Sonar interpretation - why things look the way they do on your screen

[Larry3215]

Here are some more shots that show how the overlap on sideview affects what you see on the screen in both side and downview and for comparison - 2D.

In the first drawing, this is a view from the back looking forward. Lets say there are a series of targets sitting on the bottom - rocks, fish etc. From left to ring A,B,C, D, E. For the moment - to keep things simple - lets pretend that the sideview over lap area in the center is also the downview area. In actual practice, the downview area would eb much wider as above.

Target C is dead center under the transducer. Remember the distance/depth thing? The second pic will show whats up with that.

Targets B and D are off to the sides a short distance but still within the DI cone range on both sides. Then, A and E are far enough to the sides to only be on the right or left. Lets say that in this case, the 2D cone is about the same width as the downview cone, so B and D are just within its outer edges also.



leftrighcenter.jpg

This next shot shows how those different targets would get drawn on the screen in all three views.

Target C is the closest to the transducer, so it gets drawn at the shallowest depth = closest to the center line on SI and the tallest point on down and 2D.

Targets A and E dont show up at all on down or 2D because they are too far out to the sides.

Because targets B and D are both within the left and right sideview zone, but further away from the transducer, they both get drawn on the sideview screen a little further from the center line AND because they are at the same distance from the transducer, they get drawn on top of each other.

In the down image, because they are slightly further from the transducer, B and D get drawn slightly below where the "bottom" line is and also on top of each other. Remeber that part that I have highlighted. Its important.

In the 2D shot, B and D might or might not show up in the 2D cone, so they might or might not get mixed into the bottom return below C. Ive drawn them as lower case because they would be weaker returns due to being further away, but they would be lost in the bottom line on 2D. They might get lost in the bottom return on downview as well, but that will depend on how strong the return is. I'll go into that more later.

The most important thing to note here on sideview is that things that are directly under the boat get drawn on BOTH sides of the sideview screen. Its also important that things that show up on both sides may look like they are off to the side of the boat but they really are directly under the boat.



sidedown2.jpg

[Larry3215]

Almost forgot - every example I've given so far for side and down assumes a perfectly FLAT bottom. Thats very very important. The software that is used to draw the screen images for side and downview always assumes that bottom is flat.

Always.

It cant assume anything else because it has no way to tell if the bottom is flat or not. Plus, the math is impossible unless you assume the bottom is flat OR you know the angle to the target from the transducer. The only transducers that know angles are Panoptix or other multi-beam transducers. Normal 2D/Down/Sideview transducers only know distance. They have no clue about angles.

What this means in the real world is that distances and measurements taken from side/downview screens will only be accurate if the bottom is flat. That also applies to waypoints marked off to the side of the boat. They will be off to the degree that the bottom is not flat. Waypoints marked on uphill sides of sideview will be closer than they are on the chart. Waypoints on the downhill side will be further away that they show on the chart. How far off they are will depend on how steep the slope is.

This is another reason to do your scanning across contour lines and up/down hill rather than parallel to them or along the side of a slope.

I'll get into the actual math later. Have your calculators ready. All of you do carry a scientific calculator on the boat at all times - right?

[Larry3215]

Boy, Im on a roll tonight. Must be buzzed from the meds or something... :)

Ok, back to the right/left/center/overlap thing for a bit. Here are three more screen shots showing the exact same piece of bottom structure and how it looks on 2D, down and sideview. By the way, I might slip up sometimes and use the Humminbird abbreviations SI and DI for side and downview....

This is under the Tacoma Narrows Bridge in Wa state and is looking at some of the old collapsed bridge pieces on the bottom from back in the '40's. I just happened to pass almost directly over one large chunk. Its just slightly more to the left side.

You can see how it gets drawn on both sides of the SI image but shows up as a single piece on DI. It looks like just some blob with a beard on 2D.



03MAY16_1502_00.jpg



03MAY16_1458_02.jpg





03MAY16_1459_00.jpg


This last image is how that same piece of structure looks with Panoptix PS21. You really need all four type of sonar to truly know whats under the water.

