Home

Software User Manual

1. Sentinel 3 SDP Software Development Plan SLSTR Sea and Land Surface Temperature Radiometer SME Small amp Medium size Enterprise SNO Simultaneous Nadir Overpass TBC To Be Confirmed TBD To Be Defined TOA Top Of Atmosphere TPM Third Party Mission U K United Kingdom WG Working Group WGCV Working Group on Calibration and Validation l a ed ARGANS DIMITRI MEREMSII n Ref ME MAN ARG TN SUM Date 08 05 2012 Towards the intercalibration of EO medium resolution multi spectral imagers Page 7 Software User Manual D14 Reference Documents RD 1 Bouvet M Intercomparison of imaging spectrometer over the Salar de Uyuni Bolivia Proceedings of the 2006 MERIS AATSR Validation Team Workshop RD 2 Bouvet M Ramoino F Radiometric intercomparison of AATSR MERIS and Aqua MODIS over Dome Concordia Antarctica Can J Remote Sensing Vol 36 No 5 pp 464 473 2010 RD 3 Bouvet M Goryl P Chander G Santer R Saunier S Preliminary radiometric calibration assessment of ALOS AVNIR 2 proceeding of the IGARSS 2007 RD 4 Smith D Poulsen C Latter B C
2. Marine Environment and Security for the European Area Integrated Project of the EC Framework Programme 6 MODIS Moderate Resolution Imaging Spectrometer MVT MERIS Validation Team NASA National Aeronautics and Space Administration N A Not Applicable netCDF Network Common Data Format NIR Near Infrared NOAA National Oceanic and Atmospheric Administration NPL National Physical Laboratory NRT Near Real Time ODESA Optical Data Processor of the European Space Agency OLCI Ocean and Land Colour Imager PDL Parameters Data List P Principal Investigator POLDER Polarization and Directionality of the Earth s Reflectances QA4EO a Quality Assurance framework for Earth Observation QWG Quality Working Group RAL Rutherford Appleton Laboratory RD Reference Document REASON NASA Research Education and Applications Solution Network project Rho Reflectance RMS Root Mean square RT Radiative Transfer S 3
3. Bidirectional Radiance Distribution Function CCSM Community Climate System Model CDOM Coloured Dissolved Organic Matter CEOS The Committee on Earth Observation Satellites CNES Centre National d Etudes Spatiales COM Coloured Organic Matter CRG Climate Research Group CZCS Coastal Zone Color Scanner DDS Diagnostic Data Set DIMITRI Database for Imaging Multi spectral Instruments and Tools for Radiometric Intercomparison DUE Data User Element of the ESA Earth Observation Envelope Programme II EC European Commission ECSS European Cooperation for Space Standardization EEA European Environment Agency EO Earth Observation EOSDIS Earth Observing System Data and Information System ESA European Space Agency ESDR Earth Science Data Records ESL Expert Support Laboratories EU European Union EUMETSAT European Organisation for the Exploitation of Meteorological Satellites FP6 EC Framework Programme 6 FP7 EC Fr
4. or this can be User defined 3 4 1 Sensor Selection Selections are then made for the validation site to be used the reference sensor and its processing version The available selections are determined from the DIMITRI database file The calibration sensors and processing versions can then be selected by moving the required sensor configuration into the right hand list highlighting the configuration and pressing gt gt Ref ME MAN ARG TN SUM ad Towards the intercalibration of EO medium resolution multi spectral imagers us oo ssue 1 ARGA N S DIMITRI MEREMSII Page 16 Software User Manual D14 Unwanted configurations can also be removed from the list be selecting the configuration and pressing the button 3 DIMITRI V2 0 SENSOR RECAL SETUP FOLDER AUTO REF REGION Amazon m 05 REF SENSOR AATSR_2nd_Reprocess lt AATSR_2nd_Reprocess lt b SENSOR CONFIGURATION SELECTION AATSR 2nd Reprocessing ATSR2 Reprocessing 2008 MERIS 2nd Reprocessing MERIS 3rd Reprocessing MODISA Collection 5 PARASOL Calibration 1 VEGETATION Calibration 1 ANGULAR MATCHING CRITERIA SZA 10 00 VZA 10 00 RAA 10 00 UPDATE AMC AMC 15 0000 ABSOLUTE MATCHING CRITERIA MIN TEMPORAL AND COVERAGE CRITERIA DAYOFFSET 3 CLOUD Joo REGION j8000 Figure 3 The DIMITRI Sensor Recalibration HMI 3 4 2 Angular Matching Criteria
5. uk eo esa int 8080 merci welcome do Registration http ats merci Requires ESA Cat 1 ATSR2 uk eo esa int 8080 merci welcome do Registration Requires ESA Cat 1 MERIS http merci srv eo esa int merci welcome do Registration MODIS Aqua http ladsweb nascom nasa gov Freely available PARASOL http polder cnes fr en index htm Requires registration VEGETATION 2 http www spot vegetation com index html Requires registration please note DIMITRI is currently set up to utilise VGT 2 products distributed by VITO these products are processed with an erroneous Earth Sin distance coefficient A correction provided by CNES is implemented within the DIMITRI VGT 2 processing Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium hi resolution multi spectral imagers a Issue 1 2 ARGA N S DIMITRI MEREMSII Page 10 Software User Manual D14 1 1 3 Development DIMITRI V1 0 was prototyped at ESTEC by Marc Bouvet ESA ESTEC DIMITRI V2 0 has been developed by ARGANS Ltd in collaboration with ESA through the MEREMSII project 1 1 4 DIMITRI License and code modification The DIMITRI software package is made freely available by ESA and ARGANS The Intellectual Property rights are held by ESA however modification of code is allowed The official version of original source code will be held at ARGANS and any modifications are done at the User s risk However Users are welcome t
6. Ref ME MAN ARG TN SUM E Towards the intercalibration of EO medium i i Date 08 05 2012 resolution multi spectral imagers Issue 1 2 AR GA N S DIMITRI MEREMSII Page 1 Software User Manual D14 Eme T ARGANS Towards the intercalibration of EO medium resolution multi spectral imagers DIMITRI amp MEREMSII DELIVERABLE DIMITRI Software User Manual SUM Deliverable Reference D14 Reference ME MAN ARG TN SUM Version 1 2 DATE 08 05 2012 ARGANS Towards the intercalibration of EO medium resolution multi spectral imagers DIMITRI MEREMSII Software User Manual D14 Ref ME MAN ARG TN SUM Date 08 05 2012 Issue 1 2 Page 2 Document Signature Table Name Function Company Signature Date Author Chris Kent Software Engineer ARGANS Verification satu Project Manager ARGANS Barker Approval Ine ESA Technical Officer ESA Bouvet Change record Issue Date Description Change pages 1 0 21 07 2011 First Version for DIMITRI V2 0 First version 1 1 25 10 2011 Text corrections 1 2 08 05 2012 Updated content and code appendix This is a public document available as part of the DIMITRI V2 0 package and for download on the ARGANS website www argans co uk DIMITRI For more information email dimitri argans co uk aa Towards the intercalibration of EO medium Refs ME MAN ARG TN SUM h r
7. The parameter AMC is used for geometrically matching satellite data from two different sensors The SZA VZA and RAA can be selected and used to update the AMC threshold to be utilised Any Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium S resolution multi spectral imagers uade Issue 1 2 ARGA N S DIMITRI MEREMSII Page 17 Software User Manual D14 corresponding satellite observations with an AMC value less than the threshold are selected and stored The AMC parameter is defined as 1 sy AMC SZA SZA F VZA VZA 7 RAA IRAA amp IT The User also has the option of selecting absolute angle criteria for the Viewing Zenith Angle VZA Viewing Azimuth Angle VAA Solar Zenith Angle SZA and Solar Azimuth Angles SAA 3 4 3 Temporal Spatial and Cloud Criteria The final parameter selections require the User to define the maximum allowed acquisition time difference in days between two satellite observations as well as the automated cloud percentage threshold and percentage of the ROI covered Any manual cloud screening results performed by the User override the corresponding automated cloud screening threshold checks For example if a product has been manually identified as clear non cloudy it will always be considered for matching with other satellite observations If no manual identification has been performed the product will only be used if its automated
