Biodiversity increases the resistance of ecosystem productivity to climate extremes


It remains unclear whether biodiversity buffers ecosystems against climate extremes, which are becoming increasingly frequent worldwide1. Early results suggested that the ecosystem productivity of diverse grassland plant communities was more resistant, changing less during drought, and more resilient, recovering more quickly after drought, than that of depauperate communities2. However, subsequent experimental tests produced mixed results3,4,5,6,7,8,9,10,11,12,13. Here we use data from 46 experiments that manipulated grassland plant diversity to test whether biodiversity provides resistance during and resilience after climate events. We show that biodiversity increased ecosystem resistance for a broad range of climate events, including wet or dry, moderate or extreme, and brief or prolonged events. Across all studies and climate events, the productivity of low-diversity communities with one or two species changed by approximately 50% during climate events, whereas that of high-diversity communities with 16–32 species was more resistant, changing by only approximately 25%. By a year after each climate event, ecosystem productivity had often fully recovered, or overshot, normal levels of productivity in both high- and low-diversity communities, leading to no detectable dependence of ecosystem resilience on biodiversity. Our results suggest that biodiversity mainly stabilizes ecosystem productivity, and productivity-dependent ecosystem services, by increasing resistance to climate events. Anthropogenic environmental changes that drive biodiversity loss thus seem likely to decrease ecosystem stability14, and restoration of biodiversity to increase it, mainly by changing the resistance of ecosystem productivity to climate events.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Biodiversity effects on ecosystem stability, and its resistance and resilience components.
Figure 2: Effects of biodiversity on stability measures with climate events defined over shorter or longer durations.
Figure 3: Biodiversity effects on productivity during climate events or normal years.


  1. 1

    Stocker, T. F. et al. IPCC 2013: Summary for Policy Makers (Cambridge Univ. Press, 2013)

    Google Scholar 

  2. 2

    Tilman, D. & Downing, J. A. Biodiversity and stability in grasslands. Nature 367, 363–365 (1994)

    ADS  Article  Google Scholar 

  3. 3

    Pfisterer, A. B. & Schmid, B. Diversity-dependent production can decrease the stability of ecosystem functioning. Nature 416, 84–86 (2002)

    ADS  CAS  Article  Google Scholar 

  4. 4

    Caldeira, M. C., Hector, A., Loreau, M. & Pereira, J. S. Species richness, temporal variability and resistance of biomass production in a Mediterranean grassland. Oikos 110, 115–123 (2005)

    Article  Google Scholar 

  5. 5

    Kahmen, A., Perner, J. & Buchmann, N. Diversity-dependent productivity in semi-natural grasslands following climate perturbations. Funct. Ecol. 19, 594–601 (2005)

    Article  Google Scholar 

  6. 6

    Steiner, C. F., Long, Z. T., Krumins, J. A. & Morin, P. J. Population and community resilience in multitrophic communities. Ecology 87, 996–1007 (2006)

    Article  Google Scholar 

  7. 7

    Wang, Y., Yu, S. & Wang, J. Biomass-dependent susceptibility to drought in experimental grassland communities. Ecol. Lett. 10, 401–410 (2007)

    Article  Google Scholar 

  8. 8

    Ives, A. R. & Carpenter, S. R. Stability and diversity of ecosystems. Science 317, 58–62 (2007)

    ADS  CAS  Article  Google Scholar 

  9. 9

    Griffin, J. N. et al. in Biodiversity, Ecosystem Functioning, & Human Wellbeing: An Ecological and Economic Perspective (eds Naeem, S. et al.) Ch. 6 78–93 (Oxford Univ. Press, 2009)

    Google Scholar 

  10. 10

    van Ruijven, J. & Berendse, F. Diversity enhances community recovery, but not resistance, after drought. J. Ecol. 98, 81–86 (2010)

    Article  Google Scholar 

  11. 11

    Vogel, A., Scherer-Lorenzen, M. & Weigelt, A. Grassland resistance and resilience after drought depends on management intensity and species richness. PLoS One 7, e36992 (2012)

    ADS  CAS  Article  Google Scholar 

  12. 12

    Grossiord, C. et al. Tree diversity does not always improve resistance of forest ecosystems to drought. Proc. Natl Acad. Sci. USA 111, 14812–14815 (2014)

    ADS  CAS  Article  Google Scholar 

  13. 13

    Wright, A. J. et al. Flooding disturbances increase resource availability and productivity but reduce stability in diverse plant communities. Nature Commun. 6, 6092 (2015)

    ADS  CAS  Article  Google Scholar 

  14. 14

    Hautier, Y. et al. Plant ecology. Anthropogenic environmental changes affect ecosystem stability via biodiversity. Science 348, 336–340 (2015)

    ADS  CAS  Article  Google Scholar 

  15. 15

    McNaughton, S. J. Diversity and stability of ecological communities: a comment on the role of empiricism in ecology. Am. Nat. 111, 515–525 (1977)

