Data Science, Economics, Falsifiable

Is Google Putting Money In Your Pocket?

The Cambridge Analytica scandal has put tech companies front and centre. If the thinkpieces along the lines of “are the big tech companies good or bad for society” were coming out any faster, I might have to doubt even Google’s ability to make sense of them all.

This isn’t another one of those thinkpieces. Instead it’s an attempt at an analysis. I want to understand in monetary terms how much one tech company – Google – puts into or takes out of everyone’s pockets. This analysis is going to act as a template for some of the more detailed analyses of inequality I’d like to do later, so if you have a comment about methodology, I’m eager to hear it.

Here’s the basics: Google is a large technology company that primarily makes money off of ad revenues. Since Google is a publicly traded company, statistics are easy to come by. In 2016, Google brought in $89.5 billion in revenue and about 89% of that was from advertising. Advertising is further broken down between advertising on Google sites (e.g. Google Search, Gmail, YouTube, Google Maps, etc.) which account for 80% of advertising revenue and advertising on partner sites, which covers the remainder. The remaining 11% is made up of a variety of smaller projects – selling corporate licenses of its GSuite office software, the Google Play Store, the Google Cloud Computing Platform, and several smaller projects.

There are two ways that we can track how Google’s existence helps or hurts you financially. First, there’s the value of the software it provides. Google’s search has become so important to our daily life that we don’t even notice it anymore – it’s like breathing. Then there’s YouTube, which has more high-quality content than anyone could watch in a lifetime. There’s Google Docs, which are almost a full (free!) replacement for Microsoft Office. There’s Gmail, which is how basically everyone I know does their email. And there’s Android, currently the only viable alternative to iOS. If you had to pay for all of this stuff, how much would you be out?

Second, we can look at how its advertising arm has changed the prices of everything we buy. If Google’s advertising system has driven an increase in spending on advertising (perhaps by starting an arms race in advertising, or by arming marketing managers with graphs, charts and metrics that they can use to trigger increased spending), then we’re all ultimately paying for Google’s software with higher prices elsewhere (we could also be paying with worse products at the same prices, as advertising takes budget that would otherwise be used on quality). On the other hand, if more targeted advertising has led to less advertising overall, then everything will be slightly less expensive (or higher quality) than the counterfactual world in which more was spent on advertising.

Once we add this all up, we’ll have some sort of answer. We’ll know if Google has made us better off, made us poorer, or if it’s been neutral. This doesn’t speak to any social benefits that Google may provide (if they exist – and one should hope they do exist if Google isn’t helping us out financially).

To estimate the value of the software Google provides, we should compare it to the most popular paid alternatives – and look into the existence of any other good free alternatives. Because of this, we can’t really evaluate Search, but because of its existence, let’s agree to break any tie in favour of Google helping us.

On the other hand, Google docs is very easy to compare with other consumer alternatives. Microsoft Office Home Edition costs $109 yearly. Word Perfect (not that anyone uses it anymore) is $259.99 (all prices should be assumed to be in Canadian dollars unless otherwise noted).

Free alternatives exist in the form of OpenOffice and LibreOffice, but both tend to suffer from bugs. Last time I tried to make a presentation in OpenOffice I found it crashed approximately once per slide. I had a similar experience with LibreOffice. I once installed it for a friend who was looking to save money and promptly found myself fixing problems with it whenever I visited his house.

My crude estimate is that I’d expect to spend four hours troubleshooting either free alternative per year. Weighing this time at Ontario’s minimum wage of $14/hour and accepting that the only office suite that anyone under 70 ever actually buys is Microsoft’s offering and we see that Google saves you $109 per year compared to Microsoft and $56 each year compared to using free software.

With respect to email, there are numerous free alternatives to Gmail (like Microsoft’s Hotmail). In addition, many internet service providers bundle free email addresses in with their service. Taking all this into account, Gmail probably doesn’t provide much in the way of direct monetary value to consumers, compared to its competitors.

