Hydroclimate changes across the Amazon lowlands over the past 45,000 years


Reconstructing the history of tropical hydroclimates has been difficult, particularly for the Amazon basin—one of Earth’s major centres of deep atmospheric convection1,2. For example, whether the Amazon basin was substantially drier3,4 or remained wet1,5 during glacial times has been controversial, largely because most study sites have been located on the periphery of the basin, and because interpretations can be complicated by sediment preservation, uncertainties in chronology, and topographical setting6. Here we show that rainfall in the basin responds closely to changes in glacial boundary conditions in terms of temperature and atmospheric concentrations of carbon dioxide7. Our results are based on a decadally resolved, uranium/thorium-dated, oxygen isotopic record for much of the past 45,000 years, obtained using speleothems from Paraíso Cave in eastern Amazonia; we interpret the record as being broadly related to precipitation. Relative to modern levels, precipitation in the region was about 58% during the Last Glacial Maximum (around 21,000 years ago) and 142% during the mid-Holocene epoch (about 6,000 years ago). We find that, as compared with cave records from the western edge of the lowlands, the Amazon was widely drier during the last glacial period, with much less recycling of water and probably reduced plant transpiration, although the rainforest persisted throughout this time.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Paraíso speleothem record, and comparisons with local summer insolation, tropical Atlantic SST and atmospheric CO2 concentration.
Figure 2: Comparisons of eastern Amazon and eastern China stalagmite records.
Figure 3: Comparisons of speleotherm records from the eastern and western Amazon.


  1. 1

    Baker, P. A. et al. The history of South American tropical precipitation for the past 25,000 years. Science 291, 640–643 (2001)

    ADS  CAS  Article  Google Scholar 

  2. 2

    Maslin, M. A. et al. Dynamic boundary-monsoon intensity hypothesis: evidence from the deglacial Amazon River discharge record. Quat. Sci. Rev. 30, 3823–3833 (2011)

    ADS  Article  Google Scholar 

  3. 3

    Ledru, M.-P., Bertaux, J. & Sifeddine, A. Absence of Last Glacial Maximum records in lowland tropical forests. Quat. Res. 49, 233–237 (1998)

    Article  Google Scholar 

  4. 4

    D’Apolito, C., Absy, M. L. & Latrubesse, E. M. The Hill of Six Lakes revisited: new data and re-evaluation of a key Pleistocene Amazon site. Quat. Sci. Rev. 76, 140–155 (2013)

    ADS  Article  Google Scholar 

  5. 5

    Colinvaux, P. A. et al. A long pollen record from lowland Amazonia: forest and cooling in glacial times. Science 274, 85–88 (1996)

    ADS  CAS  Article  Google Scholar 

  6. 6

    Baker, P. A. & Fritz, S. C. Nature and causes of Quaternary climate variation of tropical South America. Quat. Sci. Rev. 124, 31–47 (2015)

    ADS  Article  Google Scholar 

  7. 7

    Marcott, S. A. et al. Centennial-scale changes in the global carbon cycle during the last deglaciation. Nature 514, 616–619 (2014)

    ADS  CAS  Article  Google Scholar 

  8. 8

    Stute, M. et al. Cooling of tropical Brazil (5°C) during the Last Glacial Maximum. Science 269, 379–383 (1995)

    ADS  CAS  Article  Google Scholar 

  9. 9

    van Breukelen, M. R. et al. Fossil dripwater in stalagmites reveals Holocene temperature and rainfall variation in Amazonia. Earth Planet. Sci. Lett. 275, 54–60 (2008)

    ADS  CAS  Article  Google Scholar 

  10. 10

    Eltahir, E. A. B. & Bras, R. L. Precipitation recycling in the Amazon Basin. Q. J. R. Meteorol. Soc. 120, 861–880 (1994)

    ADS  Article  Google Scholar 

  11. 11

    Dansgaard, W. Stable isotopes in precipitation. Tellus 16, 436–468 (1964)

    ADS  Article  Google Scholar 

  12. 12

    Vuille, M. et al. Modeling δ18O in precipitation over the tropical Americas: 1. Interannual variability and climatic controls. J. Geophys. Res. 108 (D6), 4174 (2003)

