Read an Excerpt
CHAPTER 1
In Uncertain Terms
Look–Listen–Respond. That's the triptych of instructions provided on most In Case of Emergency signs. The instructions are clear and terse, but the signs assume we know what constitutes an emergency. In the case of climate change and other hazards that hit us obliquely, it can be particularly difficult to discern whether this is, in fact, an emergency. Of course, most In Case of Emergency signs are made for situations like fires, subway accidents, or plane crashes, situations in which declaring an emergency might be more straightforward. Declaring an emergency because of potentially dangerous changes in the environment is often not as clear.
Unlike disasters that slap you in the face — earthquakes and tsunamis, for example — climate change is communicated through the language of statistical probabilities and scientific uncertainty. Uncertainty can make it seem abstract. Climate change is a large, abstract problem, but what we do about it will need to be concrete.
Let's say you are in a house and a fire starts in the kitchen. You would do something, right? Perhaps you are the hero who would put the fire out single-handedly, or perhaps you are the hero who would ensure that everyone gets out of the house unscathed. Perhaps you are not the hero and would knock over Grandma in your attempt to get yourself out of the house. In all these cases, you would do something. You wouldn't just stand there.
But let's say there is not a fire; instead, radon is slowly leaking into your house. Radon is a naturally occurring radioactive gas that comes from decay of elements in rocks and soil. It is invisible and odorless. According to the US Environmental Protection Agency, "Exposure to radon in the home is responsible for an estimated 20,000 lung cancer deaths each year." You are standing in the kitchen with Grandma. All is quiet. Would you know what was happening? Would you know it was a problem? You probably wouldn't knock over Grandma to get out of a house filled with radon. You might look at the number on the radon test result and wonder whether that number was particularly bad, especially if the number was at or near the safe threshold. You might wonder, is this an emergency?
The EPA scientists cannot say for sure that if you live in a house with radon you will get lung cancer. They can only say you might get lung cancer. (Similarly, they cannot say that if there is no radon in your house, you will not get lung cancer.) The difference between knowing what will happen and what might happen is uncertainty. We are not sure what the future will look like. But we do have a good idea. The statistical uncertainty in the radon example crops up because of chaos — unpredictability because of nature's complexity. There is another type of uncertainty that is important when assessing risk posed by disasters — the uncertainty that is an inherent part of the process of science. Scientists communicate uncertainty to indicate how much we know about a situation and what we still need to learn. The latter we can decrease with more research, but it will always exist in some small amount because we will never know everything. The former — uncertainty due to chaos — we have to acknowledge and live with.
* * *
Once, I was at a local television station meeting with a TV meteorologist and several others about weather education projects. Our meeting was held at a conference table within the news studio. I was familiar with the look of the news desk and other sets from watching television, but I had never considered that on the other side of the room, off camera, there might be something as mundane as a conference table. At one point during our meeting, the meteorologist excused himself to film a series of short teasers to be broadcast during the commercial breaks of Oprah's show. I was tickled. I am not usually at meetings where someone excuses himself to be filmed. This particular meteorologist was one of those people who radiates so much charisma that when he left the table, the light seemed to dim. He strode over to the blank wall — the green screen — and a camera was trained on him. With natural confidence he spoke in quick haikuesque phrases about the evening's clear skies and full moon, bite-sized morsels of the future. He returned to the table muttering that in terms of predicting the future, at least the full moon was a sure thing.
Weather is the most common way that we experience environmental change. Weather prediction is much more accurate than it used to be, but there are still days when the future doesn't come out the way we thought it might. And when that happens, it is common to blame the meteorologist, to shake a fist at the cloud-covered sky and curse his or her name. But is that justified? The job of predicting future change is not an easy one. According to a meteorologist interviewed in The Weather Book by Jack Williams — an excellent handbook on weather and forecasting for curious novices — if there is not a cloud in the sky, you can bet there will be no rain in the next half hour. By limiting the distance into the future you forecast, you limit the number of chance events that could affect it. Basically, a prediction of no rain for half an hour under blue skies is only slightly more uncertain than a prediction of a full moon.
For a weather forecast that extends a day or a week into the future, you'll want to know what the weather is like over the whole country or even the world, because weather moves around. It helps that we can look at our planet from the vantage point of space. Weather radar and satellites are like the security systems in malls, and meteorologists are like the security guards, except that while the security guards in malls watch mallgoers and shoplifters, meteorologists watch water vapor, temperatures, and wind around the world. But even with excellent surveillance of present weather, it's a challenge to predict how weather will change.
