magicstix
09-02-14, 06:59 PM
Since there had been some questions on the various paths in Dangerous Waters, I thought I'd do a little tutorial on basic underwater acoustics.
I've put together some plots of various sound propagation paths and the sound speed profiles that give rise to them.
Hopefully it'll prove interesting to the curious and helpful to indie devs trying to make their sims more realistic (also, I wanted to do testing on the raytracer I've been writing ;P).
Convergence zone
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7685
Convergence zones need *VERY* deep water. The profile here is a fairly common one for the deep ocean, and as you can see, sound bends down, then back up to the surface. This results in "rings" of detection roughly every 65 kyd or so. The range plotted here is 80 km (80 kyd), and the water depth is ~8000 meters.
Bottom bounce
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7689
Bottom bounce is just like it sounds. The propagation is a bounce off of the bottom. In general you need a sound speed profile that is neutral (same speed everywhere) or slightly downward refracting. Like a CZ, you get rings of detection. However because the sound is reflecting off the bottom, you get much more loss. The rings also tend to be much closer to the source, and the range they appear is related to the bottom depth. Here the plot range is 13 km (13 kyd) and the bottom depth is a relatively shallow 300 m.
Surface duct
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7686
This is the type of propagation that leads to the infamous "layer" that submarines like to hide below. For a surface duct to form, there must be a "notch" or "sweet spot" of sorts where the sound speed is at a minimum, surrounded by higher speeds below the notch. Here the sound speed minimum is at 100 meters, and the target depth is 50 meters. As you can see, sound cannot pass below the layer except at extremely steep angles (which only occur at close range). The sound is forced to bend away from deeper depths, causing it to propagate long distances near the surface. This is also one of the only propagation paths that result in good detection at all ranges (in the duct), unlike CZ and bottom bounce which lead to rings of good detection with poor detection in between.
Now lets see what happens when the source moves below the layer:
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7687
This is the same setup as before, but now the source is at a depth of 250 meters, which is below the layer. Notice how all of the sound is refracting away from the layer and bending downwards. In this situation, nothing above the layer can hear what's below it. Also note how in this particular sound speed profile that receivers below the layer would also have a hard time hearing the source, since all of the energy is being refracted down into the bottom. A submarine in this situation couldn't be detected by a surface ship, but the submarine would also be almost blind to anything going on below the layer.
Half channel duct
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7688
A half channel duct is like a surface duct that goes almost to the bottom. These form in shallow waters with upwardly refracting sound speed profiles. This is probably the absolute worst place to operate if you're a submarine, since the ducting creates very good propagation out to hundreds of kilometers at all depths. Just looking at this plot, one can see that a submarine really has no place to hide and is easily detected at great range.
Keep in mind that the ocean can, and often does, support more than one type of path. This can lead to interesting multi-path effects, where you can get ghost contacts that can make a submarine (or surface ship for that matter) appear in multiple places at once.
I've put together some plots of various sound propagation paths and the sound speed profiles that give rise to them.
Hopefully it'll prove interesting to the curious and helpful to indie devs trying to make their sims more realistic (also, I wanted to do testing on the raytracer I've been writing ;P).
Convergence zone
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7685
Convergence zones need *VERY* deep water. The profile here is a fairly common one for the deep ocean, and as you can see, sound bends down, then back up to the surface. This results in "rings" of detection roughly every 65 kyd or so. The range plotted here is 80 km (80 kyd), and the water depth is ~8000 meters.
Bottom bounce
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7689
Bottom bounce is just like it sounds. The propagation is a bounce off of the bottom. In general you need a sound speed profile that is neutral (same speed everywhere) or slightly downward refracting. Like a CZ, you get rings of detection. However because the sound is reflecting off the bottom, you get much more loss. The rings also tend to be much closer to the source, and the range they appear is related to the bottom depth. Here the plot range is 13 km (13 kyd) and the bottom depth is a relatively shallow 300 m.
Surface duct
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7686
This is the type of propagation that leads to the infamous "layer" that submarines like to hide below. For a surface duct to form, there must be a "notch" or "sweet spot" of sorts where the sound speed is at a minimum, surrounded by higher speeds below the notch. Here the sound speed minimum is at 100 meters, and the target depth is 50 meters. As you can see, sound cannot pass below the layer except at extremely steep angles (which only occur at close range). The sound is forced to bend away from deeper depths, causing it to propagate long distances near the surface. This is also one of the only propagation paths that result in good detection at all ranges (in the duct), unlike CZ and bottom bounce which lead to rings of good detection with poor detection in between.
Now lets see what happens when the source moves below the layer:
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7687
This is the same setup as before, but now the source is at a depth of 250 meters, which is below the layer. Notice how all of the sound is refracting away from the layer and bending downwards. In this situation, nothing above the layer can hear what's below it. Also note how in this particular sound speed profile that receivers below the layer would also have a hard time hearing the source, since all of the energy is being refracted down into the bottom. A submarine in this situation couldn't be detected by a surface ship, but the submarine would also be almost blind to anything going on below the layer.
Half channel duct
http://www.subsim.com/radioroom/picture.php?albumid=908&pictureid=7688
A half channel duct is like a surface duct that goes almost to the bottom. These form in shallow waters with upwardly refracting sound speed profiles. This is probably the absolute worst place to operate if you're a submarine, since the ducting creates very good propagation out to hundreds of kilometers at all depths. Just looking at this plot, one can see that a submarine really has no place to hide and is easily detected at great range.
Keep in mind that the ocean can, and often does, support more than one type of path. This can lead to interesting multi-path effects, where you can get ghost contacts that can make a submarine (or surface ship for that matter) appear in multiple places at once.