Letter | Published:

Globally rising soil heterotrophic respiration over recent decades

Naturevolume 560pages8083 (2018) | Download Citation


Global soils store at least twice as much carbon as Earth’s atmosphere1,2. The global soil-to-atmosphere (or total soil respiration, RS) carbon dioxide (CO2) flux is increasing3,4, but the degree to which climate change will stimulate carbon losses from soils as a result of heterotrophic respiration (RH) remains highly uncertain5,6,7,8. Here we use an updated global soil respiration database9 to show that the observed soil surface RH:RS ratio increased significantly, from 0.54 to 0.63, between 1990 and 2014 (P = 0.009). Three additional lines of evidence provide support for this finding. By analysing two separate global gross primary production datasets10,11, we find that the ratios of both RH and RS to gross primary production have increased over time. Similarly, significant increases in RH are observed against the longest available solar-induced chlorophyll fluorescence global dataset, as well as gross primary production computed by an ensemble of global land models. We also show that the ratio of night-time net ecosystem exchange to gross primary production is rising across the FLUXNET201512 dataset. All trends are robust to sampling variability in ecosystem type, disturbance, methodology, CO2 fertilization effects and mean climate. Taken together, our findings provide observational evidence that global RH is rising, probably in response to environmental changes, consistent with meta-analyses13,14,15,16 and long-term experiments17. This suggests that climate-driven losses of soil carbon are currently occurring across many ecosystems, with a detectable and sustained trend emerging at the global scale.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Additional information

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


  1. 1.

    Köchy, M., Hiederer, R. & Freibauer, A. Global distribution of soil organic carbon—Part 1: Masses and frequency distributions of SOC stocks for the tropics, permafrost regions, wetlands, and the world. Soil 1, 351–365 (2015).

  2. 2.

    Scharlemann, J. P. W., Tanner, E. V. J., Hiederer, R. & Kapos, V. Global soil carbon: understanding and managing the largest terrestrial carbon pool. Carbon Manag. 5, 81–91 (2014).

  3. 3.

    Bond-Lamberty, B. & Thomson, A. M. Temperature-associated increases in the global soil respiration record. Nature 464, 579–582 (2010).

  4. 4.

    Hashimoto, S. et al. Global spatiotemporal distribution of soil respiration modeled using a global database. Biogeosciences 12, 4121–4132 (2015).

  5. 5.

    Friedlingstein, P. et al. Uncertainties in CMIP5 climate projections due to carbon cycle feedbacks. J. Clim. 27, 511–526 (2014).

  6. 6.

    Zhou, T., Phi, P., Hui, D. & Luo, Y. Global pattern of temperature sensitivity of soil heterotrophic respiration (Q10) and its implications for carbon-climate feedback. J. Geophys. Res. Biogeosci. 114, G02016 (2009).

  7. 7.

    Trumbore, S. E. & Czimczik, C. I. An uncertain future for soil carbon. Science 321, 1455–1456 (2008).

  8. 8.

    Giardina, C. P., Litton, C. M., Crow, S. E. & Asner, G. P. Warming-related increases in soil CO2 efflux are explained by increased below-ground carbon flux. Nat. Clim. Change 4, 822–827 (2014).

  9. 9.

    Bond-Lamberty, B. & Thomson, A. M. A global database of soil respiration data. Biogeosciences 7, 1915–1926 (2010).

  10. 10.

    Jung, M. et al. Global patterns of land-atmosphere fluxes of carbon dioxide, latent heat, and sensible heat derived from eddy covariance, satellite, and meteorological observations. J. Geophys. Res. Biogeosci. 116, G00J07 (2011).

  11. 11.

    Zhao, M., Heinsch, F. A., Nemani, R. R. & Running, S. W. Improvements of the MODIS terrestrial gross and net primary production global data set. Remote Sens. Environ. 95, 164–176 (2005).

  12. 12.

    Baldocchi, D. D. ‘Breathing’ of the terrestrial biosphere: lessons learned from a global network of carbon dioxide flux measurement systems. Aust. J. Bot. 56, 1–26 (2008).

  13. 13.

    Crowther, T. W. et al. Quantifying global soil carbon losses in response to warming. Nature 540, 104–108 (2016).

  14. 14.

    Lu, M. et al. Responses of ecosystem carbon cycle to experimental warming: a meta-analysis. Ecology 94, 726–738 (2013).

  15. 15.

    Wang, X. et al. Soil respiration under climate warming: differential response of heterotrophic and autotrophic respiration. Glob. Change Biol. 20, 3229–3237 (2014).

