Singapore developing climate model to localise findings of IPCC report

Localised projections are important as different regions will experience climate change differently. PHOTO: ST FILE

SINGAPORE - Scientists in Singapore are working on a "crystal ball" to help fine-tune and localise for the region the global climate projections in the latest report of the Intergovernmental Panel on Climate Change (IPCC).

This work on the regional climate model is being led by the Centre for Climate Research Singapore, a unit under the National Environment Agency's Meteorological Service Singapore (MSS).

"The centre has started work on the third national climate change study for Singapore, called V3, which provides localised and high-resolution climate projections derived from the latest climate models used by the IPCC," said an MSS spokesman.

Localised projections are important as different regions will experience climate change differently.

For example, even though the IPCC said global warming would cause more heat waves and higher temperatures in general, the extent to which each region will be affected varies.

The Arctic is expected to warm up twice as fast as the global average, since the loss of reflective sea ice there would worsen the warming trend, as darker land and ocean surfaces absorb more heat. As for Singapore, the heat problem is exacerbated by its highly urbanised environment since concrete traps heat.

With projections tailored to the local context, governments can respond better to the climate change threat - such as deciding how much to invest to improve drainage systems to withstand bouts of heavier rain, or determining the height of seawalls to keep out the rising tides.

Still, the IPCC was clear on one thing: No matter what the symptoms experienced by each locality are, they will get worse with every degree of warming.

The Straits Times explains the science of climate change projections.

How does the IPCC come up with climate projections?

Climate projections used in the IPCC reports are made by global climate models - a climate scientist's "crystal ball" for seeing into the future.

These models simulate the physics, chemistry and biogeochemistry of the atmosphere, land and oceans in great detail.

In such models, the earth's surface is represented by a series of tiny grid cells.

There are multiple layers of grid cells to represent the depths of the oceans and the heights of the atmosphere. Visually, then, a climate model resembles a globe made out of Lego bricks.

Mathematical equations are used to characterise how energy and matter interact in different parts of the ocean, atmosphere and land, and how energy from one grid cell flows to the next.

This allows scientists to model how the earth system responds to changes in say, the amount of planet-warming emissions that mankind release into the atmosphere.

Based on these changes, the models project how parameters such as temperature, rainfall, sea levels, or tropical cyclone activity will be affected.

Why don't the global models provide more localised projections?

This boils down to the resolution of the model, which is defined by the size of each grid cell.

Similar to how smaller "nano bricks" can build out more detailed 3D models than larger Lego bricks, smaller grid cells allow higher levels of detail in the model.

Models with more grid cells need more computing power.

In global models, each grid cell usually spans between 70km and 250km. These are useful for studying global climate and large-scale drivers and processes.

But Singapore is only about 45km across at its widest.

As a result, the Republic appears as just one grid point or not at all in these models.

With just one grid point or less, it is not possible to model certain localised processes, such as storms that develop within Singapore island.

Remote video URL

How is Singapore planning to fine-tune these projections?

The Straits Times had earlier reported that the Centre for Climate Research Singapore is working with the National Supercomputing Centre to downscale the global climate models to produce grid cells spanning from about 2km to 8km.

With more grid cells covering Singapore, the model would aid researchers in coming up with finer, more accurate predictions of how the island will experience climate change in the future.

But even as the work to build the regional model is ongoing, observational data on variables such as rainfall and temperature and sea level changes will continue to be collected.

This information is crucial to seeing how well the models simulate reality.

When a climate model is built, researchers can use it to go back in time, introduce changes that have already been seen, and see whether the output from the model corresponds with the observations of today.

Sources: Carbon Brief, ST File

Join ST's Telegram channel and get the latest breaking news delivered to you.