8. Displays the DIMITRI cloud screening widget DHMI_CS_SETUP Displays the DIMITRI cloud screening setup widget DHMI CONFIGURATION Displays the DIMITRI configuration menu widget DHMI_INGEST Displays the DIMITRI widget for ingesting I1b data DHMI_PROCESS_1 Displays the DIMITRI widget for doublet extraction intercalibration recalibration and concatenation DHMI_PROCESS_2 Displays the DIMITRI widget for BRDF computation and VEGETATION simulation GET_DIMITRI_VISUALISATION_COLOURS Returns the DIMITRI colour array for the visualisation module RSR_LEGEND_UPDATE Updates the legend in the Relative Spectral Response widget RSR_PLOT_RESET Resets the DIMITRI Relative Spectral Response object graphics plot DIMITRI RSR COLOURS Returns the DIMITRI colour array for the Relative Spectral Response module PLOT DIMITRI SENSOR RSR Displays the DIMITRI Relative Spectral Response module DHMI RSR Displays the DIMITRI Relative Spectral Response setup module DIMITRI VISUALISATION ANGLES Returns the angular data for the DIMITRI visualisation module ET Towards the intercalibration of EO medium mE TUS ws resolution multi spectral imagers pate 09 0372012 Issue 1 2 ARGAN S DIMITRI MEREMSII Page 36 Software User Manual D14 DIMITRI VISUALISATION BRDF RHO Returns the BRDF data for the DIMITRI visualisation module DIMITRI VISUALISATION POLYNOMI
9. RSR functions with User specified wavelength ranges Users can also view the associated reflectance for each sensor as the spectra is convoluted to the sensor RSR and add new site spectra to the AUX DATA spectral response USER Sites folder These files must be semi colon separated and contain two columns containing the wavelength in nm and the RSR function between 0 and 1 For example a User generated spectra file would be of the form 354 000 2 304685712e 01 355 000 2 281964719e 01 356 000 2 259108275e 01 357 000 2 233817428e 01 Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium S resolution multi spectral imagers wed dps Issue 1 2 ARGA N S DIMITRI MEREMSII Page 24 Software User Manual D14 3 10 Options The Options module allows the definition of the DIMITRI configuration parameters including plot size colour table and RGB red green blue overlays These values are then loaded into the Sensor Recalibration and VEGETATION Simulation modules as default values 3 11 Help Opens this user manual 3 12 About Displays information regarding the version of DIMITRI 3 13 DIMITRI Output DIMITRI outputs a number of jpg plots and semi colon delimited CSV files A number of intermediate results are stored as IDL save files which are restored by DIMITRI when required see Section 5 3 These files can also be restored by users with full IDL licenses if require
10. This file is of the form num observations num parameters where the parameters are decimal time and the VEGETATION reflectance bands Amazon VEGETATION Calibration 1 UCT DAT This file is also of the form num observations num parameters however contains the parameters decimal time systematic uncertainty at each VEGETATION band and the random uncertainty and each VEGETATION band Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium S resolution multi spectral imagers Date 0a U97201 Issue 1 2 ARGAN S DIMITRI MEREMSII Page 30 Software User Manual D14 6 TROUBLESHOOTING FAQ Q Why doesn t DIMITRI run when I double click on the sav file A Please ensure you have correctly unzipped the DIMITRI zip archive and that you are choosing to open the save file with IDL Q The automated cloud screening values seem incorrect when viewing the product quicklooks why is this A Automated cloud screening of L1b data is very difficult The results are likely to vary depending on which validation site you are interested in clouds over ocean are easy to detect however over snow ice and salt lakes this is very difficult The performance of the algorithms is also linked to the available wavebands for each sensor ideally wavebands in the thermal wavelengths are needed for accurate cloud detection Further updates are planned for DIMITRI to also include statistical screening using the knowledg
11. a must see website containing many pages of useful IDL hints and tips e Mpfit pro and mpcurvefit pro Created by Craig B Markwardt http www physics wisc edu craigm idl fitting html e AATSR ATSR2 product reader routines Created by Dave Smith RAL http www aatsrops rl ac uk e FILEINFO pro Created by Liam Gumley http www gumley com Ref ME MAN ARG TN SUM EE Towards the intercalibration of EO medium S resolution multi spectral imagers undc Issue 1 2 ARGA N S DIMITRI MEREMSII Page 26 Software User Manual D14 5 DATA FILES The DIMITRI software package utilises a number of different data files These can be found in the AUX data Bin and Output folders and include Database files internal IDL binary files and Auxiliary data files each outlined below 5 1 DIMITRI Database file The semi colon delimited Database file contains information regarding all ingested L1b data products It is used by a number of DIMITRI functions such as doublet extraction and VEGETATION simulation The database file can be opened in a number of text editors including Microsoft excel to allow further analysis of the ingested product data A breakdown of the column headers is provided below DIMITRI DATE The calendar date the product was ingested into DIMITRI DD MMM YY 14 Dec 10 REGION ROI folder User placed product Uyuni SENSOR Product Sensor ATSR2 PROCESSING VERSION U
12. These files must be semi colon separated and contain 2 columns containing the wavelength in nm and the RSR function Please see the other site and sensor RSR files for examples Q Why do no new IDL windows open when DIMITRI generates plots stopped DIMITRI during processing why can t generate any new IDL plots A DIMITRI utilises IDL s Z buffer for generating plots this buffer does not have a viewable window but is stored in the machines memory Plots can therefore be saved without displaying multiple IDL windows If you stop DIMITRI during processing it will still be working in the Z buffer to return to your normal IDL graphics display type WINDOWS IDL set plot win LINUX IDL set plot x under Q Why is the super sensor time series so variable A The super sensor time series includes radiometrically corrected data from a number of different sensors These sensors all have different viewing and solar geometries over the same location dependant on your chi value chosen The greater the range of geometries the better the BRDF fit Q How do I find the details of created sites A The DIMITRI file DIMITRI SITE DATA txt is a semi colon delimited file containing the site information including coordinates and type Q How do I add a new sensor to DIMITRI A DIMITRI has be designed in a modules to allow modification by users however adding a new sensor is not a simple task this requires new product readers as
13. computed BRDF model ROUJEAN_BRDF Computes the ROUJEAN BRDF model for the super sensor time series DIMITRI_INTERFACE_ROUJEAN Interfaces between the DIMITRI HMI and the BRDF routines GET_GASEOUS_TRANSMISSION Reads the hyperspectral gaseous transmission data GET_OZONE_TRANSMISSION Reads the hyperspectral ozone transmission GET_WVAP_TRANSMISSION Reads the hyperspectral water vapour transmission COMPUTE_TRANSMISSION Computes the atmospheric transmission due to ozone water vapour and gaseous components d ARGANS Towards the intercalibration of EO medium resolution multi spectral imagers DIMITRI MEREMSII Software User Manual D14 Ref ME MAN ARG TN SUM Date 08 05 2012 Issue 1 2 Page 35 PLOT SIM VEGETATION Outputs plots of the simulated VEGETATION reflectance VEGETATION SIMULATION Utilises the computed BRDF models to simulate VEGETATION Observations over a validation site CHECK ROI COVERAGE Returns a flag indicating if an ROI is fully covered by a product COMPARE DIMITRI CNES SADE Compares DIMITRI and CNES reference dataset files CONVERT SADE TO DIMITRI Converts SADE files to DIMITRI arrays COMPUTE PARASOL VIEWING GEOMETRIES Computes the PARASOL viewing geometries CONVERT TIMESERIES TO SADE Converts DIMITRI time series array to SADE input format DIMITRI SADE INTERFACE ROUTINES P1 Allows sensor recalib
