In the Colorado mountains, snowfall varies from winter to winter. The reason is not just local weather but also the distant Pacific Ocean. When water in the tropical ocean warms, the amount of snow in the mountains changes. This link between distant regions is called a teleconnection. Usually it is described through known indices, such as El Niño, but the appropriate index is not always known. The new model does not pick an index; instead, it directly uses water temperatures across the entire ocean surface and finds on its own which areas matter for Colorado snow.
How does it work? The ocean is like an orchestra, where each patch of water plays its own note. The mountains hear not individual notes but the overall melody. A typical model asks you to name a single conductor who directs everything, but such a conductor may not exist. The new model listens to the whole orchestra and picks out the main melodies on its own. One such melody the model found automatically: warm water in the central and eastern ocean, cold water in the west. This melody explains 30% of all variability in water temperatures and matches the El Niño pattern. The model did not know about El Niño in advance; it discovered it by itself.
Previously, to account for the ocean's influence, one had to choose an index in advance, such as El Niño. But such an index might not suit a particular region. Now the model finds the relevant ocean areas from the data on its own. This helps better predict how much snow will fall in winter, which is important for water supplies. This approach can also be applied to other regions where weather depends on distant oceans.
The empirical orthogonal function extracted from Pacific sea surface temperatures explains 30% of the total variability and closely mirrors the El Niño pattern: warm water in the central and eastern Pacific, cool water in the west.