how do I log the sea's depth every 2hrs (say) over 2 or 4 weeks?

[rgarside]

I agree that a depth-sounder is probably good enough for what the OP wants, provided the sampling interval is short enough to ensure that wave frequencies aren't aliased. But for accurate measurement for determining tidal constants and determination of sea-level for geodetic purposes, what I said is correct. You are correct that sea-water temperature can be compensated by measurment, but incorrect in assuming that the rate of change will be slow in shallow water, where both temperature and salinity can vary rapidly with fresh-water run-off. Shallow water is usually appreciably warmer at the end of a sunny day, as well.

The velocity error should be within +/- 3 percent, so over the 2m being measured there might be an error of +/- 6cm!!

 

[Shearwater]

Your pontoon is fixed/static so make up a marker covering the range and fit to pontoon. Fit a webcam to toe rail and point at marker. Webcam to laptop and software to take photo every 2 hours. Just got to figure out night-time shots (may not be so mportant)

I am in awe of some of the comments (sort of, as in..... I don't understand) but I really like the simplicity of a web cam; a simple answer that I should have been able to think up myself but didn't. I have an old Windows XP laptop I can use (and XP has always seemed so easy to get answers out of.) Perhaps an LED light to cycle coincident to the photo exposure. I now have to work out what to do with the data I gather; presumably associate it with atmospheric pressure at the same hour or perhaps rate of flow of the adjacent fresh water river. Please do not wait for a quick outcome !!!!! Many thanks.

 

[AntarcticPilot]

The velocity error should be within +/- 3 percent, so over the 2m being measured there might be an error of +/- 6cm!!

That's way too much for tidal or geodetic work. I've done analyses where that much error might have invalidated the whole thing. Admittedly I was trying to determine volcanic uplift from bathymetric data acquired to several different datums

All overkill for the OP, but it's an interesting topic.

 

[Bikerwookie]

Not an easy one to do I run a network of 16 gauges of differing types and the easy way is to throw money at it. That said

Cheapest solution is a pressure transducer with a d2a converter and run windmill logging software (I seem to remember running hyper terminal years ago and it getting flakey after a few days). Accuracy is not fantastic but good enough for your purposes I believe. The EA run bubbler gauges that go into the national sea level network that are subject to the same errors.

Sonar off the bottom from an echo sounder is a reasonable solution. 1m/s of velocity inaccuracy in 15m of water is 1cm error. You can work out your particular situations errors from that.

RTK gps once averaged produces good results but needs a lot of post processing.

Rolls royce solution is a vega radar head and ohmex or norcom logging but its getting to around £3k.

Id love to hear a cheap solution as I have been searching for one for 11 years.