The science of weather forecasting has changed beyond recognition, writes Jim Holland, Regional Director: Africa, Lenovo.In fact, today, thanks to supercomputers, it’s possible to make much more accurate predictions about the weather. But, how do these supercomputers make their predictions?
First, it makes sense to explain what a supercomputer is. Put simply, it is a large array of smaller computers and processing equipment aggregated to make one large, smart and very powerful computer.
Frequently found in the science, engineering, and business sectors, they can reduce the time taken to solve problems to days rather than months and they’re really demonstrating their mettle when it comes to weather predictions and climate modelling.
However, for supercomputers to make weather predictions, they need to obtain data from somewhere and this is proving to be a struggle on our continent.
In Africa, according to The Washington Post, we have just one-eighth the minimum density of weather stations recommended by the World Meteorological Organisation, which equates to a problematic lack of data about dozens of countries that are the most vulnerable to climate change.
This lack of data has meant imprecise forecasts and poor early-warning systems for people experiencing deadly cyclones, lengthy droughts and powerful floods. Researchers say the lack of data has also led to challenges in measuring the extent of climate change, making it difficult to prove global warming’s impact on the continent.
Usual data sources could include satellites, weather stations and balloons, delivering anything between 500 gigabytes and one terabyte. To put this into context, around 130,000 digital photos would require one terabyte of space – almost 400 photos every day for a year. But before the data can be used, it must be put through a process of quality control.
Once that process has been completed, mathematical models are then used to make forecasts. Known since the 19th century, these are equations that describe the state, motion, and time evolution of various atmospheric parameters such as wind and temperature.
The good news for the continent is that scientists are working on addressing our data issue.
In Narok, Kenya, for example, climate scientists, local meteorologists and farmers are working on a potential solution that could be critical to increasing data collection on the continent.
The idea is to install simple, relatively inexpensive weather stations like the one at Ole Tipis, a boarding school just outside of the town of Narok, across the continent. The Old Tipis station is one of 115 in Kenya run by the Trans-African Hydro-Meteorological Observatory (TAHMO), which has a network of 626 stations in 20 countries. If they get this right, it could positively impact Africa’s participation in understanding climate challenges as well as its ability to prepare for them.
Massive compute power
Turning data equations into accurate forecasts requires an additional factor – compute power. To understand how this works in practice, it makes sense to use a simple illustration. If the United States were divided into a mesh of 10km blocks, then a certain level of compute power would be needed to provide localised forecasts inside each block. The difficulty arises, however, when the size of the blocks is reduced. Thunderstorms, tornados and smaller scale effects are very much linked to local weather, and with a large mesh it’s easy to miss them. It’s similar to being a fisherman – a much denser net is needed to catch small fish.
Because of the enormous amounts of compute power involved in making these calculations, scientists are now looking at how other technologies like artificial intelligence can improve forecasting.
Instead of using brute-force computation to forecast weather based on present conditions, AI systems review data from the past and develop their own understanding of how weather conditions evolve. And they are already having a significant impact on forecasting. For example, the UK’s Meteorological Office recently carried out a pilot of AI technology to predict flash floods and storms.
Using radar maps from 2016 to 2018, the system was able to accurately predict patterns of rainfall in 2019 for 89% of cases. Advancements in technology mean its four-day forecast is now as accurate as its one-day forecast was 30 years ago.
Bumping up against the Butterfly Effect
New technologies will undoubtedly usher in an era of more accurate forecasting, but they will never be able to make long-term predictions about the weather with 100% accuracy. That is because the equations that are used to make weather forecasts are non-linear – they have a degree of chaos embedded in them.
As early as the 1960s, Edward Lorenz, an MIT meteorologist, was arguing that it was fundamentally impossible to predict the weather beyond ten days. Central to his argument – which later became known as chaos theory – was the claim that small differences in a dynamic system like the atmosphere could trigger completely unpredictable results. The most famous formulation of this theory was Lorenz’s 1972 academic paper ‘Predictability: Does the Flap of a Butterfly’s Wings in Brazil Set Off a Tornado in Texas?’
Setting aside chaos theory, there’s another reason why forecasting may take time to become more accurate – the science itself. Although compute power doubles every two years or so, weather science takes longer to catch up. Supercomputers were first used in the US in the 1960s and 1970s, but it took between ten and 20 years for forecasts to become much more accurate.
Still, the compute power that is now available has massively improved forecasting. When weather predictions were first made in the 1950s, the results were highly inaccurate because of the limited computational power available. To give an example of how things have moved on, a weather model that would have taken 600 years to run on computer systems in the 1960s now takes just 15 minutes on a standard Lenovo ThinkSystem server.
There is every reason to believe that as compute power increases in the next few years alongside our scientific knowledge of weather patterns, it will be possible to make even more accurate predictions. And with the ability to predict extreme weather, supercomputers have the power to save lives and make a profound impact on the world.