How to Choose a Model

The dust_extinction package provides a suite of dust extinction models. Which model to use can depend on the wavelength range of interest, the expected type of extinction, or some other property.

After initialization, all models return the A(x)/A(V) extinction values for the input x values, except for the FM90 shape model that returns E(x-V)/(E(B-V)) extinction values. The input x should be wavelengths/frequencies with astropy units.

Average Models

Simple Average Curves

These are straightforward averages of observed extinction curves. They are the simplest models and include models for the MW (RL85_MWGC, RRP89_MWGC, B92_MWAvg, I05_MWAvg, CT06_MWLoc, CT06_MWGC, GCC09_MWAvg, F11_MWGC, G21_MWAvg, D22_MWAvg; Note the different valid wavelength ranges), the LMC (G03_LMCAvg, G03_LMC2) and the SMC (G03_SMCBar, G24_SMCAvg, G24_SMCBumps).

One often used alternative to these straight average models is to use one of the parameter dependent models with the average R(V) value. For the Milky Way, the usual average used is R(V) = 3.1. See the next section.

Model

x range [1/micron]

wavelength range [micron]

galaxy

B92_MWAvg

1.3 - 2.9

0.34 - 0.78

MW

I05_MWAvg

0.13 - 0.8

1.24 - 7.76

MW

GCC09_MWAvg

0.3 - 10.96

0.0912 - 3.3

MW

G21_MWAvg

0.3125 - 1

1 - 32

MW

D22_MWAvg

0.2 - 1.25

0.8 - 4

MW

CT06_MWLoc

0.037 - 0.8

1.24 - 27.0

MW (Local)

RL85_MWGC

0.08 - 0.8

1.25 - 13.0

MW (GCenter)

RRP89_MWGC

0.08 - 1.25

0.8 - 13.0

MW (GCenter)

CT06_MWGC

0.037 - 0.8

1.24 - 27.0

MW (GCenter)

F11_MWGC

0.05 - 0.8

1.28 - 19.1

MW (GCenter)

G03_LMCAvg

0.3 - 10.0

0.1 - 3.3

LMC

G03_LMC2

0.3 - 10.0

0.1 - 3.3

LMC (30 Dor)

G03_SMCBar

0.3 - 10.0

0.1 - 3.3

SMC

G24_SMCAvg

0.3 - 10.0

0.1 - 3.3

SMC

G24_SMCBumps

0.3 - 10.0

0.1 - 3.3

SMC

Parameter Dependent Average Curves

The models that are dependent on parameters provide average curves that account for overall changes in the extinction curve shapes. For example, the average behavior of Milky Way extinction curves has been shown to be dependent on R(V) = A(V)/E(B-V). R(V) roughly tracks with the average dust grain size.

For MW type extinction, the G23 model should be considered as it spectroscopically covers the far-ultraviolet (912 A) to mid-infrared (32 micron) and is based on the spectroscopic extinction curves used for the GCC09, F19, G21_MWAvg, and D22 studies. For those who wish to bypass the python implementation of the G23 model, tables for the range of valid R(V) values with 0.1 steps are available.

A more general model is G16 as this model encompasses the average measured behavior of extinction curves in the MW, LMC, and SMC. But it only covers wavelengths between 1150 A and 3 micron. The G16 model reduces to the F99 model with fA= 1.0.

Model

Parameters

x range [1/micron]

wavelength range [micron]

galaxy

CCM89

R(V)

0.3 - 10.0

0.1 - 3.3

MW

O94

R(V)

0.3 - 10.0

0.1 - 3.3

MW

F99, F04

R(V)

0.3 - 10.0

0.1 - 3.3

MW

VCG04

R(V)

3.3 - 8.0

0.125 - 0.31

MW

GCC09

R(V)

3.3 - 11.0

0.0912 - 0.31

MW

M14

R_5495

0.3 - 3.3

0.31 - 3.3

MW, LMC

G16

R(V)_A, f_A

0.3 - 10.0

0.1 - 3.3

MW, LMC, SMC

F19

R(V)

0.3 - 8.7

0.115 - 3.3

MW

D22

R(V)

0.2 - 1.25

0.8 - 5.0

MW

G23

R(V)

0.032 - 11.0

0.0912 - 32.0

MW

Notes

The M14 models focus on refining models in the optical, and use the CCM89 models for the NIR and the UV. The M14 models use R_5495 = A(5485)/E(4405-5495), the spectroscopic equivalent to band-integrated R(V); see the paper for discussion. Because of a spurious feature in the near UV caused by smoothly tying their optical to the CCM89 UV, only the NIR and optical portions of the M14 models are provided here.

Grain Models

The models are based on dust grain models that are calculated based on dust size, composition, and shape distributions. The distributions are constrained by observations of dust extinction, abundances, emission, and polarization (usually a subset, not all). One use of these models is to provide extinction measurements at wavelengths not accessible observationally (e.g., in the extreme UV below 912 A).

Model

x range [1/micron]

wavelength range [micron]

galaxy

DBP90 MWRV31

0.00001 - 10.9

0.0918 - 100000

MW R(V)=3.1

WD01 MWRV31

0.0001 - 100

0.01 - 10000

MW R(V)=3.1

WD01 MWRV40

0.0001 - 100

0.01 - 10000

MW R(V)=4.0

WD01 MWRV55

0.0001 - 100

0.01 - 10000

MW R(V)=5.5

WD01 LMCAvg

0.0001 - 100

0.01 - 10000

LMC

WD01 LMC2

0.0001 - 100

0.01 - 10000

LMC2 Region

WD01 SMCBar

0.0001 - 100

0.01 - 10000

SMC

D03 MWRV31

0.0001 - 10000

0.0001 - 10000

MW R(V)=3.1

D03 MWRV40

0.0001 - 10000

0.0001 - 10000

MW R(V)=4.0

D03 MWRV55

0.0001 - 10000

0.0001 - 10000

MW R(V)=5.5

ZDA04 MWRV31

0.0001 - 1000

0.001 - 10000

MW R(V)=3.1

C11 MWRV31

0.00001 - 25

0.04 - 100000

MW R(V)=3.1

J13 MWRV31

0.00001 - 25

0.04 - 100000

MW R(V)=3.1

HD23 MWRV31

0.000033 - 10

0.1 - 30000

MW R(V)=3.1

Y24 MWRV31

0.00001 - 25

0.04 - 100000

MW R(V)=3.1

Shape Models

The models that focus on describing the full extinction curve shape are usually used to fit measured extinction curves. These models allow features in the extinction curve to be measured (e.g., 2175 A bump or 10 micron silicate feature). The P92 is the most general as it covers the a very broad wavelength range. The FM90 model covers the UV wavelength range and has been extensively shown to fit all known UV extinction curves. The FM90_B3 model provides a variant of the FM90 model that uses B3 instead of C3 as B3 = explicit 2175 A bump height = C3/gamma^2. G21 model focuses on the NIR/MIR wavelength range from 1-40 micron.

Model

x range [1/micron]

wavelength range [micron]

# of parameters

FM90

3.13 - 11.0

0.0912 - 0.32

6

FM90_B3

3.13 - 11.0

0.0912 - 0.32

6

P92

0.001 - 1000

0.001 - 1000

19 (24 possible)

G21

0.025 - 1

1 - 40

10