Home

(V) lidort - Planetary Aeronomy

Contents

1. Another method consists in changing the environment within VISUAL In the Solution Explorer right click on Asimut select Properties Under Debugging 19 Select Environment gt lt Edit gt Add the directory to the PATH Path Path C NProgram Files MATLAB R2007b bin win64 BE Sa Fes MATLAINOUD Te an Step 2 Indicate where the MATLAB macros located The second step to ensure the link is adding the path to the MATLAB macros ASIMUT Matlab to the PATH variable of MATLAB This link can also be done through the Directories dirMatlab in the ASI control file see ASIMUT Input File pdf for more details 20 Method 2 Runtime Be sure to Add the 2 following libraries mclmcrrt lib and the libasimutplot lib The first one mclmcrrt lib is the Matlab Runtime library to be found in the MATLAB extern lib sub directory C Program FilesNAMATLABNR2007bVexternNib extern lib The second one is the ASIMUT library containing all plotting functions It is located in the ASIMUT Lib sud directory AsimutNbinNLibNWin 4 ASIMUT Lib and its related h in AsimutbinNncludeNin64 ASIMUT_ Include 1 Import the libraries into the project In the Solution Explorer right click on Asimut choose Add gt Existing Item 1 go to the MATLAB_extern
2. 30 CO 31 2 2 7 2 General files in RESULTS juice tet oa 31 2 2 5 2 Sp cifie files T SA aissi 34 2 2 8 Examples Provided eire 35 2 2 8 1 Extracting Line Parameters 33 2 2 8 2 Nadir IR spectra of Mars atmosphere no aerosols 35 2 2 8 3 Nadir IR spectra of Mars atmosphere with aerosols 35 2 2 8 4 Analysis of a SOIR solar occultation observation 36 puo ME Dudum 37 2 3 HOW WC 38 23 How tostat 38 2 3 2 to extract spectroscopic LP data from literature database 38 2 3 3 How to limit the size of the LOG file 2 ccccccessacsscossccssscsssacedensedezesdeas 38 2 3 4 How to add a new Instrument hr crt retine d en 38 28 MEbeol don 41 2 4 1 WIN Debugging information for project exe cannot be found or does not match Binary was not built with debug information 41 242 WIN No compilation Fatal error C1033 cannot open program d tabase am 41 2 4 3 WIN UNIX The figures from MATLAB are created but are empty 41 234 WIN UNIX Problem with Files eerte inea
3. LIDORT Full scalar RT version 3 3 VLIDORT Full vector RT version 2 4RT VLIDORTF90 Full vector RT version written under F90 SPHER Calculation of aerosol atmosphere parameters needed as input for V LIDORT in case of spherical aerosol particles T MATRIX Calculation of aerosol atmosphere parameters needed as input for V LIDORT in case of non spherical aerosol particles Table 1 Scientific goals fulfilled by the programs included in the ALVL software Figure 2 illustrates the file structure of the complete ASIMUT V LIDORT project The first subdirectory contains all files related to the ASIMUT RT code It is a self contained package and can be compiled and used independently with respect to the complete ASIMUT V LIDORT project LIDORT and VLIDORT libraries are stored in folders named includes and source files are stored in sourcecode and sourcecode str respectively VLIDORT has one additional folder named LAPACK directory to store the Fortran Lapack library subroutines All files and libraries that contribute to aerosol radiation sources and RT modelling but are not parts of the original LIDORT and VLIDORT packages are designated by the prefix aux and are saved in AUX SOURCE and AUX SOURCE_V respectively VLIDORTF90 is a new addition not shown in this Figure It will replace the existing VLIDORT older version All plots are done through the Matlab interface All macros ensuring
4. Model From the atmosphere file Previous From previous fit ATMname From previous fit of ATMname molecule The last option is useful when fitting separately different isotopologues of the same molecule for example first in PASS 1 fit only the main isotopologue O3 666 then in a second PASS where isotopologue 686 is fitted indicating aPrioriMolecules O3 will Strat the fit of O3 686 from the results of the fit of O3 done in the previous pass SPn FENx ODLimit Limit of Optical Depth under which a line will not be considered SPn FENx LBLfct Type of function used for the LBL calculation using either a constant or variable step SPn FENx CrossSections cross sections XS type to be included Maximum number of cross sections that can be included in one run 20 SPn FENx FitCrossSections For each cross section indicates type of Ert Value is lt Fit only the column Value is No fit only simulation Value is gt 0 0 Fit as described in xxxXS description The decimal part of the number is the covariance for the fit if no Sa file is given 0 0 0 0 SPn FENx aPrioriCrossSections For each cross section indicates the type of a priori that will be used in the Rodgers formalism Model From the atmosphere file Previous From previous fit ATMname From previous fit of ATMname molecule SPn FENx Aerosols Aerosols AER type to be included Line by line calculations Maximum number of Aeros
5. Species Molecule ID H20 N20 CO SO NO As in HITRAN HNO OH HF HBr HI CIO OCS H CO from 1 to 42 N2 HCN CH3Cl H202 C2H2 C2H6 SFs H2S HCOOH CIONO NO HOBr C2H4 CHOH CH3Br CH3CN GeH C4H2 HC3N C3H4 From 43 to 48 molecules in GEISA but not in HITRAN HONO HO5NO N505 CHF From 49 to 68 molecules CH3CCL3 CCL4 COCLF CHE CL in CFGL but not in OCIO F134A F142B CFC113 F141B HITRAN C5HgPL PAN Table 5 Molecules recognized by ASIMUT 3 2 2 2 xxxXS options In the name of these sections xxx represent the name of the molecule ex 03 CO2 NewO2 formaldehyde XS stands for Cross Section xxxXS ATMname name to be found in the atmospheric file Must be given 60 xxxXS File name of the file containing the spectroscopic parameters Must be given xxxXS Type type of the file containing the cross sections data hitran geisa ascii If ascii is selected the following columns must be given wavenumber cm 1 N columns cm2 molec with N the number of different laboratory conditions which will be described by the next parameters Temperature Pressure air resolution resolUnit The following parameters describe the laboratory
6. The same reasoning can be applied on a Doppler limited profile with the respective equations 52 2 T v 1 A v exp _Su n exp In s 3 32 fp YD 2 aoe mz exp In s 3 39 Yo YD Su 2 A Doppler Vo v rA re 3 34 D The absorption at the center of a Voigt profile is always less than the absorption at the center of both the Doppler and the Lorentz profile from which it is calculated As a result if the line intensity at the center is less than both limits defined by Equ 3 31 and 3 34 then the line can be rejected The criterion of rejection is thus the following min A lt Aoptimit 3 3 5 Truncation distance from the line centre The determination of the distance from the line centre at which the calculation of the line profile is stopped D is based on 2 criteria Note that the value of D can also be imposed by the user in MolecLP F Bord option and when y factor is used D is defined as the width of the spectral interval on which the function is given The first criterion ensures that the line is not computed for wavenumbers outside a spectral interval defined by a minimum cut off value imposed on the normalized line profile before multiplication by the line intensity and the integrated density of the species in the corresponding layer The reasoning is based on the assumption that the line is a Voigt line profile and that tests can be done separately on conditions imposed
7. Under Debugging Select Environment gt lt Edit gt Add the 2 directories Path Path MATLAB runtime ASIMUT dll 22 2244 Under UNIX Change the following line in the makefile to indicate where the libraries can be found MATLABLIBPATH opt matlab bin glnxa64 MATLABRUNPATH opt matlab runtime glnxa64 MATLABINCPATH opt matlab extern include and ASIMUTLIBPATH bin Lib Unix ASIMUTINCPATH bin Include Unix As under WINDOWS the user can choose between two ways of importing the MATLAB functionalities through the Engine or through the Matlab Runtime library When using the Engine Use MATLABRUNTIMEE NO on the command line for the compilation To ensure direct communication between ASIMUT and Matlab under UNIX the user can use the Directories dirMatlab in the ASI control file see ASIMUT Input File pdf for more details When using the Runtime Library method Use MATLABRUNTIME YES on the command line for the compilation Makefile first lines Heat Ht ae at aE aE ae aE aE aE aE HE aE aE aE HE a a aE aE aE aE aE aE aE aE a aE aE aE aE a aE aE aE EEE EE aE Makefile for compilation of the ASIMUT LIDORT VLIDORT package created by V Letocart BIRA Belgium on 07 08 2009 last updated 14 05 2014 Heat Ht ae a HE aE aE aE aE HE aE aE aE HE EE aE aE aE aE HEE aE aE aE aE a aE aE aE aE HEE aE aE Ea ERE ERE calling examples to compile using the LIDORT option with Matla
8. endif endif COBJECTS S patsubst c 0 wildcard ASIMUT c 24 2 2 5 Using the CLAPACK library Line mixing has been included for and in near future Two methods of calculations have been implemented the first one is called 1 order approximation and the second one is the full mixing method see REF for more details on the two methods When the full mixing method is selected the CLAPACK needs to be included into the ASIMUT ALVL project For the moment this is only possible with the version of ASIMUT ALVL compiled under WINDOWS The different files are located 1n Asimut bin Lib Win64 or NWin32 z ASIMUT Lib Asimut bin Include Win64 or NWin32 z ASIMUT Include To include the CLAPACK libraries 1 Import the libraries into the project In the Solution Explorer right click on Asimut choose Add gt Existing Item go to the ASIMUT Lib directory and select the BLAS lib clapack lib and libf2c lib files 2 Specify to the compiler where to find the libraries and the related Include In the Solution Explorer right click on Asimut select Properties Under Configuration Properties gt VC Directories Select Include Directories gt lt Edit gt Add the line Asimut bin IncludeWwin64 ASIMUT_Include In the Solution Explorer right click on Asimut select Properties Under Configuration Prope
9. 2 A list containing a series of filenames that have to be treated in the same way same options of INP files but each file being analysed independently option also used with the Onion Peeling method 3 A list containing a series of filenames that have to be treated in the same way same options of INP files but all files being analysed simultaneously Occultation type of observation To follow the existing ASIMUT structures those files should be saved in the ASIMUT ASIMUT Spectra directory If you are working under SVN a Add the new files in the SVN structure in your browser under Windows right click on the file to add select SVN Add b Change the properties of the files right click on the file select SVN Properties A dialog box appears in which you add the following svn keywords LastChangedDate LastChangedRevision LastChangedBy This will allow the automatic change of the files headers see Templates when the files are modified Be sure that those headers are present Nfile NEWINSTRUMENT h brief Functions relative to NEWINSTRUMENT Created 07 03 2006 Rev 257 Last modified Date 2009 10 01 16 31 12 by Author xxx Specify the new h file to be included in ASIMUT project add the following line in the ListAllh h file Gn ASIMUTNASIM UTNcommon include Spectra NEWINSTRUMENT h The second place to change is inside the existing User Choice c and
10. DATE 01 07 2007 11 31 14 LATITUDE 34 644520 deg North LONGITUDE 151 112400 deg East ZENITH 31 720000 deg AZIMUTH 77 520000 deg SURFACE unknown ALTITUDE 0 km HOBS 833800 000000 km DISSTANCE 30 350771 km ZPT ZPT 2007070121 110662 txt UNITS cm 1 W cm2 Sr cm 1 79 followed by 2 columns with Wavenumber cm Radiance The ZPT line of the header indicates which zpt file to read The file must exist Option List is possible with this instrument The LST file must then contain only 1 column with the Name of the IASI file 80 3 3 6 7 SOIR instrument The parameters describing the instrument and spectrum or list of Spectra are described as well as the format of the different files needed SPn aotfFrequency AOTF frequency This information is also read directly from SOIR CSV and SOIR PDS type of files SPn aotfCentralWnb AOTF central wavenumber corresponding to the aotfFrequency If aotfCentralWnb is given it has the precedenc over aotfFrequency At least one of the two parameters must be present except when using the SOIR CSV and SOIR PDS3 type of files SPn CentralOrder AOTF central Order only used when reading RESOL file if no Data are specified Normally this value is read from the header of the SOIR files SPn aotfNbPixel Number of pixels on the detector SPn NumberOrders Number of adjacent orders included in the SOIR t
11. Weakly Interacting Molecular Pairs Unconventional Absorbers of Radiation in the Atmosphere 2002 Mat B et al Absolute intensities for the O2 1 27 um continuum absorption J Geophys Res 1999 104 D23 p 30585 30590 Mlawer E J et al Observed atmospheric collision induced absorption in near infrared oxygen bands J Geophys Res 1998 103 D4 p 3859 3863 Greenblatt G D et al Absorption measurements of oxygen between 330 and 1140 nm J Geophys Res 1990 95 D11 p 18577 18582 Bernath P et al The Wulf bands of oxygen Chem Phys Letter 1998 297 3 4 p 293 299 Fally S et al Fourier transform spectroscopy in the O2 Herzberg bands Absorption cross sections of the collision induced bands and of the Herzberg continuum J Mol Spectrosc 2000 204 1 p 10 20 Yoshino K et al Improved absorption cross sections of oxygen in the wavelength region 205 240 nm of the Herzberg continuum Planet Space Sci 1988 36 12 p 1469 1475 Borysow A and L Frommhold Collision Induced rototranslational absorption spectra of pairs for temperatures from 50 to 300 The Astrophysical Journal 1986 311 p 1043 1057 Boissoles J et al Theoratical calculation of the translation rotation collision induced absorption in O5 O and N5 O pairs J Quant Spectrosc Radiat Transfer 2003 82 1 4 p 505 516 Lafferty W et al Infrared collision induced absorption by nea
12. in the frame Solution Explorer or go to the menu Project Properties EEEL ELS First at the top of the window select the option configurations in the Configuration list In the Configuration Properties section Debugging Command arguments path file asi If you intend to run the program within the Visual C environment C C Preprocessor Preprocessor Definitions Add the following WIN32 or WIN64 SECURE NO WARNINGS DEBUG if you want to use the Debug options of Asimut those are not the same as the Debug options provided by Visual C other options as needed MATLAB MATLABRUNTIME FULLLINEMIXING OPERATIONAL deselect the option Inherit from parent 12 fand Leet MOLECULE Molecula Soubla decia Waves ch Flee Preprocessor Definitions Define a preprocessing symbols for your source file 20 1 If you intend to use Matlab or the LAPACK library If you need one of these options you must specify where to find the h lib for the compilation and dll or so for the run Please refer to Section 2 2 4 and 2 2 5 for additional inputs and references By default all links to MATLAB are deselected and the LAPACK library is not needed only 1 order line mixing 13 2 2 3 Compiling under UNIX A makefile is provided it produces 4 different executable fil
13. 1 0 72 1 52 T Planck cst s STP Temperarture K STP Pressure hPa Avogadro Na mol 1 Boltzmann cst J mol 1 2 KB h c Loschmidt number R TO Rgaz Na 1 0 LOSCH T0 P0 molec cm3 Gas Constant J mol 1K 1 Planck cst C1 Planck cst C2 a 1 438751926703556 W cm2 sr K cm 1 PI 180 180 Ad Earth Earth Earth radius km radius Model Atm 1 radius Model Atm 4 5 major axis Lat Lon corr minor axis Lat Lon corr REQ REQ REP REP REQ REQ REQ REQ REP REP REP REP Mars radius Venus radius km km K K K of Earth s surface of Mars s surface of Venus s S of Sun s surface Temperature Temperature Temperature Temperature Radius of Sun km dist Earth Sun 1 au distance Sun Earth au distance Sun Venus au distance Sun Mars au km 117 Annex 6 Template of file for a new instrument Template of the NEWINSTRUMENT c file EERE AE RE eoo soleo seo eee E E RRP ee dopo AE E IE AE RE A AE A AE NEWINSTRUMENT c DESCRIPTION Functions relative to NEWINSTRUMENT ae eoe Read the header Jte sp epos LE PERL ELE EEE ESO JOE LE ERLE SEL ERLE LEE EEL ELS PRESS
14. Continua N2 Formalism for N2 continuum mtckd12 MT CKD 1 2 mtckd13 MT CKD 1 3 same as 1 2 mtckd13Cor CKD 1 3 corrections mtckd25 MT CKD 2 5 64 3 2 4 Rayleigh scattering Molecular scattering is certainly the strongest mechanism causing atmospheric extinction of solar radiation in cloudless atmospheres The Rayleigh optical depth at wavenumber for the layer at the altitude z is given by V z N z o v z 3 53 where N z is the molecular number density of air in the layer o 2 is the total Rayleigh scattering cross section per molecule which may be expressed as 2 24g n 2 1 4 2 2 nv 2 where n v z is the refractive index of air at v and z F z is the King factor for the depolarization of air at v and z that accounts for the anisotropic properties of air molecules anisotropic molecules scatter more radiation at 90 than isotropic ones with the same index of refraction The F factor is calculated as the weighted sum of the partial contributions due to the molecules Ar and considering that each contribution varies as a function of wavelength The weights are given by the volume mixing ratios of the respective atmospheric constituents 24 O V Z F v z 3 54 In parallel to this implementation of the Rayleigh extinction Ciddor the user may select other methods LBLRTM the calculation of the Rayleigh cross section is very basic and
15. be fitted during the retrieval procedure b the reflection of the thermal emission sources originating from the different layers of the atmosphere specular reflection and c the reflected solar radiation attenuated through the atmosphere For measurements looking directly at the sun Jp can be either simulated by blackbody function or read from an external file allowing for example the use of the high resolution Kurucz atlas 16 or solar measurements 17 The simulation of an atmospheric spectrum can be separated into three distinctive components It first of all requires the determination of the ray path through the atmosphere and the computation of the temperature pressure and molecular densities along this path Then using this information and the line parameters or absorption cross sections relative to each absorbing species the optical depths corresponding to each layer are determined Finally the radiative transfer itself must be solved The three steps will be described in more details in the following sections 3 1 Ray tracing The determination of the radiation path through the atmosphere i e the path followed by the radiation reaching the instrument requires that the Earth s curvature and refraction be taken into account The model is based on the ray tracing program FSCATM 18 FSCATM has been improved when being incorporated into the line by line program LBLRTM 19 20 and later when included into the retrieval pr
16. by the Doppler part of the profile and of the Lorentzian part For a Lorentz profile intensity will go under the cut off limit A cutoff 1 0 10 when 1 V 3 36 cuttoff and for a Doppler profile the relation becomes 2 via y In G 37 0 y In 2 ig The interval on which the profile will be calculated is thus limited at Var AX sag 3 38 The second criterion check when the line corrected with the line intensity from HITRAN for example and the integrated density in one layer goes under a second 53 cut off limit A cutoff intensity 5 0x10 Similar sets of equations are used to define the cut off position in that case crit __ Lorentz Doppler V cutoff MAX V 2 Lom 3 3 9 with Su Lorentz m cutoffintensity Vy A YL 3 40 cutoffintensity n2 s In 3 41 d p uofintensity Doppler f D cutoffintensity Vo The final definition of the cut off position is based on the observation that the criteria defined in Equ 3 38 and 3 39 are upper limits Finally the cut off is defined as min v yen 3 42 cutoff cutoff V soi Wavenumber intervals The user defines the wavenumber spectral interval on which the simulation retrieval will be performed Limits for plotting can also be supplied This spectral interval will be extended in order to 1 ens
17. indicates the type of a priori that will be used in the Rodgers formalism Model From the atmosphere file Previous From previous fit ATMname From previous fit of ATMname molecule The last option is useful when fitting separately different isotopologues of the same molecule for example first in PASS 1 fit only the main isotopologue O3 666 then in a second PASS where isotopologue 686 is fitted indicating aPrioriMolecules O3 will Strat the fit of O3 686 from the results of the fit of O3 done in the previous pass LH The Covariance is described in each corresponding xxxLP xxxXS or xxxAl under the Sa SaType parameters R 5 4 2 Variables related to the Spectrum and its windows Several variables can be fitted wavenumber shift SPn FENx FitShift the surface temperature SPn FENx FitTs the signal baseline SPn FENx FitBaselineCst FitBaselineX and FitBaselineX2 the albedo SPn FENx FitAlbedo SPn FENx FitShift Wavenumber shift SPn FENx Shift is fitted The number given under SPn FENx FitTs Fit the surface temperatur SPn FENx Ts Different options are possible 94 3 x Fit of the surface temperature Take the SPn_FENx Ts value previously fitted in another FEN std deviation 2 No Ig However take the SPn FENx Ts value previously fitted in another FEN Fit simultaneously on another FEN in which this option will be se
18. p t and Relative Humidity are read Relative Humidity is then converted to vmr imported in the Atmopshere e Std zp t are read in the AtmosphericModels atmFile z km p mb t K Example venus day ppm dat Z km T K P mb NT cm 3 CO2 ppmv N2 ppmv CO ppmv O3P ppmv SH Z Y P N CO2 N2 CO 03 60 0 2 7278e 002 1 780 002 4 72e 018 9 65e 005 4 62 001 5 10 001 2 12e 007 61 0 2 6894e 002 1 470e 002 3 97e 018 9 65e 005 4 70e 001 4 12 001 3 03e 006 45 AtmosphericModels Density indicates how the water contribution is removed from the determination of the total air density AtmosphericModels Space definition of the altitude km of where Space begins No calculations of optical depths or contribution to the radiative budget are performed above this limit If not given the value is the highest altitude in the AtmosphericModels atmFile 3 1 3 Geometry of observation Several geometries are supported nadir off axis nadir from ground to space from space to ground solar occultation The user further needs to give some of the following parameters to fully describe the ray path the position of the observer Hops the looking direction or zenithal angle 4 the tangent height for limb measurements and the altitude of the end point of the light path H With the help of those parameters any configuration may be simulated Figure 3 The zenithal angle can be specified either
19. then become prohibitive However it can be observed that the line profile change slower at a distance from the line centre than it does near the centre The solution implemented in ASIMUT is the possibility to use a non uniform grid near the centre of the line the optimized step derived with the help of Eq 3 25 is used the step is then progressively enlarged as one goes away from the centre Our algorithm is based on the study of Fomin 37 which decomposes the spectral grid unto a series of sub intervals Let s suppose that the line shape must be calculated on the interval D sufficiently large to cover the line completely The determination of D will be described in the Truncation distance from line center sub section The limits of the centre zone are defined by v C v C with C 24 01 and a being the Doppler and Lorentzian widths of the line The grid is divided into 2L portions located at unequal intervals 51 v D v 2 C v 22 C v 2C v 2C v central zone v Cv 2C v 2C v 2 27 Cv D The number of intervals is related to D and C through the following expression 2 xC D 3 27 In the central zone the sampling is set to the optimized value determined with Eq 3 25 Then in each subsequent interval the sampling is doubled This drastically reduces the number of points on which the line profile is calculated without losing accuracy at the centre of the line At t
20. 5 4 where Se the error covariance matrix of the measurements and S the a priori covariance matrix S is chosen to be diagonal characterizing all the sources of systematic and random errors on the measured radiance The convergence is attained when the 2 following conditions are met d X4 S x Xi lt lt lt 5 5 d f x 55 with the number of parameters to retrieve m is the number of observed points and dy y f x 5 6 Ssy Se KS S 5 7 The first expression expresses the convergence condition on the fitted parameters and the second on the difference between observed and simulated spectra The matrix S S K S K 5 8 represents the total error covariance on the retrieved parameters which can be decomposed into two contributions the smoothing error which accounts for the sensitivity of the measurements forward model to the variable to be retrieved i e the measurement forward model system does not allow perfectly reproducing the true atmosphere but a smoothed value of it and the measurement error Their covariance can be expressed as 90 A 7 DS A 5 9 S GS G 5 10 measur x where G is the gain matrix and A x is the averaging kernel matrix I y is the identity matrix The basic implementation of the Rodgers algorithm in ASIMUT makes use of approximate weighting functions The main objec
