skip to primary navigation skip to content

Jean-Philippe Baudouin

Jean-Philippe Baudouin

Former PhD student


I have been working since January 2020 as a postdoctoral research at Heidelberg University, department of Environmental Physics. I am now part of the Stacy group in the project Palmod (cf. my new webpage).

I submitted my PhD in September 2020, and successfully passed the viva.

ORCID: 0000-0002-0219-8634

Also: ResearchGate, LinkedIn


My PhD was conducted as part of the TwoRains project, which investigates the resilience and sustainability of the Indus Civilisation (South Asia, about 4.000 years ago) in the face of a changing climate.

PhD Thesis: A modelling perspective on precipitation in the Indus River Basin: from synoptic to Holocene variability


The Indus River civilisation was the first urban society in South Asia. Its demise, ca. 4000 years ago, might have been caused by environmental factors, such as recurrent drought conditions. Yet, palaeoclimate archives present a fragmented picture of the climate at that time. This thesis explores the potential value of global climate models to further interpret the archaeological context. Precipitation variability is investigated at various timescales due to their different impact on human societies and their importance in the climate system: the synoptic scale, the seasonal cycle, the inter-annual to multi-decadal variability, and the multi-millennial Holocene trends. The precipitation from several climate model simulations is evaluated at each of these timescales.

Indus River Basin precipitation is first explored in observational datasets. This study highlights the quality of ERA5 reanalysis, which is used as a reference in the subsequent chapters. Statistical tools show that more than 80\% of the precipitation in the Upper Indus Basin is related to cross-barrier moisture transport along the Himalayan foothills. The climate models analysed (IPSL-CM6-A, MRI-ESM2-0 and GISS-E2-1-G) generally reproduce this process well, but the seasonality of cross-barrier moisture transport is biased, resulting in precipitation biases: the main wet season, the summer monsoon, is shorter and significantly dryer, while the second wet season, in winter, is longer and more active.

The link between winter precipitation, cross-barrier moisture transport and Western Disturbances is further explored, first in ERA5 and then compared to IPSL-CM6-A model output. Sub-daily resolution is needed to determine the origin of the positive precipitation bias in winter. This bias is related to differences in the atmospheric circulation associated with Western Disturbances. The stronger Subtropical Westerly Jet is also key to understand the precipitation overestimation. Centennial to millennial-scale precipitation variability is more difficult to evaluate due to the short length of observations and the paucity of climate records, but the results suggest that inter-annual variability in the IPSL climate model family is overly dominated by atmosphere-only processes, with a potentially large impact of the precipitation response to external forcings.

In addition to biases in the representation of mean precipitation and synoptic to inter-annual variability, climate models are also limited by the representation of other internal processes such as ocean circulation, and vegetation-dust aerosols, as well as by the uncertainty in some external forcings such as volcanic eruptions and the solar activity. Hence, at the present state, past-climate simulations do not provide the nuanced information of precipitation changes and variability that is needed to understand impact of precipitation variability on archaeological contexts, and will not do so until significant breakthroughs are achieved. Nevertheless, climate models remain a powerful tool for climatologists to investigate large-scale processes that can eventually better characterise climate variability.