Food Production and Wildlife on Farmland

What kind of agriculture most benefits biodiversity? In recent years, few questions have animated conservationists and land-use scientists more than this one. Rightly so: agricultural expansion and intensification are leading causes of wildlife declines and habitat loss,¹ and with rising demand for agricultural products, pressures are set to mount even further.²

In a 2005 paper titled “Farming and the Fate of Wild Nature,” Rhys Green and his colleagues framed the challenge in terms of two alternative strategies: land sparing and land sharing.³ In the former, agricultural intensification reduces wildlife on farmland but spares natural habitats by shrinking the overall land footprint from producing any given amount of food. In the latter, lower productivity and wildlife-friendly methods provide more suitable conditions for birds, insects, and mammals on the farmland itself, but result in more land being utilized for any given level of production, as less food is produced per hectare. Sparing and sharing occupy two ends of a spectrum of land uses, and in theory allow conservationists, farmers, and planners to find the best possible combination of food production and wildlife conservation in a given area.

Of course, in reality, things are never as black and white, and over the decade since the seminal paper by Green and others, layers of complexity have been added to the question.^4–8 But at the bottom, the choice between land sparing and sharing implies a stark trade-off: we can’t have it all.

But this notion of a strong trade-off between biodiversity and agricultural productivity has been challenged, with some arguing that it is exaggerated, or does not exist at all. “A scenario that most if not all conservationists could get behind,” says Claire Kremen at UC Berkeley, is one with “large protected areas surrounded by a relatively wildlife-friendly matrix” – matrix here referring to a farmland mosaic where species can easily move around and do their business.⁹ “A biodiversity-productivity trade-off is not a sine qua non,” argue Ivette Perfecto and John Vandermeer.¹⁰

According to this line of thinking, changing agricultural practices could circumvent the uncomfortable trade-off between food production and wildlife. “It is more likely the specific agricultural practices and suites of practices utilized, rather than the yields they produce, that determine how hospitable the shared agricultural landscape would be for elements of biodiversity,” says Kremen.⁹ In particular, agroecological or organic forms of agriculture, which, among other things, replace external chemical inputs with ecosystem services, use rotations, and eschew genetically modified seeds, are often held up as the most promising way to reach this goal.^9,11–14

This is not, strictly speaking, a question of whether there can be farmland with pretty good yields and fairly high levels of biodiversity. It is also not a question of picking either land sparing or land sharing. Rather, what is at stake is whether different forms of agricultural technology and management can provide for more wildlife at any given yield level. In other words, can the trade-off itself be mitigated – can we find solutions that are win-win?

No Food, No Space: Why High-Yield Farming Has So Little Wildlife

The answer depends on a host of factors, including which region, crop, and taxa you look at. We will start with the crops that take up the most land globally – row crops like cereals and soybean – as well as sugarcane, cotton, potatoes, and beets, all of which share relevant characteristics for this discussion.

Let us start with some first principles for how high yields are achieved in these types of crops. Several factors account for the remarkable yield improvements that have occurred in almost every world region in the last 50 years. First of all, farmers have made sure that any limiting resources like water and nutrients are in ample supply, allowing the plants to grow faster and produce bigger harvests. Second, they have tried to channel as much of these resources, as well as sunlight, into the target crop itself by planting the plants closer and closer together, making them more resilient to environmental stresses like cold, drought, and floods, and eliminating any weeds or pests that harm crop growth. Furthermore, they have maximized the amount of time during the year that plants grow, by double- and even triple-cropping where the climate so permits, and, in temperate regions, planting earlier – sometimes even during the winter.

What you end up with is a field where a single plant, the target crop itself, is extremely dominant, especially during the peak growing season. With enough water and nutrients, sunlight becomes the limiting factor, and getting the most growth out of a crop means using up every last little drop of sunlight hitting the field. This leaves little sunlight for any weeds to grow in the understory.^15,16 And if any weed were to reach up and capture some sunlight, it would be targeted with herbicides, since it would hamper crop growth. There is a catch-22 for non-crop plants: they cannot shade the crop, nor can they do well in the full shade of the crop. Given scarce weeds and an extremely simple, homogeneous crop structure, there is not a lot of food or space for organisms on the following levels in the food chain; this is further exacerbated by the required elimination of any critters that harm plants.^15,17,18 With few invertebrates to feed on, and little space for nesting, birds and mammals have a hard time making a living on the fields.¹⁷ Think of a corn field – how could much wildlife possibly fit in here?

