Agricultural productivity is the key to global food security. Modern conventional farming has substantially increased food production, but at the expense of serious environmental harm . By contrast, organic production is regarded as a suitable and more sustainable alternative owing to its seemingly reduced dependence on external inputs and its overall smaller environmental footprint. The discussion on the yield gap between conventional and organic agriculture [2,3], which is a major obstacle in implementing organic production to a greater share of land, was recently boosted by new findings claiming that organic production can narrow that gap via its inherently biologically more diverse production systems . Based on their comprehensive data analysis Ponisio et al.  state that (i) the observed yield gap of 19 (16–23)% may still be an overestimate, (ii) a biologically diversified versus chemically intensive production has the potential to promote high-yielding agriculture which is relatively environmentally beneficially and wildlife friendly compared with conventional systems, resulting in (iii) organic farming being the more sustainable way of producing food.
With the yield gap apparently approaching insignificance, this most up-to-date study warrants a closer look at system boundaries. Most of the underlying primary yield comparisons in  are derived from controlled experiments, experimental sites or paired farms, i.e. from studies covering mostly distances of only a few meters to hundreds of meters, with organic systems often being interspersed in a conventional framework. Are plot-scale organic yields and environmental benefits transferrable to a hypothetical large-scale organic agriculture without overlooking the relevant information? Only under these conditions, the above claims hold true. I argue that, for four reasons, such far-reaching conclusions must be drawn with caution and that an unbiased system comparison with comparable boundary conditions is needed.
(1) Nutrient availability is a major factor for high and stable agricultural yields. In certified organic farming, guidelines restrict the use of external nutrients, particularly of mineral fertilizers, and this downregulates productivity. Nevertheless, the published yield gap may not represent the expected yield levels under large-scale implementation of organic agriculture owing to nutrient spillover effects. Firstly, organic farmers benefit from substantial amounts of atmospheric nitrogen deposition, much of which comes from conventional agriculture. In intensively managed landscapes, N deposition often reaches tens of kilograms per hectare , most of it originating from conventional farming. Whereas this deposition is unwanted in semi-natural ecosystems, production systems with small fertilizer N inputs may benefit from it. Organic agriculture makes use of biological N fixation via legumes as a primary N source; however, owing to the many non-fixing crops in rotation, N often limits productivity. Current organic yields on plots under high N deposition may no longer represent yields should organic practices become more widespread. Secondly, nutrient import from a conventional system into an organic system is commonplace. In a survey of 63 certified organic farms in France, nutrient inflows from conventional agriculture were 23%, 53% and 73% of the total input (including biological fixation and deposition) for N, K and P, respectively . Manure, fertilizers, feedstuff and straw imports delivered most of these allowed conventional inputs. Without conventional farms nearby, the beneficial effect of nutrient imports may largely diminish (figure 1).
(2) Diversified farming systems including both conventional and organic systems may be effective in disease control, particularly through mixed cropping and crop rotations . By contrast, organic farming directives prohibit the use of most synthetic pesticides. Instead, organic farmers implement biological pest control, e.g. via introduction of predators or use of organic pesticides, in conjunction with agricultural measures such as diverse crop rotations, mechanical weeding and provision or maintenance of habitats for beneficial organisms. This may enhance landscape richness and promote biodiversity at the plot to the farm scales , although beneficial effects on species richness from organic farming weaken or diminish at higher levels of spatial aggregation . Ponisio et al.  argue that enhanced agricultural diversification is an important factor for reducing the organic-conventional yield gap. In addition, organic farmers may use a limited number of chemical agents from natural sources. Together, these measures overlap with much of the FAO-recommended integrated pest management (IPM) , but they exclude the targeted use of synthetic pesticides. It is unknown how effectively IPM without chemical agents is able to control quickly dispersing, wide-ranging pests that are abundant even in mosaic, diversity-rich landscapes. The abundant pollen beetle Meligethes aeneus, for example, can easily travel distances of a few kilometres and infest neighbouring rapeseed fields, even when they are separated by fields stocked with different crops . Owing to the lack of controlled large-scale organic implementation, it still remains unknown whether biological pest control alone is able to keep the infestation below the economic injury level in case organic fields are more abundant with a higher share of land.
(3) Diverse, lower-yielding agriculture does not always foster biodiversity at larger (square kilometre) scales . A higher per area yield (typical of land sparing) may even provide greater mutual environmental benefits compared to an area extension (typical of land sharing) . Land is one of the most limited and non-renewable resources for agricultural production. Without substantial changes in human diet or a reduction in food waste, a 19% yield reduction  may translate to a 24% larger production area that needs to be converted from current non-agricultural use, with potentially serious consequences for biodiversity. Land sparing may not necessarily follow agricultural intensification ; population growth, increases in per capita food consumption and diet changes tend to occur alongside agricultural intensification, and these societal factors lead to a higher land demand in spite of agricultural intensification. With a wider implementation of lower-yielding agriculture, appropriation of natural areas may even become more prevalent.
(4) Direct land-use change emissions of 4.3 Pg CO2-eq. per year are among the highest in the land-use sector , making any agricultural scenario with a higher land demand—from this perspective—appear inadequate. This is particularly relevant given that organic farming has, per unit output, similar greenhouse gas (GHG) emissions to conventional farming . The potential loss of biodiversity from converted natural ecosystems, and the associated substantial GHG emissions from land-use change are seldom included in a comprehensive organic farming assessment.
Two major societal returns of modern agriculture, namely food security and affordable products, seem insufficiently appreciated in the ongoing discussion. These benefits are major building blocks for a peaceful society, but they come at a price. It is unlikely that we will ever reconcile these benefits with those of a biodiverse and pollutant-free, semi-natural environment. The goal must be to keep the price low. Whereas the currently applied conventional farming has irrefutable disadvantages, we are dealing mostly with known consequences, allowing identification of viable solutions and reduction of negative externalities. Many aspects employed in organic agriculture, particularly diverse crop rotations, mixed cropping and biological pest control, have the power to reduce the environmental footprint of conventional farming without compromising its advantages. By contrast, it is not known what consequences would arise from a broad organic implementation, far beyond the currently employed less than 1% of the agricultural area. It is time to commence large-scale, self-reliant experiments to explore the true potential of organic agriculture.
The accompanying reply can be viewed at http://dx.doi.org/10.1098/rspb.2015.2913.
- Received July 6, 2015.
- Accepted November 10, 2015.
- © 2016 The Author(s)
Published by the Royal Society. All rights reserved.