Let's dive a bit more into how noise arrives at YGH and how it's processed. Seven geysers show up reliably in YGH’s infrasound or seismic data: Old Faithful, Beehive, Lion, Castle, Grand, Giant (with its neighbor Mastiff) and Oblong. Here’s what each one looks like in the traces.

In an ideal world the sound from these features would be a perfect wave of consistent frequency and arrive undistorted and with no overlap. None of those cases are true. The infrasound emitted by geysers is a mixture of frequencies ranging from less than 1Hz up to 20Hz. For a visual comparison, a sound wave of 6Hz has a wavelength of about 50 meters, this is about the same as the wavelength of the typical ocean wave breaking on a shoreline. Imagine an ocean wave moving at you at ~650 miles per hour! With waves of that length its difficult for the sensor array to determine exactly what direction the sound originates from. Given the size of the sensor array noises kind of arrive from a vague direction and so multiple features can all kind of sound like they come from the same general spot. For YGH there seems to be three primary zones from which noise arrives: Geyser Hill to the south, Grand and Castle to the West, and Giant to the northwest.

The final result of this setup is that noise calculated to be coming from Old Faithful will also show up on the noise coming from Lion, Beehive, and Giantess. The YGH Monitor calculates the noise level above background levels, and so when background levels are really quiet the noise can still be kind of separated. Noise from Old Faithful may show up on the plots for Giantess, Lion, and Beehive, but it will be loudest for Old Faithful. Under windy and noisy conditions this gets muddied and harder to separate.

Sound complicated? I guess it is, to a degree.

On a geyser-by-geyser basis its actually not that hard and we will look at each one in the following sections.

The figure above is what quiet looks like in the southern Upper Geyser Basin. At 17:15 on Thursday, 12 November 2026, nothing was making any noise. The park was closed, there was no vehicle traffic, no air traffic, no wind, and no geyser erupting. Only the Firehole River would have been making noise.

All of the features here have their sounds overlap with each other. Determining which feature is erupting needs timing and context clues.

Old Faithful is about 2,300 feet south-southeast of the array. Its eruptions leave a consistent pattern: the sound rises sharply for about 30 seconds, then slowly fades. That matches the eruption itself, which climbs to full height and strength in its first 20 to 30 seconds before slowly winding down. Once the column drops below about 20 or 30 feet, the sound fades into the background.

Old Faithful has two general types of eruptions: long and short. In both cases the eruption starts identically, climbs to its full height, and then begins dropping after about 30 seconds. The behavior after the drop is what differentiates them. In a short mode eruption the column keeps dropping back into the vent and the geyser simply stops erupting. In a long mode eruption the column drops to about 20-30 feet in height and holds there for another 2-3 minutes before gradually changing into a weak steamphase. .

The tail off of the long mode eruptions does generate infrasound, but it is generally quite low and only evident on very quiet days when no other feature is making any noise. Currently, there is no way to reliably differentiate long and short mode eruptions using the YGH Monitor.

Old Faithful is one of the quieter features in the Upper Geyser Basin, especially for it's size. It can be muffled almost completely by wind or when other features erupt. Here is an example of where Old Faithful has been distorted or muffled that covers many similar situations.

In the above example Old Faithful erupted during an already very noisy period (in this case, during an eruption of Grand Geyser, a much louder geyser!). It's infrasound trace was muffled somewhat and only rose to just over 2 times background noise. So much noise was hitting the sensors from various directions that the noise level breifly became yellow - indicating that signals may get distorted. In this case the webcam was able to confirm Old Faithful was erupting - but what if it was dark and we couldn't verify it? This is where context clues come into play.

Now we've turned on all of the infrasound traces looking south from YGH towards Geyser Hill. What we see is that the infrasound was also recorded coming from Beehive and Lion. It's important to note here that the loudest noise is coming from the Old Faithful direction, and the spikes happen at the same time. If this was a plane (which makes a similar noise trace) we would see the spikes be at different times in different directions.

