Planet-wide vocalizations: is there a natural limit preventing it?

Question

Premise

I recently recalled when the Chelyabinsk meteor crashed overhead in 2013. The pressure wave was immense. The infrasounds, as recorded by nuclear missile detectors, appeared to traverse the entirety of the Earth.

Then I began to wonder, could infrasounds from an organism also traverse the entire Earth? The key word here of course is "could." While not an easy task to undertake, let's attempt to prescribe very specific circumstances to achieve this bio-acoustic task and then objectively evaluate them pragmatically. From this, we might conclude this to be feasible, or we might learn that even under these ideal circumstances it's not feasible. Without further ado, here is my research summary on bio-acoustics and a graph I compiled using data from Wikipedia.

• Low frequency sounds travel farther (less energy lost to molecules in medium)
• Speed of sound is faster in water (molecules closer together)
• Loudest sustained sound has natural limit of 194 decibels (greatest possible oscillation of air pressure, vacuum to 2 atmospheres)
• Underwater, sounds can be louder, note marine animals on chart ^{I made this graph from scratch using javascript. Data is from Wikipedia, but other than that, credit goes to ME!! (evil laugh)}

What can be seen from the graph is that while not all loud animals are massive in size, but the one's with the lowest frequency calls are massive in size.

What I'm not sure is if it would be more realistic if I change things from the biology side:

• creature size
• power of vocal folds

Or should I try solving from the environment side:

• atmospheric conditions
• marine conditions

Question

Is there a natural barrier preventing planet-wide spanning vocalizations? As it stands, our best hope: the Sperm Whale, can communicate over 1000 miles away, but that's only 1/24th of the Earth's circumference. Natural history spans billions of years and whales are the biggest creatures to date, so I'm worried that there could be a natural barrier to going way bigger.

I'd like to stay as close as possible to known science. A few things I think might (these are just my hunches, I could be wrong) have bearing to keeping it scientifically plausible:

• square cubed law
• vocal fold to the sound's maximum distance ratio (is there a conventional understanding of this?)

Further Clarifications:

• Success Metric: Sound powerful enough to travel the circumference of the earth, which is approximately 24,000 miles. Note: anatomic diagram is not required, a general description of the theoretical organism would be fine.
• Planet info: Planet is earth-like in size only, other conditions can be modified
• Biome: Ocean or land animal is acceptable, though it seems ocean is more plausible at the moment

• Method: A vocalization in the true sense is preferred (emitted from vocal folds), however, if this is a total dead-end, try suggesting other forms of communication

• Evolutionary need: You might be thinking to yourself, why would a creature need to do this? Well, we will leave this out of scope for simplicity.

• Danger to surrounding life: Don't hold back out of concern for other life; collateral damage is acceptable. On a side note, to avert outright sterility we can suppose some portion of life has a coping mechanism, finding shelter, etc.

Answer 1

Possible, yes, but ...

The theory of long range communication by acoustics is very similar to that by radio, or indeed a number of other channels.

To reason that communication s more difficult at very long range is that some noise sources are at a more-or-less constant power density, whereas the signal power density declines with distance -- typically as $exp(-\alpha r)/r^2$, where r is the range. The term $exp(-\alpha r)$ is an exponential absorption loss, where some of the signal is absorbed by the environment, while the $1/r^2$ is the geometric term, related to the signal spreading out into space.

The "brute force" way to overcome these losses is to cram a really enormous power into the transmitted signal. However looking back at those loss terms, you can see that it can take a huge power input to achieve even a small increase in range. This is a problem, biologically: you are investing scarce resources for little gain.

And, what is that gain? Animals that can call for long distances are either creatures with a solitary habit, joining up with mates for mating season; or pack animals, co-ordinating their pack. Whales have evolved the ability to call very long distances because, as huge apex predators, their number density can be very low. If it has swum a long way in search of food, it might need to call 2400 miles to find a pod again for mating. (And "fortunately", the huge size also gives it the power to do so.) The only value in calling the other side of the world is if there are only two of you, and your only potential mate might be at the antipodes.

