Plant diversity and yield in agricultural grasslands

There is a lot of discussion about the use of multi-species mixtures within grazed grassland systems. Here, I take a look at *some* of the evidence - but note that this is definitely NOT intended to be a literature review. 

In assessing the benefits associated with multi-species mixtures, it’s important to disentangle the different and potentially confounding effects of plant richness, supply of nitrogen, and whether research has been conducted on grazed or mechanically harvested plots. Therefore, I tend to clarify these points about the different studies involved. 

Diversity promotes yield in semi-natural grasslands

Ecological research conducted in relatively species-rich and nutrient-poor systems very clearly shows that grassland yield is enhanced by species diversity (e.g. reviewed in Hooper at al. 2005, Cardinale et al. 2007), and that multiple ecological processes may be more effective when species diversity is higher. In many studies (e.g. Hector et al 1999) one sees a very rapid response (ascending limb) from 1-5 species, and a smaller but consistent response as richness increases further (see Fig. 1 below).

Note, however, that the level of yield from the more species-rich mixtures is not as high as would be obtained if other (agronomic) species were used and with added inorganic fertiliser. However, the purpose of these experiments was to demonstrate the effect of biodiversity loss within natural ecosystems i.e the consequences as one goes from higher (right hand side of Fig. 1) to lower diversity (left hand side of Fig. 1). In addition, adding even modest amounts of nitrogen fertiliser to species-rich grasslands results in a very strong reduction in plant diversity (Kleijn et al. 2009). 

Fig. 1 Aboveground biomass harvested from plots sown with known numbers of species in a common experiment across multiple EU sites as part of EU BIODEPTH experiment. Adapted from Hector et al. 1999. 

Relevance to intensively managed agricultural grasslands…

The ability of more diverse plant communities to acquire available resources and convert them into above-ground biomass has obvious relevance for agricultural systems, although the results from extensively-managed (low-nutrient) semi-natural grasslands do not necessarily extrapolate to intensively-managed grasslands. Although ecological principles from diversity-function research therefore suggest that increasing plant species diversity in agronomic systems may improve biomass production, this has not been extensively tested. The potential multiple benefits of diverse agronomic crops with more than two species have been under-researched, but could have important implications for more sustainable agricultural practices by providing sufficient crop yield while minimising environmental impacts.

More recently, a number of studies have been investigating multi-species mixtures. I discuss some of these here. 

Four-species grass/legume mixtures generally outyielded best-performing monoculture

The Agrodiversity experiment was conducted at each of 31 sites (17 European countries and Canada) (Fig. 2). Fifteen grassland communities comprising four monocultures and eleven mixtures of four functional types of species (each represented by one species) (see Finn et al., 2013 for details). The choice of species for use in multi-species mixtures can be strategically designed to include traits that maximise complementarity and interspecific interactions to improve resource utilisation and yield of above-ground biomass. Thus, we selected functional types that consisted of a fast-establishing grass, a fast-establishing legume, a temporally persistent grass and a temporally persistent legume. The selection of species varied across sites, but the most commonly used species were Lolium perenne L., Trifolium pratense L., Dactylis glomerata L., and Trifolium repens L. All plots were mechanically harvested. 

Fig. 2.  Distribution of the European sites that participated in the Agrodiversity experiment.
(Canadian site not shown.)

There was a considerable range in site productivity, reflecting the different geographical and climatic regions across the study sites. Annual averages of total yield (dry matter) per site ranged from about 18 t ha^-1 year^-1 to about 3 t ha^-1 year^-1. Across all sites, yield of sown species (not including weed biomass, and averaged across years and seed density) of mixtures exceeded that of the mean monoculture in 99.7% of mixture communities with an average (across all mixtures and sites) ratio of mixture/monoculture yield of 1.77 (Fig. 3). Transgressive overyielding (better yields than in the best monoculture) occurred in 79% of mixture communities and was significant at 71% of sites. At sixteen sites, all of the mixture communities yielded more than the best monoculture community. The yield benefit of mixtures was already evident in year 1, and persisted for the three years of the experiment. 

