LINFOR3D User Manual
Contents
1. eosfile undefined htau0 00 0E5 qmol 00 0E5 mean molecular weight teff 5770 0 grav 27500 0 copenhagen only tsurffac 0 727903 Surface temperature tsurffac Teff Reading of full files COBBOLD only isnap full 1 1 first snapshot to be read from full file s isnap full 2 9 last snapshot to be read from full file s istep full 2 Step for reading snapshots from full file s 3D mean model mavg 4 1 T average 4 T 4 average for defining 3D atmosphere 5 9 Example 37 External 1D reference model atmpath home mst idl spesyn 1Dmodels HM directory of reference model atmfile atm50_shm10c atm name of reference model NONE no reference model Line data radiative transfer Lani s line dat File with line data lutaui 7 0DO lutau2 2 0D0 dlutau 8 0D 2 tau scale defining vertical 5 model extent and resolution lctaui 5 0DO lctau2 2 0D0 dlctau 8 0D 2 universal tau scale for R integration of RT equation ntheta 0 nphi 4 number of theta and phi angles mu0 0 40 kphi 0 view angle if ntheta 0 cos theta kphi pi 2 n_chunk 1 RT is done in n_chunk slices Hbrd 207 8 option for H line broadening 0 default Cayrel amp Traving 1 BPO self reson Griem stark 2 Ali Griem self reson Griem stark ximicx 1 00 microturbulence km s 1D REF atmosphe2. The parameter nlte_flag controls whether the line transfer is performed in LTE nlte_flag 0 or in NLTE n1te flag 1 2 3 The NLTE options work only for lines with available departure coefficients which are read from a separate data file see below value meaning 0 Continuum and lines in LTE 1 Continuum in LTE line source function in LTE line opacity in NLTE 2 Continuum in LTE line opacity in LTE line source function in NLTE 3 Continuum in LTE line opacity and source function in NLTE 5 2 General paths 23 5 1 9 free flag function required type values free pointers in structures at end of program always integer 0 1 If free flag 1 then each run of LINFOR3D allocates fresh memory for the structures ff f1 fx ss s1 and sx In this case the corresponding pointers are removed at the end If you want to examine the structures after the end of execution you must have free flag O If you want to run the program several times in a row with different input parameters you should set free flag 0 in order to avoid additional memory allocation for each run 5 2 General paths 5 2 1 abupath function required type values directory where abu files and atom dat are located always string e g home mst ABU If abupath is not specified in the command file the path is taken from environment variable LINFOR3D_ABU 5 2 2 f
3. create maps ICLAMO ICLAMm m map_flag cc3d_flag 0 0 1 output of CC3 nx ny nx off on rdbb_flag 0 0 1 Read magnetic field write SIR output 0 free pointers in structures at end of program free_flag General paths abupath home mst idl spesyn Linfor_3D ABU Path to abu files and atom dat if not set abupath is read from environment variable LINFOR3D_ABU ff path work2 mst ffcache directory with cached flow fields NONE do not use cached flow fields opapath home mst opta directory with opacity tables gaspath work2 mst eos dat directory with GAS tables eospath work2 mst eos dat directory with EOS tables Model data context cobold rhdpath work1 mst dgt57g44n73 end directory with model data modelid gt57g44n73_3D c d z end data file name parfs gt work1 mst dgt57g44n73 par gt57g44n73_3D_d par parameter file xbcpath work2 mst NLTE3D_data directory of departure coefficients abuid cifist2006 abundance mixtures kiel or cifist2006 dmetal 0 0 logi0 scaling for metal abundances Z gt 3 dalpha 0 0 log10 scaling for alpha elements 0 Ne Mg Si S Ar Ca Ti nx Skip 5 ny skip 5 sampling in x y kiel cobold only more information all read from parameter file for CO5BOLD opafile undefined gasfile undefined
4. 1 Introduction LINFOR3D is based on the old code LINFOR The present version of LINFOR3D is still preliminary The aim is to develop the code in IDL and to transform it to FORTRAN for better execution efficiency after it is checked and found to work properly under IDL Hence users should be aware of the following limitations e Geometry This version is limited to spectrum synthesis from local hydrodynamical models solar type convective atmospheres In a future version it should be possible to compute the line formation in global giant models e Efficiency No effort was yet taken to really improve the execution speed of this IDL FORTRAN code In fact there are still some parts in the code which are unneces sary for the current operation Their execution should be controlled by additional flags in a future release Like the code itself this user manual is under construction too Nevertheless it will be of substantial help while creating new command and spectral line data files for LINFOR3D To get a brief overview you might want to look into the section Getting started Sect 2 first 2 Getting started First make sure that you have all files which are listed in Tab 3 and 2 These files should be put together in a directory which can be accessed under IDL Now two files have to be edited and provided in order to run LINFOR3D e linforset_cmd pro This file which is an IDL script defines the data structure cmd This
5. smallest log TRoss covered by sub model refined z grid always float e g 7 0D0 5 8 3 lutau2 function required type values largest log TRoss covered by sub model refined z grid always float e g 2 0D0 5 8 4 dlutau function required type values z spacing of sub model corresponds roughly to A log TRoss dlutau always float e g 8 0D 2 5 8 5 Ictaul function required type values smallest log Teont used for RT integration always float e g 7 0D0 gt lutaul 5 8 6 Ictau2 function required type values largest log Teont used for RT integration always float e g 2 0D0 lt lutau2 5 8 7 dlctau function required type values resolution in log Teont used for RT integration always float e g 8 0D 2 5 8 8 ntheta function required type values number of 0 angles for which spectrum is computed always integer 0 1 2 3 3 4 6 8 0 Intensity spectrum gt 0 Intensity and flux spectrum 5 8 Line data and radiative transfer 31 5 8 9 nphi function number of d angles for integration of flux spectrum required always type integer values no restriction typically 4 5 8 10 muO function view angle u cos 0 required always type float values 0 0 1 0 If the parameter ntheta 0 then the spectrum and intensity maps are compu
6. e 9 1 Molecules 55 which has the solution 1 yl 8 fi Diil ag If the molecule consists of two different atoms we have two conditions namely see Eq 110 140 Ti 0 cic Dik E tris fi 0 Te Tic Du E aped 0 141 Te From this we see that Zi vk fi fk A 142 Then we have x Di Efe 143 Le or D D 140 gt zy fy 0 144 Le Le Assuming that f gt fx and hence A gt 0 the solution for x can be written as 2 is 145 1 A Dix zeo ya t A Dix 2e0 4 fi Dik Le0 0 and for x we simply have Tio Tko A 146 For each molecule the values of zjo k obtained for the current molecule from Eq 140 or Eqns 145 146 are compared to the previous values zi P obtained from the same condi tions for a previous molecule The new estimate of is then set to the minimum of previous and current estimate a min zio dn a min 24 0 alo 147 The initial guess for the current molecule is then computed as n 1 nti T zi Dia 90 148 Te For simplicity and stability the initial guess for the electron fraction is not updated when changing the initial guesses x for the elements involved in molecule formation 9 1 5 Variable names DAB Dix Le DFOO R R FRACEL Le FRACELO fe FRACEL1 Miete fe FRACI fi FRACJ zi Sij atoms and ions j 1 4 Tik molecules FREEI zi NATMOL N 56 PNUC RIJSUM
7. ran 15 89 0 89 34 In LTE Sf S and BP a Ve 35 P B S 35 The solution of Eq 33 is Tb hine n d SP ri expirar 36 neglecting the boundary term Integration on the fixed 79 scale b Ti c 0 Ky Dio 0 1 n SPD exe G0 dr 37 0 Substituting SP from Eq 34 gives Tb c Diln 0 j 2 n 55 st exp 7a 70 d To 38 0 Ko where K Ko N Lx S and 7 are defined as a function of 79 We can also write b 70 KS Dalro 0 Sn u8 5 8D exot GO st 39 0 where 7 e QU c Qn B i563 5 is the NLTE correction to the line depression source function Integration on the log ro scale zo log 7 gives 40 zb D f a 20 in 10 s 4 74 y uh S 5D exet ri G0 44 41 In the current version of Linfor3D Eq 41 is used to compute the line depression if the parameter intline is set to 1 while Eq 36 is used if intline 2 3 4 Contribution functions The Continuum Intensity Contribution Function for a ray with inclination angle u cos 0 azimuthal angle and wavelength A is simply the horizontal average of the integrand of Eq 12 Cf 14 3 1 5 5500 19 expt r 1 a 42 H Ko XY such that Tb zb NE 33 f CEC i d A ard f In 10 ro 25 moe ta dA 43 Note that now ro u is the optical depth along the line of sight and 79 is the corresponding vertical optical depth a formal quantity in the presenc
8. 2 0D 5 1 0D 2 1 0D 2 4 00D0 0 80D 2 Test grey sf Vdop 2 D 5 eta0 1 0D 1 avgt 1 D 2 1 T 4000 000 2 0D 5 1 0D 1 1 0D 2 4 00D0 0 80D 2 Test grey sf Vdop 2 D 5 eta0 1 0D0 avgt 1 D 2 t oT 4000 000 2 0D 5 1 0DO 1 0D 2 4 00DO 0 80D 2 Test grey sf Vdop 2 D 5 eta0 1 0D1 avgt 1 D 2 1 T 4000 000 2 0D 5 1 0D1 1 0D 2 4 00DO 0 80D 2 Test grey sf Vdop 2 D 5 eta0 1 0D2 avgt 1 D 2 1 T 4000 000 2 0D 5 1 0D2 1 0D 2 4 00D0 0 80D 2 Test grey sf Vdop 2 D 5 eta0 1 0D3 avgt 1 D 2 i 7 4000 000 2 0D 5 1 0D3 1 0D 2 4 00DO 0 80D 2 Test grey sf Vdop 2 D 5 eta0 1 0D4 avgt 1 D 2 1 X 4000 000 2 0D 5 1 0D4 1 0D 2 4 00D0 0 80D 2 clam gfscale 4000 000 1 0 For the following tabulations we have defined AW log p Wr linfor3D logio Wr Eq 75 and AW logio Wr linfor3D logo W7 Eq 75 These results are obtained with intline 1 10 AW AWr dex ntheta 3 ntheta 3 ntheta 4 1 0E 02 0 000342 0 000336 0 002949 0 001690 1 0E 01 0 000331 0 000327 0 002958 0 001698 1 0E 00 20 000269 0 000271 0 003014 0 001755 1 0E 01 0 000072 0 000081 0 003203 0 001942 1 0E 02 0 000820 0 000807 0 004088 0 002831 1 0E 03 0 005441 0 005432 0 008714 0 007456 1 0E4 04 0 021428 0 021421 0 024704 0 023446 These results are obtained with intline 2 15 50 Doppler widths Avp 6 km s a 0 01 The line file used for the test calcul
