Articles / 05.14.2015
“It’s actually an existential question. If it’s at one end of the spectrum, we can adjust, but if it’s at the other end of the spectrum, we’re not going to come out of it unless we cut emissions in the next decade.”
-Krishna Palem, a computer scientist at Rice University, arguing for urgent investment in climate change modeling
One of my dearest friends – decorated special forces operative during the Vietnam War, fearless surfer, incurable prankster, cancer survivor, farmer, contrarian investor of breathtaking conviction, and general all-around badass – built a fortune on what I’ve come to call The Quantum Theory of Closeology. In a nutshell, his now surprisingly sophisticated theory means to act on observation and information that may not be technically accurate but is close enough that, practically, functionally and directionally, it may as well be. Conveniently, his theory can be reduced into a vaguely legal, and equally vaguely Latin-sounding, acronym – QTC – that has the pleasant benefit of stifling dissent.
He applied QTC to the world of oil and gas mineral rights, believing that buying cheap, undeveloped minerals without proven reserves, but that were geographically proximate to proven plays, would eventually prove profitable as exploration, development and extraction technology would evolve. Fast forward to the fracking revolution, and you can see his point.
But what, the astute reader may ask, is a blog about tight shale formation fracturing and hydrocarbon production doing in an impact investing blog? Before this reader suffers an indignation-induced aneurysm, I would simply ask that she focus on the theory rather than the application. And the application is, in my mind, quite relevant to the discipline of climate science. More specifically, to the science of climate modeling.
But first, an important but admittedly lengthy and arcane digression…
The biggest challenge with climate change modeling is its inherent imprecision. First, it is a model of the future. And the only thing we know with absolute certainty about models is that they are wrong. And, scientists being generally honest and scrupulous folk, they usually lead with some sort of disclaimer to this end, which gives the community of climate change skeptics all the room they need to cry “FOUL!!” and what would likely be an engaging debate instantly becomes a politicized, polarized argument. Meanwhile, the globe warms and science grows ever-more confident that man’s activities play a role.
But beyond the imprecision lies an even greater challenge, one that few scientists will admit, and few laypeople will understand. Put simply, we don’t yet have the computational capability to model the climate. Full stop.
To put the scale of the challenge in perspective, it is presumed that accurately modeling the future of our planet’s climate would require running an absolutely gargantuan set of calculations, for weeks on end. Just to get one model. Then, to do the sort of scenario planning that policymakers require (for example, changing the amount of carbon produced by The Netherlands through the year 2025) would require running the model again with different assumptions. Wash, rinse, repeat. For years.
A machine capable of crunching like this would need to be so much faster and more powerful than anything we currently have that it is mind-bending to contemplate. And would use more energy and generate more heat than any computer in use today. A lot more. Such a machine, were it to exist, might consume electricity equivalent to lighting, heating and cooling every single home in Reno, and might cost upwards of $20 million to operate. Annually. In this scenario, a computer built to model our planet’s climate might actually be a contributor to climate change all by itself. Bravo for life’s little ironies.
The point is that arguing about incredibly tiny details, while seductively compelling in this era of Big Data and the attendant arrogance that accompanies it is simply not critical. As my lead-in quote suggests, what is important is that we identify where we are on the spectrum of change, and what behavior is required by that placement. After all (and I sincerely hope that my deep environmentalist friends don’t shun me for the pragmatism that this comment will reveal), it actually does matter – a lot – if we are already too late, a decade from being too late, or 50 years from being too late. This will drive policy, economics, human development and migration, etc.
Seen through this lens, it matters less whether the oceans will rise 1.5m or 1.6 m (although if one’s home is located 1.6m above current sea level, this would matter a great deal) than it does whether they will rise at all, or how quickly.
And so, back to Dr. Palem. “Scientific calculations like weather and climate modeling are generally, inherently inexact,” Dr. Palem said. “We’ve shown that using inexact computation techniques need not degrade the quality of the weather-climate simulation.” In other words, intentionally embrace imprecision. Get the direction right. Employ QTC.
And, most importantly, convince the world’s leaders that inaction due to imprecision is no longer acceptable. I guess I’ll need to change my decision tree blog as well as our most recent blog on divestment.
Your hoping-we-aren’t-too-late scrivener,