Everyday Forecasting

In the old days, I never trusted the weather forecast. I’d only check to pooh-pooh how poorly they would do. Today, I’m amazed at how accurate forecasts can be, down to my local area. The story of how weather forecasting improved as it became a complex gargantuan endeavor is detailed in Andrew Blum’s The Weather Machine. Published in 2019, the book is aimed at non-experts who might be interested in how the weather wizards do what they do.

Did you know that before the World Wide Web there was the World Weather Watch (WWW)? Formed in 1963 at the instigation of an American, Harry Wexler, and a Russian, Viktor Bugaev, in the midst of the Cold War, it proposed three interlocking global systems for weather observations, data processing, and telecommunications. The core idea: “open and equal access to weather information, for operational and experimental use.”

Weather knows no borders. But measuring the weather has broad implications spanning politics, national security and global technologies. Blum astutely describes the tensions: “The only caveat written into the charter was that the WWW be used for peaceful purposes only. The UN proper might have been overwhelmed by the festering tensions of a world divided between East and West, but the weather diplomats were insistent on the borderless atmosphere. This was bold of them, given the technology on which they relied. Weather satellites were so expensive that they could be justified only on national security grounds. Mostly this limitation was technological: The innovation they required overlapped significantly with both intercontinental missiles and spy satellites. But it was also political: The jingoistic appeal of satellites was also a function of how they overflew the whole earth, without regard for the borders below – overturning the historical understanding of sovereignty and territory.”

In the book’s epilogue, Blum describes efforts by the UN’s  World Meteorological Office (WMO) in tackling climate change and its global effects especially on poorer countries. The tension between public good versus private enterprise has invaded the world of weather data collection and analysis. Global cooperation is facing increasing challenges with the rise of jingoistic world leaders and demagogues with a “me first” attitude. We are seeing this starkly in the current global pandemic involving WHO, a UN sister organization of WMO. The coronavirus knows no borders either, but yet politics, national security, public good, private choice, all come into play.

Three things jumped out at me while reading The Weather Machine. I will spend a brief paragraph on each.

(1) One chapter titled “Euro” takes the reader inside the workings of the European Centre for Medium-Range Weather Forecasts (ECMWF), which has the current best weather forecasting models and simulations. The scientists, the culture, the competitiveness, the collaboration, are fascinating aspects into the human-side of the weather enterprise, fast being taken over by computers as they crunch more data in ever more sophisticated ways. This chapter is an extended vignette among many others where Blum personally visits with the meteorologists to learn the history and inner workings of the weather business. It’s an interesting look behind the curtain to see how things work!

(2) I resonated with the discussion of Wexler’s theory that both the macroscopic and microscopic in weather observation are equally important – “a bigger picture at a higher resolution”. An accessible example might be the evolution of TVs, driven by the desire of consumers to have larger screens with yet finer resolution. I study the chemical origins of life, and we face a similar issue. On the one hand, a global bird’s eye view is important to get a handle on how life may have started, but the nitty-gritty details are also equally important. We’ll need both in increasing measure to make headway on the problem. While my research focus is on the microscopic, I ensure that my reading diet consists regularly of macroscopic views. This will be crucial for research to advance in the field.

(3) Blum makes another very astute observation when he discusses one aspect of how models and simulations work in climate change, but has broader implications to other types of simulations. “The glory of good data assimilation is that it allows for the model to compensate for places where observations are sparse. It becomes a bridge between the areas that are well observed and the areas that aren’t. The surprising result of that discrepancy between model space and real space is that the model, you might say, is more detailed than reality – or reality, at least as it is observed.” This is exactly what happens as we develop models for the origin of life. The data is sparse indeed. As a quantum chemist, I’m threading my way through the complexity but examining the in-between things that are difficult to observe experimentally.

I enjoyed The Weather Machine. I wasn’t expecting to entertain thoughts about origin-of-life simulation/modeling while reading about the weather. It’s hard to forecast when one thought will lead to another!