21. AtmosphericModels Space taken 5 also on the specific definition of the Spectra files for some instruments since this parameter could be given in their header this is the case for SOIR spectra for example SPn HEnd indicates the altitude of the ending point of the LOS km It represent the tangent height when SPn Geometry LimbZztg See also on the specific definition of the Spectra files for some instruments since this parameter could be given in their header this is the case for SOIR spectra for example SPn Angle indicates the angle 6 degrees defining the LOS See Figure 3 for the different definition of the Angle in function of the geometry type See also on the specific definition of the Spectra files for some instruments since this parameter could be given in their header this is the case for SOIR and IASI spectra for example SPn AngleType indicates if the angle 15 apparent astronomical SPn len in some case see Figure 3 two LOS can be defined by Specifying only the Hobs and HEnd altitudes If SPn len 1 it will consider the longer path if 0 the shorter one SPn Range length km of LOS when SPn Geometry Horizontal SPn Sun sza Solar zenith angle degree w r t the topocentric reference fram SPn VZA Viewing zenithal angle degree w r t the topocentric reference fram SPn SolAZ Solar azimuth angle degree w r t the topocentric ref
22. Fayt Svetlana Kochenova Tobias Kerzenmacher Michel Kruglanski Vincent Letocart Alexis Merlaud S verine Robert Ann C Vandaele Sophie Vandenbussche Yannick Willame 125 ASIMUT RELEASE 11 0 LICENSE AGREEMENT Permission to use copy modify and distribute any ASIMUT RELEASE 11 0 any documentation appertaining to this version of the VLIDORT and any results obtained using this code is hereby granted without fee and without written agreement provided that both the notice of copyright as expressed in this paragraph and the following two disclaimer paragraphs appear in all copies of the software IN NO EVENT SHALL IASB BIRA BE LIABLE TO ANY PARTY FOR DIRECT INDIRECT SPECIAL INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OF THE ASIMUT RADIATIVE TRANSFER MODEL SOFTWARE AND ITS DOCUMENTATION EVEN IF IASB BIRA HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE SOFTWARE IS WITH THE USER BECAUSE THE ASIMUT SOFTWARE IS LICENSED FREE OF CHARGE THERE IS NO WARRANTY FOR THE PROGRAM TO THE EXTENT PERMITTED BY APPLICABLE LAW EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND OR OTHER PARTIES PROVIDE THE PROGRAM AS IS WITHOUT WARRANTY OF ANY KIND EITHER EXPRESSED OR IMPLIED INCLUDING BUT NOT LIMITED TO THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE IASB BIRA HAS NO OBLIGATION TO PROVIDE MAINTEN
23. K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K FUNCTION GetNbSpec_NEWINST_LIST char FileName int SamePass T int First PURPOSE Get number of spectra from FileName containing a list of spectra INPUT FileName name of list file 119 SamePass 1 if all spectra are analysed simultaneously First index to the first spectrum OUPUT nb of spectra A oe ee ie doe tob Tob lodo bo boo dole boe int GetNbSpec NEWINST LIST char FileName int SamePass int First int char Line NL FILE hFile char str 1024 First 0 if hFile fopen FileName rb NULL sprintf str Can not open file 25 FileName set_Error ERR_FILE_NONE SEV STOPPROG LINE FILE str return 0 while fgets Line NL 1 hFile nb SamePass NON all spectra are NOT analysed simultaneously fclose hFile return nb index index of the spectrum in the list strN Complete filename to spectrum with path Desc description of the file in the list LEELECEFEL ELLE SE CES EE EE ELE EL TELS EEL EEL FUNCTION GetNameSpec NEWINST LIST char FileName int index char strN char Desc PURPOSE Get file name and description of spec
24. Name of Author Created date Rev 337 Last modified Date 2011 04 15 16 26 52 by Author EFELEELELLESE TE EERE SELES EE EERE LE CELLET EEE ERS ORO RJ include common Asimut h Instrument Instrument structure Geo Geometry of the measurement ZPTfromSpectrum 1 if ZPT information in this file FUNCTION Read NEWINST char FileName m INSTRUMENT Instrument GEOMETRY Geo int ZPTfromSpectrum PURPOSE Reading the NEWINSTRUMENT spectra from FileName INPUT FileName name of NEWINSTRUMENT file OUPUT Matrix containing the spectrum wavenb Intensity FK KK K KK K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K y Matrix Read NEWINST char FileName INSTRUMENT Instrument GEOMETRY Geo int 7pTfromSpectrum Matrix Data NULL double Matrice NULL long nbPts nbColumns 0 char str 1024 Line 256 FILE hFile NULL int onError ERR NOERROR Read the header if hFile fopen FileName rb NULL sprintf str NEWINSTRUMENT file s does not exist FileName set Error ERR FILE NONE SEV STOPPROG LINE FILE str return NULL fgets Line 255 hFile fgets Line 255 hFile 118 fgets Line 255
25. articles on the database Each portion of the cross section files i e a set with a temperature and pressure pair contains a header that points to the reference for that observation and other relevant information The sets contain absorption cross sections that are in equal wavenumber increments and the intervals can be determined by the minimum and maximum wavenumber given in the header and the number of points i e AV Vmax Vmin npts 1 where Ving is the maximum wavenumber of the set Vmin is the minimum wavenumber of the set and npts is the number of points in the set The new format is quite similar to what we had before except that the field length for the chemical symbol has been increased from 10 to 20 there is already one compound that requires 13 characters the number of points and temperature fields have been reduced from 10 to 7 and the pressure field has been reduced to 6 Previously the last 30 characters were rather undefined although there was an indicator for the reference Now we have put in the resolution of the observation For Fourier transform spectrometer observations all of the IR data this will be in wavenumber For grating spectrometer observations presently only for some UV data we employ xxxmA where xxx is a number and mA indicates milli Angstroms Since some of the experiments were done in air we have a field in which we put air blank indicates pure gas Finally there is an i
26. conditions under which the data in the different columns of the xxxXS File have been obtained xxxXS Temperature Temperature K of each column of the xxxXS File xxxXS pressure pressure torr of each column of the xxxXS File xxxXS air indicates if the wavenumber ar xpressed in air 1 or vacuum 0 xxxXS resolution resolution of each column of the xxxXS File xxxXS resolUnit unit of the resolution cm 1 or nm 0 xxxLP zfact fact factors that can be applied on the vertical profile read from the atmosphere file 61 3 2 3 Continua For some gases such as O2 CO and broad absorption features cover wide spectral regions in addition to single absorption lines These continua of absorption are due to different physical phenomenon Collision induced absorption bands resulting from the interaction of pairs of molecules such as O5 O superposition of line wings in the case of H2O and Different molecular continua are included in ASIMUT They will be described in details hereinafter In general they follow the MT CKD 1 2 formalism 20 with modifications introduced to take into account new laboratory measurements or the possibility to simulate atmospheres other than Earth 3 2 3 1 continua Fundamental band at 6 4 um Thibault et al 40 performed new measurements of the and O2 N gt absorption coefficients in the 0 20 atm and 193 293 K pressure and te
27. description of the error the file along with the function and line number in which it happens This information can be used to understand the origin of the problem file not given file not in the expected format value for requested parameter not given etc In case of problems for which the solution is not trivial do not hesitate to contact IASB BIRA see Annex 7 2 2 7 2 General files RESULTS Several files are always produced and are saved in the RESULTS directory They contain general information on the run They can be saved in ASCII or in MATLAB format MATLAB File FileQUT mat The MAT file contains 1 matrix structure s one element for each pass or each spectrum defined in file INP Each element is composed of a series of sub structures or elements nbfen 16 fen 1x16 struct geo 1x16 struct nbspecies 2 species 1x2 struct fit 1x1 struct den 1x1 struct vmr 1x1 struct where nbfen is the number of Fenetre simulated or retrieved for this pass or spectrum fen is a matrix structure defining the Fenetre 3l geo nbspecies species fit den vmr is a matrix structure defining the geometry for each of the Fenetre is the number of species retrieved non retrieved is a matrix structure defining the vertical profiles of the species included retrieved non retrieved is a structure defining the fit is a structure defining the vertical densities retrieved non
28. et al 1995 Rahman 8 8 Rahman et al 1993 Cox Munk 9 9 Cox amp Munk 1954 Gisssoil 10 13 Gisssnow 11 14 GissCoxMunk 10 Rherman 11 Breon 12 Table 6 List of BRDF kernels implemented in LIDORT VLIDORT In case of thermal spectrum there is an option to calculate the surface albedo automatically on the basis of emission values If the lower boundary of a given spectrum range is equal to exceeds 2 4 um the surface is automatically defined as Lambertian and its albedo is calculated as 1 v following the approached suggested by Wan 1999 Unfortunately BRDF surface tests have not been successful for VLIDORT 2 4RT latest version For some reason VLIDORT acts like surface albedo 1s 0 which is not correct This issue is corrected in VLIDORT v 3 5 issued in March 2011 F77 amp F90 which will be soon incorporated in ALVL 99 6 4 Comment on the Radiance units used in ASIMUT and V LIDORT In ASIMUT all radiances are expressed in W cm cm T sr whereas in LIDORT they are expressed in W m thus integrated on a small wavenumber interval The conversion factor is Radiance asnmur 10 Step cm Moreover Lidort is working on Irradiances factor of PI All outputs from the V LIDORT part have thus been converted into radiances in W cm cm EA a 6 5 V LIDORT input parameters When running V LIDORT additional parameters have to be provided by the user Some are group
29. hFile fgets Line 255 hFile fclose hFile hFile NULL Read the data for example using the Read Ascii function or the following one if Read_Ascii_from FileName amp Matrice amp nbPts amp nbColumns nbofLineinTheHeader ERR NOERROR onError zERR REPORTING errout Line LINE sprintf errout Message Problem reading NEWINSTRUMENT File Xs FileName goto TheEnd if nbColumns 2 onError SEV_STOPPROG errout Line LINE sprintf errout Message Not enough columns in NEWINSTRUMENT File 5 FileName goto TheEnd be sure that the wavenumber scale is ascending CorrectSpectra Matrice nbPts nbColumns create the MATRIX structure if Data Merge Matrix Matrice 0 Matrice 1 nbPts NULL onError ERR_REPORTING errout Line LINE sprintf errout Message Problem Merge Matrix goto TheEnd The step is also initialized here Instrument gt Step Matrice nbPts 1 Matrice nbPts 1 this flag indicate if ZPT information file name is present or not ZPTfromSpectrum NON TheEnd fF free Matdbl2 amp Matrice nbColumns if hFile NULL fclose hFile if onError ERR NOERROR return Data else free Matrix amp Data set Error onError errout Severity errout Line X FILE__ errout Message return NULL The following are not needed if the LIST option is not required KK K K K
30. in this ASI file 2 2 7 Description of the output files In the following we will use FileASI and FileOUT to represent the base names of the files ASI and the output files respectively FileASI is constructed on the name of the file ASI without its extension The FileOUT radical is by default based on the base of the name of the file INP however if a Set caption is mentioned it will define the FileOUT radical 30 For example the files mars IR asimut nadir IR test inp files will be associated to FileASI mars_nadir_ IR and FileOUT nadir IR CHA test If Set caption Nadir_Test2 is present inside the nadir IR CHA testinp file then FileOUT Nadir_Test2 2 2 7 1 LOG file A Log file is created at each run with its name FileASI yymmdd LOG It overwrites any file having the same name This file contains information on the run itself which parameters were considered default values or user given which options were selected a rough indication of the calls to subroutines indication of timing a summary of the results etc The level of details in the file is indicated by the parameter Set Verbose If Run RemoveLog is set to Remove or then the LOG file will be removed IF the programs ends without any problem If an error occurs a series of warnings and explanations will be produced and written in the LOG file To find the origin of the problem search the first warning A short
31. is the extinction coefficient of layer L km and given in kext input parameters defined on scale zFact f v is a function representing the dependence of the optical depth with wavelength Values are given in the file QextFileName 2 columns wavenumber cm 1 function values These values will be interpolated on the 66 spectrum wavenumber scale with the condition that at vo it equals 1 0 f vo 1 0 Conrath formalism Conrath The aerosol optical thickness in the layer L is calculated using Xp F Panes n AS e 3 58 with y a parameter describing the aerosols for Mars y 0 007 during a dust storm y 0 05 during polar night y 0 03 in general Psurface and pz respectively the pressure at the surface and in the layer L nz the density in the layer L and As the path length within layer The value of qo is calculated from the TotalOpacity parameter It can also be expressed in terms of scale height 2 Ay Taer CXP 7 J n As 2 sch A Where is the temperature in the layer L m is the molecular 3 59 R gas T with sch L 8 Surface radius of the planet go the acceleration of gravity at the surface weight of dry air and g is the acceleration of gravity at 21 Rp is the ext formalism Type Qext The aerosol optical thickness in the layer L is calculated using 2 T N RE ar As aer 7 3 60 where Naer is th
32. looking at the Sun solar occultation Ip is the radiance of the Sun When looking down at the planet surface nadir and off nadir modes Ip is the radiance of the surface if no solar reflection is taken into account If solar reflection on the surface is included then the radiance is calculated using Radi Transmittance Radiance Factor 3 73 3 3 5 Brightness temperature The radiance can also be saved as Brightness Temperature which is defined as the temperature in Kelvin of a blackbody that emits the observed radiance It can be expressed as C2 T amp K 3 74 c1v3 1 Radiance v where c 1 191 042 7 x 10 W m sr cm is the first radiation constant for spectral radiance and c 1 438 775 2 cm K is the second radiation constant for spectral radiance 3 3 5 1 Solar options Solar FileSolar Filename of solar irradiance Format for Filesolar For ex irradiance calculated by ACE irradiance m col 1 wavenumber col 2 IRRADIANCE at top of Earth atmosphere W cm cm Irradiance v r R B v Ts With Ts Temperature of Sun photosphere r Sun radius R distance between Earth and Sun B v TS Blackbody Radiance W cm cm sr Solar SolarAngle Solid angle through which the sun is seen from the observer sr Solar TSun Temperature of the Sun s surface K 72 3 3 5 2 Radiative options in SPn SPn source type of source
33. of the program has been one of the driving motives to write our own code Most options driven by the user s choices may be easily extended It is for example relatively easy to add a new instrument the only additional information that the user must provide are a function to read the spectra and the instrument line shape function Another example is the possibility to add other absorption line shapes besides the ones already implemented As one of the objectives was to be able to run the program either on Windows based PCs or on Unix systems graphical output possibilities are not implemented in the code However graphics are possible through the opening of a Matlab session Data and graphics commands are automatically sent to Matlab while the ASIMUT program is running if the program has detected the presence of the Matlab environment on the machine The user controls the program through a series of text files in which he can specify what has to be calculated saved and plotted radiance transmittance Jacobians retrieval results which spectra are analysed in which order It moreover describes the observation geometry corresponding to each spectrum the spectral intervals the molecules cross sections and continua to be taken into account and all parameters needed for the treatment of the species isotopes list of line parameters file for cross sections covariance matrices aerosol characteristics The ASIMUT V LIDORT ALVL software i
34. optimal resolution based on P and T in each atmospheric layers To be used only for tests SPn Step Final step for simulation only Units cm l or specified by DataXUnit SPn oversampling Oversampling If the value is lt 0 0 the oversampling will be calculated in an automatic way by ASIMUT from the optimized resolution SPn Zerofilling Zero filling factor only 2 4 and 8 are possible The number of points of the spectrum is augmented by a factor zerofilling using the well known technique in Fourier analysis SPn Filter File containing a Filter bandpass given in transmittance to apply to the radiance SPn noiseFile Name of the file containing the noise 2 columns x absolute nois same unit as the spectrum where x wavenb or wavelength depending on DataXUnit The file is in Directories Instrument 3 3 6 3 OPUS instrument The OPUS file format is used by BRUKER FTS See documentation of the FTS to have more information on this format The Opus csv option reads OPUS file converted into csv with a header containing the following information Zenith angle 153 864042 MOPD 250 000000 Apodization BX Date 20041009 Hour 7 Min 18 Sec 8 Fov 0 110685 Iris 0 850000 YMax 2 920782 NbPoints 530994 followed by 2 columns with Wavenumber cm measured spectrum Only the opus option can be used with the List option The LST file must then contain 2 columns Name of the Opus f
35. program If Run RemoveLOG yes then it is deleted if no problem occurred Run Save defines what has to be saved As many options as wished Ex save atmosphere raytrace source transmittance Run StopAfter indicates when ASIMUT will stop This option is useful for example to extract the spectroscopic parameters ExtractLP Run StopAfterError When a list of several INP is specified in List stops after the first error encountered if yes continues to the next INP file if no Run SaveResults defines th typ of files saved Ascii or and Matlab both options can be selected Run UseMatlab defines if Matlab will be used Run Plot indicates what will be plotted If something is selected a XXX file is created containing the plots of the simulation or fit results Run PlotVisible indicates if the plots are visible or not When 0 the plots are done but the focus is never set to their windows allowing further normal use of the machine while ASIMUT is running 29 Run NormalizeSpectrum indicates if the experimental spectrum has to be normalized Normalization wrt the mean of the signal same as SFIT The other options are related on how the Rodgers algorithm is applied They will be described later OnionPeeling OnionPeeling if 1 the onion peeling method will be used Directories indicate the FULL directory where different quantities will be found If these directories are
36. that has been chosen as 296 5 v is the temperature coefficient of the collision induced absorption The relative efficiency Of the colliding partners and defined by 51 B v T a 3 50 By _ y V T is practically independent of wavenumber The N air collision induced absorption can be written as By T 2 c a 3 51 where p and p are the partial pressures of and respectively Menoux et al 51 showed that the relative efficiencies of collisions with and can be accurately modeled by the following wavenumber dd relation E n T 1 294 0 45451 3 52 0 Finally the N air collision induced absorption is obtained by inserting Eq 3 52 into Eq 3 51 and using the Lafferty et al 50 model for the v T absorption 3 2 3 3 continua The formalism of MT CKD 1 2 has been implemented in ASIMUT The modeling of these continua has been extensively described elsewhere 52 3 2 3 h Continua options Continua H2O Formalism for H20 continuum mtckd12 CKD 1 2 mtckd13 MT CKD 1 3 same as 1 2 mtckd25 MT CKD 2 5 Continua CO2 Formalism for CO2 continuum mtckd12 CKD 1 2 mtckd13 MT CKD 1 3 same as 1 2 mtckd25 MT CKD 2 5 Continua O2 Formalism for O2 continuum mtckd12 MT CKD 1 2 mtckd13 MT CKD 1 3 same as 1 2 mtckd25 MT CKD 2 5 br BR IASB ULB REIMS measurements
37. the following 3 3 6 2 SPn options SPn InstrumentType Type of instrument FTS includes conventional grating spectrometer and AOTF SPn Path Sub directory where the files containing the spectra are located Full Spectra Directory is dirSpectra Path Except if Path full AND if FileName is a List the file names given in the list must be given with their full path SPn Filename Filename for observed spectrum OR name of the file containing a List of filenames Full path Normally the full path to the FileName is dirSpectra Path FileName except if Path full in that case the full path to FileName is dirSpectra FileName SPn FileType Type of file None No file given only forward model Opus OPUS file Ascii Ascii file Csv CSV file with no header only 2 columns separated Opusinfo csv CSV file with header from OPUSINFO IMG csv IMG ascii ACE ACE single spectrum Iasi IASI Level 1 spectra with IASI header SOIR csv SOIR csv with header SOIR PDS3 SOIR file PSA PDA Level 3 format SPn List Determines if FileName contains one single measurement file or a list of files This option is only valid for FileType Ascii opus img csv ACE Soir csv and Soir PDS3 No 0 Not a list default Yes 1 The FileName parameters points to a list of files that will be treated sequentially and independently Occ 2 The FileName parameters points to a list of files that will be tr
38. with the 1 order approximation method both under WINDOWS and UNIX only full mixing method under WINDOWS The full mixing requires the inclusion of the LAPACK in C library Ref Lamouroux et al 56 3 2 1 4 Molecules options Molecules fileHITRAN Localization of the general HITRAN file used if no file is specified for th molecul under MolecLP File and MolecLP Type is hitran See Annex 2 for the description of the HITRAN format Molecules fileGEISA Localization of the general GEISAfile used if no file is specified for th molecul under MolecLP File and MolecLP Type is geisa See Annex 3 for the description of the HITRAN format Molecules PartitionFct Method used to calculate the Partition function TIPS matrix or TPOLY temperature polynomials equ 8 3 2 1 5 xxxLP options In the name of these sections xxx represent the name of the molecule ex O3 CO2 NewO2 formaldehyde LP stands for Line Parameter xxxLP DBname It is possible to use a name xxx for the molecule internally defined in ASIMUT but non existing in the database usually used For example one can used the O3isol name it will be recognized by ASIMUT but not in the HITRAN database which only recognizes xxxLP DBname is the name to be found in the database xxxLP ATMname It is possible to use a name xxx for the molecule internally defined in ASIMUT but non existing in the atmos
39. 1 with E solar irradiance at the top of the atmosphere of the planet Isun where Rps distance Planet Sun radius of the Sun at the photosphere 6 96x10 m and Is is the Sun radiance see Annex 4 Lh cos sza T Z Zp 4o is the atmosphere transmittance from the top of the atmosphere z to the lower boundary 2 along the Sun Line of path defined emissivity of the planet a albedo of the planet if not specified a 1 8 The surface contribution for the atmospheric emission is characterized either as a Lambertian or specular surface reflection In the Lambertian case the surface BRDF is independent of the angle and can be expressed simply in terms of albedo which when multiplied by the downwelling radiance computed along the diffusivity ray secant 1 66 provides an good approximation for the upwelling isotropic radiance 71 at the surface The optical depth of one layer along the diffusivity ray is obtained by multiplying the nadir off nadir optical depth by the secant of the diffusivity angle In ASIMUT the user can specify the type of surface reflectivity used for the reflection of the radiance emitted by the atmosphere LambertSurfAtmEmi yes 3 3 4 Determination of the Transmittances The transmittance associated to the calculated Radiance is determined using Radiance Transmittance 3 72 0 in which Io is the radiance from the source When
40. 111 Isotope 999 666 xxxLP isoRatio Isotopic ratio for the species 0 1 0 If no number is given then it takes the isotopic ratio on Earth HITRAN values Not applicable for isotope 999 always 1 0 xxxLP limite Limit for the intensity value All line with intensity lt limit will be rejected xxxLP model Model for a priori profile can be different than the general value See 3 1 2 1 AtmosphericModels options for a complete description of the possible values xxxLP ATMfile File containing the atmospheric vertical profile for the species Must be given if model for the species is set to 0 and species not present in the AtmosphericModels model file or if this profile should supersede the one in that file Format of file 2 col Cod 5L 22 Col 2 vmr ppm The following parameters defines the Line profile to apply for this molecule xxxLP LProfile Type of line profile applied Different profiles can be selected VoigtPade VoigtHumlicek VoigtFaddeeva Rautian Galatry Lorentz xxxLP FBord Spectral interval on which the line profile is calculated Must be lt 400 If not specified internal value 25 cm of ASIMUT is used xxxLP ChiFactor Name of file containing a Chi Factor 2 col Col 1 delta wavenumber cm 1 nu nucenter Col 2 Chi fct max 1 at center The Chi factor must be given for negative and positive delta wavenum
