Model Flavors

There are four different types of models: average, R(V)+ dependent prediction, shape fitting, and grain models. 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

These models provide averages from the literature with the ability to interpolate between the observed data points. Models are provided for the Milky Way for the optical (Bastiaansen 1992), ultraviolet through near-infrared (Gordon, Cartlege, & Clayton 2009) and near- and mid-infrared (Rieke & Lebofsky 1985; Indebetouw et al. 2005; Chiar & Tielens 2006; Fritz et al. 2011; Gordon et al. 2021; Decleir et al. 2022) and the Magellanic Clouds (Gordon et al. 2003; Gordon et al. 2024).

For the Milky Way for the ultraviolet through near-infrared, one of the R(V) dependent models with R(V) = 3.1 (see next section) is often used for the Milky Way ‘average’.

(Source code, png, hires.png, pdf)

../_images/model_flavors-1.png

(Source code, png, hires.png, pdf)

../_images/model_flavors-2.png

R(V) (+ other variables) dependent prediction models

These models provide predictions of the shape of the dust extinction given input variable(s).

The R(V) dependent models include CCM89 the original such model (Cardelli, Clayton, and Mathis 1989), the O94 model that updates the optical portion of the CCM89 model (O’Donnell 1994), the F99 model (Fitzpatrick 1999) updated as F04 (Fitzpatrick 2004), and the fully spectroscopic far-UV to mid-IR G23 (Gordon et al. 2023). These models are based on the average behavior of extinction in the Milky Way as a function of R(V). The M14 model refines the optical portion of the CCM89 model (Maiz Apellaniz et al. 2014), was developed for the LMC but has been shown valid elsewhere in the Milky Way.

In addition, the (R(V), f_A) two parameter relationship from Gordon et al. (2016) is included. This model is based on the average behavior of extinction in the Milky Way, Large Magellanic Cloud, and Small Magellanic Cloud.

(Source code, png, hires.png, pdf)

../_images/model_flavors-3.png

(Source code, png, hires.png, pdf)

../_images/model_flavors-4.png

(Source code, png, hires.png, pdf)

../_images/model_flavors-5.png

(Source code, png, hires.png, pdf)

../_images/model_flavors-6.png

(Source code, png, hires.png, pdf)

../_images/model_flavors-7.png

Grain models

These models provide literature grain models interpolated between the computed data points. These dust grain models are based on fitting observed extinction curves and other observed properties of dust (e.g., abundances, IR emission). Models are provided for the Milky Way calculated for the X-ray to the submm.

(Source code, png, hires.png, pdf)

../_images/model_flavors-8.png

(Source code, png, hires.png, pdf)

../_images/model_flavors-9.png

(Source code, png, hires.png, pdf)

../_images/model_flavors-10.png

Shape fitting models

These models are used to fit the detailed shape of dust extinction curves. The FM90 (Fitzpatrick & Mass 1990) model uses 6 parameters to fit the shape of the ultraviolet extinction. Note there are two forms of the FM90 model, FM90 that implements the model as published and FM90_B3 that B3 = C3/gamma^2 as the explicit amplitude of the 2175 A bump (easier to interpret). The P92 (Pei 1992) uses 19 parameters to fit the shape of the X-ray to far-infrared extinction. The G21 (Gordon et al. 2021) models uses 10 parameters to fit the shape of the NIR/MIR 1-40 micron extinction.

(Source code, png, hires.png, pdf)

../_images/model_flavors-11.png

(Source code, png, hires.png, pdf)

../_images/model_flavors-12.png

(Source code, png, hires.png, pdf)

../_images/model_flavors-13.png