Enabling essential climate science research

Climate Change is a defining issue of our time. The UK’s top 10 warmest years have all occurred since 2002, and July 2019 saw the UK’s hottest ever recorded temperature (38.7 °C). Whilst many are enjoying the unusually non-rainy British weather, it doesn’t come without costs. Extreme weather events, such as heatwaves and flash flooding, are expected to increase in severity and frequency as our climate continues to change. Extreme weather can adversely affect the way we live our lives. Loss of food production, damaging wildlife habitats, and endangering human life are just a few impacts associated with climate change. Predicting global climate change is complex and requires vast amounts of data processing capabilities. Steps must be taken to improve the accuracy of future climate models. That’s where JASMIN - a data intensive supercomputer - comes in. 

In order to better predict climate change and thus mitigate against its effects, scientists used JASMIN to analyse data from a new generation of high-resolution global climate models. The PRIMAVERA project allows over 100 scientists from across Europe to work together to process vast numbers of climate simulations. Dr Jon Seddon, a Senior Scientific Software Engineer at the UK Met Office who is working on the project, said: “JASMIN is probably the only facility in Europe (and the world) that would allow PRIMAVERA to do its research. This is because of the large storage capability and the ability for users from multiple institutes across Europe to collaborate where the data is stored”. 

Two Petabytes of climate simulations were produced by seven different climate models on high performance computers (HPCs) across Europe for use by the project. Transferring large volumes of data between computing facilities can be a time consuming and laborious process. PRIMAVERA needed the data to be transferred to JASMIN so that they could work together to analyse it in a shared workspace. Luckily, JASMIN has a dedicated fast connection to the UK and European research networks - meaning that the data could be efficiently ‘shipped’ to JASMIN, thus allowing the collaboration.  

JASMIN not only made transferring and sharing the data easy, it also made processing huge amounts of data possible too. Dr Seddon explains how the team used JASMIN’s batch computing cluster to do this. “Due to the large volumes of data processing that we needed to undertake, it was essential that we could test our analyses and then use the parallel processing capability offered by LOTUS on the full 2PB dataset”. Running analyses in parallel can dramatically reduce analysis time - so large complex datasets can now be analysed in record time (i.e. days vs. weeks or months). Once the analyses are complete, the reprocessed datasets are curated, on the CEDA Archive, for long term reuse by other scientists. 

Dr Seddon explains how other projects not using JASMIN are struggling; “Other projects without access to JASMIN are requiring users to email their analysis code to each of the data providing institutes and then the results are emailed back. This is much slower and much less flexible than users having direct access to the data. The alternative requires a member of staff to store the data on a 10 TB disk and fly with it - not ideal when we are trying to help solve climate change, not contribute to it!”. The scale and complexity of the PRIMAVERA project means that it would not have been possible without JASMIN. 

As datasets continue to grow and higher resolution climate simulations are modelled, it is essential that computing facilities, like JASMIN, continue to support the needs of the environmental science community. The much higher resolution climate simulations that researchers can now analyse at JASMIN provide a much more detailed look at how the global climate works. Greater resolution datasets are essential for working out how climate change affects society and also allows researchers to make more accurate predictions of the future climate.

Footnotes:

JASMIN is managed jointly by STFC’s Scientific Computing Department and CEDA (Centre for Environmental Data Analysis), part of RAL Space. It is funded by the Natural Environment Research Council (NERC). 

PRIMAVERA is a Horizon 2020 project funded by European Commission under grant agreement no. 641727. The project is a collaboration between 19 European partners, led by the Met Office and the University of Reading in the UK. PRIMAVERA started in November 2015 and runs to the end of July 2020.

Contact: 

For more information about using JASMIN, please contact our helpdesk

To find out more about the PRIMAVERA project, contact the helpdesk and we can put you in touch with the researchers involved