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.

Sensitivity of global terrestrial ecosystems to climate variability

Nature volume 531, pages 229–232 (2016)Cite this article

Subjects

Abstract

The identification of properties that contribute to the persistence and resilience of ecosystems despite climate change constitutes a research priority of global relevance1. Here we present a novel, empirical approach to assess the relative sensitivity of ecosystems to climate variability, one property of resilience that builds on theoretical modelling work recognizing that systems closer to critical thresholds respond more sensitively to external perturbations2. We develop a new metric, the vegetation sensitivity index, that identifies areas sensitive to climate variability over the past 14 years. The metric uses time series data derived from the moderate-resolution imaging spectroradiometer (MODIS) enhanced vegetation index3, and three climatic variables that drive vegetation productivity4 (air temperature, water availability and cloud cover). Underlying the analysis is an autoregressive modelling approach used to identify climate drivers of vegetation productivity on monthly timescales, in addition to regions with memory effects and reduced response rates to external forcing5. We find ecologically sensitive regions with amplified responses to climate variability in the Arctic tundra, parts of the boreal forest belt, the tropical rainforest, alpine regions worldwide, steppe and prairie regions of central Asia and North and South America, the Caatinga deciduous forest in eastern South America, and eastern areas of Australia. Our study provides a quantitative methodology for assessing the relative response rate of ecosystems—be they natural or with a strong anthropogenic signature—to environmental variability, which is the first step towards addressing why some regions appear to be more sensitive than others, and what impact this has on the resilience of ecosystem service provision and human well-being.

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

Relevant articles

Open Access articles citing this article.

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

Learn more

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

Figure 1: Vegetation sensitivity index.
Figure 2: RGB composite of vegetation sensitivity index.

References

  1. Convention on Biological Diversity, Aichi Biodiversity Targets. http://www.cbd.int/sp/targets/default.shtml

  2. Scheffer, M. et al. Early-warning signals for critical transitions. Nature 461, 53–59 (2009)

    Article  CAS  ADS  Google Scholar 

  3. Solano, R., Didan, K., Jacobson, A. & Huete, A. MODIS vegetation index C5 user’s guide (MOD13 Series). 1–42 http://vip.arizona.edu/documents/MODIS/MODIS_VI_UsersGuide_01_2012.pdf (2010)

  4. Nemani, R. R. et al. Climate-driven increases in global terrestrial net primary production from 1982 to 1999. Science 300, 1560–1563 (2003)

    Article  CAS  ADS  Google Scholar 

  5. De Keersmaecker, W. et al. A model quantifying global vegetation resistance and resilience to short-term climate anomalies and their relationship with vegetation cover. Glob. Ecol. Biogeogr. 24, 539–548 (2015)

    Article  Google Scholar 

  6. Garcia, R. A., Cabeza, M., Rahbek, C. & Araujo, M. B. Multiple dimensions of climate change and their implications for biodiversity. Science 344, 1247579 (2014)

    Article  Google Scholar 

  7. Thomas, C. D. et al. Extinction risk from climate change. Nature 427, 145–148 (2004)

    Article  CAS  ADS  Google Scholar 

  8. Kharin, V. V., Zwiers, F. W., Zhang, X. & Hegerl, G. C. Changes in temperature and precipitation extremes in the IPCC ensemble of global coupled model simulations. J. Clim. 20, 1419–1444 (2007)

    Article  ADS  Google Scholar 

  9. Holmgren, M., Hirota, M., Van Nes, E. H. & Scheffer, M. Effects of interannual climate variability on tropical tree cover. Nature Clim. Change 3, 755–758 (2013)

    Article  ADS  Google Scholar 

  10. Pederson, N. et al. The legacy of episodic climatic events in shaping temperate, broadleaf forests. Ecol. Monogr. 84, 599–620 (2014)

    Article  Google Scholar 

  11. Doughty, C. E. et al. Drought impact on forest carbon dynamics and fluxes in Amazonia. Nature 519, 78–82 (2015)

    Article  CAS  ADS  Google Scholar 

  12. Holling, C. S. Resilience and stability of ecological systems. Annu. Rev. Ecol. Evol. Syst. 4, 1–23 (1973)

    Article  Google Scholar 

  13. Dakos, V. et al. Slowing down as an early warning signal for abrupt climate change. Proc. Natl Acad. Sci. USA 105, 14308–14312 (2008)

    Article  CAS  ADS  Google Scholar 

  14. Kerr, J. T. & Ostrovsky, M. From space to species: ecological applications for remote sensing. Trends Ecol. Evol. 18, 299–305 (2003)

    Article  Google Scholar 

  15. Seeman, S. W., Borbas, E. E., Li, J., Menzel, W. P. & Gumley, L. E. MODIS atmospheric profile retrieval algorithm theoretical basis document, version 6. http://modis-atmos.gsfc.nasa.gov/_docs/MOD07MYD07ATBDC005.pdf (2006).

