Scientists want to simulate different climate conditions to help prevent real risks to life on our planet. We all remember from primary school; this is the water cycle. Even if you don’t remember exactly what you learned, you probably have memories of a video or diagram that was bright and creative enough to make the day a little more fun. This is hard for some of us (myself included) to forget. Even in the adult world, it seems neat and easy to think of the three components of the water cycle as a kind of rinse-and-repeat pun: evaporation, condensation, and precipitation.

Our planet cannot survive without water, which makes up 71% of our planet’s entire surface, and the oceans contain approximately 97% of this figure. However, when you add ongoing climate change from human activities such as burning coal, as well as other daily impacts we create as a society, the process can become quite complex. As a forecaster, I know the challenges we face in predicting water-related disasters, including floods, landslides, and droughts.

However, in order for scientists to make the most accurate predictions and better understand how this cycle works, we need to obtain and examine models that include as much high-resolution data as possible. Ideally, this data should also cover every inch of the planet, from the highest mountain peaks to water buried deep within the earth.

And thanks to funding from the European Space Agency, scientists are building just that: a digital replica of Earth and its wondrous waters to explore.

“High-resolution Earth modeling is very difficult, so basically the idea is to focus on a specific target first.” — he said in a statement Luca Brocka of the Italian National Research Council. “This is the idea behind what we are developing; digital twin case studies of the Earth’s water cycle in the Mediterranean basin. Our goal is to create a system that allows non-experts, including decision makers and citizens, to run interactive simulations.”

Brocka, lead author of a detailed paper on the research, worked with colleagues to create a digital doppelganger. Thanks to this model, scientists can constantly input new data to simulate best and worst-case scenarios of natural disasters in different environments of our planet. For example, by simulating a landslide, landslide-related risks and conditions can be monitored as if they were happening in real time. This can help you prepare for potentially devastating events in the future based on what is learned from each test.

So how are these models created?

It took scientists a lot of work to collect as much satellite data as possible, collected through numerous observations made on Earth. They then combined measurements of soil moisture, precipitation, snow depth, evaporation and river flow taken at specific time intervals to paint a clear picture of the dynamics of variables on the planet. The model’s high-resolution data can then be used as an interactive tool for scientists.

“This project is a great example of the synergy between advanced satellite missions and the scientific community,” said Brocka. said. “Such collaboration, combined with investments in computer infrastructure, will be critical in managing the impacts of climate change and other human impacts.”

As with any model, it takes practice to make it perfect. But you have to start somewhere.

Brocka and his colleagues initially used a digital twin to model the Po River valley in northern Italy and other parts of the Mediterranean basin; They plan to create similar models across Europe in the future before collaborating with scientists from other continents. The main goal of this project is to help predict where floods and landslides are likely and to learn how to better manage water resources.

“We have to start with the things we know very well,” Brocka said. “The Po Valley is very complex; we have the Alps, we have the snow, which is difficult to simulate, especially in rugged and harsh terrain such as mountains. Then there is the valley, where all human activities take place (industry, irrigation). Then there is the river and extreme events: floods, droughts. More Then we moved to the Mediterranean, which is a good place to study the extremes of both too much and too little water.”

While the team’s modeling focuses on a larger-scale region, there are plans to consider more localized studies as well. But for now, scientists continue to focus on the biggest and most persistent problems in their mechanisms. For example, the complex algorithms they develop will need maintenance as large amounts of data continue to be added; They also say more ground-based observations are needed to continue validating the satellite data they use.

Likewise, to combat the uncertainties that can arise when using satellite data, Brocka hopes to incorporate artificial intelligence into its program to iron out some of the kinks. Artificial intelligence, if trained really well, can act as an extra pair of eyes, in a sense. As we see in the use of AI in weather models such as wildfire forecasts, the benefits of such collaboration include minimizing errors that can sometimes occur due to changes in atmospheric conditions when capturing images. Applying AI can also save time by allowing human engineers to focus on other problem areas.

“The collaborative efforts of scientists, space agencies, and policymakers promise a future where digital twin Earths for hydrology will provide invaluable information on sustainable water management and resilience to disasters,” Brocka said. said.