Across all sites, monocultures displayed much higher levels (and variability) of weed invasion than mixtures. The median percentage of weed biomass in the total yield of monocultures increased over time (15% in year 1, 20% in year 2 and 32% in year 3); in contrast, the median percentage of weed biomass in the mixtures remained consistently low (4% in year 1, 3% in year 2 and 3% in year 3). 

Fig. 3. Mixtures generally yielded more than the best monoculture across 31 international sites with a common field experiment comparing four-species mixtures with the four component monocultures. Average annual yield over the whole experimental duration of yield of sown agronomic species (excludes weeds) at each of 31 sites. Each open circle represents the average aboveground biomass for a specific mixture community across multiple years; horizontal lines represent the yield of the best-performing monoculture; shaded boxes represent the mean monoculture performance. Significant transgressive overyielding is indicated by an asterisk over a site at the top of the panel. From Finn et al. (2013). The site numbers correspond to individual experiments. Data available from Kirwan et al. (2014).

 The Agrodiversity experiment varied considerably in the use of inorganic nitrogen, with several sites operating under organic status, and the low-fertility sites only applied low or no inorganic nitrogen. However, the more productive sites did apply nitrogen at intermediate rates e.g. between 100 and 200 kg N per ha per year. What was absent from the experiment was a comparison with a monoculture with a high level of nitrogen that would act as a positive control i.e. to test whether the mixtures could outperform a monoculture with high N. Nevertheless, what was clear was that the magnitude of mixture benefits was sufficient for mixtures to regularly yield more than the best-performing agronomic monoculture (transgressive overyielding). This is striking for three main reasons: first, the rapidity and frequency of transgressive overyielding; second, the occurrence of transgressive overyielding across such a wide range of mixtures, and; third its occurrence across such a wide range of sites that differed in soil type, productivity and climate. 

Having demonstrated strong benefits of four-species mixtures, we are now looking at the benefits of six species mixtures with two grasses, two legumes and two herbs. The 2018 (with natural drought) total yields in plots that were mechanically harvested were as follows:

• Six-species mix 150 kg N    – 12.5 t/ha
• Ryegrass 300 kg N             – 11.2 t/ha
• Ryegrass 150 kg N             – 9.4 t/ha

This research is summarised in a recent popular article  by Guylain Grange and Saoirse Cummins  (Agriland article online). 

Grass-legume mixtures maintained yield despite substantial reductions in N fertiliser

Grass-legume mixtures in grassland forage systems can benefit from symbiotic N₂ fixation of legumes, thereby increasing total harvest yield, total N yield, and forage quality. Because legumes have access to atmospheric N₂ for their N requirements, the relative availability of soil N increases for grasses in mixtures due to ‘N sparing’ (increased availability of soil N because legumes rely on symbiotic N₂ fixation). Therefore, the use of grass-legume mixtures could allow substantial reductions in amounts of industrial N fertiliser in agricultural grassland systems without a compromise in yield. At the Swiss site of the Agrodiversity experiment, Nyfeler et al. (2009) compared monoculture and mixture yields across three levels of nitrogen (50, 150 and 450 kg N ha^-1 year^-1) (Fig. 4). Their results indicated a high potential for N-fertilizer replacement: grass–clover mixtures containing 40–60% clover and receiving 50 or 150 kg N ha^-1 year^-1 achieved the same yield as grass monocultures fertilized with 450 kg N ha^-1 year^-1 (Nyfeler et al., 2009). Diversity–productivity effects were reduced at the highest level of N fertilization and at 450 kg N ha^-1 year^-1, they virtually disappeared in the third year.

These results illustrate an example of sustainable intensification and 'more from less'; it is possible to get more yield from less input by using legumes to replace the input of nitrogen fertiliser. 

Fig. 4. Monoculture and mixture yields across three levels of nitrogen (50, 150 and 450 kg N ha^-1 year^-1), and presented in relation to the proportion of legume in the sward. Thus, the left hand side represents 100% grass and 0% legume; the mid-point represents yield with 50% grass and 50% legume, and; the right hand side represents yield with 0% grass and 100 legume. The red line represents the performance of the 100% grass sward with 450 kg/ha of nitrogen added as a reference point. From Nyfeler et al. (2009). 

