Assessing the tropical climate over the last glacial/interglacial

The world’s tropics are a region of paramount importance. They occupy around half of the Earth’s surface and nearly 40% of the world’s population live within them. This obviously means that any changes in the climate here will have a large and direct impact. However, the effect of the tropics can be felt on the climate well outside its immediate area. This effect is most noticeable in the climate changes attributed to El Niño. Here in the UK, this tropical Pacific phenomenon has entered the popular lexicon as one the biggest sources of variability in our own seasonal climate. Although we have a pretty good understanding of why the tropics behave the way they do today, when we look back to the past things are less clear. Although there are numerous observations, especially over the last glacial cycle (the last 120 thousand years), showing that the tropical climate and the El Niño variability were quite different in the past, we still can’t answer the question: why?

Our current lack of understanding is the result of the overwhelming complexity of the problem. In the tropics the behaviour of the ocean and the atmosphere are so intimately related on all timescales, from thousands of years to month to month variability, that a full complexity climate model capable of resolving these interactions must be used. Of course such a model requires considerable computing resources. These are available when looking at changes over a few tens to hundreds of years, but become so large as to be prohibitive, if we wish to study changes over thousands of years. However, one full complexity climate model has been run for many thousands of years and it is this that is the basis for this study. Initially performed in collaboration with the BBC for the series “The Incredible Human Journey” we have approximately 400,000 years of climate model data that will allow us to answer the fundamental question: what caused the tropical climate to change over the 120 thousand years of the last glacial/interglacial? 



The figure shows the sea surface temperature in the tropical Pacific Ocean and the depth of the thermocline which is the boundary between the warm surface waters and the colder deep water. We show the temperature at the present (0kyr) and 104,000 years ago (104 kyr).In the east Pacific we see the “Cold Tongue”, which is the area of cold water that extends along the equator from the coast of Peru. This “Cold Tongue” is the result of the cold deep water being closer to the surface, and you can see that the thermocline is closer to the surface here. In the west Pacific we see the “Warm Pool”, the region of the warmest sea surface temperatures. This warm water is a very deep layer, and we need to get to a depth of nearly 200m before we cross the thermocline into the colder deep waters. The surface temperature is up to 2°C cooler 104,000 years ago – we will try and answer: why?


As the model runs have already been completed a significant amount of time and headaches will be saved, allowing us the time to take a long, deep look into the model. We will use a set of analysis tools to pick apart the causes and effects of different mechanisms that we think altered the tropical Pacific climate and the variability in El Niño over the last glacial/interglacial. Using different climate model configurations, each of which highlights different forcing elements, we will begin to understand which was the most important, especially as we compare what the model tells us happened with what the observations tell us actually happened.

By the end of our study we will have gained a far deeper understanding of the mechanisms that cause the tropical climate to vary. This will not only give insights into the workings of the climate but also give us some clues as to what we might expect in the future.

Professor Paul Valdes and Dr William Roberts
University of Bristol

Paul was awarded a Research Project Grant in June 2012.