Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Letter
  • Published:

Holocene shifts in the assembly of plant and animal communities implicate human impacts

A Corrigendum to this article was published on 24 August 2016

This article has been updated


Understanding how ecological communities are organized and how they change through time is critical to predicting the effects of climate change1. Recent work documenting the co-occurrence structure of modern communities found that most significant species pairs co-occur less frequently than would be expected by chance2,3. However, little is known about how co-occurrence structure changes through time. Here we evaluate changes in plant and animal community organization over geological time by quantifying the co-occurrence structure of 359,896 unique taxon pairs in 80 assemblages spanning the past 300 million years. Co-occurrences of most taxon pairs were statistically random, but a significant fraction were spatially aggregated or segregated. Aggregated pairs dominated from the Carboniferous period (307 million years ago) to the early Holocene epoch (11,700 years before present), when there was a pronounced shift to more segregated pairs, a trend that continues in modern assemblages. The shift began during the Holocene and coincided with increasing human population size4,5 and the spread of agriculture in North America6,7. Before the shift, an average of 64% of significant pairs were aggregated; after the shift, the average dropped to 37%. The organization of modern and late Holocene plant and animal assemblages differs fundamentally from that of assemblages over the past 300 million years that predate the large-scale impacts of humans. Our results suggest that the rules governing the assembly of communities have recently been changed by human activity.

This is a preview of subscription content, access via your institution

Access options

Buy this article

Prices may be subject to local taxes which are calculated during checkout

Figure 1: Proportion of aggregated pairs over the past 300 Myr.
Figure 2: Relationship between scale and proportion of aggregated pairs.
Figure 3: Tests of possible mechanisms for decreasing spatial aggregation through time.

Similar content being viewed by others

Change history

  • 06 January 2016

    ED Table 1 was corrected (specifically the data owner for Wisconsin plant data).


  1. Botkin, D. B. et al. Forecasting the effects of global warming on biodiversity. Bioscience 57, 227–236 (2007)

    Article  Google Scholar 

  2. Gotelli, N. J. & Ulrich, W. The empirical Bayes approach as a tool to identify non-random species associations. Oecologia 162, 463–477 (2010)

    Article  ADS  Google Scholar 

  3. Gotelli, N. J. & McCabe, D. J. Species co-occurrence: a meta-analysis of J.M. Diamond’s assembly rules model. Ecology 83, 2091–2096 (2002)

    Article  Google Scholar 

  4. Marlon, J. R. et al. Global biomass burning: a synthesis and review of Holocene paleofire records and their controls. Quat. Sci. Rev. 65, 5–25 (2013)

    Article  ADS  Google Scholar 

  5. Ruddiman, W. F. & Ellis, E. C. Effect of per-capita land use changes on Holocene forest clearance and CO2 emissions. Quat. Sci. Rev. 28, 3011–3015 (2009)

    Article  ADS  Google Scholar 

  6. Zeder, M. A. & Smith, B. D. A conversation on agricultural origins: talking past each other in a crowded room. Curr. Anthropol. 50, 681–691 (2009)

    Article  Google Scholar 

  7. Smith, B. D. & Yarnell, R. A. Initial formation of an indigenous crop complex in eastern North America at 3800 B.P. Proc. Natl Acad. Sci. USA 106, 6561–6566 (2009)

    Article  ADS  CAS  Google Scholar 

  8. Connor, E. F. & McCoy, E. D. The statistics and biology of the species-area relationship. Am. Nat. 113, 791–833 (1979)

    Article  MathSciNet  Google Scholar 

  9. Connor, E. F., Collins, M. D. & Simberloff, D. The checkered history of checkerboard distributions. Ecology 94, 2403–2414 (2013)

    Article  Google Scholar 

  10. Ulrich, W. & Gotelli, N. J. Null model analysis of species associations using abundance data. Ecology 91, 3384–3397 (2010)

    Article  Google Scholar 

  11. Wagner, P. J., Kosnik, M. A. & Lidgard, S. Abundance distributions imply elevated complexity of post-Paleozoic marine ecosystems. Science 314, 1289–1292 (2006)

