That was just a quick reply, there's more to it. This is according to research and experience, but it's inly my opinion:
I get this with my Infinium, it took me a while for me to get to the bottom of this.
The salt water you're detecting is mineralised enough to cause noises. These are bits of mineralisation float in the water moving up and down as the waves move up and down. They are also lower on the conductivity scale than iron, just below, but en-mass, they cause a lot of noise, given you have a large body of water around you (over your coil) and your coil picks up signals above it as well.
This seems to be the case for all PI's according to research, but I can only speak for the Infinium. It does not seem to cause noise to VLF water detectors. When I had an Excalibur 2, it was quiet in the water, but the performance decreased (I could also hear the threshold change, both the threshold and signals became duller). Even though the Excalibur was quieter, I was still going deeper with a noisy Infinium and getting more targets. However, I never compared them side by side, but I was consistently doing better in the water with the Infinium. bare in mind, these areas I hunted only had moderate amounts of junk.
1) Find a website that shows the wind speed and wind direction for the beaches that you're detecting. You want off shore winds or very little winds at all if they're not off shore. Choose those days to hunt, as the waters will be calmer. That way you might not have to detect with any pulse delay at all or maybe a little!
2) Depending on how noisy the water is, turn the pulse delay up, but not too high. On the Infinium, the pulse delay is called "discrimination", but that is false advertising. On the sand shark it's called pulse width. The DF has one also. Don't turn it too high otherwise you will loose too much depth. Turn it too low and it might be too noisy that you'll loose targets.