But could it be a truck, or some other loud noise from the Old Faithful visitor area? Possibly! Not impossible! But there is one more critical context clue: the last Old Faithful was at 09:50 and was a short mode eruption, Old Faithful was expected to erupt at this time!

Cases FOR an eruption of Old Faithful:

1. Old Faithful was expected at this time.
2. Trace has the same shape and duration as an eruption of Old Faithful.
3. Sound comes from the south in all directions at the same time, so its not a moving object.
4. Sound is loudest when pointing at Old Faithful.

Cases AGAINST an eruption of Old Faithful:

1. A very loud truck, distant thunder, or a plane could make this noise

The odds are that this was an eruption of Old Faithful - even without visualy confirming it. More evidence would be gathered if the next eruption of Old Faithful followed at the expected time (which it did).

Giantess has not erupted since the installation of YGH. We don't know what it will look like on infrasound, but it should be loud and energetic enough to be on both the infrasound and seismic data streams. Giantess begins with a series of violent thumping (like Doublet, but with a piledriver instead of a hammer) before starting a very loud eruption. Previous eruptions have been recorded on the existing seismometers near Old Faithful. Here is hoping we find out soon enough!

Beehive is closer, about 1,600 feet away and nearly due south, and it’s very loud. Its sound shows up in the data traces coming from Lion, Giantess, and Old Faithful. In general, the magnitude of the noise is much higher. Beehive is unusual in that its strongest noises come in the middle and toward the end of its eruptions. Its rocket-like jet starts out heavy with water, then shifts partway through its four to five minute eruption into a steamy spray. That transition brings powerful bursts of pressurized steam, and a chest-pounding, bass-heavy roar you can feel as much as hear. The infrasound sensors get hit hard by it.

Occaisonally Old Faithful and Beehive erupt at the same time and their sounds overlap. We've yet to record an instance of a true simultaneous start, but we do have two instances where Beehive started near the end of Old Faithful's acoustic trace producing a distinct double-peak. What is the tell-tale sign here is that when Old Faithful erupts it's noise is higher, then when Beehive erupts it's noise becomes the louder one. You can see the switch happen on the traces.

Use caution when back calculating the start time for Beehive. Both of these instances really only provide an 'E ie' time (Electronic, in-eruption) for Beehive. We know that Beehive does not instantly jump to near full volume - it takes at least 10-30 seconds - and we know it probably didn't start simultaneously with Old Faithful or else there would not be a dip and reversal in the noise levels. If you're uncomfortable doing this then use the reversal time and report it as E, ie - include a screenshot!

Lion is the closest geyser to YGH, about 1,100 feet to the south-southwest. At times it lets out deep-throated roars of steam just before it erupts, which gave it it's current name. Lion erupts in a series with a larger initial eruption, followed by one or more smaller ones. Eruptions start with water sputtering sluggishly from the vent, then build in strength and volume. The initial eruption ends in a powerful steam phase, while the follow-ups end more quietly. All of them show up in the infrasound as the classic slow ramp of rising sound. The initial eruption stands out because its steam phase keeps the sound high for longer.

Lion Geyser is also capable of roaring without starting an eruption. The roars are usually brief, less than 10 seconds, but fairly loud. Ultimately they seem to be too breif to show up on infrasound. Below is an example of a roar which was observed on the webcam under exceptionally quiet conditions. The roar is indistinguishable from the background noise.

Sometimes a roar does turn into a small eruption with water sputtering and jetting up to 5-10 feet for as much as a minute before dying out. These typically happen near the end of series when follow up eruptions try and fail to initiate. They are visible on the infrasound traces under ideal conditions but without visual confirmation or a follow up eruptions should be considered questionable observations. Perhaps with time more data can be collected to refine observing Lion minors.

Doublet Pool is a large hot spring in the middle of Geyser Hill with two pools separated by a sinter bridge. Doublet Pool rarely erupts but it does have a cycle involving a subterranean eruption of sorts. Every 20-30 minutes enough energy forms in Doublet Pool to allow superheated steam bubbles to form and collapse created a series of rapid thumping that can be phsyically felt and audibly heard from the trail next to it. These thumps shake and rock the surface of the pool.