    Article  Google Scholar 

  16. 16

    Naeem, S. & Li, S. Biodiversity enhances ecosystem reliability. Nature 390, 507–509 (1997)

    ADS  CAS  Article  Google Scholar 

  17. 17

    Bai, Y., Han, X., Wu, J., Chen, Z. & Li, L. Ecosystem stability and compensatory effects in the Inner Mongolia grassland. Nature 431, 181–184 (2004)

    ADS  CAS  Article  Google Scholar 

  18. 18

    Tilman, D., Reich, P. B. & Knops, J. M. H. Biodiversity and ecosystem stability in a decade-long grassland experiment. Nature 441, 629–632 (2006)

    ADS  CAS  Article  Google Scholar 

  19. 19

    Balvanera, P. et al. Quantifying the evidence for biodiversity effects on ecosystem functioning and services. Ecol. Lett. 9, 1146–1156 (2006)

    Article  Google Scholar 

  20. 20

    Hector, A. et al. General stabilizing effects of plant diversity on grassland productivity through population asynchrony and overyielding. Ecology 91, 2213–2220 (2010)

    CAS  Article  Google Scholar 

  21. 21

    de Mazancourt, C. et al. Predicting ecosystem stability from community composition and biodiversity. Ecol. Lett. 16, 617–625 (2013)

    Article  Google Scholar 

  22. 22

    Polley, H. W., Isbell, F. I. & Wilsey, B. J. Plant functional traits improve diversity-based predictions of temporal stability of grassland productivity. Oikos 122, 1275–1282 (2013)

    Article  Google Scholar 

  23. 23

    Gross, K. et al. Species richness and the temporal stability of biomass production: a new analysis of recent biodiversity experiments. Am. Nat. 183, 1–12 (2014)

    Article  Google Scholar 

  24. 24

    Pimm, S. L. The complexity and stability of ecosystems. Nature 307, 321–326 (1984)

    ADS  Article  Google Scholar 

  25. 25

    Huston, M. A. Hidden treatments in ecological experiments: re-evaluating the ecosystem function of biodiversity. Oecologia 110, 449–460 (1997)

    ADS  Article  Google Scholar 

  26. 26

    Cardinale, B. J. et al. The functional role of producer diversity in ecosystems. Am. J. Bot. 98, 572–592 (2011)

    Article  Google Scholar 

  27. 27

    Tilman, D., Isbell, F. & Cowles, J. M. Biodiversity and ecosystem functioning. Annu. Rev. Ecol. Evol. Syst. 45, 471–493 (2014)

    Article  Google Scholar 

  28. 28

    Tilman, D. Biodiversity: population versus ecosystem stability. Ecology 77, 350–363 (1996)

    Article  Google Scholar 

  29. 29

    Beguería, S., Vicente-Serrano, S. M. & Angulo-Martinez, M. A multiscalar global drought dataset: the SPEIbase: a new gridded product for the analysis of drought variability and impacts. Bull. Am. Meteorol. Soc. 91, 1351–1356 (2010)

    ADS  Article  Google Scholar 

  30. 30

    Vicente-Serrano, S. M. et al. Response of vegetation to drought time-scales across global land biomes. Proc. Natl Acad. Sci. USA 110, 52–57 (2013)

    ADS  CAS  Article  Google Scholar 

Download references


This paper is a product of the STABILITY group funded by sDiv, the Synthesis Centre of the German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig (DFG FZT 118). M.L. was supported by the TULIP Laboratory of Excellence (ANR-10-LABX-41). B.S. and P.A.N. were supported by the URPP Global Change and Biodiversity of the University of Zurich.

Author information




F.I. and N.E. conceived the project; F.I., D.C., J.C., M.L., H.B., A.E., J.N.G., Y.H., A.H., P.M., S.T.M., A.M., K.E.M., S.N., C.R., E.S., M.P.T., J.v.R., A.W., W.W., B.W., and N.E. developed the project at a workshop; F.I. and M.L. defined dimensionless measures of resistance and resilience; F.I., D.C., J.C., B.S., C.B., M.B., C.B., H.B., E.d.L., Q.G., A.H., A.J., J.K., V.L., S.T.M., H.W.P., P.B.R., C.R., D.T., B.T., W.v.d.P., J.v.R., A.W., W.W., B.W., and N.E. contributed experimental data; D.C. assembled data; F.I. analysed data, with substantial input from J.C. and B.S.; and F.I. wrote the paper, with substantial input from all authors.

Corresponding author

Correspondence to Forest Isbell.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 Contrasting ecosystem productivity responses to climate events for low or high levels of resistance (Ω) and resilience (Δ).

In these stylized examples, productivity is decreased by a dry climate event during year one, is increased by a wet climate event during year 11, and is otherwise recovering back towards normal productivity levels either monotonically (black dashed lines and open triangles) or via damped oscillations (solid grey lines and filled circles). Ecosystem stability (μ/σ) depends on both resistance and resilience. See Methods for definitions of resistance and resilience.

Extended Data Figure 2 Map of study site locations (bottom) and frequency of climate events (top).

