An interpreter at high-stakes negotiations picks up not just the words but also the tone of voice. The ocean sends similar signals: warm patches on the surface, shifts in the wind. PTSTnet is a neural network that listens to these signals and predicts what El Niño will look like two years from now.
The model takes a year of data: water temperature, subsurface heat, and winds. Then it step by step projects the future, relying on its own previous forecasts. Inside, it embeds simple physical rules, such as how heat spreads through water. So it doesn't just memorize numbers; it understands why the ocean behaves the way it does.
When the model makes a mistake, its errors drift from the western Pacific toward the east, mirroring how in reality winds strengthen and cold water upwells off the coast of South America. This indicates that PTSTnet has learned the true mechanisms of the climate. Now farmers, insurance companies, and city services can prepare for droughts or floods well before they arrive.
The model was able to predict the 2015–2016 super El Niño, initializing as early as February 2015, when most dynamical models gave large uncertainty.
In Kim Stanley Robinson's novel 'Forty Signs of Rain', scientists struggle for accurate prediction of climate shifts; PTSTnet resembles a step toward such a system, where physics and data merge into a transparent tool.