    Article  Google Scholar 

  13. 13

    Lee, J.-E., Johnson, K. & Fung, I. Precipitation over South America during the Last Glacial Maximum: an analysis of the “amount effect” with a water isotope-enabled general circulation model. Geophys. Res. Lett. 36, L19701 (2009)

    ADS  Article  Google Scholar 

  14. 14

    Wanner, H. et al. Mid- to late Holocene climate change: an overview. Quat. Sci. Rev. 27, 1791–1828 (2008)

    ADS  Article  Google Scholar 

  15. 15

    Wang, X. et al. Millennial-scale precipitation changes in southern Brazil over the past 90,000 years. Geophys. Res. Lett. 34, L23701 (2007)

    ADS  Google Scholar 

  16. 16

    Cruz, F. W. et al. Orbitally driven east-west antiphasing of South American precipitation. Nat. Geosci. 2, 210–214 (2009)

    ADS  CAS  Article  Google Scholar 

  17. 17

    Cheng, H. et al. Climate change patterns in Amazonia and biodiversity. Nat. Commun. 4, 1411 (2013)

    ADS  Article  Google Scholar 

  18. 18

    Shakun, J. D. et al. Global warming preceded by increasing carbon dioxide concentrations during the last deglaciation. Nature 484, 49–54 (2012)

    ADS  CAS  Article  Google Scholar 

  19. 19

    Jaeschke, A., Rühlemann, C., Arz, H., Heil, G. & Lohmann, G. Coupling of millennial-scale changes in sea surface temperature and precipitation off northeastern Brazil with high-latitude climate shifts during the last glacial period. Paleoceanography 22, PA4206 (2007)

    ADS  Article  Google Scholar 

  20. 20

    Cook, K. H. & Vizy, E. K. South American climate during the Last Glacial Maximum: delayed onset of the South American monson. J. Geophys. Res. 111, D02110 (2006)

    ADS  Article  Google Scholar 

  21. 21

    Malhi, Y. et al. Climate change, deforestation, and the fate of the Amazon. Science 319, 169–172 (2008)

    ADS  CAS  Article  Google Scholar 

  22. 22

    Koutavas, A. & Joanides, St. El Niño-Southern Oscillation extrema in the Holocene and Last Glacial Maximum. Paleoceanography 27, PA4208 (2012)

    ADS  Article  Google Scholar 

  23. 23

    Wang, Y. J. et al. A high-resolution absolute-dated Late Pleistocene monsoon record from Hulu Cave, China. Science 294, 2345–2348 (2001)

    ADS  CAS  Article  Google Scholar 

  24. 24

    Cheng, H. et al. The Asian monsoon over the past 640,000 years and ice age terminations. Nature 534, 640–646 (2016)

    ADS  CAS  Article  Google Scholar 

  25. 25

    WAIS Divide Project Members. Precise interpolar phasing of abrupt climate change during the last ice age. Nature 520, 661–665 (2015)

  26. 26

    Salati, E., Dall’Olio, A., Matsui, E. & Gat, J. R. Recycling of water in the Amazon Basin: an isotopic study. Wat. Resour. Res. 15, 1250–1258 (1979)

    ADS  CAS  Article  Google Scholar 

  27. 27

    Winnick, M. J., Chamberlain, C. P., Caves, J. K. & Welker, J. M. Quantifying the isotopic ‘continental effect’. Earth Planet. Sci. Lett. 406, 123–133 (2014)

    ADS  CAS  Article  Google Scholar 

  28. 28

    Held, I. M. & Soden, B. J. Robust response of the hydrological cycle to global warming. J. Clim. 19, 5686–5699 (2006)

    ADS  Article  Google Scholar 

  29. 29

    Vecchi, G. A. et al. Weakening of tropical Pacific atmospheric circulation due to anthropogenic forcing. Nature 441, 73–76 (2006)

    ADS  CAS  Article  Google Scholar 

  30. 30

    Spracklen, D. V., Arnold, S. R. & Taylor, C. M. Observations of increased tropical rainfall preceded by air passes over forests. Nature 489, 282–285 (2012)

