Skip to main content

Thank you for visiting nature.com. 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:

Mycorrhiza-mediated competition between plants and decomposers drives soil carbon storage

Nature volume 505, pages 543–545 (2014)Cite this article

Subjects

Abstract

Soil contains more carbon than the atmosphere and vegetation combined1. Understanding the mechanisms controlling the accumulation and stability of soil carbon is critical to predicting the Earth’s future climate2,3. Recent studies suggest that decomposition of soil organic matter is often limited by nitrogen availability to microbes4,5,6 and that plants, via their fungal symbionts, compete directly with free-living decomposers for nitrogen6,7. Ectomycorrhizal and ericoid mycorrhizal (EEM) fungi produce nitrogen-degrading enzymes, allowing them greater access to organic nitrogen sources than arbuscular mycorrhizal (AM) fungi8,9,10. This leads to the theoretical prediction that soil carbon storage is greater in ecosystems dominated by EEM fungi than in those dominated by AM fungi11. Using global data sets, we show that soil in ecosystems dominated by EEM-associated plants contains 70% more carbon per unit nitrogen than soil in ecosystems dominated by AM-associated plants. The effect of mycorrhizal type on soil carbon is independent of, and of far larger consequence than, the effects of net primary production, temperature, precipitation and soil clay content. Hence the effect of mycorrhizal type on soil carbon content holds at the global scale. This finding links the functional traits of mycorrhizal fungi to carbon storage at ecosystem-to-global scales, suggesting that plant–decomposer competition for nutrients exerts a fundamental control over the terrestrial carbon cycle.

Access through your institution
Buy or subscribe

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

Access options

Access through your institution

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

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

Figure 1: The relationship between soil carbon and nitrogen content to a depth of one metre in AM and EEM ecosystems.
Figure 2: The relationships between soil carbon content to a depth of one metre and MAT, MAP, clay content and NPP.

Similar content being viewed by others

References

  1. Tarnocai, C. et al. Soil organic carbon pools in the northern circumpolar permafrost region. Glob. Biogeochem. Cycles 23, GB2023 (2009)

    Article  ADS  Google Scholar 

  2. Jenkinson, D. S., Adams, D. E. & Wild, A. Model estimates of CO2 emissions from soil in response to global warming. Nature 351, 304–306 (1991)

    Article  ADS  CAS  Google Scholar 

  3. Knorr, W., Prentice, I. C., House, J. I. & Holland, E. A. Long-term sensitivity of soil carbon turnover to warming. Nature 433, 298–301 (2005)

    Article  ADS  CAS  Google Scholar 

  4. Schimel, J. P. & Weintraub, M. N. The implications of exoenzyme activity on microbial carbon and nitrogen limitation in soil: a theoretical model. Soil Biol. Biochem. 35, 549–563 (2003)

    Article  CAS  Google Scholar 

  5. Allison, S. D., Gartner, T. B., Mack, M. C., McGuire, K. & Treseder, K. Nitrogen alters carbon dynamics during early succession in boreal forest. Soil Biol. Biochem. 42, 1157–1164 (2010)

    Article  CAS  Google Scholar 

  6. Lindahl, B. D., de Boer, W. & Finlay, R. D. Disruption of root carbon transport into forest humus stimulates fungal opportunists at the expense of mycorrhizal fungi. ISME J. 4, 872–881 (2010)

    Article  Google Scholar 

  7. Gadgil, R. & Gadgil, P. Mycorrhiza and litter decomposition. Science 233, 133 (1971)

    CAS  Google Scholar 

  8. Näsholm, T. et al. Boreal forest plants take up organic nitrogen. Nature 392, 914–916 (1998)

    Article  ADS  Google Scholar 

  9. Hodge, A., Campbell, C. D. & Fitter, A. H. An arbuscular mycorrhizal fungus accelerates decomposition and acquires nitrogen directly from organic material. Nature 413, 297–299 (2001)

    Article  ADS  CAS  Google Scholar 

  10. Read, D. J. & Perez-Moreno, J. Mycorrhizas and nutrient cycling in ecosystems—a journey towards relevance? New Phytol. 157, 475–492 (2003)

