2009-10-23 - 2:03 p.m.

...how science is art to me....

It's Friday, and I'm almost finished with an incredibly tedious task that I need to complete before I can submit a paper. Interestingly enough, I'm completely enjoying the tedium. It's exactly the kind of tedium that lulls me into a very relaxed, scientific rhythm.

I scroll through hundreds of papers, dutifully reading through their methods sections, pulling out information about how they determined the age of the sediments, and then providing a code for their records.

I know that there are people on this planet who would prefer shaving their heads with a cheese grater to this work. I just love it. I get to read all of these papers from all over the world, and I get to organize these data on fossil plants and animals, and put them into their proper place in a data table, assigning a number to their change. 1 means the climate was wetter. -1 means drier. 0 means it hasn't changed.

After scrolling through hundreds of papers, and disqualifying the data that do not meet a quality standard, I take all these number codes and convert them into a beautifully colored map of how things looked 20,000 years ago. The chaos of 1s and 0s and -1s turn into climate patterns that mean something (hopefully). South America was wetter; Australia was drier.

And then I compare the patterns to our model simulations of the climate from the same time period, and we sit down and think about where the climate model is right, and where it is wrong, and why. What is controlling the patterns? what makes them change? If the ocean gets colder, does that make south America wetter? If we match the climate patterns in the data, does that mean our simulation is right, and then does that mean we can place another piece in the puzzle that explains how our climate became an ice age 20,000 years ago, and how our atmosphere's CO2 levels were 50% lower than they are today?

What I like about this work is that I don't know - and don't really care - what the answer is before I code. I go through these papers blindly, asking the presence or absence of pollen to tell me something. I'm not forcing something to be a 1 or a 0. I wait until the very end, when the 1s and 0s are assembled like a impressionist painting, and then for those patterns to be revealed. Yes, it's tedious. In the same way that pointillism is.