Innovative new work from the ERC-funded TwoRains project reconstructs more than 8,000km of relic rivers, creating a step change in our current knowledge of the Ancient Indus hydrology and the environmental conditions in which this Bronze Age civilisation operated.
 |
| Multitemporal composite of vegetation indices in the area between Patiala and Bathinda [Credit: H. Orengo] |
Innovative new work by the ERC funded TwoRains project, which is based in the Department of Archaeology and the McDonald Institute for Archaeological Research at the University of Cambridge, has been addressing this issue as part of its objective to study the influence of water availability and climate change in the collapse of ancient civilisations. A journal paper by Hector A. Orengo and Cameron A. Petrie detailing the first results of this research has been selected as the cover story of the last issue of the journal Remote Sensing.
The team at Cambridge has been able to reconstruct the river network of an area equivalent to Austria or the Czech Republic. This has resulted in the mapping of more than 8000 km of relic rivers that disappeared at least in part due to climate change, but also a range of other factors.
The researchers employed more than 1700 multispectral satellite images (covering not just the visible wavelengths but those that are not visible to the human eye such as infrared), from early ones acquired in the 1980s up to 2013. These images were analysed using Google Earth Engine’s cloud high performance computing infrastructure and algorithms purposely developed to discriminate palaeorivers. These use the seasonal contrast between vegetation, soils with different moisture levels, bare terrain and soil mineral content to overcome visibility issues related to cloud cover, seasonal cultivation patterns, recent crop selection, extensive irrigation and long-term land-use patterns.
The results create a step change in our current knowledge of the Ancient Indus hydrology and the environmental conditions in which this Bronze Age civilisation operated. They also provide important insights on the influence of climate change in its eventual demise.
The code employed for the development of this study has been publicly released and is available as supplementary material to the published article. By doing so, the authors hope that their innovative method can be employed by other researchers tackling climate change or interested in past environmental conditions.
‘The methods we have employed to reconstruct past hydrology over such a large area have only recently been available to us. The research potential of the analysis of satellite Big Data collected during the last 40 years in combination with super-computing resources has only started to be understood’ said Hector A. Orengo lead author of the publication.
Cameron A. Petrie, co-author of the paper and director of the TwoRains project, remarked ‘It is wonderful to see the realisation of an approach that was effectively wishful thinking a year ago. The new methods that we have developed have the potential to transform the way archaeologists reconstruct a wide range of ancient environments'.
Source: University of Cambridge [August 02, 2017]