    Article  Google Scholar 

  11. Orwin, K. H., Kirschbaum, M. U. F., St John, M. G. & Dickie, I. A. Organic nutrient uptake by mycorrhizal fungi enhances ecosystem carbon storage: a model-based assessment. Ecol. Lett. 14, 493–502 (2011)

    Article  Google Scholar 

  12. LeBauer, D. S. & Treseder, K. K. Nitrogen limitation of net primary productivity in terrestrial ecosystems is globally distributed. Ecology 89, 371–379 (2008)

    Article  Google Scholar 

  13. Averill, C. & Finzi, A. Increasing plant use of organic nitrogen with elevation is reflected in nitrogen uptake rates and ecosystem δ15N. Ecology 92, 883–891 (2011)

    Article  Google Scholar 

  14. Lindahl, B. D., Finlay, R. D. & Cairney, J. W. G. Enzymatic activities of mycelia in mycorrhizal fungal communities. The Fungal Community: its Organization and Role in the Ecosystem 3rd edn, 331–348 (CRC Press, 2005)

    Google Scholar 

  15. Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. E. & Jarvis, A. Very high resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25, 1965–1978 (2005)

    Article  Google Scholar 

  16. Zhao, M. & Running, S. W. Drought-induced reduction in global terrestrial net primary production from 2000 through 2009. Science 329, 940–943 (2010)

    Article  ADS  CAS  Google Scholar 

  17. Randerson, J. T. et al. Systematic assessment of terrestrial biogeochemistry in coupled climate-carbon models. Glob. Change Biol. 15, 2462–2484 (2009)

    Article  ADS  Google Scholar 

  18. Torn, M. S., Trumbore, S. E., Chadwick, O. A., Vitousek, P. M. & Hendricks, D. M. Mineral control of soil organic carbon storage and turnover. Nature 389, 170–173 (1997)

    Article  ADS  CAS  Google Scholar 

  19. Cebrian, J. & Duarte, C. M. Plant growth-rate dependence of detrital carbon storage in ecosystems. Science 268, 1606–1608 (1995)

    Article  ADS  CAS  Google Scholar 

  20. Davidson, E. A. & Janssens, I. A. Temperature sensitivity of soil carbon decomposition and feedbacks to climate change. Nature 440, 165–173 (2006)

    Article  ADS  CAS  Google Scholar 

  21. Gallet-Budynek, A. et al. Intact amino acid uptake by northern hardwood and conifer trees. Oecologia 160, 129–138 (2009)

    Article  ADS  Google Scholar 

  22. Cheng, L. et al. Arbuscular mycorrhizal fungi increase organic carbon decomposition under elevated CO2 . Science 337, 1084–1087 (2012)

    Article  ADS  CAS  Google Scholar 

  23. Rygiewicz, P. T. & Andersen, C. P. Mycorrhizae alter quality and quantity of carbon allocated below ground. Nature 369, 58–60 (1994)

    Article  ADS  Google Scholar 

  24. van der Heijden, M. G. A. et al. Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396, 69–72 (1998)

    Article  ADS  CAS  Google Scholar 

  25. Talbot, J. M. & Finzi, A. C. Differential effects of sugar maple, red oak, and hemlock tannins on carbon and nitrogen cycling in temperate forest soils. Oecologia 155, 583–592 (2008)

    Article  ADS  Google Scholar 

  26. Schmidt, M. W. I. et al. Persistence of soil organic matter as an ecosystem property. Nature 478, 49–56 (2011)

    Article  ADS  CAS  Google Scholar 

  27. Cotrufo, M. F., Wallenstein, M. D., Boot, C. M., Denef, K. & Paul, E. The Microbial Efficiency-Matrix Stabilization (MEMS) framework integrates plant litter decomposition with soil organic matter stabilization: do labile plant inputs form stable soil organic matter? Glob. Change Biol. 19, 988–995 (2013)

    Article  ADS  Google Scholar 

  28. Frey, S. D., Lee, J., Melillo, J. M. & Six, J. The temperature response of soil microbial efficiency and its feedback to climate. Nature Clim. Change 3, 395–398 (2013)