    ADS  CAS  Article  Google Scholar 

  31. 31

    Edwards, R. L., Chen, J. H. & Wasserburg, G. J. 238U-234U-230Th-232Th systematics and the precise measurement of time over the past 500,000 years. Earth Planet. Sci. Lett. 81, 175–192 (1986/87)

    ADS  Article  Google Scholar 

  32. 32

    Cheng, H. et al. The half-lives of uranium-234 and thorium-230. Chem. Geol. 169, 17–33 (2000)

    ADS  CAS  Article  Google Scholar 

  33. 33

    Shen, C.-C. et al. High-precision and high-resolution carbonate 230Th dating by MC-ICP-MS with SEM protocols. Geochim. Cosmochim. Acta 99, 71–86 (2012)

    ADS  CAS  Article  Google Scholar 

  34. 34

    Cheng, H. et al. Improvements in 230Th dating, 230Th and 234U half-life values, and U-Th isotopic measurements by multi-collector inductively coupled plasma mass spectrometry. Earth Planet. Sci. Lett. 371/372, 82–91 (2013)

    ADS  Article  Google Scholar 

  35. 35

    Fairchild, I. J. & Baker, A. Speleothem Science: From Process To Past Environments p.450 (Wiley–Blackwell, 2012)

  36. 36

    Dorale, J. A., Edwards, R. L., Ito, E. & González, L. A. Climate and vegetation history of the mid-continent from 75 to 25 ka: a speleothem record from Crevice Cave, Missouri, USA. Science 282, 1871–1874 (1998)

    ADS  CAS  Article  Google Scholar 

  37. 37

    Hendy, C. H. The isotopic geochemistry of speleothems—I. The calculation of the effects of different modes of formation on the isotopic composition of speleothems and their applicability as palaeoclimatic indicators. Geochim. Cosmochim. Acta 35, 801–824 (1971)

    ADS  CAS  Article  Google Scholar 

  38. 38

    Bowen, G. J. Isoscapes: spatial pattern in isotopic biogeochemistry. Annu. Rev. Earth Planet. Sci. 38, 161–187 (2010)

    ADS  CAS  Article  Google Scholar 

  39. 39

    Johnston, V. E., Borsato, A., Spotl, C., Frisia, S. & Miorandi, R. Stable isotopes in caves over altitudinal gradients: fractionation behavior and inferences for speleothem sensitivity to climate change. Clim. Past 9, 99–118 (2013)

    Article  Google Scholar 

  40. 40

    Vuille, M. & Werner, M. Stable isotopes in precipitation recording South American summer monsoon and ENSO variability: observations and model results. Clim. Dyn. 25, 401–413 (2005)

    Article  Google Scholar 

  41. 41

    Masson-Delmotte, V. M. et al. in Climate Change 2013: The Physical Science Basis Ch. 5 (eds Stocker, T. F. et al.) 383–464 (Cambridge Univ. Press, 2013)

    Google Scholar 

  42. 42

    MARGO Project Members. Constraints on the magnitude and patterns of ocean cooling at the Last Glacial Maximum. Nat. Geosci. 2, 127–132 (2009)

  43. 43

    Marcott, S. A. et al. A reconstruction of regional and global temperature for the past 11,300 years. Science 339, 1198–1201 (2013)

    ADS  CAS  Article  Google Scholar 

  44. 44

    Thornthwaite, C. W. An approach toward a rational classification of climate. Geogr. Rev. 38, 55–94 (1948)

    Article  Google Scholar 

  45. 45

    McCabe, G. J. & Markstrom, S. L. A monthly water balance model driven by a graphical user interface. US Geol. Surv. Open-File Rep. 2007–1008 (2007)

  46. 46

    Betts, A. K. & Ridgway, W. Tropical boundary layer equilibrium in the last ice age. J. Geophys. Res. 97, 2529–2534 (1992)

    ADS  Article  Google Scholar 

  47. 47

    Kageyama, M., Harrison, S. P. & Abe-Ouchi, A. The depression of tropical snowlines at the last glacial maximum: what can we learn from climate model experiments? Quat. Int. 138-139, 202–219 (2005)

