Why Ecological Biodiversity Research Results Seldom Apply to Agriculture
Posted by Andrew McGuire | February 4, 2021
Ecologists, agroecologists, and regenerative agriculture advocates are promoting biodiversity to solve many of agriculture’s problems. They often base this on evidence of the benefits of plant biodiversity from ecological research, for example the Jena study in Europe (Buchmann et al., 2018) or the Cedar Creek Ecosystem project in Minnesota (Tilman et al., 2014). These results are used to support the use of intercropping and cover crop mixtures by farmers. However, there are reasons we should not apply this research to agriculture.
“On Average” vs. “Best vs Best”
Ecological research is done to mimic nature. In doing so, it differs from how agricultural research is conducted and how we interpret the results (see Schulze et al., 2018). Ecologists conducting biodiversity research are looking for a general pattern of higher diversity plant mixtures performing better in general than lower diversity mixtures and monocultures. Performance is most often measured by biomass production. Here is how the results often look:
The biomass production ranges widely at the monoculture level and less so as diversity increases. Ecology studies take the average of this range of production levels at each level of plant species diversity; for 1-species monocultures, 2-species mixtures, 3-species mixtures and so on, giving the orange line. As you can see, this average biomass production increases with increasing plant species diversity.
If you are looking at random mixtures for an unmanaged nature, this is fine. However, in agriculture we use management, intelligent choices based on experience and research. We know some species are top producers and so must be in a mixture. Others are not, and so we exclude them unless they bring some other benefit to the mix. Because of this management, the mixtures we assemble must compete with the best monocultures, not with the average of all the monocultures. The biodiversity standard relevant for agriculture is not the average increase with increasing plant species diversity, but comparison of the best monocultures and the best mixtures, as shown here:
In robust research, each small dot is the average of multiple plots. This allows statistical comparison of each species monoculture with each mixture at each level of diversity. Ecological research rarely makes this comparison, and so cannot be used to justify the increased use of plant mixtures in agriculture. When ecological research does use this standard, the best mixtures do not often perform better than the best monocultures (Cardinale et al. 2011). The same goes for agriculture: as Trenbath found in a 1974 review (Trenbath, 1974), and as we found in a 2020 review (Florence and McGuire, 2020), higher plant diversity is not always better in agriculture.
“Averages mislead by hiding the spread in a single number” – Hans Rosling in Factfulness.
Despite all this, agricultural researchers still often use this ecological standard of comparing averages. Land Equivalent Ratios, LERs, or Relative Yield Total, RYT, are examples of this. However, these are not the most appropriate standards to test crop mixtures in agriculture (Garnier et al., 1997). Rather, agriculture, because it includes intelligent management, should compare the best monocultures to the best mixtures, a standard called transgressive overyielding.
Agriculture is not Nature
Besides the management factor which demands a performance standard relevant for agriculture, there are other differences between agricultural and ecological research. Ecology studies naturally use natives to imitate nature, but agricultural studies use crops bred to produce food. Breeding has removed many of the characteristics of native plants from our crops, such as anti-herbivore defenses (thorns, poisonous compounds, etc.), long ripening periods, and responses to crowding. The plants used in each type of research differ, so we cannot assume they will produce the same results (Chacón-Labella et al., 2019).
Furthermore, the native plants that ecologists use are perennials, while agronomists often use annual crops. Using perennials allows ecologists to follow the performance of the mixtures for years, sometimes decades. Doing this, they have found that biodiversity effects take time to kick in, 3-5 years in many studies (Cardinale et al. 2007; Reich et al., 2012). And higher yields in more diverse mixtures are not due to any synergy in the mixtures but because the monoculture’s yields decline over time, probably due to buildup of soilborne disease (Van Ruijven et al. 2020). So, these results may apply to you if you are growing mixtures of perennial crops for 3+ years, but they do not apply to annual cropping agriculture.