Google Maps is in a similar position. There are several alternatives that are also free, like Apple Maps, Waze (also owned by Google), Bing Maps, and even the Open Street Map project. Even if you believe that Google Maps provides more value than these alternatives, it’s hard to quantify it. What’s clear is that Google Maps isn’t so far ahead of the pack that there’s no point to using anything else. The prevalence of Google Maps might even be because of user laziness (or anticompetitive behaviour by Google). I’m not confident it’s better than everything else, because I’ve rarely used anything else.

Android is the last Google project worth analyzing and it’s an interesting one. On one hand, it looks like Apple phones tend to cost more than comparable Android phones. On the other hand, Apple is a luxury brand and it’s hard to tell how much of the added price you pay for an iPhone is attributable to that, to differing software, or to differing hardware. Comparing a few recent phones, there’s something like a $50-$200 gap between flagship Android phones and iPhones of the same generation. I’m going to assign a plausible sounding $20 cost saved per phone from using Android, then multiply this by the US Android market share (53%), to get $11 for the average consumer. The error bars are obviously rather large on this calculation.

(There may also be second order effects from increased competition here; the presence of Android could force Apple to develop more features or lower its prices slightly. This is very hard to calculate, so I’m not going to try to.)

When we add this up, we see that Google Docs save anyone who does word processing $50-$100 per year and Android saves the average phone buyer $11 approximately every two years. This means the average person probably sees some slight yearly financial benefit from Google, although I’m not sure the median person does. The median person and the average person do both get some benefit from Google Search, so there’s something in the plus column here, even if it’s hard to quantify.

Now, on to advertising.

I’ve managed to find an assortment of sources that give a view of total advertising spending in the United States over time, as well as changes in the GDP and inflation. I’ve compiled it all in a spreadsheet with the sources listed at the bottom. Don’t just take my word for it – you can see the data yourself. Overlapping this, I’ve found data for Google’s revenue during its meteoric rise – from $19 million in 2001 to $110 billion in 2017.

Google ad revenue represented 0.03% of US advertising spending in 2002. By 2012, a mere 10 years later, it was equivalent to 14.7% of the total. Over that same time, overall advertising spending increased from $237 billion in 2002 to $297 billion in 2012 (2012 is the last date I have data for total advertising spending). Note however that this isn’t a true comparison, because some Google revenue comes from outside of America. I wasn’t able to find revenue broken down in greater depth that this, so I’m using these numbers in an illustrative manner, not an exact manner.

So, does this mean that Google’s growth drove a growth in advertising spending? Probably not. As the economy is normally growing and changing, the absolute amount of advertising spending is less important than advertising spending compared to the rest of the economy. Here we actually see the opposite of what a naïve reading of the numbers would suggest. Advertising spending grew more slowly than economic growth from 2002 to 2012. In 2002, it was 2.3% of the US economy. By 2012, it was 1.9%.

This also isn’t evidence that Google (and other targeted advertising platforms have decreased spending on advertising). Historically, advertising has represented between 1.2% of US GDP (in 1944, with the Second World War dominating the economy) and 3.0% (in 1922, during the “roaring 20s”). Since 1972, the total has been more stable, varying between 1.7% and 2.5%. A Student’s T-test confirms (P-values around 0.35 for 1919-2002 vs. 2003-2012 and 1972-2002 vs. 2003-2012) that there’s no significant difference between post-Google levels of spending and historical levels.

Even if this was lower than historical bounds, it wouldn’t necessarily prove Google (and its ilk) are causing reduced ad spending. It could be that trends would have driven advertising spending even lower, absent Google’s rise. All we can for sure is that Google hasn’t caused an ahistorically large change in advertising rates. In fact, the only thing that is clear in the advertising trends is the peak in the early 1920s that has never been recaptured and a uniquely low dip in the 1940s that seems to have obviously been caused by World War II. For all that people talk about tech disrupting advertising and ad-supported businesses, these current changes are still less drastic than changes we’ve seen in the past.

The change in advertising spending during the years Google is growing could be driven by Google and similar advertising services. But it also could be normal year to year variation, driven by trends similar to what have driven it in the past. If I had a Ph. D. in advertising history, I might be able to tell you what those trends are, but from my present position, all I can say is that the current movement doesn’t seem that weird, from a historical perspective.