    Article  ADS  CAS  Google Scholar 

  12. Wilf, P. & Labandeira, C. C. Response of plant-insect associations to Paleocene-Eocene warming. Science 284, 2153–2156 (1999)

    Article  CAS  Google Scholar 

  13. Lyons, S. K. A quantitative model for assessing community dynamics of Pleistocene mammals. Am. Nat. 165, E168–E185 (2005)

    Article  Google Scholar 

  14. Jackson, S. T. & Williams, J. W. Modern analogs in Quaternary paleoecology: here today, gone yesterday, gone tomorrow? Annu. Rev. Earth Planet. Sci. 32, 495–537 (2004)

    Article  ADS  CAS  Google Scholar 

  15. Ulrich, W. Pairs – a FORTRAN program for studying pair wise species associations in ecological matrices. (2008)

  16. Zachos, J. C., Dickens, G. R. & Zeebe, R. E. An early Cenozoic perspective on greenhouse warming and carbon-cycle dynamics. Nature 451, 279–283 (2008)

    Article  ADS  CAS  Google Scholar 

  17. Augustin, L. et al.; EPICA community members. Eight glacial cycles from an Antarctic ice core. Nature 429, 623–628 (2004)

    Article  ADS  CAS  Google Scholar 

  18. Bambach, R. K. Phanerozoic biodiversity mass extinctions. Annu. Rev. Earth Planet. Sci. 34, 127–155 (2006)

    Article  ADS  CAS  Google Scholar 

  19. Bowen, G. J. et al. Two massive, rapid releases of carbon during the onset of the Palaeocene–Eocene thermal maximum. Nature Geosci. 8, 44–47 (2015)

    Article  ADS  MathSciNet  CAS  Google Scholar 

  20. Blois, J. L. et al. A framework for evaluating the influence of climate, dispersal limitation, and biotic interactions using fossil pollen associations across the late Quaternary. Ecography 37, 1095–1108 (2014)

    Google Scholar 

  21. Delcourt, P. A. & Delcourt, H. R. Prehistoric Native Americans and Ecological Change: Human Ecosystems in Eastern North American since the Pleistocene. (Cambridge Univ. Press, 2004)

  22. Barnosky, A. D. Colloquium paper: megafauna biomass tradeoff as a driver of Quaternary and future extinctions. Proc. Natl Acad. Sci. USA 105 (suppl. 1), 11543–11548 (2008)

    Article  ADS  CAS  Google Scholar 

  23. Haddad, N. M. et al. Habitat fragmentation and its lasting impact on Earth’s ecosystems. Sci. Adv. 1, e1500052 (2015)

    Article  ADS  Google Scholar 

  24. Williams, M. Deforesting the Earth: From Prehistory To Global Crisis (Univ. Chicago Press, 2003)

  25. Grayson, D. K. The archaeological record of human impacts on animal populations. J. World Prehist. 15, 1–68 (2001)

    Article  Google Scholar 

  26. van Kleunen, M. et al. Global exchange and accumulation of non-native plants. Nature 525, 100–103 (2015)

    Article  ADS  CAS  Google Scholar 

  27. Capinha, C., Essl, F., Seebens, H., Moser, D. & Pereira, H. M. The dispersal of alien species redefines biogeography in the Anthropocene. Science 348, 1248–1251 (2015)

    Article  ADS  CAS  Google Scholar 

  28. Dornelas, M. et al. Assemblage time series reveal biodiversity change but not systematic loss. Science 344, 296–299 (2014)

    Article  ADS  CAS  Google Scholar 

  29. Tóth, A. B., Lyons, S. K. & Behrensmeyer, A. K. A century of change in Kenya’s mammal communities: increased richness and decreased uniqueness in six protected areas. PLoS ONE 9, e93092 (2014)

    Article  ADS  Google Scholar 

  30. Thompson, J. R., Carpenter, D. N., Cogbill, C. V. & Foster, D. R. Four centuries of change in northeastern United States forests. PLoS ONE 8, e72540 (2013)