41. 166 0 995107 O 34 1 6 0 997628 CIONO 35 1 5646 0 74957 107 2 7646 0 239694 NO 36 1 46 0 993974 HOBr 37 1 169 0 505579 2 161 0 491894 CjH 38 1 221 0 977294 2 231 0 0219595 CH OH 39 1 2161 0 98593 CH3Br 40 1 219 0 500995 2 211 0 487433 CH3CN 41 1 2124 0 973866 CF4 42 1 29 0 988890 New in HITRAN 2008 edition Molecules only present in GEISA Molecule Isot Isot Geisaid Abundance Nb code GeH4 43 1 411 26 1 00 C3H8 44 1 221 28 1 00 C2N2 45 1 224 29 1 00 C4H2 46 1 211 30 1 00 HC3N 47 1 124 31 1 00 C3H4 48 1 341 40 1 00 Molecules present in CFGL Molecule Isot Isot Cfglid Abundance Nb code HONO 49 1 1 24 1 00 HO2NO2 50 1 1 25 1 00 N205 51 1 1 26 1 00 52 1 1 29 1 00 CCL2F2 53 1 1 32 1 00 CCL3F 54 1 1 33 1 00 CH3CCL3 55 1 1 34 1 00 4 56 1 1 35 1 00 COCLF 57 1 1 37 1 00 CHF2CL 58 1 1 42 1 00 COCL2 597 1 1 43 1 00 CH3I 60 1 1 45 1 00 CHCL2F 61 1 1 48 1 00 OCIO 62 1 1 58 1 00 F134A 63 1 1 59 1 00 F142B 64 1 1 61 1 00 CFC113 65 1 1 62 1 00 F141B 66 1 1 63 1 00 C2H6PL 67 1 1 66 1 00 108 PAN 68 1 1 67 1 00 Molecule ID in CFGL H20 1 CO2 2 O3 3 N20 4 CO 5 6 O2 7 NO 8 SO2 9 NO2 10 NH
42. 182787 0 962610146433146 6 63944157657772 006 0 0789788682854946 4 12654538910255e 006 0 281796550292296 9 48605838175013 005 0 0353052533554235 2 67081772054492 005 0 232077580267617 4 72433590075699 005 Which defines the 15 parameters needed for the construction of the aotf function consisting of the sum of 5 sinc 2 AOTF 5SINC2 5 first lines fwhm of the 5 sinc then 5 lines for the displacement relative to the center of the Aotf fct finally 5 lines for the intensity factor For each parameter 2 coefficients are given defining a linear variation of the coefficients with Wn For ex the FWHM of the first sinc wil be given by 23 7537542783189 0 000164628838726562 x aotf nuCentral FileAotfFilter Type FILEPDS Same format as the one given in the PDS ESA Archive One file per binning and bin AOTF filter bandpass function Orderl binl WaveNb start WaveNb end pas value 1 value N Orderl bin2 WaveNb start WaveNb end pas pas value 1 value N Order2 binl WaveNb start WaveNb end pas pas value 1 value N Order2 bin2 WaveNb start WaveNb end pas pas value 1 value N ex AOTF TF BINNING12 TAB 101 1 100 0 100 0 0 1 1 4095e 03 1 3764e 03 1 3456e 03 101 2 100 0 100 0 0 1 2 0169e 03 2 0290e 03 2 0436e 03 SPn FileTuningFct File containing the AOTF tuning functions freq wavenb Wavenb a bxfreq tcxfreq 2 FileTuningWavenb format The format is the sam
43. 2 2 5 for more information about how to include the LAPACK library See also the xxxLP LineMixing Option Table 2 List of pre processors used by ASIMUT ALVL f neither MATLAB nor MATLABRUNTIME are present all calls to MATLAB plotting functions are non functional This is the default option no MATLAB See Section on how to link ASIMUT and MATLAB Section 2 2 4 for more details 10 2 2 3 Compiling under WINDOWS Under Windows only the functionalities of ASIMUT are available no links to those of V LIDORT or SPHER TMATRIX are possible The physical processes thus encompass only non scattering atmospheres Aerosols are included but only as extinctions The lines of code corresponding to calls to V LIDORT or SPHER TMATRIX functions are between pragma ifdef UNIX ftendif or ifdef LIDORT or ifdef VLIDORT ftendif The programme compiles without any problem but none of the calls to V LIDORT F90 or SPHER T MATRIX functions are possible In that case only the option Code ASIMUT in the input ASI file is working properly see ASIMUT Input File pdf for the complete description of all control parameters accepted by ASIMUT V LIDORT The package contains the project definition defined under MS Visual 2010 ASIMUTNASIMUTNVisualCN Asimut sln However the code does not contain any call to functions specific to the Visual compiler and it should be possible to compile it with any other compile
44. 3 11 HNO3 12 OH 13 HF 14 HCL 15 HBR 16 HI 17 CLO 18 OCS 19 H2CO 20 HOCL 21 HO2 22 H202 23 HONO 24 HO2NO2 24 N205 26 CLONO2 27 HCN 28 CH3F 29 CH3CL 30 CF4 31 CCL2F2 321 CCL3F 33 CH3CCL3 34 CCLA 35 COF2 36 COCLF 37 C2H6 38 C2H4 39 C2H2 40 N2 41 CHF2CL 42 COCL2 43 CH3BR 44 CH3I 45 HCOOH 46 2 47 CHCL2F 48 HDO 49 SF6 50 2180 51 H2170 52 CH3D 53 03668 54 03686 55 03667 56 O3676 57 OCLO 58 F134A 59 C3H8 60 F142B 61 CFC113 62 F141B 63 CH3OH 64 CH3CN 65 C2H6PL 66 PAN 67 109 Annex 2 HITRAN format HITRAN Database Format 1986 2001 The following table illustrates the format used for the spectroscopic parameters of the HITRAN database between 1986 and 2001 Field 3 9 9 HITRAN Database Format 2004 present The following table illustrates the format used for the spectroscopic parameters starting with the HITRAN 2004 Edition This format is 160 characters per record line transition Note the additional parameters compared to earlier editions Field Length Datatype Integer Integer Real Real Real Real Real Real Real Real Text Text Text Text Integer Integer Text Real Real In the folder IR XSECT are placed files of IR cross sections the definition and units being described in previous
45. 4 44 QU Wulf g Wulf Jno 3 47 with pro 3 48 n 0 where n and n are the densities of respectively and O2 This expression is valid up to 41500 cm However as stated in 46 the uncertainty on the Herzberg and Wulf continua is high above 41000 The validity of Eq 3 47 has been verified up to 41227 cm However the Herzberg continuum extends well above 42000 cm The formalism used in LBLRTM based on Yoshino s data 47 has been considered above 41227 3 2 3 2 N continua Collision induced band at 350 Based on the works of Borysov and Frommhold 48 and of Boissoles et al 49 the modeling of the collision induced absorption coefficient in air takes into account the temperature dependence of the absorption Collision induced absorption in the fundamental band Lafferty et al 50 carried out measurements of collision induced absorption of pure nitrogen in the fundamental band near 4 3 um The measurements were performed under different pressure 0 10 atm and temperature 230 300 K conditions Using available literature data in the 190 300 K temperature range Lafferty et al built a simple empirical model to compute the absorption The dependence of the absorption for pure towards temperature can be represented by the following expression 63 By Q T By y en V e 1 3 49 where By y is the absorption at the reference temperature T
46. 5 21 p 5191 5206 Rodgers C Inverse methods for atmospheric sounding Theory and practice World Scientific ed N J Hackensack 2000 University of Oxford Vandaele A C M Kruglanski and M D Mazi re Modeling and retrieval of Atmospheric spectra using ASIMUT in Proc of the First Atmospheric Science Conference ESRIN Frascati Italy 8 12 May 2006 ESA SP 628 July 2006 2006 Spurr R T Kurosu and K Chance A linearized discrete ordinate radiative transfer model for atmospheric remote sensing retrieval J Quant Spectrosc Radiat Transfer 2001 68 6 p 689 735 Spurr R J D VLIDORT A linearized pseudo spherical vector discrete ordinate radiative transfer code for forward model and retrieval studies in multilayer multiple scattering media J Quant Spectrosc Radiat Transfer 2006 102 p 316 342 Mishchenko M I and L D Travis Capabilities and limitations of a current Fortran implementation of the T matrix method for randomly oriented rotationally symmetric scatterers J Quant Spectrosc Radiat Transfer 1998 60 3 p 309 324 Bertaux J L et al A warm layer in Venus cryosphere and high altitude measurements of HF HCl H O and HDO Nature 2007 450 29 November p 646 649 doi 10 1038 nature05974 Kruglanski M et al Retrieval of formaldehyde from FTIR spectra preliminary results in NDACC IRWG meeting 2007 Puerto de La Cruz Tenerife Vandaele A C et al Composition of the Ven
47. 80 Ciddor P E Refractive index of air new equations for the visible and near infrared Applied Optics 1996 35 9 p 1566 1573 Tomasi C et al Improved algorithm for calculations of Rayleigh scattering optical depth in standard atmospheres Applied Optics 2005 44 16 p 3320 3341 Sneep M and W Ubachs Direct measurement of the Rayleigh scattering cross section in various gases J Quant Spectrosc Radiat Transfer 2005 92 3 p 293 310 U S Standard Atmosphere 1976 U S G P Office Editor 1976 Washington D C Remedios J J et al MIPAS reference atmospheres and comparisons to V4 61 V4 62 MIPAS level 2 geophysical data sets Atmos Chem Phys Discuss 2007 7 4 p 9973 10017 Goody R M Atmospheric Radiation 1964 Oxford University Press Fischer J et al Total internal partition sums for molecular species in the 2000 edition of the HITRAN database J Quant Spectrosc Radiat Transfer 2003 82 1 4 p 401 412 Gamache R R R L Hawkins and L S Rothman Total internal partition sums for atmospheric molecules in the temperature range 70 2005 K Atmospheric linear molecules J Mol Spectrosc 1990 142 2 p 205 219 Kuntz M A new implementation of the Humlicek algorithm for the calculation of the Voigt profile function J Quant Spectrosc Radiat Transfer 1997 57 6 p 819 824 Wells R J Rapid approximation to the Voigt Faddeeva function and its derivatives J Quant Spectrosc Rad
48. 93 4 448 0 00198582 5 447 0 000369280 CO 5 1 26 0 986544 2 36 0 0110836 3 28 0 00197822 4 27 0 000367867 5 38 0 000022225 6 37 0 00000413292 CH 6 1 211 0 988274 2 311 0 0111031 3 212 0 000615751 4 312 0 00000691785 7 1 66 0 995262 2 68 0 00399141 3 67 0 000742235 8 1 46 0 993974 2 56 0 00365431 106 3 48 0 00199312 SO 9 1 626 0 94568 2 646 0 0419503 NO 10 1 646 0 991616 NH 11 1 4111 0 9958715 2 5111 0 00366129 12 1 146 0 989110 OH 13 1 61 0 997473 2 8l 0 00200014 3 62 0 000155371 HF 14 1 19 0 99984425 15 1 15 0 757587 2 17 0 242257 HBr 16 1 19 0 506781 2 11 0 493063 HI 17 1 17 0 99984425 CIO 18 1 56 0 755908 2 76 0 241720 OCS 19 1 622 0 937395 2 624 0 0415828 3 632 0 0105315 4 623 0 00739908 5 822 0 00187967 H CO 20 1 126 0 986237 2 136 0 0110802 3 128 0 00197761 Q1 1 165 0 755790 2 167 0 241683 22 1 44 0 9926874 HCN Q23 1 124 0 985114 2 134 0 011076 3 125 0 00362174 CH3Cl 24 1 215 0 748937 2 217 0 23949 25 1 1661 0 994952 CH 26 1 1221 0 977599 2 1231 0 0219663 C2H6 27 1 1221 0 97699 PH 28 1 1111 0 99953283 COF 29 1 269 0 98654 SF 30 1 29 0 95018 H2S 31 1 121 0 949884 2 141 0 0421369 3 131 0 00749766 HCOOH 32 1 126 0 983898 33 1
49. ALVL ASIMUT V LIDORT Radiative transfer modelling and spectrum retrieval in a non scattering atmosphere ASIMUT and scattering atmosphere V LIDORT User Manual and Science description Version 1 2 Release 11 0 of ASIMUT Institut d A ronomie Spatiale de Belgique 3 av Circulaire 1180 Brussels Belgium Abstract ASIMUT is a modular program for radiative transfer calculations in planetary atmospheres The ASIMUT software has been developed to exploit the synergy existing between the growing number of different instruments working under different geometries The main particularities of the software are The possibility to retrieve columns and or profiles of atmospheric constituents simultaneously from different spectra which may have been recorded by different instruments or obtained under different geometries This allows the possibility to perform combined retrieval e g of a ground based measurement and a satellite based one probing the same air mass or from spectra recorded by different instruments on the same platform 1 The analytical derivation of the Jacobians The use of the Optimal Estimation method OEM using diagonal or full covariance matrices iv Its portability v Its modularity hence the ease to add future features Initially developed for the Earth atmosphere its applicability has been extended to extra terrestrial atmospheres such as those of Mars and Venus ASIMUT has been coupled t
50. ANCE SUPPORT UPDATES ENHANCEMENTS OR MODIFICATIONS TO THE ASIMUT SOFTWARE As the ASIMUT Software is provided coupled to the V LIDORT SOFTWARE please read also the V LIDORT licence agreement 126
51. NEWINST GetNbSpectra GetNbSpec NEWINST LIST GetNameSpectrum GetNameSpec NEWINST LIST ModifySpectrum ModifySpec NEWINST LIST For option 3 the Read NEWINST GetNbSpec NEWINST OCC GetNameSpec NEWINST LIST and ModifySpec NEWINST LIST functions are needed in the NEWINSTRUMENT lt c file Add the following lines inside User Choice c in the ReadSpectrum_Choice function for example just before the else else if SameString ChoiceStr NEWINST OCC IsList OCC type FILE NEWINST ReadFct Read NEWINST GetNbSpectra GetNbSpec NEWINST OCC GetNameSpectrum GetNameSpec NEWINST LIST ModifySpectrum ModifySpec NEWINST LIST 40 2 4 Troubleshooting 2 4 1 WIN Debugging information for project exe cannot be found or does not match Binary was not built with debug information To enable debugging Go to Project Project Properties On the left expand Configuration Properties On the left expand Linker On the left select Debugging On the right change Generate Debug Info to Yes Click ok Set your breakpoints Rebuild your application oe pae 242 WIN No compilation Fatal error C1033 cannot open program database Be sure that you are Administrator on your machine or that you have started VisualC with the option Run as Administrator right click 243 WIN UNIX The figures from MATLAB are created but are empty This error is also a
52. Nn EmissivityFile option SPn_FENn EmissivityFile name of the file containing the emissivity The file must contain 2 columns Col 1 Wavenumber cm 1 Col 2 emissivity The file must be located in the dirPlanet directory SPn FENn Albedo Albedo of the planet constant value If not given Albedo 1 0 Emissivity including the wavelength dependence if given for Emissivity Only constant value is valid for the moment SPn FENn NeglectThermalSource indicates if the thermal emission source in the atmospheric layers are neglected no test is performed to check if it is meaningful to neglect the thermal emission source Supersedes what is specified in SPn SPn_FENn NeglectThermalReflection indicates if the thermal emission from the atmospheric layers reflected on the surface is neglected no 73 test is performed to check if it is meaningful to neglect the thermal Source Supersedes what is specified in SPn SPn FENn IncludeSolarSource indicates if the solar source are included reflection on surface Supersedes what is specified in SPn AtmosphereEmission Function used to derive the emission from the atmosphere Supersedes what is specified in SPn Padel Pade with 1 parameter Pade2 Pade with 2 parameters Taufct More complex function considering the full dependence of B with tau Bm B A at the average temperatur SPn LambertSurfAtmEmi Indicates if the surface
53. ONS E 89 NM Torsten ME EEUU 90 5 1 General deserptionm aed pda eects 90 SAT Ta eds Dest hie 91 5 2 Determination of the 92 5 3 priori covariance matrices 92 5 3 1 Covariance options in xxxLP xxxXS and xxxAER 93 5 4 Summary of all Variables that can be 94 5 4 1 Molecules cross sections and aerosols densities 94 5 4 2 Variables related to the Spectrum and its windows 94 CS EMD OR T descripti On capter e a acia aes eee vero 96 7 8 6 1 Choice between LIDORT and VLIDORT 96 6 2 AETOSOIS nosse iit casas Gees eren Sors Mise Laudo iore hd 96 6 2 1 Modified gamma distribution both codes 97 6 2 2 Log normal distribution both codes 97 6 2 3 Power law distribution both codes 97 6 2 4 Gamma distribution both 97 6 2 5 Modified power law distribution both 97 6 2 0 Bimodal volume log norma
54. Solution Explorer right click on Asimut choose Add gt Existing Item go to the MATLAB extern lib directory and 17 select the libeng lib libmat lib and libmx lib libraries Raima Microsoft Visual e MD Express Admamectrato es ES Ede View Project Buld Debug Tools e Window MEI EI 415441555 9 Oss i Dekel REIS ZIONS Aib aa gt 913 return Molecule gt Linettixing t oLines osam 2 Specify to the compiler where to find the libraries and the related Include In the Solution Explorer right click on Asimut select Properties Under Configuration Properties gt VC Directories Select Include Directories gt lt Edit gt Add the line C Program FilesWMATLAB R2007b extern include MATLAB Include In the Solution Explorer right click on Asimut select Properties Under Configuration Properties gt VC Directories Select Library Directories gt lt Edit gt Add the line C Program FilesNAMATLABNR2007IN exterrNib extern lib 18 Asiera Microsoft Visual C Express Directories Path to use when searching for induce files while bulding a VC project Comesponds to envionment vanable 1f Molecule tinetixing AMULL return M
55. UMENT Inst GEOMETRY Geo Int ZPTfromSpectrum brief Modify Spectrum Instrument or Geometry in accordance to the string ModDesc param in Spectrum Spectrum structure param in ModDesc Descriptio of the measurement read from the list file param in Inst Instrument structure param in Geo Geometry structure param in ZPTfromSpectrum 1 if ZPT information in this file return type of error int ModifySpec_NEWINST_LIST SPECTRUM Spectrum ModDesc INSTRUMENT Inst GEOMETRY Geo Int 7pTfromSpectrum fn GetNbSpec NEWINST OCC char FileName int SamePass int First brief Get number of spectra from FileName containing a list of spectra defining an occultation param in FileName name of list file param in SamePass 1 if all spectra are analysed simultaneously param in First index to the first spectrum return nb of spectra int GetNbSpec_NEWINST_OCC char FileName int SamePass int First 124 Annex 7 Contacts and rights of use Please contact the Belgian Institute for Space Aeronomy to obtain the access rights to use ASIMUT Contacts Ann C Vandaele a c vandaele aeronomie be Tel 32 2 373 0 367 Martine De Maziere martine demaziere Q aeronomie be Director General Tel 32 2 373 0 363 ASIMUT is the results of the work of many contributors Nicolas Kumps Martine De Maziere Evelyn Dewachter Rachel Drummond Caroline
56. User Choice h Each modification will be explained in the following for all three types of NEWINSTRUMENT files First add the new instrument name in the existing list In User Choice h add the name all capitals at the end of the enum field lines 44 sq defining SpecType Type of Spectrum File enum SpecType FILE NONE FILE OPUS FILE ASCII FILE SFIT FILE CSV FILE OPUSINFOCSV FILE IMGCSV FILE IASICSV FILE SOIRCSV FILE SOIRPDS3 FILE ACE Becomes Type of Spectrum File enum SpecType FILE NONE FILE OPUS FILE ASCII FILE SFIT FILE CSV FILE OPUSINFOCSV FILE IMGCSV FILE IASICSV FILE SOIRCSV FILE SOIRPDS3 FILE ACE FILE NEWINST For option 1 only the Read NEWINST function is needed in the NEWINSTRUMENT c file 39 Add the following lines inside User Choice c in the ReadSpectrum_Choice function for example just before the else else if SameString ChoiceStr NEWINST amp amp IsList NO LIST type FILE NEWINST ReadFct Read NEWINST For option 2 the Read NEWINST GetNbSpec NEWINST LIST GetNameSpec NEWINST LIST and ModifySpec NEWINST LIST functions are needed in the NEWINSTRUMENT lt c file Add the following lines inside User Choice c in the ReadSpectrum_Choice function for example just before the else else if SameString ChoiceStr NEWINST amp amp IsList LIST type FILE NEWINST ReadFct Read
57. _lib directory and select the mclmcerrt lib library 2 repeat and go to the ASIMUT Lib directory and select the libasimutplot lib library 2 Specify to the compiler where to find the libraries and the related Include In the Solution Explorer right click on Asimut select Properties Under Configuration Properties gt Directories Select Include Directories gt lt Edit gt Add the 2 lines C Program FilesNAMATLABNR2007bVexternNnclude MATLAB Include AsimutbinNncludeNWin64 2 ASIMUT Include In the Solution Explorer right click on Asimut select Properties Under Configuration Properties gt Directories Select Library Directories gt lt Edit gt Add the 2 lines C Program FilesNAMATLABNR2007IN externNib MATLAB_extern_lib Asimu bin Lib Win64 ASIMUT_Lib 3 Specify to the compiled ASIMUT EXE where to find the dll related to the libraries The easiest method is to change the Environment Settings including into the definition of PATH where to find the different dll You need to add the 2 following sub directories 21 C Program FilesMATLAB R2007b runtime MATLAB runtime Asimut bin Runtime Win64 ASIMUT dll Path Path MATLAB runtime ASIMUT dll Another method consists in changing the environment within VISUAL In the Solution Explorer right click on Asimut select Properties
58. a list of spectra defining an occultation INPUT FileName name of list file SamePass 1 if all spectra are analysed simultaneously First index to the first spectrum OUPUT nb of spectra MEE AG A FEE sepe oo ok ooo int GetNbSpec NEWINST OCC char FileName int SamePass int First int char Line NL FILE hFile char str 1024 First 0 if hFile fopen FileName rb NULL sprintf str Can not open file Xs FileName set_Error ERR_FILE_NONE SEV STOPPROG LINE FILE str return 0 while fgets Line NL 1 hFile nb SamePass YES all spectra analysed simultaneously fclose hFile return nb 122 Template for the NEWINSTRUMENT h file PE AE ER ERT aede Ie ee eee EG sie eoo NEWINSTRUMENT h DESCRIPTION all functions relatives to NEWINSTRUMENT This program is part of the ASIMUT Project x X Seco dee dob edel deb bb deo obo dob Ie ees file NEWINSTRUMENT h brief Functions relative to NEWINSTRUMENT Created 07 03 2006 Rev 257 Last modified Date 2009 10 01 16 31 12 by Author xxx TA ifndef newinstrument h define newinstrument h fn Rea
59. aling with aerosols such as 1 multiplying the profile by a factor zFact and Fact in the INP file or imposing a total optical depth TotalOpacity 2 saving in the out files and in the optical depths files 3 fitting of the aerosol densities as a profile or as a column the full Rodgers treatment including full Sa matrix In one single run different aerosols of different types can be defined the classical ones will be calculated by ASIMUT and the Lidort or scattering ones by V LIDORT can be fitted for the moment only their density or OD The transmittance corresponding to the aerosol extinction in layer L is calculated by T uL exp fx 3 55 where Taer is the aerosol optical thickness optical depth Angstr m formalism Type Angstrom The aerosol optical thickness in the layer L is calculated using uam V x zi 3 ZE 9 3 56 with vo the reference wavenumber the optical depth at this wavenumber and in layer L of width As km is the extinction coefficient of layer L km given in kext input parameters defined on scale zFact a the Angstr m Coefficient Modified Angstr m formalism Type ModAngstrom The aerosol optical thickness in the layer L is calculated using Taer Tao f V Kio As 3 57 with the reference wavenumber the optical depth at this wavenumber and in layer L of width As km KoL
60. as the apparent or astronomical angle In the latter case an iterative procedure is applied to determine the apparent angle Off axis nadir Ground based To space 5 2 Solar occultation LimbAngle LimbZtg Figure 3 Several geometries are possible is the altitude of the observer gis the zenithal angle H is the tangent height and is the end point of the ray path The user specifies a layer structure on which the optical densities and radiative transfer calculations will be performed The ray tracing calculations are carried out on a finer grid with a 200 m step and the final results i e the effective temperature and pressure for each layer as well as the air density are obtained using the Curtis Gordon approximations 28 3 1 3 1 Geometry options All parameters describing the Geometry of the observation are indicated under the SPn section 46 SPn Geometry indicates the type of geometry of the observation horizontal path nadir looking nadir off axis limb observation LimbAngle when the angle or LimbZtg when the tangent height is Specified or looking to space SPn Refraction indicates if refraction of the atmosphere has to be considered for the Ray tracing calculations For example not necessary on Mars SPn HObs indicates the altitude of Observer km It MUST be given except for nadir and offaxis If not given for nadir and offaxis the space limit
61. at which the instrument is looking at It can be none planet or earth mars venus and sun SPn NeglectThermalSource indicates if the thermal emission source in the atmospheric layers are neglected no test is performed to check if it is meaningful to neglect the thermal emission source SPn NeglectThermalReflection indicates if the thermal emission from the atmospheric layers reflected on the surface is neglected no test is performed to check if it is meaningful to neglect the thermal source SPn IncludeSolarSource indicates if the solar source are included reflection on surface SPn AtmosphereEmission Function used to derive the emission from the atmosphere Padel Pade with 1 parameter default Pade2 Pade with 2 parameters Taufct More complex function considering the full dependence of B A with tau Bm B A at the average temperatur SPn LambertSurfAtmEmi Indicates if the surface is Lambertian for the reflection of the Atmospheric Emission 3 3 5 3 Radiative options in SPn_FENn SPn_FENn Ts temperature of the source K If a number is given it represents the temperature If the option zpt is chosen Ts will b ither given in the ZPT file if not T of the lowest altitude in the ZPT file will be considered SPn FENn Emissivity gives the emissivity value for the surface constant value It will not be used if File is given SPn_FE