    Article  Google Scholar 

  48. 48

    Broecker, W. S. Mountain glaciers: records of atmospheric water vapor content? Glob. Biogeochem. Cycles 11, 589–597 (1997)

    ADS  CAS  Article  Google Scholar 

  49. 49

    Rozanski, K., Araguás-Araguás, L. & Gonfiantini, R. in Climate Change In Continental Isotopic Records (eds Swart, P. K., Lohmann, K. C., McKenzie, J. & Savin, S. ) doi:10.1029/GM078p0001 (Am. Geophys. Union, Washington DC, 1993)

  50. 50

    Laskar, J. et al. A long-term numerical solution for the insolation. Astron. Astrophys. 428, 261–285 (2004)

    ADS  Article  Google Scholar 

  51. 51

    Lüthi, D. et al. High-resolution carbon dioxide concentration record 650,000-800,000 years before present. Nature 453, 379–382 (2008)

    ADS  Article  Google Scholar 

  52. 52

    Loulergue, L. et al. Orbital and millennial-scale features of atmospheric CH4 over the past 800,000 years. Nature 453, 383–386 (2008)

    ADS  CAS  Article  Google Scholar 

  53. 53

    NGRIP Project Members. High-resolution record of Northern Hemisphere climate extending into the last interglacial period. Nature 431, 147–151 (2004)

  54. 54

    EPICA Community Members. One-to-one coupling of glacial climate variability in Greenland and Antarctica. Nature 444, 195–198 (2006)

  55. 55

    Veres, D. et al. The Antarctic ice core chronology (AICC2012): an optimized multi-parameter and multi-site dating approach for the last 120 thousand years. Clim. Past 9, 1733–1748 (2013)

    Article  Google Scholar 

  56. 56

    Svensson, A. et al. A 60,000 year Greenland stratigraphic ice core chronology. Clim. Past 4, 47–57 (2008)

    Article  Google Scholar 

  57. 57

    Wang, X. et al. Wet periods in northeastern Brazil over the past 210 kyr linked to distant climate anomalies. Nature 432, 740–743 (2004)

    ADS  CAS  Article  Google Scholar 

  58. 58

    Anderson, R. F. et al. Wind-driven upwelling in the Southern Ocean and the deglacial rise in atmospheric CO2 . Science 323, 1443–1448 (2009)

    ADS  CAS  Article  Google Scholar 

  59. 59

    Broecker, W. S. Paleocean circulation during the last deglaciation: a bipolar seesaw? Paleoceanography 13, 119–121 (1998)

    ADS  Article  Google Scholar 

  60. 60

    Buizert, C. et al. The WAIS Divide deep ice core WD2014 chronology. Part 1. Methane synchronization (68–31ka BP) and the gas age–ice age difference. Clim. Past 11, 153–173 (2015)

    Article  Google Scholar 

  61. 61

    Genty, D. et al. Precise dating of Dansgaard-Oeschger climate oscillations in western Europe from stalagmite data. Nature 421, 833–837 (2003)

    ADS  CAS  Article  Google Scholar 

  62. 62

    Schrag, D. P., Hampt, G. & Murray, D. W. Pore fluid constraints on the temperature and oxygen isotopic composition of the glacial ocean. Science 272, 1930–1932 (1996)

    ADS  CAS  Article  Google Scholar 

  63. 63

    Schrag, D. P. et al. The oxygen isotopic composition of seawater during the Last Glacial Maximum. Quat. Sci. Rev. 21, 331–342 (2002)

    ADS  Article  Google Scholar 

  64. 64

    Horita, J. & Wesolowski, D. J. Liquid-vapor fractionation of oxygen and hydrogen isotopes of water from the freezing to the critical temperature. Geochim. Cosmochim. Acta 58, 3425–3437 (1994)