14. extraction routines and allows comparison of satellite data based on User defined cloud screening parameters as well as temporal spatial and geometric matching DIMITRI is a database containing the so called remote sensing TOA reflectance values from 2002 until the present day for ATSR2 ESA AATSR ESA MERIS ESA MODIS Aqua NASA PARASOL POLDER 3 CNES and VEGETATION CNES over eight predetermined validation sites see Table 1 DIMITRI is supplied with all L1b data pre loaded giving instant access to time series of data which totals more than 5 terabytes Additional data for other validation sites or more recent acquisitions can be ingested into DIMITRI to allow even greater temporal and spatial analysis 1 1 2 Purpose and principle of DIMITRI Within DIMITRI a reference sensor relates to a user selected sensor for all other sensors to be intercalibrated against A calibration sensor refers to a chosen sensor which will be compared against the reference sensor and recalibrated to its radiometric scale A number of calibration sensors can be compared against one reference sensor A further function of DIMITRI allows to intercalibrate all calibration sensors to the reference sensor and to generate radiometrically consistent TOA reflectances from all sensors This allows computation of a TOA Bidirectional Reflectance Distribution Function BRDF RD 14 over a selected validation site this is then utilised to simul
15. 4 DIMITRI ROUTINES iso ei Hae PN PARRA SEES PRRME INIHI IE NEP EMPMIDI PIENE 25 AA JROULIMENOCATION E DO D 25 42 Recompllatioficnsuctiositt tunisie iuis p Ee tei IR LUE cete EER 25 4 3 External ROULINGS 2 2 2 pere ia taba ta ba bo Ea ba ba ba Ee be be be be i ba Ge be bebe ba ba baba EE r i pe be bebe be be bebe e bbb ee E E GE 25 5 DATA Ec 26 51 DIMITRI Database fileu iiie ioo ero err err ert errore r eret E Ear rR beers r 26 52 Auxiliary Catan lS osos coat esestoteha nanna treo tole aea aola d eoe eoo 26 5 3 intemal SAV DinanieS Ca i Ren RCHD Rp ED RP DRRES RD hee RS RD RODEO D HORDED RD DSD HD RE ADRESSE HORE DEP eres 27 5 341 Extracted TOA Reflectance S niis usce edtic istas spud uento cenas suceda coena oaa eden 27 5 9 2 Enacted GOUDICLS otacactniacactcutacn c aqui don caQua acu sacu0dcu ea 27 9 3 3 E DIT qieidicicii inii ETI SE EERETISEEESETISE EISE TII AAAA 28 534 RECODO ON noris EIEII ITIITI EI I IITIN 28 T ARODUEANJBRDE a 29 5 3 6 E E E E A E E E E E E E E E indie E nthe nthe i 29 6 TROUBLESHOOTING FA Q Gus sci eve EM RU RS UR NM a a aa UE 30 APPENDIX A DIMITRI IDL ROUTINES 555553222 6545503 6a a 3E epa F2R Tasa E SFR aao aS IR eK a a Rete r sara Pda Ege 32 List of Figures Figure 1 Schematic diagram of the DIMITRI folder structure sees 12 Figure 2 The DIMITRI HMI GUI start page on windows top and Linux bottom 14 Figure 3 The DIMITRI Sensor R
16. AL Returns the polynomial data for the DIMITRI visualisation module DIMITRI VISUALISATION RECALIBRATION Returns the recalibration data for the DIMITRI visualisation module DIMITRI VISUALISATION REFLECTANCE Returns the reflectance data for the DIMITRI visualisation module DIMITRI VISUALISATION SUPER RHO Returns the super sensor data for the DIMITRI visualisation module DIMITRI VISUALISATION VEGETATION RHO Returns the VEGETATION reflectance data for the DIMITRI visualisation module DIMITRI VISUALISATION Displays the DIMITRI visualisation object graphics module DHMI VISU Displays the DIMITRI visualisation setup module DIMITRI DATABASE PLOTS WD Displays the DIMITRI acquisition plots module DIMITRI DATABASE STATS WD Displays the DIMITRI database statistics module DIMITRI DOWNLOAD WD Displays the DIMITRI download module DIMITRI NEW SITE WD Displays the DIMITRI new site module DIMITRI V2 Loads the DIMITRI HMI e ARGANS Towards the intercalibration of EO medium resolution multi spectral imagers DIMITRI MEREMSII Software User Manual D14 Ref ME MAN ARG TN SUM Date 08 05 2012 Issue 1 2 Page 37 End of Document
17. ARGANS DIMITRI MEREMSII i Page 14 Software User Manual D14 3 HMI FUNCTIONALITY The DIMITRI HMI has a series of primary and secondary functions all of which have been designed for command line usage as well as through the DIMITRI HMI A list of all the routines is included in Appendix 1 described here are the main functions of DIMITRI Figure 2 shows the DIMITRI HMI as it appears under Windows and Linux All HMI figures will be extracted from the Windows version of DIMITRI unless otherwise stated The following sections explain the functionalities the buttons of the HMI GUI 3 DIMITRI 2 0 Mgestbat Process Visualise mese Figure 2 The DIMITRI HMI GUI start page on windows top and Linux bottom Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium S resolution multi spectral imagers add dS Issue 1 2 ARGA N S DIMITRI MEREMSII Page 15 Software User Manual D14 3 1 Add L1b Data The Add L1b data button starts the ingestion interface for adding L1b data products to the DIMITRI database A specific site or sensor combination can be requested or DIMITRI will automatically search for any L1b data not currently ingested into the database All data is automatically cloud screened according to the sensors associated cloud screening algorithm see Section 3 6 Only products located in the correct Input folders can be ingested int
18. EGETATION_PRODUCT Ingests VEGETATION data into DIMITRI DIMITRI_INTERFACE_INGEST Interfaces between the DIMITRI HMI and ingestion routines COMPUTE_AMC Computes the angular AMC value COMPUTE_AMC_THRESHOLD Computes the AMC threshold given angular geometries EXTRACT_DOUBLETS Extracts observation doublets between reference and calibration sensor time series DIMITRI_INTERFACE_DOUBLET DIMITRI_POLYNOMIAL_FIT_FUNCTION INTERCALIBRATE_DOUBLETS Interfaces between the DIMITRI HMI and doublet extraction routine Returns the DIMITRI polynomial fit model Intercalibrate doublets between a reference sensor and calibration sensors DIMITRI_INTERFACE_INTERCALIBRATION Interfaces between the DIMITRI HMI and the intercalibration routine RECALIBRATE_DOUBLETS Recalibrates doublet data using the polynomial fit computed during intercalibration CONCATENATE_TOA_REFLECTANCE Concatenates recalibrated doublet data into a super sensor time series DIMITRI_INTERFACE_RECALIBRATION Interface between the DIMITRI HMI and the recalibration and concatenation routines ROUJEAN_BRDF_KERNEL_F2 Computes the F2 ROUJEAN BRDF kernel ROUJEAN_BRDF_KERNEL_F1 Computes the F1 ROUJEAN BRDF kernel ROUJEAN_BRDF_COEF Computes the ROUJEAN K coefficients ROUJEAN_BRDF_COMPUTE_RHO Computes the reflectance given BRDF model parameters and solar viewing angles ROUJEAN_BRDF_PLOTS Outputs plots of the
19. ENING The cloud screening interface routine GET AATSR AUX FILES Returns AATSR auxiliary data file names GET AATSR L1B REFLECTANCE Returns AATSR l1b reflectance GET AATSR LAT LON Returns AATSR geolocation ET Towards the intercalibration of EO medium Ref ME MAN ARG TN SUM S resolution multi spectral imagers Pate 090372012 Issue 1 2 ARGAN S DIMITRI MEREMSII Page 33 Software User Manual D14 GET_AATSR_QUICKLOOK Outputs AATSR quicklook GET_AATSR_VIEWING_GEOMETRIES Returns AATSR viewing geometries GET_AATSR_TIMESERIES_PLOTS Outputs AATSR time series plots INGEST_AATSR_PRODUCT Ingests AATSR products into DIMITRI GET_ATSR2_AUX_FILES Returns ATSR2 auxiliary data file names GET_ATSR2_L1B_REFLECTANCE Returns ATSR2 l1b reflectance GET_ATSR2_LAT_LON Returns ATSR2 geolocation GET_ATSR2_QUICKLOOK Outputs ATSR2 quicklook GET_ATSR2_VIEWING_GEOMETRIES Returns ATSR2 viewing geometries GET_ATSR2_TIMESERIES_PLOTS Outputs ATSR2 time series plots INGEST_ATSR2_PRODUCT Ingests ATSR2 products into DIMITRI GET_MERIS_AUX_FILES Returns MERIS auxiliary data file names GET MERIS ECMWF HUMIDITY Returns the MERIS ECMWF parameter GET MERIS ECMWF OZONE Returns the MERIS ECMWF parameter GET MERIS ECMWF PRESSURE Returns the MERIS ECMWF parameter GET MERIS ECMWF WIND Returns the MERIS ECMWF paramet
20. GETATION geometries and the BRDF models e correction for water vapour ozone and gaseous transmission e interpolation to hyperspectral wavelengths e re addition of atmospheric transmission e Convolution to the VEGETATION bands Please note the VGT 2 simulation methodology can be performed using only MERIS BRDF models in this instance the 1 6 micron band is extrapolated from the MERIS 900nm band and should therefore be disregarded 3 5 4 Simulated VEGETATION TOA reflectance uncertainty The final uncertainties output are the random and systematic uncertainties associated with the VEGETATION simulated observations Starting from the uncertainties associated to the super sensor observations we need to add the methodology uncertainties These are described in details in RD 21 The final uncertainties associated with the simulated VEGETATION observations are respective a systematic uncertainty of at least 8 and a random uncertainty of at least 11 30 The systematic and random uncertainties output can be higher RD 21 3 6 Cloud Screening An important development in DIMITRI V2 0 is the addition of automated cloud screening during ingestion of L1b satellite data The following algorithms have been implements for each sensor e Landsat ACCA RD 15 ATSR2 AATSR MODIS Aqua e Globcarbon MERIS RD 16 MERIS PARASOL e VGT operational VEGETATION The computed cloudiness of each product over the validation site is stored in