In summary, it looks like the expected value for the average person from Google products is close to $0, but leaning towards positive. It’s likely to be positive for you personally if you need a word processor or use Android phones, but the error bounds on advertising mean that it’s hard to tell. Furthermore, we can confidently say that the current disruption in the advertising space is probably less severe than the historical disruption to the field during World War II. There’s also a chance that more targeted advertising has led to less advertising spending (and this does feel more likely than it leading to more spending), but the historical variations in data are large enough that we can’t say for sure.

Literature, Model

Does Amateurish Writing Exist

[Warning: Spoilers for Too Like the Lightning]

What marks writing as amateurish (and whether “amateurish” or “low-brow” works are worthy of awards) has been a topic of contention in the science fiction and fantasy community for the past few years, with the rise of Hugo slates and the various forms of “puppies“.

I’m not talking about the learning works of genuine amateurs. These aren’t stories that use big words for the sake of sounding smart (and at the cost of slowing down the stories), or over the top fanfiction-esque rip-offs of more established works (well, at least not since the Wheel of Time nomination in 2014). I’m talking about that subtler thing, the feeling that bubbles up from the deepest recesses of your brain and says “this story wasn’t written as well as it could be”.

I’ve been thinking about this a lot recently because about ¾ of the way through Too Like The Lightning by Ada Palmer, I started to feel myself put off [1]. And the only explanation I had for this was the word “amateurish” – which popped into my head devoid of any reason. This post is an attempt to unpack what that means (for me) and how I think it has influenced some of the genuine disagreements around rewarding authors in science fiction and fantasy [2]. Your tastes might be calibrated differently and if you disagree with my analysis, I’d like to hear about it.

Now, there are times when you know something is amateurish and that’s okay. No one should be surprised that John Ringo’s Paladin of Shadows series, books that he explicitly wrote for himself are parsed by most people as pretty amateurish. When pieces aren’t written explicitly for the author only, I expect some consideration of the audience. Ideally the writer should be having fun too, but if they’re writing for publication, they have to be writing to an audience. This doesn’t mean that they must write exactly what people tell them they want. People can be a terrible judge of what they want!

This also doesn’t necessarily imply pandering. People like to be challenged. If you look at the most popular books of the last decade on Goodreads, few of them could be described as pandering. I’m familiar with two of the top three books there and both of them kill off a fan favourite character. People understand that life involves struggle. Lois McMaster Bujold – who has won more Hugo awards for best novel than any living author – once said she generated plots by considering “what’s the worst possible thing I can do to these people?” The results of this method speak for themselves.

Meditating on my reaction to books like Paladin of Shadows in light of my experiences with Too Like The Lightning is what led me to believe that the more technically proficient “amateurish” books are those that lose sight of what the audience will enjoy and follow just what the author enjoys. This may involve a character that the author heavily identifies with – the Marty Stu or Mary Sue phenomena – who is lovingly described overcoming obstacles and generally being “awesome” but doesn’t “earn” any of this. It may also involve gratuitous sex, violence, engineering details, gun details, political monologuing (I’m looking at you, Atlas Shrugged), or tangents about constitutional history (this is how most of the fiction I write manages to become unreadable).

I realized this when I was reading Too Like the Lightning. I loved the world building and I found the characters interesting. But (spoilers!) when it turned out that all of the politicians were literally in bed with each other or when the murders the protagonist carried out were described in grisly, unrepentant detail, I found myself liking the book a lot less. This is – I think – what spurred the label amateurish in my head.

I think this is because (in my estimation), there aren’t a lot of people who actually want to read about brutal torture-execution or literally incestuous politics. It’s not (I think) that I’m prudish. It seemed like some of the scenes were written to be deliberately off-putting. And I understand that this might be part of the theme of the work and I understand that these scenes were probably necessary for the author’s creative vision. But they didn’t work for me and they seemed like a thing that wouldn’t work for a lot of people that I know. They were discordant and jarring. They weren’t pulled off as well as they would have had to be to keep me engaged as a reader.