9. 6 44 2 0 35 s 6 54 3 3 0 36 s 6 68 4 0 35 s 6 51 5 i 0 35 s 6 49 TOUS xi ert ssi E L E total 5 33 s 100 00 The file is also saved during a running LINFOR3D process Thus time statistics are available even after aborting the process The statistics show the system time needed for individual computation steps routines of LINFOR3D and their contribution to the total time in percent For the case that the same operation is performed several times e g doing the radiative transfer for more than one model snap shot the total of all calls the average time and the duration for each individual step is given see example above 9 IONDIS 51 9 1 Molecules Table 4 Small molecular network 5 atoms 8 molecules H CIN TOT Mg H Ho CH NH OH MgH C CH Co CN CO N NH CN O OH CO Mg MgH Table 5 Large molecular network 10 atoms 14 molecules uouot m Mg Ti Cr Fe H Li C N O F Mg Ti Cr Fe H LIH CH NH OH FH MgH CrH FeH LiH Lid CH Cy CN CO NH CN OH LiO CO TIO FH MgH TiO CrH FeH 9 1 1 Some definitions Ne Nkern gt N Po Ne kT fi Ni N amp ern Ti N Nkem Lik Ni k NKern Le Ne Nkern Total number density of nuclei of element i includ
10. C4 0 then use Griem Phys Rev 165 258 1968 and Cowley Obs 91 139 1971 approximation C13 Alog Yraa Enhancement factor for natural line broadening C14 Craa Natural line broadening 1079544 if Craa lt 0 use classical formula Yrag 2 22 101 12 1 s where A is in C15 AA A Line profile is computed from Ay AA to Ay AA 44 6 LINE DATA FILE LINE DAT C16 A A Spacing of wavelength points for spectrum synthesis C17 Wo mA total equivalent width of this blend see below This format has not yet been used or tested and no example files are available As in the cases of format 0 1 and 2 it should also be possible to enter an equivalent width Wo in m now in column C17 For this purpose nbl must be negative with nbl being the number of blend components The gf value producing this equivalent width Wo is returned in result gflg01 average 3D atmosphere and result gflgOx 1D reference atmosphere 6 2 6 Single line calculations complete line data format 7 This data format was designed for simple test calculations where the line profile is fixed i e the line parameters are depth independent see also Sect 3 5 This format has a maximum of 7 columns Description of entries Row 1 Header identifies the meaning of the columns for data in row 5 Row 2 Two integers kline and ktotal kline number of line calculations requested in this file ktotal is the total number
11. Dividing by NkKern we obtain the fractional molecule abundance BrE m _ Nem Tn Tro _ Pe a D zs 107 where we have defined ce Dik Pe ky 108 We note that D p depends only on T and P but not on absolute number densities and hence is constant during the iteration For all elements involved in molecule formation i fimo we have the following conservation equation fi zti y Tik 1 54 109 k or i Di 1 05 fi 11 ur b bia f 110 where is the Kronecker symbol accounting for the correct counting of atoms in molecules with two identical components The electron density is given by N NE Y NZ NZ 111 ie imo ina where the mean degree of ionization of element i Z is defined as 5 j Nig 5 Nu u gt j Nig 5 jS Y j Sij Sir 112 a j 1 3 j 1 3 Nij 3 j 1 3 Here index j runs over the 4 ionization stages j 0 3 of element i For elements forming negative ions the sum includes the negative ion j 1 For such elements Z may become negative Note that Z depends only on T and P but not on the degree of molecule formation Dividing Eq 111 by Nem we obtain E cs SN supe Y Ui 113 i imo t timo Defining 5 ti Zi 5 fi Zi const 114 i mol i imol 9 1 Molecules 53 we have finally Y Aie 115 Elimol Y Combining Eq 110 and we have the following vector equation X F X R 116 where AX oc DEG CEN RIS 117
12. SAHAJ ZEFF NKern kT m Pw Fi Sig atoms and ions j 1 4 Kx molecules 9 IONDIS Index Contribution functions Grey test case IDL IONDIS Line Data line dat LINFOR3D 6 Main program Calling sequence Parameter input linfor_setemd Radiative transfer Structures structures flow field ray system spectrum Transfer equation for the continuum intensity a Transfer equation for the line depression 9 Transfer equation for the line intensity 57
13. Wi u el CF To 4 9 d To 57 1 2 oe In 10 ro 20 CT 70 26 Hs 9 d 20 5 where c Di u Nd X CNE IS 6 X dX en For the flux spectrum we have CH m f ORANAN 59 and 1 70 and WE E cf eias 60 1 2 mop ff ma G0 CH ma with Dr X dX BON P T DEQO FEQD AN Irrespective of the parameter intline the structures contf cfd3i and contf cfd3f hold the contribution functions CP ro 410 bo A0 and CP o A0 while CW ro u0 bo and CW ro are stored in contf cfw3i and contf cfw3f respectively 61 3 5 Grey test case If cmd context is set to grey a 3D nz ny 10 hydrostatic atmosphere is constructed instead of reading a 3D model The temperature stratification on the Rosseland optical depth scale is given by 1 3 1 4 T TRoss Teg 2 E 4 TRoss 62 3 5 Grey test case 13 and the source function is linear in TRoss o 1 3 Sond ST G 3 Thess 63 The Eddington Barbier relation is strictly correct in this case For any inclination u cos 0 the emergent continuum intensity is given by I T5 5 4 64 In particular at disk center u 1 the continuum intensity is 50 Ilm 1 Iz To and the flux is i F 2m ld dp o Thr 66 0 Comparison of the results obtained from LINFOR3D for continuum intensity and flux for TEFF 5000 00 GRAV 316 200 LUTAU1 8 0000000 LUTAU2 2 0000000 DLUTAU 0 0800000 OPAFILE t5000g250mm30_ma
14. files COPBOLD only 2 22 222 202 28 9 5 1 isnapfulll less 28 5 9 2 ISN ap ull 2 uuu wre a Bee ae CE In n Dun ai en D NIS eee 28 55 3 Jstep tulll zoete ok A e ha qe ele a eode 28 EO 29 ee BOG UR Genes a E GEI DEEP IA REUS EE Ge er 29 5 7 External 1D reference model 29 5 7 1 atmpath 5659 zm a a ana OR OE US RA 29 Dz atte PPP 29 5 8 Line data and radiative transfer 2222 2 2 Coon m nn 29 5 8 1 life s uomo a te OS ee ee ee 29 AE aE bk ORA eem Rp ke 30 AA AA 30 Pda Bee Se Be de dar E BO eR ER GS 30 530 Ita ooo 28 82 a rd et ded 30 AN O A RN 30 das Get ak Set e A a AI o aa eee aw 30 5 8 8 MEET Las ae dere A Sa A A Qu Gu ag erg dod 30 F89 APAI ge meh RR ar ade ee ene a a de te Ee ae Pee 31 5 8 10 IA enne ed er Ek he da eed 31 soe ED de de Shas Ae Ae a pe ene ai BE oh en A 31 5 8 12 Hbrdh suas ma an u Ka ee dox b RR ae le Ma me AS 31 TL 32 bia E Sn au e ds hae qd ea SOC X ESPERE A a aaa c OASIS A 32 AM A E O a RATS 32 EE Bk As a a BA 32 A EIA 33 5 3 18 3xumaCo our en 33 A II E are am ee 33 5 89 20 VIC la aos aa BG a A A a a a 33 ee en eee ERE caine Sr eh ae aay ee Eher vet et Ge ee 33 bod goer Seve Wea a eed ads E ae arth eee ae as he ee 33 Busines eh gerela a ee ti A geek he aes ee Gt 34 D 8 24 3n DM sus dob oue Sade rn PR oe we SU EE oe SS E oe a 34 AE AAA 34 A ROA AN 34 5 8 27 intmode y sg Ge a a A Ba eS ee 34 5 8 28
15. for upper level C11 AA A Line profile is computed from Ag AA to Ao AA C12 SA A Spacing of wavelength points for spectrum synthesis C13 Wo mA total equivalent width of this blend see below Row 6 Description for data in row 6 Row 7 clam and gfscale see Sect 6 1 In this case Ar ag is computed from LU DIU LO DIO Function rrca As before the Stark broadening due to collisions with electrons is neglected C4 0 Radiative damping aa is treated in the classical approximation In the case of a single blended line the line dat file looks as follows Example Mult namj ei alam gflg dlgC6 lu diu lo dio dlam ddlam d 8 O I ApJ Line 1 67 6158 17 10 741 1 140 3 2 44 67 800 10 741 6158 15 1 985 1 0 1 99 2 0 0 67 800 10 741 6158 17 1 140 1 0 i 0 0 2 0 0 67 800 10 741 6158 19 0 553 1 0 1 0 02 0 0 4 D 1 4 D 3 clam gfscale 2000 0 1 0 Here kline 1 ktotal 3 nbl 3 incode 1 1 1 Note that only the last of the rows describing the blend need entries C11 and C12 As in the case of format 0 it is possible to enter an equivalent width Wo in m in column C13 For this purpose nbl must be negative with nbl being the number of blend components The gf value producing this equivalent width Wo is returned in result gflg01 average 3D atmosphere and result gflgOx 1D reference atmosphere Example unblended line Mult namj ei alam sflg dlgC6 lu diu lo dio dlam ddlam WO i d 42 6 LI
16. given in the following sections An example follows in Sect 5 1 Program execution flags The user can control the program execution by setting the flags run flag nlte flag cvi flag cv2 flag cv3 flag plt flag maps flag cc3d flag rdbb flag free flag which are ex plained in more detail below 5 1 1 run flag function program mode required always type integer values 2 1 0 1 2 3 usually 3 This parameter determines the general function of LINFOR3D Setting run_flag 2 allows you to compute 3x3 file for the external reference model only Setting run flag 1 allows you to restore old results and replace the results of the pre vious 1D external atmosphere with those of a different 1D external atmosphere Setting run flag O similar to run flag 1 allows you to quickly compare the 3D spectra with another external 1D reference atmosphere Finally the results are saved in files linfor 3D 1 idlsave and linfor_3D_2 idlsave Rarely used setting Setting run flag 1 is used for plotting the structure of the input model on the origi nal grid No radiative transfer calculations are done Setting run_flag 2 is used for plotting the structure of the input model on the re duced refined grid No radiative transfer calculations are done Setting run_flag 3 is the usual case After construction of the 3D atmosphere on the reduced refined grid and of the 1D mean atmospher