While in reality the situation might not be as extreme as described above, there are good reasons to think that the yields themselves, and the exclusion of non-crop life they entail, are an important – if not the most important – determinant of farmland biodiversity. As John Krebs has noted, “Intensification is about making as great a proportion of primary production as possible available for human consumption. To the extent that this is achieved, the rest of nature is bound to suffer.”¹⁹ For example, the loss of European and American farmland birds in the last few decades has not been driven so much by direct mortality from pesticide applications as by the loss of feed sources and nesting sites in high-yield agricultural systems.^18,19 In fact, in many developed countries, pesticide applications have stabilized or declined, and the overall toxicity has gone down rapidly.²⁰ Similarly, it is not the fertilizers themselves that kill the critters, but – again – the loss of living room that results from the increasing biomass dominance of the crop.¹⁷

The challenges in raising yields are not that different among farming systems, be they “conventional,” organic, or something else. Fertilizers and water have to be supplied in adequate amounts. Pests and weeds have to be eliminated. The plants need to capture as much of the sunlight that falls on a field as possible during the growing season, and crops need to be growing during a larger share of the year, even virtually all the time, as in double- and triple-cropping systems. These are simple biophysical components of yield grow that there is not much of a way around.

An Unavoidable Trade-off?

In so far as yields and not individual practices determine farmland biodiversity, alternative farming systems like organic are not at an inherent advantage: raising yields in organic systems will most likely reduce biodiversity on the fields, just as lowering yields in conventional systems can increase it. European farming was about as “conventional” in the 1970s as it is today, especially in terms of external inputs like chemical fertilizers and pesticides. Yet populations of many farmland bird species, for instance, were as much as 50% higher at that time.²¹ While this has been driven in part by the loss of field margins and hedgerows, the changes in cropping itself have most likely contributed substantially.^19,22,23

Organic farming – the best-defined of alternative or agroecological farming methods – is consistently associated with more farmland wildlife. In a comprehensive review of existing evidence, Janne Bengtsson and colleagues found that organic farmland has on average 30% higher species richness and 50% higher abundance as compared with typical non-organic practices.²⁴ But organic farming also has consistently lower yields compared to more conventional systems. Several studies have found the yield gap between organic and conventional farming to be around 20 to 25%.^25–28 When looking at only the most comparable organic and conventional systems, Verena Seufert and her colleagues found the yield difference to be as high as 34%.²⁶ That implies that producing one unit of food with organic methods takes about 50% more land. The exact yield gap depends on the crop type, with cereals having a larger gap and legumes like soybean a lower one.²⁸

This implies that the difference in biodiversity between organic and conventional, for any given yield level, might not be that big. Few studies have looked simultaneously at yields and biodiversity, but one by Doreen Gabriel and colleagues, focusing on cereals in England, is an exception. They concluded that “the higher biodiversity levels in organic compared to conventional farming observed in many studies may simply reflect the lower production levels rather than the more wildlife-friendly farming methods per se.”²⁹ Another study by Paul Donald and others found that cereal yields alone explained a large share of the decline in farmland bird populations in Europe.²³

What this means is that, should the yield gap between organic and conventional farming narrow, the difference in biodiversity would probably also shrink. It is plausible, in fact, that if organic and conventional farming had the same yields, they would also have the same levels of biodiversity. It also means that efforts to increase farmland biodiversity – in any system – are very likely to require more cropland for any given level of food demand. The resulting cropland expansion usually takes place at the expense of natural habitats like forests, which are often host to more threatened species.^8,30 What wildlife is gained on farms might thus be outweighed by losses elsewhere.

Biodiversity on European farms has decreased due to the loss of field margins and hedgerows like those pictured above, along with changes in cropping to increase yields.

Wiggle Room: The Potential For Win-Win Practices

What we have seen so far points to a strong trade-off between yields and biodiversity. However, as Ben Phalan, a zoologist with the University of Cambridge, remarked, “There is probably quite a lot of wiggle room available from where we are at the moment.” This wiggle room not only points to some promising opportunities to mitigate the trade-off, but also to the limitations of viewing organic and conventional farming as two static, dichotomous farming systems.

Limiting or avoiding tillage, for example, can be a boon for biodiversity, since it reduces disturbance and leaves plant residues on the ground for invertebrates and other organisms to feed on, although this effect may be canceled out if no-till requires greater herbicide applications.^31,32 Cover crops can have a similar beneficial effect outside the growing season, but can cause a yield penalty if they deplete water stored in the soil, says Andrew Kniss, an agronomist at the University of Wyoming. Furthermore, according to Phalan, cover crops are often dominated by a single species, with limited benefits for plant biodiversity.

Kniss describes how some agronomists are experimenting with planting forage crops as understory in corn fields early in the growing season. The forage crops get a little peek at the sun before the corn canopy closes, then just barely make it through the peak growing season, but finally get some more sun once the corn plants start drying out again. Even this little bit of extra plant diversity can make a difference, as long as they don’t compete with the crop itself for sunlight.

Intercropping – where more than one crop is planted in each field – may also create a little more diversity, giving a small boost to the entire food chain.³³ Some techniques for intercropping can be compatible with large-scale mechanical harvesters, so they do not require a complete overhaul of the farming system, according to Kniss.