    Article  ADS  CAS  Google Scholar 

  29. Pinheiro, J., Bates, D., DebRoy, S. & Sarkar, D. & the R Development Core Team nlme: Linear and Nonlinear Mixed Effects Models R package version 3. 1–109, http://cran.r-project.org/web/packages/nlme/nlme.pdf (2013)

  30. Mayes, M. A., Heal, K. R., Brandt, C. C., Phillips, J. R. & Hardine, P. M. Relation between soil order and sorption of dissolved organic carbon in temperate subsoils. Soil Sci. Soc. Am. J. 76, 1027–1037 (2012)

    Article  ADS  CAS  Google Scholar 

  31. Grossman, R. B. & Reinsch, T. G. 2002. in Methods of Soil Analysis, Part 4: Physical Methods (eds Dane, J. H. & Topp, C. ) 201–203 (Soil Society of America, 2002)

    Google Scholar 

  32. Gee, G. W. & Or, D. 2002. in Methods of Soil Analysis, Part 4: Physical Methods (eds Dane, J. H. & Topp, C. ) 255–293 (Soil Society of America, 2002)

    Google Scholar 

  33. Whittaker, R. H. Communities and Ecosystems 2nd edn, 111–191 (Macmillan, 1975)

    Google Scholar 

  34. Mazerolle, M. J. AICcmodavg: Model selection and Multimodel Inference based on (Q)AIC(c) R package version 1.31,. http://cran.r-project.org/web/packages/AICcmodavg/index.html (2013)

  35. Tofallis, C. Least squares percentage regression. J. Mod. Appl. Stat. Methods 7, 526–534 (2008)

    Article  Google Scholar 

  36. Magee, L. R2 measures based on wald and likelihood ratio joint significance tests. Am. Stat. 44, 250–253 (1990)

    Google Scholar 

  37. Aleksandra, M. lmmfit: Goodness-of-Fit-Measures for Linear Mixed Models with One-Level-Grouping R package version 1.0,. http://cran.r-project.org/web/packages/lmmfit/index.html (2011)

  38. Fox, J. & Weisberg, H. S. An {R} Companion to Applied Regression 2nd edn (Sage Publications, 2011)

    Google Scholar 

  39. Zuur, A. F., Ieno, E. N. & Elphick, C. S. A protocol for data exploration to avoid common statistical problems. Methods Ecol. Evol. 1, 3–14 (2010)

    Article  Google Scholar 

  40. Yee, T. W. The VGAM package for categorical data analysis. J. Stat. Softw. 32, 1–34 (2010)

    Article  Google Scholar 

Download references

Acknowledgements

We thank L. Nave and the International Soil Carbon Network for access to their database. C. Hawkes provided feedback during data collection and initial analyses of C storage. C. Iversen, J. Powers and M. Vadeboncouer provided unpublished data that contributed to this analysis. D. Jacquier provided the Australian soil database and E. Carlston helped to extract data from the Australian soil database. C. Shaw provided the Siltanen soil carbon database and the Forest Ecosystem Carbon Database of Canadian soils. T. Baisden provided scans of the California Soil-Vegetation Survey. E. Brzostek, N. Fowler, P. Groffman, E. Hobbie, B. Schlesinger and B. Waring provided feedback on earlier versions of this manuscript. The Center for Tropical Forest Science (CTFS) and Smithsonian Institution Geo-observatories (SIGEO) provided funding for the collection and analysis of soil profile data at large forest dynamics plots, and we thank the many collaborators, field assistants and laboratory technicians who assisted in the collection and analysis of soil profile data. This work benefited from extensive data contributions to the International Soil Carbon Network from both the USDA Natural Resources Conservation Service, National Cooperative Soil Survey, and the US Geological Survey. C.A. was supported by a fellowship from the University of Texas at Austin and by the National Science Foundation Graduate Research Fellowship Program (grant DGE-1110007). A.C.F. was supported by NSF grant number DEB 07-43564 and DOE grants 10-DOE-1053 and DE-SC0006916. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Author information

Authors and Affiliations

  1. Department of Integrative Biology, Graduate Program in Ecology, Evolution and Behavior, University of Texas at Austin, Austin, 78712, Texas, USA