Weather models use math equations to describe all sorts of processes that happen on earth every day — how sunshine heats land and water, how clouds form, and how differences in air pressure cause winds to blow. The more we know about how the planet works, the better our math, the better our models, and the better our predictions. If you feed a model the current weather, it can extrapolate into the future based on what it is programmed to know about how our planet works. You can extend beyond tomorrow to the next day and the day after. However, there are limits to how far out in time the weather model can predict. That's why you see a seven- or ten-day forecast instead of a one-year forecast. The farther into the future you are looking at weather, the more uncertainty there is because of the possibility of chance events — of chaos.
If there's an antihero in the world of weather prediction, it's chaos. It limits our ability to see the future. Chaos theory is the idea that when added up over time, small things affect large things. It makes predictions uncertain.
The father of chaos theory, Dr. Edward Lorenz of the Massachusetts Institute of Technology, explained chaos in the atmosphere with a metaphor: a butterfly flapping its wings may cause a tornado thousands of miles away. He wasn't blaming the butterfly. There might be another butterfly with flapping wings that would stop a tornado from forming. And neither butterfly would be trying to change the weather. They would just exist, as butterflies do. Extrapolated into larger and more human terms, it's a good example of how we all make a difference in the world even if we don't feel we're accomplishing anything.
With chaos, small changes in the starting conditions can lead to a wide variety of outcomes. Much like a Choose Your Own Adventure story, in which there are many possible stories in one book, chaos means that there can be a whole variety of outcomes when a hurricane hits the coastline, a tornado forms, or an earthquake rattles.
Put another way, chaos theory is like the difference between a day's to-do list and what one actually does that day. Each morning over breakfast, I scribble on a notepad a list of things that I intend to do, but I know there is a chance that my plans will change. My morning list is a prediction of the future. When things come up during the day that weren't in my prediction, that's chaos. It might be something good, like free ice cream. It might be something bad, like a car accident. Either way, it's a little slice of chaos. Sometimes the things that come up are so large that they make the rest of my list irrelevant. Other times they cause only slight changes to my list. Anything can happen. Often, it does.
If you don't like the weather, just wait a few minutes. Mark Twain once said this about New England, and Will Rogers once said it about Oklahoma. Countless other people have said it about countless other places. I've noticed that it's often said with a boastful tone about a specific place, as if the weather in that place is far more cunning than weather anywhere else. But weather changes everywhere. That's what weather does. And chaos is often at the root of the changes. And while chaos affects everything, weather is often the part of our planet where it is most obvious. It affects us every day, and most of us try to prepare for it, wearing sweaters, coats, and galoshes and then finding ourselves inappropriately dressed when a butterfly flaps its wings and the predicted cold front moves in a different direction. Weather is especially affected by chaos compared with other parts of the planet, because it happens in the atmosphere, a giant sloshing fluid that is prone to perturbations.
* * *
In 2009, the National Oceanic and Atmospheric Administration (NOAA) started to describe uncertainty due to nature's chaos in weather forecasts, including the probability that specific weather would occur. For example, instead of a forecast stating that the low temperature in central Florida will be 36°F tonight, the new forecast includes that there is a 30 percent probability that the temperature will dip below freezing, 32°F. With a forecast that includes the uncertainty in the form of the probability of a freeze, Florida orange growers can decide whether they need to take precautions to keep their groves from freezing. NOAA issued a statement to explain how they would transition from providing forecasts as a single value of the most likely scenario to reporting both the most likely value and the probabilities of other values. "NOAA's constituents are now requesting uncertainty information for weather, water, and climate scenarios for better risk-based decision making," read the statement.
In 2011, the United Kingdom Meteorological Office (the Met Office) did the same, describing the probability of precipitation in weather forecasts. This was done in an effort to provide transparency, to let people who hear the forecast come to their own conclusion about whether to bring an umbrella. However, this change was met with outrage from a UK organization called the Plain English Campaign, a group that lobbies against what they call "gobbledygook" in communications of all sorts. In 2011 the Plain English Campaign gave the Met Office their Golden Bull award for the year's best example of gobbledygook. According to the Plain English Campaign, the award was given for "empowering people to make their own decisions by using the technical systems for the probabilities of precipitation." I'm fond of alliteration, so I'm a fan of the phrase "percent probability of precipitation," but I can understand that this seems confusing to some. "Often people want to make a decision, such as whether to put out their washing to dry, and would like us to give a simple yes or no. However, this is often a simplification of the complexities of the forecast and may not be accurate," the Met Office explained in a rebuttal. "By giving [probabilities of precipitation], we give a more honest opinion of the risk and allow you to make a decision depending on how much it matters to you."
