical Physics and Astrophysics, Wuertzburg Portugal LIP - Laboratorio de Instrumentacao e Fisica Experimental de Particulas, Lisbon Switzerland Observatoire de Neuchatel, Neuchatel UK University of Leeds, Leeds Japan RIKEN - Institute of Physical and Chemical Research, Tokyo ICRR - Institute for Cosmic Ray Research, Tokyo ISAS - Institute of Space and Astronautical Science KEK - High Energy Accelerator Research Organization NAO - National Astronomical Observatory USA MSFC - Marshall Space Flight Center, NASA, Huntsville NSSTC - NASA, Huntsville UAH - University of Alabama in Huntsville, Physics Dept., Huntsville UCB - University of California at Berkeley, Berkeley UCLA - University of California at Los Angeles, Physics Dept., Los Angeles UT - University of Texas, Physics Dept., Austin VU - Vanderbilt University, Physics & Astronomy Dept., Nashville

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/6

11.3.2 The EUSO Science Steering Committee (ESSC) The EUSO Science Steering Committee (ESSC) is established to represent the interests of the EUSO Consortium in the mission and to support the EUSO Principal Investigator in the management of the Consortium activities. The ESSC is responsible for: • providing reference and advise to the EUSO-PI in the management of the Consortium activities,

• refining of the mission objectives and requirements in consultation with the EUSO Science Objectives Working Group (SOWG),

• instrument complement recommendations,

• the recommendation of the criteria used to determine the scientific observation program,

• the recommendation of a policy establishing the handling and distribution of the scientific data. The ESSC advice must be sought for any action or modification of the program that could have an impact on the science objectives. By means of the links to the national funding agencies, the ESSC will verify and guarantee that each member of the EUSO Consortium fulfils its obligations toward the mission. The ESSC is chaired by the EUSO Principal Investigator and is formed by: • the EUSO Principal Investigator (EUSO-PI),

• the EUSO Instrument Manager (EUSO-IM),

• the national representatives for the EUSO Consortium,

• the chairperson of the EUSO Science Objectives Working Group (SOWG),

• the chairperson of the EUSO Education and Public Outreach Working Group (EPOWG),

• the ESA Study Scientist (as an observer).

11.3.3 The EUSO Working Groups

11.3.3.1 The Instrument Working Group (IWG) The EUSO Instrument Working Group (IWG) is chaired by the Instrument Manager (EUSO-IM) assisted by the Deputy-IM, and it is composed of the responsible persons of each subgroup involved in the Instrument development as well as the ESA Study Manager (acting as an observer). During the Phase A Study, the EUSO-IWG has been composed by several sub-groups reporting to the Instrument Manager, namely: System, Structure and Thermal Control (ST), Optics Module (OM), Focal Surface (FS), Analog Front End Electronics (AFEE), Digital Front End Electronics (DFEE), Trigger Electronics & On-brad read-out (TEO), Atmosphere Soundind (AS), Simulations (SIM), Scientific Data Analysis (SDA), Science Operation and Data Center (SODC), Support Experimental Activities (SEA). The IWG has been responsible for conducting the technological Phase A Study of the EUSO Instrument. Its advice must be sought for any action or modification that could have an impact on the Instrument technology and development. The IWG is responsible for: • towards the EUSO-PI via the EUSO-IM for the development of the EUSO Instrument,

• the definition of the criteria and guidelines for the Instrument resources distribution among the different subsystems,

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/6

• the Instrument calibration during the entire mission,

• providing all the information and data needed for the definition and implementation of the observing strategy and scientific program.

11.3.3.2 The Science Objectives Working Group (SOWG) The Scientific Objective Working Group (SOWG) is charged with defining and updating the Science objectives of the EUSO Mission. The SOWG is composed of scientists who are specialists in scientific areas related to EUSO and is chaired by one of the members; the EUSO Deputy-PI acts as scientific secretary of the SOWG.

11.3.3.3 The Education and Public Outreach Working Group (EPOWG) The Education and Public Outreach Working Group (EPOWG) is charged with responsibility for EUSO education and public outreach related activities and mainly devoted to make EUSO in circles as large as possible. The development of an EUSO-EPO plan, devoted to the no-scientific community, has been defined during Phase A study, taking into account the different reality (school plans) in the EUSO participating countries. Chaired by one of its members, the EPOWG includes national representative persons for the EUSO Consortium, as well as the responsible of the EUSO Web Official Site.

