A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Ian Vaughn
Date: 2021 Feb 15, 18:34 -0800
I'd leave an authoritative answer on precession to others, but might I suggest a more modern reason for the discrepency? 2.4 degrees sounds plausible for a projection / datum shift. I calculated the difference between UTM grid north and true north as about 1.5 degrees west at Bejing, but could have made a math mistake-- and Google Earth is infamous for using goofy datums/projections. If it is a projection issue, I'd expect different errors on east/west vs. north/south lines.
Either way, I'd bet on a Chinese survey team from 600 years ago over Google Earth's datum management. This issue comes up often enough ESRI posted an article about it.
Disclaimer: I'm not really a GIS person, I just sometimes play one in real life.