Forty percent probability of precipitation means that if today were repeated ten times, rain, snow, graupel, or hail would fall from the sky during four of those times. The weather forecasters run their models many times to figure out these odds. Because we all have different tolerances for risk, some people might carry an umbrella if there was 30 percent probability of precipitation. Others might not carry an umbrella unless the probability was over 50 percent, or only if they were wearing a shirt with a Dry Clean Only tag. Knowing about the tendency for rainy weather in the UK, and my own aversion to risk, I'd carry an umbrella all the time.
Researchers Rebecca Morss, Jeff Lazo, and Julie Demuth have found that some people interpret weather uncertainty incorrectly. Some interpret a forecast of 40 percent probability of precipitation as rain falling during 40 percent of the day or rain falling over 40 percent of the land area. In contrast, most people interpret uncertainty in temperature forecasts correctly as being a range of degrees instead of one average number.
Providing only an average hides the amount that something can differ from the average. For example, if you were to describe what dogs look like to someone who had never seen a dog — a visiting alien, perhaps — you'd need to communicate that there is a wide range in size, shape, color, fur texture, and friendliness. If you only described the average dog, the alien might imagine a four-footed animal that's about thirty pounds. When encountering a two-pound Chihuahua tucked into someone's purse or a two hundred–pound Saint Bernard pulling a sled, the alien would not think that either was a dog. In the case of dog breeds, the variability is not natural; it's because we have chosen to breed dogs that have certain characteristics. In the case of a weather forecast, variability in the prediction is largely due to nature's chaos, and this uncertainty is communicated as a probability. While it is unlikely that visiting aliens will ever ask you to describe dogs, it is likely that you will encounter statistical uncertainty in other realms.
In the case of weather, models are run over and over again to identify the amount of variability in outcome. Climate models are looking at a broad scale instead of details, so they are not affected by chaos, but there are some chaotic elements — the timing of El Niño events, for example. Climate models are also run multiple times to help us understand the range of possible changes in temperature that we are facing in the future. However, most of the range in temperatures is not because of chaos and uncertainty. It's because of us. We don't know how much greenhouse gas we'll be adding to the atmosphere. We don't know whether we'll change our ways quickly enough to stop catastrophic levels of climate change. The outcome is contingent upon our actions now and in the future. Just as the looks and temperaments of dog breeds are due largely to human decisions about which dogs to breed over many generations, our planet's warming climate is due largely to the decisions we make about whether to burn fossil fuels.
* * *
There is another source of uncertainty that affects every branch of science. This uncertainty is due to the fact that we don't know everything.
Scientists work at the outskirts of our understanding, on the border between known and unknown. On one side of this boundary is all the science that we know. On the other side of the boundary is an endless plain of knowledge that we do not yet know. And the science that researchers are exploring now is right on the boundary.
Over a century ago, the idea that carbon dioxide traps heat was on the outskirts of our understanding, but today it is far inland, because many scientists have tested it with different methods. About seventy years ago, the idea that humans were changing the amount of carbon dioxide in the atmosphere was at the boundary of our understanding, but today that's understood, too. For decades, scientists around the world have been measuring the amount of carbon dioxide in air collected each day. They have also been measuring carbon dioxide from ancient air bubbles extracted from glaciers and the amount of carbon dioxide we emit into the atmosphere. Over time, understanding of our ability to change carbon dioxide levels and warm the planet has grown. That is no longer on the border of what's known.
Today, scientists are working at the outer reaches of our understanding, learning details that will let us make better predictions of future climate, such as the rate that methane gas is released as frozen soils melt in the Arctic and the impact of different types of clouds on climate. These topics and many others are on the boundary of what's known. The answers will help us know more specifically how much climate change we should expect and how climate change will affect other parts of the planet.
(Continues…)
Excerpted from "Tales From An Uncertain World"
by .
Copyright © 2018 L. S. Gardiner.
Excerpted by permission of University of Iowa Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