Bottom: locations for all 46 studies (yellow triangles) and an example of spatial variation in water balance, where SPEI-12 was classified as in the bottom panel. August 2005 was chosen for this example because many experiments were underway and harvested during this particular month of this particular year (Extended Data Table 1). The spatial patterns of wet and dry climate events shown on this map would differ at other times (that is, during a different month or year) and for climate events defined over other durations (that is, based on water balances aggregated over more or fewer than the preceding 12 months). There were multiple experiments at some sites (Extended Data Table 1), thus some symbols completely overlap on this map. Top: cutoffs for bins correspond to events occurring every 1 in 4 years (±0.67) or every 1 in 10 years (±1.28).

Extended Data Figure 3 Classification of extreme dry, moderate dry, normal, moderate wet, and extreme wet years for each year of the 46 experiments.

The 12-month version of the SPEI is shown, where positive values indicate wetter than normal water balances (precipitation minus potential evapotranspiration) during the 12-month time interval preceding and including the month of peak biomass harvest. For example, if peak biomass was harvested in September, then SPEI-12 accounts for the water balance from the previous October to September. Drought index values are based on month-by-month variations in climate over the past century (January 1901 to December 2011), based on monthly means of measurements made at more than 4,000 weather stations worldwide, and provided on 0.5 degree × 0.5 degree grids globally. Dashed lines show cutoffs for 1 in 4 (±0.67) or 1 in 10 (±1.28) year events. Seven experiments that included only normal years (Agrodiversity Germany a, Agrodiversity Ireland a, Czech Republic) or that did not include any normal years (Agrodiversity Poland a, Agrodiversity Spain a, Iowa BioGEN, North Dakota a) were excluded from subsequent analyses because it was not possible to compare perturbed with normal productivity levels for these studies.

Extended Data Figure 4 Productivity during and after both climate events and normal years for monocultures and mixtures of 16 species.

Values shown are predicted means and 95% confidence intervals from the mixed-effects model. Productivity tends to be decreased during dry events and increased during wet events. This trend is reversed during the year after climate events. This pattern of overshooting normal levels of productivity during recovery 1 year after climate events is consistent with damped oscillations, rather than monotonic recovery (Extended Data Fig. 1). Relatively high productivity after extreme droughts could be due to increased nutrient availability and/or decreased abundance of herbivores as a result of reduced plant productivity during the drought. This might be especially true for low-diversity communities, which have the lowest productivity during drought, possibly explaining why biodiversity increases resilience after extremely dry years (Fig. 1c). Similarly, relatively low productivity after extremely wet years might be due to decreased nutrient availability and/or increased abundance of enemies as a result of increased plant productivity during the wet event. This might be especially true for high-diversity communities, which have the highest productivity during wet years, possibly explaining why biodiversity decreases resilience after extremely wet years (Fig. 1c). Dashed horizontal lines show normal productivity levels.

Extended Data Figure 5 Biodiversity–productivity relationships for each year of each study, including normal years and climate events.

Points are plot-level values and lines are mixed-model fits (Fig. 3).

Extended Data Figure 6 A marginally significant interaction between biodiversity and intensity (moderate or extreme).

Table 1 indicates that productivity was marginally more resistant to moderate than to extreme climate events, especially at high biodiversity. All other interactions were non-significant (P > 0.10). Axes are logarithmic.

Extended Data Figure 7 Biodiversity effects on the resistance of productivity to climate extremes.

Shown for each study for which there were observations of productivity during both normal () and climate event (Ye) years (Extended Data Fig. 3). Points are plot-level values and lines are mixed-model fits (Fig. 1b). Axes are logarithmic.

Extended Data Figure 8 Biodiversity effects on the resilience of productivity to climate extremes.

Shown for each study for which there were observations during normal (), climate event (Ye), and post-climate event (Ye + 1) years. Quantifying resilience requires more information (that is, Ye + 1) than quantifying resistance, thus we were unable to quantify resilience for eight of the studies shown in Extended Data Fig. 7. Specifically, we were unable to quantify resilience for studies where the only climate event occurred during the last year of the study (Extended Data Fig. 3) because in this case Ye + 1 is unknown, and for studies where the only normal year was also the only post-event year (Extended Data Fig. 3) because in this case  = Ye + 1 and resilience is undefined. Points are plot-level values and lines are mixed-model fits (Fig. 1c). Axes are logarithmic.

Extended Data Figure 9 Biodiversity effects on the resistance of productivity to climate events that were preceded either by a climate event (green lines) or by a normal year (black lines).

The significant interaction shown here indicates that biodiversity increased resistance more during climate events preceded by years with climate events than during climate events preceded by normal years (F1,64.8 = 7.21, P < 0.01). Axes are logarithmic. The sequence of climate events at each site is shown in Extended Data Fig. 3.

Extended Data Table 1 Study details

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Isbell, F., Craven, D., Connolly, J. et al. Biodiversity increases the resistance of ecosystem productivity to climate extremes. Nature 526, 574–577 (2015).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.