    Colin Averill

  2. Smithsonian Tropical Research Institute, Apartado 0843-03092, Balboa, Ancon, Republic of Panama,

    Benjamin L. Turner

  3. Department of Biology, Boston University, Boston, 02215, Masachusetts, USA

    Adrien C. Finzi

Authors
  1. Colin Averill
    View author publications

    Search author on:PubMed Google Scholar

  2. Benjamin L. Turner
    View author publications

    Search author on:PubMed Google Scholar

  3. Adrien C. Finzi
    View author publications

    Search author on:PubMed Google Scholar

Contributions

C.A. and B.L.T. collected the data. C.A. performed all statistical analyses. C.A. and A.C.F. conceptualized the work and wrote the manuscript. All authors contributed to revisions.

Corresponding author

Correspondence to Colin Averill.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 Soil C versus N in the first 50 cm of mineral soil.

Purple symbols are EEM observations and black symbols are AM observations. Plotted lines represent univariate regression lines of the respective subsets of the data. We note that plotted lines are univariate regressions of data subsets and are included for visualization purposes only. Removal of the surface organic horizon did not qualitatively change the interpretation of the data. Both the full model and the best AICc-selected model had a significant interactive effect between mycorrhizal type and soil N on soil C storage, with EEM systems storing 1.6 times more C per unit N than AM systems (P < 0.0001).

Extended Data Figure 2 Soil C versus N excluding boreal observations.

Purple symbols are EEM observations and black symbols are AM observations. Plotted lines represent univariate regression lines of the respective subsets of the data. We note that plotted lines are univariate regressions of data subsets and are included for visualization purposes only. Both the full model and the best AICc-selected model showed a significant interactive effect of mycorrhizal type and soil N on soil C storage, with EEM systems storing 1.6 times more C per unit N than AM systems (P = 0.0014).

Extended Data Figure 3 Soil C versus N limiting data set to observations with less than 3.5 kg N m−2.

Purple symbols are EEM observations and black symbols are AM observations. Plotted lines represent univariate regression lines of the respective subsets of the data. We note that plotted lines are univariate regressions of data subsets and are included for visualization purposes only. Both the full model and the best AICc-selected model found a significant interactive effect of mycorrhizal type and soil N on soil C storage, with EEM systems storing 1.4 times more C per unit N than AM systems (P = 0.0304).

Extended Data Figure 4 Soil C versus N excluding grassland observations.

Purple symbols are EEM observations and black symbols are AM observations. Plotted lines represent univariate regression lines of the respective subsets of the data. We note that plotted lines are univariate regressions of data subsets and are included for visualization purposes only. Both the full model and the best AICc-selected model found a significant interactive effect of mycorrhizal type and soil N on soil C storage, with EEM systems storing 1.5 times more C per unit N than AM systems (P = 0.0023).

Extended Data Figure 5 Soil C versus N restricting the analysis to temperate and tropical forest observations only.

Purple symbols are EEM and black symbols are AM observations. Plotted lines represent univariate regression lines of the respective subsets of the data. We note that plotted lines are univariate regressions of data subsets and are included for visualization purposes only. Both the full model and the best AICc-selected model incorporated the interactive effect of mycorrhizal type and soil N on soil C storage, with EEM systems storing 1.3 times more C per unit N than AM systems, although the effect was marginally not significant (P = 0.0690). We re-emphasize that the full model incorporates biome type, and weights observations by the inverse of their C values, to prevent undue influence of large observations on the estimated effect size.

Extended Data Table 1 Mineral soil (0–50 cm) analysis regression output from the best AICc-selected model
Full size table
Extended Data Table 2 Removing boreal forests analysis from the best AICc-selected model
Full size table
Extended Data Table 3 Restricting range of N content analysis from the best AICc-selected model
Full size table
Extended Data Table 4 Removing grasslands analysis from the best AICc-selected model
Full size table
Extended Data Table 5 Temperate and tropical biomes only, from the best AICc-selected model
Full size table

PowerPoint slides

Source data

Rights and permissions

Reprints and permissions

About this article

Cite this article

Averill, C., Turner, B. & Finzi, A. Mycorrhiza-mediated competition between plants and decomposers drives soil carbon storage. Nature 505, 543–545 (2014). https://doi.org/10.1038/nature12901

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue date:

  • DOI: https://doi.org/10.1038/nature12901

This article is cited by

Comments

Commenting on this article is now closed.