Yield benefits of mixtures persist under grazing

Many experiments on multi-species mixtures have been conducted under conditions where the forage has been harvested by mowing (although variety testing for grasses and clovers in pastures is harvested in the same way). Do the benefits of mixtures that are observed under mowing also prevail under grazing? There are many experiments with grass-clover (2-species) mixtures under grazing; however, comparisons of more species-rich combinations are less common. 
A recent study investigated whether grazing modifies the benefits of mixtures on total N yield compared to mowing (Huguenin-Elie et al., 2016). The design included N_2-fixing and non-fixing species, as well as shallow- and deep-rooting species. Lolium perenne (Lp) monoculture and mixtures with Cichorium intybus (Ci), or/and Trifolium repens (Tr) and Trifolium pratense (Tp) were compared under grazing or mowing for their N yield and capture of fertilizer and atmospheric N₂. Mixtures of the N₂ fixing and the non-fixing species with 145 kg N ha^-1 yr^-1 yielded as much N as the L. perenne monoculture fertilized with 350 kg N ha^-1 yr^-1, showing the considerable benefit of mixtures for N efficiency. The benefits of the mixtures on N yield were similar under grazing and mowing. Grazing did not modify the proportion of N derived from fertilizer and symbiotic N₂ fixation in the plants (Huguenin-Elie et al., 2016).

In a two-year grazing experiment (with 75 kg N ha^-1 yr^-1) in France, an increase of botanical complexity from one to five species (two grasses, two clovers and chicory) resulted in positive effects on animal performance (Roca-Fernández et al., 2016). They distinguished between monocultures of perennial ryegrass, ‘mixed swards’ of grass and clover, and ‘multi-species swards’ of grasses, clovers and chicory. Compared to mixed swards, multi-species swards improved production of milk (+0.8 kg/day) and milk solids (+0.04 kg/day), which was attributed to enhanced sward quality and increased dry matter intake (+1.5 kg DM/day).

In a two-year grazing experiment in the northeastern USA (no nitrogen fertiliser was applied), four mixture communities were compared: two species (one grass, one legume), three species (one grass, one legume, and chicory), six species (three grasses, two legumes and chicory), and nine species (four grasses, four legumes and chicory). In a dry year, the two-species mixture (4800 kg ha^-1 dry matter) yielded less than the other mixtures (7600 kg ha^-1 dry matter); there was no difference in dry matter yields (9800 kg ha^-1 dry matter) in the year with plentiful rainfall (Sanderson et al. 2005).

More recently in Ireland, the Smartgrass project established farmlets with perennial ryegrass (Lolium perenne) only, receiving 163 kg N ha⁻¹ year⁻¹ (PRG); a perennial ryegrass and white clover (Trifolium repens) sward (PRGWC); a six species sward containing two grasses, two legumes and two herbs (6S); and a nine species sward containing three grasses, three legumes and three herbs (9S), and each of the latter three treatments receiving 90 kg N ha⁻¹ year⁻¹. Lambs grazing the multispecies 6S and 9S swards had greater liveweight gain and body condition score than those in the PRG sward, and required fewer anthelmintic treatment than lambs on the PRG or PRGWC swards (Grace et al. 2019b). Grace et al. (2018) concluded that “Multispecies swards did not differ in terms of annual DM production despite lower N inputs compared to PRG swards over a two year study. Sward chemical composition was largely similar between sward types. However, herb content of the swards decreased over the duration of this study.”

Overall, these experiments point to the ability of more complex (>2 species) mixtures to improve yields for livestock production, and with less input of inorganic nitrogen. Several studies have reviewed the contribution of grass-legume mixtures to the nutrition and production of livestock (e.g. Lüscher et al. 2014, Dewhurst et al. 2009).

As have many other authors, Grace et al. (2019a) stated that “Multispecies swards show potential for higher DM production from lower N inputs compared with PRG‐only swards. However, maintaining the proportions of key species may affect the productivity of such swards." This is an important point: to achieve the benefits of multispecies mixtures under intensive management while maintaining the multiple species that provide the benefits. This is where systems-level research can help to investigate management to promote species’ persistence (duration of grazing rotation), and implement effective techniques to re-introduce species (e.g. oversowing) that may be declining in the sward. 