Differences Between Ecological and Agricultural Research
Managed to produce food
Average biomass yields at each diversity level
Best mixtures vs. Best monocultures
Plants used in research
High anti-herbivore defenses
High % of inedible biomass
Lower anti-herbivore defenses
High % of edible food
High biodiversity in space
High biodiversity in time (crop rotations of annual crops)
Soil nutrient status
Low to moderate
High to moderate
Export of food/biomass
Biodiversity in Agriculture
Biodiversity has a place in agriculture. We have long-known that simple cereal-legume mixtures enhance biomass production on low nitrogen soils (Ofori and Stern, 1987). Some intercrops make sense because of their pest control benefits, even with lower yields. And diversity in time through crop rotation is a time-honored practice. Beyond these examples, we must test biodiversity in agriculture by the standards that make sense for agriculture, not those of ecology. When presented with evidence supporting the use of more crop mixtures in agriculture, ask whether it is ecological research focused on natural, unmanaged ecosystems, or if it is focused on agriculture, using annual crops and comparing the best monocultures to the best mixtures.
Buchmann, T., J. Schumacher, A. Ebeling, N. Eisenhauer, M. Fischer, et al. 2018. Connecting experimental biodiversity research to real-world grasslands. Perspectives in Plant Ecology, Evolution and Systematics 33: 78–88. doi: 10.1016/j.ppees.2018.06.002.
Cardinale, B.J., K.L. Matulich, D.U. Hooper, J.E. Byrnes, E. Duffy, et al. 2011. The functional role of producer diversity in ecosystems. Am. J. Bot. 98(3): 572–592. doi: 10.3732/ajb.1000364.
Cardinale, B.J., J.P. Wright, M.W. Cadotte, I.T. Carroll, A. Hector, et al. 2007. Impacts of plant diversity on biomass production increase through time because of species complementarity. PNAS 104(46): 18123–18128. doi: 10.1073/pnas.0709069104.
Chacón-Labella, J., P. García Palacios, S. Matesanz, C. Schöb, and R. Milla. 2019. Plant domestication disrupts biodiversity effects across major crop types. Ecology Letters 0(0). doi: 10.1111/ele.13336.
Florence, A.M., and A.M. McGuire. 2020. Do diverse cover crop mixtures perform better than monocultures? A systematic review. Agronomy Journal 112(5). doi: 10.1002/agj2.20340.
Garnier, E., M.-L. Navas, M.P. Austin, J.M. Lilley, and R.M. Gifford. 1997. A problem for biodiversity-productivity studies: how to compare the productivity of multispecific plant mixtures to that of monocultures? Acta Oecologica 18(6): 657–670. doi: 10.1016/S1146-609X(97)80049-5.
Ofori, F., and W.R. Stern. 1987. Cereal–Legume Intercropping Systems. In: Brady, N.C., editor, Advances in Agronomy. Academic Press. p. 41–90
Reich, P.B., D. Tilman, F. Isbell, K. Mueller, S.E. Hobbie, et al. 2012. Impacts of Biodiversity Loss Escalate Through Time as Redundancy Fades. Science 336(6081): 589–592. doi: 10.1126/science.1217909.
Schulze, E.D., O. Bouriaud, U. Weber, C. Roscher, D. Hessenmoeller, et al. 2018. Management breaks the natural productivity-biodiversity relationship in forests and grassland: an opinion. Forest Ecosystems 5(1): 3. doi: 10.1186/s40663-017-0122-y.
Tilman, D., F. Isbell, and J.M. Cowles. 2014. Biodiversity and Ecosystem Functioning. Annual Review of Ecology, Evolution, and Systematics 45(1): 471.
Trenbath, B.R. 1974. Biomass Productivity of Mixtures. In: Brady, N.C., editor, Advances in Agronomy. Academic Press. p. 177–210
van Ruijven, J., E. Ampt, D. Francioli, and L. Mommer. 2020. Do soil-borne fungal pathogens mediate plant diversity–productivity relationships? Evidence and future opportunities. Journal of Ecology.
Picasso, D.V.D. 2018. The “Biodiversity–Ecosystem function debate”: An interdisciplinary dialogue between Ecology, Agricultural Science, and Agroecology. Agroecology and Sustainable Food Systems 42(3): 264–273. doi: 10.1080/21683565.2017.1359806.
“The “biodiversity-ecosystem function debate” is considered one of the most heated recent scientific issues within the discipline of Ecology. However, it can be better understood as an interdisciplinary dialogue between Ecology, Agricultural Science, and Agroecology.”