21. X TO WAVELENGTH Converts instrument Band ID to DIMITRI wavelength CONVERT WAVELENGTH TO DINDEX Converts DIMITRI wavelength to Band ID GET DIMITRI CONFIGURATION TEMPLATE Returns the DIMITRI configuration template GET DIMITRI CONFIGURATION Reads the DIMITRI configuration file GET SENSOR BAND INDEX Returns the band ID for a given sensor and DIMITRI band GET SITE COORDINATES GET SITE TYPE Returns the coordinates of a validation site Returns the site type of a validation site SAVE DIMITRI CONFIGURATION Saves the DIMITRI configuration file SENSOR BAND INFO Returns the number of bands for a sensor SENSOR PIXEL SIZE Returns the pixel resolution for a sensor GET DIMITRI TEMPLATE Returns the DIMITRI database template SAVE DIMITRI DATABASE Saves the DIMITRI database UPDATE DIMITRI DATABASE Updates the DIMITRI database file DIMITRI CLOUD SCREENING BRIGHT RHO Computes if a pixel is bright based on the operational MERIS bright test CS BAND INFO Returns the DIMITRI band ID s for a given cloud screening algorithm CLOUD MODULE GLOBCARBON The GLOBCARBON cloud screening algorithm CLOUD MODULE GLOBCARBON P The GLOBCARBON cloud screening algorithm modified for PARASOL bands CLOUD MODULE LCCA The Landsat Cloud Cover Algorithm LCCA CLOUD MODULE VGT The operational VEGETATION cloud screening algorithm DIMITRI CLOUD SCRE
22. alibration Status of the AATSR Reflectance Channels MERIS AATSR workhop 2008 proceedings RD 5 Miesch C Cabot F Briottet X Henry P Assimilation method to derive spectral ground reflectance of desert sites from satellite datasets Remote sensing of environment 2003 vol 87 no2 3 pp 359 370 RD 6 P Henry and A Meygret Calibration of HRVIR and VEGETATION cameras on SPOT4 Adv Space Res vol 28 pp 49 58 2001 RD 7 Proba V Mission Requirement Document ESA document RD 8 Landnet sites http calval cr usgs gov sites catalog ceos sites ph RD 9 GEO http www earthobservations org about geo shtml RD 10 GSICS http www star nesdis noaa gov smcd spb calibration icvs GSICS RD 11 http www star nesdis noaa gov smcd spb calibration icvs GSICS LLvisiblelnfrared ph RD 12 http www eumetsat int Home Main Access to Data IntercalibrationServices index htm lzen RD 13 European Cooperation for Space Standardisation ECSS E ST40C RD 14 Roujean J L Leroy M and Deschamps P Y 1992 A bidirectional reflectance model of the Earth s surface for the correction of remote sensing data Journal of Geophysical Research 97 DIS 20 455 20 468 RD 15 Irish R R Landsat 7 Automatic Cloud Cover Assessment See http landsathandbook gsfc nasa gov handbook pdfs ACCA SPIE paper pdf RD 16 Plummer S E 2008 The GLOBCARBON Cloud Detection System for the Along Track Scanning Radiometer ATSR Sensor Series IEEE Transactions on Geoscience and Rem
23. amework Programme 7 FR Final Report FVR Full Validation Report GIS Geographic Information System GMES Global Monitoring for Environment and Security GOCI Geostationary Ocean Color Imager GOMOS Global Ozone Monitoring by Occultation of Stars GOOS Global Ocean Observing System GSC GMES Space Component GSC CQC GMES Space Component Coordinated Quality Control Hermes Data portal for Ocean Colour Data Users http hermes acri fr HMI Human Machine Interface IOPs Inherent Optical Properties IPCC Intergovernmental Panel of Climate Change Ref ME MAN ARG TN SUM E Towards the intercalibration of EO medium i Date 08 05 2012 1 resolution multi spectral imagers I 2 12 ARGANS DIMITRI MEREMSII iud Software User Manual D14 IVOS Infrared and Visible Optical Sensors Subgroup of WGCV LUT Look Up Table MEaSUREs Making Earth Science Data Records for Use in Research Environments MEREMSII MEdium REsolution Multi Spectral Imagers Intercalibration MERIS Medium Resolution Imaging Spectrometer MERSEA
24. are the decimal time and the error between the actual radiometric differences and the polynomial approximation of the differences e JCOEF SITE SENSOR PROCVER REF SENSOR PROCVER BAND dat This binary contains the 3 polynomial coefficients for the specified calibration sensor reference sensor and band e JUCRT SENSOR PROCVER REF SENSOR PROCVER BAND dat This binary contains the uncertainty values associated to the polynomial coefficients and contains the systematic error the random error and a flag indicating that the random error has been taken from the polynomial fit rather than the original sensor time series 5 3 4 Recalibration A number of internal binaries are output during the recalibration and super sensor time series generation These include e RECAL SITE SENSOR PROCVER REF SENSOR PROCVER dat RECAL REF SITE SENSOR PROCVER dat These SAV files are of the form num parameters num observations where the parameters are as those described in Section 5 3 1 and the observations are the recalibrated and reference sensor reflectances e SSEN SITE SENSOR PROCVER BAND DAT Ref ME MAN ARG TN SUM E Towards the intercalibration of EO medium S resolution multi spectral imagers Date 00 0537201 Issue 1 2 AR GAN S DIMITRI MEREMSII Page 29 Software User Manual D14 These files contain the Super Sensor observations generated for the specific band using the SENSOR value as the reference sensor T
25. ate VEGETATION TOA reflectance using the VEGETATION products viewing and solar geometries All stages of DIMITRI intercalibration BRDF modelling and TOA simulation include an attempt to propagate uncertainties providing detailed information on the processing stages performed The computed systematic and random uncertainties are provided in the DIMITRI outputs RD 21 The standard DIMITRI TOA Reflectance at any given wavelength is defined as L P F d2 Cos 3 ARGANS Towards the intercalibration of EO medium resolution multi spectral imagers DIMITRI MEREMSII Software User Manual D14 Ref ME MAN ARG TN SUM Date 08 05 2012 Issue 1 2 Page 9 Where p reflectance L TOA Radiance Fo Solar Irradiance Flux Solar Zenith Angle and d a correction factor for the Earth Sun distance Table 1 Geolocation values for the 8 predefined DIMITRI validation sites Site Name North Lat N South Lat N West Lon E East Lon E Uyuni Salt Lake 20 00 20 16 68 05 67 45 Libya 4 29 05 28 05 22 89 23 89 Dome C 74 90 75 30 122 90 123 90 Tuz Golu 38 80 38 70 33 25 33 40 BOUSSOLE 43 45 43 25 7 80 8 00 Amazon Forest 1 33 1 00 57 00 56 50 SPG 31 00 31 50 137 5 137 0 SIO 30 00 30 50 80 50 80 00 Table 2 DIMITRI Sensor data location information Sensor Data Location Access http ats merci Requires ESA Cat 1 AATSR s
26. athe 16 3 4 3 Jemporal Spatial and Cloud Criteria ss ioo ise cuo ettet eoe deese ne eee etae nu eoe 17 344 FING SteDS cuc tete tut aetans tote tetutetetatutetutetut edel tabedeledietabedebadededierabsrebetetes 17 34 5 S per Sensor observation ncertainty 2 c ot con nantun haenar nego deno gena naaien 18 35 VEGETATION Simulation eicere Eee Eee eoo Eee ERE sagesaee sadebaes EEEE 18 3 5 1 Super Sensor Selection sessi eese nnne n ener ener enne nennen 19 35 2 BRDFEand VEGETATION Criteria s iioii ei andes 19 3 5 3 SEIDOLSIEDS UU Po ee or eee eee I EIS 20 3 5 4 Simulated VEGETATION TOA reflectance uncertainty esses 21 3 6 Cloud Screening sisi EE a PR ENEEEIM NS S MP IS INDE eas SR nan SRL n SOS EGER 21 34 MisWalisdblOITs memi sect Sete eee ce moie Epio See ace See Gece See ace See ce eee Gece Eu ie Cece Soe cote pet eee epe ace teed 22 B28 Database statistiGs tette tette bebe btt tbt tbe 23 3 9 Relative Spectral Response viewer essen nennen nnne nnns nna 23 3 10 loei cm 24 3 11 HEID ecr e E E E E EE EEE 24 3 12 Yee mmm 24 3 13 DIMITRIAQUUDUL 5 555 5 58 8018 R2 SR SEER RASA REIR SER REER RR ERR RER RR ER RR RR ERROR AER MRER REC RRN RR SC PRP P PRSE E ER ER 24 aa Towards the intercalibration of EO medium Beh ME MAN ARG TN SUM i Date 08 05 2012 resolution multi spectral imagers Issue 1 2 ARGAN S DIMITRI MEREMSII e age 4 Software User Manual D14