    Article  ADS  CAS  Google Scholar 

  31. Ulrich, W. Pairs – a FORTRAN program for studying pair-wise species associations in ecological matrices., (2010)

  32. Stone, L. & Roberts, A. The checkerboard score and species distributions. Oecologia 85, 74–79 (1990)

    Article  ADS  Google Scholar 

  33. Gotelli, N. J. Null model analysis of species co-occurrence patterns. Ecology 81, 2606–2621 (2000)

    Article  Google Scholar 

  34. Ulrich, W. & Gotelli, N. J. Null model analysis of species nestedness patterns. Ecology 88, 1824–1831 (2007)

    Article  Google Scholar 

  35. Gotelli, N. J. & Ulrich, W. Statistical challenges in null model analysis. Oikos 121, 171–180 (2012)

    Article  Google Scholar 

  36. Efron, B. Bayesians, frequentists, and scientists. J. Am. Stat. Assoc. 100, 1–5 (2005)

    Article  MathSciNet  CAS  Google Scholar 

  37. Wickham, H. ggplot2: Elegant Graphics for Data Analysis (Springer, 2009)

  38. Kidwell, S. M. & Holland, S. M. The quality of the fossil record: implications for evolutionary analyses. Annu. Rev. Ecol. Syst. 33, 561–588 (2002)

    Article  Google Scholar 

  39. Behrensmeyer, A. K., Kidwell, S. M. & Gastaldo, R. A. Taphonomy and paleobiology. Paleobiology 26, 103–147 (2000)

    Article  Google Scholar 

Download references


We thank G. Dietl for comments that improved the manuscript. Support for this research was provided by a National Museum of Natural History Program grant to the Evolution of Terrestrial Ecosystems Program (ETE) and NSF-DEB 1257625. This is ETE publication 338.

Author information

Authors and Affiliations



All authors provided input into the final manuscript. S.K.L. helped design the study, contributed and analysed data, and wrote the paper. K.L.A., J.L.B., J.T.F., and W.U. were involved in the study design, and contributed and analysed data. A.T. was involved in the study design, contributed and analysed data, and created Supplementary Fig. 1. A.K.B., A.B., W.A.D., J.T.E., N.J., C.L., C.V.L., S.W., and B.R. were involved in the study design and contributed data. H.A., J.A., and D.W. contributed data. M.D. and J.H.M. were involved in the study design, analysed data, created figures and wrote text for the supplementary information. A.D. was involved in the study design and analysed data. G.R.G., B.M., D.P., S.P.-M., R.T., R.P., and A.V. were involved in the study design. N.J.G. designed the study and helped write the paper.

Corresponding author

Correspondence to S. Kathleen Lyons.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 Map showing distribution of fossil data sets.

Polygons enclose the localities for each fossil data set included in our analyses. Mammals are in blue, plants are in green. Dark colours represent data sets that are older. This map was created using ArcGIS software by Esri and can be found at ArcGIS and ArcMap are the intellectual property of Esri and are used herein under license. Copyright © Esri. All rights reserved. For more information about Esri software, please visit

Extended Data Figure 2 Breakpoint analysis of the composite data.

The analysis was performed on all data including the islands (see Fig. 3 main text), showing the mean estimate (red point; 103.778 years) and 95% confidence interval (error bar at base of plot; 101.606, 105.951 years) of the initiation of reduced percentage of aggregated species pairs, as well as the mean and confidence intervals around the change in slope of the two resulting linear models. The breakpoint analysis was run using all the data resolved to the best possible dates to allow for the greatest amount of power in detecting where the switch occurred. However, the results were similar when island data were excluded.

Extended Data Figure 3 Weighted Loess curve with and without modern data.

Loess curve weighted by number of sites with shaded 95% confidence intervals illustrates the reduction in the proportion of aggregated species pairs towards the present. Data are analysed with (black line and shading) and without (red line and shading) the modern data. Colours indicate continent: North America (green), Eurasia (purple), Australia (dark grey), South America (dark blue), Africa (orange). Point shapes indicate type of data: pollen (square), mammals (triangle), macroplants (circle). Data on terrestrial communities from ref. 2 are diamonds. Only mainland assemblages were included in the calculation for the weighted Loess curve and the density plots here and in Fig. 1.

