A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Geoff Hitchcox
Date: 2020 Oct 31, 21:51 -0700
I recently made a bespoke Tide Clock for my sister's birthday and to celebrate her new house renovations in Queen Charlotte Sound, New Zealand.
The clock was enjoyable to assemble, but how to SET the clock ? - that now became the more interesting part of the project.
Reading various "Owners Manuals", videos and website instructions, I realised that basically there are TWO types of setting instructions.
Type #1, Simply set the Tide Clock to HIGH at time listed in Official local Tide Table for "today".
Type #2, Set the clock to HIGH as above, but only on the day of Full or New Moon.
How accurate are these instructions? and can we do better now that many countries publish the Official Tide in a machine readable form? I started by writing a computer algorithm, that performs an optimised iterated search to find the "best-fit" to the Official Tide, for a Tide Clock - whose hour hand rotates at a FIXED rate of 12:25:14.
I also wanted to be able to compare tides from different sites, to ascertain where it was worth using a Tide Clock, and where the local tide drifted too far from a semi diurnal form.
The approach I have taken so far (I still keep tinkering) is to sort the difference between the "best-fit" and Official Tide into three bins.
Bin_1 = Difference less than 30 minutes in time (% of total).
Bin_2 = Difference between 30 and 60 minutes in time (% of total).
Bin_3 = Difference exceeding 60 minutes in time (% of total).
For example: looking at my local beach for Nov+Dec 2020, I get:
85.59%, 14.41%, 0.00% at New Brighton Beach, Christchurch, New Zealand.
This means (if I set the Tide Clock using "best-fit") the Tide Clock will be within 30 minutes of the published tide times for 86% of the High Tides, and 14% will be between 30 and 60 minutes difference, over the two month period.
Most of my experimenting has been with New Zealand Tides, however I can derive the "best-fit" tide for any place that has the Harmonic Constituents. So once I have downloaded the tide from either NOAA, NIWA, LINZ or using XTIDE, the algorithm gives the "best-fit" and daily setting times and accuracy summary - in less than a second. The computer is doing a large amount of maths - that would "drive anyone to drink" if they had to do it manually !
Testing the "conventional setting instructions" for three different sites, over the same period of time (Nov+Dec 2020).
Newport, Narragansett Bay, Rhode Island, USA.
21.19%, 22.88%, 55.93% - Setting type #1
63.56%, 36.44%, 0.00% - Setting type #2 (Full Moon)
67.80%, 30.51%, 1.69% - "best fit"
St. Helier, Jersey, Channel Islands
06.78%, 11.02%, 82.20% - Setting type #1
64.41%, 18.64%, 16.95% - Setting type #2 (Full Moon)
68.64%, 14.41%, 16.95% - "best fit"
New Brighton Beach, Christchurch, New Zealand
33.90%, 38.14%, 27.97% - Setting type #1
67.80%, 32.20%, 0.00% - Setting type #2 (Full Moon)
85.59%, 14.41%, 0.00% - "best fit"
Setting a TIDE CLOCK at "any" day of the Month (type #1) is clearly not a good idea, and is the reason I suggest many folk have a low opinion of a Tide Clock. Setting at High Tide at Full Moon (type #2), is much better - if you can wait that long ;-)
For me, it is really nice to finally see some "statistics" as to how well a Tide Clock can perform - versus the Official Harmonic Tide.
It means a Tide Clock deserves a place of prominence in my home!
I noticed that on the 16 Nov 2020, the difference between High and Low tides at St Helier is 11.05 metres (36.25 feet). A rather large difference that I'm sure my good friend Carole knew very well as a young girl, living "on the sea" in the area a number of years ago.
When I presented the Tide Clock to my sister, I said that she was not the first person in the world (by a long way) to now own a Tide Clock, but that perhaps her one is in better sync with her local tide - than any other Tide Clock in the area.
There are many beautiful Tide Clocks around, but I do have to wonder, how well are they "set" - to match their local tide ?
Oh, and just a reminder, even set with a "best-fit" solution, a Tide Clock should not be used for Navigation or berthing the Titanic. However they are very useful and enjoyable to have. I can tell at a glance, when to either go for a swim (high tide) or walk on the hard sand (low tide) at my local beach!
Whilst writing the new algorithm, I wondered whether I could change my 100 year old family heirloom Ansonia Kitchen clock, to run (at a more leisurely rate in its retirement) as a Tide Clock, by increasing the pendulum period by:
M2 constituent of the Harmonic Tide
Speed = 28.984104 (degrees per hour)
Period = 360 / Speed = 12 Hours 25 Minutes 14 Seconds
Increase Pendulum Period by (12:25:14 / 12:00:00) = 1.035046
There is now only 1mm of thread left on the pendulum shaft, but I was very pleased to be able to regulate the clock to within the thermal constraints of a steel pendulum rod !
The heirloom Ansonia Clock now has a new life telling me the local Tide (set accurately of course by the "best-fit" algorithm).
Christchurch, New Zealand.
- The "best-fit" for a TIDE CLOCK @ Newport_RI for Nov+Dec 2020
- The bespoke Tide Clock I made for my sister's birthday in 2020.
- Variation in Moonrise at Okiwa Bay over 38 years - what is the average?