  16. Mu, Q., Zhao, M. & Running, S. W. Improvements to a MODIS global terrestrial evapotranspiration algorithm. Remote Sens. Environ. 115, 1781–1800 (2011)

    Article  ADS  Google Scholar 

  17. Ackerman, S. et al. Discriminating clear-sky from cloud with MODIS: algorithm theoretical basis document (MOD35), version 6.1. http://modisatmos.gsfc.nasa.gov/_docs/MOD35_ATBD_Collection6.pdf (2010).

  18. Sala, O. E., Gherardi, L. A., Reichmann, L., Jobbgy, E. & Peters, D. Legacies of precipitation fluctuations on primary production: theory and data synthesis. Phil. Trans. R. Soc. Lond. B 367, 3135–3144 (2012)

    Article  Google Scholar 

  19. Richard, Y. & Poccard, I. A statistical study of NDVI sensitivity to seasonal and interannual rainfall variations in southern Africa. Int. J. Remote Sens. 19, 2907–2920 (1998)

    Article  Google Scholar 

  20. Intergovernmental Panel on Climate Change. Climate Change 2013: The Physical Science Basis. (Cambridge Univ. Press, 2013)

  21. Macias-Fauria, M., Forbes, B. C., Zetterberg, P. & Kumpula, T. Eurasian Arctic greening reveals teleconnections and the potential for structurally novel ecosystems. Nature Clim. Change 2, 613–618 (2012)

    Article  ADS  Google Scholar 

  22. Myers-Smith, I. H. et al. Shrub expansion in tundra ecosystems: dynamics, impacts and research priorities. Environ. Res. Lett. 6, 045509 (2011)

    Article  ADS  Google Scholar 

  23. Clark, D. A., Piper, S. C., Keeling, C. D. & Clark, D. B. Tropical rain forest tree growth and atmospheric carbon dynamics linked to interannual temperature variation during 1984 –2000. Proc. Natl Acad. Sci. USA 100, 5852–5857 (2003)

    Article  CAS  ADS  Google Scholar 

  24. Doughty, C. E. & Goulden, M. L. Are tropical forests near a high temperature threshold? J. Geophys. Res. 113, G00B07 (2008)

    ADS  Google Scholar 

  25. Williams, J. W., Jackson, S. T. & Kutzbach, J. E. Projected distributions of novel and disappearing climates by 2100 AD. Proc. Natl Acad. Sci. USA 104, 5738–5742 (2007)

    Article  CAS  ADS  Google Scholar 

  26. Lenton, T. M. et al. Tipping elements in the Earth’s climate system. Proc. Natl Acad. Sci. USA 105, 1786–1793 (2008)

    Article  CAS  ADS  Google Scholar 

  27. Barbosa, H. A., Huete, A. R. & Baethgen, W. E. A 20-year study of NDVI variability over the northeast region of Brazil. J. Arid Environ. 67, 288–307 (2006)

    Article  ADS  Google Scholar 

  28. Harris, A., Carr, A. S. & Dash, J. Remote sensing of vegetation cover dynamics and resilience across southern Africa. Int. J. Appl. Earth Obs. Geoinf. 28, 131–139 (2014)

    Article  ADS  Google Scholar 

  29. Hirota, M., Holmgren, M., Van Nes, E. H. & Scheffer, M. Global resilience of tropical forest and savanna to critical transitions. Science 334, 232–235 (2011)

    Article  CAS  ADS  Google Scholar 

  30. Lehner, B. & Döll, P. Development and validation of a global database of lakes, reservoirs and wetlands. J. Hydrol. (Amst.) 296, 1–22 (2004)

    Article  ADS  Google Scholar 

  31. Huete, A. et al. Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sens. Environ. 83, 195–213 (2002)

    Article  ADS  Google Scholar 

  32. Cleugh, H. A., Leuning, R., Mu, Q. & Running, S. W. Regional evaporation estimates from flux tower and MODIS satellite data. Remote Sens. Environ. 106, 285–304 (2007)

    Article  ADS  Google Scholar 

  33. Zuur, A. F., Ieno, E. N. & Smith, G. M. Analyzing Ecological Data. (Springer, 2007)

  34. R Core Team. R: A language and environment for statistical computing. http://www.R-project.org (2015).

  35. Hijmans, R. J. raster: geographic data analysis and modeling. R package version 2.4-20. http://CRAN.R-project.org/package=raster (2015).

  36. Pinheiro, J., Bates, D., Debroy, S., Sarkar, D. & Team, A. T. R. D. C. nlme: linear and nonlinear mixed effects models. R package version 3.1-122. http://CRAN.R-project.org/package=nlme (2013).