11.4 Procurement Philosophy

11.5 Industrial Management

11.6 Development Philosophy and Schedule

11.7 Evaluation of Cost to the EUSO Consortium for Phase B to Mission Completion

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 1/7

12. References

Scarsi L. et al., F2/F3 proposal Scarsi L. et al., Accommodation Study IIDD last version

In Chapter 1 – Scientific Case

Abu-Zayyad T., et al. (HiRes Collaboration), 2002, e-print astro-ph/0208243; e-print astro-ph/0208301 Afanasiev B.N. et al., 1996, Proc. Int. Symp. On EHECR, Univ. Tokyo, Nagano, N. ed., Vol 32 Aharonian F.A. and Atoyan A.M., 1999, Astron. & Astroph. 351, 330 Aloisio R., Berezinsky V., and Blasi P., in preparation Alvarez-Muniz J., Engel R., and Stanev T., 2001, Astrophys. J. 572, 185 Auger P. & Maze R., 1938, Académie des Sciences 207, 228 Auger P., Maze R., Grivet-Meyer T., 1938, Académie des Sciences 206, 1721 Bahcall J. and Waxman E., 2001, Phys. Rev. D 64 Barbot C. and Drees M., 2002, Phys. Lett. B 533, 107 Berezinsky V. and Gazizov A., in preparation Berezinsky V. and Vilenkin A., 2000, Phys. Rev., D 62, 083512 Berezinsky V. and Kachelriess M., 2001, Phys. Rev. D 63, 034007 Berezinsky V.S. and Smirnov A.Y., 1975, Ap.Sp.Sci 32, 461 Berezinsky V.S. and Zatsepin G.T., 1969, Phys. Lett B 28, 423 Berezinsky V.S. and Zatsepin G.T., 1970, Soviet Journal of Nuclear Physics 11, 111 Berezinsky V.S., Bulanov S.V., Dogiel V.A., Ginzburg V.L. , Ptuskin V.S., 1990, “Astrophysics of Cosmic Rays”, North-Holland, Amsterdam Berezinsky V.S., 1977, Proc. of ``Neutrino-77'', 1, 177 Bertone G., Isola C., Lemoine M., and Sigl G., 2002, Phys. Rev. D 66, 103003 Bird D. J. et al., 1993, Phys. Rev. Lett. 71, 3401 Blanton M., Blasi P., and Olinto A.V., 2001, Astropart. Phys. 15, 275 Blasi P., Burles S., and Olinto A.V., 1999, Astrophys. J. 514, L79 Blasi P., Epstein R. and Olinto A.V., 2000, Astrophys. J. Lett. 533, 123 Boldt E. and Ghosh P., 1999, MNRAS 307, 491 Boldt E. and Loewenstein M., 2000, MNRAS 316, 29 Bottai S. & Giurgola S., EUSO-SIM-REP-002- Catanese M. and Weekes T.C., 1999, e-print astro-ph/9906501, invited review, PASP 111-764, 1193 Chen P., Tajima T., and Takahashi Y., 2002, Phys. Rev. Lett. 89, 1101