I wonder if a similar process is what caused the changes that the Sad Puppies are now lamenting at the Hugo Awards. To many readers, the sexualized violence or sexual violence that can find its way into science fiction and fantasy books (I’d like to again mention Paladin of Shadows) is incredibly off-putting. I find it incredibly off-putting. Books that incorporate a lot of this feel like they’re ignoring the chunk of audience that is me and my friends and it’s hard while reading them for me not to feel that the writers are fairly amateurish. I normally prefer works that meditate on the causes and uses of violence when they incorporate it – I’d put N.K. Jemisin’s truly excellent Broken Earth series in this category – and it seems like readers who think this way are starting to dominate the Hugos.

For the people who previously had their choices picked year after year, this (as well as all the thinkpieces explaining why their favourite books are garbage) feels like an attack. Add to this the fact that some of the books that started winning had a more literary bent and you have some fans of the genre believing that the Hugos are going to amateurs who are just cruising to victory by alluding to famous literary works. These readers look suspiciously on crowds who tell them they’re terrible if they don’t like books that are less focused on the action and excitement they normally read for. I can see why that’s a hard sell, even though I’ve thoroughly enjoyed the last few Hugo winners [3].

There’s obviously an inferential gap here, if everyone can feel angry about the crappy writing everyone else likes. For my part, I’ll probably be using “amateurish” only to describe books that are technically deficient. For books that are genuinely well written but seem to focus more on what the author wants than (on what I think) their likely audience wants, well, I won’t have a snappy term, I’ll just have to explain it like that.


[1] A disclaimer: the work of a critic is always easier than that of a creator. I’m going to be criticizing writing that’s better than my own here, which is always a risk. Think of me not as someone criticizing from on high, but frantically taking notes right before a test I hope to barely pass. ^

[2] I want to separate the Sad Puppies, who I view as people sad that action-packed books were being passed over in favour of more literary ones from the Rabid Puppies, who just wanted to burn everything to the ground. I’m not going to make any excuses for the Rabid Puppies. ^

[3] As much as I can find some science fiction and fantasy too full of violence for my tastes, I’ve also had little to complain about in the past, because my favourite author, Lois McMaster Bujold, has been reliably winning Hugo awards since before I was born. I’m not sure why there was never a backlash around her books. Perhaps it’s because they’re still reliably space opera, so class distinctions around how “literary” a work is don’t come up when Bujold wins. ^

Falsifiable, Physics, Politics

The (Nuclear) International Monitoring System

Under the Partial Test Ban Treaty (PTBT), all nuclear tests except for those underground are banned. Under the Non-Proliferation Treaty (NPT), only the permanent members of the UN Security Council are legally allowed to possess nuclear weapons. Given the public outcry over fallout that led to the PTBT and the worries over widespread nuclear proliferation that led to the NPT, it’s clear that we require something beyond pinky promises to verify that countries are meeting the terms of these treaties.

But how do we do so? How can you tell when a country tests an atomic bomb? How can you tell who did it? And how can one differentiate a bomb on the surface from a bomb in the atmosphere from a bomb in space from a bomb underwater from a bomb underground?

I’m going to focus on two efforts to monitor nuclear weapons: the national security apparatus of the United States and the Comprehensive Test Ban Treaty Organization (CTBTO) Preparatory Commission’s International Monitoring System (IMS). Monitoring falls into five categories: Atmospheric Radionuclide Monitoring, Seismic Monitoring, Space-based Monitoring, Hydroacoustic Monitoring, and Infrasound Monitoring.

Atmospheric Radionuclide Monitoring

Nuclear explosions generate radionuclides, either by dispersing unreacted fuel, as direct products of fission, or by interactions between neutrons and particles in the air or ground. These radionuclides are widely dispersed from any surface testing, while only a few fission products (mainly various radionuclides of the noble gas xenon) can escape from properly conducted underground tests.