  1. Web Admin

    This comment is posted on behalf of Dr Joseph Craine:

    Averill et al. (ref 1) analyze a global dataset of soil C and N contents of soils and conclude that soils under ectomycorrhizal and ericoid mycorrhizal (EEM) plants contain 70% more carbon per unit nitrogen than soils under arbuscular mycorrhizal (AM) plants. Moreover, they report little influence of climate or soil texture on C and N storage. A summary of the research in the same issue2 raised questions about non-linearities in relationships and potentially spurious relationships, but ultimately reinforces that the type of mycorrhizal fungus influences global patterns of C storage in soils. In light of these reports, I have reanalyzed their data and show there is no significant effect of mycorrhizal association on soil C storage.

    In analyzing the data on soil C and N, first, only temperate and tropical forest biomes should be used in the analysis. There were no examples of soils from boreal forests under plants that used arbuscular mycorrhizal fungi, nor grasslands that used ectomycorrhizal or ericoid mycorrhizal fungi. As such, they would provide no evidence on the changes in soil C and N when mycorrhizal types were to change within these biomes.

    Second, biome identity should not be included as a categorical factor along with climate in a regression model. The variance inflation factor threshold of 10 the authors used is an arbitrary guideline. The variance inflation factors for climate parameters and biome in regression models are > 9, which signifies a poor ability of regression models to separate the predictors. Significant differences in soil C and N among biomes were likely due to the effects of climate.

    Third, as there is no dependent-independent relationship between soil C and N levels and similar levels of error in measuring them, soil N should not be used to predict soil C in a regression model. This was a central part of their analyses.

    Fourth, both soil C and N are log-normally distributed and non-linearly related. Not log-transforming C and N data over-emphasizes the importance in regressions of a relatively few high C and N soils. Both soil C and N should be log-transformed prior to analyses.

    In all, a better way to test whether mycorrhizal type affects the amount of C and N stored in soils is to first generate relationships between log-transformed soil elemental concentrations and both climate and clay concentrations for each mycorrhizal type for just the temperate and tropical forests, which both have arbuscular and ectomycorrhizal plants to be compared. Then, soil C:N can be compared at a common climate and clay concentration to examine if different mycorrhizal types are associated with different degrees of C storage.

    Using predictive relationships from regressions of log-transformed C and N (Table 1), soil C and N contents are responsive to mean climate parameters and increase with greater amounts of clay. Typically, hot, dry sites would have less soil C and N than cold, wet sites. Given the relationships between climate and soil C and N, C:N ratios were then predicted for mean climate (MAT = 16.8 �C, MAP = 1774 mm) and clay concentrations (31.3% clay) for each mycorrhizal type. Predicted C:N ratio for AM forests was 12.9 � 1.4. Predicted C:N ratio for EEM forests was 14.8 � 2.9. Although the trend is for EEM forests to have 15% more C per unit N, this is much less than the 70% stated by the authors, and not statistically significant.

    1 Averill, C., Turner, B. L. & Finzi, A. C. Mycorrhiza-mediated competition between plants and decomposers drives soil carbon storage. Nature 505, 543-545, (2014).

    2 Bradford, M. A. Ecology: Good dirt with good friends. Nature 505, 486-487, (2014).

    3 Craine, J. M., Morrow, C. & Fierer, N. Microbial nitrogen limitation increases decomposition. Ecology 88, 2105-2113, (2007).

    4 Nemergut, D. R. et al. The effects of chronic nitrogen fertilization on alpine tundra soil microbial communities: implications for carbon and nitrogen cycling. Environ Microbiol 10, 3093-3105, (2008).

    5 Hagedorn, F., Spinnler, D. & Siegwolf, R. Increased N deposition retards mineralization of old soil organic matter. Soil Biol. Biochem. 35, 1683-1692, (2003).