  37. Pebesma, E. J. Multivariable geostatistics in S: the gstat package. Comput. Geosci. 30, 683–691 (2004)

    Article  ADS  Google Scholar 

  38. Bivand, R., Keitt, T. & Rowlingson, B. rgdal: bindings for the geospatial data abstraction library. R package version 0.9-3. http://CRAN.R-project.org/package=rgdal (2015).

  39. Warnes, G. R., Bolker, B. & Lumley, T. gtools: various R programming tools. R package version 3.5.0. http://CRAN.R-project.org/package=gtools (2015).

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

    Article  Google Scholar 

  41. Pope, N. corHaversine function. http://stackoverflow.com/questions/18857443/specifying-a-correlation-structure-for-a-linear-mixed-model-using-the-ramps-pack (2013).

Download references

Acknowledgements

This work was funded by Statoil ASA, Norway, Contract number 4501995279 (K.J.W., A.W.R.S., D.B.), and by the European Commission LIFE12 ENV/UK/000473 (K.J.W., D.B. and P.R.L.). P.R.L. was also supported by an Oxford Martin School Fellowship. M.M.-F. was supported by a Natural Environment Research Council Independent Research Fellowship (NE/L011859/1) and A.W.R.S. was supported by a Research Council of Norway Postdoctoral Fellowship within a FRIMEDBIO project grant (FRIMEDBIO-214359) during analysis and write-up of this work.

Author information

Author notes

  1. Alistair W. R. Seddon and Marc Macias-Fauria: These authors contributed equally to this work.

Authors and Affiliations

  1. Department of Biology, University of Bergen, Allégaten 41, Bergen, N-500, Norway

    Alistair W. R. Seddon & Kathy J. Willis

  2. School of Geography and the Environment, South Parks Road, University of Oxford, Oxford OX1, 3QY, UK

    Marc Macias-Fauria

  3. Department of Zoology, Long-Term Ecology Laboratory, Biodiversity Institute, Oxford Martin School, South Parks Road, University of Oxford, Oxford, OX1 3PS, UK

    Peter R. Long, David Benz & Kathy J. Willis

  4. Royal Botanic Gardens, Kew, Richmond, TW9 3AB, Surrey, UK

    Kathy J. Willis

Authors
  1. Alistair W. R. Seddon
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Marc Macias-Fauria
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Peter R. Long
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. David Benz
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Kathy J. Willis
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

All authors designed the study. D.B. and P.R.L. prepared and downloaded the remote-sensing data and A.W.R.S. and M.M.-F. carried out the data analysis. A.W.R.S., M.M.-F. and K.J.W. co-wrote the paper, with contributions from D.B. and P.R.L.

Corresponding author

Correspondence to Alistair W. R. Seddon.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Additional information

Remote sensing data are uploaded in the ORA repository (http://www.bodleian.ox.ac.uk/ora, DOI:10.5287/bodleian:VY2PeyGX4.

Extended data figures and tables

Extended Data Figure 1 Study Design.

Flow chart of the algorithm used to estimate the vegetation sensitivity index.

Extended Data Figure 2 RGB composite of climate weights.

RGB composite global map of the mean climate coefficient weights from monthly multiple regressions between vegetation productivity (defined as EVI), vegetation productivity at t−1 and three climate variables (temperature, red; water availability, blue; and cloudiness, green). Areas with dominant barren land (mean EVI < 0.1 for all months) and permanent ice are shown grey. Pixel resolution, 5 km; period, 2000–2013.

Extended Data Figure 3 Global map of the t−1 coefficient.

Global map of t−1 (AR1) coefficient weight from a monthly multiple regressions between vegetation productivity (defined as EVI), vegetation productivity at t−1 and the three climate variables. Areas with dominant barren land (mean EVI < 0.1 for all months) and permanent ice are shown grey. Wetland areas, as identified by the Global Lakes and Wetlands Database30, are mapped in blue. Pixel resolution, 5 km; period, 2000–2013. Continental outlines were modified from a shapefile using ArcGIS 10.2 software (http://www.arcgis.com/home/item.html?id=a3cb207855b348a297ab85261743351d). ArcGIS and ArcMap are the intellectual property of Esri and are used herein under license.

Extended Data Figure 4 EVI variability in areas of low total annual precipitation.

Time series plots of the mean EVI (green) and mean EVI monthly anomalies (blue) for six different dryland/water-limited regions across the world. Time series are calculated by finding the mean monthly value for all 5-km pixels with a 1° grid cell (total pixels = 400). The light green shading in the mean EVI plots represents the upper and lower two standard deviations. a, North American temperate grassland (pixel centre 99.5 W, 47.5 N). b, Eurasian temperate grassland (30.5 E, 48.5 N). c, Eurasian temperate grassland (115.5 °E, 44.5 °N). d, Caatinga forests, woodlands and scrub (37.5 W, 8.5 S). e, Sahel subtropical savanna and shrubland (10.5 E, 13.5 N). f, Australian desert (127.5 E, 27.5 N). The map in the main panel insert represents areas with t−1 and water limitation linear regression coefficients within the upper quartile (see Methods). Red, t−1; dark blue, water limitation; light blue, both).

