The M-3 Conductivity Maps, and previous ones, were "starting points", based on very little data, especially local variations. That's why it says "estimated". Much of the more detailed data was based on Class I-A and Class I-B measured contours, like the 15 mS/m region between the WLS site and St. Louis, and early directional antenna proofs of performance in the 1940s and 1950s. The Class III and Class II DA Data rarely went out more than 20 or 30 miles. In more recent years, many conductivity studies have been done for stations to upgrade. In many parts of Michigan, which I am most familiar with, the conductivity is much much less than M-3. These are not "fake", they are empirically accurate from driving and observing signals for decades, all over the state. More recently, I have been able to also confirm this using an FIM-41. To be sure, some may be "fake" elsewhere, as the late Glen Clark told me. He only trusted certain consultants doing the measurements.
In Colorado, a 250 watt Class D station has more recently been able to upgrade directionally, to 50000 watts Class D.
David is also very familiar with the areas of the NW Lower Peninsula of Michigan, and of Lake Michigan, where stations across the Lake have signals stronger than the ones just 20 or so miles inland. WTCM 1400, a Class IV when David spent a lot of time there, was able to move to 580 with 2500 watts, and with measured studies, was able to increase to 50000 watts Daytime. One of David's early jobs was at WCCW 1310, when it was just 1000 watts, then soon 5000 watts. They upgraded to 15000 watts Daytime from another site to go full-time, and had a CP for 50000 watts Daytime. They recently moved back to their original site with 15000 watts nondirectional, Class D, and looking through all the applications, they probably could be 50000 watts nondirectional from that original site. Much of the Lower Peninsula is shown as 8 mS/m, whereas the measured values are 0.1 mS/m in many areas, and in some rare small areas, up to 3 mS/m.
Recently, on Long Island, all 0.5 mS/m on M-3, a study near Patchogue showed radials where all the points were well below the 0.1 mS/m curve on the ground wave graph, the lowest on the graphs.
The Alaska and Hawaii conductivity maps show 4 mS/m near the coasts, and 2 mS/m inland, as I recall. There is very little detail.
There is a previous Conductivity Map in the Fourth Edition of the NAB Engineering Handbook, from 1949.