62. ata obtained for different planetary atmospheres In the case of SOIR data it consisted in generalizing the ray tracing module to another planet but also in including the possibility to simulate echelle spectra in particular the existence of overlapping diffraction orders This paper describes the different parts constituting the ASIMUT software The simulation module consists in a ray tracing unit units devoted to the calculation of the optical depths and of the radiative transfer and a final component related to instrumental effects The retrieval module is based on the Optimal Estimation Method 7 and has the peculiarity to use the Jacobians determined analytically during the forward radiative transfer modelling The number of runs of the forward model during the retrieval process is therefore much reduced Recently the ASIMUT software 8 has been coupled to the V LIDORT 9 10 package as well as to the SPHER TMATRIX 11 software to correctly include scattering effects in the Radiative transfer calculations In this paper we present the different blocks defining the new code The forward modeling as well as retrieval capabilities of ASIMUT will be described and discussed in a series of papers in preparation or already submitted and related to specific scientific research objectives 12 15 2 General Description of the software 2 1 Implementation The ASIMUT program has been written in the C programming language The modularity
63. b gt make USE LIDORT MATLAB YES without Matlab gt make USE LIDORT MATLAB NO to compile using only ASIMUT with Matlab gt make USE ASIMUT MATLAB YES without Matlab but using gcc and gfortran gt make USE ASIMUT MATLAB NO CC gcc FC gfortran do not forget to use make clean when you change the compiling option indicate here the C and Fortran compilers CC supported icc gcc FC supported ifort gfortran if you use other compilers do not forget to define the options CFLAGS for CC FFLAGS for FC Default values 23 FC ifort USE LIDORT MATLAB NO MATLABRUNTIME YES 3 indicate here where the MATLAB libraries are located MATLABLIBPATH opt matlab bin glnxa64 MATLABRUNPATH opt matlab runtime glnxa64 MATLABINCPATH opt matlab extern include ASIMUTLIBPATH bin Lib Unix ASIMUTINCPATH bin Include Unix ifeq MATLAB NO MCFLAGS DNOMATLAB endif ifeq MATLAB YES ifeq MATLABRUNTIME NO MCFLAGS DMATLAB I MATLABINCPATH 5 MFFLAGS L MATLABLIBPATH lmx lmat leng lut W1 rpath S MATLABLIBPATH else MCFLAGS DMATLABRUNTIME IS MATLABINCPATH I ASIMUTINCPATH MFFLAGS L MATLABLIBPATH L MATLABRUNPATH L ASIMUTLIBPATH lasimutplot lmwmclmcrrt Wl rpath MATLABLIBPATH
64. be implemented in the near future 97 Particle size distributions 40 1 Spheres 0 025 um In o 0 2 H 2 Spheroids r 0 04 um In o 0 2 2 0 0 INL err Ven 0 00 0 02 0 04 0 06 0 08 0 10 0 12 0 14 Radius r uum Figure 6 Example size distributions of Martian dust particles Size distribution parameters were taken from Dlugach et al 2002 Both SPHER and T MATRIX have been slightly modified to become part of ALVL In particular their input parameters now come from outside the codes instead of being specified inside as before Also a special function was added to calculate the boundaries of a radius integration range on the basis of two accuracy parameters specified by the user For example let s consider the size distributions shown in Figure 6 There is an option to cut the tails of the distribution as shown with the red line The minimum and the maximum radii rl and r2 will be then calculated as values at the cross points of the cut line and the size distribution curves The integration will be performed over a smaller range which will result in a significant decrease in calculation time The two accuracy parameters which the user has to specify are the value of cutoff for a given n r and the accuracy of rl and r2 calculation Aerosol parameters extinction cross sections SSA and expansion coefficients do not change with wavenumber as rapidly as gas absorption parameters and therefo
65. behaviour the opacity far from the line centre is less than that predicted by a Lorentzian profile whereas H2O shows a super Lorentzian behaviour The x function is usually defined on large spectral intervals and must be provided by the user 3 2 1 1 Grid and resolution used for the OD simulation Sampling of the lines The grid on which the profile must be determined must be fine enough so that the narrowest line be adequately represented In the upper atmosphere the Doppler width is the limiting factor for selecting an adequate sampling value As Doppler width depends on temperature it varies with altitude the line profile must therefore be sampled with different steps as the altitude varies ASIMUT determines for each layer of temperature T and pressure P the optimized sampling step as Av P T aX P T 3 25 with ap the Doppler width obtained for a molecule of mass 20 0 and the Lorentzian width obtained for a molecule characterized by no self broadening a foreign broadening of 0 04 cm and a temperature coefficient of 0 5 However for some particular temperature and pressure conditions this sampling step might be too large compared to the desired final resolution In that case the sampling is set to the value of final resolution 15 Line By Line LBL calculations often require the computation of a large number of line shapes over large to very large spectral intervals The number of points might
66. ber can be asymmetric xxxLP Collision Collision parameter used for Rautian profile xxxLP shift Constant shift applied on all lines cm The following parameters supersede the values found the spectroscopic file xxxLP gammaself Self broadening parameter cm 1 atm 2 Do nothing take HITRAN GEISA values as they are Take value of GammaAir 0 0 All GammaSelf 0 0 20 0 Default value when HITRAN GEISA are 0 0 xxxLP GammaAir Default value when HITRAN GEISA are 0 0 cm atm xxxLP ShiftSelf Default value when HITRAN GEISA are 0 0 cm atm xxxLP ShiftAir Default value when HITRAN GEISA are 0 0 cm atm xxxLP ShiftSelf Default value when HITRAN GEISA are 0 0 cm t atm 58 xxxLP nSelf Default value when HITRAN GEISA are 0 0 xxxLP nAir Default value when HITRAN GEISA are 0 0 xxxLP zfact fact factors that can be applied on the vertical profile read from the atmosphere file xxxLP LineMixing Indicates if Line Mixing has to be included and what type of line mixing 0 n no No line mixing 1 y yes sine mixing with 1st order approximation full Line mixing full treatment only possible if the FULLLINEMIXING option is used at compilation and if the LAPACK library has been added to the project For the moment only implemented for WINDOWS runs xxxLP LineMixingDir Path where the data used for the line mixing calculation are located relative to Directories
67. ccompanied with the following warnings during the execution of the program 777 Undefined function or method PlotZPT for input arguments of type double Undefined function or method AddString for input arguments of type double 777 Undefined function or method PlotSpecSimul for input arguments of type double Check that ASIMUT MATLAB directory is included into the Path of MATLAB and saved To do that select the File menu then Set Path click on Add with Subfolders and then select X ASIMUT MATLAB Then Save Windows Note that you have to be Administrator on your machine to be authorized to save If you are not logged as Administrator you can right click on MATLAB and select the Run as Administrator option You will be asked to enter your Administrator login Another easier solution is to mention the ASIMUT MATLAB directory under the directories dirMatlab in the ASI file Directories dirMatlab C MMyProjectsNAsimutMatlab 244 WIN UNIX Problem with Files If you encounter a problem with files that apparently are in line with the input expected to be read from them check 1 That they are not open by another program If it is the case close that program and start ASIMUT again 2 Check that you have the right access permission to the files and directories including write permission for the output files 41 3 Check the endings of the line under Windows i
68. ck on the Green arrow or select Start debugging Under Windows outside the VISUAL environment Open a cmd window Change directory to where the executable is located Type Asimut fullPath file ASI 2 3 2 How to extract spectroscopic LP data from literature database In the ASI file set Run save lp Run StopAfter ExtractLP Under List specify one INP file representative of what will be done same number of FEN and the limits that will be used In the INP file do not forget to remove or place in comments using at the start of the line the lines defining the name of the files in the xxxLP sections Run Asimut the MOD files will be created with all the information needed for this specific call to Asimut The two files mars nadir Ip asi and IR CH4 test lp inp in the Example directory give you an example of such calling 2 3 3 How to limit the size of the LOG file Set Run verbose to a lower value 234 How to add a new Instrument To include a new instrument is easy but requires the modification of several files Fist of al specific NEWINSTRUMENT c and accompanying NEWINSTRUMENT h file has to be built following the template proposed in Annex 6 Template of file for a new instrument The content of this file will depend on what type of input files are read Several options are available 1 Only one file containing one observation spectrum 38
69. ctrum of the Mars atmosphere under nadir viewing CO2 CH4 and H20 are included as Line by Line molecules LP as well as two types of aerosols illustrating Angstr m type aerosols and LidortG type The examples have been run under ASIMUT windows and under UNIX with the ASIMUT and LIDORT options When using the ASIMUT option only the extinction of the aerosols are included no scattering effect in ASIMUT With the type Mode2 LidortG aerosol scattering due to these aerosols also included when using the LIDORT option For each INP file 3 sets of output files are created one using ASIMUT under Windows one using ASILID with the Run Code ASIMUT optin files ending by A and one using ASILID with the Run Code LIDORT optin files ending by 17 The plot RESULTS Comparison Aerosols fig and Comparison Aerosols png illustrates the comparison of the radiances obtained in the different runs 35 2 2 8 4 Analysis of a SOIR solar occultation observation Different test cases are given simulation of an AOTF like spectrum reading of a PDS SOIR spectrum reading of a CSV SOIR spectrum analysis of a series of individual spectra combined analysis of a series of spectra onion peeling Venus 149 asi 341 149 1 onlySimulation inp No observed spectrum is provided all information on the instrument and observation has to be specified in the INP file Venus 149 asi 341 149 1 singlePDSspectrum inp A SOIR PDS type of file is
70. d NEWINST char FileName INSTRUMENT Instrument GEOMETRY Geo int 7PTfromSpectrum brief Reading the IMG spectra from FileName param in FileName name of IMG file param in Instrument Instrument structure param in Geo Geometry of the measurement param in ZPTfromSpectrum 1 if ZPT information in this file return Matrix containing the spectrum wavenb Intensity ui Matrix Read NEWINST char FileName INSTRUMENT Instrument GEOMETRY Geo int ZPTfromSpectrum fn GetNbSpec NEWINST LIST char FileName int SamePass int First brief Get number of spectra from FileName containing a list of spectra param in FileName name of list file param out SamePass 1 if all spectra are analysed simultaneously param out First index to the first spectrum return nb of spectra int GetNbSpec NEWINST LIST char FileName int SamePass int First fn GetNameSpec NEWINST LIST char FileName int index char strN char Desc brief Get file name and description of spectrum index param in FileName name of list file param in index index of the spectrum in the list param in strN Complete filename to spectrum with path param in Desc decsription of the file in the list return type of error int GetNameSpec NEWINST LIST char FileName int index char strN char Desc 123 fos Vfn ModifySpec NEWINST LIST SPECTRUM Spectrum char ModDesc INSTR
71. d VLIDORT are capable of simulating plane parallel and pseudo spherical atmospheres The pseudo spherical approximation means that the solar beam attenuation is calculated in a curved atmosphere scattering effects are calculated in a plane parallel atmosphere Both LIDORT and VLIDORT are monochromatic codes Since ASIMUT operates with a range of wavenumbers a wavenumber loop was implemented to run LIDORT or VLIDORT at a chosen resolution Simulations at fine resolution require more time To have more information of the method used by LIDORT to calculate the radiative transfer budget and radiances please consult its User Manual lidort_3p3_userguide_v3_OInov07 pdf 6 1 Choice between LIDORT and VLIDORT Both codes have their own advantages and disadvantages The choice between them should mostly depend on the simulated spectrum VLIDORT is capable of accounting for the polarization of radiation in the atmosphere LIDORT in its turn is not able to account for polarization but it consumes less operational memory then VLIDORT The effects of polarization are known to be quite significant in UV visible and near IR while thermal emissions become important for wavelengths in excess of 2 4 um for Earth Thus it is advisable to use VLIDORT for the spectrum range before 2 4 um for Earth and LIDORT in all other cases 6 2 Aerosols LIDORT and VLIDORT can account for scattering and absorption events caused by the presence of aerosol
72. dirLP final path will b Directories dirLP LineMixingPath 59 3 2 2 Cross sections Heavy molecules give rise to generally dense absorption spectra which can not be resolved into single transitions Similarly absorption in the visible and the ultraviolet is also characterized by broad unresolved structures In those cases one defines the absorption through the cross section which depends on temperature and pressure ASIMUT can import cross section data from the HITRAN 38 and GEISA 39 databases as well as data given by the user As those data may be rather scarce in some cases in terms of different temperature and pressure conditions it was decided to use the data corresponding to the nearest couple T P present in the databases without any interpolation The contribution of the absorption cross sections to the absorption coefficient of a layer characterized by temperature T and total pressure P is the following c 3 43 where is the absorption cross section of species j whose cumulated density in the layer is n molec cm 3 2 2 Cross sections catalogues Absorption cross sections can be read from the HITRAN and GEISA compilations as well as from user defined files in which different data columns can be associated to different temperature and pressure conditions The import of cross section data is not limited to the species found in the HITRAN and GEISA databases
73. e binning option of the observation The N first columns are the wavenumber cm 1 and the following N ones contain the transmittance The MATLAB macro ConvertPDS ASCII m used to convert PDS file to CSV and to create some additional files such as the LST file giving the 83 list of all files converted is also provided in the ASIMUT MATLAB subdirectory The difference between SOIR csv and SOIR csv occ is that all files are analysed individually in the first case or all together in the second See also the examples provided SOIR PDS3 format These files correspond to the PSA ESA archive Level 3 data Their format is explained in the Archive in details One file contains all the data concerning one occultation observation On lin gives all the information and data on one Spectrum obtained on one bin The different information are the date time bin binning alt pointing angle degree distance to Venus tilted angle of the slit slit height latitude longitude local solar time lst speed of VEX wrt Sun speed of Venus wrt Sun Speed of Venus on Orbit error on the Altitude AOTF frequency integration time number of accumulation pixel wavenumber conversion coefficients 5 values maximum Transmittance 320 values error on Transmittance 320 values followed by 16 parameters describing the housekeeping 84 3 3 6 8 SPn FENx options SPn FENx Pass gives a PASS number to
74. e as the PSA PDS one Calibration freq AOTF wavenb TuningFct file forma cf ARCHIVE WN TAB with explanation of format in AOTF F WN LBL F 2WN binning bin number Param 1 Param 2 Param 3 Param 4 Param 5 WN F binning bin number Param 1 Param 2 Param 3 Param 4 Param 5 d Cf ex WN TAB F gt WN 3 1 0 0000e 00 0 0000 00 2 4852e 07 1 4577 01 3 5373 02 F gt WN 3 2 0 0000e 00 0 0000 00 2 4935 07 1 4563e 01 3 5526 02 SPn FileResolution File containing the resolution cm 1 of the SOIR instrument FileResolution format The format is the same as the PSA PDS one Resolution cm 1 file forma cf ARCHIVE CALIB RESOL BINNING12 TAB with explanation of format in AOTF F WN LBL 82 Order resolution in bin 1 Resolution in bin 2 ex RESOL BINNING12 TAB 101 1 0956e 01 1 0749e 01 102 1 1058e 01 1 0855e 01 103 1 1161e 01 1 0961e 01 104 1 1264e 01 1 1067e 01 SPn FileCalibWavenb File containing the parameters for calibrating the wavenumber scale The parameters needed for the conversion pixel wavenumber and wavenumber pixel are given Normally these parameters are also found in the headers of the SOIR PDS and SOIR CSV files In that case the later have priority ASIMUT will only read the PIX gt WN line and will compute the parameters for the Wn gt PIX conversion from it FileCalibWavenb
75. e compile link with the Preprocessor Definition NOMATLAB The program then works fine but of course no links to MATLAB are possible When using VC 2010 follow the next steps 1 Download and install the Windows Software Development Kit version 7 1 Visual 2010 Express does not include a 64 bit compiler but the SDK does A link to the SDK http msdn microsoft com en us windowsserver bb980924 aspx 2 Choose the X64 project configuration This is obtained by a Changing the project configuration Go to Properties of your project On the top of the dialog box there will be a Configuration drop down menu Make sure that selects Configurations There will also be a Platform drop down that will read Win32 Finally on the right there is a Configuration Manager button press it In the dialog that comes up find your project hit the Platform drop down select New then select x64 Now change the Active solution platform drop down menu to x64 When you return to the Properties dialog box the Platform drop down should now read x64 b Changing the toolset In the Properties menu of your project under Configuration Properties General change Platform Toolset from v100 to Windows7 1SDK 16 3 sure to add WIN64 preprocessor definition and to remove the NOMATLAB one 4 sure either to use a user ADMIN account or to start VC 2010 with the option Run as Administrator Then first lau
76. e observations Can be Cst Constant value either SPn FENx snr or SPn FENx snrabs fromSpectrumFile Noise is read from the Spectrum file only possible now for SOIR type of spectra fromFile From an external file whose name is Specified under SPn noiseFile SPn FENx snr snrabs Signal to noise ratio and absolute noise level At least one of them must be given used in the Rodgers formalism The Absolute snrabs must be given in the same units as the spectrum SPn FENx Sediag Indicates if Se noise on measurement is diagonal If not diagonal the non diagonal elements of the Se matrix are calculated by applying a Gaussian function with a width equals to the resolution of the instrument The following parameters describ which molecules LBL or cross sections continua or aerosols are included in the window SPn_FENx molecules Molecules LP type to be included Line by line calculations 85 Maximum number of Molecules that can be included in one run 129 number of recognized isotopologues SPn FENx FitMolecules For each molecule indicates type of fit Value is lt Fit only the column Value is No fit only simulation Value is gt 0 0 Fit as described in xxxLP description The decimal part of the number is the covariance for the fit if no Sa file is given 0 0 0 0 SPn FENx aPrioriMolecules For each molecule indicates the type of a priori that will be used in the Rodgers formalism
77. e particule density of the aerosol in layer L rep is the effective radius of the aerosols mm Qext is the extinction calculated by Rayleigh or Mie theory must be provided by the user Lidort formalism If the SSA OD and the aerosol scattering matrix coefficients are not given then ASIMUT will read the supplied Aerosol file fileAerosols and will call SPHER TMATRIX to calculate those quantities The aerosol optical thickness in the layer L is calculated using the cross section XS which is read from the file fileAerosolOD using XS N aer L AS 3 61 with Naer is the particule density part m of the aerosol in layer L If kext are specified then values of the aerosol optical depth for layer L are normalized to the value which is the extinction coefficient of layer L km and given in kext input parameters defined on scale zFact 67 LidortG formalism Henyey Greenstein One file containing SSA g and OD wrt wavenumber is read In that case the Legendre coefficients are calculated from the g parameters using the Henyey Greenstein formula The aerosol optical thickness in the layer L is calculated using the OD which is read from the file fileAerosol OD using T OD fromFile Niet aer L As 3 62 with is the particule density of the aerosol in layer L If kext are specified then values of the aerosol optical depth for layer L are normaliz
78. eINP rad forw dat and FileINP rad final dat If SPn DataType Radiance WaveNumber Simulated Radiance Observed Radiance Io Radiance Solar Radiance Source Radiance If SPn DataType Transmittance WaveNumber Simulated_Transmittance Observed_Transmittance Simulated_Radiance Observed_Radiance Io Radiance Solar Radiance Source Radiance The wavenumber column will be expressed in cm default or nm depending on the SPn DataXUnit flag All radiances are expressed in W cm cm sr If no Spectrum is provided by the user the columns Observed Radiance and or Observed Transmittance disappear Io Radiance radiance used for the calculation of the Transmittance It depends on the geometry for solar occultation SUN radiance for NADIR 1f no Sun reflection Surface emission o 1f Sun reflection included SUN radiance x SolarAngle Solar Radiance radiance of SUN if SUN is included in the RT calculation Source Radiance radiance of the Source indicated in the SPn section For solar occultation 2 SUN radiance for NADIR always Surface emission including emissivity The definition of Simulated_Transmittance depends on the geometry For solar occultation Simulated_Radiance SUN radiance for NADIR o ifno Sun reflection Simulated Radiance Surface_radiance o if Sun reflection Simulated Radiance x x SUN irradiance Radiance Factor 34 2 2 8 Examples Provided Different example
79. eated simultaneously as an occultation set Only valid for all 3 The FileName parameters points to a list of files that will be treated simultaneously Differences between List Occ Onion Peeling 76 If Run OnionPeeling no If List yes or 1 all the files are analysed sequentially on an independent manner This is an easy way to analyse a series a similar files with identical fit or simulation parameters If List occ or 2 all files listed are part of single observation one single solar occultation and are all fitted Simultaneously Only valid for an occultation set the altitudes of the tangent heights will define the RT altitude scale that will replace the Set zScale except for the altitudes above the highest tangent altitude If Run OnionPeeling yes If List yes all the files are analysed sequentially but using the onion peeling method The spectrum corresponding to the highest altitude is analysed first then its results are used as input for the analysis of the second one etc SPn DataType Observed spectrum is either a radiance or a transmittance SPn DataXUnit Units of the X scale of the spectrum Can be cm 1 Wavenumber nm Wavelenght For AOTF instrument only cm 1 is possible SPn Date SPn Time date and time of observation SPn Lon SPn Lat Longitude 360 Lon 360 and Latitude 90 Lat 90 of observation SPn FenList Index to the select
80. ecify the C compiler Default icc FC ifort gfortran Specify the Fortran compiler Default ifort clean c Remove all o file coming from c files clean f Remove all o file coming from f or f90 files in LIDORT VLIDORT VLIDORTF90 AUX SOURCE AUX SOURCE V and AUX SOURCE VLF90 clean bak Remove all bak files clean Remove all bak o files Table 3 List of calling convention under UNIX option VLIDORT and VLIDORTF90 are only accessible through the SVN repository of IASB BIRA 14 Examples of usage Within IASB BIRA to compile using the LIDORT option with MATLAB and the icc and ifort compilers gt make USEZLIDORT MATLAB YES to compile using only ASIMUT no MATLAB using the gcc and gfortran compilers gt make USE ASIMUT CC gcc FC gfortran Outside IASB BIRA to compile using the LIDORT option make f Makefile ToGive USEZLIDORT to compile using only ASIMUT make f Makefile ToGive USEZASIMUT Do not forget to modify the makefile to correspond to your own system You can specify the C and Fortran compilers that will be used For the moment the makefile supports gcc icc ifort and gfortran Change also the Path indicating the place where the Matlab libraries can be found See also Section 2 2 4 for more details on the linking to MATLAB Even if the code has been compiled with the LIDORT VLIDORT or _VLIDORTF90_ options it is possible to run the softwa
