China cannot rob us of Brahmaputra

 Brahmaputra, Assam, China, Nilanjan Ghosh, Water Governance, Teesta, ORF, Strategic Studies

Brahmaputra in Assam

Media has long been reporting on China’s plan of northward rerouting of the Brahmaputra waters (known as Yarlung Tsangpo in Tibet) from the Tibetan borders through constructions of dams. This has emerged as a prime point of contention with China-India strategic relations. Brahma Chellaney, one of the foremost strategic thinkers of India, described the Chinese design of taking control over Brahmaputra water as “most dangerous”. The fear of drying up of the Brahmaputra has become widespread in Indian public psyche, especially in Assam. This hypothesis of perceived fear is termed in this article as the “Brahma hypothesis”.

The growing water demand in Tibet and the option available in principle to China of building water storage and transfer projects on the Yarlung have given birth to such fears in India.


The apprehension is this can affect Bangladesh further downstream. The concern has aggravated with the news of Chinese plans to build a 1,000-km-long tunnel to divert water from the Brahmaputra River in Tibet to the parched Xinjiang region. It has been reported in sections of the media that the perceived Chinese threats to divert the river’s water prompted the Centre to call an inter-ministerial meeting recently to discuss proposed projects on Brahmaputra.

Amidst the clamour about Chinese projects on Brahmaputra, there has hardly been an objective data-based analysis of the popular “Brahma hypothesis”. These contentions deserve to be examined through data, hydrological regimes, upstream interventions and their downstream implications.

Identifying the flow

The Brahmaputra is identified as the flow downstream of the meeting of three tributaries, namely Luhit, Dibang and Dihang, near Sadiya. The link of Brahmaputra with Yarlung Tsangpo, which originates from the Angsi glacier near Mt. Kailash, was discovered rather recently. Out of the total length of the Brahmaputra of 2,880 km, 1,625 km is in Tibet flowing as Yarlung Tsangpo, 918 km is in India known as Siang, Dihang and Brahmaputra and the rest 337 km in Bangladesh has the name Jamuna till it merges into Padma near Goalando.

As a trans-Himalayan tributary, Yarlung is substantially fed by snow and glacial melts, in addition to rainfall. The normalised melt index (defined as the volumetric snow and glacier upstream discharge divided by downstream natural discharge) of the Brahmaputra is merely in the range of 0.15-0.2, signifying that snow and glacial melt, the main source of run-off in the Tibetan region, contributes negligibly to the total flow.

Making the discourse realistic

The Tibetan region lies in the rain shadow with the Himalaya acting as the barrier to the rain-laden monsoon. The annual precipitation in the trans-Himalaya Tibet averages about 300 mm annually. As the tributaries cross the Himalayan crest line, the annual average precipitation reaches about 2000 mm.

A very large component of the total annual flow of Brahmaputra is generated in the southern aspect of the Himalaya in India by tributaries from Buri Dihing in the East to Teesta in the west.


Data published by Chinese scholar Jiang and team show that the total annual outflow of the Yarlung River from China is estimated to be about 31 BCM while the annual flow of Brahmaputra at Bahadurabad, the gauging station near the end of the sub-basin in Bangladesh, is about 606 BCM. These figures do not support the linear thinking that the flow in a river is proportional to its length inside a country.

Further, while the peak flows during monsoon at Nuxia and Tsela Dzong in Tibet, a measuring station at the great bend in the Tibetan plateau, are about 5,000 and 10,000 cumecs, as presented by Vijay Singh and colleagues, the peak flow at downstream Guwahati is around 40,000 cumecs and the one at Bahadurabad in Bangladesh is approximately 50,000 cumecs.

During the lean season, the flow in Nuxia, as identified from a hydrograph given in Rivers and Lakes of Xizang (Tibet) (in Chinese), is 300-500 cumecs, while the one at Pasighat is to the tune of 2000-odd cumecs, the one at Guwahati is around 4000-odd cumecs, and Bahadurabad is about 5000 cumecs, all these being peer-reviewed data.

This data shows that the Brahmaputra gets fatter and mightier as it flows further downstream. This is more so because of the flow contribution of the various tributaries like Dibang, Luhit, Subansiri, Manas, Sankosh, Teesta to name a few.


This can be noted from the fact that at Guwahati (Pandu), the percentage annual yield of the main river course from Pasighat is barely 34 per cent, while the tributaries like Dibang, Luhit, Subansiri, as also the tributaries joining between Pasighat and Guwahati contribute the remaining 66 per cent. Further downstream, the mainstream contribution diminishes further.

Another concern relates to the impact of the projects on the sediment flow. Can water diversion affect sediment flow? The flow volume and discharge in the Yarlung River is not sufficient to generate and transport carry the very large sediment load as in prevalent in the downstream Brahmaputra.

The annual suspended sediment load near Nuxia in Tibet is around 30 million metric tonnes, (as suggested in a 2016 volume titled River Morphodynamics and Stream Ecology of the Qinghai-Tibet Plateau by Wang and colleagues), which is miniscule as compared to same load measured as 735 million metric tonnes at Bahadurabad.

Therefore, the large amount of suspended sediment load that gets deposited in the downstream to form a fertile Jamuna floodplain cannot be carried by the Yarlung-Tsangpo stretch. It is created further downstream in India, where precipitation is almost 12 times higher than the rain shadow Tibet.

A popular hypothesis

Prima facie, it can be said that the impacts of water diversion (or even hydropower like the Zangmu Dam) in the Yarlung-Tsangpo cannot have substantial impact on the flow regime in the Indian boundary, especially in the Assam floodplains and Bangladesh. The concern of many in India has been based on the perception that structural interventions always reduce downstream flows, which, in case of Brahmaputra, is not true.

Based on the hydro-meteorological data, it seems highly improbable that a cloudburst can occur in the rain-shadow Tibet so as to cause floods in Assam.


Therefore, the “Brahma hypothesis” or the myth spread in the media does not stand the test posed by scientific data and knowledge. Informed science should inform public perceptions, policy, hydro-politics, and water governance, rather than jingoistic emotions or linear, reductionist logic.

This commentary originally appeared in Business Line.




Nilanjan Ghosh