What We Consume Matters. So Does How We Produce It.

I enjoyed reading Linus Blomqvist’s recent essay on wildlife and food production, and the response by Claire Kremen. Blomqvist provides a good overview of the reasons that high-yielding farming—even if it is organic or based on agroecological principles—is unlikely to provide habitat for more than a handful of the original species present in an area. When humans make maximum use of space, light, and water to divert as much as possible of the primary production in an area for human consumption, populations of many other species will suffer. Kremen acknowledges this, but argues that it would be more productive to focus on tackling demand-side issues including meat consumption, population growth, and food waste.

Both perspectives have considerable merit. I agree with Blomqvist’s conclusion that trade-offs between agricultural yields and biodiversity are the norm. While there are opportunities to improve both yields and biodiversity outcomes in many places where farmland currently has little value for either, maximizing both yields and conservation value on the same plots of land is likely to be unattainable in most (if not all) contexts.

I also agree with Kremen that many of the most important challenges revolve around consumption. Where I disagree is with the implication that reductions in demand will make choices about how to produce food along the land sharing-sparing continuum obsolete. Even if current trends are reversed, and global consumption of food declines, land sparing will be important in creating space for habitat restoration and rewilding,¹ helping to reduce extinction debt. We must explore the value of both supply-side and demand-side interventions, as recent studies have begun to do.^2,3 As part of those explorations, a continued focus on the implications of different strategies for allocating and managing land use is warranted, not least because it matters what land is spared, and where.⁴ If we believe that other species have intrinsic value⁵—a point on which I expect Kremen, Blomqvist, and myself are all in agreement—then we conservationists have an ethical obligation to understand what we can do to minimize the harm inflicted on them by how our species uses land.

What has the land sharing-sparing framework told us about this question? We have learned that the majority of species, specialists and generalists alike, are negatively affected when their habitats are converted to agricultural use, even to apparently benign uses such as diverse landscape mosaics with agroforestry plots and fallows. A few species do well, including some not found in the original native vegetation types. We have learned that species richness—the only biodiversity metric used in many studies arguing for land sharing—is a poor way to measure these changes, because it fails to detect the replacement of restricted-range species and forest or grassland specialists by widespread generalists, and because it fails to detect dramatic declines in population abundance.

We have learned that the farming systems that hold the greatest conservation value, such as the lightly grazed semi-natural grasslands of the South American pampas,⁶ are so low yielding as to make little meaningful contribution to food supply. If we wish to preserve these systems, we are better to do so primarily for their ecological and cultural values, rather than positioning them as a model of production and conservation in harmony. We have also learned that while on-farm biodiversity can make an important contribution to food production, and can in many cases support sustainable increases in yields, actions to promote such biodiversity are insufficient to conserve those species that have little direct service value.⁸ Such species—probably the majority of life on Earth—will benefit if we can produce the same amount of food on less land, while conserving and restoring native vegetation elsewhere.

The land sharing-sparing framework does not tell us that high-yield farming will result in land being spared for nature: it is an analytical framework for understanding the “biophysical option space,” to borrow a term,³ not a framework for predicting land-use change. The basic framework is a starting point, not the last word, and has already been modified to accommodate objectives beyond biodiversity and food production, and to incorporate more complex land-use scenarios. However, it can tell us how species might respond if we can find ways to make land sparing happen, and it exposes the substantial limitations of land-sharing strategies for reconciling conservation and production objectives. These various conclusions seem to be consistent, so far, across all studied taxa and different geographies (for a review, see endnote 8).

These observations require us to think very differently than before about the role of wildlife-friendly farming in conservation. Their implication is that the priority for conservationists must be to minimize the land area devoted to production, and to increase that devoted to conservation. The other millions of species we share the planet with need space, and if we are to limit the losses underway in what some have termed the “sixth extinction,” we must conserve and restore large areas of native vegetation around the world. There are undoubtedly sound reasons to promote some biodiversity on farmland, for its functional and cultural roles, but those are mostly about the needs of our own species. When we look at the needs of other species, especially those at greatest risk of extinction, what most of them need from us are bigger, higher-quality, better-connected areas of their natural habitats, and protection from additional threats such as hunting, logging, and invasive species.⁹ If proponents of alternative agriculture really want to help biodiversity, it is not enough to provide on-farm resources for a few (typically widespread and generalist) species. They too must think beyond the farm, to how their activities can support the objectives of halting and reversing habitat loss and degradation in the wider landscape.

It is also clear from studies that have applied the land sharing-sparing framework that efforts to reduce consumption and minimize the amount of land needed for food production can enlarge the “biophysical option space,” and help with this objective of making more space for wild species. Here, Kremen and I are in full agreement. For example, in our 2011 paper on land sparing and sharing in Science, one of the conclusions my co-authors and I reached was as follows:

Measures to reduce demand, including reducing meat consumption and waste, halting expansion of biofuel crops, and limiting population growth, would ameliorate the impacts of agriculture on biodiversity.¹⁰

Today, I would add reducing dairy consumption to that list.¹¹ The production of virtually all animal products is more land demanding than their plant-based counterparts, and while there is an important role for livestock in subsistence societies, the imperative for reducing consumption of livestock products in wealthy countries could not be clearer.¹² Reversing support for crop-based biofuels in Europe and North America offers another good opportunity to reduce global agricultural expansion, because it is a use of land that was largely created by government policies, with few beneficiaries and dubious environmental benefits.¹³

Because of what I have learned in the course of my work on land sparing and sharing, I have become mostly vegan, and have been involved in advocacy to halt the use of crops for biofuels. Shifting to more plant-based diets, and reducing the use of land for producing fuel, are two of the most promising ways for our species to increase food yields on a smaller land base.¹⁴ While influencing human behavior and changing policy are complex and take time, both can help to make more space available for other species.

Looking to the future, I have a vision that differs in some respects from Kremen’s. I would like to see more landscapes where agriculture is confined to a few productive zones surrounded by a matrix of natural vegetation types—a mosaic of forests, wetlands, grasslands, and shrublands. It is a vision in which humans see themselves as one species living alongside others, embedded in a larger ecosystem, rather than seeing nature as something permitted to continue only within island-like protected areas and in the interstices of human-dominated land use. For now, the closest thing to such landscapes might only be found in a few indigenous territories, with low population densities and often a culture of respect for nature. However, if we take the best of agroecological and agronomic knowledge, our growing ability to restore native vegetation in different parts of the world, and most importantly, our increasing recognition of humanity’s ethical responsibilities towards other species, I believe we could replicate it in other parts of the world with higher population densities. It is a vision that aligns with ideas expressed by E. O. Wilson in Half-Earth¹⁵ and George Monbiot in Feral,¹⁶ and is consistent with an ecocentric ethic that values all life, not just that of humans.¹⁷

What is encouraging is that much of what needs to be done to achieve any of these visions is the same: developing alternatives to the paradigm of constant economic growth; changing what (and how much) we consume; building on existing policies, regulations, and incentives (and developing new ones) to end agricultural expansion; expanding protection for native vegetation and wild species on public, private, and customary lands; improving and scaling up habitat restoration techniques; and developing methods for systems of high-yielding agriculture that respect both human communities and the ecosystems of which they are part.

Acknowledgments

I thank Andrew Balmford, David Williams, and Erasmus zu Ermgassen for comments on a draft. The opinions expressed, and any errors, are mine.