17. is the vector of unknown number fractions and E fis fo N fe const 118 is the known constant right hand side N is the number of chemical elements included in the molecular network Equation is a system of N 4 1 nonlinear algebraic equations which can be solved for X by Newton Raphson iteration The first step it to find a suitable starting vector for the iteration Xo This is done as described below The correction 5X giving the next improved estimate of X is computed as follows Assume after n iterations we have Xn FX Rn 119 Then we require that Xn X F X 0X R 120 or Xn 6X F Xn JT X R 121 hence JT 1 6X R Rr 122 The elements of the Jacobian J are defined as OF 123 Jia 55 123 Since we know that F X Y Dipag 1 6i4 fori 1 N 124 Te iN and Fyyi X Ci Zi 125 i 1 N we can readily evaluate Ji j We find from Eqs 124 and 125 OF Ti m NE fori 1 N andi j 126 a Dy ey Di 3 Dia L fori 1 N 127 Ti kzi Ze h 1 N Te OF J N41 g 5 gt Di k Le 1 051 fori 1 N 128 Ox Te IN OF Nais forj 1 N 129 Ox 54 9 IONDIS OF INN 0 130 Le With this information we can solve Eq 122 for 6X and obtain the next estimate Ka Xn X 131 Once the iteration has converged the molecule densities can be computed from Eq 107 9 1 3 Criterion for convergence The criterion for convergence is curr
18. nx ny nlam kline ndata only present if maps_flag 2 see Sect 16 W3LAM equivalent width maps for all models and lines Note e Intensities are given in units of erg cm s sr 1 e The file formerly called linfor_3D idlsave was renamed to linfor_3D_1 idlsave 7 2 Foreshortening Note that the maps include foreshortening effects A model with a quadratic cross section becomes a rectangle when viewed off center If nthetaz 0 the flux spectrum is computed as before and the intensity maps show the vertical view as before Keyword view added to plotting routine linfor plot3 If given the intensity and equiv alent width maps show the foreshortened view 50 8 TIMING STATISTICS 8 Timing statistics At the end of a run of LINFOR3D timing statistics are presented which are also saved to the file linfor_timing txt in the current directory This file could look like this TIMING STATISTICS Routine linfor_find_ff total 0 028 C 0 30 Restoring structure FF total 0 08s 1 55 average 0 01 s C 0 26 0 0 03 s 0 55 4 1 0 01s 0 19 2 0 01 8 C 30419 4 3 0 01s 0 19 4 0 01s 0 19 5 3 0 01s 0 25 Reading 3D model Routine linfor_ionopa_3d Saving structure EE a oenen Radiative transfer for 3D model total 2 08s 39 07 average 0 35 s 6 51 0 0 34s 6 41 1955 0 34s
19. of spectral lines including blends in this case kline 1 ktotal 1 Row 3 String identifier of the line calculation Row 4 Integer nbl integer array incode nbl nb number of blend components for this line calculation 1 incode integer array identifying the input format for each of the blend components 7 Row 5 Cl Central wavelength of blend component A C2 Doppler broadening in units of c vp c C3 Mo Kline Kcont at line center C4 a parameter for Voigt profile a y 2 Awp ratio of half half width of dispersion profile and Doppler widths of Gaussian C5 AA A Line profile is computed from Ay AA to Ay AA C6 A A Spacing of wavelength points for spectrum synthesis C7 Wo mA total equivalent width of this blend see above Row 6 Description for data in row 6 Row 7 clam and gfscale see Sect 6 1 Example alam Vdop eta0 avgt dlam dd lam 1 1 Test grey sf Vdop 2 D 5 eta0 1 0D0 avgt 1 D 2 Lp 7 4000 000 2 0D 5 1 0DO 1 0D 2 0 90D0 0 90D 2 clam gfscale 4000 000 1 0 6 2 7 Multiple Line Calculations It is also possible to process a whole set of lines in a single run The requirement is however that all lines have the same central wavelength continuum wavelength This mode was designed for parameter studies e g investigating the granulation abundance corrections as a function of line excitation potential 6 3 Conversion of line broadening parameters 45 Example 8 unblended N I lines of diffe
20. pro S Integration of RT equation linfor refatm pro S Define 1D reference atmosphere from 3D flow field linfor regrid pro S Cut out surface layers from original model re define grid linfor tauinfo pro S Prints information about optical depth scales linfor ztau pro S Prepares bundle of inclined rays on monochromatic tau linfor_plot0 pro S Plots flow field linfor_plot1 pro S Plots spatially resolved line profiles linfor plot2 pro S Plots averaged line profiles linfor_plot3 pro S Plots monochromatic granulation images alpha line pro F Computes a parameter for VOIGT function eta0 pro F Computes no the opacity at line center of metal lines rrca pro F Computes mean square orbital radius of electron Uns ld vdop pro F Computes thermal turbulent Doppler velocity cs linfor_timing pro S Prepares and gathers timing statistics linfor timing print pro S Print timing statistics Table 3 List of all IDL modules the table shows the file name the type Subroutine or Function and its description 20 5 PARAMETER INPUT LINFOR_SETCMD PRO 5 Parameter Input linfor setcmd pro The input parameters except for those defined in line dat see Sect 6 are basically specified by editing the routine linfor_setcmd pro In this way the user defines the structure cmd see Table 1 The order of entries is irrelevant Parameters which are not required may be omitted A detailed explanation of the various input parameters and their possible values is
21. s required only needed if context cobold type integer values us UE 5 5 2 isnap full 2 function last snapshot to be read from full file s required only needed if context cobold type integer values MEL 5 5 3 istep full function step for reading snapshots from full file s required only needed if context cobold type integer values zw 5 6 3D mean model 29 5 6 3D mean model 5 6 1 mavg function mode of averaging 3D T structure on TRoss required always type integer values 1 4 value meaning 1 T 3p 7 Ross T3p TRoss 4 Trap TRoss Th TRoss 1 4 5 7 External 1D reference model 5 7 1 atmpath function directory with 1D model atmospheres required always type string values e g home mst atm 5 7 2 atmfile function name of 1D reference model required always type string values e g NONE falc atm Note No external 1D reference atmosphere will be used if atmfile NONE In this case the parameter atmpath has no meaning 5 8 Line data and radiative transfer 5 8 1 linfs function name of line data file required always type string values e g Li67 line Note If linfs is not specified the default value line dat is assumed 30 5 PARAMETER INPUT LINFOR_SETCMD PRO 5 8 2 lutaul function required type values
22. structure contains all necessary information except for spectral line data like e g paths and names of model file s See Sect 5 for more details e line dat This file contains all data for spectral line such as e g oscillator strength and broadening parameters The usual file name is line dat but it might be given another name which then has be to entered in linfor_setcmd pro See Sect 6 for more details After done so you might start IDL and start LINFOR3D by simply typing IDL gt r linfor_3D pro Several output files are created It is also possible to directly use the results under IDL See Sect 7 for more details Note that LINFOR3D stores the flow field in temporary cache files which are automatically restored if the same calculation is repeated 3 Basic Equations of Radiative Transfer 3 1 Transfer equation for the continuum intensity ae K I KS SS 1 together with the definition of the optical depth along the ray dry K ds 2 reads ar T 3 The solution of Eq is b RED f RE el rar BCR exot rt 79 4 TX where T is the continuum optical depth at the lower boundary The emergent continuum intensity is Tb ise 0 SSC exptorjar 150 expli 5 0 Defining uy IX Sy 6 we have the transport equation dus e Asi 7 dr UA drs 7 The solution for u is found by replacing S by d S d 7 in Eq 4 ro SED ata a dr vi est 8 The e
23. 19 He 0 gt 14 8 sse opa ar Km st Q4 Integration on the log ro scale zo log To gives a HG ma0 mo 23 149554 xlr dh Al PAG 2 where 26 is the minimum log optical depth Alternatively in analogy to Eq 9 we obtain X dS Kin 0 o SAG rar uio expt rt 20 where we have defined u Ik Sy 27 which in the diffusion approximation may be written as dS dr KG KS dS b or h n A SO Lb u n ig 28 On the universal optical depth scale 7 we obtain from Eq 26 Kin 0 f S9 rara hi ein 09 In LTE where S S the integral in pu differs from the integral in Eq 15 only by the exponential factor which involves the total optical depth 7 instead of the continuum optical depth 7 The absolute line depression is then calculated as D I r 0 kl 30 In nz current version of Linfor3D Eq 25 is used if the parameter intline is set to 1 and Eq 26 is used if intline is set to 2 3 3 Transfer equation for the line depression We may analyse the transfer equation for the absolute line depression defined in Eq 30 SIS x H S 31 EF de As Ky ty STU A 2 A 31 or dD c c L of de Ky Sr Dy HOD SEDD Ky Sy 32 or d Dy d 74 ie 33 10 3 BASIC EQUATIONS OF RADIATIVE TRANSFER where the line depression source function is D 7 ce _ca _ 7 c CY c CL S 77 5 3
24. 3 5948 530 3 130 1 0 90 00 0 0 13 80 0 0 1 0 5 D 1 5 D 3 clam gfscale 2000 0 1 0 Here kline 1 ktotal 2 nbl 2 incode 2 2 2 Note that only the last of the rows describing the blend need entries C12 and C13 As in the cases of format 0 and 1 it is possible to enter an equivalent width Wo in m in column C14 For this purpose nbl must be negative with nbl being the number of blend components The gf value producing this equivalent width Wo is returned in result gflgO1 average 3D atmosphere and result gf1g0x 1D reference atmosphere No examples are given since the necessary modification the the data format should be obvious 6 2 5 Single line calculations complete line data format 3 This data format has a maximum of 17 columns It differs from format 2 only in the way the van der Waals broadening parameters are specified Columns C7 with Ar a2 is replaced by the four columns C7 LU Orbital quantum number of valence electron of lower level C8 DIU excitation energy eV of parent term for lower level C9 LO Orbital quantum number of valence electron of upper level C10 DIO excitation energy eV of parent term for upper level as in format 1 The remaining columns are as in format 2 but shifted by 3 C11 Alog C4 Enhancement factor for Stark line broadening C12 log C4 Stark broadening constant if log C4 lt 0 typically log C4 1 0 then Cy 0 if log