The most interesting thing about it is that its actually about 45 ft long. It sure doesnt look that long on side or downview does it? I'll get into why in a bit....


03MAY16_1458_00.jpg

[Larry3215]

Ok, here is another look at how things get shown on the screen in downview.

First, here is another image showing how the down "cone" looks fairly wide side to side and very narrow front to back.



down1.jpg

Next, imagine you are driving over some structure on the bottom. On the left side is a view from the rear. In the center is a view from the top down. On the right side is how it looks on the downview screen.

Call the three tall arrows tree stumps that are sticking more or less straight up - one of which has branches. The green and dark red shapes are rocks on the bottom. Then there is a brown log that sits across the path and is almost as long as the beam is wide.

Notice how they are spread out front to back and side to side.

Then look at the right side of the drawing to see how those shapes show up in the downview screen.

The first thing to notice is that they all look like they are lined up perfectly under the boat in a nice straight line - but that is not the case at all. Only the green stump and the center of the log are directly under the boat. Everything else is off to one side or the other.

Notice how all the items that are off to the sides in the cone appear sunken down into the bottom return to one degree or another. How far down they are is based on how far to the side they are in the cone. Even two of the tree stumps are partly sunk into the bottom return.

Finally, see how the log got bent into an inverted V shape and is mostly down in the bottom return line.

Depending on how strong the return signal is from those items, and what the bottom is like, the parts down in the bottom return may or may not be visible. So you may or may not be able to tell if something is off to the side or not.




In the next post I'll show you some actual downview screen shots that show this happening to one degree or another.


downsqueeze1.jpg

[Larry3215]

Take a look at the 2D, down and sideview images below. You can see how things that are off to the sides of the boat can get buried in the bottom return on downview and wont show up on 2D at all. Also, things off to the sides in sideview all get squeezed together so they look like they are in the center on downview.











02MAR17_2048_00.jpg02MAR17_2046_00.jpg02MAR17_2041_02.jpg02MAR17_2041_00.jpg02MAR17_2046_01.jpg

[Larry3215]

ok, now imagine you drive over another set of objects on the bottom, but in this case, they are more or lees in a straight line from left to right across the width of the beam.

In this situation, they will all get picked up at the same time as the cone passes over them.

More importantly, they all get drawn on the screen - all at the same time - on top of each other. Also note that things that are further to the side get drawn lower on the screen because they are further away from the transducer.




That can make for some very cluttered, chaotic screen images that can be virtually impossible to understand. You can have fish on top of logs on top of rocks on top of trees on top of brush all mixed in with each other and the bottom return.

It gets even worse when there is a slope and worse again if you are not traveling in a dead straight line and dont have your heading perfectly in line with COG.

Im actually kind of amazed anyone ever gets good downimages. If the bottom is at all busy, or sloped or you are crabbing, its very hard to get good images. So dont feel too bad if your are not the greatest :)downsqueeze2.jpg

[Larry3215]

I dont think I covered this before, so I better put it in now.

I just said that you get distorted images on downview if your Heading isnt aligned perfectly with your COG. What I mean is the boat needs to be traveling - over the bottom - in the exact same direction the bow is pointed. Heading must match COG or Course Over Ground.

They will never match perfectly if there is any wind or current from the side. Any side wind or side current will cause crabbing to one degree or another.

I should point out that when I say Heading, I really mean the direction the transducer is pointed. If your transducer is on the trolling motor, it may not be pointed in the same direction the bow is pointed. Actually, most of the time it wont be.

The distortion that Im talking about is actually worse for sideview images but it effect downview as well. This next drawing shows what happens to the screen images when you are crabbing. Less crabbing equals better images. More crabbing = worse images. Going dead sideways = garbage on the screen.







skew1.jpg

skew2.jpg


Thats a sideview screen, but downview images also get offset incorrectly and stretched similarly. That just adds to the confusion when things get drawn on top of each other.

[Larry3215]

Here is an example of an actual screen shot. The downview image makes it look like those weeds where very thickly packed when they were actually spaced apart quite a ways. It was early spring and those weeds were still pretty thin - maybe 1 to 2 ft spacing between them, but the stacking makes it look much thicker.




clutter.JPG

[RangerRic]

Larry - in the pic in post #88, I assume 2d would have showed the weeds as 'thin'?
If so, then this is an example of how using split screens gives more complete/accurate info ...