27. ation where both MERIS and AATSR have been separately used as reference sensors Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium S resolution multi spectral imagers eer Issue 1 2 ARGA N S DIMITRI MEREMSII Page 19 Software User Manual D14 3 5 1 Super Sensor Selection The output folders containing the Sensor Recalibration results can be found using the and buttons If the required data is found in the folder the processing configurations become available for selection in the corresponding selection boxes If no data is available the value n a is presented The routine then determines if any VEGETATION data is available corresponding to the site provided in the selected MERIS folder All outputs are placed in this MERIS output folder 3 5 2 BRDF and VEGETATION Criteria You are then able to select if automatic BRDF plots should be generated note for large time series this can take a long time the BRDF binning period how many days should cover one bin period the acquisition limit how many super sensor observations are required to make the BRDF model valid for each bin and the cloud and ROI thresholds for the VEGETATION sensor data recommended to be the same as the doublet extraction values i ARGANS Towards the intercalibration of EO medium resolution multi spectral imagers DIMITRI MEREMSII Software User Manual D14 Ref ME MAN ARG TN SUM Da
28. ation HMI module 4 Band centre index This file contains the relative indexes for each sensor band against the defined DIMITRI wavelengths It is used throughout a number of DIMITRI functions and should not be modified There are 30 defined DIMITRI bands covering the wavelength range 400 12000nm for each sensor the band index is linked with the corresponding DIMITRI band to allow comparison against other sensors which also have bands matching that DIMITRI band This comparison is performed internally and allows specification of comparing different bands for example MODISA has two band setting land and ocean which are defined in the Band centre index auxiliary file 5 3 Internal SAV binaries 5 3 1 Extracted TOA Reflectances For each sensor over each site an internal SAV stores the extracted time series data the syntax for these files are SENSOR_TOA_REF dat and have dimensions num_of_parameters num_of_observations num_directions Where the parameters are decimal_time VZA VAA SZA SAA Ozone Pressure Humidity Zonal_WIND Meridional_WIND Water Vapour Mean RHO Band O Band n STDEV RHO Band O Band n ndicates mean value and standard deviation num directions are the number of different views from the sensor e g 1 for MERIS 2 for AATSR Where more than one viewing direction is available the observations are ordered as obs 1 dir 1 0bs 1 dir n obs 2 dir 1 NB for MODIS Aqua the reflectance
29. bands are stored with indexes 0 14 as the 1Km bands 15 19 as the 500m bands and 20 21 as the 250m bands 5 3 2 Extracted doublets The extracted doublet internal binaries are of the syntax ED_SITE_SENSOR1_PROCVER1_SENSOR2_PROCVER2 dat and contain the extracted doublets for Ref ME MAN ARG TN SUM ER Towards the intercalibration of EO medium S resolution multi spectral imagers dabas Issue 1 2 ARGA N S DIMITRI MEREMSII Page 28 Software User Manual D14 SENSOR1 when it has been extracted against SENSOR2 given the User selected matching parameters These files are of the form num of parameters num of observations Where the parameters are the same as those listed in section 5 3 1 but with the added parameter s of Number of pixels automated cloud screening manual cloud screening tmatching type AMC The number of observations is dependent on the matching criteria selected 5 3 3 Intercalibration A number of internal binaries are created during the intercalibration of sensor doublet information These include e JCDIF SITE SENSOR PROCVER REF SENSOR PROCVER BAND dat This has the dimensions num observations num params where the parameters are decimal time reflectance bias to reference sensor and the polynomial bias to the reference sensor e JCERR SITE SENSOR PROCVER REF SENSOR PROCVER BAND dat This has the dimensions num observations num params where the parameters
30. cloud screening percentage is below the cloud threshold set by the User If the input ROI percentage coverage is less than 10096 the expected number of pixels for the selected sensor is compared against the actual number for each observation If however the values is set as 10096 a different test is used only products in which the four corners of the defined ROI are covered are kept this utilises the ROI COVER flag within the DIMITRI database Any satellite observations from two separate sensors or configurations which are within the defined temporal and spatial matching criteria including cloud coverage are known as doublets 3 4 4 Final Steps Once all parameters have been selected press the START button to begin processing Sensor Recalibration comprises of doublet matching between the reference sensor and all calibration sensors generation of a polynomial to fit the temporal variations of the radiometric difference between these sensors and recalibration of all calibration sensor data to the radiometric scale of the reference sensor These steps are known as doublet extraction intercalibration and recalibration and output a so called super sensor time series of sensor observations over the same location with different viewing and solar geometries During intercalibration a polynomial fit is applied to the temporal differences between the calibration and reference sensor for a given band The following mod
31. d by typing IDL gt restore sav filename Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium hi resolution multi spectral imagers i a Issue 1 2 ARGA N S DIMITRI MEREMSII Page 25 Software User Manual D14 4 DIMITRI ROUTINES 4 1 Routine location All DIMITRI routines can be found in the source folder All routines developed by ESA and ARGANS contain a code header providing information on what the routine does how it is called what are the outputs and the modification history If any routines are updated by Users please ensure the code header is updated accordingly 4 2 Recompilation DIMITRI has been developed in IDL V7 1 and is open to users with full IDL licenses to modify the code and develop it further note ESA and ARGANS are not responsible for any damage this may cause to your computer All routines can be found within the Source folder including a shell script called compile dimitri which recompiles all routines A compiled SAV file can then be generated by typing IDL compile dimitri The created save file can then be used by the IDL runtime version which does not require an IDL license 4 3 External Routines The DIMITRI software package has been developed by ESA and ARGANS using IDL However some functions have been included which were developed by other people These include e FSC field pro Created by David Fanning http www idlcoyote com
32. download DIMITRI Email address will only be used to provide notifications of updates to DIMITRI and to obtain User feedback User information will only be stored at ARGANS Ltd on behalf of ESA and will not be passed on to any other organisation All Users must comply with the following Service Level Agreement SLA 1 Acknowledgement to ESA and ARGANS is required for any presentations or publications however written permission is not required 2 The DIMITRI package cannot be distributed by anyone other than ESA and ARGANS 3 Any modifications to the DIMITRI code are appropriately detailed in the code header information and are performed under the responsibility of the User 4 ESA and ARGANS are not responsible for any damage to computer systems as a result of using the DIMITRI software package Analysis and results using DIMITRI will be uploaded to the CalVal Portal http calvalportal ceos org Registration on this site is necessary to see these data 2 2 Installation The DIMITRI software package is downloaded as a compressed archive To install DIMITRI V2 0 unzip the folder into the desired installation location It is recommended that either 7zip http www 7 zip org or Filzip http www filzip com are used for extraction of the zip archive Please note it is recommended to install DIMITRI with administrator or root privileges This is to allow the creation of all required files and folders Linux Users can install t