Other links:
Multi-species mixtures promote yield stability (under drought conditions)

Four-species mixtures increased weed suppression in intensively managed grasslands

Acknowledgements

Much of the above content has been taken from a conference paper by Finn et al. and submitted in 2017 to the 54^th Annual Meeting of the Brazilian Society of Animal Science.

 References

Cardinale B J, Wright J P, Cadotte M W, Carroll I T, Hector A, Srivastava D S, Loreau M, Weis J J. 2007. Impacts of plant diversity on biomass production increase through time because of species complementarity. PNAS 104: 18123-18128.

Dewhurst, R. J., Delaby, L., Moloney, A., Boland, T., and Lewis, E. (2009). Nutritive value of forage legumes used for grazing and silage. Irish Journal of Agricultural and Food Research, 167-187.

Finn et al. (2013), Ecosystem function enhanced by combining four functional types of plant species in intensively managed grassland mixtures: a 3-year continental-scale field experiment. Journal of Applied Ecology, 50: 365–375.

Grace, C., Boland, T.M., Sheridan, H., Lott, S., Brennan, E., Fritch, R. and Lynch, M.B., 2018. The effect of increasing pasture species on herbage production, chemical composition and utilization under intensive sheep grazing. Grass and Forage Science, 73(4), pp.852-864.

Grace, C., Boland, T.M., Sheridan, H., Brennan, E., Fritch, R. and Lynch, M.B., 2019a. The effect of grazing versus cutting on dry matter production of multispecies and perennial ryegrass‐only swards. Grass and Forage Science.

Grace, C., Lynch, M.B., Sheridan, H., Lott, S., Fritch, R., Boland, T.M. 2019b. Grazing multispecies swards improves ewe and lamb performance. Animal 13: 1721-1729.

Hooper D.U., Chapin F.S., Ewel J.J., Hector A., Inchausti P., Lavorel S., Lawton J.H. et al. (2005) Effects of biodiversity on ecosystem functioning: A consensus of current knowledge. Ecological Monographs 75, 3-35.

Huguenin-Elie O., Husse S., Buchmann N. and Lüscher A. (2016) Mixtures provided similar benefits to nitrogen yield under grazing and under mowing. Grassland Science in Europe, 21: 609-611.

Kirwan et al. 2007. Evenness drives consistent diversity effects in intensive grassland systems across 28 European sites. Journal of Ecology 95: 530-539.

Kirwan et al. 2014. The Agrodiversity Experiment: three years of data from a multi-site plant diversity experiment in intensively managed grasslands. Ecological Archives, 95: 2680.

Kleijn, D., Kohler, F., Báldi, A., Batáry, P., Concepción, E.D., Clough, Y., Diaz, M., Gabriel, D., Holzschuh, A., Knop, E. and Kovács, A. 2009. On the relationship between farmland biodiversity and land-use intensity in Europe. Proceedings of the Royal Society of London B: Biological Sciences 276: 903-909.

Lüscher, A., Mueller-Harvey, I., Soussana, J.F., Rees, R.M. & Peyraud, J.L. (2014) Potential of legume-based grassland–livestock systems in Europe: a review. Grass and Forage Science, 69, 206-228.

Nyfeler, D., Huguenin-Elie, O., Suter, M., Frossard, E., Connolly, J. & Lüscher, A. (2009) Strong mixture effects among four species in fertilized agricultural grassland led to persistent and consistent transgressive overyielding. Journal of Applied Ecology, 46, 683-691.

Nyfeler, D., Huguenin-Elie, O., Suter, M., Frossard, E. & Lüscher, A. (2011) Grass-legume mixtures can yield more nitrogen than legume pure stands due to mutual stimulation of nitrogen uptake from symbiotic and non-symbiotic sources. Agriculture Ecosystems & Environment, 140, 155-163.

Roca-Fernández, I, Peyraud, J. L., Delaby, L. and R. Delagarde. 2016.  Pasture intake and milk production of dairycows rotationally grazing on multi-species swards. Animal, 10, 1448-1456. 

Sanderson, M., Soder, K., Muller, L., Klement, K., Skinner, R., Goslee, S., 2005. Forage mixture productivity and botanical composition in pastures grazed by dairy cattle. Agron. J. 97, 1465-1471.