For the purposes of minimizing fallout, underground tests are obviously preferred. But because they only emit small amounts of one particular radionuclide, they are much harder for radionuclide monitoring to detect.

Detecting physical particles is relatively easy. There are 80 IMS stations scattered around the world. Each is equipped with an air intake and a filter. Every day, the filter is changed and then prepared for analysis. Analysis involves waiting a day (for irrelevant radionuclides to decay), then reading decay events from the filter for a further day. This gives scientists an idea of what radioactive elements are present.

Any deviations from the baseline at a certain station can be indicative of a nuclear weapon test, a nuclear accident, or changing wind patterns bringing known radionuclides (e.g. from a commercial reactor) to a station where they normally aren’t present. Wind analysis and cross validation with other methods are used to corroborate any suspicious events.

Half of the IMS stations are set up to do the more difficult xenon monitoring. Here air is pumped through a material with a reasonably high affinity for xenon. Apparently activated charcoal will work, but more sophisticated alternatives are being developed. The material is then induced to release the xenon (with activated charcoal, this is accomplished via heating). This process is repeated several times, with the output of each step pumped to a fresh piece of activated charcoal. Multiple cycles ensure that only relatively pure xenon get through to analysis.

Once xenon is collected, isotope analysis must be done to determine which (if any) radionuclides of xenon are present. This is accomplished either by comparing the beta decay of the captured xenon with its gamma decay, or looking directly at gamma decay with very precise gamma ray measuring devices. Each isotope of xenon has a unique half-life (which affects the frequency with which it omits beta- and gamma-rays) and a unique method of decay (which determines if the decay products are primarily alpha-, beta-, or gamma-rays). Comparing the observed decay events to these “fingerprints” allows for the relative abundance of xenon nuclides to be estimated.

There are some background xenon radionuclides from nuclear reactors and even more from medical isotope production (where we create unstable nuclides in nuclear reactors for use in medical procedures). Looking at global background data you can see the medical isotope production in Ontario, Europe, Argentina, Australia and South Africa. I wonder if this background effect makes world powers cautious about new medical isotope production facilities in countries that are at risk of pursuing nuclear weapons. Could Iran’s planned medical isotope complex have been used to mask nuclear tests?

Not content merely to host several monitoring stations and be party to the data of the whole global network of IMS stations, the United States also has the WC-135 “Constant Phoenix” plane, a Boeing C-135 equipped with mobile versions of particulate and xenon detectors. The two WC-135s can be scrambled anywhere a nuclear explosion is suspected to look for evidence. A WC-135 gave us the first confirmation that the blast from the 2006 North Korean nuclear test was indeed nuclear, several days before the IMS station in Yellowknife, Canada confirmed a spike in radioactive xenon and wind modelling pinpointed the probable location as inside North Korea.

Seismic Monitoring

Given that fewer monitoring stations are equipped with xenon radionuclide detectors and that the background “noise” from isotope production can make radioactive xenon from nuclear tests hard to positively identify, it might seem like nuclear tests are easy to hide underground.

That isn’t the case.

A global network of seismometers ensures that any underground nuclear explosion is promptly detected. These are the same seismometers that organizations like the USGS (United States Geological Survey) use to detect and pinpoint earthquakes. In fact, the USGS provides some of the 120 auxiliary stations that the CTBTO can call on to supplement its fifty seismic monitoring stations.

Seismometers are always on, looking for seismic disturbances. Substantial underground nuclear tests produce shockwaves that are well within the detection limit of modern seismometers. The sub-kiloton North Korean nuclear test in 2006 appears to have been registered as equivalent to a magnitude 4.1 earthquake. A quick survey of ongoing earthquakes should probably show you dozens that have been detected that are less powerful than even that small North Korean test.

This probably leads you to the same question I found myself asking, namely: “if earthquakes are so common and these detectors are so sensitive, how can they ever tell nuclear detonations from earthquakes?”

It turns out that underground nuclear explosions might rattle seismometers like earthquakes do, but they do so with characteristics very different from most earthquakes.

