The countries of the Horn of Africa are suffering the full brunt of climate change. Devastating floods or droughts are jeopardizing the lives of the populations. Shruti Nath, a climatologist at the Department of Physics at the University of Oxford in England, believes that artificial intelligence could help better predict extreme weather conditions and save lives in the region.

Using satellite data on cloud banks and cloud-top temperatures, the AI makes predictions and tries to identify upcoming extreme weather conditions.

“For the AI, there is a part where it predicts. But to predict, it must first be trained. And we train it using historical observational data. The input data for our AI system are weather forecasts. The AI takes these forecasts and trains itself to match them with the observed reality based on cutting-edge satellite observations and the station data we have,” she explains.

Like all AI models, this one is constantly improving. The code rewards accurate predictions and penalizes inaccurate ones.

"So, as the training progresses, the model learns more and more, and it is rewarded when it provides accurate predictions that match the observed reality. If it doesn't, it is penalized."

In countries like the United Kingdom, supercomputers are used to predict the weather. For example, the Met Office supercomputer performs 16,000 trillion calculations per second.

The cost of these supercomputers, data collection stations, and radar banks that support them is high, and they are not available in developing countries.

In contrast, the AI prediction code produced by the Department of Physics at the University of Oxford can be used on a laptop.

“That’s the beauty of it, because the computation is so affordable that we literally have people running it on their laptops. So you can have this model, and once it’s trained, you just need to run it on your laptop to produce 50 predictions about all possible weather outcomes, which, as you said, would otherwise require banks and banks of supercomputers,” explains Nath.

The project is still in its pilot phase, but the results are impressive according to those on the ground in the Horn of Africa.

Isaac Obai, head of food systems at the World Food Programme, is in the United Kingdom for the project’s biannual meeting.

He states: “If nothing is done to disseminate information or early warning messages about these extreme weather conditions, we will see more people becoming vulnerable, more people exposed to these extreme weather conditions, and we will have many lives lost, many lives affected, as well as livelihoods. In the end, the severity of the weather conditions intensifies and their frequency increases. And I think that due to climate change, they are also becoming more aggressive. If nothing is done, many people will be affected, lives will be lost, livelihoods will be impacted, and the level of poverty will increase significantly.”

It is notoriously difficult to predict the weather in this region of Africa due to its changing nature and the lack of weather stations that observe and record data.

Thanks to the 48-hour alert provided by the AI weather forecasts, people in danger can be warned via text messages, emails, and even radio and television broadcasts. Maslin Gudoshava is a meteorologist at IQPAC (African Regional Forecasting Organization).

“We are convinced that AI models can really improve the prediction system. Not only in the short term, but also in terms of the short-term timeframe. By short term, I mean a few days, but also the entire season. If we could really achieve this, I believe it could significantly enhance our early warning systems,” she says.

The pilot program is currently being implemented in Kenya and Ethiopia, but there are plans to expand it throughout the region. If it proves successful, it could be used in other parts of the world where extreme weather conditions, due to climate change, are devastating.

The program is the result of a collaboration between the World Food Programme, the Department of Physics at the University of Oxford, the Kenya Meteorological Department, and IQPAC (the regional meteorological service).

Article source: africanews