Friday, January 26, 2007

‘Wildlife-friendly farming’ versus ‘Land sparing’

I recently came across a two year old paper by four biologists published in the journal Science. Brazenly titled “Farming and the Fate of Wild Nature,” the authors less than subtlety suggest that there is a lot at stake. There is! They report that 50% of natural habitat on arable land has been cleared for agriculture, and while forests are regrowing in the developed world, deforestation continues apace in the developing world, which houses most of the world’s biodiversity. Using data from an international database of threatened birds, they also demonstrate that agriculture adversely affects biodiversity. As both population and affluence continue to rise for the foreseeable future, agriculture will continue to pose a serious threat to wilderness.

Two general strategies have been proposed to mitigate the adverse affects of agriculture:
  1. Wildlife-friendly farming – Reduced use of chemicals and planting of buffer zones may make farmland more wildlife-friendly. Supporters of organic agriculture, for example, often promote its biodiversity enhancing properties. The trade-off is lower yield, which the authors argue is a real phenomenon, citing that farmers in Europe generally do not switch to less intensive agriculture voluntarily, but are persuaded to by conversion subsidies.
  2. Land sparing – Intensive agriculture (high chemical inputs, irrigation, machinery) creates an environment harsh to most forms of life, but the high yields it generates potentially preserve wilderness that would otherwise be converted to agriculture.
The optimal strategy is not intuitively obvious. The authors put it bluntly: “Identifying the key parameters that can resolve this trade-off requires a model.” The key parameter in their model turns out to be the shape (concave or convex) of the density-yield function. A concave density-yield function means that as yield rises, the population density of a given organism initially declines slowly. Imagine a hypothetical butterfly species that persists at high population levels in undisturbed habitat, but persists at moderate density in low yield farms with buffer zones, few chemicals, etc. Conversely, a convex density-yield function says that population density initially declines precipitously with increasing yield. For the aforementioned butterfly, substitute a large tree that cannot survive at appreciable densities outside wild nature.

The authors’ primary purpose with this article is to provoke an empirical research program, not generate conclusions. Nonetheless, they tepidly support the land sparing strategy on the premise that many species, especially those of conservation concern, seem to exhibit a convex density-yield function. If correct, the results would advocate the position of the green revolution: high yield, high input agriculture is optimal both for people and the environment.

The authors anticipate three potential limitations of their approach, one of which I address here. Their model does not incorporate effects on wildlife populations on nonfarm land, a widely cited example being the dead zone in the Gulf of Mexico caused by chemical runoff from farms. Incorporating such externalities might shift the balance in favor wildlife-friendly farming. This is an important concern, but the street goes both ways. External effects of agriculture are a product of both intensity and area. Intensive farming has high external environmental costs per unit area, but they must be weighed against reduced external costs per area spread over a larger land area. For example, an abundance of low yield farms could increase habitat fragmentation.

I suspect there may be two more general criticisms that arise, not necessarily to this paper in particular, but to this approach to conservation, especially if it turns out that the convex function is more common, favoring a land sparing approach. First, a critic might claim that this is “just a model,” too abstract for the complexity of real ecosystems and overly reductionist, breaking ecosystems down into constituent species and functions. Such criticisms are wrongheaded. The real dichotomy is not abstraction/realism, reductionism/holism, but rather those who acknowledge their working models and those who do not. It is trite, but worth repeating, that our perception of the world is not an unadulterated version of sensory input. Rather, we all use models, conscious or otherwise, to make sense of reality. Mathematical models formalize this innate process and bring it the forefront of our consciousness, permitting critical evaluation. The alternative is blind surrender to our preconceptions.

Another complaint might be that the paper establishes false premises and ignores a possible third way. In particular, the small-scale, labor intensive, beyond organic, local farming movement. Relying on greater local knowledge and increased inputs of manual labor, such farms have high, sustainable yields without large inputs of chemicals or machinery. At least, that is the anecdote, as I have not seen a systematic study that bears this out. Despite the growth in local farming, it represents a small portion of total agricultural output for obvious reasons: labor is expensive compared to capital, and knowledge workers are not drawn to farm work in large numbers. Taken together, this suggests that for local farming to ever reach a broad consumer base, people must be willing to spend a lot more on food, which I doubt will occur. Improbability aside, I am skeptical, as I have expressed in earlier posts, that local farming actually has any environmental benefits over industrial agriculture.

Beyond utilitarian considerations, the wildlife-friendly farming/land sparing debate raises philosophical questions about protecting biodiversity. Though I am not that familiar with the debate, it seems reminiscent of the 19th century schism between preservationists and conservationists. The former advocated unadulterated nature for its own sake, while the latter argued that nature should be managed for human benefit. In the present context, land sparing might permit larger areas of unadulterated nature in the form of natural parks and wildlife preserves, while wildlife-friendly farming would permit land use that is beneficial to humans at minimum cost to biodiversity. I see the validity of both perspectives and do not offer a solution. Nevertheless, it is interesting how contemporary environmentalists sort out along philosophical lines. In general, those with a biological inclination lean toward wildlife-friendly farming, while developmental economists and agriculturalists advocate land sparing. I suspect divergence between the two groups arises from the duality of humanity, existing simultaneously in and above nature. Evolutionary theory shows us our place in the tree of life, while our advanced civilization has brought us out of nature. Natural scientists, steeped in evolution and ecology, may tend to perceive us as part of nature, while the social scientists, the experts on civilization, may tend to view us as outside of nature. Thus, while the former sees man as integral with and therefore capable of tinkering with his environment, the latter does not mind a clear demarcation. Bringing such cultural and philosophical differences to the forefront can promote constructive dialogue that evaluates agricultural policy not solely upon preserving biodiversity, but also on the type of nature we ultimately desire.

1 comment:

tibor hartel said...

Interesting paper - agree with your points. I have some further concerns regarding the so called more drastic loss of biodiv in developing countries, than in developed countries stated by authors:

"Several kinds of data suggest that, although it is an important driver almost everywhere,
the effect of agricultural change on wild nature is now greatest in developing countries."

It is, of course true. Because...there is a very rich biodiversity here, and from this level, only decline is possible....

Tibor Hartel