Worldwide Eratosthenes stick experiment, May 14, 15, 16 2021

[reficius]

Hi all,.
We're running a worldwide Eratosthenes stick experiment on the Reddit sub r/flatearth. This is the experiment where you measure the shadow of a stick at local noon, and compare with other places in the world, and derive the size & shape of the earth.

Details here:


Source: https://www.reddit.com/r/flatearth/comments/n8lm9l/worldwide_eratosthenes_stick_experiment_may_14_15/?utm_source=share&utm_medium=web2x&context=3

 

[JMartJr]

Neat. At some later point you might do one that takes the measurement at the same time (obviously that would leave out that half of the globe that is experiencing night!) and plot the angles and bearings. Plotted onto a globe, if the Earth is a globe with a distant Sun you would expect lines that are essentially parallel. Plotted onto a plane, if the Earth were flat with a "local Sun," they would be expected to converge on a single nearby point. Plotting the data onto the incorrect model should not provide the predicted results.

Thanks for the option of leaving data here, since I don't Reddit!

Now to pick out the perfect stick...

PS: Would it be helpful to you if I were to share this more widely, or do you anticipate having all the responses you want to spend time entering?

Edited for typos.

 

[reficius]

Neat. At some later point you might do one that takes the measurement at the same time (obviously that would leave out that half of the globe that is experiencing night!) and plot the angles and bearings. Plotted onto a globe, if the Earth is a globe with a distant Sun you would expect lines that are essentially parallel. Plotted onto a plane, if the Earth were flat with a "local Sun," they would be expected to converge on a single nearby point. Plotting the data onto the incorrect model should not provide the predicted results.

Thanks for the option of leaving data here, since I don't Reddit!

Now to pick out the perfect stick...

PS: Would it be helpful to you if I were to share this more widely, or do you anticipate having all the responses you want to spend time entering?

Edited for typos.

Please do share this. More is better!

 

[JMartJr]

Will do. One advantage of my weird kite festival lifestyle is I have a lot of friends in furren lands, and they all have some sticks. Can't promise they'll come through, but I'll put it in front of them.

 

[JMartJr]

Need to clarify, I think I misread -- do we need to enter the data through Reddit? I had firt read it as being able to leave data in THIS thread on MB, but re-reading I think that was in error.

 

[Mick West]

Need to clarify, I think I misread -- do we need to enter the data through Reddit? I had firt read it as being able to leave data in THIS thread on MB, but re-reading I think that was in error.

On Reddit. That's an embedded Reddit post, not a post here.

 

[reficius]

It's ok if you post your data here, if the mods don't mind. I'm not really expecting a flood of responses.

 

[Mick West]

It's ok if you post your data here, if the mods don't mind. I'm not really expecting a flood of responses.

No Problem. I'll try to remember to give it a go. Weather looks good.

 

[reficius]

Here's my data, for an example...

Stick height: 118 inches; Shadow length: 54 1/2 inches; Latitude: 43.0°

Taken today about 5 minutes before my solar noon.

 

[M Bornong]

May 14
35.35 N
12:52 PDT (UTC-7) solar noon and within 1 minute of measurement
96 inch stick
28 11/16ths inch length of shadow

Hope that helps.

 

[Easy Muffin]

Not the clearest of shadows thanks to haze and clouds but we wound up with

Stick: 44 cm
Shadow: 27 cm
Lat: 50.6 N

 

[reficius]

Here's some preliminary data. We only had 4 participants, which is enough, but more would always be better.

First, a graph of the Sun's angular altitude vs latitude of the observer:

This should show a straight line, and the Y intercept should be 90° + the Sun's declination, or about 109°. The linear fit's intercept is about 107.7°, so not too bad. The Sun's declination, for the uninitiated, is equivalent to the latitude at which the Sun is directly overhead at solar noon. It amounts to about +19° during the time of the experiment, since we're well past the March equinox.

Next, a graph of the calculated height of the Sun vs the latitude of the observer. Since the Sun was due south for all measurements, and since flat earthers use a linear interpretation of latitude, we can just assume a linear number of miles per flat earth ° of latitude, which amounts to about 110 km, and just apply a little trig to get the Sun's height:

Here, we should see a straight horizontal line if the flat Earth/nearby Sun hypothesis is true... the Sun should be at the same height for all measurements. The mean value approximates that, but the best linear fit is rather strongly biased toward some other kind of model.

Any thoughts on where to go with the analysis from here?

 

[Mick West]

Sorry I forgot!

 

[JMartJr]

Was clouded out. Presumably my data would have fallen on the lines!

 

[AmberRobot]

Here's some preliminary data. We only had 4 participants, which is enough, but more would always be better.

First, a graph of the Sun's angular altitude vs latitude of the observer:

View attachment 44397

If the sun were close and above a flat earth this line wouldn't be straight. It's easy to derive that formula and it fails to fit any data.

the flat earth model also spectacularly fails to get the Sun's azimuth through the day correct. And also never gets the sun close to the horizon for any time of day, let alone predict an actual sunset.