The National Institute for Standards and Technology (NIST) broadcasts a standard time signal from which any surveyor with a shortwave radio that receives their signal and a digital stopwatch can get UT1, a time scale based on the rate of the Earth's rotation, to use in timing astronomical observations, including solar observations for azimuth. The broadcast time signal is on a time scale known as UTC kept by atomic clocks so accurate that they measure the small changes in UT1 that occur occasionally.
NIST Radio Time Signal Frequencies
Per NIST:
"If you need UT1 with an uncertainty of 0.1 s, you can apply a correction to UTC. UT1 corrections are encoded into the broadcasts by using doubled ticks during the first 16 s of each minute. You can determine the amount of the correction (in units of 0.1 s) by counting the number of doubled ticks. The sign of the correction depends on whether the doubled ticks occur in the first 8 s of the minute or in the second 8 s. If the doubled ticks are in the first 8 s (1-8) the sign is positive. If the doubled ticks are in the second 8 s (9-16) the sign is negative. For example, if ticks 1, 2, and 3 are doubled, the correction is +0.3 s. This means that UT1 equals UTC plus 0.3 s. If UTC is 8:45:17, then UT1 is 8:45:17.3. If ticks 9, 10, 11, and 12 are doubled, the correction is -0.4 s. If UTC is 8:45:17, then UT1 is 8:45:16.6. If none of the ticks are doubled, then the current correction is 0."
NIST Discussion of Time Signal Broadcast Format
The field method is straight forward. You tune the radio to the NIST broadcast time signal and on the minute mark start the stopwatch running. Note the broadcast time (which will be UTC) and note the stopwatch time of the same instant. If you started the stopwatch at 23:30:00 UTC, then:
23:30:00 UTC = 0:00:00.00 Stopwatch
Wait for the next minute and, using the lap time function, check the Stopwatch time when the minute hack is broadcast.
On February 18, 1994 as I stood beside a Zeiss one-second theodolite set up near a windmill on a 12,000-acre ranch in the middle of a large, empty West Texas county, I got this comparision:
23:31:00 UTC = 0:01:00.02 Stopwatch
I used the same lap time function on the stopwatch to get the times of each of six pointings I made on the trailing edge of the Sun, alternating Face Left and Face Right to make a set and checking directions to the backsight target between the three sets. Then, after taking the directions to the backsight after the last set on the Sun, I compared the stopwatch to the time signal again and got:
23:48:00 UTC = 0:17:59.98 Stopwatch.
The equation between UTC and UT1 coded into the time signal that day was:
UT1 = UTC + 0.1"
So, I had all the information necessary to calculate the UT1 time of each of the six pointings on the Sun and, knowing the latitude and longitude of the station where the theodolite was set up sufficiently accurately, calculate the azimuth of that trailing edge of the Sun at each instant.
Fine work. And now we could have discussions about the quality of the stopwatch, reaction times of the instrument operator and the timekeeper, relationship of the gun quality to the time quality to the quality of the field crew.
Isn't GPS a great tool! 😉
> And now we could have discussions about the quality of the stopwatch, reaction times of the instrument operator and the timekeeper, relationship of the gun quality to the time quality to the quality of the field crew.
As to the reaction time or "personal equation", as it is known in the surveying literature, what my experience has been is that as long as the observer is handling the stopwatch, the reaction times largely cancel out since nominally the same time lag between hearing/seeing event and pressing stopwatch lap time button applies to timings.
For some observations, timings aren't particularly critical, but for observations for azimuth on the Sun by the hour angle method they are and the observer needs to also be timekeeper.
As for the quality of instruments, my own experience is that when the Sun is casting sharp, well-defined shadows and is up at least 20 degrees from the horizon, one should expect to be able to point at the trailing edge with an accuracy comparable to pointing at any good surveying target. The key elements to realizing the full accuracy of the measurement are:
(a) stability of the tripod (joints adjusted and set up on a firm base),
(b) exceptional care in leveling instrument.
With an optical theodolite, that careful leveling is done using the zenith angle compensator via zenith angle readings as the instrument is oriented as it would be to be leveled using the plate vial. That procedure would be much simpler with an electronic instrument that displays its leveling. I prefered to use a one-second optical theodolite because
(a) it gave excellent results,
(b) the solar filter was inexpensive and
(c) I didn't have to worry about burning out any elements of the EDM when pointing at the Sun.
:good:
