Why Paleoclimate Modelling?

I am interested in paleoclimate modelling. What does that mean and why should anyone attempt it?

There are a few things I need to discuss before I attempt to answer that question. 

• The concept of "climate" is like the air we breathe. We already have systems in place which depend on it, but as long as it is everywhere and fresh, and our systems work smoothly, we are not really required to give much thought to it. However, when trouble arises, we are required to develop an understanding of it, however alien and complicated this task might seem.
  
  
  
• "Climate" does not mean "weather"; a climatologist and a meteorologist are not the same, nor can they possibly be. These are two different trades requiring different backgrounds, skills and motivations. Weather is short-term, climate is long-term. We would be wasting our time by mistaking the two for each other.
  
  
  
• Climate models are useful and a good thing. Since unfamiliarity can also breed contempt, people often imagine climate models as black boxes/crystal balls used for trying to scare society. If you don't understand something, it is natural and in fact partly rational that you want nothing to do with it; and if that something is giving bad news, it is human to want to disbelieve it. At their core, climate models are simply sets of equations a computer solves. These equations encapsulate what is known about the atmosphere, oceans, land etc. The result is numbers about things that we care about, including climate variables like temperature and precipitation. A little more familiarity can be gained through this excellent article by Carbon Brief.
  
  
  
• Climate models are a necessary thing. As the scientists Knutson and Tuleya put it, "..if we had observations of the future, we obviously would trust them more than models, but unfortunately observations of the future are not available at this time." Simply put, if one is at all interested in planning for the future, and is wise enough to not disregard the climate in this planning, one has to rely on climate modelling. There is no other way, and this is independent of the quality of our current models.

Now, having developed some vague idea that the climate deserves studying and climate models are a useful tool for this study, we can try to understand the point of paleoclimate modelling.

"Paleo" refers to the past. Paleontology is the study of past life, paleoecology is the study of past ecosystems, paleobotany is the study of past plants, paleoanthropology is the study of past humans. (Similarly, there are paleobiology, paleozoology, paleoentomology, paleogeography, paleogeomorphology -- you name it.) The past can be studied on various timescales. The Paleozoic Era (a geological time unit -- 541-252 million years ago) was the time of ancient animals. The Paleocene Epoch (another geological time unit -- 66-56 million years ago) was the time before the Eocene (yet another geological time unit). The Paleolithic Period (an archaeological time unit -- 3.3 million to 12 thousand years ago) was the time of old and simple stone tools. A paleo diet mimics the diet of past humans. Moving forward, we shall stick to Paleoclimatology, or the study of past climates.

The idea that past climates deserve to be studied has another idea implicitly embedded in it -- that past climates were different from present climate. It is true. Our planet has been a completely different world in its past. If you don't believe the science, perhaps you believe Hollywood. Dinosaurs roamed the Earth during the Cretaceous, which was so warm that there were forests in the South Pole. Mammoths roamed the Earth during the Quaternary, some parts of which were so cold that New York was buried under an ice sheet. The central idea is that we cannot take our current climate for granted, that the climate depends on a lot of other things (external forcings) and is certainly not a boring, stationary thing (internal variability). If climate is known to change -- sometimes drastically, sometimes regularly -- hopefully the change can be studied, understood, anticipated and perhaps even managed (and survived).

Paleoclimate modelling attempts to model the climates of the past. In doing so, it takes climate models outside the "familiar" realm of the last one or two centuries and delves into the geological past -- in the order of hundreds, thousands, even millions of years ago. It asks awkward questions like "Can our climate models detect climate change of the past?" before we assume an answer to "Can our climate models detect climate change of the future?". In trying to answer these awkward questions, paleoclimate modellers use the same climate models that are widely used to study the climate of the present and project the climate of the future. Thus, there are no specific "paleoclimate models", forcing us to aspire to the very ambitious principle of understanding the climate so well that we can recognize all facets of its personality. The hope is that trying to model every chapter of the Earth's geological history will teach us something useful either about the climate or about our way of modelling it.

How do paleoclimate modellers go about asking these questions? How do they go about answering them? And have they learnt anything useful yet? -- I will be exploring these questions in subsequent blog posts. Till then, suggested reading: 

• How do climate models work? - Carbon Brief
• Climate Change in the Context of Paleoclimate - NCEP, USA

=====

If you like such posts, you may subscribe here.