If you don’t think long range forecasting is hard, ask a meteorologist. Or a climatologist. By some of the old predictions, parts of New York City, California, and the Netherlands were supposed to be underwater by 2010. Obviously that didn’t happen, and that gets into more than just climate models. Predicting the effects of forecast temperature is no less tricky. If the average global temperature rises by, say, 2º C, what does that mean? If only there was some way to take out the guesswork, some hard data we could look at and know exactly what will happen if temperature rises to a certain point.

Fortunately, there is. The earth’s climate has varied in the past, and the study of it is known as paleoclimatology. Paleoclimatologists cast a wide net to determine what the climate was like in the past, including written accounts ;tree rings; pollen and seeds in sediment and dig sites; the remains of algae, insects, and animals; and remnants of seashores. Really, it’s limited only by the cleverness of the scientist. This is good because written records can be sketchy and only go back so far, and climate variation goes all the way back to when the forecast was for lava with a chance of scattered meteors.

Of course, for predicting the effects of climate change,  we’re more interested in recent times, geologically speaking, when continents were already where they are now, and the critters were recognizable, excluding the megafauna. Here paleoclimatologists have us covered, with more data coming in all the time. How that helps climate prediction is that we can look at how things were at such and such a temperature,  then know what will happen if we see those temperatures again.

Let’s say at one location there’s a lot of grass pollen in sediment when temperatures were at a certain level.  If those temperatures return, you can look for conditions that support grasslands. Or let’s say there’s ancient barrier islands inland, and you know they date from such-and-such a time. If the area’s been geologically stable for all those years and you know what the temperature was when this was the shoreline, it’s a pretty good bet that the seas will rise to that point if temperatures get to those levels again.

Dovetail this with other data, such as archeology and history, and you can even see how changes affected people. The Medieval Warm Period was a good deal for Europe and North America. The southwestern part of North America had enough problems that a lot of people packed up and left, but the Medieval Warm Period saw the rise of the Mound Builders. When the Little Ice Age hit, Europe saw crop failures, famine, war, and disease, while in North America the Mound Builder culture collapsed.

There’s been big changes on both continents since then, and temperature shifts would affect people differently. Yet, like looking at flood records to know what to expect when the river crests at a certain point, paleoclimate data can tell us what to expect in terms of rainfall and what plants thrive and what plants won’t if temperatures change.

What comes out of this might not be as sensational as the tabloid style predictions of the past, but they’ll be grounded in hard data. And if the climate predictions turn out to be wrong and the world cools, paleoclimatology shows us what to expect from that, too.

Even science fiction writers can benefit from a little paleoclimatology research. Obviously Waterworld was overblown, but what about a story that has the seas rising to a certain level, and such-and-such local conditions? Data about past climate effects can help a writer create an accurate world.

It’s also a good way to calibrate the skept-o-meter. Such as various predictions that land high and dry in 2015 would have been underwater by now, Was it the last time temperatures reached the predicted level? If not, it won’t if we see those temperatures again.

Maybe they should have asked a paleoclimatologist first.