25. LINFOR3D User Manual version 5 1 4 Matthias Steffen Hans G nter Ludwig Sven Wedemeyer B hm October 21 2014 N Contents 1 Introduction 2 Getting started 3 Basic Equations of Radiative Transfer 3 1 Transfer equation for the continuum intensity 3 2 Transfer equation for the line intensity 3 3 Transfer equation for the line depression 3 4 Contribution functions 3 5 Grey test case 4 Program Files and Data input files 4 1 Main program flow 4 2 Structures in Common Block linfordata 4 3 IDL Files 5 Parameter Input linfor_setemd pro 5 1 Program execution flags 5 1 1 5 1 2 5 1 3 5 1 4 run_flag cvl_flag cv2_flag cv3 flag 5 1 5 pit flag 5 1 6 maps flag 5 1 7 cc3d_flag 5 1 8 nlte flag 5 1 9 free flag 5 2 General paths 5 2 1 abupath 5 2 2 ff path 5 2 3 opapath 5 2 4 gaspath 5 2 5 5 3 Model data 2 9 1 eospath context 5 3 2 rhdpath 5 3 3 modelid 5 3 4 parts 5 3 5 xbcpath 9 3 6 9 3 9 5 4 1 5 4 3 5 4 4 htau0 5 4 5 qmol 5 4 6 Tefl 5 3 10 ny skip 5 4 More model information all read from parameter file for CO BOLD data opafle cen abuid 5 3 7 dmetal 5 3 8 dalpha nx skip eosfilel CONTENTS CONTENTS 3 DA OT OM une Pede de u eae m RR Banane Bea qb Eee er ee ie 27 bus tsubtiac du sadas a ooh x YE RIP Rok e we A Hae eee we Ps do Gow a 27 5 5 Model data reading of ful
26. NE DATA FILE LINE DAT O I ApJ Line 2 92 6300 30 0 000 9 773 i 1 92 800 0 000 6300 30 9 773 1 0 1 0 0 2 0 0 4 D 1 4 D 3 7 00 clam gfscale 2000 0 1 0 Example blended line Mult namj ei alam gflg dlgC6 lu diu lo dio dlam ddlam WO i 3 O I ApJ Line 1 67 6158 17 10 741 1 140 3 4131 67 800 10 741 6158 15 1 985 1 0 1 0 0 2 0 0 67 800 10 741 6158 17 1 140 1 0 1 0 0 2 0 0 67 800 10 741 6158 19 0 553 1 0 1 0 0 2 0 0 4 D 1 4 D 3 10 00 clam gfscale 2000 0 1 0 6 2 4 Single line calculations complete line data format 2 For a single unblended line the this form of the line dat file looks like this Example Mult namj ei alam gflg dlgC6 drrcai dlgC4 CAlg dlggr Crad dlam 1 1 Si I AA Line 5 16 5948 540 5 0823 1 130 390 03 11 80 1 86 1 2 16 1400 5 0823 5948 540 1 130 1 0 390 03 0 0 11 80 0 0 1 0 B D 1 clam gfscale 2000 0 1 0 Description of entries Row 1 Header identifies the meaning of the columns for data in row 5 Row 2 Two integers kline and ktotal kline number of line calculations requested in this file ktotal is the total number of spectral lines including blends in this case kline 1 ktotal 1 Row 3 String identifier of the first line calculation Row 4 Integer nbl integer array incode nbl nb number of blend components for this line calculation 1 ddlam 5 D 3 incode integer array identifying the input format for each of the blend components 2 Row 5 Line data
27. Y NDATA KLINE NLAM MODELID MUO PHIO CLAM LINEID DV3 ICLAMO ICLAM2 LONG LONG INT INT LONG STRING FLOAT FLOAT FLOAT STRING FLOAT FLOAT FLOAT 140 140 12 1 11 Array 12 1 00000 0 00000 3966 34 Array 1 Array 11 Array 140 140 12 Array 140 140 11 1 12 Depending on the value of the control parameter maps flag see Sect 5 1 there might be a tag named ICLAM1 ICLAM1 FLOAT Array 140 140 12 1 instead of ICLAM2 or even both might be missing if maps flag 0 7 2 Foreshortening 49 Description of entries nx ny x y dimensions of the 2D images ndata number of models for which spectrum synthesis was done kline number of lines for which spectrum synthesis was done clam central continuum wavelength for all maps modelid model identifier 0 ndata 1 lineid line identifier O kline 1 ICLAMO continuum intensity maps for all models at clam dimensions nx ny see Sect ICLAM1 emergent intensity maps for all models and lines including line absorption at clam window center dimensions nx ny ndata kline only present if maps flag 1 see Sect 6 ICLAM2 emergent intensity maps for all models and lines including line absorption at all wavelengths within the wavelength window of width 2 dlam around the central wavelength clam A clam dlam i ddlam where clam alam dclam alam is the wavelength of the main blend component as defined in line dat dimensions
28. ations is 3 BASIC EQUATIONS OF RADIATIVE TRANSFER 16 no AW AWr dex ntheta 3 ntheta 3 ntheta 4 1 0E 02 4 0 000334 4 0 000328 0 002957 0 001698 1 0E 01 4 0 000324 4 0 000320 0 002965 0 001706 1 0E4 00 0 000261 0 000263 0 003022 0 001762 1 0E 01 0 000063 0 000074 0 003211 0 001950 1 0E 02 0 000825 0 000814 0 004097 0 002838 1 0E 03 0 005447 0 005438 0 008720 0 007463 1 0E 04 0 021432 0 021425 0 024708 0 023450 17 4 Program Files and Data input files In this section all the program files making up the LINFOR3D package are listed First an overview on the program flow and the structures in common block linfordata is given The format of the data input files is described since the primary way the user controls the program execution is via the control parameters read from linfor_setcmd pro see Section 5 The line parameters are specified in the input file line dat see Section 6 4 1 Main program flow Basically the calling sequence is as follows incomplete listing of linfor_3D pro Read input parameters linfor_setcmd pro Initialize atomic data linfor_atom pro Read line data linfor_rdline pro Initialize ionopa abundances opacity tables and EOS tables Set constants linfor_init Define ff type linfor flowfield linfor_flowfield_define pro Define 1 type linfor flowfield linfor_flowfield__def
29. by the linfor_3D package Structure Defined in Description atom linfor_atom pro Atomic weights amp ionization potentials const linfor_init pro Physical amp model constants cmd linfor_setcmd pro Input parameters controlling program execution line linfor_rdline pro Line data derived from line dat gas linfor_init pro GAS tables initialized by tabinter_rdcoeff eos linfor_init pro EOS tables initialized by tabinter rdcoeff result linfor_init pro Basic results for computing abundance corrections Table 1 List of all structures in common block linfordata the table shows the name of the structure the routine where it is defined and a description 4 3 IDL Files Table 3 shows a list of all source files necessary to run LINFOR3D Finally you will also need the files which are listed in Tab File name Type Description blam pro F Computes the Kirchhoff Planck function monocubic pro F Performs monotonic piecewise cubic interpolation ms_int pro F Integrates a given function over optical depth Table 2 List of additional IDL modules which not unique to LINFOR3D 4 3 IDL Files 19 File name Type Description linfor_3D pro S main program linfor flowfield define pro S Definition of flow field structure linfor spectrum define pro 5 Definition of spectrum structure linfor raysys define pro S Definition of ray system s
30. c Stark broadening The required data is however not always available and must be converted from other broadening parameters e g y4 In the particular case of the Vienna Atomic Line Database the broadening is provided as log y4 N for a temperature of T 10 K Please note that here and in Linfor3D in general the parameters C n 4 6 are defined via C Aw 79 r whereas the definition by Uns ld is C Aw 27 Z 80 r The Linfor parameters C are thus a factor 27 larger than in the definition by Unsold 6 3 1 Quadratic Stark effect The broadening parameter 4 for the quadratic Stark effect can be written as ya 11 37 C2 vuU Ne 81 where Ure is the relative velocity between the regarded atom and the perturber i e the colliding particle 2 ART 1 1 Urel ET a rus 82 46 6 LINE DATA FILE LINE DAT A and Asa are the atomic weights in atomic mass units e g Aa 1 for a colliding hydrogen atom and A 56 for iron atoms and Ag 1 1837 me my for electrons For Stark broadening with electrons as perturbers the following good approximation can be made 1 1 1 A Ag gt x 1837 1 gt gt Ag As A A 837 my me 83 With this Eq 81 can be written as log B log11 37 log ore log vrai 1 2 84 e 2 1 T dosis er ios 85 3 6 T Me 2 1 1 056 5 log Cy 1 931 log T 86 2 1 1 T 1 056 log C4 1 931 log 10 log _ ger ENT EIR io 88 Wit
31. d in ionopa routines required always type float values eg 0 0 0 4 The logarithmic enhancement factor to be applied to all a elements O Ne Mg SiS Ar Ca Ti 5 3 9 nx skip function sampling of model in x direction required if context cobold kiel muram type integer values 1 4 10 1 If both nx_skip and ny_skip see Sect 5 3 are negative the original data are re binned from nx ny to nx abs nx_skip ny abs ny_skip In the usual case that both nx_skip and ny_skip are positive the original data are re sampled skipping by nx_skip in x and by ny_skip in y direction nx nx_skip and ny ny_skip should preferably be an integer If nx_skip and ny_skip have different signs an error message is printed and the program is stopped The value 1 has no effect 5 3 10 ny skip function sampling of model in x direction required if context cobold kiel muram type integer values 1 4 10 1 For details see description of nx_skip Sect 5 3 5 4 More model information all read from parameter file for CO BOLD data The parameters in this section are ignored in the case of COPBOLD data and instead read from the specified CO BOLD parameter file 5 4 1 opafile function name of opacity file binned opacity tables required not needed if context cobold type string values eg g2v opta 5 4 Mo
