NLTE grids#
Non Local Thermal Equilibrium (NLTE) calculations are important to accuarately fit certain lines. PySME supports them using pre-computed grids of NLTE departure coefficients, which need to be created for every element. For common elements PySME provides grids (see below) via the LFS (see Large File Server). If any of these grids are used, please kindly take care to cite the papers describing the NLTE models and departure coefficient calculations.
NLTE calculations need to be specified for each element they are supposed to be used for individually using sme.nlte.set_nlte(el, grid) (the grid can be omitted if there is a grid in lfs).
Similarly they can be disabled for each element using sme.nlte.remove_nlte(el), where sme is your SME structure.
If no element is set to NLTE in the structure PySME will perform
LTE calculations only.
Fields of NLTE object#
elements: The elements for which NLTE has been activatedgrids: The grid file that is used for each active elementsubgrid_size: A small segment of the NLTE grid will be cached in memory to speed up calculations. This sets the size of that cache by defining the number of points in each axis (rabund, teff, logg, monh).flags: After the synthesis all lines are flaged if they used NLTE
Grid interpolation#
The grid has 6 dimensions.
teff: Effective Temperature
logg: Surface Gravity
monh: Overall Metallicity
rabund: relative abundance of that element
depth: optical depth in the atmosphere
departure coefficients: The NLTE departure coefficients describing how much it varies from the LTE calculation
We then perform linear interpolation to the stellar parameters we want to model. And we then perform a cubic spline fit to the depth scale of the model atmosphere we specified (See Atmosphere).
We then use the linelist to find only the relevant transitions in the grid, and pass the departure coefficients for each line to the C library.
NLTE flags in line list#
PySME provides information on whehter a line is synthesized in NLTE through the nlte_flag column in the line list.
1: this line was synthesized with in NLTE.0: this line was synthesized in LTE.-1: this line was not included in the current synthesis pass (see Line Filtering).
Recommended and default grids#
Element |
Grid name |
Zenodo version |
Citation |
|---|---|---|---|
H |
|
3 |
|
Li |
|
3 |
|
C |
|
2 |
|
N |
|
3 |
|
O |
|
3 |
|
Na |
|
3 |
|
Mg |
|
3 |
|
Al |
|
3 |
|
Si |
|
3 |
|
S |
|
7 |
|
K |
|
3 |
|
Ca |
|
3 |
|
Ti |
|
5 |
|
Mn |
|
3 |
|
Fe |
|
4 |
|
Cu |
|
6 |
|
Ba |
|
3 |
All recommended/default grids are available in Zenodo.
Deprecated grids#
H
marcs2012_H2018.grd
Li
marcs2012_Li.grd (Lind et al. 2009)
marcs2012_Li2009.grd (Lind et al. 2009)
nlte_Li_multi.grd
C
marcs2012_C.grd
N
marcs2012_N.grd
O
marcs2012p_t1.0_O.grd (Sitnova et al. 2013)
marcs2012_O2015.grd (Amarsi et al. 2016b)
marcs2012_O.grd
Na
marcs2012p_t1.0_Na.grd (Mashonkina et al. 2001)
marcs2012_Na.grd (Lind et al. 2011)
marcs2012_Na2011.grd (Lind et al. 2011)
nlte_Na_multi_full.grd
nlte_Na_multi_sun.grd
Al
marcs2012_Al.grd
marcs2012_Al2017.grd
Mg
marcs2012_Mg2016.grd (Osorio et al. 2016)
marcs2012_Mg.grd
Si
marcs2012_Si2016.grd (Amarsi & Asplund 2017)
marcs2012_Si.grd
K
marcs2012_K.grd
Ca
marcs2012s_t2.0_Ca.grd (Mashonkina et al. 2007)
marcs2012p_t1.0_Ca.grd (Mashonkina et al. 2017)
marcs2012_Ca.grd
Ti
marcs2012s_t2.0_Ti.grd (Sitnova et al. 2016)
Mn
marcs2012_Mn.grd
Fe
marcs2012_Fe2016.grd (Amarsi et al. 2016a)
marcs2012s_t2.0_Fe.grd (Mashonkina et al. 2011)
nlte_Fe_multi_full.grd
marcs2012_Fe.grd
Ba
marcs2012p_t1.0_Ba.grd (Mashonkina et al. 1999)
Eu
nlte_Eu.grd