|
Size: 4803
Comment:
|
← Revision 25 as of 2025-09-23 13:13:33 ⇥
Size: 7417
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 3: | Line 3: |
| VALD-3 binary record format (in February 2011 we decided to expand the level designation fields to 86 characters extending the total record length to 270 bytes): | = VALD Team internal pages = <<Include(FrontPageValdTeam, "Navigation Inside VALD", titlesonly)>> == VALD-3 binary record format == Total record length: 270 bytes |
| Line 7: | Line 12: |
| || species || int*4 || || || VALD species code, see ~vald/data/VALD_list_of_species || | || species || int*4 || || || VALD species code, see [[VALD list of species]] || |
| Line 16: | Line 21: |
| || gamst || real*4 || (s*Ne)^-1|| 0 || log Quadratic Stark damping constant at l0000 K (per perturber) || || gamvw || real*4 || (s*NH)^-1|| 0 || log van der Waals damping constant at l0000 K (per perturber) || |
|| gamst || real*4 || (s*Ne)^-1|| 0 || log Quadratic Stark damping constant at l0000 K (where Ne is the electron number density) || || gamvw || real*4 || depends on value || 0 || van der Waals damping parameter: if <0: log gamma_6 in units of (s*NH)^-1 at l0000 K (where NH is the number density of neutral hydrogen); if >0 : two parameters as defined in Barklem et al. 2000A&AS..142..467B : integer part is cross-section sigma, fractional part is velocity parameter alpha || |
| Line 22: | Line 27: |
| || source || char*7 || || blank || reference to the oscillator strength source || || accuracy_flag || char*1 || || blank || type of the accuracy value (see below) || || accuracy || char*6 || || blank || accuracy of the oscillator strength || |
|| source || char*7 || || blank || reference to the oscillator strength source^1^ || || accuracy_flag || char*1 || || blank || type of the accuracy value (see below)^2^ || || accuracy || char*6 || || blank || accuracy of the oscillator strength^2^ || |
| Line 30: | Line 35: |
| || || || || E || Forbidden: Magnetic octupole transition (M3)|| | || || || || E || Forbidden: Electric octupole transition (E3)|| || || || || F || Forbidden: Magnetic octupole transition (M3)|| |
| Line 47: | Line 53: |
| accuracy_flag values:<<BR>> ' ' ... no value<<BR>> |
1. Special SOURCE filed content. To store multiple references to the wavelengths, oscillator strengths and/or isotopic splitting a special form of the SOURCE field is used. The first byte is the mask. An indicator of special form is that the first byte has a value less than 48 (decimal value of character '0').<<BR>> The least significant bits of the first byte are treated is indicators for additional references as following: 00033221. Bit "1" set to 1 means that that the following 2 bytes contain an integer number of the VALD reference that is assigned in the VALD_ref.bib file. In this case it is the reference for the oscillator strength. This reference comes first. The flags for two other (optional) references "11" and "22" can have values of 0 (absent), 1 (2nd reference for gf), 2 (reference for isotopic shift), or 3 (another reference for the wavelength, which is also assumed to be applicable for the level energies). This is why two bits are required for 2nd and 3rd flags. The following 3x2-byte sequences are interpreted as integer numbers (unless the corresponding flag is 0) matching the VALD references as explained above. Typical combinations are: gf+iso+wl, gf+gf+iso, gf+wl+wl, gf+iso+iso. Multiple references usually means averaged measurements of wavelengths or isotopic shifts. 2. Accuracy_flag values:<<BR>> ' ' or '_' ... no value<<BR>> |
| Line 50: | Line 59: |
| 'E' ... error in dex<<BR>> | An estimated accuracy is listed for each transition strength, indicated by a code letter as given in the table below: <<BR>> AAA ≤ 0.3%<<BR>> AA ≤ 1%<<BR>> A+ ≤ 2%<<BR>> A ≤ 3%<<BR>> B+ ≤ 7%<<BR>> B ≤ 10%<<BR>> C+ ≤ 18%<<BR>> C ≤ 25%<<BR>> D+ ≤ 40%<<BR>> D ≤ 50%<<BR>> E > 50%.<<BR>> 'E' ... absolute error in dex<<BR>> |
| Line 52: | Line 75: |
| Cancellation factor (CF) is defined as follows. Calculations of the line transition probability are made by summation of contributions of eigenvector components of wave functions of the lower and upper states that come with different signs. All positive and negative contributions are summed separately in partial sums S+ and S−, so that the line strength is given by S = (S+) + (S−). Then CF is defined as CF = [(S+) − abs(S−)]/[(S+) + abs(S−)]. Thus, CF ends up in the range between [−1, 1], and “strong cancellation” occurs when |CF| is close to zero. <<BR>> |
VALD Team internal pages
VALD-3 binary record format
Total record length: 270 bytes
Field short name |
Type |
Unit |
Default |
Description |
wl |