Doublet Pool's thumps send out relatively high frequency vertical seismic waves which YGH readily picks up. It is visible on the Seismic Viewer with the Doublet preset checked. Under very quiet conditions Doublet Pool's thumping will be visible in the infrasound data. It is strongest in the direction of Giantess - which is almost perfectly inline with Doublet Pool - and weaker on the Old Faithful and Beehive bearings.

To the south and west of YGH are two large and loud geysers and a cluster of energetic but small geysers between them. The two large ones, Castle and Grand, are two of the stars of the Upper Geyser Basin; Grand Geyser is probably the tallest most frequently active geyser ever known. These two features readily show up on infrasound. Between them is the Sawmill Complex, which includes both Spasmodic and Sawmill Geysers. These two geysers are quite close to YGH, but not enough for them to be reliably heard by the sensor.

Castle is about 2,000 feet to the southwest. Its eruptions start suddenly and pulse with sound and energy. It has two kinds of eruptions. All eruptions begin with a water phase which pulses between periods of strong jetting up to 80 feet and near total quiesence. Some of these pauses can last multiple minutes. Minor eruptions happen when one of these pauses never restarts. Major eruptions occur when the quiet pulses are replaces by increasingly powerful blasts of steam. After about 20 minutes this gives way into a steady and loud steam phase. Both kinds show up clearly in the infrasound, and, as with Lion, the steam phase of a major eruption is plain to see.

Castle Geyser can have very long pauses during it's eruption. Essentially it seems like Castle tries to stall into a minor but fails and rebounds back into a major eruption. These are visible in the infrasound; the pauses typically have a very abrupt onset and ending. It is possible for there to be false pauses in the infrasound data if some other, louder, feature makes noise in a different direction. Most commonly this is Beehive or Old Faithful Geysers, whose eruptions make Castle's noise level relative to background drop.

Grand is almost due west, about 1,700 feet away. It and the features beyond it to the north are harder to tell apart. Grand and Giant are only about 25 degrees apart as seen from the array, close enough that each one’s sound leaks a little into the other’s beam. Grand’s sound shows up faintly on Giant’s line, and Giant’s on Grand’s. Grand is still clearest in its own direction, and its eruptions stand out plainly. The sound level closely follows the energy of the eruption, including the moments when Grand seems to slow down. Grand is also unusual in being able to stop and restart its eruptions in a series of distinct bursts. The pauses between bursts can bring activity to a complete stop for up to a minute, and the infrasound drops almost to nothing with it.

The Grand Checker is an application within the YGH Monitor which is an experimental tool for trying to determine what dips in Grand's sound are real pauses versus eruption slow downs or interference from other features. A growing body of observed and verified single and multiburst eruptions are being used to build a model of what both look like on the infrasound monitor to compare to new eruptions which can not be observed due to weather, nightfall, or the camera being pointed elsewhere.

The Sawmill Complex is a cluster of interconnected springs and geysers south of Grand and more or less in between Grand and Castle (as far as YGH is concerned). This complex is highly energetic and active with something erupting in it about half of the time. The groups is (acoustically) dominated by Sawmill Geyser and Spasmodic Geyser. Sawmill Geyser erupts for up to 4 hours with repetitive violent and concussive explosions of water. Spasmodic Geyser erupts from a dozen vents with a combination of heavy bursting and jetting. This high amount of energy is apparently one of the problems with the group - it just always makes noise and as such it seems to fall into the backgorund. It doesn't help that a ridgeline is between the features and YGH which partially blocks the noise. Under ideal conditions the noise from Sawmill Geyser has been faintly visible, but there is no reliable 'signal' from Sawmill as yet. I've included it here because Sawmill is an important feature and I have been asked about its visbiility in infrasound.

