Basic ASW Tactics

[SeaQueen]

Originally Posted by Mr.Grinch

Could someone please explain to me, without going into the minutea of the sensor equipment the very basic steps they take when conducting an ASW search. Let's say for example you've just loaded up a Dangerous Water “Quick Mission”, you're in an improved Los Angeles class and there's empty ocean all around you...what do you do?

Assuming auto-crew for everything, (I turn it on and off as needed) Do you sit at the computer and wait for a sound contacts, do you spin and check your baffles?...how do you start your search?

Hehe, search tactics... you asked for it.

It depends on which type of quick mission you're performing. Is it the "area search" or is the "barrier search." If it's the "area search" then you want to plot a course through the area such that the area covered by your primary search sensor (usually your towed array) is equal to the size of the area. Usually you can do this by driving a ladder pattern. The spacing of the various waypoints depends on how far you think you can see. You can guestimate it from experience or you can do what I do, and use the mission editor. If you look in missions I design, I actually try to include that information as part of the briefing.

It doesn't really matter much what you do when you're in the box, statistically, though, because complicating factors such as target motion, maneuvering for TMA, flight ops, and unmasking weapons, means that you the area covered will almost always overlap itself a little bit no matter how carefully you plan the search. That has the effect of making almost every search equivilent to random searching. The formula for the cumulative probability of detection as a function of time is given by the Koopman random search equation.

Pd = 1-exp(-2R v t / A)

where R is the sensor range, v is your speed, t is time, and A is the size of the area you are searching. It's just the exponential distribution. You can find detailed derivations of it in Wagner's Naval Operations Analysis, or Koopman's Search and Screening. You might also be able to find it in SAR manuals. Search tactics for ASW and search tactics for search and rescue are basically the same ideas. You want to find something where you have only a rough idea of it's location as efficiently as possible. From that you should be able to make a pretty good estimate of how long it'll take you to find him, on average.

For the case of the barrier quick mission, you just need to drive back and fourth between the two waypoints. The probability of you succeeding in detecting a target depends on your speed, your sensor range, the length of the barrier and the target's speed. In the literature I just mentioned, they give formulas for the Pd of barriers as well. By making assumptions about the target's speed and your sensor range, you can often optimize your speed to maximize your probablity of detection, given the constraint of wanting to minimize your noise level.

Checking your baffles isn't necessarily the first thing you do, but it can't hurt. I actually have a random number generator in an Excel spreadsheet, to generate things like random times and bearings, so that I don't get too predictable. Some people do it regularly. One of the good things about your towed array, though, is that it gives you pretty good coverage of every place but the endfire forward. The baffles clearing thing is mostly an issue with submarines that lack a towed array.

Now you've got a sound contact, let's say you're picking up the lower frequencies from you TA, now what?...do you turn towards it and try to get closer so your hull/sphere array can pick him up?

That's one possible answer. The first thing you need to do if it's a TAPNB contact is figure out it's bearing. You do that by turning and waiting for the towed array to stabilize again. Notice how the target's relative bearing changes as it stabilizes. That should allow you to deduce the direction of the target.

Next, I'd start trying to figure out how far away he is in a vague sense. Not so much his exact range, but whether he's "near" or "far." If it's a low bearing rate target, then there's two possiblities, he's either on a collision course with me or he's a long way away. If I turn AWAY from him (don't turn towards him because if he's close you might bang into him), and the bearing rate rapidly increases after the towed array stabilizes, then he was on a collision course and may have even avoided hitting him. I had that happen once, and promised myself to never do it again. It was a sad. If the bearing rate stays relatively unchanged, then he's far away.

Next I'd figure out his direction of relative motion (DRM). This is just highschool geometry.

NOW I'm ready to plan my approach...

Somewhere in the process of approaching my target, I end up widdling down the area of uncertainty surrounding the contact to something I think is acceptable, then shooting at him.

Hopefully he doesn't get a sniff of me or my torpedoes, and counterfire at me so I can sneak away without having to evade anything.

Edit:

Found a paper on search tactics discussing computer modeling and historical results using U-boats in the Bay of Biscay. It's air searchers using the naked eye, instead of a sensor, but much of the essentials remain the same and there's a sample of search patterns in it.

http://www.informs-cs.org/wsc03paper...om%20search%22