32. d in the classical approximation In the case of a single blended line the line dat file looks as follows Example 40 6 LINE DATA FILE LINE DAT Mult namj ei alam gflg dlgC6 drrcai dlam ddlam t 2 Fe I 0 00 eV Fe II 3 00 eV 2000 A 2 00 9999 2600 0 000 2000 0 6 441 1 0 10 0 9999 2601 3 000 2000 0 4 550 1 0 10 0 1 5D 1 1 5D 3 clam gfscale 2000 0 1 0 Note that it is not necessary that the blend components belong to the same ion Here kline 1 ktotal 2 nbl 2 incode 0 0 Note that only the last of the rows describing the blend need entries C8 and C9 With a slight modification it is possible to enter an equivalent width Wo in m in column C10 For this purpose nbl must be negative with nbl being the number of blend components The gf value producing this equivalent width Wo is returned in result gflg01 average 3D atmosphere and result gf1lg0x 1D reference atmosphere Example unblended line Mult namj chik alam sflg dlgC6 drrcai dlam ddlam WO d 1 N I Fictitious Line 1 0 000 5500 0 7 6914 1 00 10 00 75 00 1 0 9999 700 0 000 5500 0 7 6914 1 00 10 00 3 00E 01 3 00E 03 75 00 clam gfscale 2000 0 1 0 Example blended line Mult namj chik alam gflg dlgC6 drrcai dlam ddlam WO 1 2 Fe I 0 00 eV Fe II 3 00 eV 2000 A 2 00 9999 2600 0 000 2000 0 6 441 1 0 10 0 9999 2601 3 000 2000 0 4 550 1 0 10 0 1 50D 1 1 50D 03 100 00 clam gfscale 2000 0 1 0 6 2 3 Single line calculations
33. data mst model 5 3 Model data 25 5 3 3 modelid function required type values name of 2D 3D model file always string e g gtb7g44n66 3Dgz end Note A list of files can be specified by using wildcards e g chro3D04 full 5 3 4 parfs function full path to parameter file rhd par required CO BOLD only type string values e g data mst model par gt57g44n66 par 5 3 5 xbcpath function required type values full path to xbc files xbc CO BOLD only string e g data mst NLTE3D data model Note xbc files are necessary for NLTE line formation calculations Presently limited to Li for selected CO BOLD models 5 3 6 abuid function Abundance mixture to be used in ionopa routines required always type string values kiel cifist2006 special abuid identifies the solar abundance mix which is then modified according to dmetal and dalpha see below The corresponding tables kiel abu cifist2006 abu or special abu must be located in diectory abupath 5 3 7 dmetal function required type values metallicity M H log o to be used in ionopa routines always float e g 0 0 0 5 2 0 The logarithmic abundance of all elements beyond Li N gt 3 is changed by dmetal 26 5 PARAMETER INPUT LINFOR_SETCMD PRO 5 3 8 dalpha function alpha enhancement to be use
34. e the line formation calculations are done and the results are plotted linfor_plot1 spatially and temporally resolved line profiles and bisectors linfor_plot2 surface and time averaged line profiles and bisectors Finally the results are saved in files linfor_3D_1 idlsave and linfor_3D_2 idlsave 5 1 Program execution flags 21 run_flag value control of program flow 1 restore results compute 1D ref atmosphere amp spectrum save results restore results compute 1D ref atmosphere amp spectrum plot2 save results compute 3D 1D atmospheres 1 plotO1 stop compute 3D 1D atmospheres 1 2 plot02 stop compute 3D 1D atmospheres 1 2 line formation ploti plot2 save results 5 1 2 cvl flag function required type values enforce pu 0 always integer 0 1 The parameter cv1 flag controls whether or not the z component of the velocity field is adjusted to ensure zero mass flux in z direction 0 no 1 yes Default 0 5 1 3 cv2 flag function required type values enforce p uy 0 always integer 0 1 The parameter cv2_flag controls whether or not the y component of the velocity field is adjusted to ensure zero mass flux in y direction 0 no 1 yes Default 0 5 1 4 cv3 flag function required type values enforce pu 0 always integer 0 1 The parameter cv3_flag controls whether or not
35. e of horizontal inhomogeneities 3 4 Contribution functions 11 The Continuum Flux Contribution Function at wavelength X is consequently 27 1 CS 1 A wl C ro i 8 Nau ag 44 such that b Tb Z Bro 79 3 Opad f ma0 ml Chlla Na 45 Note that the horizontal averaging in Eq 42 works only because the transfer equation is inte grated on the fixed universal optical depth scale 7 The contribution functions C 7o to Po Ao and C 7o Ao are saved in contf cfc3i and contf cfc3f respectively Corresponding con tribution functions are also computed for the 3D model and saved in contf cfcii and contf cfcif respectively and for the external 1D reference atmosphere contf cfcxi and contf cfcxf Similarly we can also write down the Line Intensity Contribution Function as the horizontal average of the integrand of Eq 24 1 CHA 1 2 1 S ro n tn 46 xy such that the intensity at a given wavelength in the line profile is b To 20 Blo 0 9 Camodd mao ml Ciis 6 X da AT The Line Flux Contribution Function at wavelength A is 27 1 Choa f Cir dx arad 48 such that b Tb 2 gites Xs n Ott A dr f Hannei aed 49 Cf To Ho do Ay and CL 1o Ao are stored in contf cf13i and contf cf13f respectively and similarly for the 1D atmospheres in contf cf11i contf cfllilf contf cflxi and contf cflxf Formally a Line Depression Contribution Funct
36. ently jot S091 lt 1 10 f 132 i and n n je Da Tk 1 4 biz fi lt 1 107 fi 133 k n Te for all elements 7 The maximum number of iterations is 15 9 1 4 Initial guess The initial concentrations of free atoms and ions of elements involved in molecule formation z are computed as follows First we assume that no molecules are formed and so the initial x are set to fi Tio fi 134 for all elements From this the electron fraction x is computed as Teo Max Z fe 5 Ti 2 135 i imol where Zemin 1 10 79 Using this value for ze we compute the molecule concentrations Tik according to Eq 107 If the resulting Tik lt 1 107 min zi k 136 the formation of this molecule is considered negligible and no correction of r z and x x is necessary If 1 107 min zi g lt Tik MIn k 137 molecule formation is no longer negligible but also not exhaustive In this case the molecule concentrations must be iterated but the initial guesses for k r and x need not be changed Finally if Tik m nqd sk 138 then molecule formation is exhaustive and the initial guesses for x y x and x are changed We compute x and x as the equilibrium values that would result if only this particular molecule was present If the molecule consists of two atoms of the same element the condition is see Eq 110 q 2 Dii Ti fi 0 139
37. ergent line profile from Eq or 71 At disk center we have _ 3 no H a v 5 19 H a v 1 Di p 1 Te u 1 Rr 73 and for flux 1 no H a v 2 no H a v 1 Clearly the emergent line profiles are no longer Voigt profiles due to saturation effects The reduced disk center equivalent width is obtained from numerical integration of the emergent line profile Dp F Rp 74 z oo Wr Ri v no a dv 75 oo and Wr 5 6 Wr An analytical curve of growth W no a 0 01 is shown in Fig 1 The equivalent width in mA is obtained from the reduced equivalent width by W m 1000 AoA W 76 100 00 10 00 Lul 1 00 Lul 0 10 Lul 0 01 E lt v l l l l l l l 3 0 1 0 10 1 00 10 00 100 00 1000 00 10000 00 eta0 Figure 1 Analytical curve of growth showing the reduced equivalent width integrated from v 100 to v 100 as a function of ro assuming a 0 01 Black disk center red flux The dashed lines have slopes 0 5 and 1 0 Diamonds show the numerical results obtained with LINFOR3D integration from v 50 to v 50 The results of a number of test calculations are listed below The wavelength resolution was chosen to be 1 10 of the Doppler width 6X 0 1 Ao Avp c The wavelength range was set to 3 5 Grey test case shown below alam Vdop eta avgt dlam ddlam T 7 Test grey sf Vdop 2 D 5 eta0 1 0D 2 avgt 1 D 2 1 7 4000 000
38. f integration in routines ms int tau and ms int exp which can be linear 5 8 Line data and radiative transfer 35 0 or monotonic and cubic 1 standard 5 8 28 intline function mode of integrating the line transfer equation required always type integer values 1 2 1 2 Determines the method of integrating the line transfer equation see Section 3 for details Default value is intline 1 value meaning 1 Line depression on fixed log 7 scale Eq 2 Line depression on monochromatic 7 scale Eq 36 1 Line intensity on fixed log 7 scale Eq 2 Line intensity on monochromatic 7 scale Eq 26 5 8 29 nchunk function rad transfer is done in n chunk slices required always type integer values eg 2 Default is nchunk 1 i e the whole model is processed as one block For large models it may be necessary to split the computation into several chunks to save memory 36 5 PARAMETER INPUT LINFOR_SETCMD PRO 5 9 Example pro linfor_setcmd common linfordata Program execution flags nlte_flag 0 0 1 2 3 LIE or NLTE for lines with xb run_flag 3 execution mode 2 1 0 1 2 3 cvi_flag 1 0 1 enforce lt rho vi gt z 0 off on cv2_flag 1 O 1 enforce lt rho v2 gt z 0 off on cv3_flag 0 0 1 enforce lt rho v3 gt z 0 off on plt_flag 1 1 0 1 plotting off bisectors off on maps_flag 1
39. f path function directory to be used for reading and writing cached flow fields required always type string values e g data mst ffcache 24 5 PARAMETER INPUT LINFOR_SETCMD PRO 5 2 3 opapath function required type values directory with opacity tables opta files always string e g home mst RHD opa dat 5 2 4 gaspath function required type values directory with GAS tables gas_ eos files always string e g home mst RHD eos dat 5 2 5 eospath function required type values directory with EOS tables eos_ eos files always string e g home mst RHD eos dat 5 3 Model data 5 3 1 context function source of input model required always type string values e g cobold value meaning cobold 3D CO BOLD copenhagen N amp S 3D code kiel Kiel 2D HDW Code muram MURAM 3D MHD Code grey construct grey 3D nz ny 10 hydrostatic atmosphere for test purposes The TRoss grid of the grey atmosphere is defined by the parameters cmd lutaul cmd lutau2 cmd dlutau The atmospheric parameters must be specified as cmd Teff and cmd grav The opacity table must be specified as cmd opafile and the equation of state as cmd eosfile and cmd gasfile 5 3 2 rhdpath function required type values directory with 2D 3D model atmospheres end full files always string e g
