Climate science

Water's past revisited to predict its future

A reconstruction of 1,200 years of water's history in the Northern Hemisphere, based on proxy data, fuels the debate about whether anthropogenic climate change affected twentieth-century precipitation. See Letter p.94

Predicting future climate is one of science's great challenges1. Climate models are key to these predictions, but such models must be validated through comparison with, and integration of, measured or proxy-based climate data2. On page 94 of this issue, Ljungqvist et al.3 take on this challenge by reconstructing the past 12 centuries of water history in the Northern Hemisphere from proxy records. Their reconstruction for the twentieth century diverges sharply from climate-modelling results, and calls for greater development of water-sensitive proxy data and their integration into modelling efforts.

Estimating future water availability in a warming world is crucial, for various socio-economic and ecological reasons4. To compensate for the absence of measured data for future times, climate models are often assessed using twentieth-century data and, in many cases, inferred data from periods that pre-date climate measurements. Palaeoclimate data are derived from proxies of measured data, through the analysis of climate-sensitive archives such as ice cores, lake sediments and mineral deposits in caves (Fig. 1).

Figure 1: Calcite columns in Lechuguilla Cave, New Mexico.

Michael Nichols/National Geographic Creative

Cave mineral deposits can provide proxy data on past precipitation patterns. Ljungqvist et al.3 have reconstructed the water history of the Northern Hemisphere for the past 1,200 years, based on proxies, including mineral deposits.

Ljungqvist and colleagues undertook a formidable task in this area of research. They compiled various water- and temperature-sensitive proxy data for the past 1,200 years, assessed age controls for the data, compared each data set across regions and continents, and then compared their proxy results with previously reported climate-model results. Their reconstruction of hydroclimate is generally consistent with model results for the period concerned — for example, it indicates a higher percentage of wetter-than-average land (relative to the average before the twentieth century) during the ninth to eleventh and the twentieth centuries, and the opposite during the twelfth to nineteenth centuries. Persistent precipitation dipoles (north–south patterns of precipitation anomalies that have opposing signs), such as that observed now in the western United States5, are also evident. However, the data do not capture the climate trends6 known as the Little Ice Age and the Medieval Climatic Anomaly.

Perhaps the most notable finding is that the proxy-based reconstruction diverges from model outputs for the twentieth century. Specifically, the proxy results indicate no difference in twentieth-century water dynamics compared with the pre-industrial era, whereas model results indicate the development of a substantial bimodal distribution in precipitation between extreme wetness and extreme dryness — dry gets drier and wet gets wetter (DDWW).

The DDWW idea is not new7. This concept is based largely on the expectation that higher global temperatures will intensify the water cycle, increasing the average rainfall and/or rainfall intensity, as well as amplifying net evaporation. This intensification is related to the Clausius–Clapeyron equation7, which states that moisture content in the lowest part of the atmosphere will rise with higher temperatures. Already-wet regions might therefore receive more-frequent and/or more-intense precipitation. Conversely, dry regions will become drier as warmer temperatures expand the subtropical dry zones and intensify areas of high pressure (low precipitation). But this picture is an oversimplification and effects are likely to manifest in unforeseen ways8.

Whether twentieth-century water-cycle intensification and the associated DDWW response have been detected is still controversial9. The divergence of Ljungqvist and colleagues' hydroclimate reconstruction from model results for the twentieth century certainly adds fuel to the fiery debate. Do their results invalidate current predictive models? Certainly not. But they do highlight a big challenge for climate modellers, and present major research opportunities both for modellers and for climate scientists who work with proxy data10.

For example, the authors highlight the continued importance of combining proxy-based results with climate models. This is more easily said than done. Not every site contains the same types of proxy, and the response of sources of proxy data to climate can change over time. Moreover, the climatic sensitivity of proxy-data sources varies from type to type and site to site.

To address these limitations, palaeoclimate scientists must standardize best practices for proxy sampling, reconstructions, dating and statistical analyses. One priority should be to select key sites for proxy sampling and analysis. For example, are there crucial geographical gaps that require attention and that strike an urgent socio-economic or ecological chord4,11? Fortunately, several initiatives are already focusing on improving efforts to integrate models and proxy results, for example the Paleoclimate Modelling Intercomparison Project and the Past Global Changes project.

Ljungqvist et al. were, of course, constrained by the data available for analysis — indeed, their efforts reveal a shocking lack of data. For example, Figure 1 of their paper3 highlights the vast geographical gaps between proxy sites. Immense areas of the Northern Hemisphere still require exploration for proxy development, many in highly populated regions. The current analysis should therefore be revisited as proxy records from these regions become available.

Nevertheless, this research is a crucial first step in the use of models and proxy data to reconstruct and explain the history of water — and not just of temperature change — through time. Future research efforts should improve, test and extend Ljungqvist and colleagues' results. Global warming will undoubtedly change Earth's water cycle, so the more that is known about the cycle's past behaviour through proxy records, and the better those changes can be modelled, the more confidence we will have in predictive models as we forge ahead into the twenty-first century.Footnote 1


  1. 1.

    See all news & views


  1. 1

    Kirtman, B. et al. in Climate Change 2013: The Physical Science Basis. Working Group I Contribution to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (eds Stocker, T. F. et al.) 953–1028 (Cambridge Univ. Press, 2013).

  2. 2

    Ault, T. R., Cole, J. E., Overpeck, J. T., Pederson, G. T. & Meko, D. M. J. Clim. 27, 7529–7549 (2014).

  3. 3

    Ljungqvist, F. C. et al. Nature 532, 94–98 (2016).

  4. 4

    Conway, D. et al. Nature Clim. Change 5, 837–846 (2015).

  5. 5

    Wise, E. K. Geophys. Res. Lett. 37, L07706 (2010).

  6. 6

    Mann, M. E. et al. Science 326, 1256–1260 (2009).

  7. 7

    Trenberth, K. E., Dai, A., Rasmussen, R. M. & Parsons, D. B. Bull. Am. Meteorol. Soc. 84, 1205–1217 (2003).

  8. 8

    Cook, B. I., Smerdon, J. E., Seager, R. & Coats, S. Clim. Dyn. 43, 2607–2627 (2014).

  9. 9

    Prein, A. F., Holland, G. J., Rasmussen, R. M., Clark, M. P. & Tye, M. R. Geophys. Res. Lett. 43, 1272–1279 (2016).

  10. 10

    Schmidt, G. A. J. Quat. Sci. 25, 79–87 (2010).

  11. 11

    Myers, N., Mittermeier, R. A., Mittermeier, C. G., da Fonseca, G. A. B. & Kent, J. Nature 403, 853–858 (2000).

Download references

Author information

Correspondence to Matthew E. Kirby.

Related links

Related links

Related links in Nature Research

Climate change: Human influence on rainfall

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kirby, M. Water's past revisited to predict its future. Nature 532, 44–45 (2016).

Download citation

Further reading

  • European warm-season temperature and hydroclimate since 850 CE

    • Fredrik Charpentier Ljungqvist
    • , Andrea Seim
    • , Paul J Krusic
    • , Jesús Fidel González-Rouco
    • , Johannes P Werner
    • , Edward R Cook
    • , Eduardo Zorita
    • , Jürg Luterbacher
    • , Elena Xoplaki
    • , Georgia Destouni
    • , Elena García-Bustamante
    • , Camilo Andrés Melo Aguilar
    • , Kristina Seftigen
    • , Jianglin Wang
    • , Mary H Gagen
    • , Jan Esper
    • , Olga Solomina
    • , Dominik Fleitmann
    •  & Ulf Büntgen

    Environmental Research Letters (2019)


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.