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 2/7

Chung D.J., Kolb E.W., and Riotto A., 1999, Phys. Rev. D 59, 023501 Cour-Palais B.G., 1969, "Meteoroid Environment Model [Near Earth to Lunar Surface]", NASA SP-8013 Cronin J.W., 1992, Nucl. Phys. B (Proc. Suppl.) 28B, 213; see also The Pierre Auger Observatory Design Report (ed. 2), March 1997 Divine N., 1993, J. Geophys. Res., 98-E9 Domokos G. and Kovesi-Domokos S., 1999, Phys. Rev. Lett. 82, 1366 E.Witten,1985, Nucl. Phys. B 249, 557 Fargion D., Mele B., and Salis A., 1999, ApJ 157, 725 Feng J. L. and Shapere A.D., 2002, Phys. Rev. Lett. 88-021303 Fodor Z., Katz S.D., and Ringwald A., 2002, Phys.~Rev. Lett. 88, 171101 Gaisser T.K., 1998, “Cosmic Rays and Particle Physics”, Cambridge University Press, Cambridge Gandhi R., Quigg C., Reno M.H., and Sarcevic I., 1996, Astropart.Phys. 5, 81 Gandhi R., Quigg C., Reno M.H., and Sarcevic I., 1998, Phys. Rev. D 58-093009 Gorbunov D., Tinyakov P., Tkachev I., and Troitsky S., 2002, Astrophys. J. Lett. 577, 93 Gorbunov D.S., Raffelt G.G., and Semikoz D.V., 2001, Phys. Rev. D 64-096005 Grasso D. and Rubinstein H., 2001, Phys. Rept. 348, 163 Greisen K., 1966, Phys. Rev. Lett. 16, 748 Grün E., Zook H. A., Fechtig H. and Giese R. H., 1985, Icarus, 62, 244 Halzen F. and Hooper D., 2002, Rept. Prog. Phys. 65, 1025 Hayashida H. et al., 1999, Astropart. Phys. 10, 303 Hess V.F., 1912, Phys. Z. 13, 1084 Hill C.T., Schramm D.N., and Walker T.P., 1987, Phys. Rev. D 36, 1007 Hillas A.M., 1984, Ann. Rev. Astron. Astrophys. 22, 425 Isola C. and Sigl G., 2002, Phys. Rev. D 66-083002 Jenniskens P., 1994, Astron. Astrophys. 287, 990 Jones F.C. and Ellison D.C., 1991, Space Sci. Rev. 58, 259 Kalashev O.E., Kuzmin and Semikoz D.V., 1999, e-print astro-ph/9911035 Kuzmin V. and Tkachev I., 1999, Phys. Rept. 320, 199 Lawrence M.A., Reid R.J.O., Watson A. A., 1991, J. Phys. G : Nucl. Part. Phys. 17, 773 Lee S., 1998, Phys. Rev. D 58-043004 Levinson A., 2000, Phys. Rev. Lett. 85, 912 Linsley J., 1963, Phys Rev. Lett. 10, 146 Linsley J., 1982, Proc. Workshop on “Very High Energy Cosmic-Ray Interactions”, University of Pennsylvania, Cherry, M. L., Lande, K., and Steinberg, R. I. (Eds.) , 476 Mannheim K., Protheroe R.J., Rachen J.P., 2001, Phys. Rev. D 63-023003 MASTER model (Meteoroid And Space Debris Terrestrial Environment Reference Model Sdunnus, H., Meteoroid and Space Debris Terrestrial Environment Reference Model "MASTER", Final Report to ESOC Contract 10453/93/D/CS, 1995.

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 3/7

Medina-Tanco G. and Ensslin T., 2001, Astropart. Phys. 16, 47 Nagano M. & Watson A.A., 2000, Rev. Mod. Phys. 72-73, 689 Narayan R., Paczynski P., and Piran T., 1992, Astrophys. J. Lett. 395, L83 NASA90 model [Anderson, B. J., Review of Meteoroids/Orbital Debris Environment, NASA SSP 30425, Revision A, 1991]. Norman C.A., Melrose D.B., and Achterberg A., 1995, Astrophys.J. 454, 60 Ong R.A., 1998, Phys. Rept. 305, 95 ORDEM96 model, also referenced as the NASA96 model [Kessler, D. J., J. Zhang, M. J. Matney, P. Eichler, R. C. Reynolds, P. D. Anz-Meador, and E. G. Stansbery, A Computer-Based Orbital Debris Environment Model for Spacecraft Design and Observations in Low-Earth Orbit, NASA Technical Memorandum 104825, 1996.] Protheroe R.J. and Stanev T., 1996, Phys. Rev. Lett. 77, 3708; erratum, ibid. 78 (1997) 3420 Protheroe R.J. and Biermann P.L., 1996, Astropart. Phys. 6, 45 Puget J.L., Stecker F.W., and Bredekamp J.H., 1976, Astrophys. J. 205, 638 Ryu D., Kan, H., and Biermann P.L., 1998, Astron. Astrophys. 335, 19 Sarkar S. and Toldra R., 2002, Nucl. Phys. B 621, 495 Scully S.T. and Stecker F.W., 2002, Astropart. Phys. 16, 271 Shinozaki K. et al. (AGASA collaboration), 2002, Astrophys. J. 571, L117 Sigl G., Lee S., Schramm D., and Coppi P.S., 1997, Phys.Lett.B 392, 129 Sigl G., Schramm D.N., and Bhattacharjee P., 1994, Astropart.Phys. 2, 401 Stanev, 1997 Staubach P., Grün E. and Jehn R., 1996, 31-th COSPAR Sci. Assembly, Birmingham, UK Stecker F.W. & Salamon, M. H., 1999, Ap. J . 512, 521 Stecker F.W., 2000, Astropart. Phys. 14, 207 Szabo A.P. and Protheroe R.J., 1994, Astropart. Phys. 2, 375 Takeda M., et al., 1998, Phys. Rev. Lett. 81, 1163 Takeda M., et al., 2002, astro-ph/0209422, 13 Nov. 2002 Teshima M. et al., 2001, 27th ICRC, 1, 337 Tinyakov P.G. and Tkachev I.I., 2001, Pisma Zh. Eksp. Teor. Fiz. 74, 3 [JETP Lett. 74, 1] Tinyakov P.G. and Tkachev I.I., 2001, JETP Lett. 74, 445 Venkatesan A., Aparnar , Miller M.C., and Olinto A.V., 1997, Astrophys. J. 484, 323 Vietri M. and Stella L.,1998, Astrophys. J. Lett. 507, 45 Vietri M., De Marco D., and Guetta D., 2003, preprint astro-ph/0302144 Vietri M., 1995, Astrophys. J. 453, 883 Waxman E. and Bahcall J., 1999, Phys. Rev. D. 59 Waxman E., 1995, Phys. Rev. Lett. 75, 386 Weiler T.J., 1982, Phys Rev. Lett. 49, 234 Weiler T.J., 1999, Astropart. Phys. 11, 303 and 317

