This could happen at two degrees of global warming
Moderate global warming does not rule out consequences associated with a three- or four-degree warmer world, according to a new study.
Droughts in important agricultural areas are estimated to occur 10 per cent more often in a world that is two degrees warmer on average, but the increase could be as high as 50 per cent.(Photo: André Künzelmann / UFZ)
Jana Sillmann at CICERO Center for International Climate Research, together with colleagues in Germany and Switzerland, has examined the risks of moderate global warming.
They focused on three types of events: flooding, drought in key agricultural areas, and forest fires.
Large climate reports, such as those from the Intergovernmental Panel on Climate Change (IPCC), typically summarise results from many climate models to estimate how much warmer it will get and what the consequences will be.
Annonse
Jana Sillmann explains that typically, an average is taken from many different climate models.
Within this average, there is considerable variation.
The new study shows that individual simulations may predict stronger changes at two degrees of warming than the average projections for a world that is three or four degrees warmer.
Jana Sillmann is a research group leader at CICERO.(Photo: Private)
Cannot simply be dismissed
"The main point of the study is that we cannot just look at the average. Even at two degrees of warming, the climate may produce extreme rainfall or drought that could have a very large impact," says Sillmann.
"Each model represents a realistic and plausible development of the climate in the future," she points out.
This means that more severe outcomes cannot simply be dismissed as incorrect, even if they are less likely.
In the new study, the researchers used simulations from a large international collaboration known as CMIP6. These are the same climate models used in UN climate reports.
They examined three potentially serious effects of a two-degree temperature increase:
Drought in agricultural areas, intense rain in urban areas, and fire-prone weather conditions in forest areas.
Why it matters
Jana Sillmann explains why they chose to look at food security, flooding, and wildfires:
"Food security affects people all over the world. Society is globalised, and we get our food from different parts of the world. Global projections are important to see how strongly different parts of the world will be affected by heatwaves that, for example, can destroy crops."
When it comes to wildfires, there have been many serious forest fires in recent decades, says Sillmann.
"They have destroyed large forest areas that we actually need to absorb CO2 from the atmosphere."
Floods can lead to fatalities and destroy infrastructure in urban areas.
"We chose these because they are socially relevant and have already affected societies worldwide."
"Much higher than the average for four degrees"
The study highlights the wide range of outcomes projected by different climate models for a two-degree increase in temperature.
First, food security:
The researchers looked at how often drought occurs in regions critical to global food production, such as areas that grow wheat, soy, rice, and maize.
Annonse
On average, models suggest droughts will occur up to 10 per cent more frequently at two degrees of warming compared to pre-industrial levels.
At three or four degrees of warming, the increase is estimated to be around 15 per cent, according to the study.
But the models do not fully agree. Some models predict that in the worst case, drought frequency could increase by 20, 30, or even 50 per cent in a two-degree warmer world.
"This is much higher than the average for four degrees," says Sillmann.
On the other hand, some models indicate only a small increase of a few per cent, lower than the average suggests.
Both the best and worst
For flooding, the researchers examined maximum rainfall over a five-day period in urban areas.
The average across models suggests an increase of about 8 per cent at two degrees of warming.
But a few models project increases of up to 14 per cent.
"That's closer to what we would expect at three degrees of warming," she says.
If the worst projections prove accurate, there will be more intense rainfall in urban areas, including in India, parts of Europe, and East Asia.
Other models suggest a more modest increase of around 4 per cent in the same regions.
"If we are to assess the risk, we need all the information, including both the worst and best scenarios," says Sillmann.
Important for risk assessment
There are also major differences when it comes to fire-prone weather in forest areas. In the most extreme scenario, the increase is four times higher than in the most optimistic estimate.
The most severe projections even exceed the average expected for a world that is three degrees warmer.
The researchers see the largest changes in rainforest regions, as well as in North America and northern Europe and Asia.
"To properly assess risk, we need to take all model outcomes into account. We cannot rely only on averages, since severe consequences may still occur at lower levels of global warming than anticipated," says Sillmann.
"This kind of information is crucial for decision-makers, as accurate and comprehensive data is needed to make reliable risk assessments," she adds.
Have used a 'democratic principle'
Cecilie Mauritzen, a climate researcher at the Norwegian Meteorological Institute who was not involved in the study, considers the study both strong and relevant.
Future climate projections, such as those used by the IPCC, are calculated by dozens of model centres around the world, she writes in an email to Science Norway.
"Because each projection carries uncertainty, and no single model has been considered more reliable than others, researchers have traditionally followed a 'democratic principle' by averaging all results."
However, growing knowledge about why models differ now makes it possible to identify which ones are better suited for specific purposes, Mauritzen explains.
"An important improvement"
She elaborates:
"For example, if you are interested in how precipitation will change in the future, it's important to exclude models that do not produce realistic storm tracks today."
If storm systems do not reach Europe in simulations of the current climate, then the models will not produce much rainfall there – and it is unlikely they will estimate future precipitation accurately either, she explains.
On the other hand, if the focus is on temperature development, storm tracks are less important. Instead, it may be necessary to exclude models that struggle with radiation and clouds.
"This study has done exactly that. The researchers have excluded models that behave unrealistically in relation to the questions they are addressing, such as food security. In doing so, they have moved away from the democratic approach and towards a new framework that uses scientific insight to evaluate and rank climate models," she writes.
According to Mauritzen, they are not the first to make this choice, and certainly won't be the last, because this represents an important improvement in climate projection research.
"By selecting only parts of the large set of future projections, the average values will naturally change. This means the estimated risk of extreme outcomes may be either higher or lower than the 'democratic' average. In turn, this improves our understanding of what could happen in the future," she writes.
