GENIE:ESci glossy text

Letting the GENIE out of the bottle
GENIE: Grid ENabled Integrated Earth system modelling

The GENIE project’s key challenges were to:

•	develop a software framework for flexibly constructing computationally-efficient Earth system models

•	build more comprehensive Earth system models including the carbon and other nutrient cycles, ice sheets and ocean sediments

•	use these models both to understand past climate change and to project future change

•	improve confidence in model projections by taking statistically sound approaches to tuning models and better quantifying uncertainty

Background

The most sophisticated Earth system models include complex interactions between different components of the Earth system: the oceans, atmosphere, land-surface, ice sheets and life. Individual computer models represent each of these components. Researchers working on the GENIE project used the greatly enhanced power of distributed computing to develop a new Earth system modelling framework that allows much more flexibility in how these different components interact, making it easier to use different configurations of models.

Key outcomes

•	Simulations of climate and the carbon cycle over the last ice age cycle

•	Projections of climate change and sea level rise on a millennial timescale

•	Quantification of uncertainties in future projections

•	New software to couple together Earth system models

•	Advanced tuning of Earth system models toward observations

•	User-friendly toolkit to run models and analyse the results

Worlds of the past

Over the last 125,000 years the Earth has been plunged into and has recovered from a great ice age. Understanding the causes of such natural climate changes, and the corresponding variations in greenhouse gases, is an essential prerequisite to improving confidence in future predictions. Together with NERC's Quantifying and Understanding the Earth System (QUEST) programme, we are using GENIE models and Grid computing resources to simulate the climate and carbon cycle through the last ice age cycle. By comparing the models with paleo-data this is helping constrain the models' representation of key processes and thus improve confidence in future projections.

Climate in year 3000

How much warmer will the world be at the next millennium? Could Britain become a sub-tropical oasis with a reshaped coastline? These were questions addressed by the GENIE team in a study commissioned by the UK Environment Agency through the Tyndall Centre for Climate Change Research. The study revealed that if we burn the known reserves of coal, oil and gas the world will get 6°C warmer with sea levels rising 3-4m. If unconventional fossil fuel sources are exploited it could get over 10°C warmer causing a complete melt of the Greenland ice sheet, collapse of the global ocean conveyor, and over 10m of sea-level rise. By globally achieving the UK target of a 60% reduction in carbon dioxide emissions by 2050 followed by a complete stop in emissions by 2200, we might protect Greenland from melting and restrict global warming to 1.5°C.

To be or not to be circulating

Some of the great circulation engines in nature can be either ‘on’ or ‘off’. The global ocean conveyor is the most famous example. But some recent studies have questioned whether it really has two stable states? GENIE scientists have used computers from all around the UK, including the National Grid Service, to map out the possible states of the ocean circulation in Earth system models of differing complexity. In all models there are certain combinations of ocean and atmospheric parameters for which both ‘on’ and ‘off’ states of the circulation are stable. To their surprise, they found that in a model where the oceans and atmosphere fully interact this extends the range of parameters where the ocean exhibits two stable states. In these cases, if the ocean conveyor is switched off, the atmosphere makes it harder for it to switch back on again.

Anticipating the tipping point

Can we detect an approaching tipping point, such as a collapse of the global ocean conveyor, before we reach it? If so, an early warning system would be of obvious value to policy makers and wider society. In principle an early warning is possible, because a system approaching a critical transition becomes more sluggish in its response to perturbations, including natural variability. GENIE researchers have developed a method that extracts information about a system's response time from time-series data. In model simulations the method can successfully anticipate a collapse of the ocean conveyor belt. It also detects the transition that marks the end of the last ice age in ice core records. However, the GENIE model world reveals that an effective early warning system for the real world would require a much longer record of ocean circulation than currently exists.

Figures to include:

Figure 1: Andy Ridgwell’s schematic of GENIE

Figure 2: Millennial timescale climate projections (CO2, T, sea level ocean circulation) or an attention grabbing image

Figure 3: Two states of the ocean conveyor (bi-stability diagram GENIE-1 and -2) or a cartoon of the ocean conveyor

Sidebar details:

Project: Grid ENabled Integrated Earth system modelling (GENIE)

Website: www.genie.ac.uk

Contact details: Professor Tim Lenton, School of Environmental Sciences, University of East Anglia Tel: 01603 591414 Email: t.lenton@uea.ac.uk