This code provides a method for recovering gas-phase metallicity gradients from integral field spectroscopic (IFS) observations of barely resolved galaxies.

The approach here is based on [Carton et al. 2017](https://ui.adsabs.harvard.edu/abs/2017MNRAS.468.2140C and employs a forward modelling approach to fit the observed spatial distribution of emission-line fluxes, accounting for the degrading effects of seeing and spatial binning. The method is flexible and is not limited to particular emission lines or instruments, nor one set specific set of photoionization models.

While this model can be fit in many ways, we adopt a Bayesian approach with a robust likelihood.

It is advisable to install this in a separate virtual environment. The code currently requires

• numpy
• scipy
• pymultinest
• h5py
• astropy

and it is advisable to also install matplotlib and getdist for visualisation and jupyter to use the notebooks and nbconvert to ensure the notebook is possible to run on other installations.

If you use conda, a possible installation method would be

  > conda create -n metaldisc  matplotlib numpy scipy h5py astropy  jupyter nbconvert pip
   <... Various output ...>
  > conda activate metaldisc
  > pip install pymultinest
  > pip install getdist
  > git clone https://github.com/jbrinchmann/metaldisc.git
  > cd metaldisc
  > pip install . 
  OR
  > pip install -e . 

Note in particular that pymultinest and getdist are most easily installed using pip so that is what is indicated here.

In the example directory, there are two scripts:

• fit_from_sfrmap.py - shows how to fit the example data
• model_from_sfrmap.py - provides a visualisation of the model output

The example directory also contains two notebook versions of this. The fitting notebook uses getdist to visualise the results in contrast to fit_from_sfrmap.py which uses the code distributed with pymultinest.

If you use this code, you should cite Carton et al. 2017, where the method is explained.

You may also wish to cite Carton et al. 2018, where we provide further discussion and improvements.