81. ed in a specific file using the format imposed by V LIDORT file LIDORT flags They are fully described in the ASIMUT Input File pdf file Some parameters must be given in the file INP file and those used by SPHER TMATRIX and defining the aerosols characteristics or optical properties are specified in an additional file also described in ASIMUT Input File pdf 6 5 1 Parameters defined in the file INP SPn FENn fileLIDORTflags Name of a file containing flags used to control RT SPn FENn fileLIDORTlog Name of output log file generated by LIDORT VLIDORT only in case of error warning SPn_FENn RTstreams Number of quadrature streams in the cosine half space 0 1 it must be less than or equal to the symbolic dimension MAXSTREAMS Suggested values 1 no aerosol 8 with aerosol in IR SPn FENn Nstokes Number of Stokes vector components choice between scalar and vector modes 1 for LIDORT 3 for VLIDORT SPn_FENn transOnly Flag for ignoring scattering and accounting for transmission only SPn_FENn rayleighOnly Flag for simulating only Rayleigh Scattering SPn FENn SetSolartol If SetSolartol 1 then the Solar spectrum is set to 1 0 only for testing purposes 100 7 Conclusions The ASIMUT code performs forward modelling as well as retrieval of vertical information Initially developed for nadir looking instruments such as IASI on board METOP A and for Earth observations ASIMUT has been extended t
82. ed to the value which is the extinction coefficient of layer L km and given in kext input parameters defined on scale zFact 3 2 5 1 xxxAER options In the name of these sections xxx represent the name of the aerosol ex Model soot AER stands for AERosol xxxAER Type Type of aerosol Angstrom Angstrom formalism ModAngstrom Modified Angstrom formalism Qext From given Qext Dirac distribution Conrath Conrath formalism Lidort V LIDORT formalism LidortG V LIDORT formalism xxxAER TotalOpacity Total opacity of aerosols integrated on the whole atmosphere Lidort Conrath amp Angstrom formalism If TotalOpacity lt 0 not taken into account Must be given for Conrath formalism xxxLP zfact fact factors that can be applied on the vertical profile read from the atmosphere file xxxAER kextType For Angstrom formalism Type of extinction coefficients km 1 Values Values vl are given in kext File Values are given in a file filename kext xxxAER kext For Angstrom formalism vi Extinction coefficients km corresponding to the altitudes given in zfact Filename Filename containing 2 1 z kext in dirAerosol xxxAER angstromCoeff Value of the Angstrom coefficient xxxAER angstromNuRef Value of the reference wavenumber used in the Ansgtrom formalism cm 1 If not given then it is taken as the mid value of wavenumber of the FEN inter
83. ed windows for spectrum SPn Example In case of 1 2 25 Windows SP1 FENI 5 1 FEN2 and SPl FEN25 will be selected Several parameters describe the instrument Some are relevant only for FTS SPn fov tilt focal mirrorsize iris Other are common to all instruments types SPn ILS SPn FilelLS Instrumental Line Shape and associated file containing the information if required Boxcar Triangle Trapeze Happ Genzel Blackmann Harris 3P Blackmann Harris 4P Norton Beer Weak Norton Beer Medium Norton Beer Strong Filename ils function in the spectrum space Filename for eap eah correction 3 columns opd modulation phase if only 2 columns no phase Gaussian ILS fct of ACE No ILS applied SPn Resolution Instrument or final resolution Can be either a Value or a code to indicate which File Format to read For the moment 77 the only option is SOIRPDS to read RESOL xxx TAB kind of files from the SOIR PDS Units cm l or nm specified by DataXUnit The Bruker definition for the resolution of FTS is used 0 9 MOPD SPn FileResolution Name of the file containing the information relative for the determination of the resolution For the moment only for SOIR format from the PDS Archive SPnj Resolution SOIRPDS SPn HighResolution High resolution used for the calculation of the Optical Depths Units cm 1 or nm specified by DataXUnit Normally should not be imposed ASIMUT determines th
84. elf broadened HWHM cm 1atm 1 296K for water S temperature dependence coefficient n of the self broadening halfwidth for water S estimated accuracy on the temperature dependence coefficient n of the self broadening halfwidth for water T self pressure shift of the line transition cm 1atm 1 296K for water T estimated accuracy on the self pressure shift of the line transition cm 1atm 1 296 for water U temperature dependence coefficient n of the self pressure shift for water U estimated accuracy on the temperature dependence coefficient n of the self pressure shift for water 111 And for the Cross Sections Absorption cross sections data format Wavenumber Absorption cross Molecule Atmospheric section cm mol identification code conditions code 1 10 3 3 Associated file for the atmospheric conditions code format Atmospheric Temperature Presssure Resolution Internal GEISA code for conditions code K Pa data identification 9 12 7 1 aer I3 F9 1 F12 2 F7 3 A1 112 Annex 4 Radiance irradiance Radiance energy emitted per unit area m per unit time s per unit wavelength nm or um or per incremental solid angle sr Unit of radiance W m um sr Ex blackbody function B A T 2 Incremental solid angle incremental surface area on a unit sphere dO dAg rs with dAs an incremental surface a
85. ensity values Retrieved density values a priori values if not fitted Errors on retrieved density values 0 0 values if not fitted Each element of s 1 vmr contains names z ap rt er Names of species Altitudes A priori vmr values Retrieved vmr values a priori values if not fitted Errors on vmr density values 0 0 values if not fitted ASCII File FileOUT out The ASCII file contains all data read or computed during the run and corresponding to the parameters from file INP The file is divided into BLOCKs and GROUPs For example all data concerning one Pass is saved into one BLOCK called Pass n where n is the Pass number usually 1 Different GROUPS are defined all containing a series of Parameters PARAM data axis AXIS and data DATA Matlab macros are provided in blockreader and binput to import easily the data contained in this file into Matlab or to perform searches on selected variables Plot File FileOUT ps If plots were requested a single PostScript file is created containing different plots radiance transmittance averaging kernels etc 29 2 2 7 3 Specific files in SAVE Files can be saved in ASCII or in MATLAB format files are created with a header indicating the date and time of creation and the content and units of each column Radiances and transmittances see also Annex 4 Depending on the SPn DataType flag the following data will be saved in the Fil
86. eoretical models of the effect the soft and hard collision theories lead to two slightly different line profiles Under the soft collision hypothesis 33 the added parameter introduced to describe the narrowing effect is equal to 6 P where is the narrowing parameter cm atm l This parameter should be compared to the dynamical friction Big Which is deduced from the mass diffusion coefficient D32 k T HMM 3 19 Bay 2zcmD The line shape is described by a Galatry function 33 1 1 4 0 Re 1 d 5 z 3 20 S y ix DS 2z _ with z vin2 321 D and M is the confluent hypergeometric function 3 b bb 1 2 bb D 2 3 3 22 When the hard collision theory is used the line shape is described by the Rautian profile 34 W x y 2 R 3 23 4 0 223 50 where W x y is the complex error function already defined in Eq 3 23 Both the Galatry and Rautian profiles are implemented in ASIMUT In the case of the H O and CO molecules sub and super Lorentzian line profiles have been observed 35 36 which are characterized by a symmetric or asymmetric y function respectively PV 3 24 This correction factor has been introduced to take into account the fact that far from the line centre the line displays marked deviations from a Lorentzian behaviour Typically displays a sub Lorentzian
87. erence fram SPn ObsAZ Viewing azimuth angle degree w r t the topocentric reference fram 47 3 2 Optical depths The absorption coefficient a v z is in general the sum of several contributions single absorption lines of different gaseous species absorption cross sections and continua contribution and the Rayleigh scattering term a v z Y S D V v T P p ino G T P p Gs us Vz 3 4 i j c where P and are the pressure and temperature at altitude z n and p the density and the partial pressure of species i at altitude z The first term of the right hand side of the equation represents the absorption due to single absorption lines The sum on j represents the sum on all transitions participating in the absorption of light The second term represents the broadband absorption of light due to molecules which do not present resolved absorption structures And the last term is not in se an absorption term as it corresponds to the Rayleigh scattering but it is nevertheless taken into account at this point as it is similar to a diminution of light through the atmosphere These terms will be discussed in more details in the following 3 2 1 Line by line calculation The absorption coefficient kj cm molecule for a particular line j of species i is given as P p ST x Oy yj 3 5 where is the normalized line shape f av 1 3 6 line The inten
88. es ASILID asimut AVL or AVL_F90 depending on the command line options see Table 3 The version of the software package for outside IASB BIRA only includes the possibility to link ASIMUT with LIDORT only the USE ASIMUT and USE LIDORT options are available For outside IASB BIRA the makefile provided is called Makefile ToGive USE LIDORT will compile and link the LIDORT code ASILID executable Default USE ASIMUT will compile ONLY the ASIMUT part of the code asimut executable USE VLIDORT will compile and link the VLIDORT code AVL executable not available outside IASB BIRA USE VLIDORTF90 will compile and link the VLIDORT F90 code AVL_F90 executable not available outside IASB BIRA MATLAB NO YES Allows to indicate if compilation will be done with or without MATLAB embedded NO is the Default value MATLABRUNTIME Allows to indicate if the compilation linking will be YES NO done with the MATLAB Runtime library If NO is selected then the links to MATLAB are done through the MATLAB Engine YES is the Default value OPERATIONAL YES Allows to indicate the value of the pre processor NO OPERATIONAL NO is the Default value FULLLINEMIXING Allows to indicate the value of the pre processor YES NO FULLLINEMIXING NO is the Default value Full treatment of the Line Mixing is not yet implemented under UNIX only 1 approximation CCzicc gec Sp
89. ffraction combine on the detector Special care must be taken to accurately simulate the overlapping of successive orders and take into account all instrument characteristics 3 3 6 1 Set options Set Caption String describing the retrieval without spaces If not given it will be set to the Filaname of the INP file without Path and extension Set zType Set Range Description of the type of altitude scale on which the Radiative Transfer calculations will be computed Set zType can take the different values Range the 3 values zmin zmax deltaz km are given in Set Range File Ascii file with 1 column containing the z values km Filename given in Set Range FilSFIT2 Ascii file with 1 column containing the values km but the first line indicates the number of levels Filename given in Set Range Values the values z1 z2 km are given in Set Range Zpt z scale will be the same as in the zpt file Horizontal the output will be given on the Hobs Set zScale String describing the retrieval without spaces If not given it will be set to the Filaname of the INP file without Path and extension Set Occ_zmin Value for tangent height minimum all files corresponding to a lower altitude are rejected km 75 Set Occ zmax Value fof tangent height maximum all files corresponding to a higher altitude are rejected km Set nbSpectra Number of spectra which are described in
90. format Calibration pixel wavenb file format cf ARCHIVE CALIB PIX_WN TAB with explanation of format in PIX WN LBL if PIX gt WN binning bin number Param 1 Param 2 Param 5 WN PIX binning bin number Param 1 Param 2 Param 5 ex PIX WN TAB PIX gt WN 12 1 0 000 0 0000 4 8623e 008 5 8565e 004 2 2348e 001 PIX gt WN 12 2 0 000 0 0000 6 1550e 008 5 8211e 004 2 2348e 001 WN gt PIX 12 1 0 000 0 0000 2 2366e 002 1 1703e 004 1 4983e 005 WN gt PIX 12 2 0 000 0 0000 2 8226e 002 1 4331 004 1 7931 005 SOIR csv and SOIR csv occ formats The files are in Ascii CSV format comma separated Each file starts with a header date 2006 05 20 TangHeight 87 478120 87 442389 DistVenus 1211 655717 1211 655717 Angle 122 303598 122 304126 AOTFfrequency 25709 000000 25709 000000 Lon 21 693378 21 697435 Lat 82 686342 82 686230 LST 17 193611 17 193333 PixWwn bini 000000000000000 4 396840000000000e 010 1 801220000000000e 007 6 232410000000000e 004 2 234470000000000e 001 PixWwn bin2 000000000000000 5 209850000000000e 010 3 006560000000000e 007 5 479830000000000e 004 2 234840000000000e4001 tiltAngle 0 199909 0 041842 Hslit 2 319432 2 319432 spdVEXSun 8 688690 8 688690 spdVenusSun 129398 0 129398 spdVenusOrbit 2 741840 2 740380 errorAlt 0 113087 0 113089 followed by N x 2 columns where N is the number of bins corresponding to th
91. forward modelling able to simulate the measurements in the probed air mass and taking into account the characteristics of the instrument instrumental line function field of view Initially developed for nadir looking measurements in the thermal infrared IR performed by instruments such as IASI Infrared Atmospheric Sounding Interferometer on board METOP A 1 ASIMUT has been further extended to accommodate different observation geometries and to the simulation of atmospheric spectra for atmospheres other than that of the Earth For example ground based Sun looking measurements recorded by the Fourier transform infrared spectrometer operated by the Belgian Institute for Space Aeronomy IASB BIRA at Uccle or at Ile de la R union 2 are exploited to investigate tropospheric and stratospheric composition but also for satellite validation purposes 3 IASB BIRA is also involved in the Venus Express mission of ESA 4 through the SOIR instrument 5 6 This instrument which combines an echelle grating spectrometer and an Acousto Optical Tunable Filter for the selection of the appropriate grating orders performs solar occultation measurements in the IR region 2 2 4 3 um at a spectral resolution of 0 15 cm As the techniques involved in the recording and simulation of and in the retrieval from SOIR spectra are similar to those used by instruments dedicated to Earth observation it was decided to develop the same software to interpret d
92. he end the absorption line shape is interpolated to correspond to the sampling used for the determination of the optical depth OD The latter is chosen by considering the step asked for the radiance or transmission simulation and an oversampling factor either provided by the user or chosen such that the OD wavenumber step is a factor 10 lower than the final radiance step 3 26 Rejection of lines Any spectral line whose center intensity is less than a certain absorption level is rejected This level is by default 1 0x107 but can be adjusted by the user with the SPn FENx ODLimit parameter The rejection criteria is based on the assumption that the line is a Voigt line profile and that tests can be done separately on conditions imposed by the Doppler part of the profile and of the Lorentzian part For Lorentz profiles the transmission 7 v in one specific layer is defined as Suy a v vw 7i where 5 is the line intensity from HITRAN is the integrated density in the layer vo is the position of the line center This can also be rewritten as T v 1 A v 3 28 Su In 1 A v 3 29 z v v 5u If the line is faint A 1 and In 1 A A leading to Su 3 30 vy At the center of the line the intensity is Su Aroren Vo e 3 3 1 L A Lorentz line having an intensity Azorent Vo lower than the OD limit will not be computed in the corresponding layer
93. i et al 24 As Ciddor s equation gives somewhat better results on a broader range of wavelengths and under extreme environmental conditions of temperature pressure and humidity it has been adopted by the International Association of Geodesy as the standard equation for calculating the index of refraction In fact the Edlen equation includes approximations for the water vapour term that are only accurate at 20 C and therefore could be in error at high temperature and humidity The default subroutine used to determine the index of refractivity is thus the Ciddor one at least for the Earth This code is not at all adapted for other planets atmospheres Different other possibilities have thus been included Lbirtm 19 20 same expression as used in the LBLRTM code which is expressed in terms of A red and Ae including a small correction for the water vapour Sneep 25 determination of the total refractive index of air by considering the individual contribution of CO O2 and Argon The general expression of this function makes it applicable for all planets atmospheres Magellan data for Venus the index of refraction can be determined using the measurements performed by Magellan Mars parameterization using only the altitude as variable 3 1 1 1 Planet Options These options can be selected in the INP file by imposing the Planet Refractivelndex parameter General description of the planet characte
94. iat Transfer 1999 62 1 p 29 48 Varghese P L and R K Hanson Collisonal narrowing effects on spectral line shapes measured at high resolution Applied Optics 1984 23 14 p 2376 2385 Rautian S G and Sobel man The effect of collisions on the Doppler broadening of spectral lines Sov Phys Usp 1967 9 p 701 716 Clough S A F X Kneizys and R Davies Line shape and the water vapor continuum Atmos Research 1989 23 3 4 p 229 241 Pollack J B et al Near infrared Light from Venus nightside A spectroscopic analysis Icarus 1993 103 p 1 42 103 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 Fomin Effective interpolation technique for line by line calculations of radiation absorption in gases J Quant Spectrosc Radiat Transfer 1995 53 6 p 663 669 Rothman L S et al The HITRAN 2004 Molecular Spectroscopic Database J Quant Spectrosc Radiat Transfer 2005 96 2 p 139 204 Jaquinet Husson N et al The 1997 spectroscopic GEISA databank JOSRT 1999 62 2 p 205 254 Thibault F et al Infrared collision induced absorption by near 6 4 for atmospheric applications measurements and empirical modeling Applied Optics 1997 36 3 p 563 567 Hermans C et al Absorption cross section of the collision induced bands of oxygen from the UV to the NIR in NATO Advanced Research Workshop
95. ile Name of the corresponding ZPT file 78 3 3 6 4 IMG instrument The file is in Ascii csv comma delimited It contains a header DATE 27 12 1996 12 13 53 LATITUDE 4 8 deg LONGITUDE 23 7 deg SURFACE Sea ALTITUDE 600 0 m ECMWF era 1996122702 38135 annc UNITS cm 1 W cm2 Sr cm 1 followed by 2 columns with Wavenumber cm Radiance The ECMWF line of the header indicates which zpt file to read The file must exist Option List is possible with this instrument The LST file must then contain 2 columns Name of the Opus file Name of the corresponding ZPT file 3 3 6 5 ACE instrument ci The file containing the ACE spectra are in ascii 2 columns Wavenumber cm Transmittance ACE spectra can be used with the LIST option The LST file must then contain the following information Path BaseName Index of spectrum 1 Tangent height of spectrum 1 Index of spectrum 2 Tangent height of spectrum 2 Index of spectrum n Tangent height of spectrum n The name of the file containing each of the spectra in the list will then be reconstructed using Path BaseName spectNNN dat where NNN is the index of the spectrum The same LST file can be used with the ACE occ option All spectra listed are then analysed simultaneously 3 3 6 6 IASI instrument The IASI files are Ascii csv comma delimited They contain a header MISSION IASI METOP
96. in the atmosphere provided by the user but is required for example to calculate the index of refraction by the Sneep method CO2 CO No Oo H20 are needed if not present in the user defined atmosphere values will be taken from the Default hard coded one AtmosphericModels atmFile AtmosphericModels atmFileType indicate the name and type of the file containing the atmosphere Several types are supported Col ascii columns with VMR values expressed ppm altitudes ar xpressed in km the first line must start with S and contain the label of each of the columns lines starting with are comments Example venus day ppm dat Z km NT cm 3 CO2 ppmv N2 ppmv CO ppmv O3P ppmv Z 2 N2 03 6 0000 001 4 7200e 018 9 6500e 005 4 6200 001 5 1000 001 2 1200 007 6 1000 001 3 9700e 018 9 6500 005 4 7000e 001 4 1200e 001 3 0300e 006 Sfit SFIT type of file provided by the REFMOD program e Mipas format file AtmosphericModels zptType AtmosphericModels zptFile indicate the type and name of ZPT file This file contains information on the pressure and temperature structures versus altitude Several types are supported Ascii ascii file with 3 columns z km p mb t K e Ncep NCEP file format used by SFIT2 defined on 41 levels Ecmwf ECMWF file format only the 3 first columns z and t are read e EcmwfRH ECMWF file format z
97. ine define define define define define define define define define define PIX2 PI2 PIS PIDEMI SORTPI ISORTPI SQRT2 CONST1 CONST2 CLUM HPL TO PO AVOGAD B RADCN2 LOSCH GCAIR RGAZ PLANCKC1 PLANCKC2 DEG2RAD RAD2 DEG Planet Radius RE EARTH REM1 REM4 REQ REP E2 EP2 RE MARS RE VENUS TEMPEARTH TEMPMARS TEMPVENUS TEMPSOL SUNRADIUS AU SUN EARTH SUN VENUS SUN MARS 3 1415926535897932384626433832795 PI PI 2 0 1 2 PI 9 869604401089358 PI 2 31 0062766802998 lt PI 3 1 570796326794897 1 PI 2 1 772453850905516 1 sqrt 0 5641895835477563 1 sqrt PI 1 414213562373095 1 sqrt 2 0 8 325546111576977e 001 lt 1n 2 CONST1 ISQRTPI 1n 2 pi 2 99792458e408 ligth velocity m s 6 62606957e 34 1 273 15 1 1013 25 1 6 02214129e423 1 1 3806488e 23 FEIK 1 4387752 2 686780515696967 19 1 380648792449704e 19 8 3144621 2 0e4 HPL CLUM CLUM HPL CLUM 1 0e2 KB 0 01745329251994330 1 lt 57 2957795130823200 1 lt calculation 6371 23 1 6378 39 1 6356 91 6378137 6356752 3 1 6 694384442042606e 003 lt 6 739501254387102e 003 lt 3389 9 1 6051 8 288 0 215 0 1 733 0 5780 0 1 lt 696000 0 1 149599000 0 1 0
98. is Lambertian for the reflection of the Atmospheric Emission Supersedes what is Specified in SPn 74 3 3 6 Instrumental functions The resulting radiance coming either directly from the RT part of ASIMUT or from LIDORT is further processed to take into account the instrumental line shape ILS conventional ILS of Fourier transform spectrometers Boxcar Norton Beer strong apodization are implemented in the program Specific ILS corresponding to the IMG 55 IASI 1 ACE FTS 56 instruments are also included The field of view effect as well as the distortion of the ILS by off axis incident light 57 are included It is also possible to use a user specified ILS or to correct the ILS with a modulation efficiency and a phase error 58 Spectra obtained by conventional grating spectrometer are also treated by ASIMUT Spectra can be simulated either using a wavenumber scale cm or a wavelength scale nm ASIMUT is also able to simulate spectra recorded using the SOIR instrument 5 6 on board the Venus Express mission of ESA 4 The SOIR channel that is part of the SPICAV SOIR instrument uses a new instrument concept It combines an echelle grating spectrometer with an Acousto Optical Tunable Filter for the selection of the appropriate grating orders This instrument performs solar occultation measurements in the IR region 2 2 4 3 um at a spectral resolution of 0 15 cm In such spectrometers different orders of di
99. l distribution SPHER only 97 6 3 ard a 99 6 4 Comment on the Radiance units used in ASIMUT and V LIDORT 100 6 5 OOEIDORT 1HDUC eed hse et ode te serons iy epi eis 100 6 5 1 Parameters defined in the file INP 100 Cone USTODS 101 Referentes 102 List of Annexes Annex 1 List of molecules supported and isotopic abundances 106 Annex 2 HITRAN FORMA ocn easet dieere rna UR i dicun tob 110 Annex 3 GEISA f rMat soie eritis ie D at pto Gara cue iun quan peo PE eU 111 Annex 4 Radiance 1 1 113 Ann x 5 Physical Constants i 117 Annex 6 Template of file for new 118 Annex 7 Contacts and rights 125 List of Tables Table 1 Scientific goals fulfilled by the programs included in the ALVL software 8 Table 2 List of pre processors used by ASIMUT ALVL 10 Table 3 List of calling convention under UNIX option VLIDORT and VLIDORTF90 are only accessible through the SVN reposi
100. layer D f T 3 65 where 7 is the optical depth of layer t is the transmittance of layer Equ 3 64 and T is temperature of the upper boundary of the layer wrt to the light propagation direction and 7 is the average temperature of layer a Old version Bm emission is calculated as a Planck function at the average temperature of the layer this is on fact only valid if the optical depth of the layer is small enough Bae 3 66 b Tau dependent ref Clough et al IEEE Trans Geosc and Remote sensing 44 5 1308 1323 2006 By BG BO BG F z 3 67 3 68 l t 70 For small optical depth F z Pad 2 parameters ref Clough et al JGR 97 D14 15761 15785 1992 B 7 7 7 BG ar bz BG 1 az br 3 69 with a 0 193 and b 0 013 d Pad 1 parameter ref Clough et al JGR 97 D14 15761 15785 1992 Boy BG ar BG 1 3 70 with a 0 2 3 3 3 Surface contribution The surface is characterized by a temperature Ts and an emissivity It is considered Lambertian at least when using only ASIMUT different types of surface Hapke can be selected when using Lidort When the reflection of the solar radiation on the surface is included the lower boundary condition at the surface is the following L5 SE ty TZ Zp ko 3 7