Giant is, well, giant. It’s the second-tallest active geyser in the world, after Steamboat, with eruptions reaching about 250 feet, and the sheer volume of water is hard to describe. Both the eruption column, with its roar of escaping steam, and the tons of water crashing back to the ground make enough sound for the infrasound sensors. There’s no mistaking a Giant eruption, whether on the sensors or in person. At about 3,000 feet to the northwest it’s the farthest geyser from the array, and also the loudest.

Next to Giant is Mastiff, a key player in the buildup to Giant’s rare eruptions. Mastiff is a significant geyser in its own right, with two vents reaching up to 50 feet and arcing 20 to 30 feet across. It sometimes erupts in the minutes before Giant, and it often, though not always, shows up in the infrasound. Of the four verified eruptions of Mastiff, three were visible in infrasound data. It sits on the same platform as Giant, so the array hears the two from the same direction.

Oblong is probably the hardest geyser to observe from YGH. It’s about 2,600 feet to the west-northwest, and its activity is a long-term indicator for Giant: when Oblong erupts more often, Giant usually becomes more active. Oblong isn’t a particularly large or loud geyser. Its eruptions are mostly heavy boiling and surging, so the infrasound sensors don’t reliably pick it up. But like Doublet Pool, Oblong thumps violently at depth during its eruptions, as large steam bubbles form and collapse in its plumbing. On the nearby trail you can feel it. Some thumps are strong enough that you feel them travel up through your feet and legs into your spine, and you can even hear the pounding. Those thumps travel through the ground as strong side-to-side shaking that reaches the YGH seismometer. Even when the infrasound can’t hear Oblong, the seismometer feels it.

Oblong is also unqiue in that it is probably the only feature where the seismometer provides the most accurate and consistent eruption start and end time - even over direct observation. The eruption of Oblong starts during one of Oblong's periodic episodes of superheated boiling. This boiling surges slightly and a flood of water goes across the platform and is accomapnied by the start of the heavy thumping. The eruption also ends when the thumping and boiling ceases. Under ideal conditions these two events are unmistakable in the field but during bad weather or when the boardwalks are full of people they can be difficult to determine. The situation for the webcam is at a severe disadvantage because the webcam can't actually see Oblong's pool and instead relies upon the sudden increase in steam which occurs in the last few seconds before water is visible. Even then, only when water is visible and near its full height will infrasound reach YGH. The seismic waves produced by the thumping - which only occurs during an eruption - reach the sensors up to 40 seconds before the webcam visually is able to recognize that Oblong may be starting an eruption, and up to 2 minutes or more before the infrasound arrives.

Oblong has a good chance of erupting while Doublet Pool is thumping. That’s not because the two are connected: Doublet simply thumps almost a quarter of the time. Looked at directly, the seismic record mixes the two together. Pull them apart and they’re distinct. Doublet’s vibration is high-pitched, at roughly 9 to 20 hertz, and mostly up and down. Oblong’s is a much lower 3 to 5 hertz and side to side. Filter for Oblong’s lower frequencies and its horizontal motion, and its eruptions stand out from Doublet’s thumping.
As an interesting aside, these waves are called Love Waves. If you were at YGH and were facing Oblong and could feel them you'd notice that the waves would rock you side to side - not towards and backwards from Oblong. These slow moving but high amplitude side to side Love Waves are the destructive waves felt at long distances from the epicenters of major earthquakes.

There are sources of noise which can quite effectively mimic geyser eruptions on infrasound. The most common ones are planes and noises from the Old Faithful Visitor Complex.

Aircraft routinely fly above and around the Upper Geyser Basin. Jetliners at cruising altitude, helicopters and light planes over the park, all generate infrasound. Their sounds tend to reach YGH as an increase and decrease in noise, often not entirely unlike a geyser erupting.

Planes are a moving object, and as they move the direction their sound comes from changes. In general, a plane will show up on the infrasound as a series of peaks in sound which share the same general shape but show a progression in peaks in time.

Loud noises from trucks and heavy equipment on the road can sometimes come through as pulses from the general direction of the intersection. Sometimes we don't really know what the source of the sound is, only that is probably didn't come from a geyser.