Channels

Fortunately, there are much smarter ways to increase your signal:noise ratio, and nature has found several of them. One of them is to transmit into a channel which stops your signal spreading out. For a pipe-like channel this can get rid of that $1/r^2$ term, giving huge gains. For a sheet-like channel (which is more common in practice), it changes it to a $1/r$ term, which is still much better. Sheet-like channels are in fact found in the oceans between thermoclines and haloclines, they are actually used, and they are part of the reason whale song can be detected at thousands of miles.

A potential disadvantage of a channel is that if the receiver is outside it, they get less signal. However, sheet-like channels tend to be arranged horizontally, so being in the channel layer can be as simple as having a preferred depth.

There is another important acoustic channel that we know about, which Nature never seems to have exploited: upper atmosphere thermoclines. These, in fact, are part of the reason that infrasound from meteors can be detected globally. The existence of these channels was kpet secret for a long time, because they were (and probably still are) being used to detect nuclear detonations.

These channels have another important advantage over marine sound channels, which is related to the second method to increase range: your signal:noise ratio gets better if there is less noise. These upper atmosphere channels are, apparently, deathly quiet. Those detection systems we know about tended to use balloons, with the rigging specially designed to eliminate creaking.

So why aren't they used by animals? Well, maybe some do: bar-headed geese have been detected calling to each other at nearly 28,000 feet. But generally speaking, to really exploit an acoustic channel like this, you also need a lot of acoustic power, and that needs a big animal, and that makes flight unlikely.

Directional transmission

Another method of improving received signal strength is directionality: instead of being the signal out in all directions, you focus it toward the receiver; or the receiver turns its ears toward the sender. Even modest amounts of directionality can easily double the signal strength, but the upper limit is defined only by the ratio of the physical size of the directing structure (mouth, or ears) to the wavelength. Hence, a highly directional signal can be easily achieved in the ultrasonic region, but to get directional infrasound would require an impractically immense structure.

For mobile stations -- which includes most animals -- a disadvantage is that you need to know what direction to use. One solution is for each end to have a brief, high power "hailing signal" without directionality, which allows everyone to turn towards each other, and then continue the conversation from there.

Signalling to the antipodes is a special case. You still have to deal with the $exp(-\alpha r)$ absorption term, but provided the signal is trapped in the atmosphere (all acoustic signals are), all radiating paths re-converge at the antipodes; you get the advantage of directional transmitters and receivers without turning in any particular direction.

That's the simple stuff. Now we get subtle

For ultra-long range radio communications -- not pole to pole, but Earth to Voyager -- or for radio communications in a non-permissive environment (i.e. someone is jamming you), you need to change the signal itself, so that it becomes intelligible at lower signal:noise ratios. There are a number of ways to do this, of increasing technical sophistication. They can all be considered special cases of modulation methods.

• Slowing down the transmission;

• Repeating the message;

• Sending the same message on multiple frequencies, directions, or times (the latter is the same as repeating);

• Encoding the message so that partial reception can enable reconstruction of the whole (error correcting codes);

• Compressing the signal energy into the time slot, frequency or space the receiver expects. (In the case of space compression, this is the same as directionality.)

• Modulating the signal carrier so that a tuned filter can distinguish it better from noise.

The last is the most interesting -- it is commonly called "spread spectrum transmission", and it can achieve amazing virtual signal gains by effectively multiplexing huge signal bandwidths.

Now, here's the kicker: my wife's lovebird used several of these methods. A tiny bird, it was capable of "calling all lovebirds" across a very wide area, by using very efficient signalling techniques including spectrum spreading (chirp modulation), time slot diversifaction, and power pulsing.

An additional advantage of spread spectrum modulation is that for receivers with a tuned receiver (your own species) it improves reception; for those without the correct tuning (your predators, you hope), it degrades it.