First, the waveform is different. Imagine you’re holding a slinky and a friend is holding the other end. There are two mains ways you can create waves. The first is by shaking it from side to side or up and down. Either way, there’s a perspective from which these waves will look like the letter “s”.

The second type of wave can be made by moving your arm forward and backwards, like you’re throwing and catching a ball. These waves will cause moving regions where the slinky is bunched more tightly together and other regions where it is more loosely packed.

These are analogous to the two main types of body waves in seismology. The first (the s-shaped one) is called an S-wave (although the “S” here stands for “shear” or “secondary” and only indicates the shape by coincidence), while the second is called a P-wave (for “pressure” or “primary”).

I couldn’t find a good free version of this, so I had to make it myself. Licensed (like everything I create for my blog) CC-BY-NC-SA v4.0.


Earthquakes normally have a mix of P-waves and S-waves, as well as surface waves created by interference between the two. This is because earthquakes are caused by slipping tectonic plates. This slipping gives some lateral motion to the resulting waves. Nuclear explosions lack this side to side motion. The single, sharp impact from them on the surrounding rocks is equivalent to the wave you’d get if you thrust your arm forward while holding a slinky. It’s almost all P-wave and almost no S-wave. This is very distinctive against a background of earthquakes. The CTBTO is kind enough to show what this difference looks like; in this image, the top event is a nuclear test and the bottom event is an earthquake of a similar magnitude in a similar location (I apologize for making you click through to see the image, but I don’t host copyrighted images here).

There’s one further way that the waves from nuclear explosions stand out. They’re caused by a single point source, rather than kilometers of rock. This means that when many seismic stations work together to find the cause of a particular wave, they’re actually able to pinpoint the source of any explosion, rather than finding a broad front like they would for an earthquake.

The fifty IMS stations automatically provide a continuous stream of data to the CTBTO, which sifts through this data for any events that are overwhelmingly P-Waves and have a point source. Further confirmation then comes from the 120 auxiliary stations, which provide data on request. Various national and university seismometer programs get in on this too (probably because it’s good for public relations and therefore helps to justify their budgets), which is why it’s not uncommon to see several estimates of yield soon after seismographs pick up on nuclear tests.

Space Based Monitoring

This is the only type of monitoring that isn’t done by the CTBTO Preparatory Commission, which means that it is handled by state actors – whose interests necessarily veer more towards intelligence gathering than monitoring treaty obligations per se.

The United States began its space based monitoring program in response to the Limited Test Ban Treaty, which left verification explicitly to the major parties involved. The CTBTO Preparatory Commission was actually formed in response to a different treaty, the Comprehensive Test Ban Treaty, which is not fully in force yet (hence why the organization ensuring compliance with it is called the “Preparatory Commission”).

The United States first fulfilled its verification obligations with the Vela satellites, which were equipped with gamma-ray detectors, x-ray detectors, electromagnetic pulse detectors (which can detect the electro-magnetic pulse from high-altitude nuclear detonations) and an optical sensor called a bhangmeter.

Bhangmeters (the name is a reference to a strain of marijuana, with the implied subtext that you’d have to be high to believe they would work) are composed of a photodiode (a device that produces current when illuminated), a timer, and some filtering components. Bhangmeters are set up to look for the distinctive nuclear “double flash“, caused when the air compressed in a nuclear blast briefly obscuring the central fireball.

The bigger a nuclear explosion, the larger the compression and the longer the central fireball is obscured. The timer picks up on this, estimating nuclear yield from the delay between the initial light and its return.

The bhangmeter works because very few natural (or human) phenomena produce flashes that are as bright or distinctive as nuclear detonations. A properly calibrated bhangmeter will filter out continuous phenomena like lightning (or will find them too faint to detect). Other very bright events, like comets breaking up in the upper atmosphere, only provide a single flash.

There’s only been one possible false positive since the bhangmeters went live in 1967; a double flash was detected in the Southern Indian Ocean, but repeated sorties by the WC-135s detected no radionuclides. The event has never been conclusively proved to be nuclear or non-nuclear in origin and remains one of the great unsolved mysteries of age of widespread atomic testing.

