Minutes of LIME discussions in Edinburgh, March 10

Introduction

Molecular excitation and radiative transfer code
Given a density, temperature, abundance profile, it computes the NLTE RT in full 3D
Written in C, compiled at run time
Demo

Better interface: have an easy way to create a grid of model.
WIP in Leiden and Toulouse
How to get the exact same grid twice? Feature in LIME that allows to do this already.
Save the voronoi cells as well? No much control on how the grid is built; may be a problem for some problem.
Make the gridding part of the model.c (the user can define it).
Preview can be used to check the grid

Performance: problem that we can not tackle because of speed/performance?
MCMC search of best fit parameters
Running multiple instance of code of a cluster is difficult; need to compile the code only one
Python branch has this, but it needs to be tested
Interface:
WIP in progress to include LIME in CASA
Well defined SF models to make predictions for e.g. proposals
These models could used as benchmark as well
Benchmark; time for a new RT workshop to compare code results?

Integrate with chemistry code?
Would add much complexity; better to used specialized code (khrome, astrochem)
Integrate with other codes (e.g. hydro)
Maintain a cookbook (on the wiki) on how to use chains of code (hydro -> chemistry -> LIME)
FIR lines (JWST). For this we need:
Central source
Scattering
Large range in velocity gradient
UV (could be computed with another code and used as input)
Lot of work