Warmer, wetter, hotter, drier? How to choose between climate futures

Some climate models show that, under high greenhouse gas emissions, Sydney could be up to 4.8C hotter and have 20% less rainfall by 2090. Others show the Harbour City could warm only by 2.3C and become 25% wetter. How do we deal with such large uncertainty?

The online Australian Climate Futures tool released this week by CSIRO and the Bureau of Meteorology is designed to show projections from up to 40 climate models for different regions, years and emission scenarios. The tool helps select future scenarios so people can plan for climate change.

The tool is part of a new website, with updated climate projections for Australia, that gives unprecedented access to climate change projections data. You can explore the key changes expected in your area through the Regional Climate Explorer.

Climate Futures web tool from CSIRO on Vimeo.

No single future

Climate change projections are presented as a range of possibilities. This occurs because different models produce different outcomes. Even though they are based on the same physical laws, such as conservation of mass, moisture and energy, each climate model treats regional processes in the oceans and atmosphere slightly differently.

These differences pose problems for decision makers who are keen to prepare for climate change. How can we plan for this range of possibilities? Can we limit the number of future climate scenarios? Are some models more reliable than others?

Choose your climate future

If you want to assess the impact of climate change, you need to look at the full range of possibilities.

A projection from a single climate model will not represent the full range of possible futures and impacts. So we need to look at a range of data sets and models. This might seem a daunting task when confronted by up to 40 models, up to four emission scenarios, multiple years and multiple climate variables.

The Climate Futures web tool simplifies the process. It is underpinned by the Climate Futures Framework and the most extensive, independently peer-reviewed climate model evaluation ever undertaken in Australia.

The tool considers three cases: “best case”, “worst case” and “maximum consensus”. It also allows us to compare two different climate variables, such as temperature and rainfall, at the same time. We can further simplify these scenarios into “hotter and drier” or “warmer and wetter”.

For instance, below we have the climate futures of southwest Western Australia, comparing changes from 40 global climate models in temperature and rainfall around 2050 under an intermediate emissions scenario. Most models show a warmer and drier climate (1.5-3.5C warmer, 5-15% drier). This is the “maximum consensus” future, with almost half (45%) of the models producing this outcome.

CSIRO

But it is also helpful to think about other climate futures that are less likely but could have a much greater impact. A worst-case scenario for some sectors might be much drier and hotter (more than 15% less rainfall and 1.5-3C hotter). This future is simulated by 5% of the models.

A best-case scenario for some sectors might be wetter and warmer, also simulated by 5% of the models.

For Sydney, Australian Climate Futures shows that by 2030 under high greenhouse gas emissions, more than half the models simulate Sydney’s climate becoming warmer with little change in rainfall (0.5 to 1.5C warmer and less than 5% change in rainfall).

However, more extreme changes – such as warmer and much drier, or warmer and much wetter – are simulated by a few models. These may be very important in terms of impacts.

The tool also has a feature that helps users select a small number of climate models that represent the “best case”, “worst case” and “maximum consensus” futures. This reduces the effort involved when assessing the impact of climate change. The tool also provides information about model reliability, and warns against using some models in some areas.

Planning for the future

The natural resource management (NRM) community is already using the tool to assist in incorporating climate change into regional strategic plans.

NRM groups are able to use the Climate Futures approach to ensure a range of plausible futures is considered when making, for example, decisions about on-ground options for biodiversity management, or when assessing the opportunities provided by carbon farming schemes to improve regional landscapes.

For example, under the worst-case climate future, an endangered vegetation community may be projected to disappear altogether. To address this would require a large investment – perhaps to translocate species to newly emerging suitable habitats. Under the best case, though, it may be projected to shrink but still remain. This may require lower levels of investment to enhance existing management.

For the maximum consensus future, the expected impacts (and treatment costs) lie between these extremes. Thus planners can weigh this information with social, economic and environmental factors to make informed investment decisions.

The tool and framework have also been successfully applied in the Asia-Pacific region, supporting neighbouring countries to simplify the processes of undertaking climate impact assessments.

Through an extensive Pacific-Australian climate science collaboration, countries like Samoa have used the Pacific Climate Futures tool and associated climate projections to review and update important national infrastructure.

Recently, Pacific Climate Futures was used, along with detailed extreme sea-level modelling, to enable a “climate-proof” design for the new Samoa Parliament Complex.

This article is the second in a series on climate change in Australia, coinciding with the release of new climate websites by CSIRO. Read more:

• A new website shows how global warming could change your town

John M Clarke will be on hand for an author Q&A between 11 am and 12 pm AEST on Friday April 10. Post your questions in the comments section below.

This research was partially funded by the Department of Environment. As employees of CSIRO the authors have undertaken research for various State/Territory and Australian government departments in recent years.

This research was partially funded by the Department of Environment. As employees of CSIRO the authors have undertaken research for various State/Territory and Australian government departments in recent years.

This research was partially funded by the Department of Environment. As employees of CSIRO the authors have undertaken research for various State/Territory and Australian government departments in recent years.