  2. Web Admin

    This comment is posted on behalf of Colin Averill:

    Mycorrhizal effects on soil carbon maintained

    Colin Averill (1), Benjamin L. Turner (2) and Adriend C. Finzi (3)

    (1)Department of Integrative Biology, Graduate Program in Ecology, Evolution and Behavior, University of Texas at Austin, Austin, TX 78712, USA
    (2)Smithsonian Tropical Research Institute, Apartado 0843-03092, Balboa, Ancon, Republic of Panama
    (3)Department of Biology, Boston University, Boston, MA 02215, USA

    Craine has reanalyzed our data and finds no significant effect of mycorrhizal association on soil carbon (C) storage. However, his analysis cannot test for an interaction between nitrogen (N) and mycorrhizal status, the central aim of our study (Ref 1). We conclude that our original analysis is robust and that mycorrhizal association influences soil C storage.

    Craine argues that boreal and grassland sites should be excluded from our analysis, because they contain only a single type of mycorrhizal association. He then argues that since biome and climate are confounded, biome should be excluded from the analysis. These two options are mutually exclusive; either boreal and grassland sites are excluded and biome is retained, or biome is excluded and grasslands and boreal forests are retained. Reanalyzing our data set without biome maintains a significant mycorrhizal effect; MAT and MAP are retained in the final model based on AIC selection criteria but are not significant, while NPP and clay effects are statistically significant but explain only a small fraction of the variation. As biome is not included in this analysis, Craine's discussion of variance inflation factors is no longer relevant.

    Craine states that there is no basis for a dependent-independent relationship between soil C and N content, yet there is overwhelming experimental evidence that changes in soil N affect soil C (Refs 2,3,4). In our original article we provided theoretical and empirical evidence for a C-dependent relationship with N, including the potential for mycorrhizal control. An argument against the dependent-independent relationship between C and N must be made on theoretical grounds, but Craine provides no such argument.

    Craine claims that soil C and N are log-normally distributed and non-linearly related, yet Figure 1 in our original article shows a linear relationship between C and N. Sample variance commonly increases as a function of the sample mean (i.e., increasing variability along the regression lines in Figure 1), but this does not mean that the data are log-normally related. The log transformation is by definition variance-stabilizing (Ref 5), and is applied as a patch to fit models using ordinary least squares (OLS); it does not imply an underlying causal relationship between variables. The limitations and assumptions imposed by OLS are obviated in our original analysis by the use of least-squares percentage regression (Ref 6); the residuals of the model we employed have homogeneous variance, are normally distributed, and do not require the assumption of a log-normal distribution. This approach was validated with extensive Monte Carlo simulation, demonstrating that the analysis is not sensitive to a few soils with high organic matter content.

    Finally, and most importantly, Craine's re-analysis tests only for a main effect of mycorrhizal type and soil C storage. It cannot test for an interaction between mycorrhizal type and soil N, the central aim of our approach. This means that Craine tests for a fundamentally different relationship between soil C and N, which explains why his estimate depends on the common climate selected.

    We conclude that our original approach was robust and appropriate to test the proposed interaction between mycorrhizal status and soil N. Although there is no doubt that climate influences soil carbon, our article highlights the under-appreciated significance of biological processes and mycorrhizal association (Ref 1).

    References

    1. Averill, C., Turner, B.L. & Finzi, A.C. Mycorrhiza-mediate competition between plants and decomposers drives soil carbon storage. Nature 505, 543-545.
    2. Mack, M.C., Schuur, E.A.G., Bret-Harte, M.S., Shaver, G.R. & Chapin, F.S. Ecosystem carbon storage in arctic tundra reduced by long-term nutrient fertilization. Nature 431, 440-3 (2004).
    3. Allison, S.D., Gartner, T.B., Mack, M.C., McGuire, K. & Treseder K. Nitrogen alters carbon dynamics during early succession in boreal forest. Soil Biol. Biochem. 42, 1157-1164 (2010).
    4. Janssens, I.A. et al. Reduction of forest soil respiration in response to nitrogen deposition. Nature Geoscience 3, 315-322 (2010).
    5. Everitt, B.S. The Cambridge Dictionary of Statistics (4th Edition). Cambridge University Press (2010).
    6. Tofallis C. Least squares percentage regression. J. Mod. Appl. Stat. Methods, 7, 526-534, (2008).

Search

Advanced search

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