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 4/7

Yoshida S. & Teshima M., 1993, Progr. Theor. Phys., Vol. 89 Yoshida S., Sigl G., and Lee S., 1998, Phys. Rev. Lett. 81, 5505 Zatsepin G.T. & Kuzmin V.A., 1966, Pis'ma Zh. Eksp. Teor. Fiz. 4, 114; JETP. Lett. 4, 78

In Chapter 2 – Observational Approach Berezinsky V.S. et al., 1990, “Astrophysics of Cosmic Rays”, North Holland, Amsterdam, 1990

In Chapter 3 – AS Scientific Rationale EUSO-AS-REP-002-1, "EUSO-LIDAR, The scientific rationale to have an atmosphere sounding system coupled to the EUSO detector “, 10 July 2003, G. Fiocco, R. Viola ON/EUSO/1.00/2003/18_July_2003, “Comparison between lidar techniques for AS-EUSO application by numerical performance simulation”, 18 July 2003, Observatoire de Neuchâtel, Neuchâtel, Switzerland

In Chapter 5 – Instrument Takahashi Y., Lamb , Hillman L., 1999 Zuccaro A., 2003, “Wide Angle Large Refractive Optics for Astronomical Observations: Design and Testing of Fresnel Lenses,” PhD Dissertation, Università degli Studi di Trieste, 2003 Gambicorti L., Mazzinghi P., Pace E., 2003, “UV filtering optical adapters to enhance the collection efficiency of MAPMT R8900”, DASS Tech Rep. TR4-2003. EPOTEK, Epoxy Technologies Inc., Billerica, MA 01821 RDFS1

V. Gracco and A. Petrolini, Use of multi-anode photomultipliers for the AirWatch photon-detector, EUSO-FS-note-002, 23/10/1998;

The AirWatch Collaboration, Workshop on Observing Giant Cosmic Ray Air Shower, AIP Conf. Proc., 433, 353-357, (1998);

The AirWatch Collaboration, Proc. SPIE 3445, 486, (1998);

R. Stalio et al., Proc. 26th ICRC 2, 403 (1999);

M. Ameri et al., Study report on the EUSO photo-detector design, Report INFN/AE-01/04, (2001);

M. Ameri et al., EUSO (Extreme Universe Space Observatory): the focal surface photo-detector, 27th ICRC (2001);

H. M. Shimizu et al., AIP Conf. Proc., 566 (2001) 381. RDFS2

Optimization of OWL-AirWatch Optics & Photo-Detectors, K. Arisaka, UCLA-COSMIC/1999-02, (1999). RDFS3

MAPMT R5900/R7600 series, Hamamatsu Photonics K.K.. RDFS4

M. Ameri et al., Proc. 27th ICRC 856 (2001). RDFS5

EUSO-FS-REP-005, PMT PROCURMENT SPECIFICATION. RDFS6

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 5/7

EUSO-FS-REP-006, Characters of MAPMT.