  16. 16.

    Zhou, L. et al. Interactive effects of global change factors on soil respiration and its components: a meta-analysis. Glob. Change Biol. 22, 3157–3169 (2016).

  17. 17.

    Melillo, J. M. et al. Long-term pattern and magnitude of soil carbon feedback to the climate system in a warming world. Science 358, 101–105 (2017).

  18. 18.

    Bond-Lamberty, B., Wang, C. & Gower, S. T. A global relationship between the heterotrophic and autotrophic components of soil respiration? Glob. Change Biol. 10, 1756–1766 (2004).

  19. 19.

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

  20. 20.

    Hursh, A. et al. The sensitivity of soil respiration to soil temperature, moisture, and carbon supply at the global scale. Glob. Change Biol. 23, 2090–2103 (2017).

  21. 21.

    Vargas, R. et al. On the multi-temporal correlation between photosynthesis and soil CO2 efflux: reconciling lags and observations. New Phytol. 191, 1006–1017 (2011).

  22. 22.

    Anav, A. et al. Spatiotemporal patterns of terrestrial gross primary production: a review. Rev. Geophys. 53, 2015RG000483 (2015).

  23. 23.

    Warszawski, L. et al. The Inter-Sectoral Impact Model Intercomparison Project (ISI-MIP): project framework. Proc. Natl Acad. Sci. USA 111, 3228–3232 (2014).

  24. 24.

    Falge, E. et al. Seasonality of ecosystem respiration and gross primary production as derived from FLUXNET measurements. Agric. For. Meteorol. 113, 53–74 (2002).

  25. 25.

    Pilegaard, K., Ibrom, A., Courtney, M. S., Hummelshøj, P. & Jensen, N. O. Increasing net CO2 uptake by a Danish beech forest during the period from 1996 to 2009. Agric. For. Meteorol. 151, 934–946 (2011).

  26. 26.

    Urbanski, S. P. et al. Factors controlling CO2 exchange on timescales from hourly to decadal at Harvard Forest. J. Geophys. Res. 112, G02020 (2007).

  27. 27.

    Keenan, T. F. et al. Increase in forest water-use efficiency as atmospheric carbon dioxide concentrations rise. Nature 499, 324–327 (2013).

  28. 28.

    Adamczyk, S., Adamczyk, B., Kitunen, V. & Smolander, A. Monoterpenes and higher terpenes may inhibit enzyme activities in boreal forest soil. Soil Biol. Biochem. 87, 59–66 (2015).

  29. 29.

    Le Quéré, C. et al. Global carbon budget 2016. Earth Syst. Sci. Data 8, 605–649 (2016).

  30. 30.

    Vargas, R., Paz, F. & de Jong, B. Quantification of forest degradation and belowground carbon dynamics: ongoing challenges for monitoring, reporting and verification activities for REDD+. Carbon Manag. 4, 579–582 (2013).

  31. 31.

    Wayson, C. A., Randolph, J. C., Hanson, P. J., Grimmond, C. S. B. & Schmid, H. P. Comparison of soil respiration methods in a mid-latitude deciduous forest. Biogeochemistry 80, 173–189 (2006).

  32. 32.

    Reichstein, M. et al. Modeling temporal and large-scale spatial variability of soil respiration from soil water availability, temperature and vegetation productivity indices. Glob. Biogeochem. Cycles 17, 1104 (2003).

  33. 33.

    Osborn, T. J. & Jones, P. D. The CRUTEM4 land-surface air temperature data set: construction, previous versions and dissemination via Google Earth. Earth Syst. Sci. Data 6, 61–68 (2014).

  34. 34.

    Dorigo, W. et al. Evaluating global trends (1988–2010) in harmonized multi-satellite surface soil moisture. Geophys. Res. Lett. 39, L18405 (2012).

  35. 35.

    Hengl, T. et al. SoilGrids1km — global soil information based on automated mapping. PLoS ONE 9, e105992 (2014).

  36. 36.

    Ito, A. A historical meta-analysis of global terrestrial net primary productivity: Are estimates converging? Glob. Change Biol. 17, 3161–3175 (2011).

  37. 37.

    Frankenberg, C. et al. New global observations of the terrestrial carbon cycle from GOSAT: patterns of plant fluorescence with gross primary productivity. Geophys. Res. Lett. 38, L17706 (2011).

  38. 38.

    Joiner, J. et al. Global monitoring of terrestrial chlorophyll fluorescence from moderate-spectral-resolution near-infrared satellite measurements: methodology, simulations, and application to GOME-2. Atmos. Meas. Tech. 6, 2803–2823 (2013).

