Why we love supermodels
Monday’s highlight
The trouble with variables is that they’re so, well, variable. The trouble with the real world is that it’s full of variables. Scientists have spent decades refining ways of peeling back the layers, stripping a system down to its bare minimum. Doing so allows them to find out what makes something tick. But that’s all changing now.
Synergy is an oft-heard term at this year’s conference. Small impacts from different sources combine to create a whammy that no simple model could predict. The problem in trying to take this into account is that by increasing their complexity, models can become so complicated as to be meaningless…
Synergistic effects have the unfortunate characteristic that they rarely work against each other. Multiple insults don’t cancel each other out. Rather, it’s more like dealing with a sustained rain of blows: even if getting hit by a left would straighten you out after taking a right, the chances that you would be happy with your lot are slim.
For corals, the ecological blows are so different that it’s less like being hit by a left-right combination than being pummeled and then set on fire.
Coral reef scientists face a dilemma. Keep plugging away with those single variable, nicely predictive models, or live in the real world. If the Great Unwashed are believe the science, and they must if coral reef conservation is to succeed, they have to be fed data that reflect the same coral reefs they snorkel, fish and dredge amongst. For the scientists, that perhaps means going against the grain and building models and experimental systems capable of exploring the effects of more than one variable.
Within the conference climate change symposium, voices for this multivariate approach are being heard.
Josh Madin’s study of water velocity and acidification is a case in point. Increased water acidity affects the strength of coral structures, making them vulnerable to the impact of the more frequent storm lashings predicted by the latest climate change models.
He started with a simple model of how storm surges affect the structure of coral formations, comparing lumpy, middling and big branchy forms. Unsurprisingly, more elaborate forms were more susceptible to the weather, suffering devastating levels of damage with each passing hurricane. However, when acidification was factored in to the mathematical model – no flimsy simulations here, it’s pure maths analysis – the already weakened skeletons went on to become… dead skeletons.
The real world isn’t simple. Nor should models of coral reef survival be. They don’t face stresses in isolation, and it’s becoming clear that we won’t understand how they will deal with multiple stresses until we start modeling them.
Talks throughout the session in which Josh talked have given an inspiring preview of the way things can be. By moving the maths out into the real world, we stand a chance of holding onto it.
Written with SeaWeb’s Christian Reilly
Leave a Reply