RDEL1

O. Catalano, AIRWATCH FROM SPACE: progress report, IFCAI/CNR Internal Report, March 1999;

O. Catalano, Extreme Universe Space Observatory—EUSO: an innovative project for the detection of extreme energy cosmic rays and neutrinos, IL NUOVO CIMENTO Vol. 24 C, N. 3, Maggio-Giugno 2001.

RDEL2

G. Bosson et al., phase A Design and Prototype studies for the EUSO Front-End Electronics (FEE), EUSO-FEE-SP-001

ASDR [RD6 and RD13]

EUSO-SDA-REP-012-1 " The possible use of InfraRed cameras in EUSO ", 13 May 2003, P. Abreu, A. Anzalone, O. Catalano, M. Pimenta RDPS1

CAEN, BATTERY ON-BOARD STUDY FOR EUSO EXPERIMENT, Apr. 2003. RDPS2

LABEN, MODULAR SOLUTION FOR POWER DISTRIBUTION AND CONVERSION OF THE EUSO EXPERIMENT, DOCUMENT No.: TL19343, Dec. 2002;

CAEN, Sextant Centralized Power Supply Solution for EUSO, CAENA/EUSO/T-01, Issue 1, 31/01/2003.

[IIDD] quale ?

In Chapter 6 – Expected Performance: Agostinelli S., et al., 2003, NIM A 506, 250 Allen C.W, 1973, “Astrophysical Quantities”, Athlone Pub. Allen H.J. et al., 1963, NASA Tecnhical Report TR R-185, Washington D.C. Baldwin B & Shaeffer Y., 1971, J.Geophys.Res. 76, 4653 Baltrusaitis et al., 1987, J. Phys. G Nucl. Phys. 13 Berat C. et al., 2003, Proc. ICRC2003, Tsukuba, Japan, and references therein Bernlohr K. ??? Bottai S., 2001, Proc. 27th ICRC, Hamburg, Germany, HE 180, 848 Bronshten V.A., 1983, “Physics of Meteoric Phenomena”, D.Reydel Publication Co Brun R. et al, 2001, http://root.cern.ch Bunner A.N., 1964, Ph.D Thesis, Cornell University, Ithaca, NY Catalano 0. et al., 2002, Nuclear Instruments & Methods, NIM-A 480, 2-3, 547

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 6/7

Catalano O. et al., 2001, Proc. 27th ICRC, Hamburg, Germany, HE 498 Ceplecha Z. et al., 1998, Space Science Reviews 84, 327 Ceplecha Z., 1992, Astronomy and Astrophysics 263, 361 Cinzano P. et al., 2001, astro-ph/0108052 Engel J. et al., 1992, Phys. Rev. D46 Engel J. et al., 1999, Proc. 26th ICRC, Salt Lake City, Utah, 1 FIWAF 2002, Proc. First Int. Workshop on Atmospheric Fluorescence, Salt Lake City, Utah Fletcher R.S. et al., 1994, Phys Rev D50 Giarrusso S. et al., 2001, Proc. 27th ICRC, Hamburg, Germany, HE 684 Grimmer G., 1948, Journal of Applied Physics 19-10, 947 Heck D. et al., 1998, Report FZKA6019, Forschungszentrum Karlsruhe Hedin A.E., 1991, J.Geophys.Res., 96-A2, 1159-1172 Hillas A.M., 1982, J. Phys. G Nucl. Phys. 8 Hills J.G. and Goda M.P., 1993, Astron. J. 105-3, 1114 ISCCP, 1999, http://isccp.giss.nasa.gov/ Kakimoto F. et al., 1995, Nucl. Instr. and Meth. for Phys. Research. A372, 244 Kalasev E. et al., 2002, hep-ph/0205050 Kalashev O.E. et al., 1999, astro-ph/9911035 Kneizys F.X. et al., 1996, "The MODTRAN 2/3 Report and LOWTRAN7 Model", L.W.Abreu and G.P. Anderson Eds. Lebrun D., 2002, ISN-Grenoble Report 02-37 Leinert Ch. et al., 1998, Astron. & Astrophys. Suppl. Series 127,1-99 Levedinsky A.I. et al., 1965, Space Research, 77, Nauka, Moscow, Linsley J., 2001a, Proc. 27th ICRC, Hamburg, Germany, HE 502 Linsley J., 2001b, Proc. 27th ICRC, Hamburg, Germany, HE 683 Liu V.C., 1978, Geophys.Res.Lett. 5,309 Maccarone M.C. et al., 2001, Proc. 27th ICRC, Hamburg, Germany, HE 852 Mannheim K., 1995, Astroparticle Physics 3, 295 Nagano M. et al., 2003, astro-ph0303193, March 2003 O’Keefe J.D. & Ahrens T.J., 1982, in “Geological implications of impact of iarge asteroids and comets on the earth”, L.T.Silver and P.H.Schultz Ed., Geol.Soc.Am.SP 190, 103 Passey Q.R & Melosh H.J., 1980, Icarus 42, 211 Picone M. et al., http://nssdc.nasa.gov/space/model/atmos/nrlmsise00.html Protheroe R.J. & Johnson P.A., 1995, Astroparticle Physics 4, 253 Protheroe R.J. & Stanev T., 1996, Phys. Rev. Lett. 77, 3708 Sciutto S., 1999, Auger Project Report GAP-99-044 Sigl G. et al., 1997, Phys. Lett. B 392, 129