Definitely appreciate you taking the time for this - great stuff.

[Larry3215]

Let me see if I grabbed a 2D shot from the same time....

[Larry3215]

I only got one good shot that shows 2D and down at the exact same time with those weeds/pads. This is the same lake in a slightly different place where the pads were starting to thin out just at the edge of deeper water.

The 2D shot matches up to the right hand side of the down shot. That white vertical line is at the same time/place in both.

2d.JPGCapture2.JPG

[Larry3215]

How different the 2D and down screen will look as far as clumping or clutter will of course depend on the relative cone widths. Most of the time, the 2D should look cleaner than the down image just because of the narrower cone angle used on 2D in most cases.

[Larry3215]

Unless someone has something to add or some questions, lets move on to sideview.

I think sideview is the most confusing sonar for most people. The screen seems to show one thing but the reality under the boat is very different.

For starters, this is a good video showing how the 3D situation under the boat gets drawn on the screen.



After watching that video, the things I went over in post 81 as far as how things get drawn on the screen, should make more sense. The key thing to start off is that the point on the bottom directly under the boat gets split between the two sides of the screen in sideview.





leftrighcenter.jpg
sidedown2.jpg

The next important detail is the distance/depth scale on the sideview screen. In that image above, it looks like the bottom, at point C, is around 150 ft to the side of the boat. In reality, that point is directly under the boat at a depth of 150 ft or so.

So obviously, that scale isnt a good way to tell how far something is to the side of the boat. Points B and D look like they are 250 ft to the side, but they are not much further than point C.

What that scale actually is, is the straight line distance from the transducer to the target or to that point on the bottom. Humminbird calls this the slant line distance.

[Larry3215]

Hold those thoughts and lets look at just the water column under and around the boat. Thats the black area in those screen shots. Somehow, the 3D volume of water under and to the sides of the boat that is scanned by sideview, gets compressed into the flat 2D section of screen on either side of the center line.

Whats going on here is much like what happens on downview with the targets that get scanned off to the sides being all crammed into the center on the screen.

For starters, lets imagine you are in very deep water - say 51 ft or more, and you set your range on sideview to less than the depth - say 50 ft. If you do that no bottom return will show up on the screen. All you will see is the "water column". This next drawing is what that "water column" actually is. Its not really a column at all.



sideview3.jpg

Lets say all the blue fish are 30 ft from the transducer and all the green fish are 45 ft from the transducer and all the red fish are over 50 ft away from the transducer. All of the fish are at different depths.

All of the blue fish will be drawn in the same place 30 ft from the center line on the screen - even though they are at different depths.

All the green fish get drawn 45 ft from the yellow line - even though they are at different depths.

None of the red fish show up at all.

Here is another drawing.



sideview2.jpg

Every fish on the arc that is the same distance from the transducer gets drawn in the same place on the sideview screen.

The important point here is that targets you see in the water column can be at any depth. The only thing you know for sure is they are on the left or right side. You have no idea of the depth or distance to the side.

There are two ways you can get a vague clue - shadows and split images - and I will go over those next.

[Larry3215]

Here is another example of how SI can fake you out when it makes you think something is way off to the side when it really isnt.

This is a screen shot of a large herring ball on sideview. Schools of fish show up pretty well on SI and DI even though individual fish not so much.

In this first shot, it looks like that herring ball starts about 100 ft to the side of the boat out to 175 ft or so and is touching the bottom.



03MAY16_1613_00.jpg


Thats not even close to being true.

That herring ball is actually almost directly under the boat but mostly on the right side. Its depth goes from about 100 ft to about 175 ft down, but it does not actually touch the bottom at all.

Here is a DI picture of that same herring ball taken at the same time. Its actually a little higher above the bottom than the downview shows. Because its off to one side a bit it shows up a little deeper than it really is.