    ADS  CAS  Article  Google Scholar 

Download references


This work was supported by a Singapore National Research Foundation (NRF) Fellowship (NRFF2011-08) and a Gary Comer Fellowship to X.W.; US National Science Foundation (NSF) grants 1103404 and 1317693 to R.L.E. and H.C.; a Brazil National Council for Scientific and Technological Development (CNPq) grant (540064/01-7) to A.S.A.; grants from the China National Basic Research Program (NBRP; 2013CB955902) and the National Natural Science Foundation of China (NSFC; 41230524) to H.C.; and a grant from the São Paulo Research Foundation of Brazil and US NSF Dimensions of Biodiversity joint program (FAPESP/NSF; 2012/50260-6) to F.W.C. Field travelling funds were partially supported by a National Geographical Society grant, 7574-03. We acknowledge the help of colleagues from the Grupo Bambuí de Pesquisas Espeleológicas with cave mapping and sampling. We thank R. Fonseca, S. Yuan, Y. Lu and Y. Djamil for assistance with the figures concerning wind fields and regional rainfall, and B. Wohlfarth and S. Hemming for discussions during manuscript preparation.

Author information




X.W., R.L.E. and A.S.A. designed the project. X.W., A.S.A. and J.A.D. performed the fieldwork and sampling. X.W. and H.-W. C. carried out the uranium/thorium dating. X.W., X.K. and Y.W. contributed to the oxygen-isotope measurements. X.W. wrote the manuscript, which was edited by R.L.E. and other authors. All authors discussed the results and implications and commented on the manuscript at all stages.

Corresponding author

Correspondence to Xianfeng Wang.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Additional information

Reviewer Information Nature thanks M. Bush, J. Shakun and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Extended data figures and tables

Extended Data Figure 1 Cave locations and moisture pathways.

a, The locations of Paraíso Cave in the eastern Amazon (red rectangle), and of Diamante cave17 (blue rectangle) and Tigre Perdido cave9 (purple rectangle) in the western Amazon. Paraíso Cave is located between Belém and Manaus, next to the Tapajós River. Also shown are easterlies, which carry moisture to the lowlands from the tropical Atlantic. The Amazon basin and the Andes are shown in green and brown, respectively. b, 72-hour back-trajectories of moisture arriving at Paraíso and the western Amazonian cave sites (white stars), during the wet season (in red) and the dry season (in blue), averaged over 1981 to 2010. The background topographical map was created with grid files from the global multi-resolution topography (GMRT) synthesis (http://www.marine-geo.org/tools/GMRTMapTool). Moisture trajectories were derived using the US National Oceanic and Atmospheric Administration (NOAA) Hysplit model (http://ready.arl.noaa.gov/HYSPLIT.php). The moisture at the Paraíso Cave site is predominantly from the tropical Atlantic, whereas precipitation received in the western Amazon has largely endured recycling in the lowlands.

Extended Data Figure 2 Climatology of tropical South America.

a, Depiction of horizontal winds over South America at 850 hPa (vectors, in metres per second), based on data from the National Centers for Environmental Prediction (NCEP) Climate Forecast System Reanalysis (CFSR) (1981–2010; http://cfs.ncep.noaa.gov/cfsr/atlas/). Also shown is precipitation (blue shading, in millimetres per day) from the Tropical Rainfall Measuring Mission (TRMM) 3B43 dataset (1998–2010; http://trmm.gsfc.nasa.gov/3b43.html). Winds and precipitation are averaged over December to March. b, As in a, but for June to September. c, Monthly averaged temperature, precipitation and rainfall δ18O over Belém (blue dots) and Manaus (green triangles). The local climate at Paraíso Cave shares the same characteristics as those of Belém and Manaus. Data are from the International Atomic Energy Agency (IAEA) Global Networks of Isotopes in Precipitation (GNIP) database (http://www-naweb.iaea.org/napc/ih/IHS_resources_gnip.html).

Extended Data Figure 3 A Paraíso calcite stalagmite, and age models.

a, Image of a Paraíso sample. The Paraíso calcite stalagmites typically have a high uranium concentration (up to 40 p.p.m.) but a low thorium concentration (<1 parts per billion, p.p.b.), almost ideal for uranium/thorium-based age determination. b, Age models for samples PAR01, PAR03, PAR06, PAR07, PAR08, PAR16 and PAR24. The chronology of the samples is established by linear interpolation between successive uranium/thorium dates. Dates are shown in black dots. Age uncertainties (2σ) are also included (most of the error bars are smaller than the symbols).