40. h T 104 K which is assumed for data in VALD we derive Ya 2 log 3 654 log C 89 og N 3 og C4 89 and finally the conversion formula log C4 1 5log 24 5 4805 90 Ne For instance a value of 5 491 from VALD gives log C4 13 717 The parameter C4lg is thus set to 13 717 6 3 2 Van der Waals broadening The broadening parameter Ye for the van der Waals effect can be written as Ve 8 08 Gen Ye Ny 91 The perturbing particles are mostly hydrogen atoms with Aa 1 We now make the approxima tion i i i A1 gt A2 x 1 92 1 2 gt 7 a 92 With this the relative velocity of the particles Eq reduces to gt 8kT Urol m 93 We can thus rewrite Eq log 2 amp 1088 08 log p log v4 94 H 2 3 8kT 0 907 log Ce 1 T 9 SEE 10 T MH 2 2 0 907 log C 2 497 M log T 96 2 3 3 T 0 907 log Co 2 4 log 10 log z log Cs 97 jg 108 0 10 8 19K 97 6 3 Conversion of line broadening parameters 47 With T 10 K which is assumed for data in VALD we derive Y6 2 log 4 604 log Ci 99 8 Nn g log Ce 99 and finally the conversion formula log Cs 2 5 log x 11 510 100 H For instance a value of 7 619 from VALD gives log Cg 30 558 Unfortunately there is no parameter C61g which could be set directly in the present code version Instead the van der Waals broadening can be
41. i amp Griem 1966 Phys Rev 144 366 BPO Barklem Piskunov and O Mara 2000 A amp A 363 1091 A08 Allard et al 2008 A amp A 480 581 G Griem 1960 ApJ 132 883 with corrections to approximate the Vidal Cooper amp Smith 1973 ApJS 25 37 profiles Note option Hbrd 2 has an effect only on Ho Hf and Hy and Hbrd 3 affects only Ha all other hydrogen lines are treated according to option Hbrd 1 unless Hbrd 0 5 8 13 ximicx function required type values isotropic Gaussian microturbulence velocity km s for external 1D reference model added quadratically to thermal velocity always float e g 1 0 5 8 14 ximicl function required type values isotropic Gaussian microturbulence velocity km s for 3D mean models added quadratically to thermal velocity always float e g 1 0 5 8 15 ximic3 function required type values isotropic Gaussian microturbulence velocity km s for 2D 3D models added quadratically to thermal flow velocity always float e g 1 0 5 8 16 ximacx function required type values Isotropic Gaussian macroturbulence velocity km s for external 1D reference model additional line broadening after line formation always float e g 1 6 5 8 Line data and radiative transfer 33 5 8 17 ximacl function required type values Isoptropic Gaussian macroturbulence ve
42. in format 2 11 2 columns C1 Multiplet number for information only C2 Identifier of atom or ion e g 2601 mean Fell C3 Excitation potential of lower level in eV C4 Central wavelength of blend component C5 log gf value of blend component C6 Alog Cg Enhancement factor for van der Waals line broadening CT Ar az Difference of mean square electron orbital radii C8 Alog C4 Enhancement factor for Stark line broadening C9 log C4 Stark broadening constant if log C4 lt 0 typically log C4 1 0 then Cy 0 if log C4 0 then use Griem Phys Rev 165 258 1968 6 2 Line Data Formats 43 and Cowley Obs 91 139 1971 approximation C10 Alog Yraa Enhancement factor for natural line broadening C11 Cyaa Natural line broadening 107554 if Craa lt 0 use classical formula yraa 2 22 10 A 1 s where A is in A C12 AA A Line profile is computed from Ag AA to Ay AA C13 6A Spacing of wavelength points for spectrum synthesis C14 Wo mA total equivalent width of this blend see below Row 6 Description for data in row 6 Row 7 clam and gfscale see Sectl6 1 In the case of a single blended line the line dat file looks as follows Example Mult namj ei alam gflg dlgC6 drrcai dlgC4 CAlg dlggr Crad dlam ddlam i 2 Si I Si II blend 16 5948 540 5 0823 1 130 390 03 11 80 1 86 2 22 16 1400 5 0823 5948 540 1 130 1 0 390 03 0 0 11 80 0 0 140 16 1401 0 082
43. ine pro Define fx type linfor flowfield linfor_flowfield__define pro Define ss type linfor spectrum linfor spectrum define pro Define s1 type linfor spectrum linfor spectrum define pro Define sx type linfor spectrum linfor spectrum define pro Read model data into ff structure linfor_rduio pro Recompute model on refined z grid linfor_regrid pro Compute ionopa quantities pe kappa zeta and monochromatic tau for 3D model linfor_ionopa_3d pro Construct 1D reference atmosphere from ff store in f1 linfor_refatm pro Compute ionopa quantities pe kappa zeta and monochromatic tau for 1D reference at mosphere linfor_ionopa_3d Do radiative transfer calculations for 3D model linfor_dort pro Do radiative transfer calculations for averaged 3D atmosphere linfor_dort pro Store results for later evaluation linfor_eval ss s1 nf kl Make Plots of line profiles and bisectors linfor_plot1 pro Do radiative transfer calculations for 1D reference atmosphere linfor_dort pro Store results for later evaluation linfor_evalx pro Create postcript file s linfor_plot2 pro Generate output files linfor_3D_1 idlsave linfor_3D_2 idlsave Free pointers to structures ff f1 fx ss s1 and sx if free flag 1 see Sect 5 1 linfor_flowfield_free pro 18 4 PROGRAM FILES AND DATA INPUT FILES 4 2 Structures in Common Block linfordata Table 1 shows a list of the structrues in common block linfordata used
44. ing nuclei bound in diatomic molecules Number density of nuclei of element 4 not bound in diatomic molecules Number density of neutral nuclei of element i not bound in diatomic molecules Total number density of diatomic molecules made up of one nucleus of element 7 and one nucleus of element k Total electron number density Total number density of nuclei of all elements diatomic molecules counting as 2 nuclei Elektron pressure Fractional abundance of element i constant Fractional abundance of free nuclei of element 2 x fi const for elements not involved in molecule formation i 44 41 For elements forming molecules i imoi v is the variable to be iterated Fractional abundance of molecule i k Fractional abundance of free electrons iterated quantity 9 1 2 Equations The Saha equation provides the relation between Nao and Nj Nio Sio Ni 104 52 9 IONDIS where the Saha factor S depends on temperature and electron pressure and on the ionization potentials and the partition functions of the different ionization stages Molecule partial pressures are given by the relation P Pr Kik Pik 105 where K x is the dissociation constant for the neutral diatomic molecule ik composed of one nucleus of elements 7 and k each P and Py are the partial pressures of the neutral atoms of elements 7 and k respectively Since P N kT molecule densities are kT Nio Nko Nin eo 106
45. intlime u aa aa wam wak RR a a ROG TRA 35 5 8 29 mebunk eed ra ad Coe 35 ee Be ana sne re Ae EO Eee rn 36 6 Line Data File line dat 38 Poa te Ai a Cre 38 6 11 clami es a s ee ok ok ee 9 mR ee en A 38 Tr 38 6 2 Line Data Formats a a lll ss 38 6 2 1 Continuum only 38 6 2 2 Single line calculations line data format 0 39 6 2 3 Single line calculations line data format 1 40 6 2 4 Single line calculations complete line data format 21 42 4 CONTENTS en 43 re 44 fa bbe TOR A a Bh ds Godot amp age de de hed 44 etd dated EN 45 6 3 1 Quadratic Stark effect 2 ee 45 6 3 2 Van der Waals broadening 22 22 En nn nn 46 6 3 3 Natural line broadening on nn nn 47 48 ee et R GR Ge ee ee ee la 48 2 boreshortening a e maisos ok a x A RUE a eee be eae s 49 50 9 IONDIS 51 0 1 Molecules sarea 2 2 4 8 oom aa ce en oko Wo et RO ee 4 51 9 1 1 Some definitionsl eee 51 we ee oh que wae Rae hc ence Row ROS RUN Sa d S Pe EU eos 51 wok bdo ee Bea RE Sa 6 Rok Ee eR RR c 54 9 1 4 Initial guess eee 54 9 1 5 Variable names LIST OF FIGURES 5 List of Figures 1 Analytical curve o growth eee 14 List of Tables fe hae we A AR OS 18 2 List of additional IDL modules gt o sa srete nn 18 3 List of all IDL modules 2 2 oo nn 19 PRIMERA 51 5 Large molecular network 10 atoms 14 molecules ls 51 6 2 GETTING STARTED
46. ion could be defined as GP o et N m m exp r8 mo 1 14 B pta an 80 H Ko x y such that the absolute line depression at any wavelength in the line profile is b Tb 2 Dita estu do f OP s ud X d sl f In 10 ro 25 eaa 51 Note however that P does not have the desired physical meaning because the factor 1 1 8 exp 7 i becomes negative when T is small rf is the optical depth due to the line opacity only and ii it is non zero also in layers where the line opacity vanishes For this reason P is not considered useful and hence is not computed in the current version of LINFOR3D A much better way to define the Line Depression Contribution Function is to consider Eq 39 and to define it as mn E nimi n B So epla 52 ay 12 3 BASIC EQUATIONS OF RADIATIVE TRANSFER Note that this contribution function vanisches whereever the line opacity 7 8 is zero For the flux spectrum we define as before 27 1 CP CP ron d aul ao 53 Then the absolute line depression at any wavelength in the line profile is Tb zb Dr ro 0 H 0 A n CP Th H A dry j In 10 Tol zp CP To 2b H 0 A dz gt 54 and b Tb yA Drie n CP Ady n In 10 eh CP rol 3 d 24 55 for the intensity and flux spectrum respectively TheEquivalent Width Contribution Function is computed as CH 10 158 CP m NAN 56 and 1 To W p