The most promising avenue for win-wins, however, lies in pest control. Many insects and other invertebrates are not pests, and can coexist with crops if they can find the right resources to survive in the fields. Yet many of them currently fall prey to the use of pesticides, which often kill species other than the target pests.¹⁸ Some insects even help with pest control, being natural predators to the pests.³⁴

The challenge when it comes to pest control is precision: being as effective as possible in killing the pests while harming as few other species as possible. Fortunately, there are many options. New high-tech tools allow farmers to better monitor pest outbreaks and then only spray pesticides exactly when and where they are needed.³⁵ Innovations in synthetic pesticides have allowed them to become more targeted and less toxic overall.²⁰ And new GM traits like Bt, where the plant – usually corn, soy, and cotton – produces its own insecticide, makes it unnecessary to do any spraying at all. Studies have shown non-target invertebrates to be more abundant in Bt cotton and corn fields than in fields managed with conventional insecticides.³⁶

Another set of options is typically associated with agroecological or organic farming, but could just as well be adopted in any other system. Most importantly, crop rotations – where fields alternate between two or more crops over time – make it far harder for most pest species to persist from year to year, while allowing for a broader diversity of non-pest species.^33,37 In this sense, rotations are a preemptive strategy that can reduce the need for reactive applications of pesticides once there is an outbreak. When outbreaks do occur, organic systems no longer have an advantage, since they rely on a small number of often highly toxic pesticides.³⁸ This often – but far from always – creates a situation, according to Kniss, where organic farmers as defined today use good preventative tools but bad reactive tools, and conventional farmers use good reactive tools but often fail to use preventative ones. How those two balance out is not clear, says Kniss, but the way forward is clear: combining organic’s preemptive rotations with the diversity and flexibility offered by synthetic pesticides and GM traits like Bt.

Crop rotations – where fields alternate between two or more crops over time – make it far harder for most pest species to persist from year to year, while allowing for a broader diversity of non-pest species.

What About Other Crops?

The challenges and opportunities in boosting biodiversity described above apply, as mentioned at the outset, to most row crops like corn, soybean, or cotton. The situation might, however, be different for other crops, especially perennial ones like coffee, cocoa, oil palm, or rubber trees. By virtue of being taller and sometimes allowing both some understory and overstory, at least some of these trees can foster quite a bit more diversity in flora and fauna than, say, a cornfield. Cocoa and coffee, in particular, have been touted as examples of crops where high yields and biodiversity can coexist.^9,12,39 This is true – but there is still a fairly strong trade-off. Cocoa and coffee can both be grown under a selectively thinned canopy in native forests. But the shade does come at a cost for the growth of the crops, and, in general terms, the less shade they have, the higher the yields.^40,41 The very highest yields are typically recorded where there is no shade at all, at which point these plantations start resembling the more species-poor plantations of, for example, oil palm.⁴⁰

Broadly speaking, more shade means more structural diversity, and thus more niches for other life. So when a native forest is thinned to make way for a cocoa or coffee plantation, forest-dwelling species usually suffer; one study found the number of forest species to decline by 60% upon this initial conversion to cocoa agroforestry.⁴² Going from partly shaded to full-sun systems likely involves a similar drop. But in between, along some part of the shade gradient, there might be opportunities for raising yields marginally without losing a lot of biodiversity, although studies that show this for cocoa haven’t accounted for the composition, only the number, of species.⁴³ The biodiversity friendliness might also decline with distance from intact forest fragments and over time, as native trees are replaced by planted, often non-native trees.^40,43,44

The orange-billed nightingale-thrush is an insect-eating bird that lives on Costa Rican “shade” coffee plantations.

Can’t Have the Cake and Eat It

All in all, there clearly exist some opportunities to reduce the trade-off between yields and biodiversity on farmland – although there is no guarantee that the species that will benefit are those of conservation concern, as opposed to widespread generalists. The examples mentioned above are not exhaustive; there are more out there. A lot of effort among conservationists has gone into comparing the trade-off between land sparing and sharing in different locations, but less into studying how it can be mitigated in the first place. Claire Kremen is right when she says that “focusing on how specific agricultural practices or suites of practices relate to both yields/profits and biodiversity” is what can provide scope for management interventions.⁹ This is an agenda that not only unites land sparers and land sharers, but that should also make it more obvious that conservation scientists and agronomists need to work more closely together.

Nevertheless, it is also clear that we cannot proceed on the assumption that these trade-offs will be more than marginally reduced, let alone eliminated. What sort of strategy each country or locale adopts – be it sparing or sharing or something in between – is ultimately up to the democratic will of these constituencies. Organic farming can be a perfectly legitimate choice, even though, due to its inherent limitations, it will likely remain lower yielding than non-organic farming for the foreseeable future.

But there will be consequences that extend beyond that single place. When Europeans choose to convert increasing areas to organic farming or implement certain forms of agri-environmental measures in the context of a shrinking overall area for agriculture, the lost production will be picked up somewhere else, more likely than not in a biodiverse tropical region. We can make the best of the situation by locating agriculture in the places where the biodiversity losses are smallest and the yield gains are the greatest, and steer any expansion that does happen into the least sensitive areas. But by and large, when it comes to biodiversity and farming, we cannot have the cake and eat it.