Book review – The Weather Machine: How We See Into the Future

6-minute read

Checking the weather forecast is like flushing your toilet. A banal activity we all engage in a few times a day. But does anyone of us really know what goes into making it? Andrew Blum is fascinated by infrastructures. His previous book Tubes explored the physical infrastructure that keeps the internet running. Here he delves into the infrastructure that enables weather predictions. Most of us might have an inkling it involves satellites and computer models, but that does not begin to describe the globe-spanning collaborative network that hides under the bonnet.

The Weather Machine: How We See Into the Future, written by Andrew Blum, published in Europe by The Bodley Head (a Vintage imprint) in June 2019 (hardback, 207 pages)

Blum’s particular mission statement does have one corollary: the weather itself is almost absent from this book. If you are interested in the basics of weather patterns, where rain comes from, how storms develop, what the deal is with areas of high and low pressure, etc. you will have to look elsewhere. Of course, seeing how big of a topic meteorology is, this omission is understandable.

For this book, Blum takes the opening of the first telegraph line in the USA in 1844 as the starting point of his history of modern meteorology. Their sensitivity to foul weather and the instantaneous exchange of information meant its operators were the first to collect weather observations over larger areas. With it came the almost spontaneous realisation that weather was not a local phenomenon, but moved across the country.

From these early beginnings, Blum introduces one of the founding fathers of weather modelling, the Norwegian Vilhelm Bjerknes. His work at the turn of the 19th century is still relevant today, underlying our current models. Visions of a “forecast factory” filled with humans working in parallel to compute weather patterns were floated, but the reality lagged behind. The manual calculations were too time-consuming to be of practical use and there was a chronic lack of observations to do the calculations with.

“Visions of a “forecast factory” filled with humans working in parallel to compute weather patterns were floated, but the reality lagged behind.”

Though this first part of the book makes for fascinating reading, a moment’s reflection suggests that it is somewhat lacking in its coverage. A pioneer such as Francis Beaufort – he of the wind scale – is barely mentioned. The deeper history of weather observations and early instruments, before the 1800s, is left out. The humble weather balloon, which is still let up in droves to this day, receives scant mention. And what of the ubiquitous weather station?

Delving into Bjerknes’s history entails a trip to Norway, where Blum visits the island of Utsira, home to one the nation’s oldest weather stations. His reporting here is full of atmosphere but surprisingly thin on factual information. What instruments do you actually find in a modern ground-based weather station? What do they measure, and how? How have they changed over time? Blum remains vague on the specifics and interested readers might want to consult e.g. Setting Up a Weather Station and Understanding the Weather or the more in-depth The Weather Observer’s Handbook.

“[…] models have grown in complexity and power to the point we have virtual models of the planet that produce reliable forecasts a week into the future.”

Fortunately, the book hits its stride when Blum moves on to the invention of both rocket technology in the wake of World War II, and the satellite technology of the Cold War. As with many scientific disciplines, war and military funding provided much of the impetus for technological development. This seems a topic closer to Blum’s heart, and he explores the different types of satellites, the logistical challenge of updating and moving around this space fleet, and the different agencies responsible for this. Despite living in Europe myself, I was previously only familiar with the US agency NOAA (the National Oceanic and Atmospheric Administration) but not its European counterpart EUMETSAT (the European Organisation for the Exploitation of Meteorological Satellites). Blum visits the German headquarters, witnessing data collection from a satellite in action, while back in the US, he travels to the Jet Propulsion Laboratory to marvel at a new type of satellite to measure soil moisture, but misses out on the launch due to (ironically) bad weather.

The final stretch of reportage sees Blum delve into the weather models and forecasts. As also argued in A Vast Machine, data and models go hand in hand. With increasing amounts of data from all corners and layers of the globe, models have grown in complexity and power to the point that we have virtual models of the planet that produce reliable forecasts a week into the future.

“Meteorology seems to be one of the few areas where nations put aside their differences to work towards a greater good […]”

In the US, Blum talks to scientists at the National Center for Atmospheric Research (NCAR), while in Europe he visits the undisputed leader in this field, the European Centre for Medium-Range Weather Forecast (ECMWF). The latter is an especially fascinating organisation, both in the way it is run, as what it is running on (some of the world’s fastest supercomputers). Lastly, there is The Weather Company in the US who provides forecasts-on-demand over the internet to all the major players (Google, Facebook etc.) and has become increasingly reliant on models over humans.

Alongside all the gadgetry, Blum takes the time for interesting reflections. I mentioned the continued entanglement of military and meteorological interests. But more importantly, there is the continuous cooperation between all of the world’s countries. Meteorology seems to be one of the few areas where nations put aside their differences to work towards a greater good without any financial gain involved. Even so, the increasingly expensive and complicated infrastructure sees power and knowledge becoming concentrated in the hands of fewer players. Entirely different developments that have meteorologists scratching their heads are the spectres of both crowd-sourced and private data in the hands of large technology corporations seeking to profiteer from it.

Overall, I think The Weather Machine is somewhat lacking when it comes to the early history of meteorology but hits its stride when it gets to the satellite era and beyond. Blum features a pleasant mix of accessible information with on-the-ground interviews and impressions, and I breezed through this book in a day. If like me, you know little about modern meteorology, this absorbing book provides a fascinating glimpse into a world largely hidden from view. But do not be surprised if new questions arise after reading this book.

Disclosure: The publisher provided a review copy of this book. The opinion expressed here is my own, however.

You can support this blog using below affiliate links, as an Amazon Associate I earn from qualifying purchases:

The Weather Machine hardback, paperback, ebook or audio CD