Welcome Prerequisites Registration Venue Programme Accommodation Science Plan Software Stack Contact

Welcome

The 2026 UK km-scale hackathon will take place at the Met Office Collaboration Building, Exeter, on 18-22 May 2026. A hierarchy of km-scale Met Office Unified Model simulations, from regional to global, will enable analysis of atmospheric phenomena over convective to planetary scales. It is part of the World Climate Research Programme (WCRP) effort to push the boundaries of climate system modelling and digital innovation.

Prerequisites

During the hackathon climate data from km-scale simulations with the Met Office Unified Model will be provided in Healpix format at different zoom levels. Jupyter Notebooks will be provided to explain the basics of the data handling. Participants can then develop their own Jupyter Notebooks or python scripts to analyze and plot the data.

Participants are expected to have basic knowledge of:

  • Linux terminal
  • Python, including Matplotlib, Numpy, Xarray
  • Concepts of data visualization
  • Handling of global 2D and 3D data in Climate science or similar fields

Registration

Registration is now open! The registration deadline is Friday 17th April 2026.

Please use this Microsoft Form to confirm you are attending and which days you will join for. The questions will help us to make sure everything is arranged for the hackathon to run smoothly. Note that this event is free - there is no registration fee. The Met Office will provide lunch and snacks during the working day, we hope to plan a group meal on evening of Wed 20 May.

Venue

The event will take place at the Met Office Collaboration Building in Exeter - we look forward to welcoming you.

Programme

The 2026 UK K-Scale hackathon will run from lunchtime on Monday May 18 to lunchtime on Friday 22 May. Lunches will be provided at the Met Office Collaboration building each day. Further details of the programme for the week will follow.

Accommodation

Please organise your own accomodation for the week. The hackathon will take place in the Met Office collaboration building which is on the opposite (east) side of the M5 to Met Office HQ and part of the Exeter Science Park. The collaboration building is approximately a 20 min walk from Met Office HQ so staying at the Holiday Inn or Premier Inn (both of which are next to Met Office HQ is one option). Alternatively there is a new hotel at the Exeter Science Park, Voco Zeal, which is approx. 5 min walk from the collaboration building. Some may want to make use of the multiple hotel options in Exeter City Centre, the 4A bus from the city centre goes to/from the science park but allow additional time during the morning and evening rush hour.

Science Plan

The science topics for all nodes fall under two umbrella themes: 1) organised travelling convection and 2) circulation, which are closely related. As part of the investigations, there are working groups on tracking (of TCs, MCSs, MJO); characteristics of precipitation, e.g. large scale versus convective, spectra, diurnal cycle; atmospheric blocking; air-sea and land-atmosphere coupling, etc. The UK plans to explore a hierarchy of simulations from the Met Office K-scale project, from global to regional, to investigate:

  1. Convection under a range of simulation approaches, e.g. the dependence on resolution, domain size, type of convective parametrisation
  2. Weather systems, such as TCs and MCSs, and their interaction with the large scale circulation
  3. Upscale effects of a number of phenomena, with initial focus on TCs, and how they affect the large scale circulation
  4. Cloud tracking across models using state-of-the art tools (e.g. PyFLEXTRKR, tobac) and will expand on analysis of cloud spatial structure using multifractal and computer vision approaches.
  5. The Oxford group will also work on global km-scale ICON runs with explicit aerosols conducted within NextGEMS.

Science Questions:

  1. Are weather systems, and their interactions with the large-scale flow, significantly different, depending on the use of convective parametrisation in the various DYAMOND-3 simulations?
  2. Can we detect significant contributions to the large-scale flow by the existence of MCSs and TCs, including the ones in the top categories?
  3. What are the characteristics of precipitation under different treatments of resolution, parametrisation (including aerosol), coupling?
  4. Having established the degree of simulation fidelity, based on the analyses of phenomena and processes above, are any of the participating models good/valuable candidates for ML training?
  5. What new science emerges robustly across the models taking part in DYAMOND-3? Examples:
    1. strength and process-chain for UPSCALE effects emerging from analysis of TCs and MCSs
    2. quality of the precipitation fields as revealed by comparison with latest observations, and in preparation for EarthCARE
    3. are there universal lessons on the value of simulation at these scales, e.g. in terms of results that are robust across model families?

Approach: A number of data sets will be made available on the JASMIN object store, in HEALPix format, and accessible remotely. Notebooks to perform the analyses above, at least at an initial stage, will be provided to all, including catalogues for data discovery. Observational products will also be provided, e.g. ERA5, IMERG, CERES. Many of the more complex data sets, e.g. TC tracks, MCS tracks, will be provided before the event. Multiple working groups of 5 or more individuals will be formed around the topics above, and will work together at the venue, as well as identify equivalent groups at other nodes to exchange information with on a daily basis.

Software Stack

Here are some steps to get you started for the hackathon. These should be done before the hackathon begins.

1. Hackathon python environment

In order to be able to run the notebooks provided for the hackathon and conduct data analysis using the tools for the hackathon, the official global hackathon software stack needs to be installed. To do this, please follow this link and download the environment.yaml file. Then install the python environment with the following command (note: you need to have conda installed and the base environment activated to do this):

conda env create --name hackathon -f environment.yaml

Activate the environment with:

conda activate hackathon

Instead of hackathon you can also choose any other name for your environment.

2. Notebook kernel

If you are planning on running your code on the Jasmin Jupyter server open the terminal on the Jasmin notekook web interface, activate your environment from step 1 and then type:

python -m ipykernel install --user --name=name-of-environment

(as explained in step 2 here)

This will allow you to use your environment in your notebooks (you might have to reload the page to see the change).

3. Test your environment

To make sure everything is working and to familiarize yourself with the available tools run the demo notebook in this link.

It shows you how to access the data by using the intake catalog and creates some simple plots.

4. Further reading

If you would like to learn more about the healpix format in which all the model data is stored please follow this link.