Extended Data Figure 4 Scatterplots of oxygen and carbon isotope ratios for the Paraíso stalagmites.

a, Relationship between the δ18O and δ13C data for Holocene Paraíso stalagmites. b, As in a, but for glacial Paraíso samples.

Extended Data Figure 5 Estimation of monthly water balance in the region.

a, Monthly averaged precipitation (solid dots and triangles) and actual evapotranspiration (AET, open dots and triangles) over Belém and Manaus. We used the water-balance model44 as implemented in the US Geological Survey (USGS) Thornthwaite model45 to calculate monthly AET. b, As in a, but for LGM conditions. We assume that the cave temperature was ~21 °C during the LGM. Rainfall in the region was ~60% of today’s in each month, as calculated in Extended Data Table 1. The LGM and present-day patterns are essentially the same.

Extended Data Figure 6 Comparisons of the Paraíso record with local insolation curves.

The cave δ18O record spans about 46,000 years, long enough to cover two precessional cycles. However, no obvious correlation can be observed between the cave record with local insolation in the months of January (blue), April (cyan), July (dark blue) and October (dark cyan). Insolation data are from ref. 50.

Extended Data Figure 7 Comparisons of the Paraíso δ18O record with atmospheric concentrations of greenhouse gases.

Changes in atmospheric CO2 (blue) and CH4 (dark blue) concentrations are recorded in Antarctic ice cores51,52.

Extended Data Figure 8 Comparisons of the Paraíso cave record and ice-core records.

a, The Paraíso δ18O record is compared with ice-core records from Greenland53 (dark blue; North Greenland Ice Core Project (NGRIP)) and from Antarctica54 (blue; EPICA Dronning Maud Land (EDML) Ice Core) during the time interval from 25 kyr bp to 45 kyr bp. The NGRIP ice-core data are plotted in the Antarctic ice-core chronology 2012 (AICC12) timescale55, which is identical to the annual-layer-counted Greenland ice-core chronology 2005 (GICC05) timescale56 for the studied time interval. The EDML ice-core data are plotted in the AICC12 age scale55. D/O events are marked on the NGRIP record. The strong correlations between the Paraíso record and the ice-core records confirm the existence of rapid air–sea interactions between the high latitudes and the tropics on millennial timescales57,58, probably through the so-called bipolar seesaw mechanism59. b, As in a, but the Paraíso record is compared with ice-core records from Greenland53 (dark blue; NGRIP) and from Antarctica25 (blue; West Antarctic Ice Sheet Divide Ice Core (WDC)). The NGRIP and WDC data are plotted in the West Antarctic Ice Sheet Divide (WD) 2014 timescale25. The slightly enhanced correlations between the Paraíso record and the ice-core records, albeit visually, support the chronological method adopted in ref. 60. VSMOW, Vienna standard mean ocean water.

Extended Data Figure 9 Paraíso δ13C record.

Contrary to the stalagmite δ18O record, the Paraíso δ13C record does not show an obvious shift from the last glacial period to the Holocene. In fact, the δ13C value reaches as low as about −10‰ during the LGM, similar to the observed minimum value in the Holocene. This suggests that the type of vegetation in the region has not undergone dramatic changes, remaining dominated by C3 plants37,61. The rainforest in the eastern Amazon might have become an open forest when the precipitation decreased substantially during the LGM. However, it was not replaced by savanna or grassland—that is, it has not become dominated by C4 plants. The δ13C spikes were probably caused by individual air–water–rock interactions during calcite precipitation.

Extended Data Table 1 Calculations of water vapour loss over the eastern Amazon

Supplementary information

Supplementary Table 1

This table contains the Paraíso speleothem U-Th dating results, and δ18O and δ13C data. (XLS 458 kb)

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wang, X., Edwards, R., Auler, A. et al. Hydroclimate changes across the Amazon lowlands over the past 45,000 years. Nature 541, 204–207 (2017). https://doi.org/10.1038/nature20787

Download citation

Further reading


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.


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