By the time of this (possible) false positive, the bhangmeters had also detected 41 genuine nuclear tests.

The Vela satellites are no longer in service, but the key technology they carried (bhangmeters, x-ray detectors, and EMP detectors) lives on in the US GPS satellite constellation, which does double duty as its space-based nuclear sentinels.

One last note of historical errata: when looking into unexplained gamma-ray readings produced by the Vela satellites, US scientists discovered gamma-ray bursts, an energetic astronomical phenomenon associated with supernovas and merging binary stars.

Hydroacoustic Monitoring

Undersea explosions don’t have a double flash, because steam and turbulence quickly obscure the central fireball and don’t clear until well after the fireball has subsided. It’s true that radionuclide detection should eventually turn up evidence of any undersea nuclear tests, but it’s still useful to have a more immediate detection mechanism. That’s where hydroacoustic monitoring comes in.

There are actually two types of hydroacoustic monitoring. There’s six stations that use true underwater monitoring with triplets of hydrophones (so that signal direction can be determined via triangulation) which are very sensitive, but also very expensive (as hydrophones must be installed at a depth of approximately one kilometer, where sound transmission is best). There’s also five land based stations, which use seismographs on steeply sloped islands to detect the seismic waves underwater sounds make when they hit land. Land based monitoring is less accurate, but requires little in the way of specialized hardware, making it much cheaper.

In either case, data is streamed directly to CTBTO headquarters in Vienna, where it is analyzed and forwarded to states that are party to the CTB. At the CTBTO, the signal is split into different channels based on a known library of undersea sounds and explosions are  separated from natural phenomena (like volcanos, tsunamis, and whales) and man-made noises (like gas exploration, commercial shipping, and military drills). Signal processing and analysis – especially of hydrophone data – is a very mature field, so the CTBTO doesn’t lacks for techniques to refine its estimates of events.

Infrasound Monitoring

Infrasound monitoring stations are the last part of the global monitoring system and represent the best way for the CTBTO (rather than national governments with the resources to launch satellites) to detect atmospheric nuclear tests. Infrasound stations try to pick up the very low frequency sound waves created by nuclear explosions – and a host of other things, like volcanos, planes, and mining.

A key consideration with infrasound stations is reducing background noise. For this, being far away from human habitation and blocked from the wind is ideal. Whenever this cannot be accomplished (e.g. there’s very little cover from the wind in Antarctica, where several of the sixty stations are), more infrasound arrays are needed.

The components of the infrasound arrays look very weird.

Specifically, they look like a bunker that tried to eat four Ferris wheels. Each array actually contains three to eight of these monstrosities. From the CTBTO via Wikimedia Commons.



What you see here are a bunch of pipes that all feed through to a central microbarometer, which is what actually measures the infrasound by detecting slight changes in air pressure. This setup filters out a lot of the wind noise and mostly just lets infrasound through.

Like the hydroacoustic monitoring system, data is sent to the CTBTO in real time and analyzed there, presumably drawing on a similar library of recorded nuclear test detonations and employing many of the same signal processing techniques.

Ongoing research into wind noise reduction might eventually make the whole set of stations much more sensitive than it is now. Still, even the current iteration of infrasound monitoring should be enough to detect any nuclear tests in the lower atmosphere.

The CTBTO has a truly great website that really helped me put together this blog post. They provide a basic overview of the four international monitoring systems I described here (they don’t cover space-based monitoring because it’s outside of their remit), as well as pictures, a glossary, and a primer on the analysis they do. If you’d like to read more about how the international monitoring system works and how it came into being, I recommend visiting their website.

This post, like many of the posts in my nuclear weapon series came about because someone asked me a question about nuclear weapons and I found I couldn’t answer quite as authoritatively as I would have liked. Consequently, I’d like to thank Cody Wild and Tessa Alexanian for giving me the impetus to write this.

This post is part of a series on special topics in nuclear weapons. The index for all of my writing on nuclear weapons can be found here. Previous special topics posts include laser enrichment and the North Korean nuclear program.