some latitude/longitude hardcoded in sites.json not accurate

[taichiuyama]

Description

For the coordinates of the Mauna Kea observatories, I found inconsistencies between sites.json and IRTF reports.
https://irtfweb.ifa.hawaii.edu/observing/telescopeCoordinates.php
At least the IRTF one is consistent with Subaru; https://subarutelescope.org/Observing/Telescope/Parameters/

Additional context

I need accurate latitude and longitude to calculate precession-corrected parameters.

[hamogu]

Hi @taichiuyama thanks for checking. The data for the sites.json is contributed by users like you. It would be great if you could open a pull request to improve whatever you found.
Originally, this database was meant for observation planning to get the time of sunset/sunrise etc and no great precision was needed. It's only later that people started using it outside of the original use case. So, we would very much welcome a pull request to update information as needed.

[taichiuyama]

thank you @hamogu for clarifying the data format, I thought the astropy databases were accurate.
will check other coordinates and open a pull request later.

[kgolap]

Alma coordinates are inaccurate especially in height or elevation.
The better values are available on wikipedia
https://en.wikipedia.org/wiki/Atacama_Large_Millimeter_Array

[hamogu]

I would be great to have a pull request for someone who has detailed knowledge of ALMA. I'm more of an X-ray observer and don't really know what even the definition of the "location" for an array of antennas is that are so far apart in some configurations and that are moved to different locations every few weeks. Is there a defined "center" or "zero point" for ALMA?