Higher species richness enhances yield stability in intensively managed grasslands with experimental disturbance

Climate models predict increased frequency and severity of drought events. At an Irish and Swiss site, experimental summer droughts were applied over two successive years to grassland plots sown with one, two or four grassland species with contrasting functional traits. Mean yield and plot-to-plot variance of yield were measured across harvests during drought and after a subsequent post-drought recovery period. At both sites, there was a positive relationship between species richness and yield. Under rainfed control conditions, mean yields of four-species communities were 32% (Wexford, Ireland) and 51% (Zürich, Switzerland) higher than in monocultures. This positive relationship was also evident under drought, despite significant average yield reductions (−27% at Wexford; −21% at Zürich). Four-species communities had lower plot-to-plot variance of yield compared to monoculture or two-species communities under both rainfed and drought conditions, which demonstrates higher yield stability in four-species communities. At the Swiss but not the Irish site, a high degree of species asynchrony could be identified as a mechanism underlying increased temporal stability in four-species communities. These results indicate the high potential of multi-species grasslands as an adaptation strategy against drought events and help achieve sustainable intensification under both unperturbed and perturbed environmental conditions.


Drought treatment
A summer drought event of nine to ten weeks was simulated at each site over two years (2013 and 2014 at Wexford; 2012 and 2013 at Zürich). During drought periods, precipitation was excluded completely from one randomly selected split-plot in each experimental main-plot using tunnel-shaped rain-out shelters (see Fig. S4). Shelters consisted of steel frames (3 m × 5.5 m and a height of 140 cm) covered with a transparent plastic foil (SunMaster SuperThermic, 150 μm, XL Horticulture, UK, at Wexford and Gewächshausfolie UV5, 200 μm, Folitec Agrarfolien-Vertrieb, Germany, at Zürich). To ensure minimal changes to ambient temperature and relative humidity, shelters had a ventilation opening of 35 cm over the entire length at the top and at both sides at the bottom and were left open at each end. At Zürich in year one, the drought period had to be restarted after 5 weeks due to a heavy thunderstorm.
Microclimatic parameters were measured under rainfed control and drought conditions during the drought period (Fig. S3). Relative humidity and air temperature were measured every 10 minutes at Wexford (temperature and relative humidity sensor, Voltcraft, Switzerland) and hourly at Zürich (temperature and relative humidity sensor, Decagon, US). Precipitation and other meteorological data were collected by respective national meteorological stations at a maximum distance of 1.4 km from each site.

Site management
Harvesting and fertiliser applications at each site were conducted in accordance with local management practices. Aboveground biomass was harvested five and six times annually at Wexford and Zürich, respectively, from a central strip of 5 m × 1.5 m in each split-plot. Biomass was cut at a height of 5 cm in Wexford using a Haldrup plot combine (HALDRUP GmbH., Germany) and at 7 cm in Zürich using a Hege 212 plot harvester (Wintersteiger GmbH, Austria).
Dry matter content of each plot yield was determined by drying a subsample of bulk mass until constant weight (40 °C for 72 h for Wexford samples and at 100 °C for 24 h for Zürich samples).

Appendix S2 Soil moisture content measurements and determination of the threshold of plant-available soil water
At both sites soil moisture content (SMC) was measured weekly in the plots with equiproportional mixtures at two depths under control and drought conditions. Measurements were recorded at 10 cm and 40 cm soil depth at Wexford (n = 3 per depth, PR2 Probe, Delta-T Devices, Cambridge, UK), and at 5 cm and 40 cm soil depth at Zürich (n = 3 per depth, 5TM sensor, Decagon, USA) under drought and control conditions (Figs. 2, S2). While useful, one limitation of SMC is that this metric does not reflect the physical characteristics of the soil 2,3 . To permit more informative intra-and inter-site comparisons of drought effects and severity, the approximate threshold of plant-available soil water was determined at both sites. Intact soil-cores (5 cm diameter) were removed from each of the three plots with equal species proportions under rainfed control and drought conditions at 10 and 40 cm deep (n = 2 replicates per depth). Using a standardised pressure plate method 4 , soil water retention curves (the relationship between SMC and soil matric potential) were determined for each of these plots and soil depths, providing a metric to quantify water stress. For comparisons of drought effects, we refer to a soil matric potential of -1.5 MPa because this value gives an estimation of the lower limit to which a plant can extract water from a specific soil, and varies depending on soil physical and chemical characteristics. In our experiment, differences between sites at the threshold of -1.5 MPa were strongly driven by soil physical properties, particularly at 40 cm depth (Fig. S2), which the SMC alone does not reflect. This highlights the efficacy of including measurements of the soil matric potential to aid quantification of drought stress in accompaniment to SMC measurements 2 .
Fertiliser was split into four and six applications per year in Wexford and Zürich, respectively.

Table S1
Summary of mean yields and standard deviations (SD, in parentheses) across all six harvests. Mean yields (μjk) and corresponding SDs (σjk) for monocultures, two-species mixtures, and four-species mixtures were calculated based on regression analysis (eqn. 1, with j and k as defined in the main text). Note that this data is a reorganisation of the data presented in Fig. 4.
For richness levels within each of the site x treatment combinations, values with different superscripts (yield: upper case letters; SD: lower case letters) are significantly different at P < 0.05, except SD under drought at Zürich, which is at P < 0.1.

Mean yield (t ha −1 )
Wexford Zürich    Grey shading indicates the periods when rain was excluded from drought-treatment plots. The horizontal reference line is the soil moisture content that corresponds to a soil matric potential of -1.5 MPa, which is the approximate threshold of plant-available soil water. C. D.