Extended Data Figure 4 Results of PAIRS analyses of two Pleistocene–Holocene fossil data sets.

a, Mammal data for three periods: late Pleistocene (40,000–10,000 years ago), Holocene (10,000–500 years ago), and modern (present, literature survey data). Note the switch from >50% aggregated pairs to <50% aggregated pairs occurs in the Pleistocene versus Holocene data sets. b, Results for large and small mammals separately. There is a significant difference (P < 0.001) between the Holocene and the Pleistocene for all mammals (blue bars) and for large mammals (purple bars) only (P = 0.015). However, the direction of the shift was different. For all mammals, there were fewer positive associations in the Holocene, whereas, for large mammals only, there were more positive associations in the Holocene. c, North American pollen data from the past 21,000 years (modified from ref. 20). Data are from cores resolved into 1,000-year time slices. The size of the circle is related to the number of sites in the data set. The point at 0 represents a period from the present to 1,000 years ago, but is sampled from the top of the pollen cores using the same methodology as the older time slices. Note the trend of decreasing percentage of aggregations towards the present, especially in times with the largest numbers of sites (after 14,000 years). A weighted regression that takes into account the number of sites in each time slice is significant (dashed green line; P = 0.04, adjusted R2 = 0.15). The final time slice at 0 records a shift from a dominance of aggregated pairs to a dominance of segregated pairs. The sampling methods and data structure are the same for all time slices. Grey dashed line is at 50% in each panel.

Extended Data Figure 5 Relationship between the proportion of aggregated pairs and scale.

The proportion of significant pairs that are aggregated is not the result of temporal or spatial scale of data. Arcsine transformation of the proportion of significant pairs that are aggregated plotted as a function of spatial (a, b) or temporal (c, d) grain (b, d) or extent (a, c). Linear regressions are non-significant and adjusted R2 values are extremely low.

Extended Data Figure 6 Climate variability measured during the temporal extents of the fossil data sets.

Proportion of significant pairs that are aggregated shows no relationship with climate variability within a time interval. a, b, Climate variability was quantified as the standard deviation of all climate proxy data for that time interval (a), or the standard deviation of the first differences in climate across the interval (b). c, d, Climate variability (standard deviation of first differences) and age of data sets show no relationship (c), suggesting no trend in climate variability sampled by the fossil data sets across the Phanerozoic. There is a significant, but weak, positive relationship (d, dashed line) between climate variability and decreasing age of the data sets when the linear model is weighted by sample size of climate proxy data enveloped by the temporal window of the fossil data sets (P = 0.007, adjusted R2 = 0.0998).

Extended Data Figure 7 Relationship between proportion of aggregated pairs and fixed-width time intervals.

High-amplitude Pleistocene climate variability oscillating between glacial and interglacial cycles may have imposed its own novel pressures on floral and faunal communities. Furthermore, ecological impacts may lag behind climate episodes themselves, complicating efforts to quantify climatic links to changes in the proportion of aggregated species pairs over time. Thus, limiting our measure of climate variability to the temporal span of the data sets themselves may potentially not account for important (and possibly ecologically significant) climatic variability from the previous millennia. To incorporate this possibility, we re-analysed the relationship between the proportion of aggregated species pairs and climate variability of each data set, but included climate across the preceding 100,000 years, 10,000 years (not shown), and 1,000 years (not shown). As in the more restrictive analysis (Fig. 3a), there is consistently no relationship between climate variability and the proportion of aggregated species pairs.

Extended Data Table 1 Raw data for Fig. 1
Extended Data Table 2 Effect of taxonomic resolution
Extended Data Table 3 Proportion of aggregated pairs across critical intervals

Supplementary information

Supplementary Information

This file contains Supplementary Text and Data comprising: Dataset criteria, Modern data; Matrix Fill; Mammal Datasets; Plant Datasets. (PDF 13451 kb)

PowerPoint slides

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kathleen Lyons, S., Amatangelo, K., Behrensmeyer, A. et al. Holocene shifts in the assembly of plant and animal communities implicate human impacts. Nature 529, 80–83 (2016).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


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.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing