How to create research graphics in MatLab, Python and Ferret

By Pearse Buchanan, Ramkrushn Patel, Saurabh Rathore

When we read a paper with great graphics, graphics that convey the story well, the research is more easily understandable. The author’s conclusions can even be more powerful. A picture is definitely worth a thousand words.

But, creating great figures takes time. Many of us have spent days making small changes to a figure that needed to be just right. Everyone does this, to a greater or lesser extent.

For those who are just starting out though, making beautiful figures can be a daunting task. You might not even know where to start. You have probably heard of lots of different programs, but each has a different syntax, pros and cons.

When I started my PhD, I had very little coding skill. Sure, I’d had a go on MatLab, but I was by no means confident in my programming. Needless to say, I felt daunted when I read great papers with beautiful graphics.

So, with the burgeoning climate scientists in mind, some of us here at the ARC Centre of Excellence for Climate System Science have pulled together what we have learned over the years in our respective figure-making-programs of choice.

The following figures and their source code were produced in MatLab (Ramkrushn Patel), Python (Pearse Buchanan) and Ferret (Saurabh Rathore). For each program, we have provided a few examples complete with the code that was used to produce them. We hope that these short sections of our work are helpful.

Follow the links to find code and figures developed in:

For any questions or comments, please feel free to contact.

Pearse Buchanan –

Ramkrushn Patel –

Saurabh Rathore -

NB: The code that is provided is only that required to make the figures. No data is included, or any pre-processing that was undertaken to prepare the data for plotting.

UNSW logo ANU logo Monash logo UMelb logo UTAS logo