33. e that validation sites by definition should be generally radiometrically homogeneous Q My computer crashed whilst ingesting new products is there a backup of the DIMITRI database file A Yes the latest copy of the DIMITRI database can be found in the folder Bin DB backup Q Why can t I access the uncertainty data through the visualisation module A The visualisation module is designed to allow quick visual inspection of the general DIMITRI outputs For uncertainty analysis please use the output CSV files Q Why are some MODIS bands showing a strong cosine dependence A A number of the ocean bands available for MODIS Aqua can saturate over land DIMITRI extracts all data no matter which validation site is used however only intercalibrates recalibrates the Land bands for Land validation sites Q What is the best tool for reading utilising the output CSV files A The semi colon delimited output CSV files can be opened in most text editing software the data can also be read into Microsoft Excel for further analysis e g plots and statistics Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium S resolution multi spectral imagers ee ee Issue 1 2 ARGA N S DIMITRI MEREMSII Page 31 Software User Manual D14 Q How do I add extra RSR site spectra A User generated site RSR spectra will be automatically read by DIMITRI if placed in the AUX DATA spectral response USER Sites folder
34. ecalibration HMI sessi 16 Figure 4 Process 2 setup module HMI seesssssseesseseseee een nna 20 Figure 5 Manual Cloud Screening HMI esses nini 22 Figure 6 Example of the VISU Module sess nnnnn nnns nene nnn nnns 23 List of Tables Table 1 Geolocation values for the 8 predefined DIMITRI validation sites eeeeeeeessss 9 Table 2 DIMITRI Sensor data location information cesses eene ener enne 9 Ref ME MAN ARG TN SUM E Towards the intercalibration of EO medium i Date 08 05 2012 resolution multi spectral imagers I 1 2 ARGANS DIMITRI MEREMSII ind Software User Manual D14 Acronym List AATSR Advanced Along Track Scanning Radiometer AC Atmospheric Correction AD Applicable Document ADEOS Advanced Earth Observation Satellite AIMES Analysis Integration and Modelling of the Earth System ATBD Algorithm Theoretical Baseline Document AVHRR Advanced Very High Resolution Radiometer BEAM Basic ERS and Envisat A ATSR and MERIS Toolbox BOUSSOLE Bou e pour l acquisition de S ries Optiques Long Terme BPAC Bright Pixel Atmospheric Correction BRDF
35. el is used for the polynomial fit y Ax Bx C Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium hi resolution multi spectral imagers bad Issue 1 2 ARGA N S DIMITRI MEREMSII Page 18 Software User Manual D14 The coefficients A B and C are computed using a least squares regression fit and are output in the ICAL CSV files No weighting is applied to the polynomial model computation In addition to the polynomial coefficients the covariance matrix is also output in the ICAL CSV file The matrix is a 3x3 element array with columns A B and C and rows 1 2 and 3 the output values can therefore be replaced into a matrix given the values header information For example CVAR A1 refers to the covariance value matrix 1 1 starting from 1 3 4 5 Super Sensor observation uncertainty The final output of the process previously described is the time series of super sensor observations All sensors observations have been rescaled to the reference sensor radiometric scale This does not mean that the super sensor observations have the same systematic uncertainty than the reference sensor because the methodology has introduced additional uncertainties both random and systematic To evaluate these uncertainties it is assumed in DIMITRI V2 0 that all standard satellite TOA reflectance values have both systematic and random uncertainties of respectively 396 and 3 36 see RD 21 In addition t
36. er GET MERIS L1B DETECTOR INDEX Returns MERIS detector index GET MERIS L1B FO Returns MERIS solar irradiance values GET MERIS L1B RADIANCE SF Returns MERIS radiance scaling factors GET MERIS L1B RADIANCE Returns MERIS radiance values GET MERIS LAT LON GET MERIS QUICKLOOK Returns MERIS geolocation Outputs a MERIS quicklook GET MERIS SOLAR FLUX RR GET MERIS VIEWING GEOMETRIES Returns MERIS solar flux values Returns MERIS viewing geometry GET MERIS TIMESERIES PLOTS Outputs MERIS time series plots INGEST MERIS PRODUCT Ingests MERIS products into DIMITRI CONVERT EMISSIVE TO BTEMP Converts MODISA emissive data to brightness temperature GET MODISA AUX FILES Returns MODISA auxiliary data filenames GET MODISA DATE INFO Returns MODISA date information GET MODISA VIEWING GEOMETRIES Returns MODISA viewing geometries GET MODISA L1B EMISSIVE Returns MODISA emissive data values GET MODISA L1B EMISSIVE LINUX Returns MODISA emissive data values for Linux GET MODISA L1B REFLECTANCE Returns MODISA reflectance values GET MODISA L1B RADIANCE Returns MODISA radiance values GET MODISA L1B REFLECTANCE LINUX Returns MODISA reflectance values on Linux GET MODISA L1B RADIANCE LINUX Returns MODISA radiance values on Linux GET MODISA LAT LON Returns MODISA geolocation GET MODISA QUICKLOOK Outputs MODISA quic
37. esolution multi spectral imagers Date 00057201 ARGANS DIMITRI MEREMSII id dd Software User Manual D14 Table of Content 1 INTRODUCTION 8 LL DIMITR ES 8 11 1 DIMITRI Software Package rosnie neinei a aoia r Ea ainn 8 T1212 jPurposeund principleor DIMITR re AEEA eas 8 ENSE MEET 10 1 1 4 DIMITRI License and code modification cesses eese 10 1 2 SYSTEM REGUIFEIMENTSS sees cos eap et n RRRERERERESERRA ER E PASA E Esa a EE Pea ARE S 10 2 SETUP e ON 11 241 Whereto acquire DIMITRI ssisieisaesoat es sedo opo oen esee aiu po teae besoin in Ee feda a aaan aaa a 11 2 2 IhstallatiOIzs m qd meo eee ER eee REDDE Robe Robo bep bt ob beo tp tton 11 DES A TAAL CT T ETT TTC 11 2 4 Folder Structure ee pere re Pre ere EEEE EEEE eee aen een aene aane aen eee EEE een ut 12 2 4 1 VEGETATION Folder Structures i cadatad edad ta dedu dada data adatta datae dada dabat 13 3 HIM PFUNSCHONADDTY S ia I EVA KO RR ORDRE DUE OA EUN SERIEN C PDA ar P RR FE 14 3 1 AddlIb RE eee ee m 15 3 2 INeW SIte sisi yrs Six Y YE Y EYES EX RR ER NER EY ERR RR YRR ER RR E YER ERR sabes RR ENSE ERR VERE MR E YRR RN XRR T ERE YR VERE 15 BS DEPO aeris 15 34 Sensor RECALIDFAt OM se E E EEE E EEE Ay 15 241 SCNSOMSE CCHON vixinsoseautiait tastiest iauteenddeadientdandiedteaetvastnastyastdastyastdastyastoastvastnasteanaves 15 34 2 Angular Matching Criteria aeaa etude iod Este SESSEL MEL aa beet tact l
38. gh the visualisation module and saved as a CSV file e TOA RHO Plots Sensor doublet observations Super Sensor and simulated VEGETATION time series e RECAL RHO Plots the recalibrated time series data for each calibration sensor used e RHO Bias Plots the reflectance bias difference to reference sensor for each calibration sensor e POLY Bias Plots the polynomial bias difference to reference sensor for each calibration sensor e VZA VAA SZA SAM Plots the doublet angular information e AMC Plots the computed AMC values between observations Ref ME MAN ARG TN SUM EE Towards the intercalibration of EO medium hi resolution multi spectral imagers Date 00057201 Issue 1 2 ARGA N S DIMITRI MEREMSII Page 23 Software User Manual D14 3 DIMITRI 2 0 VISU REF SENSOR MERIS 2nd_Reprocessing t REF MER 2nd REF ATS 2nd o o Reflectance o gt Figure 6 Example of the VISU Module 3 8 Database statistics The Database Stats button starts the database statistics module which provides up to date statistics on the current DIMITRI database file Further to this plots can also be viewed highlighting the number of products ingested for each product over each site Plots can be saved in the JPG PNG and CSV formats 3 9 Relative Spectral Response viewer The Relative Spectral Response RSR viewer allows generation and visualisation of plots of sensor
39. he DIMITRI package by typing tar xf DIMITRI V2 0 tar gz 2 3 Quickstart Following extraction DIMITRI is now ready to be utilised this can be achieved by e On Windows Double clicking the DIMITRI V2 0 sav file or running IDL runtime and selecting the file e On Linux Typing idl vm DIMITRI_V2 0 sav Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium S resolution multi spectral imagers uade Issue 1 2 ARGAN S DIMITRI MEREMSII Page 12 Software User Manual D14 These commands will load the DIMITRI Human Machine Interface HMI See Section 3 for details on the functionality of the HMI If using a full IDL license all DIMITRI routines can be used through the command line This requires compilation of the routines which can be performed by running the compile DIMITRI IDL script located in the Source folder IDL gt cd Source IDL gt compile_DIMITRI 2 4 Folder Structure The installed DIMITRI directory has the following folder structure The Input folder will contain all of the Level 1 data for all sensors across all sites please note for distribution and size restrictions only the quicklooks and stored TOA reflectance data are available for download e All site folders will be indentified through the syntax Site e g Site Uyuni and allows User defined sites to be added to DIMITRI It is recommended that all new sites are generated
