|
Size: 2654
Comment:
|
Size: 2135
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 1: | Line 1: |
| ## Please edit system and help pages ONLY in the master wiki! ## For more information, please see MoinMoin:MoinDev/Translation. ## IMPORTANT NOTE: ## When you use this page as a template for creating your project page: ## * please remove all lines starting with two hashes (##) ## * except the acl line, please keep that, but remove one hash, so it reads #acl ... ## * fix the acl line so it has the correct page instead of the sample Project/...Group ##acl Project/AdminGroup:admin,read,write,delete,revert Project/ReadWriteGroup:read,write Project/ReadGroup:read ##master-page:Unknown-Page ##master-date:Unknown-Date |
#acl ValdTeamGroup:read,write,delete All:read |
| Line 13: | Line 4: |
| VALD3 stores vacuum wavelengths of all the transitions. This allows uniform handling of extraction across the whole spectral range. On the other hand the selection tools in VALD3 include options for returning the air wavelengths. Furthermore, some of the original lines list include wavelengths measured in the air. This calls for conversion tools. Such tools must be uniformly accurate and reversible across the whole spectral range. | = Wavelength conversion air <-> vacuum = VALD3 stores vacuum wavelengths of all the transitions. This allows uniform handling of extraction across the whole spectral range. On the other hand, the selection tools in VALD3 include options for returning the air wavelengths. Furthermore, some of the original line lists include wavelengths measured in the air. This calls for conversion tools. Such tools must be uniformly accurate and reversible across the whole spectral range. |
| Line 19: | Line 13: |
| The conversion is than: λ,,air,, = λ,,vac,, / n. | The conversion is then: λ,,air,, = λ,,vac,, / n. |
| Line 27: | Line 21: |
| [[air2vac.gif|Here]] is the comparison of the Morton and my inverse transformation between 2000 Å and 100000 Å | [[attachment:air2vac.gif|Here]] is the comparison of the Morton and my inverse transformation between 2000 Å and 100000 Å. |
| Line 29: | Line 23: |
| The conversions are implemented as a FORTRAN90 module functions air2vac and vac2air that take single scalar or vector parameters of type REAL*4 or REAL*8 and return the result of the same type. The module name is AIR_VAC. | The conversions are implemented as a FORTRAN90 module in the functions air2vac and vac2air that take single scalar or vector parameters of type REAL*4 or REAL*8 and return the result of the same type. The module name is AIR_VAC. |
Wavelength conversion air <-> vacuum
VALD3 stores vacuum wavelengths of all the transitions. This allows uniform handling of extraction across the whole spectral range. On the other hand, the selection tools in VALD3 include options for returning the air wavelengths. Furthermore, some of the original line lists include wavelengths measured in the air. This calls for conversion tools. Such tools must be uniformly accurate and reversible across the whole spectral range.
For the vacuum to air conversion we used the formula from Donald Morton (2000, ApJ. Suppl., 130, 403) for the refraction index which is also the IAU standard:
n = 1 + 0.0000834254 + 0.02406147 / (130 - s2) + 0.00015998 / (38.9 - s2), where s = 104 / λvac and λvac is in Ångströms.
The conversion is then: λair = λvac / n.
This formula comes from Birch and Downs (1994, Metrologia, 31, 315) and applies to dry air at 1 atm pressure and 15ºC with 0.045% CO2 by volume. The corrections to Edlén (1953, J. Opt. Soc. Am., 43, 339) are less than 0.0001 Å at 2000 Å and less than 0.001 Å at 30000 Å.
The opposite conversion (air-to-vacuum) is less trivial because n depends on λvac and I did not find readily available formulas with sufficient precision. So I did the fitting and came up with the following solution:
n = 1 + 0.00008336624212083D + 0.02408926869968 / (130.1065924522 - s2) + 0.0001599740894897 / (38.92568793293 - s2), where s = 104 / λair and the conversion is: λvac = λair * n.
Here is the comparison of the Morton and my inverse transformation between 2000 Å and 100000 Å.
The conversions are implemented as a FORTRAN90 module in the functions air2vac and vac2air that take single scalar or vector parameters of type REAL*4 or REAL*8 and return the result of the same type. The module name is AIR_VAC.
The corresponding IDL functions are also available.