Our climate by its very nature is variable, over the course of days, weeks, seasons and beyond. Some of these embedded variations can occur over very long periods of time in specific regions and have an influence on the global climate.

The tropical and South Pacific Ocean regions are home to a number of long-term natural variations of this type. Two of the more influential variations in this region include:

  • The El Niño Southern Oscillation – which produces La Niña and El Niño events that bring prolonged dryness or rain to many parts of the world – and
  • The South Pacific Decadal Oscillation which, as its name suggests, produces a long-term influence arising from the Pacific Ocean that can affect the climate across the world for years to decades at a time.

Because of their importance, researchers use complex computationally demanding climate models to understand these two variations and help forecast future changes.

CLEX researchers and colleagues explored whether a simpler, less demanding model could achieve results that were similar enough to these more complex models to achieve useful results. To do this, they used a simple model – called a simple linear inverse model – and included the processes and variations found in the upper part of the Pacific Ocean.

They found that the inclusion of upper South Pacific Ocean variability significantly improved the predictions of those modes in simple models. This suggests processes just below the surface of the South Pacific Ocean maintain and enhance these big and low-frequency variations in a way that potentially makes them predictable over longer timescales.

This has implications for our capacity to forecast these events further ahead than we currently can, while using computer models that demand less time and computational expense.

  • Paper: Lou J, TJ O’Kane and NJ Holbrook, 2020: A linear inverse model of Tropical and South Pacific seasonal predictability. Journal of Climate, 33, 4537-4554, doi:10.1175/JCLI-D-19-0548.1.