03MAY16_1613_01.jpg

There is a clue in the first picture about where the herring ball is but its hard to see in my screen shot. If you look closely on the left side, under the depth numbers, you can see a faint mirror image of the herring ball. That tells you that part of the ball was in the left side cone, but probably still on the right side of the boat.

Thats what I mean by the split image. If something shows up on both sides of the sideview screen, then its more or less under the boat.

The reason the herring ball looks like it touches the bottom in the SI image, is because that part of the herring ball is further from the transducer than the bottom thats directly under the boat.

[MikeF]

Larry
Thanks for all your time and effort on this thread.
Well done!

If you set the down image to narrow beam, would that not produce an image similar to the 2D with less crowding of the weeds?
Thanks

[Larry3215]

If the cone angles are similar, then yes. Most of the time, down image cone angles are going to be quite a bit larger than 2D cone angles though. Its all going to depend on the transducer.

For example, the GT51 has 2D cone angles that go from CHIRP Traditional (24-13); ClearVü/SideVü (2.0x51 @ 260 kHz) & (1.4x29 @ 455 kHz)

The GT52 has similar angles and frequency choices - Traditional (24-16); ClearVü/SideVü (2.0x50 @ 455 kHz) & (1.0x30 @ 800 kHz)

In both cases, the widest 2D cone is still narrower than the narrowest down cone. They do get pretty close though.

This is one of the reasons why higher frequencies in downview make for better images. The narrower cone angles make for less clutter brought in from the sides. Thats in addition to better resolution from the higher frequency.

[mosin]

it seemd that those quadrabeam / humminbird 47 got some correct ideas back then. are there any consumer multibeam SI tranducers being produced?

[Larry3215]

Quadrabeam was a bit of an odd ball. It wasnt the same "multibeam" I have been talking about.

Lots of sonar brands have had what they called "multibeam" sonar setups. I guess you could break it down into two basic types.

One type is any sonar transducer/mfd that has more than one beam. Pretty much any transducer that does more than one frequency at a time or that has sideview is 'mulibeam' by that definition. That just means it has more than one beam. But they dont interact with each other except for maybe being shown in a combo screen or possibly even a chirp setup.

The other main type for sport fishermen, are phased array sonar units with multiple, narrow, electronically steerable beams. Panoptix and the new Furuno equivalent thats just getting released, are two examples of that type of multibeam. Simrad also has had a forward looking setup thats more for seeing bottom ahead of the boat but works basically the same way. There are also commercial variations that cost in the $100k range that are in this category.

Garmin bought a company called Interphase a while back and thats where Panoptix comes from. They patented one type of phased array multibeam sonar that Garmin then developed into Panoptix. I have no real clue how they did it, but Panoptix is light years ahead of what Interphase had.

Quadrabeam was multi-beam only in so far as it had 4 beams. It was really a split frequency 2D sonar with added side lobes that were very crude side scan beams. The two center beams were I think 83 and 200 khz. Just their normal 2D sonar but with both running at the same time. One wide cone and one narrow cone. The side lobes were 455 khz(?) but they were not narrow front to back beams. They were very wide front to back which made them more like 2D than sidescan as far as image quality and detail.

You can get a "quadrabeam" type display - with much better quality on the sidescan part - on any modern mfd, by spliting the screen with sideview on the bottom and chirp 2D on top.

Traditional sidescan and the the Panoptix type multibeam are not even close to the same thing.

As far as a mutibeam sidescan - there are at least two or three different 'multibeam' technologies being used on the hi-end oceanographic type equipment - like the stuff NOAA uses for example. I havent kept up with that stuff but IIRC one of them uses some sort of back scatter technology in addition to or along with phased arrays. Its been several years since I read up on that stuff, so Im way out of date Im sure.

[SHOTGUN RUSS]

Larry, Just to give a quick challenge to your interpretation on the SI using DI as a control. Why would you assume that the DI would give a better representation of the bait ball's distance from the bottom? In reality that ball is 3D and who knows what it actually looks like. Both the SI and DI can Lie in different ways. Yes the SI will depict a better representation of what side the bait ball is on but 2D Sonar with a narrow cone would have a better chance of showing the bottom proximity of the bait ball.