“Would this heatwave or severe rainfall have happened without climate change?” This is the kind of question that policymakers and businesses are asking scientists.
Climate attribution science can help us understand the answers to questions like this, as the probability and intensity of extreme weather changes in a warmer world. This will help countries, communities and businesses become more resilient to a changing climate.
Earlier this month we looked at what is meant by Loss and Damage.
Climate attribution science plays a part in the evidence base needed in this complex area, and here the Met Office’s Dr Fraser Lott – a Climate Monitoring and Attribution Scientist – explains how attribution works and the challenges it presents.
What is an extreme weather and climate event, and how does it differ from a ‘normal’ event?
Extreme weather and climate events affect all aspects of society, and impacts can include large economic costs, displacement and loss of life. An extreme weather event – such as a heatwave or extreme rainfall – is defined as an event which ranks above a threshold, or is near the upper or lower ends of the historical range of values; making it significantly different from the usual weather pattern. For instance, this could be temperatures above a certain threshold for a sustained period. Some extreme events would have happened within natural variability, but their likelihood will change with climate change. Climate attribution science can help us to identify this difference.
How does climate attribution help us to understand whether extreme weather is caused by climate change?
Climate-related attribution studies look at individual extreme weather events such as a heatwave and ask: “Would this have happened without climate change?” And if so: “How much hotter or longer is this heatwave because of climate change?” And: “How much more likely are these heatwave conditions because of climate change?” There is also climate trend attribution, which look at trends beyond individual extreme events.
Attribution science compares modelled worlds with and without the human influence on greenhouse gas emissions to understand the impact of climate change on the intensity and frequency of specific weather events. It can then say for instance: “climate change has made this heatwave 30 times more likely”. This is called an attribution statement.
How accurate is climate attribution?
Climate attribution studies rely on a lot of information, which means they have historically taken years to analyse whether a specific extreme event is linked to climate change. The last 20 years have seen huge developments in climate attribution; meaning that now scientists are able to conduct rapid climate attribution in a few days to weeks after an event. These are timescales relevant to the public and disaster recovery programmes.
Due to the speed of analysis, the scientists behind these rapid attribution studies do not always have time to conduct the extra simulations that would be part of a longer-term study of the specific details of the event. This leads to a trade-off between speed and level of confidence in the result, meaning we may need to make more general statements in a rapid study than we would in a scientific paper.
Confidence in the attribution studies also depends on the quality of observations going into the model, the ability of the models to simulate a particular type of extreme event, and how well the physical processes causing this extreme event are understood (as well as how they may change with climate change).
For instance, current climate models are good at replicating conditions which lead to extreme heat and cold, but it is harder to simulate conditions which lead to severe convective storms and tropical cyclones. This means confidence in attributing extreme events such as heatwaves to climate change is higher than attributing cyclonic activity to climate change.
Can climate attribution be applied to extreme weather across the globe?
The requirements for making a confident attribution statement mean that climate attribution is not yet used in all situations. For instance, regions that lack a reliable and sufficient set of historical observations (a problem encountered particularly in developing countries), will struggle to provide a dataset to base their model on. Not all climate models perform well over all regions of the globe. For example, some climate models do better in mid-latitudes compared to tropical ones, which can be a barrier to doing attribution studies in regions where climate models struggle to simulate the relevant physical processes.
Developing countries – including those more vulnerable to extreme weather – can also lack resources, tools and training to perform studies. This can create a bias with attribution studies focussing on more developed nations in the global north, rather than being truly representative of extreme events which have occurred globally. This also means that scientists from the global north often need to conduct attribution studies for a region in which they are not experts.
The Met Office has been supporting the development of expertise in other parts of the globe, as well as helping improve observations, through work such as that delivered by the Weather and Climate Science for Services Partnership programme, supported by the UK Government’s Department for Science, Innovation and Technology.
Carbon Brief has an interactive map of up-to-date attribution studies, and UK-based attribution studies can be found on the Met Office website.
In the next in our series of blog posts around the theme of Loss and Damage, we will be exploring how climate attribution science helps increase understanding and provide an evidence-base in relation to this complex topic. Follow this blog or #GetClimateReady on Twitter to keep up to date.