# Computational Earth Sciences meetings

Computational science (taken broadly and which could include developing new software, significant use of compute resources, or complex data management) is an increasingly important part of the Earth sciences, but we often lack the opportunity to discuss the details of the computational methods and tools we use. The aim of these meetings is to provide space to discuss these details across the traditional research areas in the Earth sciences. Ideally the science background for these talks will be approachable by anybody in the Department, but the computational side will be more detailed than what is typically covered outside individual research groups. We also include some broader discussion of ideas and the literature as well as some training activities.

See https://compearthsci.github.io/meetings/2023/09/18/schedule.html

## Next term

optimal transport - towards the end of term (after 9th Nov)
Green computing - external speaker
Research software engeneering as career path?
Something else about supercomputers - Frank and Tom (scheduling needed... hard)

## Ideas for other activities

Other things that would be interesting to do:

* Managing (python) environments - best practice
* 1 minute presentations on your favorite library
* Blogs that are worth keeping track of. These include [Nick Higham](https://nhigham.com/blog/) on applied maths, numerical linear algebra and software, [It Will Never Work in Theory](https://neverworkintheory.org/) on empirical research on software engineering, [Walking Randomly](https://walkingrandomly.com/) by Mike Croucher, and RSE who now works for Mathworks, and [The Third Bit](https://third-bit.com/blog/), Greg Wilson's contributions on how we teach programming in HE (and other things).
* Training resources: [ARC at Oxford](https://www.arc.ox.ac.uk/training)
* Basic bash (from zero to, say, submitting an ARC job)
* Tips for better bash (tab compleation and history, loops, functions...)
* Intro to version control with git
* Intro to parallelising code (basic concepts, range of approaches, pointers to further information and a python example)
* GPUs (at what point in the next 5 years should we care)
* Julia
* Better python memory management
* Python optimisation
* (Quantum computing)
* How to save data
* timestepping methods, finite element methods, discrete element methods
* intro to the "departmental network"
* ssh tricks
* modules and encapusalation in python
* testing 101
* Green algorithms https://www.green-algorithms.org/

## Mailing list

To avoid overfilling your mail boxes, further information about these meetings will be circulated via a dedicated mailing list. In order to self subscribe to the mailing list for computational geoscience, please send an email to [`[e-mail]`](mailto:[e-mail]). There is no need for this email to contain any text, and the mail list system should generate a reply asking for confirmation.