In response to Wright and Muller-Landau’s paper on the future of tropical forests (which suggests that declining rural populations can allow forest recovery), Sloan pointed out reduced rural population often leads to increased deforestation. Really that’s not a huge surprise – peasant farmers tend to have limited labour to clean and plant land, and being capital-limited they tend not to be able to switch to mechanised agriculture. Thus, they are likely to leave a substantial portion of the landscape either forested – either ‘still’ forested (because they haven’t cleared the land yet) or reforested (secondary forest grows rapidly on abandoned agricultural land so long as fire is excluded). Larger operations are less likely to be constrained by these problems. Cattle ranching, one of the major land uses behind the frontier, uses less labour and more land. Cattle ranchers are usually better capitalised, and are often supported by state subsidies. Fire and grazing prevent forest regeneration, and does limitation in seed supply (the bigger the opening, the greater the distance the seed has to travel) and competition from grasses. The other alternative, mechanised agriculture (as practised in Mennonite colonies in Bolivia, for example) also maintains large areas free of forest, and since they repeatedly crop the same land and depend on chemical inputs, there is no opportunity for forest recovery.
That said, there are areas of the world where post-agricultural landscapes have reverted to forest. The classic examples are New England and Puerto Rico. In the case of Puerto Rico, forest cover went from around 7% to over 35%. So what was different in Puerto Rico? One aspect that has gotten a lot of attention was the role of the US. The US mainland acted as a “safety valve”, allowing migration of “surplus” population. Another factor was industrialisation – the Puerto Rican economy went from being primarily agricultural to being primarily industrial over a very short period of time. (See Rudel et al. 2000 for a more detailed discussion.)
So here’s the thing: Puerto Rico doesn’t seem to fit either model.
Of course, there are at least two patterns in Puerto Rico – one in the mountains and one in the lowlands. In the mountains, there’s little benefit to consolidation – while I have seem cattle pastured on some pretty steep slopes, it’s far from ideal. As for mechanised agriculture – not a good choice. I suspect that in mountainous areas, forest cover is likely to be inversely related to population density. The nature of the crops grown is going to make a big difference too – shade coffee is very different from row crops.
In the lowlands, on the other hand, there was the option for intensification. After the American conquest of the island large sugar cane plantations dominated the lowland. How that changed the land tenure and forest cover is something I don’t know. But as agriculture declined, there was a shift from sugar to cattle. While urbanisation and migration to the metropole are likely to have depopulated the countryside, a shift from sugar to cattle would have reduced labour demands (as would, obviously, a shift from manual cane cutters to mechanical harvesting). Cattle were a stop-gap – as one land owner in Puerto Rico told me, farming is a hobby, you can’t make a living off of it. Grass-fed cattle can’t compete with CAFO-raised cattle fed on subsidised corn. While it seems odd that it’s cheaper to import pineapples from Hawaii than it is to produce them locally, high labour costs are an important element in the decline of Puerto Rican agriculture.
So while rural populations declines may contribute to reforestation, I think it’s only true where industrial agriculture isn’t viable.
Rudel T.K., Perez-Lugo M. and Zichal H. 2000. When fields revert to forest: development and spontaneous reforestation in post-war Puerto Rico. Professional Geographer 52: 386–397.