"Erik's Horns" or "Railroad Tracks" in EM swath data near nadir (testing ISSUES workflow) about community

[kjerram]

Adding a little more context... Again, from the EM/SIS notes and EX2101 report:

The system appears to be detecting high-amplitude returns from the specular reflection (i.e., near-nadir seafloor that is directly facing the echosounder) and tracking this return on RX sidelobes at a fairly constant range (probably limited to within the TX pulse length) out to several degrees on port and starboard.

The Penetration Filter may compound this artifact by encouraging the bottom detection process to select shallower candidates for each RX beam near nadir, such that the reported bottom detections follow the constant-range trend as it moves above the seabed out to several degrees on each side. Subsets below show Erik’s Horns (top) with normal runtime parameters (top) and after tilt and filter adjustments (bottom).

[kjerram]

Proposed solution, again from the aforementioned notes and EX2101 report:

One method of reducing this artifact is to slightly reduce the nadir reflection strength by tilting the TX direction forward or aft by up to a few degrees. The highest-amplitude return is directed slightly away from the sonar, allowing the bottom detection process to more clearly/quickly distinguish between the seafloor (at the correct range for each RX beam main lobe) and the high amplitude of the first return (at a constant range across many RX beam sidelobes near nadir).

[kjerram]

Looks better! Closing the issue and posting the solution on the Troubleshooting page:
https://github.com/oceanmapping/community/wiki/Troubleshooting-(Issues)#railroad-tracks--eriks-horns