40. he doublet matching process has then been estimated to introduce a 396 systematic uncertainty with respect to the reference sensor and a 396 3c random uncertainty see reference RD 1 and RD 2 The super sensor observation radiometric systematic uncertainty is thus 396 with respect to the reference sensor radiometric scale The combined random uncertainty associated to the super sensor observations is the quadratic sum of the random uncertainties of the calibration sensor reference sensor methodology uncertainty This amounts to 5 296 36 The consistency of this figure is checked against the value of the RMSE of the polynomial fit to the radiometric differences between the calibration and reference sensor doublets If the polynomial s RMSE fit is greater than the combined random uncertainty then this value becomes the random uncertainty for the calibration sensor time series the super sensor observations can therefore have different uncertainties for each day and wavelength depending on which calibration sensor data has been used The propagated uncertainties are output in both internal IDL save files and semi colon delimited CSV files Please note updated uncertainty propagation is foreseen for future releases of DIMITRI 3 5 VEGETATION Simulation VGT Simulation starts the setup modules for BRDF model generation and simulation of VEGETATION TOA reflectance This process will only work with outputs from Sensor Recalibr
41. hese files have the form num_parameters num_observations where the parameters are decimal time VZA VAA SZA SAA Ozone Pressure Humidity Zonal WIND Meridional WIND Water Vapour Mean RHO Band Systematic uncertainty Random uncertainty Poly err flag and a flag indicating which sensor configuration the observation originates 5 3 5 ROUJEAN BRDF The following internal SAV files are generated during the BRDF computation ROUJEAN ER SITE REF SENSOR PROCVER dat ROUJEAN K1 SITE REF SENSOR PROCVER dat ROUJEAN K2 SITE REF SENSOR PROCVER dat ROUJEAN K3 SITE REF SENSOR PROCVER dat These files are al of the form num params number bins where the parameters are decimal time the number of sensor observation within the bin and the corresponding ROUJEAN coefficient for all bands The number of bins will depend on the bin size selected for processing ROUJEAN UC SITE REF SENSOR PROCVER dat This file is of the form num params number bins where the parameters are time num sensor observations VZA min VZA max SZA min SZA max RAA min RAA max poly err flag the systematic uncertainty for each band and the random uncertainty for each band 5 3 6 VGT Simulation The VEGETATION simulation module outputs the following IDL SAV binaries Amazon_VEGETATION_Calibration_1 DAT This file has the same format as the extracted sensor L1b file in Section 5 3 1 Amazon_VEGETATION_Calibration_1_SIM DAT
42. klook GET MODISA QUICKLOOK LINUX Outputs MODISA quicklook on Linux GET MODISA TIMESERIES PLOTS Outputs MODISA time series plots INGEST MODISA PRODUCT Ingests MODISA products into DIMITRI ET Towards the intercalibration of EO medium Ref ME MAN ARG TN SUM S resolution multi spectral imagers pate 09 0372012 Issue 1 2 ARGAN S DIMITRI MEREMSII Page 34 Software User Manual D14 INGEST_MODISA_PRODUCT_LINUX Ingests MODISA products into DIMITRI on Linux GET_PARASOL_L1B_PIXEL_STRUCTURE Returns PARASOL pixel structure GET_PARASOL_L1B_HEADER Returns PARASOL header information GET_PARASOL_L1B_DATA Returns PARASOL pixel data GET_PARASOL_QUICKLOOK Outputs PARASOL quicklook GET_PARASOL_TIMESERIES_PLOTS Outputs PARASOL time series plots INGEST_PARASOL_PRODUCT Ingests PARASOL products into DIMITRI GET_VEGETATION_HEADER_INFO Returns VEGETATION header information GET_VEGETATION_L1B_REFLECTANCE Returns VEGETATION reflectance data GET_VEGETATION_LAT_LON Returns VEGETATION geolocation GET_VEGETATION_OZONE Returns VEGETATION ozone values GET_VEGETATION_WVAP Returns VEGETATION water vapour values GET_VEGETATION_VIEWING_GEOMETRIES Returns VEGETATION viewing geometries GET_VEGETATION_QUICKLOOK Outputs VEGETATION quicklook GET_VEGETATION_TIMESERIES_PLOTS Outputs VEGETATION time series plots INGEST_V
43. o DIMITRI If limited hard drive space is available products can be removed from the input folder following ingestion however the quicklook should remain for the manual cloud screening module Ingestion can take several minutes for each product depending on system performance and which sensor as well as if child or parent products are used 3 2 New Site New Site starts the new site creation tool A new site can be added to the DIMITRI database through selection of Name Type and basic geolocation coordinates North South East and West Once selected the input folders for the site are automatically generated there is no need to create any folders manually Following creation products can then be placed in the correct input folders for ingestion into DIMITRI 3 3 Data Download The Data Download module provides quick links to the associated sensor websites for retrieval of products Please note that not all sensor data is freely available and some data access requires usernames and passwords 3 4 Sensor Recalibration Sensor Recalibration starts the setup module for intercomparison between a reference sensor and a number of calibration sensors This intercomparison is based on the identification of acquisitions made between the two sensors at similar time and under similar geometries The output folder can be left as auto to generate an automatic folder name e g SIO 20110426 REF MERIS 2nd Reprocessing
44. o contact the DIMITRI developers at dimitri argans co uk with suggestions for improvement and if verified and approved by ESA these may be included in later versions of DIMITRI See section 4 for more information on DIMTIRI routines and compilation 1 2 System Requirements A full IDL license is NOT required for DIMITRI V2 0 the freely available IDL Virtual Machine available at http www ittvis com ProductsServices IDL IDLModules IDLVirtualMachine aspx will allow use of the pre compiled DIMITRI package and use of the full functionalities accessible from the HMI DIMITRI has been developed to be compatible on both Linux and Windows based systems however MAC compatibility cannot be guaranteed DIMITRI has been developed for use with IDL 7 1 or higher the minimum requirements required for IDL 7 1 are therefore the minimum requirements for running DIMITRI A full IDL license http www ittvis com will allow command line usage modification of routines and recompilation of the software package Ref ME MAN ARG TN SUM ER Towards the intercalibration of EO medium S resolution multi spectral imagers buddies Issue 1 2 ARGA N S DIMITRI MEREMSII Page 11 Software User Manual D14 2 SETUP 24 Where to acquire DIMITRI The DIMITRI software package is made available by ESA on the ARGANS website www argans co uk dimitri This webpage provides information on the DIMITRI package and allows registration to
45. ote Sensing 46 6 1718 1727 RD 17 Santer R 2010 OLCI Pixel Classification ATBD S3 L2 SD 03 CO1 LISE ATBD Version 2 0 RD 18 D L Smith and C T Mutlow and C R N Rao 2002 Calibration Monitoring of the Visible and Near Infrared Channels of Along Track Scanning Radiometer 2 ATSR 2 using Stable Terrestrial Sites Applied Optics 41 no 3 515 523 RD 19 Rao C R N and J Chen 1995 Inter satellite calibration linkages for the visible and near infrared channels of the Advanced Very High Resolution Radiometer on NOAA 7 9 and 11 spacecraft International Journal of Remote Sensing 16 1931 1942 RD 20 Smith D 2006 Update on AATSR Visible Channel Long Term Trends See http envisat esa int pub ESA DOC ENVISAT AATSR Visible Channel Update TN 0552 pdf RD 21 Bouvet M 2011 Simulating VGT from Super Sensor observations version 3 technical note Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium hi resolution multi spectral imagers uade Issue 1 2 ARGA N S DIMITRI MEREMSII Page 8 Software User Manual D14 1 INTRODUCTION 1 4 DIMITRI 1 1 1 DIMITRI Software Package The DIMITRI software package contains a suite of IDL routines for the intercomparison of Top Of Atmosphere TOA radiance and reflectance values within the 400nm 4um wavelength range this is generally known as Level 1b Earth Observation EO satellite data The package includes product reader and data
