Discussing China’s Dams on the Mekong

Stimson scholars consider China's dams on the Mekong, evidence of their impact downstream, policy motivations, and more

By  Brian Eyler  • Yun Sun

In China

Ed note — This discussion between Yun Sun, Co-Director of the East Asia Program and Director of the China Program, and Brian Eyler, Director of the Southeast Asia Program, references the April, 2020 feature story, How China Turned Off the Tap on the Mekong River and comes in the context of widespread reporting on low Mekong River water levels in downstream countries in 2019.

Yun Sun

In the discussion of the water holding by the Lancang cascade during the 2019 wet season, one factor that has been highly disputed is the water volume from the upper stream. According to official Chinese report, water flow was 40% higher during the dry season in April and May of 2019, but was 30% lower than the annual average during the wet season. [Ed note: Linked article is in Chinese. See English translation.] Because the price of electricity generated during the dry season is three times the price during the wet season, the Chinese power company chose to maximize the power generation before June, suppressing the water level of Xiaowan and Nuozhadu to unprecedented low levels, 1170 meters and 765 meters, close to or at their dead storage levels, 1166 meters and 765 meters, respectively. At these levels, the Chinese official claim of “historically lowest levels” for the reservoir volumes of these dams is correct, although an astute analyst would point out the hypocrisy of that claim as the suppression of water to such low levels was a Chinese decision to boost profits. How are we taking these factors into consideration?

Brian Eyler

These factors help lay out a much fuller picture of China’s upstream operations than what was previously known. Relevant stakeholders in China did not share the information contained in the CSG article with downstream countries nor the Mekong River Commission.  The CSG article outlines the steps CSG and Huaneng Hydrolancang took in early 2019 that created conditions for an unprecedented restriction of water held behind China’s dams. These decisions led to great economic benefit for CSG and Hydrolancang under the auspices of clean energy development but led to severe costs downstream. There are several cost components to address. Namely, how those restrictions exacerbated drought conditions along the mainstream in the lower basin and how the severe alteration of natural flow impacted important ecosystem processes and fisheries migration downstream. Further, the unnatural release of extra water downstream during the dry season has been documented to cause considerable economic damage to communities in northern Thailand dry season releases are also documented to have profound effect on the geological integrity of the Mekong’s floodplain in Cambodia and Vietnam.  With knowledge of dry season electricity generation, market prices and sales volume, it is now possible to conduct a basin-wide cost benefit analysis of China’s decision to reconfigure its upstream cascade operations throughout the entity of 2019.

Yun Sun

This gets to the issue of the intention of China’s holding of water. Without detailed analysis on the conditions and realities of Chinese actions, it is not justified to draw conclusions based on speculations over China’s geopolitical aspirations or the capriciousness of the water elites. The reality in 2019 was that the dam operations released extreme amounts of water during the end of dry season (particularly in March and April), therefore they needed a large volume of the natural flow in the wet  season to replenish the level of the reservoirs. Since the snowmelt cycle in the Himalayas occurred earlier than usual, and this water was released to  profit  from  hydroelectric production in April (much of the snow melt occurred in the upper basin in April, before the traditional  wet season). Chinese operators used much of the flow in the wet season to replenish the level of the reservoirs that were depleted in the prior months. They stated that flow was reduced by 30% during the wet season, however, it appears the Lancang cascade chose to store the inflow to bring the water level back to normal. This economic calculation offers the most logical explanation unless there is evidence to suggest the Chinese used the stored water for another purpose. However, these Chinese data about the cascade’s operations in 2019 do not support the claim that the wet season water flow into the cascade was higher than usual or the claim that the cascade restricted more water from the downstream. Since the conclusion was that China was holding back an abnormally high level of water during the wet season, (which need substantiation at the first place), what did the Chinese use the water for?