real*8 |
Å |
|
wavelength in vacuum |
species |
int*4 |
|
|
VALD species code, see VALD list of species |
loggf |
real*4 |
|
|
log statistical weight times oscillator strength |
e_low |
real*8 |
cm^-1 |
|
energy of the lower level |
j_low |
real*4 |
|
|
orbital quantum number for the lower level |
e_upp |
real*8 |
cm^-1 |
|
energy of the upper level |
j_upp |
real*4 |
|
|
orbital quantum number for the upper level |
lande_low |
real*4 |
|
99 |
Lande factor for the lower level |
lande_upp |
real*4 |
|
99 |
Lande factor for the upper level |
gamrad |
real*4 |
s^-1 |
0 |
log Radiative damping constant |
gamst |
real*4 |
(s*Ne)^-1 |
0 |
log Quadratic Stark damping constant at l0000 K (where Ne is the electron number density) |
gamvw |
real*4 |
depends on value |
0 |
van der Waals damping parameter: if <0: log gamma_6 in units of (s*NH)^-1 at l0000 K (where NH is the number density of neutral hydrogen); if >0 : two parameters as defined in Barklem et al. 2000A&AS..142..467B : integer part is cross-section sigma, fractional part is velocity parameter alpha |
term_flag_low |
char*2 |
|
blank |
level coupling type (e.g. LS) |
char*86 |
|
blank |
lower level term designation |
|
term_flag_upp |
char*2 |
|
blank |
level coupling type (e.g. LS) |
char*86 |
|
blank |
upper level term designation |
|
source |
char*7 |
|
blank |
reference to the oscillator strength source1 |
accuracy_flag |
char*1 |
|
blank |
type of the accuracy value (see below)2 |
accuracy |
char*6 |
|
blank |
accuracy of the oscillator strength2 |
Transition type |
char*1 |
|
blank |
Allowed transition |
|
|
|
A |
Autoionizing transition |
|
|
|
B |
Forbidden: Electric quadrupole transition (E2) |
|
|
|
C |
Forbidden: Magnetic dipole transition (M1) |
|
|
|
D |
Forbidden: Magnetic quadrupole transition (M2) |
|
|
|
E |
Forbidden: Electric octupole transition (E3) |
|
|
|
F |
Forbidden: Magnetic octupole transition (M3) |
Extra information |
char*1 |
|
blank or 0 |
No extended information |
|
|
|
1 |
Extended van der Waals |
|
|
|
2 |
Extended Stark |
|
|
|
3 |
Extended van der Waals & Stark |
|
|
|
4 |
Transition has hfs extension |
|
|
|
5 |
Extended van der Waals & hfs |
|
|
|
6 |
Extended Stark & hfs |
|
|
|
7 |
Extended van der Waals & Stark + hfs |
Not assigned |
char*1 |
|
blank |
Not used |
Not assigned |
char*1 |
|
blank |
Not used |
comment |
char*16 |
|
blank |
additional comments (e.g. wavelength accuracy) |
Notes:
Fields without Default value are mandatory.
1. Special SOURCE filed content. To store multiple references to the wavelengths, oscillator strengths and/or isotopic splitting a special form of the SOURCE field is used. The first byte is the mask. An indicator of special form is that the first byte has a value less than 48 (decimal value of character '0').
The least significant bits of the first byte are treated is indicators for additional references as following: 00033221. Bit "1" set to 1 means that that the following 2 bytes contain an integer number of the VALD reference that is assigned in the VALD_ref.bib file. In this case it is the reference for the oscillator strength. This reference comes first. The flags for two other (optional) references "11" and "22" can have values of 0 (absent), 1 (2nd reference for gf), 2 (reference for isotopic shift), or 3 (another reference for the wavelength, which is also assumed to be applicable for the level energies). This is why two bits are required for 2nd and 3rd flags. The following 3x2-byte sequences are interpreted as integer numbers (unless the corresponding flag is 0) matching the VALD references as explained above. Typical combinations are: gf+iso+wl, gf+gf+iso, gf+wl+wl, gf+iso+iso. Multiple references usually means averaged measurements of wavelengths or isotopic shifts.
2. Accuracy_flag values:
' ' or '_' ... no value
'N' ... quality class (NIST)
- An estimated accuracy is listed for each transition strength, indicated by a code letter as given in the table below:
AAA ≤ 0.3%
AA ≤ 1%
A+ ≤ 2%
A ≤ 3%
B+ ≤ 7%
B ≤ 10%
C+ ≤ 18%
C ≤ 25%
D+ ≤ 40%
D ≤ 50%
E > 50%.
'E' ... absolute error in dex
'C' ... cancellation factor
Cancellation factor (CF) is defined as follows. Calculations of the line transition probability are made by summation of contributions of eigenvector components of wave functions of the lower and upper states that come with different signs. All positive and negative contributions are summed separately in partial sums S+ and S−, so that the line strength is given by S = (S+) + (S−). Then CF is defined as CF = [(S+) − abs(S−)]/[(S+) + abs(S−)]. Thus, CF ends up in the range between [−1, 1], and “strong cancellation” occurs when |CF| is close to zero.
'P' ... predicted line