101. lines only used under UNIX calls to LIDORT VLIDORT SPHER TMATRIX _ASIMUT_ to specify that only ASIMUT has to be considered _LIDORT_ to specify if ASIMUT has to be linked with LIDORT _VLIDORT_ to specify if ASIMUT has to be linked with VLIDORT VLIDORTF90 to specify if ASIMUT has to be linked with VLIDORT F90 LIDORTF90 to specify if ASIMUT has to be linked with LIDORT F90 DEBUG when running under DEBUG configuration In that case all in out from functions are written in the DebugFile MATLAB to indicate that all calls to MATLAB functions will be done through the use of the MATLAB Engine This option is to be selected when MATLAB is installed on the computer but without the runtime library MATLABRUNTIME To indicate that the calls to MATLAB will be done through the runtime library of MATLAB and a specific ASIMUT library containing all plotting functions OPERATIONAL Operational mode no messages no files are saved no plots are possible Only the file OUT is created with almost full content The LOG file contains only final results FULLLINEMIXING Line mixing calculation using the 1 approximation does not require any specific action from the USER at compile linking time However if the user wants to use the full line mixing approach the LAPACK library need to be included into the project The Default option is 1 ORDER LINE MIXING which does not require the inclusion of any additional library Please refer to Section
102. listed ASIMUT Input File pdf Spaces and tabs are automatically removed so that Parameter value is equivalent to Parameter value Parameters names are case insensitive PARAmeter value is equivalent to parameter value Series of related parameters are regrouped under different Section whose names are indicated within brackets Section For example all parameters dealing with the characterization of the solar spectrum are located under the Solar section Within one section the order of appearance of the Parameters is not important Section PARAmeterA value PARAmeterB value PARAmeterC value 28 is equivalent to Section PARAmeterC value PARAmeterA value PARAmeterB value The symbol at the start of the line is used for comments and to de select some parameter definitions For example parameter value will not be considered Almost all parameters have DEFAULT values taken if the corresponding line is not present in the ASI or INP files See ASIMUT_Input_File pdf for the full definition of the possible parameters and their possible and default values 2 2 6 2 Description of the ASI file Run Verbose defines the level of comments in the LOG file 0 only results in LOG 1 description of important variables choices of functions 2 everything useful to locate a potential problem Run RemoveLOG defines if the LOG file is removed after normal completion of the
103. mentioned file names in the following INP files ar defined relative to these directories If not given file names in INP must contain full path information dirInput Dir where the INP files are located dirSpectra Dir where spectra are located dirInstrument Dir where instrumental functions are located dirAtmosphere Dir where Atmosphere data are located dirZzPT Dir where ZPT files are located dirPlanet Dir where planetary files are loaded Emissivity topography dirSolar Dir where solar data are located dirHitran Dir where general Hitran Geisa files are located dirLP Dir where local mod hitran geisa files are located dirApriori Dir where a priori Sa TikonovFile data are located dirAerosol Dir where the data relative to the aerosols are located dirLidort Dir containing a V LIDORT input flag file and the SPHER T MATRIX aerosol input file dirMatlab Dir where the macros for plotting with Matlab are located Normally this directory should also be added to the PATH variable in Matlab Option exists in the case the user has not the rights to change this path see Section 2 4 Troubleshooting RadiativeCode Code Type of Radiative Transfer calculation Asimut Lidort Two other options Vlidort and VlidortF90 exist but they have not been tested fully it is thus highly recommended NOT TO USE them List gives the list of the INP files that will be processed using the same options defined
104. mperature ranges Moreover they made use of available experimental data in the 190 360 K temperature range to build a simple empirical model From measurements of and collision induced absorption they calculated the O air absorption coefficient Po Py Bo T o V T The temperature wavenumber dependences were then modeled using the simple empirical law y T 3 44 B V T V exp as V 5 3 45 05 air 2 in which the reference temperature 296 was retained The values given in 40 are expressed in Amagat B v T expression has therefore to be slightly O air modified to be compatible with the units used in ASIMUT Finally the absorption coefficient for the layer at altitude z is given by PT 1 Qo air VZ fig B TC 3 46 V 2l z 3 46 with Po and the standard pressure and temperature 1013 25 hPa 273 15 Nz the Loschmidt number 2 686 777 x10 and n o 1 the O2 cumulated density in the layer z Near infrared and visible bands Two different calculations of the collision induced absorption coefficients for the near infrared and visible O bands are implemented in ASIMUT The first one is based on the MT CKD 1 2 continua implemented in the latest version of LBLRTM and the second is based on more recent measurements of these bands 41 The 1 27 um band is
105. n Emissivity gives the emissivity value for the surface constant value It will not be used if a File is given SPn FENn EmissivityFile option SPn FENn EmissivityFile name of the file containing the emissivity The file must contain 2 columns Col 1 Wavenumber cm 1 Col 2 emissivity The file must be located in the dirPlanet directory 87 4 Onion peeling method The onion peeling method was implemented to coherently treat a series of spectra recorded during one occultation In this method one starts the analysis in the uppermost layer i e with the first spectrum containing absorption structures due to the constituents of the atmosphere deriving concentrations in that layer and progressively goes deeper into the atmosphere taking into account the results from the layers above Vertical profiles of several key species of the Venusian atmosphere have been obtained by applying this technique as will be demonstrated hereafter For the sake of clarity we will consider in the following that the second right term in the general equation describing the radiative transfer through the atmosphere I v I v e B v T s a v s e ds 4 1 is negligible VEX Exo atmosphere Tangent point Laver Laver 1 Layer 2 Venus Figure 5 Geometry of solar occultation measurements and definition of the onion peeling method The observed transmittance 7r corresponding to Ray 1 passing through the uppe
106. nPeeling OnionPeeling if 1 the onion peeling method will be used Set Occ zmin Value for tangent height minimum all files corresponding to a lower altitude are rejected km Set Occ zmax Value for tangent height maximum all files corresponding to a higher altitude are rejected km 89 5 Retrieval module Two main objectives in the development of ASIMUT have been to use the Optimal Estimation Method OEM 7 coupled to the analytical calculation of the Jacobians and to allow simultaneous retrieval from a series of spectra obtained under different observation geometries with the possibility to combine spectra recorded by different instruments 5 1 General description The general forward radiative transfer equation can be written as y f xb e 5 1 where y is the measurement vector the measured radiance x is the state vector the vertical profiles to be retrieved b represents the additional parameters used by the forward model f The forward function f describes the complete physics of the measurement including the description of the instrument In the case of a moderately nonlinear problem the best estimate of the solution of Eq 5 1 is found by solving iteratively Xi 1 Xa Gily f x Ki xi x4 5 2 where x is the a priori constraint The Jacobian matrix is defined by the following expression of K 5 3 x 5 3 and the Gain matrix G is defined by Gi Sz KS K Kis
107. nch VC 2010 and then import the Asimut_vc2010_Win64 sin project Do not double click on the project file 2245 Two ways to link ASIMUT and MATLAB There are two ways to link ASIMUT and MATLAB the first one uses the Matlab Engine and the second one uses the runtime Matlab library The choice between the 2 ways is based on the version of Matlab that has been installed on your PC In all cases the libraries needed to use the Matlab Engine are present in some cases the runtime library is also installed Once you have checked which installation is present on your system you need to tell ASIMUT which version to use TIP If the Matlab Runtime library is installed on your system then you should find a sub directory like C Program FilesAMATLABNR2007bNruntime 3 MATLAB runtime Containing the 1 4 the numbers might change depending on your version of Matlab Use the MATLAB precompiler if only the Engine is installed 2 Method 1 Engine Use the MATLABRUNTIME precompiler if the Matlab Runtime library is installed Method 2 Runtime Method 1 Engine Step 1 Add the MATLAB Engine libraries Be sure to Add the 3 following libraries Jibeng lib libmat lib and libmx lib from the correct MATLAB sub directory containing the libraries for external use This should be like C Program FilesNAMATLABNR2007bVexternNib extern lib 1 Import the libraries into the project In the
108. nsity of the line in cm l molecule cm 2 296K C Air broadening pressure halfwidth HWHM cm 1atm 1 296K D Energy of the lower transition level cm 1 E Transition quantum identifications for the lower and upper state of the transition F temperature dependence coefficient n of the air broadening halfwidth G identification code for isotope as in GEISA I identification code for molecule as in GEISA J Internal GEISA code for the data identification K Molecule number as in HITRAN L isotope number 12most abundant 2 second etc as in HITRAN M transition probability in debye2 N self broadening pressure halfwidth HWHM cm 1atm 1 296 for water air pressure shift of the line transition cm latm 1 296K P accuracy indices for frequency intensity and halfwidth Q indices for lookup of references for frequency intensity and halfwidth R temperature dependence coefficient n of the air pressure shift estimated accuracy cm 1 on the line position estimated accuracy on the intensity of the line in cm 1 molecule cm 2 estimated accuracy on the air collision halfwidth HWHM cm 1atm 1 F estimated accuracy on the temperature dependence coefficient n of the air broadening halfwidth estimated accuracy on the air pressure shift of the line transition cm latm 1 296K estimated accuracy on the temperature dependence coefficient n of the air pressure shift N estimated accuracy on the s
109. nteger pointer at the end for the references which we will manage in the HAWKS software as well as providing an acrobat file Documentation ReferenceTable pdf for those not using HAWKS Following is the format of the headers in the current edition 2000 Cross section Header format Chemical symbol Wavenumber Number Temp Press Maximum Res Common Name Not Bro Ref Minimum Maximum Of pts X section 10 20 30 40 50 60 70 80 90 100 Note Chemical Symbol is right adjusted Res is resolution in cm for FTS measurements and in milli Angstroms for grating measurements in the UV xxxm 110 Annex 3 GEISA format Format description for the GEISA 2003 database KKK K K K K K K K K K KK K K K K K K K K KK K K K KK K K K K K K K K K K KK K KK K K K K K K K Parameter A B C D 2 F G I J K L M N 0 1 6 10 9 9 9 914 333121 10 5 8 136 F12 6 1PD11 4 OPF6 4 10 4 9 A9 A9 A9 F4 2 I3 I3 I2 1PE10 3 OPFS 4 F8 6 16 A B lw o RINI s s 6 10 n 6 6 15 4 8 8 6 4 10 6 1PD11 4 OPF6 4 4 2 F8 6 F6 4 5 4 F4 2 4 2 F8 6 F8 6 F4 2 F4 2 F12 6 D11 4 F6 4 F10 4 A36 F4 2 13 I3 A3 D I1 E10 3 F5 4 F8 6 13 I6 F6 4 F10 6 D11 4 F6 4 F4 2 F8 6 F6 4 F5 4 F4 2 F4 2 F8 6 F8 6 F4 2 F4 2 A B C D KL BCFORN 5 ST TU U A wavenumber cm 1 of the line B inte
110. o SPHER TMATRIX and V LIDORT codes to include the complete treatment of the scattering effects into the radiative transfer calculations Table of Contents ls dnitoduclon iui MI d 6 2 General Description of the 7 2 1 mpl mentati oiin n aceite eden 7 22 Compiling and running ASIMUT ALVL eese 10 2 2 1 Pre processor definitions soe ed bunte pia saca te o ta usas aid 10 2 2 2 Compiling under WINDOW loe ca 11 ZG Create the project ote nn MA EUM I ODE S nese ences 11 2443 Compiling nde r UNIX 14 22 4 Linking ASIMUT ALVL and MATLAB 16 2 2 4 1 Under Windows 32 bits ioco o votum petiti besito dini onde de e 16 242 Under Windows G4 DIST oap noe 16 2245 Two ways to link ASIMUT and MATLAB 17 224A Under UNIX Et OREL ceed cate 23 2 43 Using the CLAPACGCK 25 2 2 6 Running ASIMEET ALNY Ici sesion se ttem 26 2 2 6 1 Description of the input files iro rtt ceti accen aser 28 220 2 Description of the ASI Bleu irri 29 2 2 7 Description or the output TIIeR eos esti Foto
111. o different geometries off axis nadir looking mode solar occultations from space and from the ground The latter option will be used for validation purposes using ground based Fourier transform infrared FTIR instruments involved in the Network for the Detection of Atmospheric Composition Change NDACC http www ndacc org The internal structure of ASIMUT has been made such that it is easily adapted to simulate planetary atmospheres other than Earth Mars and Venus The forward model includes the possibility to simulate LBL molecular absorption but also the Rayleigh scattering the absorption in the form of cross sections and continua of absorption due to H20 and The latter are essentially treated with the MT CKD 1 2 formalism 20 with slight modifications to take into account new laboratory measurements LBL simulation uses different line profile functions Lorentz Voigt Galatry and Rautian It is moreover possible to introduce a symmetric or asymmetric factor defining sub or super Lorentzian profiles Spectroscopic parameters and absorption cross sections can be directly read from HITRAN CDSD and GEISA databases or from user defined files The retrieval unit of ASIMUT uses the Optimal Estimation Method and is based on the analytical calculation of the Jacobians during the forward model run ASIMUT is continuously being updated and improved In the near future the treatment of solar lines present in direct sun measu
112. of the uncertainty in photon energy is small Collision broadening is caused by secondary interaction with other gas molecules and atoms Absorption lines under collisional regime are described by a Lorentzian function Doppler broadening describes the statistical distribution of frequency shifts due to thermal motion This effect is characterized by a Gaussian function and is most significant under low pressure conditions when the collision broadening effect is small The Doppler profile of a line centred on voj wavenumber joa E e a 4 G y yp D is characterized by its half width which depends on the molecular mass M of the absorbing species as well as on the temperature 21n2k T Vs MEC 3 10 The half width of the Lorentzian profile jr a 3 11 v v ty is composed of a self broadening contribution and of a foreign broadening contribution which are pressure and temperature dependent P 3 12 and Lan 7 T N foreign yr 3 13 where p is the partial pressure of species i and are the reference pressure and temperature and and are the temperature coefficients of respectively the self and the foreign broadenings The values of cm atm and Nye Nforeign unit less are tabulated values in the databases Pressure induced shifts are also taken into account in the determinati
113. ogram SFIT 21 However some modifications were included in the ASIMUT implementation essentially dealing with the determination of the index of refraction 43 3 1 1 Index of refraction Ray tracing performed in FSCATM uses the equation of Edlen 22 including a correction for the presence of water vapour As discussed in Ciddor 23 and in Tomasi et al 24 new measurements of dispersion and of absolute refractive index have since been made Ciddor 23 developed a new set of equations based on the most recent equations for air density and air dispersion in the visible and near infrared regions The moist air refractivity n v P T e C at nonstandard temperature and pressure P conditions and taking into account the presence of water vapour with partial pressure e and of with volume concentration C is calculated as the sum of a first term giving the refractivity of dry air e 0 containing CO multiplied by its density normalized to that of standard dry air conditions 7 2288 15K 1013 25 hPa and a second term giving the refractivity of pure water vapour n o v T e multiplied by its density normalized to that defined at standard conditions T 293 15 and 1333 Pa 0 pO T 0 C ple T e 0 P T e C 1l es th NSD a p e T e 0 n v P T 0 C 1 3 3 The detailed description of how to determine all the parameters appearing in Eq 3 3 can be found in Ciddor 23 and Tomas
114. olecule Linesixing NbLines Report to the LOG file 3 Specify to the compiled ASIMUT EXE where to find the dll related to the libraries The easiest method is to change the Environment Settings including into the definition of PATH where to find the different dll You need to add the following sub directory C Program FilesMATLAB R2007b bin win64 MATLAB bin Path Path MATLAB bin CE gt Control Pane p System end Secunty p System v p If Search Content Pan View basic information about your computer Windows Windows 7 Professional Copyright 2009 Microsoft Corporation rights reserved Computer Name Hardware Advanced System Protection Remote You must be logged on as an Admnustratorto make most df these changes Pedomence Maus eflects processor scheduling memory usage and vitual memory Stm User varables for aen Variable Value C Program Fies intel WFibin VC Prog USERPROFILE SS MopData Local Temp SUSERPROFILE VopData Local Temp Sustup end Recovery ac MIO Locis Lotta System startup system fahre and debugging information System variables Variable Value 3 BURN AUTOPLAY C Program Files x86 Roxio Roxio Burn Ci windows gystem 32 DEFLOGDIR ProgramOata McAfee IDesktopProtec LEMCAUTOPLAY C Program x86 Common Fies R _ J
115. ols that can be included in one run 10 SPn FENx FitAerosols For each Aerosol indicates type of fit Value is lt Fit only the column Value is No fit only simulation Value is gt 0 0 Fit as described in xxxAER description The decimal part of the number is the covariance for the fit if no Sa file is given 0 0 0 0 SPn FENx aPrioriAerosols For each Aerosol indicates the type of a priori that will be used in the Rodgers formalism Model From the atmosphere file 86 Previous From previous fit ATMname From previous fit of ATMname molecule SPn FENx Continua indicates which continua will be included Maximum number of Aerosols that can be included in one run 5 SPn FENx Rayleigh indicates if Rayleigh diffusion will be included A series of parameters describe the background of the spectrum SPn FENx BaselineCst indicates the Baseline constant term If set 1 1 the value is automatically determined from the measurement file SPn FENx BackgroundWavenumbers BackgroundValues defines the background size of the two vectors must be the same A series of parameters describe the radiative properties to be used in the window SPn FENn Ts temperature of the source K If a number is given it represents the temperature If the option zpt is chosen Ts will b ither given in the ZPT file if not T of the lowest altitude in the ZPT file will be considered SPn FEN
116. on of the Sun radiation on the planet s surface Cf for ex Haus and Titov Haus 2000 2110 Lower boundary condition equ 15 without diffusion 115 a Gacet d E Hy TCZ Zp Hy EB T with E solar irradiance at the top of the atmosphere of the planet T r R p Isun Rsun Lh cos sza distance Planet Sun a albedo of the planet if not specified a 1 emissivity of the planet Rsun Isun iWav Fenetre gt AngleSolar Where Fenetre gt AngleSolar 7 t Rps 4 Definition of the Transmittance Radiance Transmittance 0 In the normal case Ip radiance from the source radiance from the sun if looking at the sun solar occultation of the planet if looking at the planet in Nadir mode when no solar reflection is taken into account If solar reflection on the surface is included Radiance Transmittance T 19 7 RadianceFactor 116 Annex 5 Physical constants The physical constants used throughout ASIMUT are defined in the file asimut h They have been updated according to the updates given on the NIST website In June 2012 the constants used in ASIMUT were the following define define define define define define define define define define define define define define define define define define define define define define define define define define define define define define define def
117. on of the line absorption The central wavenumber is displaced by 6 n self T T reign D T 3 14 if self and foreign Orig shifts are considered n pand m are the temperature coefficients of respectively the self and the foreign shifts The shifts 6 can be found in the databases as well but the temperature dependences are not available yet and has been taken as being identical to the values affecting the broadening coefficients Nsei and Nforeign When both broadening effects must be considered the absorption line is best reproduced by a Voigt profile which is the convolution of the Doppler and Lorentzian profiles 49 In2 1 v y 3 15 Yp where y K x y 2 dt Re W x 3 16 3E i in which I2 uta d 3 17 Yd W x y is complex error function W x ne ja 3 18 1 Different methods to calculate Voigt profile are implemented Pade approximation Humlicek 31 or Fadevaa 32 Intramolecular collisions have as already explained a broadening effect on the line shape But they also perturb the translation motion of molecules which implies a reduction of the broadening related to the Doppler effect This effect is often referred to as the Dicke narrowing of the line shape To model this effect a further parameter has to be included and the line shape is modified Two th
118. onal matrices diagonal matrices with off diagonal terms added as Gaussian 5 D JS 6 08 G Dexp 5 5 18 or exponential functions PC T ET lz z S j 4S G S CJ j exp a 5 19 to account for correlations between the concentrations at different altitudes z is the altitude of level is the vertical correlation length 5 3 1 Covariance options in xxxLP xxxXS and xxxAER All the following parameters are valid for molecules LP cross sections XS and aerosols AER and indicate how the fit will be done and which covariance will be selected xxxMM zfit altitudes on which the fit will be performed Several options are possible Zfact same altitudes as those given in xxxMM zfact zRay same altitudes those used for the Radiative Transfer calculations Set zScale Z1 2255 altitudes xxxMM SaType description of the Sa values None No given decimal part of FitMolecules FitCrosssections or FitAerosols for all diagonal Sa values Values Values given xxxMM Sa gauss Values given in xxxMM Sa Gaussian correlation Expon Values given in xxxMM exponential correlation File Sa values in file 1st col altitudes DiagFromFile Take only diagonal values of Sa in file GaussFromFile Diagonal values of in file Gaussian correlation ExponFromFile Diagonal values of Sa in file exponential correlation xxxMM Sa Sa definition c
119. ontent depends on xxxMM SaType Can be a FileName or values xxxMM SaH Mixing height km used with the option Gauss and Expon xxxMM SaZ Altitudes km at which Sa diagonal values are given in Sas xxxMM UseTikonov indicates if the Tikonov regularization method is used xxxMM TikonovFile Filename containing the Tikonov matrix which should be a square matrix defined on the same altitude grid as the one given by xxxMM zfit 93 5 4 Summary of all Variables that can be fitted Several variables can be fitted obviously the densities of molecules cross sections and aerosols columns and or profiles the background signal the surface temperature This section gives an overview of all fit possibilities and the calling convention options in the INP files 5 4 1 Molecules cross sections and aerosols densities Columns and profiles can be fitted by specifying the following parameters SPn FENx FitMolecules SPn FENx FitCrossSections SPn FENx FitAerosols For each molecule cross section or aerosol indicates type of fit Value is 0 0 Fit only the column Value is 0 0 No fit only simulation Value is gt 0 0 Fit as described in xxxLP xxxXS or XXXAER description The decimal part of the number is the covariance for the fit if no Sa file is given SPn FENx aPrioriMolecules SPn FENx aPrioriCrossSections SPn FENx aPrioriAerosols For each molecule cross section or aerosol
120. orte ena ende 64 324b saos rec ui eid DE UL 65 3 2 4 1 Continua DOpHODSSuo cedro qus lord a dus 65 SIS MEE TC e 66 3 2 5 1 so 68 3 3 Radiative transter model o e ipei ts eui ep EN UE E NIE 70 3 3 1 Determination of the ooo eaaet uae Edom Ie E 70 3 3 2 Atmosphere emissiOTl isses t ei es ue dan dar pae e EIS e 70 2 9 9 Surface CODLEIDUDOEE soe ete ats i do HUN 71 3 3 4 Determination of the 22 3 3 5 Brightness termiperabirte NA oe 72 3 3 5 1 cubo ditus oc ee 72 3 55 2 Radiative options in SPN sc p eR ROS EDO Ip ORNARE a 73 3 3 5 3 Radiative options in SPn_FENN user tetas 73 2 505 Instrumental EURCUOTIS o0 dit aduer tacit o Ru E 75 3 3 6 1 ptos 75 3 3 6 2 ODUDHS 76 3903 oe en aa ee 78 21049 4 IMG unie I ES oed nee ee eine 79 2009 ACE Cenn nep a Pf 79 3 9 0 0 TAST 79 A MO LO SOLR ANSE rUMENE oes tisque gues npe 81 SPA FENZ QODHOIS oras teda caida 85 4 peeling method rriro toto ite condi sed tec dae 88 4 1 Onion peeling OPUL
