And that assumes a worst case scenario of your transducer’s acoustics being perfectly directional. Airmar’s DST800 installation instructions say “Best performance on hull deadrise angle through 7°. Accommodates up to a 22° deadrise angle.” I’d take that to mean the signal transmission and detection operate in a 7 degree cone, so you can subtract 7 degrees from the actual deadrise in your cosine calculation (if all transducers have a similar beam pattern).Nice catch @Mike Oldak . When you stop and see it in real life it's not that big a deal as long as the bottom is realtively horizontal.
IMO, a simpler way of looking at this error in depth is to go back to basic trig :
Just multiply what you're seeing on your depth instrument by the cosine of the angle of your transducer face to horizontal (which never changes) and you'll have your true depth, in this example COS 20° = 0.94.
The OP’s case is probably on the more extreme end of installation locations. 22 degrees in one axis and 17 in the other is a net deflection of about 27 degrees (according to my friendly AI helper). Subtracting 7 degrees for a transducer’s acoustic cone leaves us with a 20 degree angle to compensate for (when the boat isn’t heeling). The OP has a Hunter 34 in the Chesapeake, so he probably has a shoal draft of 4.25 feet. The transducer is probably mounted 1.25 feet below the surface. So when the boat is aground an ideally installed transducer will read 3’, but an uncompensated transducer would read 3.2’. I suppose there probably aren’t too many anchorages or channels where that 3 inches is going to make all the difference, but I guess if you can be more precise why not try to.