In the early 1800s, shortly after the start of the industrial era, carbon dioxide began to skyrocket, while temperature remained flat. Temperature began to spike only about 30 years ago. In contrast, temperature changes preceded carbon dioxide changes at Vostok until the beginning of the industrial era. Scientists believe these natural changes in carbon dioxide were a feedback response to initial, small changes in temperature, and that those changes – along with other responses – amplified the original temperature shifts.

The other responses included changes in the levels of other greenhouse gases, primarily methane; changes in the area covered by polar ice sheets and sea ice, which reflect sunlight back into space and cool the planet; changes in the levels of dust and airborne aerosols, which also cool by reflecting sunlight (the “parasol effect”); and changes in the mix of grassland, desert, and forest, which affect the reflectivity of the land.

The history of these factors is known. Besides the information about greenhouse-gas levels from the trapped air bubbles at Vostok, a sediment core from the bottom of the Red Sea indicates changes in sea level, which in turn give an approximation of ice sheet area. (The ice sheets grew and thereby drained the oceans during cold times; they melted to refill them during warm times.)

Using these and other geological records, Hansen can calculate Earth’s temperature at any given time in the past 420,000 years. He plugs in the data for greenhouse levels, sea level, and so on to produce a temperature estimate for the corresponding time. And as he showed his audience last December, these calculations match temperatures as recorded by the deuterium and hydrogen in Vostok’s ice quite precisely over the entire 420,000-year span.

Global-warming deniers like to complain that scientists base their predictions on faulty computer models. But Hansen’s calculations show that we don’t need a computer to know how temperature will respond to a given change in the greenhouse – or a change in dustiness, or forest cover, or the amount of ice on the Arctic Ocean. Solid geological field data give us everything we need – and provide a check for computer models. And lend credibility to Hansen’s predictions.

Besides demonstrating his firm grasp of the power of these various factors to change temperatures, this remarkable matching of theory to real-world data also tells us just how ornery the climate beast may be: the orbital changes that paced the ice ages were incredibly small. They had little effect on the total amount of sunlight reaching Earth in a single year – only its distribution over seasons and latitudes. Nevertheless, these minute redistributions led to swings in temperature of about 5 ºC and variations in sea level of more than 100 meters.

Greenhouse-gas levels, on the other hand, are more like a knob controlling the brightness of the sun. And the turning up of the rheostat that humanity has accomplished by adding about a trillion tons of carbon dioxide to the atmosphere thus far in the industrial era dwarfs the redistributions in sunlight that once switched the planet back and forth between glacials and interglacials. We are poking the climate beast in a way it has not been poked in the entire era of cyclical ice ages – at least two million years. As Hansen told his audience last December, “Humans now control global climate, for better or worse.”