121. parameterized SHDOM extract from the SHDOM software package similar to LIDORT but developed for Mars It only contains the contribution from CO2 and O2 which are the main constituents of the Martian atmosphere SNEEP based on the formulas of Sneep and Ubachs 53 giving Rayleigh and King contributions for different species CO2 O2 N2 Two other very basic parameterizations VENUS and MARS 3 24 Continua options Continua Rayleigh Formalism for Rayleigh scattering mtckd MT CKD 1 2 Hansen Hansen and Travis 1974 from pressure Shdom shdom valid only for Mars Venus Sneep Sneep and Ubachs JOSRT 92 293 310 2005 Ciddor Ciddor AO 35 1566 1573 1996 Continua RayleighDepolarisation Rayleigh depolarisation ratio Default value for Earth Young AO 1980 Default value for Mars amp Venus 0 0774 SHDOM model If 20 0 no scattering effect but still absorption 65 3 2 5 Aerosols Aerosols are included in the ASIMUT code either as extinction ASIMUT or full absorbing and scattering species ALVL through the call to V LIDORT Different types of parameterization are included The aerosols calculated by V LIDORT are defined just as another type of aerosols In that case ASIMUT is just reading the input parameters and letting V LIDORT do the calculations or call the SPHER T MATRIX However as they are aerosols they have access to the whole range of routines already defined in ASIMUT de
122. pass from W m nm 5 to W em cm 1 sr used in ASIMUT B d v B dv pan WB W 2 W 10 m TE uua w rden tm _ B qs W i vi em How to pass from W cm nm lsr to W cm cm sr used in ASIMUT 114 B d v B dv B W B W B W 107m B cy m E 2 21 55 E Wl 2h V cm am em V 10 V cm A0 m B 10 Ed 4 v In ASIMUT Reading the input file Input file Solar FileSolar Filename This file must contain at least 2 columns V irradiance at the top of Earth atmosphere W cm cm irradiance not radiance at top of the Earth atmosphere Irradiance v r R B v Ts with Ts Temperature of Sun photosphere r radius R distance between Earth and Sun B Blackbody Radiance W cm cm ysr Example of files Solar irradiance ACE dat or Solar irradiance SOLSPEC dat built using ACE irradiance m 2 Building the Isun vector int Sun double Io double Wave long NbWave double resol FENETRE Fenetre double Tsun Reads the irradiance from the file specified in the input file and divides it by EARTHSOLANGLE or Planet which is r Rg sY with distance Planet Sun Isun rradiance from File EARTHSOLANGLE rradiance from File r Res ad B v Ts 3 In the RT part When taking into account the reflecti
123. pheric files usually used For example one can used the O3isol name it will be recognized by ASIMUT but not in the US Standard database which only recognizes O3 xxxLP ATMname is the name to be found in the atmospheric file xxxLP File name of the file containing the spectroscopic parameters xxxLP Type type of the file containing the spectroscopic parameters hitran geisa cfgl for the main databases See Annex 2 and Annex 3 for the description of the HITRAN and GEISA formats respectively It can also be ascii or grat The latter is the internal format of ASIMUT binary code If ascii is selected the following columns must be given s Table 4 xxxLP isotope definition of the isotopologue of the molecule If 9997 all isotopologues will be considered It 999 xxx take all isototopologues except xxx Be sure that the spectroscopic parameters are then corrected by the isotopic ratios this is the case for hitran geisa cfgl For grat files it is also the case if the files were created with the 999 option To specify a given isotopologue use the HITRAN denomination see Error Reference source not found Examples 1 all isotopologues of O3 together O3LP Isotope 999 2 two isotopologues of O3 1 and 3d ones O3i1LP Isotope 666 27 03i3LP Isotope 686 3 all isotopologues of O3 except the main one 03
124. provided all information on the instrument and the Observation are read from the header This file format is identical to the one used in the SOIR PSA ESA archive Venus 149 asi 341 149 1 singleCSVspectrum inp A SOIR CSV type of file is provided all information on the instrument and the observation are read from the header The MATLAB macro ConvertPDS ASCII m used to convert PDS file to CSV and to create some additional files such as the LST file giving the list of all files converted is also provided in the ASIMUT MATLAB subdirectory Venus 149 asi 341 149 1 severalCSVspectra inp Individual analysis of the spectra listed in the 20070820 101 149 Sub Lst list This list has been created from the full list of spectra 20070820 101 149 Lst by selecting one spectrum every 5 km spectra are analysed sequentially and independently Venus 149 asi 341 149 1 severalCSVspectraTogether inp Same as previous but all spectra are analysed simultaneously only ONE fit of the CO2 profile considering the 6 spectra all together Venus 149 OnionPeeling asi 341 149 1 severalCSVspectra onionPeeling inp Same as previous but all spectra are analysed using the Onion Peeling method Only ONE CO2 profile is reconstructed from the analysis of the 6 spectra 36 2 29 On line help A web based online help is also provided It contains general information on the ASIMUT ALVL project but also a full description of all the h files all st
125. ption of the input files ASIMUT ALVL uses a series of input files file ASI and file INP when calling only ASIMUT When using V LIDORT additional file is mandatory file LIDORT flags If calls are made to SPHER TMATRIX then a further file describing the aerosols has to be provided file aerosol INP The extensions of these files are not imposed and can be changed However in the following we will consider ASI and INP the definition of the two types of files required by ASIMUT The ASIMUT Input File pdf file describes in details the formats and contents of the two input files used or needed by ASIMUT ALVL The first file asi or asimut by convention contains a series of parameters definitions for the general run conditions For ASIMUT a second file inp by convention contains the details of the simulations or retrievals to be performed When calling ASIMUT ALVL a specific file containing LIDORT parameters is also needed And when calling SPHER TMATRIX a file is needed to describe the aerosol included In this section we will briefly describe the ASI file The content of the INP file will be described in the next sections in parallel to the description of the physics and mathematics of the algorithms files are constructed on the same canvas Values of parameters are specified by lines such as Parameter value The parameter name must be a recognized name and the possibilities are described in the following and
126. r Two dedicated files explain in details how to install ASIMUT on a PC running under Windows ASIMUT Screenshots Install pdf and ASIMUT FileExtension pdf Special care has to be taken to ensure the link between ASIMUT and Matlab to be effective see 2 2 4 2 2 2 Create the project Create a new sub directory You can either use an existing sub directory such as the VisualC or create a new one ex ASIMUT ASIMUT Local Open the Visual environment VC2008 or VC2010 Be sure to be Administrator of your PC or to start VisualC with the option Run as Administrator for VC2010 no such problem encountered with VC2008 Create a new project based on existing files Select File New Project from existing files You are then proposed a series of screens in which step by step you will define the new project Screen 1 What type of project Visual C Screen 2 Project File location Browse to select the NASIMUTVASIMUTNLocal 11 Project Name Asimut Click on the Add Button and select the ASIMUT ASIMUT directory deselect the ASIMUT ASIMUT Local directory check that the option Add files to the project from these folders is selected In the field File types to add to the project leave only c h Screen 3 Project Console Application project Screen4 Finish Modify the properties of the new project Right click on Asimut
127. r 4 3 um for atmospheric applications measurements and empirical modeling Applied Optics 1996 35 30 p 5911 5917 Menoux V et al Collision induced absorption in the fundamental band of temperature dependence of the absorption for and N5 O pairs Applied Optics 1993 32 3 p 263 268 Mlawer E J et al Collision induced effects and the water vapor continuum in Proc of the 8th Atmospheric Radiation Measurement Science Team Meeting 1998 Tucson Arizona USA Magalhaes J A J T Schofield and A Seiff Results of the Mars Pathfinder atmospheric structure investigation J Geophys Res 1999 104 E4 p 8943 8955 Tashkun S et al CDSD 1000 the high temperature carbon dioxide spectroscopic databank J Quant Spectrosc Radiat Transfer 2003 82 p 165 196 104 55 56 57 58 59 60 Kobayashi H et al Development and evaluation of the interferometric monitor for greenhouse gases a high throughput Fourier transform infrared radiometer for nadir Earth observation Applied Optics 1999 38 p 6801 6807 Bernath P F et al Atmospheric Chemistry Experiment ACE Mission overview Geophys Res Lett 2005 32 p doi 10 1029 2005GL022386 Kauppinen J and P Saarinen Line shape distortions in misaligned cube corner interferometers Applied Optics 1992 31 1 p 69 74 Hase F T Blumenstock and C Paton Walsh Analysis of the instrumental line shape of high resolu
128. re there is no need to calculate the aerosol parameters at the same fine wavenumber resolution as the gas parameters There is an option in ALVL allowing the user to calculate the aerosol parameters at a coarse resolution and then to interpolate them to a finer resolution specified for absorption and RT calculations 98 6 3 Surface LIDORT VLIDORT simulate Lambertian and BRDF Bidirectional Reflectance Distribution Function surfaces In case of a Lambertian surface the user has to input the surface albedo In case of a BRDF surface the user is asked to choose up to 3 different BRDF kernels from those incorporated in the codes The BRDF function will be then calculated as a linear combination of chosen semi empirical kernel functions The list of BRDF kernels implemented in LIDORT and VLIDORT is shown in Table 6 The index is an internal code parameter which nevertheless has to be specified by the user Unfortunately the list of references for these kernels is not complete yet as some kernels are stated implemented inside the codes but the corresponding user s guides do not have any record of them The list will be updated after further research Name Index Index Reference LIDORT VLIDORT Lambertian 1 1 Ross thin 2 2 Wanner et al 1995 Ross thick 3 3 Wanner et al 1995 Li sparse 4 4 Wanner et al 1995 Li dense 5 5 Wanner et al 1995 Hapke 6 6 Hapke 1993 Roujean 7 7 Wanner
129. re with all V LIDORT functionalities disabled pure ASIMUT code by forcing Code ASIMUT in the input ASI file Please beware the contrary is not true compiling with the _ASIMUT_ option will disable all V LIDORT calls Likewise compiling with the _LIDORT_ option will deactivate the VLIDORT and VLIDORTF90 calls and similarly with VLIDORT and _VLIDORTF90_ the calls to the other options except only ASIMUT will be disabled All tests have been performed on the LIDORT version IT IS HIGHLY RECOMMENDED TO USE ONLY THE USE LIDORT OPTION or ASIMUT if no scattering is needed Only the LIDORT code is provided outside IASB BIRA VLIDORT and VLIDORTF90 are only accessible through the IASB BIRA svn repository 15 224 Linking ASIMUT ALVL and MATLAB plots are done through a MATLAB interface Therefore the link between ASIMUT ALVL and MATLAB should be ensured The link has been successfully tested under Windows 32 and 64 and UNIX but some limitations have been found in relation to the compatibility existing between Visual and MATLAB versions 2 2 4 1 Under Windows 32 bits No problems encountered neither with VC 2008 nor with VC 2010 Be sure that the WIN32 precompiler is set and the NOMATLAB one is not present except if you do not want to use Matlab or if Matlab is not installed on your PC 2 2 4 0 Under Windows 64 bits There seems to be an incompatibility between the new DLLs furnished by MATLAB and VC 2008 Pleas
130. rea radius of the true sphere r sinOd pN s Fs gt y 3 I spectral radiance at a given point in space regardless of the source of the radiance Unit W m um sr spectral actinic flux integral of the spectral radiance over all solid angles of a sphere dE zL dO if I isotropic E 4x D Unit W m um 113 F spectral irradiance or net flux or energy flux vertical component of the radiant energy propagating from all directions across a flat surface per unit area time and wavelength Integral of the spectral radiance over the hemisphere above the x y plane For szaz0 z0 irradiance impinging the x y plane is maximal Sza 90 irradiance is 0 0 dF cos 8 dQ if I isotropic isotropic spectral irradiance x isotropic radiance Unit W m um Effect of the distance from the Sun radiance at top of atmosphere radiance emitted at the Sun s photosphere B A with T the temperature of the Sun s photosphere 5796 K Fpinetj incident solar radiation at the top of the planet s atmosphere Depends on solar radiance The planet Sun distance Radiance decreases proportionally to the inverse square distance from source FPianet GRY Fon With r radius of the Sun at the photosphere 6 96x 10 m R Planet Sun distance Units of radiance irradiance NN EC prs eles How to
131. rectories where the so can be found The call to the script is the following Run alvl sh EXENAME ASI file Where EXENAME is the name of the executable to use asimut ASILID AVL and ASI file is the first input file needed by ASIMUT ALVL The script automatically points to the right library path Run alvl sh bin bash export MATLABLIBPATH opt matlab bin glnxa64 export MATLABRUNPATH opt matlab runtime glnxa64 export JAVALIBPATH opt matlab sys java jre glnxa64 jre lib amd64 server export ASIMUTRUNPATH dirname 0 bin Runtime Unix qf OM woh o help H then echo 0 run ALVL binary gt amp 2 echo Usage 0 NAME OF EXE ASI file gt amp 2 echo NAME OF EXE ASILID or ASILIDF90 or asimut 262 exit 0 fi 26 export LIBRARY PATH SMATLABLIBPATH SMATLABRUNPATH SASIMUTRUNPATH SJAVALIB PATH LD LIBRARY PATH exec dirname 0 1 2 Several output files are created depending on what the user has asked One file is ALW AYS created 15 a LOG file which contains information on the run of ASIMUT It contains also the major results and if the program stopped why and where in the code it failed Reading problem encountered this file in case of problem gives in general the reason of the See also Section Troubleshooting for more information on Error Termination with ASIMUT 27 2 2 6 1 Descri
132. rements and the effect of direct sun reflection on the surface for example will be implemented Based on the vertical optical depths obtained with ASIMUT this algorithm uses an OEM to retrieve both the aerosol optical thickness and the surface temperature assuming a thin horizontal layer of a mixture of two aerosol types located at an arbitrary altitude in the troposphere 101 10 11 12 13 14 15 16 References Cayla and P Javelle JASI instrument overview Int Soc Opt Eng Proc SPIE 1995 2583 p 271 281 Senten C et al Technical Note New ground based FTIR measurements at Ile de La R union observations error analysis and comparisons with independent data Atmospheric Chemistry and Physics 2008 8 p 3483 3508 De Mazi re M et al Validation of IASI Atmospheric Chemistry Products for CO HNO3 N20 and with FTIR Ground based network data in Ist EPS MetOp RAO Workshop 2006 ESRIN Fracsti Italy ESA Publications SP 618 Svedhem H et al Venus Express The first European mission to Venus Planet Space Sci 2007 55 12 p 1636 1652 Mahieux A et al In flight performance and calibration of SPICA V SOIR on board Venus Express Applied Optics 2008 47 13 p 2252 65 Nevejans D et al Compact high resolution space borne echelle grating spectrometer with AOTF based on order sorting for the infrared domain from 2 2 to 4 3 micrometer Applied Optics 2006 4
133. retrieved is a structure defining the vertical mixing ratios retrieved non retrieved Each element of s i fen contains Ts Tsap Tserr B Bap Berr nu obs clc dif Brightness temperature result of fit K Brightness temperature a priori value K Error on b Brightness temperature K 4 parameters defining the background result of fit background B 1 B 2 nu B 4 B 3 nu B 4 2 idem a priori values idem error on fitted values except B 4 Wavenumbers Observed spectrum Simulated spectrum Observed Simulated spectrum Each element of s i geo contains Hobs Hend Hmin type sza beta phi Re lon lat Observer altitude km End altitude of light path km Tangent height km 0 HORIZ 1 NADIR 2 OFFAXIS 3 LIMBANG 4 LIMBTG 5 TOSPACE 6 GENERAL Solar zenith angle Angle at HObs Angle at the Planet Angle at Hend Planet radius Longitude Latitude Each element of s i species contains name 7 value err den vmr Name of the species Altitudes used for fitting the species Fitted values around 1 Error on fitted values Fitted densities Fitted vmr Each element of s i fit contains n Number of fitted parameters 32 m A m Sn dof rms Number of spectral points Averaging kernels Smoothing error Experimental error Degree of freedom Rms if fit Each element of s 1 den contains names 7 rt er Names of species Altitudes A priori d
134. ribution function normalized to unity as follows 96 s n r dr 6 1 Here n r dr is the fraction of particles with equivalent sphere radii between and r dr and rl and 72 are the minimum and the maximum equivalent sphere radii in the size distribution Mishchenko amp Travis 1998 6 2 1 Modified gamma distribution both codes a 1 Oe where the gamma function aand yare constants and rc is the radius value approximately corresponding to the peak of the distribution 6 2 2 Log normal distribution both codes In r In rg mae exp 6 3 2 In og where r and o are the mean and standard deviations of r 6 2 3 Power law distribution both codes 2 1 I Dr lt r lt r3 r2 6 4 0 otherwise For this distribution the codes automatically calculate the minimum and the maximum radii r and on the basis of supplied effective radius re and effective variance veff which are analytically defined as n r rnr dr 6 5 1 Ver suce n r r r g nr dr 6 6 where n r nr dr 6 7 6 2 4 Gamma distribution both codes 1 3b jn where 5 the gamma function Pe ind b are constants 6 2 5 Modified power law distribution both codes 2 1 2 lt v 0 lt lt y 5 forr lt r lt r 6 9 r 12 amp n rx iE 0 gt 6 2 6 Bimodal volume log normal distribution SPHER only This distribution will
135. ristics are to be supplied under Planet Planet PlanetRadius km DistToSun AU Topography Defaults values are taken into account depending on the choice of the Planet Planet parameter Earth Mars or Venus Other planets or bodies can be simulated but then respectiv values of the parameters must be supplied 44 nu o T E 1 3 1 2 Atmospheres The user may either choose one of the built in atmospheres US Standard Atmospheres 26 MIPAS Model Atmospheres 27 or propose his own volume mixing ratio vmr vertical profiles as well as the vertical profiles for temperature and pressure 3 1 2 1 AtmosphericModels options AtmosphericModels Model The user can specify one of the hard coded atmospheres for the Earth or a user defined file Model 0 There are also hard coded models for the atmospheres of Venus VIRA day model and Mars typical atmosphere South Winter but those do not cover a wide series of species for ex only CO and in the case of Venus For these planets the user is advised to give a user defined atmosphere For Earth 6 different atmospheres have been hard coded They correspond to the Tropical Mid latitude summer Mid latitude winter Subarctic summer Subarctic winter of the US standard Model as well as the US Standard 1976 Model Model 1 to 6 respectively The AtmosphericModels Default correspond to the hard coded atmosphere that is used when a species is not found
136. rmost layer Layer 1 in Figure 5 is then given by Tn v exp a v As 4 2 where v stands for a T P v and As is the length of the ray path in Layer 1 obtained by the raytracing procedure In this expression the only unknowns are the concentration N of each species and the aerosol loading in Layer 1 which are 88 retrieved from the analysis of this first layer Transmittance observed for Ray 2 will result from the combination of the absorption of light in Layer As and Layer 2 As If we moreover consider the atmosphere as spherical and homogeneous we can further write Tr v exp VAs 457 4 3 in which the only unknowns are the concentration of each species and the aerosol loading in Layer 2 By going down progressively the vertical profiles of the interacting species can be derived The onion peeling method 59 has been implemented in ASIMUT for solar occultation measurements This method derives vertical information on the retrieved species starting at the most distant top layer and going down toward the lowest taking into account the information obtained the way down Spectra recorded by the SOIR instrument 5 6 on board the Venus Express mission of ESA 4 have successfully been investigated with ASIMUT 12 15 The ACE FTS instrument 56 also measures atmospheric absorptions by solar occultation and the same method can be used in this case 4 1 Onion peeling options Onio
137. rties gt VC Directories Select Library Directories gt lt Edit gt Add the line Asimu bin Lib Win64 ASIMUT_Lib 2 Specify to the compiler the FULLLINEMIXING option Be sure to add the FULLLINEMIXING pre processor option 25 2 26 Running ASIMUT ALVL Under Windows if you are not using the Visual C environment WINDOWS may require some additional software to be installed to be able to run executable programs You will then need to download and install the Microsoft NET Framework package The latest version can be downloaded from microsoft website http www microsoft com net The calling convention is gt Asimut X fullpath FileASLasimut where Asimut is the executable The call being done in the directory where asimut exe is located if not you need to specify the path where this file has been saved Do not forget to specify where to find the dll used by the project if you want to use the functionalities of MATLAB see 2 2 4 for more details Under UNIX and if using the makefile provided with the project this executable is named asimut ASILID or AVL when using the ASIMUT LIDORT and VLIDORT options respectively gt asimut X fullpath FileASLasimut gt ASILID X fullpath FileASLasimut gt AVL X fullpath FileASLasimut A script is also provided run_alvl sh which is particularly useful when using the plotting functionalities which require the specification of the di
138. ructures variables functions being part ofthe code The ASIMUTNDocumentationNASIMUT Documentation html contains a WINDOWS link to the index of the web based online help which is ASIMUTNDocumentationNitmlNndex html Remark under WINDOWS the link contains the FULL path to the index html file please do not forget to change the full path in the Properties of the ASIMUTNDocumentationNASIMUT Documentation html file Under UNIX start the online page by selecting the index file ASIMUT Documentation html index html The on line help is also accessible on the Planetary Aeronomy webpage http planetary aeronomie be en asimut_documentation html index html 37 2 3 Howto In this section we will give some advice how to perform specific operations This covers for example How to start ASIMUT How to extract spectroscopic data etc but also issues on how to add specific code inside ASIMUT These are the most asked questions about how ASIMUT runs and works 2 3 1 How to start ASIMUT Under UNIX Change directory to be where the ASILID executable is located or provide the full path to the ASILID executable in the following line ASILID fullPath file ASI Under Windows within the VISUAL environment Select the mode in which you want to run ASIMUT Debug or Release Change the Properties Configuration Properties Debugging set the Command arguments to fullPath file ASI Cli
139. s a combination of 5 scientific programs and 2 auxiliary blocks written in C and Fortran see Figure 1 and Table 1 ASIMUT interacts with the V LIDORT codes with the help of a special interface used to transfer variables from C to Fortran and vice versa The aerosol processing SPHER and T MATRIX codes are linked to the ASIMUT code directly Input files required for LIDORT VLIDORT and changes in the ASIMUT input related to the incorporation of the new RT are explained in detail in ASIMUT Input File pdf ASIMUT Input files p including OEM __ gt Output files BRE ean T inC MEC T E YO ae Interface ALV to link C and Fortran files i gos Sie Interface AL URS aeree to link C and rudes EO Fortran files checking SPHER amp a TEES Output files checking E SPHER amp T MATRIX HIDORT YOIDORT T MATRIX in Fortran in Fortran in Fortran in Fortran A A Input files Input files p 4 1 t Figure 1 Scientific programs included in the ALVL software package 7 Program Purpose ASIMUT under Full independent Radiative Transfer Code no scattering Windows and UNIX When used with V LIDORT Calculation of molecular atmosphere parameters needed as input for V LIDORT simulation of instrument properties OEM Optimal Estimation Method
140. s are provided with the code A general one in the EXAMPLE directory illustrating different options of the forward model with and without LIDORT A second series of examples in the EXAMPLE SOIR directory is more specific to the SOIR instrument is also given and illustrate the simulation and retrieval of SOIR spectra In each case the full set of ancillary data is provided as well as output files LOG OUT and data files to check that the runs are done correctly 2 2 8 1 Extracting Line parameters Uses mars nadir IR calling nadir IR CHA lp inp Shows how to call ASIMUT ALVL to extract the spectroscopic line parameters for a specific run ASIMUT will stops after the extraction of the spectroscopic files which will be used by Example 2 2 2 8 2 Nadir IR spectra of Mars atmosphere no aerosols Uses mars nadir IR asi calling nadir CH4 noAerosol inp This example shows how to call ASIMUT ALVL to simulate an IR spectrum of the Mars atmosphere under nadir viewing No aerosol are included in the run only CO2 CH4 and H20 are included as Line by Line molecules LP 2 2 8 3 Nadir IR spectra of Mars atmosphere with aerosols Uses mars nadir IR asi calling nadir 4 AerosolModel inp or nadir CH4 AerosolMode2 inp Also uses mars nadir IR Lidort asi when running under UNIX with the ASILID version link between ASIMUT and LIDORT This example shows how to call ASIMUT ALVL to simulate an IR spe