47. lam ddlam 1 1 Fe I 5500 A 0 00 eV 1 0 0000 2600 0 000 5500 0 6 000 1 0 10 0 5 5D 1 5 5D 3 clam gfscale 2000 0 1 0 Description of entries Row 1 Header identifies the meaning of the columns for data in row 5 Row 2 Two integers kline and ktotal kline number of line calculations requested in this file ktotal is the total number of spectral lines including blends in this case kline 1 ktotal 1 Row 3 String identifier of the first line calculation Row 4 Integer nbl integer array incode nbl nb number of blend components for this line calculation 1 incode integer array identifying the input format for each of the blend components 0 Row 5 Line data in format 0 7 2 columns C1 Multiplet number for information only C2 Identifier of atom or ion e g 2601 mean FelI C3 Excitation potential of lower level in eV C4 Central wavelength of blend component C5 loggf value of blend component C6 Alog Cg Enhancement factor for van der Waals line broadening C7 Ar a Difference of mean square electron orbital radii C8 AA A Line profile is computed from Ag AA to Ag AA C9 5A A Spacing of wavelength points for spectrum synthesis C10 Wo mA total equivalent width of this blend see below Row 6 Description for data in row 6 Row 7 clam and gfscale see Sect 6 1 In this case the Stark broadening due to collisions with electrons is neglected C4 0 Radiative damping yraa is treate
48. line data format 1 For a single unblended line the this form of the line dat file looks like this Example Mult namj ei alam gflg dlgC6 lu diu lo dio dlam ddlam i 1 O I ApJ Line 2 92 6300 30 0 000 9 773 i 1 92 800 0 000 6300 30 9 773 1 0 1 0 0 2 0 0 4 D 1 4 D 3 clam gfscale 2000 0 1 0 Description of entries Row 1 Header identifies the meaning of the columns for data in row 5 Row 2 Two integers kline and ktotal 6 2 Line Data Formats 41 kline number of line calculations requested in this file ktotal is the total number of spectral lines including blends in this case kline 1 ktotal 1 Row 3 String identifier of the first line calculation Row 4 Integer nbl integer array incode nbl nb number of blend components for this line calculation 1 incode integer array identifying the input format for each of the blend components 1 Row 5 Line data in format 1 10 2 columns C1 Multiplet number for information only C2 Identifier of atom or ion e g 2601 mean Fell C3 Excitation potential of lower level in eV C4 Central wavelength of blend component C5 log gf value of blend component C6 Alog Cg Enhancement factor for van der Waals line broadening C7 LU Orbital quantum number of valence electron of lower level C8 DIU excitation energy eV of parent term for lower level C9 LO Orbital quantum number of valence electron of upper level C10 DIO excitation energy eV of parent term
49. locity km s for 3D mean models additional line broadening after line formation always float e g 1 6 5 8 18 ximac3 function required type values Isotropic Gaussian macroturbulence velocity km s for 2D 3D models additional line broadening after line formation always float e g 1 6 5 8 19 vfacx function required type values the x component of the hydrodynamical velocity field of the 2D 3D models is multiplied by this factor always float e g 0 0 1 0 5 8 20 vfacy function required type values the y component of the hydrodynamical velocity field of the 2D 3D models is multiplied by this factor always float e g 0 0 1 0 5 8 21 vfacz function required type values the z component of the hydrodynamical velocity field of the 2D 3D models is multiplied by this factor always float e g 0 0 1 0 5 8 22 micro function required type values controls microturbulence in 1D curve of growth always integer 0 1 Determines whether or not different microturbulence values should be used when computing the 1D curve of growth 0 only one value given by ximicx and ximici respectively 1 sequence 34 5 PARAMETER INPUT LINFOR_SETCMD PRO of microturbulence values defined by parameters xi_a xi_b xi_d see below 5 8 23 xia function determines
50. mergent intensity can also be obtained from Eq 8 R 0 Siti eoe PO pt ar A Now we define a fixed central wavelength Ao with the corresponding fixed universal optical depth scale 79 which is equidistant in log To and may used alternatively for all integrations On this optical depth scale Eq 4 becomes b Kim DSC expt rl 6 47 IKE expt xS r8 7 8 10 To Ko giving the continuum intensity at wavelength A as a function of optical depth 7 Note the factor 5 Kg under the integral The intensity at the lower boundary T amp 78 can be computed from the diffusion approximation ko ASS ro Sx T0 0 e ar T0 11 but the boundary term may also be neglected at least for the emergent intensity because the exponential factor is usually very small For the emergent intensity we have from Eq 5 b C To ES c Cc c c Dro 0 f ne SAO exp 7 60 dro IX 79 exp 7h 79 12 0 8 3 BASIC EQUATIONS OF RADIATIVE TRANSFER Similarly Eq 8 becomes To d S c C c c Cc en 79 exp 7 1 TX To dro us ro expi r r0 75 70 13 uj ro TO Note the absence of the factor x r under the integral in this case u5 7 is obtained from the diffusion approximation C C fio een KS d To 79 14 b u To The emergent intensity can be computed from Eq 13 as I r 0 f EED rana 0D ont 15 In the latest version of Linfo
51. r3D the continuum intensity is calculated from Eqs 8 and 9 at 3 different wavelengths Ag AA Ao and Ay AA where AA is specified by the parameter dclam We ensure that the derivative d 5 d ro fulfills the condition mas 7 1 dr Si ra Sir 16 Ti TA The reason for using Eqs 8 and 9 instead of Eq 5 is that the quantity u r is needed to compute the line depression source function see Sect 3 3 We have checked that the usual trans fer equation Eq 5 gives numerically very closely the same results for the emergent continuum intensity as Eq 9 3 2 Transfer equation for the line intensity In the presence of lines the transfer equation at wavelength A reads dl C C C s gt DARA KS Su 17 The line source functions St may be different from the LTE continuum source function SY With the definition of the total optical depth an eX dsedr dr 18 g and the total source function Bic Ki SS Di SS 14 P se 19 KS e KS KHP Ly ds where jc T gt BILDEN _ Le Zum B dee ky Sx 20 d Xie K KS KSI C we can write i dI In LTE 5 S The solution of Eq 21 is Tb I r 0 f S eva ICH ext or 22 3 8 Transfer equation for the line depression 9 In analogy to Eq 12 we can also obtain the emergent line intensity by integration on the universal optical depth scale 79 ls 9 f A Len Sn exon Il exon 23 or substituting 4 from Eq
52. rcs_idmean3xRT3 opta GASFILE gas_cifist2006_m30_a04_15 eos EOSFILE eos_cifist2006_m30_a04_15 eos with the above theoretical results yields LINFOR3D 3 1 3 ratio ntheta numerical analytical 1 2 3 3 4 I linfor3D I Eq 65 1 0005573 1 0005573 1 0005573 1 0005573 1 0005573 Br 1 0004105 1 0148776 1 0079553 1 0004507 1 0050481 If the ratio y of line opacity kp and continuum opacity Ke is constant with optical depth n kefke the intensity in the line is simply o 1 3 u Llu T do E 67 the absolute line depression is c 3 n Di u Te u Le m Tos i I 68 and the relative line depression at disk center is 3 7 Di p U E n l Ell 69 The absolute line depression for flux is x 1 7 De Fi Fe 2e p Di u du a rn 70 and the relative line depression for flux is L 7 Dp F 1 n 21 n 71 The ratio between the relative line depression in flux and at disk center is therfore 5 6 and the same ratio holds for the equivalent widths 14 3 BASIC EQUATIONS OF RADIATIVE TRANSFER The local absorption line profile is now defined by n a v No H a v gt 72 where v A Ao AAp and a Aly 2 AAp AAp Doppler width AAy full width at half maximum of the Lorentzian damping profile The Voigt function H a v is normalized such that fora 1 H a v 0 1 Assuming that no a and AAp are constant we can compute the em
53. re ximici 1 00 microturbulence km s 1D AVG atmosphere same 0 00 microturbulence km s 3D RHD atmosphere ximacx 1 60 macroturbulence km s 1D REF atmosphere ximaci 1 60 macroturbulence km s 1D AVG atmosphere ximac3 0 00 macroturbulence km s 3D RHD atmosphere vfacx 1 00 fudge factor for 3D x velocity vfacy 1 00 fudge factor for 3D y velocity vfacz 1 00 fudge factor for 3D z velocity micro 0 compute microturbulence sequence 0 1 xi_a 0 0 xi_b 2 0 xi_d 0 1 start stop delta of micro sequence dclam 0 0 variation of continuum from clam dclam clam dclam A intmode 1 integration mode linfor_msint intline 1 line integration depth 1 2 I 1 2 end 38 6 LINE DATA FILE LINE DAT 6 Line Data File line dat There are several different formats for historical reasons to specify line data which are described in Sect Note that all formats were extended in version 1 5 0 and now do have to contain the two lines clam gfscale 2000 0 1 0 at the end These parameters are explained in Sect Some helpful remarks concerning the conversion of line broadening parameters are given in Sect 6 1 Parameters in Line Data File 6 1 1 clam function continuum wavelength in A also center of wavelength window required always type float values e g 2000 0 clam defines the wavelength where the continuum opacities are compu
54. re model information all read from parameter file for CO BOLD data 27 5 4 2 gasfile function name of GAS file P T gt p e required not needed if context cobold type string values e g gas mm00 1 eos 5 4 3 eosfile function name of EOS file p e gt P T required not needed if context cobold type string values e g eos_mm00_1 eos 5 4 4 htau0 function opacity scale height cm at top of 3D model required not needed if context cobold type float values e g 60 055 5 4 5 qmol function mean molecular weight of neutral gas required not needed if context cobold type float values e g 1 301855 5 4 6 Teff function effective temperature of 3D model required not needed if context cobold type float values e g 5770 0 5 4 7 grav function surface gravity cm s of 3D model required not needed if context cobold type float values e g 27500 0 5 4 8 tsurffac function surface temperature T 0 of 3D model is tsurffac Ter required not needed if context cobold type float values e g 0 727903 28 5 PARAMETER INPUT LINFOR_SETCMD PRO 5 5 Model data reading of full files CO BOLD only The parameters in this section are only needed for reading snapshot from CO BOLD data files 5 5 1 isnap_full_1 function first snapshot to be read from full file