Source data

Extended Data Figure 5 t−1 and water limitation against total annual precipitation.

a, b, Plots of the t−1 (a) and water limitation coefficients (b) from the AR1 linear regression model (see Methods) plotted against total annual precipitation (mm) calculated as the sum of the WorldClim monthly precipitation data40. A random subsample of 1,000 points were taken from dryland areas, defined here as having total annual precipitation between 100 – 800 mm, and between 50 N and 50 S. After removing no-data values from the random subset (that is, unresponsive pixels from the VSI calculation), the total number of samples was 795. A linear model was fit to both data sets independently using generalized least squares in the ‘nlme’36 package in R34. An exponential spatial error term using geographic distance was used to account for spatial autocorrelation in the residuals in the model41. There was a negative significant effect on the size of the t−1 coefficient with increasing total annual precipitation (−0.0003 ± 0.00003, significant at P < 0.01), with a smaller, positive effect of total annual precipitation on water availability (0.0001 ± 0.00003, significant at P < 0.01).

Source data

Extended Data Figure 6 Cloudiness index.

Example output of the cloudiness index derived from the MOD35_L2 Cloud Mask product for June 2005. High values indicate more cloud-free days. Note the large number of cloud-free days in dryland regions, and the large number of cloudy days in southeast Asia as a result of the seasonal monsoon. Pixel resolution, 5 km.

Extended Data Figure 7 Number of months with a significant (P < 0.1) coefficient in the principal components regression.

Number of months with a significant (P < 0.1) coefficient in the principal components regression between vegetation productivity (EVI), and climate (temperature, water availability, and cloud cover), and a t−1 vegetation variable. Areas with dominant barren land (mean EVI < 0.1 for all months) and permanent ice are shown grey. Wetland areas, as identified by the Global Lakes and Wetlands Database30, are mapped in blue. Pixel resolution, 5 km; period, 2000–2013. Continental outlines were modified from a shapefile using ArcGIS 10.2 software (http://www.arcgis.com/home/item.html?id=a3cb207855b348a297ab85261743351d). ArcGIS and ArcMap are the intellectual property of Esri and are used herein under license.

Extended Data Figure 8 Mean–variance relationships.

a–d, Plots of the mean–variance relationships for EVI (a) and the three climate variables derived from MODIS data (ground temperature (b), water availability (c) and cloud cover (d)). Owing to the large number of pixels (7,200 × 3,000), these plots are made using 1,000 randomly sampled points from across the Earth surface for clarity.

Source data

Extended Data Figure 9 Mean standard error of the MODIS EVI observations.

Mean standard error of the MODIS EVI observations, calculated on a monthly basis over the period 2000–2013 as the standard deviation of all EVI observations per 5 km pixel divided by the square root of the number of observations. Areas with dominant barren land (mean EVI < 0.1 for all months) and permanent ice are shown grey. Wetland areas, as identified by the Global Lakes and Wetlands Database30, are mapped in blue. Continental outlines were modified from a shapefile using ArcGIS 10.2 software (http://www.arcgis.com/home/item.html?id=a3cb207855b348a297ab85261743351d). ArcGIS and ArcMap are the intellectual property of Esri and are used herein under license.

Extended Data Figure 10 Normalized confidence interval amplitudes.

Normalized confidence interval amplitudes (NCIA) for the regression coefficients in the EVI versus external forcings (temperature, water availability, cloudiness) and memory effects (EVI t−1) regression. Larger NCIA values correspond to larger uncertainty in the coefficient estimates. Amplitudes were normalized by the mean coefficient value in each 5 km pixel (that is, a value of 2 corresponds to a total uncertainty twice as big as the coefficient value). Only significant coefficients in the original PCA regression were accounted for, and hence no coefficient crosses zero in any pixel. Areas with dominant barren land (mean EVI < 0.1 for all months) and permanent ice are shown grey. Wetland areas, as identified by the Global Lakes and Wetlands Database30, are mapped in blue. a, Water availability; b, temperature; c, cloudiness; d, EVI t−1. Continental outlines were modified from a shapefile using ArcGIS 10.2 software (http://www.arcgis.com/home/item.html?id=a3cb207855b348a297ab85261743351d). ArcGIS and ArcMap are the intellectual property of Esri and are used herein under license.

PowerPoint slides

Source data

Rights and permissions

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Seddon, A., Macias-Fauria, M., Long, P. et al. Sensitivity of global terrestrial ecosystems to climate variability. Nature 531, 229–232 (2016). https://doi.org/10.1038/nature16986

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nature16986

This article is cited by

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.

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