skip to primary navigation skip to content


Project overview

The Indus Valley is an area of contrast, with winter rains and summer monsoon precipitation overlapping in an otherwise arid environment (Fig 1). Hence, the ancient Indus population appear to have adapted to the diversity of the environment. Their success at this is evidenced by the rise of the Indus Civilisation (c. 4600-3900 BP), which was home of the first cities in South Asia (Fig 2). The TwoRains project has been designed to give new insight into the Indus Civilisation by investigating questions of sustainability and resilience.

While the Indus Civilisation spanned an immense area, only a few urban centres are known, and only five appear to have been cities: Dholavira, Mohenjo-daro, Ganweriwala, Harappa, and Rakhigarhi (Fig 1). Indus society remains complex nevertheless, with distinctive forms of urban organisation manifested in planned streets, coordinated drainage systems, fired and mud brick buildings, and a range of sophisticated technologies, including copper-based metallurgy, standardized weights and measures, and intricately engraved stamp seals. It persisted until c.3900 BP, by which time its cities were largely depopulated and smaller rural settlements predominated. This de-urbanisation process appears to have coincided with a significant change in rainfall, and especially the occurrence of a centennial weakening of Indian summer monsoon around 4100 BP. Such an event would have threatened food security and therefore triggered dramatic changes in water management and subsistence strategies, of which migration and de-urbanisation could have been a result.

To test different hypotheses regarding the relationship between climate change and the Indus civilisation, the TwoRains project is continuing the extensive program of environmental and archaeological analysis begun by the Land, Water, and Settlement project (2007-2014). TwoRains will identify the importance of varied agricultural and food production techniques, and involves satellite imagery analyses, on the ground survey, climate proxy analysis and environmental reconstructions based on computational modelling. This latter element will lead to agent-based modelling simulations, which will be performed to assess the sustainability of Indus society’s practices. All this work will finally sharpen understanding of how people in the Indus area succeeded of failed to reshape their landscape and build resilient new ways of life in the face of a changing climate

Our role

At the department of Geography, we are tackling the question of rainfall variability in the Indus River catchment. The area sits at the edge of the Indian summer monsoon domain, and today receives a high but heterogeneous amount of precipitation from July to September, leading sometimes to widespread floods. However, in contrast to the rest of India, it also receives rainfall in winter, due to the interaction with western disturbances. This winter precipitation produces snowfall at higher altitudes in January and February, which melt later in the season concomitant of an increasing risk of thunderstorms and flash floods.

There is clear evidences that centennial droughts occurred is the past, especially between 4.5 and 3 kyr BP, which is sometimes reported as the 4.2 ka event. Interestingly, this ‘event’ overlays a decreasing trend in the monsoon intensity, from 6 to 3 kyr BP, which appears to be related to changes in orbital forcing. However, questions remain concerning the possible drivers of the centennial droughts and the influence of the different orbital forcing over Asian summer monsoon and winter rain

To address those questions, our work aims first to push forwards the understanding of the current precipitation regime and its variability in the Indus catchment area, and then to transpose it to our period of interest. Specific focus is put on the large scale interaction between mid-latitude and tropical features, and on the different oceanic drivers that might explain the yearly or decadal variability. Existing millennium and multi-millennium numeric experiments will then be used to understand the impact of the orbital forcing on the precipitation regime and to investigate any multi-decadal droughts possibly produced. Finally, possible feedback of the land cover will be examined.

The results of this study will allow the analysis of the water supply variability and the evolution of floods and droughts occurrence, which will be of considerable interest for the archaeologists who try to understand the rise and demise of the Indus Civilisation. Furthermore, better theoretical understanding of winter and summer rainfall variability in the area will help to evaluate their possible evolution in a changing climate.

Figure 1

Figure 1: Relief in the Indus Valley and its surrounding. The Indus catchment area in drawn in dark blue, whereas its main rivers are highlighted in light blue. The 300 mm contour line of the winter (summer) precipitation is drawn in white (black), with the white (black) shaded area representing the zone of higher precipitation.

Figure 2

Figure 2: Composite satellite image of the same area as Figure 1, with the same representation of the hydrological network. The red line represent the limit of the extend of the Indus Civilisation (c. 4600-3900 BP), with its 5 main cities.