We greatly appreciate Tedersoo’s1 interest in our recent publication2 concerning temperature effects on continental scale diversity of microbial communities in forest soils2. The Correspondence raises several questions, primarily regarding the approaches used in our study. While we welcome this debate, we disagree with the claims regarding our study design, statistical analyses and interpretations.

The first concern raised in the correspondence is related to the sampling design. In contrast to most traditional microbial biogeographic studies, we sampled microbial communities at multiple spatial scales. Specifically, we focused on many replicates at a small scale, which resulted in a tradeoff for fewer samples at a large scale. This design is appropriate for addressing our research questions because soil environments are highly heterogeneous, which could mask the effects of environmental variables such as temperature on microbial distributions at continental scales. Increased sampling effort with many replicates within a site is the most effective way to mitigate the impacts of such heterogeneity on detecting patterns of microbial community diversity. We have now conducted a simulation analysis (Supplementary Table 1), which shows that large within-site sample sizes are necessary for capturing the trends between microbial diversity and temperature. Also, the experimental design with extensive sampling within a site allows us to appropriately assess the large within-site variation due to heterogeneity, which is also critical for revealing differences between microbial communities across sites. To further demonstrate this point, a similar simulation was performed using well-known published data on plant diversity and stability3. Diversity effects on stability could only be observed when the within-treatment variation was captured by more than 19 out of the 30 replicates (Supplementary Table 2). In addition, the range of temperatures (2.5–27 °C) of our six forest sites is comparable to many other continental surveys for plants, animals and phytoplankton, and should be broad enough to detect the general patterns of temperature effects on ecological communities. However, we do acknowledge that the number of sites is limited, and more sites within and/or across continents should be examined to evaluate whether our results are applicable to other forest ecosystems. Thus, we pointed out this potential caveat in the original paper2.

The correspondence1 also questions our regression analyses. In the original version of the study, we used univariate linear regression, multiple linear regression, nonlinear univariate regression and more complex models (original Supplementary Tables 7–9)2. We have now also conducted partial correlation analysis (Supplementary Table 3). All of these analyses indicated that temperature is generally a better predictor of microbial diversity than other environmental variables. We also performed univariate and multiple hierarchical linear model (HLM) analyses (Supplementary Tables 4 and 5), where site was used as grouping factor to account for our nested sampling design, and the degrees of freedom for the variable(s) and intercept were 4 and 120, respectively. Since heteroscedasticity and non-normality of residuals were found in many cases of our HLM analysis, variance weights were introduced to allow different variances for various sites, and the significance test was based on bootstrapping rather than parametric tests. The HLM analyses showed the same trends as univariate and multiple linear regression (original Table 9 and Supplementary Table 6, ref. 2), which further supports our original conclusion.

The third point is related to the use of Chao1 estimation. In this study, we did consider sequencing artefacts by removing singletons from raw sequencing data before resampling, which could improve sequence quality and thus subsequent analysis4. Also, our conclusions were supported by various diversity indexes showed the same pattern as Chao1, including Shannon, Inverse Simpson, Faith’s phylogenetic diversity and the net relatedness index (original Supplementary Tables 6 and 7, ref. 2), as well as the Hill number5 (Supplementary Tables 6 and 7). Therefore, we do not believe the Chao1 estimation introduced significant bias in our conclusions.

Another critique is that this study did not cite all of the relevant literature. This was a result of space limitations, although we note that we were aware of all of this literature before publication of our original paper. Importantly, none of this literature provides explicit evidence that temperature is more important than other proposed environmental drivers in shaping microbial diversity, particularly in forest soils. Thus, this literature does not invalidate our original statement that ‘This is the first demonstration that temperature plays a more primary role in shaping variation in microbial diversity in the forest soils than other proposed environmental drivers.’ (ref. 2).

In conclusion, we believe that (i) the approaches used in our study were appropriate for addressing our research questions of interest, (ii) the experimental results were cautiously interpreted and (iii) the conclusions were sound and solid.

Additional information

How to cite this article: Zhou, J. et al. Reply to ‘Analytical flaws in a continental-scale forest soil microbial diversity study’. Nat. Commun. 8, 15583 doi: 10.1038/ncomms15583 (2017).