141. s in an atmosphere However both codes need to be supplied with already calculated specific aerosol parameters such as aerosol optical depth single scattering albedo SSA and expansion in scalar mode Legendre coefficients Such an approach is not always convenient for the user who faces the necessity to find a way to calculate these parameters outside the codes Moreover the majority of publications describe aerosols in the form of particle types and distributions and not in the form of expansion coefficients To avoid this kind of possible inconvenience it was decided to link both LIDORT and VLIDORT with aerosol processing codes The list of best aerosol processing software available on the scientific market includes the SPHER and T MATRIX codes which are free to download from http www giss nasa gov staff mmishchenko These codes were developed by Dr M Mishchenko NASA Goddard Institute for Space Studies New York USA SPHER calculates scattering and absorption properties of polydisperse homogeneous spherical particles on the basis of the Lorenz Mie theory Mishchenko et al 2002 while T MATRIX is designed to be used for polydisperse randomly oriented particles of identical axially symmetric shapes Mishchenko et al 2002 Mishchenko amp Travis 1998 Both SPHER and T MATRIX handle several particle size distributions The analytical formulas defining these distributions are presented below In these formulas n r is the dist
142. sity S T exhibits a temperature dependence which can be described as OE La 5 07 5 07 200 1 3 7 QT where is the second Planck constant Ac kg 1 4387 cm with the Planck constant kg the Boltzmann constant and c the speed of light vo is the central wavenumber of the j transition E is the energy of the lower state cm 5 and S To is the intensity at the reference temperature Ty cm molecule cm Q T and Q To are the total partition functions under local thermodynamic equilibrium conditions at temperature 7T and respectively These functions describe the temperature dependence of the line intensity of the transition The values of these functions can be determined by ASIMUT by two means see Molecules PartitionFct parameter 1 default The latest TIPS table 29 based on HITRAN2008 has been implemented and is read by ASIMUT in function of the molecule isotopologue and temperature The values are given for temperatures ranging from 70 K to 500 K with a step of 1 K For the moment no interpolation is used the value corresponding to the closest temperature is used 2 The partition functions are approximated by Q T a aT a T 3 8 where do a and are tabulated coefficients 30 48 ASIMUT allows the user to select between different line profiles the Voigt profile being the default one Natural broadening of the lines which occurs as a consequence
143. t must be CR LF under UNIX only CR n 4 Under UNIX check that the name of the directories and files contain only no are allowed This is not a problem under WINDOWS 42 3 Radiative Transfer simulation The ASIMUT program simulates the atmospheric transmittance and radiance in cases where scattering can be neglected and under conditions corresponding to local thermodynamic equilibrium The model assumes a one dimensional atmosphere with all parameters varying only with the vertical variable the altitude ASIMUT can be used to simulate measurements recorded under various geometries nadir off axis nadir down and up looking limb looking with the instrument located at ground level inside the atmosphere or in space The equation describing the radiative transfer through the atmosphere can be written as Le 9 f B v T s av sje ds 3 1 where represents the light intensity at the starting point of the ray path situated at the distance z from the observer a v s is the absorption coefficient B v T is the Planck function and T V 5 5 lt av s ds 3 2 is the optical depth along the path between the points s and s2 The integration in Eq 3 1 occurs along the ray path which must be known beforehand For nadir looking viewing mode 1 is composed of several terms a the radiation emitted by the ground simulated by a blackbody function whose temperature might
144. t to 1 x 0 No Fit l x Fit the surface temperature std deviation SPn FENx FitBaselineCst Fit the baseline constant term Different options are possible Fit of the baseline constant term Take the SPn FENx FitBaselineCst value previously fitted in another FEN std deviation 2 No fit However take the SPn FENx FitBaselineCst value previously fitted in another FEN zd Fit simultaneously on another FEN in which this option will be set to 1 x 0 No Fit l x Fit the baseline constant term x std deviation SPn FENx FitBaselineX Fit the baseline nu term Set to 0 if no fit is requested Set to 1 x to fit with x std deviation SPn FENx FitBaselineX2 Fit the baseline nu 2 term Set to 0 if no fit is requested Set to 1 x to fit with x std deviation SPn FENx FitAlbedo Fit the albedo of the surface SPn FENx albedo Different options are possible Fit of the albedo Take the SPn FENx albedo value previously fitted in another FEN x std deviation 2 No fit However take the 5 FENx albedo value previously fitted in another FEN 1 Fit simultaneously on another FEN in which this option will be set to 1 x 0 No Fit lx Fit the albedo x std deviation Albedo is fitted only if the geometry allows the reflection of Solar radiation on the surface and is asked by the user 95 6 V LIDORT description LIDORT an
145. ters which have to be present in the user file Molecules included in the HITRAN and GEISA databases cover a wide range of atmospheric species as shown in Table 5 which indicates the molecule ID for each possible molecule The isotopologue ID has been chosen as the one defined in HITRAN directly in relation to their relative abundance in Earth atmosphere For some species the GEISA database reports more isotopologues than HITRAN This is the case for the O0 C 0 OCS and CPCH species which are included in ASIMUT The older versions of the HITRAN databases with a slightly different format than the current version are also supported Parameter Symbol Unit Molecule ID mol Isotopologue ID 150 Central wavenumber Vo cm Line intensity Si To cm molec cm Self broadening pressure half width ym cm atm Air broadening pressure half width gor cm atm Temperature dependence of the self broadening Nself coefficient Temperature dependence of the air broadening Nair coefficient Self pressure wavenumber shift cm atm Air pressure wavenumber shift atm Temperature dependence of the self pressure shift ny Temperature dependence of the air pressure shift Collision narrowing parameter p atm Energy of the lower transition level Ej Table 4 Line parameters required by ASIMUT 55 3 2 1 3 Line Mixing to be completed For the moment only for CO2
146. the most intense of the three near infrared collision induced bands The MT CKD 1 2 is based on measurements performed by Mat et al 42 using sample densities from 1 to 10 times that of ideal gas under standard conditions at temperatures of 253 273 and 296 for and O N mixtures From their measurements they derived the collision induced absorption coefficient for a mixture 62 of 7996 and 21 These data are however corrected to take into account observations of the 1 27 um band in atmospheric spectra 43 In the same study the 1 06 um continuum was fitted using the sum of two Lorentzians centered at 9375 and 9439 cm with exponential damping on their low wavenumber sides The MT CKD 1 2 makes also use of the data obtained by Greenblatt under high pressure conditions 44 The second implementation is based on new measurements 41 who reported values for all O O collision induced absorption structures existing between 7000 and 30000 1 cm Herzberg and Wulf continua Measurements of Bernath et al 45 and of Fally et al 46 of the Herzberg continuum and of the Wulf bands of the oxygen molecule have been implemented They cover the wide spectral range from 34000 cm to 41500 cm These measurements were performed at a resolution of 2 cm To obtain the absorption cross section cm molecule of the continuum in this region the following expression has been used 95 o V 9
147. the plotting of input data of intermediate calculations and of final results are located in the ASIMUT Matlab subdirectory They have been compiled to form a Matlab library which is located in lib and so files ASIMUTNDinNLIDNWin64 or NWin32 or NUNIX all necessary files to include in ASIMUTNbinNncludeNWWin64 or Win32 or UNIX respectively runtime dll or so ASIMUT bin Runtime Win64 or NWin32 or NUNIX Asimut gt svn ASIMUT Aerosol AtmModels o Continua crosssection o doc e Fit ForwardModel P instrument Jacobian Lidort o Loading d Local Math molecular Plot radtrans Rodgers Saving Set Spectra eo UserFct VisualC AUX SOURCE Include AUX SOURCE F90 P ex AUX SOURCE V AUX SOURCE VLF90 E M bin Include Unix Win32 Win64 gt Lib P Runtime gt Documentation LIDORT includes sourcecode gt Matlab gt amp SPHER Figure 2 File structure of the ASIMUT V LIDORT code 9 2 2 Compiling and running ASIMUT ALVL 2 2 1 Pre processor definitions The following pre processor definitions are used by ASIMUT ALVL WIN32 Indicates that the program is compiled and run on a Windows 32 machine WIN64 Indicates that the program is compiled and run on a Windows 64 machine default _UNIX_ to indicate
148. the window 11 window having the same PASS number will be fitted simultaneously otherwise the fit is sequential For example if SP1 FEN2 Pass 1 SP2 FEN44 Pass 1 and 5 1 FENI1 Pass 2 the spectra corresponding to the window 2 of spectrum 1 and window 44 of spectrum 2 will be fitted together then the spectrum of window 1 of spectrum 1 will be fitted A series of parameters described the wavenumber wavelength limits of the window The unit in which those limits ar xpressed depend on the SPn DataXUnit flag SPn FENx wavemin wavemax waveminplot wavemaxplot SPn FENx NbdeWeight SPn_FENx deWeightedLimits It is possible to indicate regions inside the limits which will be fitted with a different weight as the rest of the window This is usefull to remove from the fit a bad pixel or known saturated line The two parameters indicate the number of such deweigthed regions and their limits SPn FENx shift constant shift to apply on the wavenumber scale of the spectrum SPn FENx Calibration for FTS multiplicative correction factor applied on the wavenumber scale The value given is indeed the valu of the multiplicative factor 1 0 All wavenumbers will be multiplied by the factor 1 0 SPn_FENx Calibration A series of parameters described the instrument There supersede th values given in SPn fov tdt focal mirrorsize iris oversampling SPn FENx noiseType Type of noise on th
149. tion Fourier transform IR spectrometers with gas cell measurements and new retrieval software Applied Optics 1999 38 15 p 3417 3422 Goldman A and R S Saunders Analysis of Atmospheric Infrared Spectra for Altitude Distribution of Atmospheric Trace Constituents I Method of Analysis J Quant Spectrosc Radiat Transfer 1979 21 2 p 155 161 Kruglanski M A C Vandaele and M De Mazi re Detection of tropospheric aerosols from satellite nadir sounding in the thermal infrared in EAC 2005 Ghent Belgium 105 Annex 1 List of molecules supported and isotopic abundances Isotopic Abundances Used for HITRAN based on P De Bievre N E Holden and I L Barnes Isotopic Abundances and Atomic Weights of the Elements J PAys Chem Ref Data 13 809 891 1984 Molecule Isot Nb Isot code Abundance 1 1 161 0 997317 2 181 0 00199983 3 171 0 000371884 4 162 0 000310693 5 182 0 000000623003 6 172 0 000000115853 CO 1 626 0 984204 2 636 0 0110574 3 628 0 00394707 4 627 0 000733989 5 638 0 0000443446 6 637 0 00000824623 7 828 0 00000395734 8 827 0 00000147180 9 838 0 00000004446 10 837 0 0000000165354 3 1 666 0 992901 2 668 0 00398194 3 686 0 00199097 4 667 0 000740475 5 676 0 000370237 N20 4 1 446 0 990333 2 456 0 00364093 3 546 0 003640
150. tive is to save computation time In this approach the Jacobians are calculated for the first guess Ko and are not updated at each iteration K Ki S Ko 5 11 This also implies that Gia Ep m Go 5 12 and Xi 1 Goly f xi Ko xi x 5 13 In that case the solution converge to x satisfying the equation Xo Xa Goly f x Ko xo x4 5 14 Convergence will be defined by Go K Kool Xj O x Xo 5 15 This approach is correct if the forward model is linear between x and x which is a valid hypothesis for moderately nonlinear problems In such cases it is only necessary to calculate and at the first and last iteration respectively and to use in all equations defining the error on the fit ASIMUT however offers the possibility to update the K at each iteration when the problem cannot be considered linear enough Tikonov regularization method is also implemented 5 1 1 Run options Run Rodgers imposes a constraint on the fitted x when thes ar 0 0 Nothing normal way to use Rodgers No constraint Fitted x can be 0 0 Note that in ASIMUT the concentrations not their logarithm are fitted X values 0 0 are thus meaningless Also note that x 0 0 leads to an error if using V LIDORT Zero any values 0 0 are set to 0 0 Apriori set th negativ fitted x values to their apriori xa values Run Derivatives Jacobians are supposed to
151. tory of IASB BIRA 14 Table 4 Line parameters required by 55 Table 5 Molecules recognized by ASIM LIT ie FEX qn 60 Table 6 List of BRDF kernels implemented in LIDORT VLIDORT 99 List of Figures Figure 1 Scientific programs included in the ALVL software package 7 Figure 2 File structure of ASIMUT V LIDORT code 9 Figure 3 Several geometries are possible Hops is the altitude of the observer the zenithal angle H is the tangent height and is the end point of the ray path 46 Figure 4 Spectral limits defined and used by ASIMUT 54 Figure 5 Geometry of solar occultation measurements and definition of the onion ju dpi E C 4 88 Figure 6 Example size distributions of Martian dust particles Size distribution parameters were taken from Dlugach et al 2002 98 1 Introduction The number of instruments dedicated to the characterization of the Earth atmosphere composition either from the ground or from space even on board balloons or aircraft is steadily growing Flexible tools are needed that can be easily adapted to different instruments and different observation geometries Retrieval is based on accurate
152. trnten botes e titio tates 41 3 Simulation module usc aereo tete den tuae Error Bookmark not defined 3 1 A EH 43 Index of refrati rc 44 3 1 1 1 PLanet eoo oae un Fat e a UM RUIN EU ORE S UU ue E 44 3 1 2 Atmospheres 45 3 1 2 1 AtmosphericModels OpfOS eee eite 45 3 1 3 Geometry of ODSerVatlOn ui essaie sta pe opea IR phos ea aee Hae o tlie V Re era EUER 46 3 1 3 1 rw ODIOSA IMITARI din i 46 3 2 OTE aL ONS s obo 48 3 2 1 ILuesbysImecale latiOrt sos odes erbe aas tu Iu recul fortis eitis 48 3 2 1 1 Grid and resolution used for the OD simulation 51 3 2 1 2 Spectral line catalogues oa pee tetra 55 21 3 Eme MPH Feci Bele 56 3 2 1 4 ODUODSs 5 57 3 2 1 5 XxxDLP iso n SA 57 3 2 2L ME TOSS SECTIONS as efus edid s c uA 60 3 2 2 1 Cross sections catalogues 60 3 2 2 2 amp ae cease n e loisi sca aetati 60 32 62 ah XOYCONUDUS os orationi dies sedeo 62 3 2 3 2 ON COMUEMMUA eoe eee eee aeos 63 3 2 5 9 CO and continua eese teo edite ete 64 3254 Contzinua OptiOnS eet eh ap
153. trum index INPUT FileName name of list file X 0X HF KF KR OH OUPUT type of error ies qop OE OE olo ob de ooo eto ee te oh ort int GetNameSpec NEWINST LIST char FileName int index char strN char Desc nb i char str 1024 Line NL path NULL base NULL stnb NULL FILE hFile int onError ERR NOERROR if hFile fopen FileName rb NULL sprintf str Can not open file s FileName return set_Error ERR_FILE_NONE SEV STOPPROG LINE FILE str 120 for i 0 i lt index i fgets Line NL 1 hFile stnb trim ext Line if allocini String Desc stnb ERR NOERROR onError SEV_STOPFCT errout Line LINE sprintf errout Message Problem Desc goto the name of the next spectrum is the first thing read strN Extract_1s stnb free String amp stnb if allocini_String strN str ERR_NOERROR onError SEV_STOPFCT errout Line LINE sprintf errout Message Problem strN goto TheEnd if hFile NULL fclose hFile free String amp stnb if onError set REPORTING errout Severity errout Line FILE errout Message return onError Inst Instrument structure Geo Geometry structure ZPTfromSpectrum 1 if ZPT information in this file OUPUT
154. type of error TE ERE ea ete EEFE ie ae eh ah heh oboe de oboe se FUNCTION ModifySpec NEWINST LIST SPECTRUM Spectrum ModDesc INSTRUMENT Inst GEOMETRY m int ZPTfromSpectrum E PURPOSE Modify Spectrum INstrument or Geometry in accordance to the string ModDesc INPUT Spectrum Spectrum structure T i ModDesc Descriptio of the measurement read from the list file FK KK K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K K int ModifySpec_NEWINST_LIST SPECTRUM Spectrum char ModDesc INSTRUMENT Inst GEOMETRY Geo Int ZPTfromSpectrum char str int nb if allocini String amp str ModDesc ERR NOERROR return set EPrror ERR REPORTING errout Severity LINE FILE _ errout Message index useful information from the ModDesc string 121 7pPTfromSpectrum NON free String amp str return ERR NOERROR The following is not needed if the OCC option is not required LEELELEFEL LEE E Feb ELLE SEL EE EELS TERETE CELLS CELLS EE ETRE dob FUNCTION GetNbSpec NEWINST OCC char FileName int SamePass T int First PURPOSE Get number of spectra from FileName containing
155. ulations The computation is based on the separability of the functions representing on the one hand the radiative forward model F x b and on the other hand the instrument I b y f xib e I0b 5 16 in which and b are respectively the model parameters needed for the radiative transfer part and the instrument description respectively Note that implicitly Eq 5 16 means that it is not possible to retrieve instrumental parameters such as modulation or phase error with ASIMUT From Eq 3 1 the following derivatives can be deduced OF Ay Ty e 92 oF 2 OF Or OT OT 5 17 The OE derivatives are calculated along the path using relations similar to Eq T 2 Dt 5 47 The 3 L are computed for each species at each altitudes taking into account the different contributing terms LBL term cross section continua For H20 and the derivatives of the Rayleigh scattering are included Moreover the continua expressions depends on the density and vice versa and the corresponding cross terms are also included 5 3 A priori covariance matrices The a priori covariance matrices S should represent the variability of the gas aerosol considered This information is not always available ASIMUT can manage different types of S matrices full matrices provided by the user and obtained 92 with the aid of climatologies diag
156. ure that no information will be lost when performing the convolution adding Wc an interval whose width is calculated as being 4 Res where Res is the final resolution of the spectrum 2 ensure that all lines near the extremities but not within the chosen spectral interval will be taken into account adding Wserect at least 25 0 cm to both extremities this value can be extended by using the MolecLP FBord option for the molecules included in the simulation When y factor is used the width of the spectral interval on which the function is given is considered These limits are summarized in Figure 4 Welect Welect 1 Wc We B 3 3 8 S S WaveSimulMax 3 gt 9 WaveSimulMax A S S B 5 S 2 2 cM gt Range use for retrieval Yt Simulated range internally Figure 4 Spectral limits defined and used by ASIMUT 54 3 2 1 2 Spectral line catalogues Spectroscopic parameters such as line position intensity and broadening coefficients as well as absorption cross sections are taken for practical reasons from existing spectroscopic databases ASIMUT can directly import data from the two widely used HITRAN 38 and GEISA 39 databases but also from the CDSD 54 and CFGL used by the SFIT community database whose format is similar to that of HITRAN However the program can also read user provided Ascii files in which the line parameters are given Table 4 describes the parame
157. us mesosphere measured by SOIR on board Venus Express J Geophys Res 2008 p doi 10 1029 2008JE003140 Wilquet V et al Line parameters for the 01111 00001 band of 12069180 from SOIR measurements of the Venus atmosphere J Quant Spectrosc Radiat Transfer 2008 109 p 895 905 Kurucz R L et al Solar flux atlas from 296 nm to 1300 nm in Natl Sol Obs Atlas 1 1984 Harvard Univ Cambridge Mass 102 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Thuillier G et al The solar spectral irradiance from 200 to 2400 nm as measured by the SOLSPEC spectrometer from the ATLAS and EURECA missions Solar Physics 2003 214 p 1 22 Gallery W O F X Kneizys and S A Clough Air mass computer program for atmospheric transmittance radiance calculation FSCATM H A Air Force Geophysical Laboratory MA Editor 1983 Air Force Clough S A et al Atmospheric spectral transmittance and radiance FASCODIB Proc Soc Photo Opt Instrum Eng 1981 277 p 152 166 Clough S A et al Atmospheric radiative transfer modeling a summary of the AER codes J Quant Spectrosc Radiat Transfer 2005 91 p 233 244 Meier A et al Improvements to air mass calculations for ground based infrared measurements J Quant Spectrosc Radiat Transfer 2004 83 1 p 109 113 Edl n B The refractive index of air Metrologia 1966 2 2 p 71
158. val xxxAER Gamma Gamma parameter in the Conrath formalism xxxAER reff Effective radius um 68 xxxAER Naer For Qext 1 formalisms Densities of aerosols part cm corresponding to the altitudes given in zFact see XXX This parameter is mandatory for Qext formalism xxxAER QextFileName File containing the Qext values 2 col nu cm 1 Qext Qext must be calculated for the given reff in dirAerosol 69 3 3 Radiative transfer model 3 3 1 Determination of the Radiance The radiative transfer calculation in ASIMUT is performed in each layer following the radiation path The radiation exiting one layer j v is obtained from the radiation entering it v through 110 21 021 02 B Vk i ew 3 63 where B j v k is the emission function for the layer and is the transmittance of the layer tn t v se ze 3 64 where the sum is taken over all absorbing processes occurring in the layer o are the absorption coefficients cm molecule and n are the cumulated densities the layer molecule cm Different contributions to the Radiance are considered e Direct contribution from the source Thermal emission from the surface Thermal emission from the atmosphere Reflection of this thermal atmospheric emission on the surface Reflection of the Sun on the surface Lambertian 3 3 2 Atmosphere emission The emission from each layer is given by T s
159. vary smoothly wrt the fitted parameters In that case they are calculated only once at the beginning of the retrieval and once at the end to calculate the error If this hypothesis is not true set Run Derivatives each 91 and the Jacobians will be recalculated at each step of the iterative Rodgers algorithm Run OptimizeRodgers With that option the Rodgers algorithm will be the following use the approximate Jacobians calculated once at the first iteration after convergence recalculate the new KO0 Jacobians and redo a full iterative process Repeat until convergence is achieved on both criteria x and f This is a mix between the approximate Jacobians and the full Rodgers updated at each iteration Run InterpolMethod When fitting a vertical profile see XX the user can specify levels on which the fit will be done These levels must not be the same as the ones used for the Radiative Transfer calculations The Run InterpolMethod flag indicates how to extrapolate the fitted factors to all levels If set to 0 the extrapolated fitted factors on layers above or below the fitted levels are set to the top and bottom fitted values respectively If set to 1 all factors in layers above and below the fitted levels ar set to 1 0 not fitted use the a priori profile 5 2 Determination of the Jacobians Jacobians used by the retrieval module are computed analytically during the forward model calc
160. ype spectra SPn aotfbinningOpt Binning option on the detector This will impose also the Maximum number of bins for binning 3 4 8 bins for binning 12 16 2 bins SPn aotfbinNb Bin number to consider For example using the binning 12 option two bins are recorded simultaneously SPn AOTFFunction Type of AOTF function to use It can be Sinc2 Simple sinc2 5Sinc2 Sum of 5 sinc2 File From ASCII File 2 columns nu aotf filePDS From PDS file When a file file or filePDS options is required its name is given in SPn FileAotfFilter SPn FileAotfFilter File containing the AOTF band pass characterization either as parameterization or the complete function FileAotfFilter format Type FILE One file per bin AOTF filter bandpass function LI param 1 pl p2 p3 12 param 2 pl p2 p3 in param n pl p2 p3 Values of the parameters are calculated using a polynomial relation 1 2 x aotf nucentral p3 x aotf nucentral 2 using maximum 3 coefficients pl p3 ex Zo Created by get all work miniscan m Date 12 Sep 2007 10 23 23 Bin 1 Binning 12 23 7537542783189 0 000164628838726562 81 11 3345861384054 0 00251115825898408 150 41787923001 0 0302391265256812 104 843380735229 0 0219931132766153 134 798482794379 0 0285662006934839 0 0 12 1817220123877 0 00174231790834488 21 5151538373025 0 00129500896606321 83 5570832000095 0 00754897676743825 36 8957781658948 0 00461531397

Download Pdf Manuals

Related Search

Related Contents

mForms 2.4 - openXdata documentation  ASPI-PODO 1 Y 2  Putty Lab 80 TDS  Microdata User Guide  Case Logic 20 Capacity CD Visor  Vortex Media Clock TimeLord-Net Master Clock User's Manual    Utica Boilers SSC Tax Certification  

Copyright © All rights reserved. Failed to retrieve file