It would be interesting to know what type of tree that was.
In southern Canada we have a lot of different soil conditions, it's an area where the trees go from deciduous in the south to mostly coniferous. I have sandy moraines, or even solid granite with a few minutes drive.
Because of the lakes, historically, there can be clay in weird places. I have one site that has modern clay near the water 6" down, or you can go to other areas 20' away and encounter soft wetlands that would consume almost anything. At the highest point in town I can still find decomposing clay 8" down from where the water was 10,000 years ago.
In addition to the animals already mentioned, the yearly freezing and thawing cycle does cause the coins to sink in soil (and up on sandy moraines ), the ground breaths, it expands with the freeze, shrinks in the summer. especially saturated soils.
This action could additionally flip a lot of coins on edge. A coin on edge has a higher PSI and would sink faster then a flat coin. I wouldn't be surprised if a coin tumbles as it sinks through the ground over time.
Some forested areas have soft, loamy fluffy soil, perhaps with sand mixed in. I would speculate things here would disappear quicker, even if it's just because of the addition of rotting sticks and leaves over the years.
So if you can find areas that have a clay base, you will probably find that the upper soil is very clay-like compared to soils that are composed mostly of organic matter. These clay rich soils will hold the coins shallower, longer IMO, but that doesn't mean they are all shallow targets. They will still drop through the first 6" comparable to other sites.