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    Correspondence: Analytical flaws in a continental-scale forest soil microbial diversity study. Nat. Commun. 8, 15572 (2017).

  2. 2.

    et al. Temperature mediates continental-scale diversity of microbes in forest soils. Nat. Commun. 7, 12083 (2016).

  3. 3.

    , & Biodiversity and ecosystem stability in a decade-long grassland experiment. Nature 441, 629–632 (2006).

  4. 4.

    et al. Reproducibility and quantitation of amplicon sequencing-based detection. ISME J. 5, 1303–1313 (2011).

  5. 5.

    , & Unifying species diversity, phylogenetic diversity, functional diversity, and related similarity and differentiation measures through Hill numbers. Annu. Rev. Ecol. Evol. Syst. 45, 297–324 (2014).

Download references

Author information

Affiliations

  1. State Key Joint Laboratory of Environment Simulation and Pollution Control, School of Environment, Tsinghua University, Beijing 100084, China

    • Jizhong Zhou
    •  & Yunfeng Yang
  2. Institute for Environmental Genomics, Department of Microbiology and Plant Biology and School of Civil Engineering and Environmental Sciences, University of Oklahoma, Norman, Oklahoma 73019, USA

    • Jizhong Zhou
    • , Ye Deng
    • , Lina Shen
    • , Chongqing Wen
    • , Qingyun Yan
    • , Daliang Ning
    • , Yujia Qin
    • , Kai Xue
    • , Liyou Wu
    • , Zhili He
    • , James W. Voordeckers
    •  & Joy D. Van Nostrand
  3. Earth Science Division, Lawrence Berkeley National Laboratory, Berkeley, California 94270, USA

    • Jizhong Zhou
  4. CAS Key Laboratory for Environmental Biotechnology, Research Center for Eco-Environmental Science, Chinese Academy of Sciences, Beijing 100085, China

    • Ye Deng
  5. Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, Arizona 85721, USA

    • Vanessa Buzzard
    • , Sean T. Michaletz
    •  & Brian J. Enquist
  6. The Santa Fe Institute, 1399 Hyde Park Rd, Santa Fe, New Mexico 87501, USA

    • Brian J. Enquist
  7. EEB Graduate Program, Department of Biology, University of Oklahoma, Norman, Oklahoma 73019, USA

    • Michael D. Weiser
    •  & Michael Kaspari
  8. Smithsonian Tropical Research Institute, Balboa 0843-03092, Republic of Panama

    • Michael Kaspari
  9. Department of Biology, University of New Mexico, Albuquerque, New Mexico 87131, USA

    • Robert Waide
    •  & James H. Brown

Authors

  1. Search for Jizhong Zhou in:

  2. Search for Ye Deng in:

  3. Search for Lina Shen in:

  4. Search for Chongqing Wen in:

  5. Search for Qingyun Yan in:

  6. Search for Daliang Ning in:

  7. Search for Yujia Qin in:

  8. Search for Kai Xue in:

  9. Search for Liyou Wu in:

  10. Search for Zhili He in:

  11. Search for James W. Voordeckers in:

  12. Search for Joy D. Van Nostrand in:

  13. Search for Vanessa Buzzard in:

  14. Search for Sean T. Michaletz in:

  15. Search for Brian J. Enquist in:

  16. Search for Michael D. Weiser in:

  17. Search for Michael Kaspari in:

  18. Search for Robert Waide in:

  19. Search for Yunfeng Yang in:

  20. Search for James H. Brown in:

Contributions

All authors contributed intellectual input and assistance to this reply. J.Z., D.N., Y.D. and S.T.M. developed ideas for additional data analysis. D.N. and Y.D. performed the data analysis. J.Z. and D.N. wrote this reply with help from Y.D., S.T.M., B.J.E., M.K. and R.W. All the authors read and approved this reply.

Competing interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to Jizhong Zhou or Ye Deng.

About this article

Publication history

Published

DOI

https://doi.org/10.1038/ncomms15583

Comments

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.