The last post posed the question…why isn’t there greater public-, policymaker-, and private-sector demand for the information Earth scientists and science-based services have on offer? There seemed to be four sobering possibilities (thankfully, a reader has raised a fifth, which isn’t quite so humbling; it arguably puts us in a somewhat better light).
Perhaps it’s worthwhile to drill down on each of these (including that happier fifth one).
Here we tackle the first.
Maybe…just maybe…some of that stuff we “know” isn’t quite so true, or so universally true, or so useful, as we might hope. Maybe we don’t know so much as we think we do.
In science, this shortcoming comes in two flavors. The first is malignant. Here the problem’s not that I’m ignorant; it’s what I know that isn’t so. [That’s the point of the Darwin quote to the right of the blog homepage…] Science has a history here. When, back in the second century AD, Galen stated that the function of the heart is to heat the body, he set generations of physicians off on a wild goose chase. It took fourteen hundred years to turn things around. [Other examples abound. For example, the 17th-century phlogiston theory of combustion falls in the same category.] Fortunately, this shortcoming is less prevalent than it once was. It’s more a feature of scientists version 1.0.
Today’s shortcomings are preponderantly of a second, more benign type. Frequently we “know” things that turn out to be only approximately true. Take Newtonian mechanics. Isaac Newton laid out the foundations in the 17th century. In our direct human experience, with large objects moving at slow speeds (a billiard ball, car, an airplane, even a rocket or a bullet), the theory works quite well…but get something moving at nearly the speed of light, and Newton’s laws begin to break down. And though our direct human experience is Newtonian, a large number of our day-to-day gadgets – computers, smartphones, GPS, flat-screen TV’s – work in that other realm.
Why do today’s problems tend to be less severe? They’re the product of scientists version 2.0 (and higher); scientists who have done their mathematics and experimented. The equations and the laboratory measurements or field observations alert scientists when they start to go astray.
Let’s take a look at an atmospheric problem. Our ancestors noted early on in human history that pollution – the smoke from a fire or the sludge in water – was local. Get away a little distance, and the effects were barely detectable. The atmosphere and large bodies of water appeared able to assimilate any amount of garbage.
Or so it seemed at first. But just as Newtonian mechanics held only for slow speeds, the idea that the planet could readily accommodate all human waste held only for small numbers of people, and slow rates of waste generation. As our numbers have increased, and our economic activity has accelerated, we’ve found that pollution has become regional (think of the smog in the Los Angeles basin, or acid rain, or effluent blowing across the Pacific from Chinese coal-fired power plants, or the brown cloud over southeast Asia). As we developed our science, measurement technologies, and data-analysis technique, we started seeing our impacts were/are global (think carbon emissions and global warming). And just as Galen’s reasonable-enough but incorrect hypothesis about the function of the heart clouded the scientific and public discussion for centuries, it’s taking a while to get the climate change discussion on track. New science still takes time to work its way across a global society, undergo scrutiny, and gain acceptance.
Let’s turn now to the predictability of the weather. And let’s start with another little bit of history. Back in the 1880’s the U.S. Army Signal Service, which was responsible for weather predictions of the time, used to claim an accuracy of 80% or more for its forecasts. Wow! But here’s how they determined that. They divided the entire country of the time into nine regions. They would predict rain or no rain for each of those nine regions for the coming day. If they predicted rain for a region, and rain fell in any amount anywhere in that region over the 24-hour period of the one-day forecast, they gave themselves full credit.
Seen from this historical perspective, we’ve made enormous strides. In terms of atmospheric parameters – wind speeds, pressures, temperatures – our forecast accuracy has improved year on year for decades. But in terms of society’s needs, we are barely keeping pace. For example, fifty years ago, a nine-hour hurricane landfall notice was enough time to evacuate Sanibel Island, just off Fort Myers, Florida. Today, because of the population growth in that area, 30-40 hours’ notice is required.
Or click on this video depicting Fed Ex aircraft flights into their Memphis hub during a thunderstorm, and contemplate the highly-localized nature of weather observations and predictions that are needed to support air traffic control today versus the air traffic of fifty years ago.
And that’s at just one airport. Globally, we have 50,000 airports. That translates into a bit more than one new airport a day for the past century. Something like 2000 thunderstorms are underway at any given time worldwide. A lotof opportunities for trouble; ask any summer traveler. So even as new techniques for Earth observations, and greater capacity for numerical weather prediction come on line, they struggle to match pace with the growing needs of aviation for weather support.
And that’s just aviation. As food surpluses decline worldwide, and the work to increase yields intensifies, agribusiness needs more detailed weather information to support decisions of every type: when to plant? When to harvest? How about irrigation? Or the application of herbicides and pesticides when they’ll help the crops before they’re washed away by rains? What’s the outlook for crops worldwide? Even as our knowledge of agricultural meteorology has grown, the needs have become more urgent and complex.
Then there’s renewable energy. How much energy will the world’s solar power plants collect tomorrow? What will be the cloud cover? When will the clouds develop? How about the winds over those farms of wind turbines? Oh…and it’s not good enough to have a general idea. To get the most out of those 100-m turbines, it’s necessary to know details of the wind fluctuations over the farms over periods of minutes, to know the variation over the wind field terrain not just at the surface but up a couple of hundred meters.
The same story applies to water resource management.
So how much do we know? A lot more than we knew ten years ago. We’re approximately correct. From that perspective, it’s tempting to think we’ve made progress. But in the zero-margin future world of energy, water, and food, today’s capability, today’s approximation, often isn’t good enough. It turns out we know less than we thought…or certainly less than we need to.
We’re smarter than we were, but not so smart as we need to be.
But even if we were smarter, we still wouldn’t be rich. More on that in the next post.