The linked article shows evidence that the Maya had specialization, as in flint workers and cotton spinners.
Then, after suggesting that settlement was disperse, and tended towards ridges where soil and drainage are good, they write,
Our results at once provide a basis for understanding the ancient Maya land use strategies and a criterion for protecting the most sensitive areas from development.
slipping the modern bias for central planning and 'protecting sensitive areas into an otherwise excellent report.
The researchers developed a predictive model for finding ancient settlements based on soil quality, drainage, streams, and slope. It sounds pretty good.
Our model has another important potential. It could prove a major asset in the management of cultural resources of the Maya forest and give probability strategies for locating ancient settlements.
So the best land may become off limits. Maybe silverseed is onto something: go to where there hasn't ever been agricultural development.
The prevailing view is that the present forests of the Maya area are essentially anthropogenic, that they are a result of human interaction with the forest, manipulating the species composition to increase the value of the forest for the economy of the culture. By actively nurturing and providing preferential attention of some species over others in the forest and planting and caring for economic plants in the garden, there is a blur in the distinction between the wilderness and the garden.