55. rent excitation potential Mult namj chik alam gflg dlgC6 drrcai dlam ddlam WO 8 8 N I Fictitious Line 1 0 000 5500 0 7 6914 1 00 10 00 75 00 1 0 9999 700 0 000 5500 0 7 6914 1 00 10 00 3 00E 01 3 00E 03 75 00 N I Fictitious Line 2 2 000 5500 0 5 7282 1 00 10 00 75 00 1 0 9999 700 2 000 5500 0 5 7282 1 00 10 00 3 00E 01 3 00E 03 75 00 N I Fictitious Line 3 4 000 5500 0 3 8298 1 00 10 00 75 00 T Q 9999 700 4 000 5500 0 3 8298 1 00 10 00 3 00E 01 3 00E 03 75 00 N I Fictitious Line 4 6 000 5500 0 1 9876 1 00 10 00 75 00 1 0 9999 700 6 000 5500 9876 1 00 10 00 3 00E 01 3 00E 03 75 00 N I Fictitious Line 5 8 000 5500 0 0 1961 1 00 10 00 75 00 1 0 9999 700 8 000 5500 1961 1 00 10 00 3 00E 01 3 00E 03 75 00 N I Fictitious Line 6 10 000 5500 0 1 5485 1 00 10 00 75 00 eL Q 9999 700 10 000 5500 5485 1 00 10 00 3 00E 01 3 00E 03 75 00 N I Fictitious Line 7 11 000 5500 0 2 4046 1 00 10 00 75 00 l0 9999 700 11 000 5500 4046 1 00 10 00 3 00E 01 3 00E 03 75 00 N I Fictitious Line 8 12 000 5500 0 3 2510 1 00 10 00 75 00 st 0 9999 700 12 000 5500 0 3 2510 1 00 10 00 3 00E 01 3 00E 03 75 00 clam gfscale 2000 0 1 0 Sd l j o gt o N Note that now kline 8 and ktotal 8 since all lines have one blend component only 6 3 Conversion of line broadening parameters The line broadening can be specified in different ways e g as logC4 for quadrati
56. specified via the difference of mean square electron orbital radii Ar az where ag is the Bohr radius log Ar ag log Cs 32 3867 101 We finally derive a conversion formula 20 877 2 5 log xe Ar ag 10 102 The exemplary value of 7 619 from VALD thus gives 67 437 for the parameter drrcal In addition d1gC6 should be set to 0 unless you want to apply an additional enhancement of the broadening 6 3 3 Natural line broadening The broadening parameter Jrag can be converted like this Crag LE DN 103 For instance log Yrag 7 877 would give Crag 0 753 The parameter Crad is thus set to 0 753 and dlggr is set to 0 0 48 7 Output files 7 OUTPUT FILES LINFOR3D generates the following output files in the LINFOR3D directory name linfor_3D_1 idlsave content IDL structures cmd const line result linfor_3D_2 idlsave IDL structure maps see below for more details linfor_timing txt Timing statistics see Sect linfor_3D_1 ps Postscript file local line profiles plus average linfor_3D_2 ps Postscript file line profiles for 1D reference atmosphere and time averaged 1D and 3D spectra granulation abundance correction 7 1 Files containing intensity maps The output file linfor_3D_2 idlsave contains a structure MAPS An example of this structure is Structure lt 83027f4 gt 11 tags length 11289820 data length 11289820 refs 1 NX N
57. start value for microturbulence sequence required always type float values e g 0 0 default 0 5 5 8 24 xi_b function determines end value for microturbulence sequence required always type float values eg 2 0 default 1 5 5 8 25 xid function determines intervals of microturbulence sequence required always type float values e g 0 1 default 0 125 The microturbulence sequence is computed as xi i xi0 xi_a i xi d i20 im where xi0 is ximicx and ximici respectively and im xi b xi a xi d 5 8 26 dclam function determines the variation of the continuum required always type float values e g 20 0 default 0 if dclam 0 the continuum is treated as constant default Otherwise the continuum is computed at 3 wavelength points clam dclam clam clam dclam where clam is the central wavelength in A of the computed spectral range see Sect 6 and dclam is half the width of the specified spectral range in A The continuum is computed by parabolic interpolation inside the spectral window If the spectral range of the specified synthetic spectrum which is defined by the parameters of the line file see Sect 6 exceeds a few A dclam should be set to match half the total spectral range 5 8 27 intmode function mode of integration in routines ms int tau and ms int exp required always type integer values 0 1 Determines the mode o
58. ted and also defines the cen ter of the window for which spectrum synthesis is done The window extends from A clam dlam to A clam dlam depending on the value of dlam specified for the particular line From Version 3 1 2 a negative clam indicates that the continuum source function is to be set to the wavelenth integrated Planck Function S c T 7 and the continuum opacity is set to the Rosseland mean opacity Kcont KRoss 6 1 2 gfscale function global scaling factor for oscillator strengths required always type float values eg 1 0 Note The value 1 has no effect 6 2 Line Data Formats 6 2 1 Continuum only It is possible to do pure continuum calculations In this case the line dat file looks like this Example Some text header i 1 Continuum 2000 A 4 clam gfscale 2000 0 1 0 Description of entries Row 1 Header identifies the meaning of the columns for data in row 5 Row 2 Two integers kline and ktotal both of them must be 1 6 2 Line Data Formats 39 Row 3 String identifier of the continuum calculation Row 4 Two integers nbl 1 incode 1 Row 5 Description for data in row 6 Row 6 clam and gfscale see Sect 6 1 All the line parameters remain undefined 6 2 2 Single line calculations line data format 0 For a single unblended line the simplest form of the line dat file looks like this Example Mult namj ei alam gflg dlgC6 drrcai d
59. ted for inclination angle muO cos 0o mu0 1 0 corresponds to vertical rays i e disk center view mu0 0 0 corresponds to the very limb but a value of mu0 0 0 will clearly not work 5 8 11 kphi function view angle required always type integer values 2 132 3 The parameter kphi determines the direction from which the model is viewed value meaning 0 rays emerge parallel to the x axis i e the model is viewed somewhere on the equator between the left limb and disk center 1 rays emerge parallel to the y axis i e the model is viewed somewhere on the meridian between the lower limb and disk center 2 rays emerge anti parallel to the x axis i e the model is viewed somewhere on the equator between the right limb and disk center 3 rays emerge anti parallel to the y axis i e the model is viewed somewhere on the meridian between the upper limb and disk center Other integer values of kphi are allowed but give no new results increasing kphi by one increases phi by 7 2 5 8 12 Hbrd function controls broadening of hydrogen lines required always type integer values 0 1 2 3 32 5 PARAMETER INPUT LINFOR_SETCMD PRO value meaning 0 Cayrel Traving 1960 default 1 Resonance broadening AG Stark boradening G 2 Resonance broadening BPO Stark boradening G 3 Resonance broadening A08 Stark boradening G AG Al
60. the z component of the velocity field is adjusted to ensure zero mass flux in z direction 0 no 1 yes Default 0 5 1 5 plt flag function required type values plotting of bisectors always integer 1 0 1 The parameter plt flag controls if line bisctors should be plotted or not 0 no 1 yes If plt flag is set to 1 all plotting is suppressed 22 5 PARAMETER INPUT LINFOR_SETCMD PRO 5 1 6 maps flag function controls output of intensity maps required always type integer values s 41 2 The parameter maps flag controls the output of intensity maps which are provided in the IDL structure MAPS value meaning 0 Continuum images only Create map ICLAMO 1 Continuum images ICLAMO plus images at the centre of the wavelength window ICLAM1 all at wavelength A clan 2 Continuum images ICLAMO plus images ICLAM2 at all wavelengths within the wavelength window of width 2 dlam around the central wavelength clam A clam dlam i ddlam see Sect 6 5 1 7 cc3d flag function output of 3D contribution function required always type integer values 0 1 The parameter cc3d flag controls whether the 3D continuum intensity contribution function should be saved in structure contf3d or not 0 no 1 yes 5 1 8 nlte flag function output of 3D contribution function required always type integer values 0 1 2 3
61. tructure linfor_atom pro S Defines atomic data linfor setwts pro S Defines weights for angle quadrature linfor setcmd pro S Command file parameter input linfor_rdxatm pro S Reads 1D reference atmosphere calling linfor_rdatmos or linfor rdatlas9 linfor rdatlas9 pro S Reads ATLAS9 1D atmosphere atm dat linfor rdatmos pro S Reads ATMOS 1D atmosphere atm dat linfor_rdf15 pro S Reads a sequence of FOR15 snapshots from 2D Kiel hydro simulations FOR15 linfor_rdsav pro S Reads 3D snapshot from Copenhagen code savfs linfor_rduio pro S Reads 3D snapshot from CO BOLD uio output files linfor rdvog pro S Reads 3D snapshot from Voegler MHD code linfor findff pro S Finds cached flow fields linfor rdline pro S Reads line data line dat linfor init pro S Initializes ionopa EOS Opacities several constants linfor bisector pro S Computes line bisector positions called by linfor_plot1 and linfor_plot2 linfor_convol pro S Convolves line profile with Gauss kernel called by linfor_plot1 and linfor_plot2 linfor_dort pro S Computes spectrum from flow field main RT module calling several lower level routines linfor eval pro S Evaluates mean spectrum abundance corrections linfor evalx pro S Evaluates reference spectrum abundance corrections linfor incline pro S Inclines 3D flow field called by linfor_ztau linfor_ionopa_3d pro S Calculates electron pressure ionization fractions and monochromatic optical depth for given flow field linfor rad3
Download Pdf Manuals
Related Search