EUSO Instrument DOCUMENT: EUSO-PI-REP-002 ISSUE: -1 Report on the Phase A Study REVISION: - DATE: - JULY 2003 PAGE: 7/7

Sigl G., 2000, Lect. Notes Phys. 556, 259 USSA-76, http://nssdc.gsfc.nasa.gov/space/model/atmos/us_standard.html Whipple F.L., 1950, “The Collected Contributions of Fred L Whipple – Volume One: Meteors and Comets and the Interplanetary Complex”, SAO Cambridge, MA Christian H.J., 1999, http://thunder.msfc.nasa.gov/otd/

EUSO-AS-REP-001-1.1, "EUSO-LIDAR, The scientific baseline to have an atmosphere sounding system coupled to the EUSO detector", 10 October 2002, G. D'Alì Staiti, G. Fiocco, E. Plagnol, R. Viola , (plus Annexes A and B by L. Scarsi and Y. Takahashi) EUSO-SDA-REP-002-0, "Cluster Analysis and Line Fits to reconstruct EUSO tracks", 24 April 2002, M.C. Maccarone EUSO-SDA-REP-008-1.1, "ESAF - EUSO Simulation and Analysis Framework", 13 May 2002, D. De Marco and M. Pallavicini EUSO-SDA-REP-013-1, "Meteors Observation by EUSO: General Overview and Scientific Objectives", 19 February 2002, P. Scarsi EUSO-SDA-REP-015-1.1, “SLAST – Shower Initiated Light Attenuated to the Space Telescope”, 13 March 2003, D. Naumov EUSO-SDA-REP-016, “ EUSO Reconstruction Algorithms”, 11 July 2003, P. Colin, D. Naumov, P. Nedelec EUSO-SEA-REP-001-1, "ULTRA - Uv Light Transmission and Reflection in the Atmosphere - a supporting experiment for the EUSO project", 4 June 2002, O. Catalano, P. Vallania, D. Lebrun, P. Stassi, M. Pimenta, C. Espirito Santo EUSO-SIM-REP-006-2, “Background sources ", 6 February 2003, P. Colin EUSO-SIM-REP-008-1, "Meteors Observation by EUSO. From Meteoroids to Meteors: the general Process", 10 March 2003, P. Scarsi EUSO-SIM-REP-009-1, " EUSO Duty Cycle: Moon and Sun Light Effects ", May 2003, F. Montanet EUSO-SIM-REP-010, “EUSO acceptance and counting rates”, draft version, March 2003, E. Plagnol

In Chapt 10 – Communication and Outreach: EUSO-EPO-DD-001-1, “The EUSO EPO Program – Preliminary Design Document”, 20 June 2003 F. Vannucci, M.C. Maccarone, C. Pennypacker

Main EUSO Web Sites: “The Official EUSO Web Site”, http://www.euso-mission.org/ “The EUSO Livelink Archive at ESA”, http://astro.esa.int/livelink/