EGU2020-16898, updated on 12 Jun 2020
EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Rainfall seasonality changes in northern India across the 4.2 ka event

Alena Giesche1, Sebastian F.M. Breitenbach2, Norbert Marwan3, Adam Hartland4, Birgit Plessen5, Jess F. Adkins6, Gerald H. Haug7, Amanda French4, Cameron A. Petrie8, and David A. Hodell1
Alena Giesche et al.
  • 1Department of Earth Sciences, University of Cambridge, Cambridge, UK
  • 2Department of Geography and Environmental Engineering, Northumbria University, Newcastle upon Tyne, UK
  • 3Potsdam Institute for Climate Impact Research, Potsdam, Germany
  • 4Environmental Research Institute, School of Science, Waikato University, Hamilton, New Zealand
  • 5Helmholtz-Centre Potsdam, German Research Centre for Geosciences, Potsdam, Germany
  • 6California Institute of Technology, Pasadena, USA
  • 7Max-Planck-Institute for Chemistry, Mainz, Germany
  • 8Department of Archaeology, University of Cambridge, Cambridge, UK

Despite intensive research efforts by archaeologists, geomorphologists, and palaeoclimatologists, the climatic and environmental changes accompanying the societal changes in the wider Indus/Thar region c. 4000 years ago remain puzzling. In particular, rainfall seasonality might be an important determinant for societal well-being. A major hurdle to a more detailed understanding of climate-human interaction is the relative scarcity of well-dated and highly resolved proxy records.

We present a multi-proxy record from aragonitic stalagmite DHAR-1 collected in Dharamjali Cave, Uttarakhand, India, that spans c. 1600 years between c. 4.25 and 2.6 ka BP.  The stalagmite has been dated with 13 U/Th dates with average uncertainties of <18 years (2σ). In addition to c. 1600 oxygen and carbon isotope samples, element ratios (X/Ca) were measured using high resolution μXRF and laser ablation ICPMS at 25 μm resolution.  

The DHAR-1 record represents the most precisely dated speleothem record to date from northern India, covering the mid-Holocene 4.2 ka BP event and the millennium thereafter. The attained sub-decadal to seasonal resolution allows robust assessment of both regional and local hydrological changes, and changes in amount and temporal distribution of summer and winter rainfall. 

The speleothem record reveals decadal-scale trends that can be related to changes in seasonality. The δ18O record reveals a 220-year period of weakened ISM from 4.2 to 3.98 ka BP. A contemporaneous increase in δ13C, and decrease in U/Ca, Ba/Ca, and Sr/Ca point to increased prior aragonite precipitation (PAP) resulting from increased aridity above the cave extending throughout the dry season. The ISM intensified after c. 3.7 ka BP while dry seasons remained dry, with a resultant increase in seasonality. Lower PAP after c. 3.4 ka BP can be interpreted as sign of reduced rainfall seasonality.

We compare the results with available records from the wider region, and discuss potential implications of the suggested changes in seasonality for agriculture-based societies.

How to cite: Giesche, A., Breitenbach, S. F. M., Marwan, N., Hartland, A., Plessen, B., Adkins, J. F., Haug, G. H., French, A., Petrie, C. A., and Hodell, D. A.: Rainfall seasonality changes in northern India across the 4.2 ka event, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-16898,, 2020

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 2 – uploaded on 28 Apr 2020
I just replaced a typo
  • CC1: Comment on EGU2020-16898, Addison Rice, 04 May 2020

    Hi! I'm very much interested in seasonality and just finished a little project in India. This is a really interesting record and some beautiful data!

    I'm wondering if it's possible to disentangle precipitation in the cold-dry season (Nov-Feb) from the hot-dry season (Mar-May) in the record.  Have you looked into this?

    • AC1: Reply to CC1, Sebastian F.M. Breitenbach, 04 May 2020

      Hi and thanks for your comment. that is exactly what we are after! the d18O signal is mainly a warm summer monsoon signal, bc the winter rains do not contribute to effective infiltration. The PAP proxies however tell us about the dry season intensity and length (unfortunately, we cannot distinguish intensity and length) - the drier and/or longer the dry season the more PAP. This way we can check on changes in seasonal contrast in moisture regime. however, looking more closely there are tricky bits to it. We are working on that... :-)

      • CC2: Reply to AC1, Addison Rice, 04 May 2020

        Very interesting! Thank you!

  • CC3: Comment on EGU2020-16898, Camilla Francesca Brunello, 04 May 2020

    @Seb. (to better articulate my chat question) I know d18O is an integrated signal of source+amount+season. But if I am correct, in your poster, lower d18O values are still interpreted as stronger ISM, while in this region you could obtain the same values simply with a “different” ISM, with different source contributions (ArabianSea vs BayogBengal vs Contiental recycling vs Westerlies) rather than amount changes. In particular, the role of continental recycling is crucial when changes in agricultural practices are involved, as irrigation is enhancing surface water availability and continental recirculation. Thus, the question is not only how seasonality affect agriculture but also, how agriculture may have affected relative contributions from different moisture sources. Meaning that some of the correlation may come from feedback loops. In the end we all agree that to disentangle the d18O signal it would be ideal to identify independent proxies which are exclusively (or at least predominantly) sensitive to either moisture source changes or rainout amount and then play around with statistics. Excellent session btw!

Presentation version 1 – uploaded on 28 Apr 2020 , no comments