Why was it preserved?
By examining the relative thicknesses of the rock units at the mine, such as the Herrin coal and overlying Energy shale, as well as noting the nature of the contact (either gradational or abrupt) between the afore mentioned rock units and adding in other observations from the mine, ISGS geologists John Nelson and Scott Elrick and Peabody geologist Phil Ames discovered that the ancient forest had grown in an estuary that was tectonically fault controlled. In other words, the fossil forest was growing in a low lying area that was 'low lying' because a fault was allowing part of the land to slowly sink.
More importantly however, was the realization from examining deposits above the fossils and the fossils themselves that this area had not only been sinking slowly, but the fault responsible for the slow tectonic sinking had also drowned the forest by way of an abrupt earthquake... quickly dropping the land surface below sea level. The extreme southern end of a nearby known fault called the Royal Center fault in Indiana was hypothesized to be the culprit.
An intriguing modern day analogue to this kind of phenomenon of earthquake created lowlands is Reelfoot lake in Tennessee.
Like many other fortunate fossil finds, especially ones that have such great preservation, this locality had the happy coincidence of being in the 'right place' at the 'right time'. A catastrophic event (earthquake) provided the opportunity for a large segment of forest to be preserved and reveal to us ecologic subtleties of Pennsylvanian age peat mires that we had guessed and inferred before, but can now see.