  39. 39.

    Köhler, P., Guanter, L. & Joiner, J. A linear method for the retrieval of sun-induced chlorophyll fluorescence from GOME-2 and SCIAMACHY data. Atmos. Meas. Tech. 8, 2589–2608 (2015).

  40. 40.

    Chapin, F. S. et al. Reconciling carbon-cycle concepts, terminology, and methods. Ecosystems 9, 1041–1050 (2006).

  41. 41.

    Fernandes, R., Leblanc, S. G. Parametric (modified least squares) and non-parametric (Theil–Sen) linear regressions for predicting biophysical parameters in the presence of measurement errors. Remote Sens. Environ. 95, 303–316 (2005).

  42. 42.

    Xu, M. & Shang, H. Contribution of soil respiration to the global carbon equation. J. Plant Physiol. 203, 16–28 (2016).

  43. 43.

    Ito, A. et al. Photosynthetic productivity and its efficiencies in ISIMIP2a biome models: benchmarking for impact assessment studies. Environ. Res. Lett. 12, 085001 (2017).

  44. 44.

    Kolby Smith, W. et al. Large divergence of satellite and Earth system model estimates of global terrestrial CO2 fertilization. Nat. Clim. Change 6, 306–310 (2016).

  45. 45.

    Rosenzweig, C. et al. Assessing inter-sectoral climate change risks: the role of ISIMIP. Environ. Res. Lett. 12, 010301 (2017).

  46. 46.

    R Development Core Team. R: A Language And Environment For Statistical Computing. Version 3.3.3 (2017).

Download references


We are indebted to the thousands of researchers who measured and published the data collected here. This work used eddy-covariance data acquired and shared by the FLUXNET community (see Methods). This research was supported by the US Department of Energy, Office of Science, Biological and Environmental Research as part of the Terrestrial Ecosystem Sciences Program. The Pacific Northwest National Laboratory is operated for DOE by Battelle Memorial Institute under contract DE-AC05-76RL01830. R.V. acknowledges support from NASA-CMS (80NSSC18K0173) and USDA (2014-67003-22070). C.M.G. received additional support from the National Science Foundation Division of Environmental Biology, Award 1353908.

Reviewer information

Nature thanks A. Konings, K. Ogle and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Author information


  1. Pacific Northwest National Laboratory, Joint Global Change Research Institute at the University of Maryland–College Park, College Park, MD, USA

    • Ben Bond-Lamberty
    •  & Min Chen
  2. Biological Sciences Division, Pacific Northwest National Laboratory, Richland, WA, USA

    • Vanessa L. Bailey
  3. Department of Biology, Virginia Commonwealth University, Richmond, VA, USA

    • Christopher M. Gough
  4. Department of Plant and Soil Sciences, University of Delaware, Newark, DE, USA

    • Rodrigo Vargas


  1. Search for Ben Bond-Lamberty in:

  2. Search for Vanessa L. Bailey in:

  3. Search for Min Chen in:

  4. Search for Christopher M. Gough in:

  5. Search for Rodrigo Vargas in:


B.B.-L. conceived this study, and together with C.M.G. and R.V. designed the analysis. M.C. provided SIF data processing and expertise, and with V.L.B. furnished key insights. B.B.-L. wrote the manuscript in close collaboration with all authors.

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to Ben Bond-Lamberty.

Extended data figures and tables

  1. Extended Data Fig. 1 Effect of dataset temporal span on RH:RS trend.

    The model summarized in Table 1 was repeatedly fitted to the SRDB dataset, as filtered with various first and final year assumptions (that is, the earliest and latest measurement allowed, to test whether the results were an artefact of the 1989–2014 period chosen, indicated by a black diamond in the figure). Colour shows the significance of the RH:RS temporal trend.

  2. Extended Data Fig. 2 SRDB coverage in terrestrial climate space.

    Blue dots show MAT and MAP of the entries in the database; lighter points indicate more recent data. Grey background indicates global climate distribution in the HadCRUT433 data, with darker shades indicating more land area, for 1991–2010.

  3. Extended Data Fig. 3 Respiration-to-production model residuals over time.

    The model examines how the dependent variable—the ratio of respiration to GPP or SIF—is related to the independent variables: climate, land cover, disturbance and SOC content; for more details, see Methods. Two respiration fluxes (RH and RS), two GPP sources (the statistically upscaled MTE dataset, and the remotely sensed MODIS product), and two SIF sources (SCIAMACHY and GOME-2) are shown. Grey bands show 95% confidence intervals. Blue lines indicate least-squares trend, while grey bands show 95% confidence intervals.

