The Texas Freeze: A Failure to Conceptualize Risk
Author: Ron Dembo
The big freeze now sweeping through Texas and the Southern states betrays a fundamental inability to manage the growing uncertainties of climate change.
The Fundamental Flaw with Forecasting
Using statistical modelling to forecast future events works well over short time periods. The science of meteorology, for example, has advanced significantly in the 21st century — to the point where we were able to predict the arrival of this arctic blast several weeks in advance. Better still, we can say that the likelihood of the weather patterns that would cause this can be known a year in advance.
What is broken is that we seek to plan using a definitive forecast. This is not only our grid planners but everything from our central banks to our governments to our journalists who write primarily in the language of forecasting.
Texas exemplifies this failure. Its grid operators forecast how much energy residents would need over this winter. Using those parameters, they allowed several power plants to shut down for maintenance while other equipment was left overdue for upgrades. As the cold front hit, household power usage jumped dramatically as residents struggled to keep warm — it jumped outside the range that the operators had forecast, triggering rolling blackouts. Then conditions worsened further, disabling power plants and exacerbating the disaster.
Not only was the forecast wrong, but the operators had failed to prepare for the possibility that they could make a mistake.
The Problem Is Radical Uncertainty
That climate change is happening is a verifiable fact. Carbon dioxide concentration in the atmosphere is higher than it has been for over 3 million years. By burning fossil fuels, we have emitted nearly 2.5 trillion metric tons of CO2 into the atmosphere since the Industrial Revolution began, raising concentrations by 67%. This dirty energy is cooking our planet — its surface temperature has jumped more than 1.8°F (1.0°C) over the last 115 years, and the average rate of change now is twice what it was in 1980.
But the exact manifestations of climate change are radically uncertain. The Texas snowstorm was the result of a polar vortex — a massive expanse of swirling cold air in the polar regions — breaking much further south than normal.
There is solid science suggesting that global warming causes these arctic outbursts. Rising temperatures in the Northern Hemisphere appear to be weakening the circulating jet stream that holds the vortex in place. When heat in the stratosphere punctures that stream, Arctic air spills south.
But it is almost impossible to predict exactly when this puncture will occur, how severe it will be, and how exactly it will interact with other macro and micro climate factors across the United States. There is no way to predict that an event exactly like we are experiencing now in Texas will occur next January. You might as well be throwing a dart at a rock on Mars.
What is possible is to say that, most likely, given global warming scenarios for the future, there will be more events like this. We can say that, most likely, Sacramento will experience many more hot days (temperatures above 100 degrees Fahrenheit) in the coming years. We can say that The West Coast of Canada is more likely to get dryer and the East Coast wetter as the century progresses. We know that bigger and more frequent hurricanes are creating more destructive flood events and storm surges.
We can also say, with significant certainty, that if the concentration of carbon dioxide in the atmosphere keeps on increasing as it has over the last 50 years, there will be much more volatility in the weather. But, we cannot tell for certain whether Southern England will be a tropical paradise or more like Scandinavia by the end of the century.
As Alice Hill, a climate scientist on the National Security Council during the Obama administration, states, “We are colliding with a future of extremes.”
Learning to “Risk Think” for the Future
If one thing is clear, it is that the past is no longer a guide to this uncertain future. We must alter our current mode of risk management, which depends on forecasting to predict a singular state of the world, and move to a method that can comprehend a spectrum of possible scenarios about the future. And we must strategize accordingly.
We need to change our language about the future. We need to think like “risk thinkers” when describing a future — using the language of scenarios and planning for mitigation and adaptation (hedging).
Scientists and activists frequently demonstrate that we have the capacity for this. For example, many have long since advocated for decentralizing the power system. Microgrids would enable communities to generate and store their own green energy using rooftop solar panels and wind turbines nearby. With the correct infrastructure, this would allow communities to keep power running to essential services when extreme weather events strike.
This is the correct form of a solution to an uncertain future: it doesn’t rely on a singular forecast; it hedges against multiple outcomes by building resiliency.
Ultimately, we need to accept that our view of the future is like the view through our windshield of our car on a dark night. We only see what the headlights show us. And, as we get closer to objects we see them better and can revise our driving strategy. But the radical uncertainty is whether a moose is about to run out of the darkness into our path. So how to hedge? We slow down so that in the event the poor animal runs across the road we and the animal will not be gravely hurt. This metaphor highlights the difference between drivers that speed through the night, even with poor visibility, and those that risk think. We don’t want to land up in a pile up like the 100+ car collision that happened in Texas last week.