46. ration using CEOS IVOS dataset files DIMITRI SADE INTERFACE ROUTINES P2 Allows VGT 2 simulation using CEOS IVOS dataset files DIMITRI INTERFACE EXTRACT TOA NCDF Interfaces with the NCDF extraction routines GET DIMITRI BAND NAME TEMPLATE Defines the DIMITRI band name template GET DIMITRI EXTRACT NCDF DATA STRUCTURE Returns the NCDF data structure GET DIMITRI EXTRACT TOA NCDF NAMES Defines the NCDF variable names GET DIMITRI VISUALISATION COLOURS BRDF Defines the colour codes for the visualisation module GET PARASOL DIRECTION GEOMETRY Returns the PARASOL direction geometry GET PARASOL VIEWING GEOMETRIES Returns the PARASOL viewing geometries GET SENSOR BAND NAME Returns the sensor band name string GET VGT CORRECTION FACTOR READ DIMITRI EXTRACT TOA NCDF READ SADE PRODFILE TEXT Returns the VITO VGT 2 reflectance correction factor Reads the DIMITRI NCDF file Reads a CEOS IVOS reference dataset product file READ SADE TXT REMOVE BAD SADE SENSOR DIRECTION INFO Reads a SADE input format file Removes manually screened products from the CEOS IVOS dataset Returns the number of directions available for each sensor UPDATE_DIMITRI_EXTRACT_TOA_NCDF Updates the DMITRI NCDF files VGT_CORRECTION_TEMPLATE Defines the VGT 2 correction file template WRITE_DIMITRI_EXTRACT_TOA_NCDF Outputs data to the DIMITRI NCDF files DHMI_CLOUD_SCREENING
47. ser defined processing version of product 2nd Reproc YEAR Year of product acquisition 2002 MONTH Month of product acquisition 3 DAY Day within month of acquisition 30 DOY Day Of Year relating to YEAR MONTH and DAY values 89 DECIMAL YEAR Decimal year of data acquisition 2002 244 FILENAME Product filename without path reference ROI COVER Integer raised if all corners of the corresponding ROI are covered by the product NUM ROI PX Number of pixels within the associated ROI 1068 AUTO CS Automated ROI Cloud Screening result in percent 50 0 MANUAL CS Integer flag indicating 1 not performed O clear 1 cloudy AUX_DATA 1 10 A string of auxiliary data used to process the L1b product 5 2 Auxiliary data files A number of auxiliary data files are used by the DIMITRI software package The main aux files are Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium S resolution multi spectral imagers nadie Issue 1 2 ARGAN S DIMITRI MEREMSII Page 27 Software User Manual D14 1 Sensor info This file contains information on the available satellite sensors within the DIMITRI software package 2 Site info this file contains information on each of the DIMITRI validation sites as well as any user defined sites 3 Configuration file this file contains the user s DIMITRI configuration settings e g plot sizes and RGB quicklooks and is updated by the configur
48. te 08 05 2012 Issue 1 2 Page 20 3 DIMITRI V2 0 VGT SIM SETUP MERIS FOLDER DomeC 20111006 REF AATSR 2nd Reeprocessing MERIS CONFIG NZ AATSR FOLDER DomeC 20111006 REF AATSR 2nd Reprocessing AATSR CONFIG SITE VGT CONFIG AATSR 2nd Reprocessing DomeC Calibration 1 USER PARAMETERS BRDF PLOTS ON OFF BRDF BIN PERIOD DAYS 5 BRDFOBS THRESHOLD VGT CLOUD THRESHOLD 0 000000 VGT ROI THRESHOLD Z 90 0000 ABSOLUTE ANGLE CRITERIA 3 5 3 Final Steps When all parameters are selected press the START button to begin processing DIMITRI then utilises the super sensor time series to generate the ROUJEAN BRDF model RD 14 for each band and each binning period Once complete the BRDF models are concatenated and used to simulate Figure 4 Process 2 setup module HMI the VEGETATION TOA reflectances This involves e extraction of VEGETATION observations corresponding to each BRDF bin e the checking of BRDF models at each MERIS band plus the AATSR 1 6 micron band any missing MERIS bands are computed from the closest nominal BRDF model Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium hi resolution multi spectral imagers indie Issue 1 2 ARGA N S DIMITRI MEREMSII Page 21 Software User Manual D14 e the generation of TOA reflectance using VE
49. the DIMITRI database file and used with the selected cloudiness thresholds for Sensor Recalibration Section 3 4 and VEGETATION Simulation Sections 3 5 In addition to the automated cloud screening Users can open the manual cloud screening module This allows visualisation of a products quicklook for manual determination if it is cloudy or clear or if the product contains errors suspect The manual cloud screening results always override the automated results during Sensor Recalibration and VEGETATION Simulation Ref ME MAN ARG TN SUM a Towards the intercalibration of EO medium S resolution multi spectral imagers err Issue 1 2 ARGAN S DIMITRI MEREMSII Page 22 Software User Manual D14 13 DIMITRI V2 0 CLOUD SCREENING Filename MER RR 1PRMAP20080102 083031 O00000542075 00150 35775 D0001 N1 Region Libya Sensor MERIS Proc Ver 3rd Reprocessing Num ROI PX 8857 Auto CS 0 00 ManualCS N A CLEAR CLOUDY i SUSPECT Figure 5 Manual Cloud Screening HMI 3 7 Visualisation The visualisation module provides a quick and easy opportunity to view any output results from either Sensor Recalibration or VEGETATION Simulation Each menu is dynamically created depending on which data is available in the output folder selected Plots can be generated and saved as JPG s or PNG s or as a semi colon delimited CSV file Statistics on plotted data can also be viewed throu
50. using the DIMITRI HMI New Site module e All instrument folders will be identified through the defined values such as MERIS MODISA and AATSR e All processing versions will be indentified through the syntax Proc for example Proc 1 Reprocessing for the y reprocessing The Output folder will contain folder names based on the syntax site date ref sensor e g Domc 20100922 ref MERIS 2nd Reprocessing This syntax allows users to identify different processing options runs performed User defined folders will be available through the main DIMITRI HMI Year EN Figure 1 Schematic diagram of the DIMITRI folder structure ed ARGANS Towards the intercalibration of EO medium resolution multi spectral imagers DIMITRI MEREMSII Software User Manual D14 Ref ME MAN ARG TN SUM Date 08 05 2012 Issue 1 2 Page 13 Important A number of routines require the explicit syntax of certain files and folders It is therefore critical that the folder names are NOT modified as this will result in a loss of functionality 2 4 1 VEGETATION Folder Structure Due to the structure of VEGETATION products the following filename convention must be used for VEGETATION products ONLY Input 2Site 2VEGETATION 2Proc Version 2Year Product Folder 20001 data files T Ref ME MAN ARG TN SUM Towards the intercalibration of EO medium i Mr resolution multi spectral imagers Davee Os 005 l ale
51. well as updates to the DIMITRI auxiliary data files located in the AUX DATA and Bin folders It is recommended that the DIMITRI auxiliary data files and database are fully backed up before adding any new sensor code a Towards the intercalibration of EO medium Ref ME MAN ARG TN SUM Ws resolution multi spectral imagers pate 090372012 Issue 1 2 ARGAN S DIMITRI MEREMSII Page 32 Software User Manual D14 APPENDIX A DIMITRI IDL ROUTINES Routine Short Description GET DIMITRI BAND INDEX TEMPLATE Returns the DIMITRI band index template EXTRACT DIMITRI RSR Extracts Relative Spectral Response data over defined wavelengths CONVERT TIME TO JDAY Converts ENVISAT Time to Julian Day GET DIMITRI LOCATION Returns the location of different DIMITRI files and folders GET PRODUCT IDENTIFIERS Returns the products identifiers used to search for data products GET HISTOGRAM Outputs a histogram JPG GET ENDIAN SIZE Computes the machine endian size GREAT CIRCLE DISTANCE Computes the great circle distance between two points GET DIMITRI RSR TEMPLATE Returns the DIMITRI Relative Spectral Response template GET DIMITRI SENSOR DATA TEMPLATE Returns the DIMITRI Sensor data template GET DIMITRI SITE DATA TEMPLATE Returns the DIMITRI Site data template DIMITRI ANGLE CORRECTOR Corrects angular information to the DMIITRI standards CONVERT INDE

Software User Manual

Contents

Download Pdf Manuals

Related Search

Related Contents