  4. Extended Data Fig. 4 Difference between FLUXNET GPP and remotely sensed GPP and SIF over time.

    Trends are not significant for the satellite data from GOME-2 SIF (P = 0.22), MODIS GPP (P = 0.07) or SCIMACHY SIF (P = 0.49). Blue lines (solid line is statistically significant, dashed line is non-significant) lines indicate least-squares trend. The temporal trend for MTE GPP decreases significantly (P = 0.02) over time, that is, MTE GPP increasingly underpredicts GPP measured by eddy covariance tower.

  5. Extended Data Fig. 5 Sensitivity analysis of how much GPP satellites would have to be missing to render the GPP–respiration trends shown in Fig. 2a non-significant.

    We ran a sensitivity analysis that assumed a conservative (that is, high) GPP trend of 0.5% yr-1, and repeatedly re-fitted the Fig. 2a models assuming that satellites missed 0%, 10%, 20%,...100% of this gain. This figure plots the ‘missed’ percentage (x axis) versus the P-value of the linear regression (y axis) for each respiration flux (RH and RS) and GPP or SIF data product (‘GPPadjusted’, that is, adjusted for the assumed ‘missed’ percentage). Even if satellites are missing 100% of this large assumed GPP increase, most of the temporal trends shown in Fig. 2a trends remain highly significant.

  6. Extended Data Fig. 6 Ratio of NEEnight to GPP in the FLUXNET 2015 ‘Tier 1’ dataset.

    Panels show different International Geosphere-Biosphere Programme (IGBP) land cover type: crop (CRO), closed shrubland (CSH), deciduous broadleaf forest (DBF), evergreen broadleaf forest (EBF), evergreen needleleaf forest (ENF), grassland (GRA), mixed forests (MF), open shrublands (OSH), herbaceous savannahs (SAV), snow and ice (SNO), wetlands (WET) and woody savannahs (WSA).

  7. Extended Data Fig. 7 Bootstrap analysis examining effect of ratio of significant to non-significant data on likelihood of observing a trend.

    What is the likelihood that the FLUXNET data subset (that included GPP measurements made in the same study site and year as site-specific RS measurements in the SRDB) is too small to detect signals of rising RH, given site-to-site variability in climate and carbon dynamics? Each point is a different random draw from the original dataset, the comparison of RS to MODIS GPP, with 1,000 bootstrap draws per fraction of artificial no-trend data. The horizontal dashed line is P = 0.05. Note log scale (to base e, which R defaults to) of y axis. Boxplots visualize five summary statistics (the median, the first and third quartiles, and two whiskers; see https://ggplot2.tidyverse.org/reference/geom_boxplot.html); points are semi-transparent to indicate data density at each point.

  8. Extended Data Fig. 8 Bootstrap analysis examining effect of dataset size on likelihood of observing a trend.

    What is the likelihood that the n = 106 FLUXNET dataset is too small to detect signals of rising RH, given site-to-site variability in climate and carbon dynamics? Each point is a different random draw from the original dataset, the comparison of RS to MODIS GPP, with 1,000 bootstrap draws per fraction of no-trend data. The vertical dashed line shows the size of the original RS:FLUXNET dataset, about 5% of the RS:GPPMODIS data, while the horizontal line is P = 0.05. Note log scale (to base e) of y axis. Boxplots visualize five summary statistics (the median, the first and third quartiles, and two whiskers; see https://ggplot2.tidyverse.org/reference/geom_boxplot.html); points are semi-transparent to indicate data density at each point.

  9. Extended Data Fig. 9 Distribution of the longitudinal site data in climate change space relative to the main SRDB dataset.

    The longitudinal data (Fig. 2c, coloured dots below) in climate change space (1991–2010 HadCRUT4 changes in air temperature and precipitation) are shown over the distribution of the main dataset (n = 1,852, grey dots). The longitudinal data cover −0.03 to +0.08 °C yr−1 and −12 to +16 mm yr−1; the main dataset covers a much broader climate change space of −0.07 to +0.17 °C yr−1 and −35 to +57 mm yr−1.

  10. Extended Data Table 1 Coefficients for general model18 of relationship between RS and RH

Supplementary information

  1. Supplementary Information

    References for FLUXNET site data used in this analysis. See http://fluxnet.fluxdata.org/data/data-policy/ . Citation numbers refer to references listed in the Methods.

About this article

Publication history







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.