I think it likely that the difference is test error or a salinity difference. Getting such a large difference between the new salt water and the tank would require a lot of alk and calcium demand for an extended period with no magnesium additions.
It is not too hard to put a conservative estimate on that effect, so lets do it, then you can assess whether your scenario might be real or not.
Suppose you are dosing a fairly high alk rate of 2 dKH per day. That means calcium is being consumed at about 14 ppm per day, and magnesium at no more than 1.4 ppm per day.
Assuming that you add no magnesium at all, this is what would happen at fixed salinity over time when doing a 30% water change each month. Thus, I do not see it getting to the value you quote.
Month Magnesium
0 1350
1 1320
2 1298
3 1283
4 1273
5 1266
6 1261
7 1257
8 1255
9 1253
10 1252
11 1251
12 1250
I'm dosing roughly 14ml of brs alk and ca a day. But I see your math and reasoning. So either user error in testing, ****** test kits are resulting in inaccurate results or bad batch of salt, unlikely.