Brian Eyler

The water was held back to recharge the Xiaowan and Nuozhadu reservoirs after generating a higher than average amount of electricity during the 2019 dry season. The article describes how China’s Lancang hydropower cascade managers strategized in early 2019 to optimize power production by developing monthly targets of reservoir reductions during the dry season to generate a maximum amount of electricity. This planning intentionally drove reservoir levels at Xiaowan and Nuozhadu to extremely low levels close to their dead storage. When the wet season came, electricity production tapered off to allow the reservoirs to recharge. Satellite image analysis of these reservoir levels through the 2019 dry season (January-May) confirm a steep drop in the two reservoirs levels during that period. Also the gauge measurements at Chiang Saen during March and April detected extremely high levels of flow during the end of the dry season. Moreover, satellite images also confirm those reservoirs recharged to near normal levels during the 2019 wet season and in order to do so, very little water was released into the downstream during this time. Remote sensing from several datasets show higher than normal precipitation and wetness in the upper basin during the wet season, while the lower basin was experiencing drought. It appears that the excessive release of water during the first months of 2019, left the dam operator in a position to severely restrict the amount of water that would be available downstream, since their agenda was to recharge reservoirs that had been deleted in the prior months. The advantage of regulating the river in this unnatural flow regime optimized the profit of the hydroelectric production. The Chiang Saen gauge confirms abnormally low levels of water flowed out of China during the wet season as those reservoirs recharged.

Yun Sun

In the era of heightened competition with the U.S. for strategic influence regionally and globally, the argument that China deliberately holds water and antagonizes the Mekong countries for no good reason fits within the accusatory narrative, but it does not hold water with the given evidence or the logical deduction. Sophisticated observers tend to draw nuanced distinctions between the Chinese government and Chinese companies, and between the foreign political policy apparatus and the foreign economic policy apparatus, instead of drawing blanket conclusions about China as a monolithic player. In this specific case of Lancang-Mekong hydropower development, what are the internal conflicts among different groups and what internal intervention and mediation mechanisms are there to resolve the conflicts and rein in various players?

Brian Eyler

I have not accused China of antagonizing downstream countries nor any kind of nefarious actions. My team’s interpretation of the Eyes on Earth’s findings stated that China’s water 2019 water restrictions delivered devastating effects downstream. My team explored several different motivations drawn from past observation and knowledge of the way China has operated dams on the Mekong and other rivers as well as derived from the way China’s water elite have talked about upstream water use and hydropower operations. We can only speculate, but the operational manipulation of the river’s hydrological cycle vis a vis Huaneng Hydrolancang and China Southern Grid could have created tension among those in China who have pledged to release water during times of drought downstream – a pledge frequently heard from China’s Lancang Mekong Water Resources Center in Beijing. There could have been elite in Beijing who sought to relieve drought downstream during the 2019 wet season but these calls could have been trumped by Huaneng and CSG’s operational preferences and their perception of market opportunities into 2020. Additionally, since most of the available water in the cascade is stored in the Xiaowan reservoir and Nuozhadu reservoir, if those reservoirs had reached their dead storage levels, there simply would have been zero to little water to share. Dead storage is located below a dam’s outlet and is technically difficult to release since it cannot be released by gravity.

Yun Sun

Interpretation of satellite imagery has been a highly controversial matter because different analysts can draw entirely different conclusions based on the same imagery due to different criteria, analytical backgrounds, and specialized knowledge. The challenge of accuracy is tremendous. As mentioned by the Mekong River Commission, “reservoir water surface from satellite imagery helps understand how Upper Mekong dams are managed and operated….However, this requires careful selection of thresholds to obtain sufficient accuracy.” With this in mind, does satellite imagery allow for different interpretations? If so, how conclusive will the study based on satellite imagery could be? What level of confidence interval can be established and how?

Brian Eyler

Reliable data shared with downstream stakeholders in a regular and transparent manner obviates the need for satellite imagery analysis. Satellite imagery can only supplement analysis. Now equipped with the knowledge of Huaneng and CSG’s 2019 “grand strategic platform” as described by an abbreviated English language translation of the CSG article linked above, satellite imagery fully corroborates a narrative that the Xiaowan and Nuozhadu reservoirs dropped to historically low levels in order to generate a large amount of electricity in the 2019 dry season and gradually filled throughout the 2020 wet season in Yunnan. Satellite images also show the reservoirs dropping during the 2020 dry season (now that it has concluded) but not to the extent of 2019. Independent satellite observations can provide objective data, which allows transparency of data between all parties within the riparian basin. In conclusion, collaboration using independent, quantitative and near real time data is essential for cooperative arrangements for water resource monitoring and distribution. This also allows for an effective drought and flood management program.  

Ultimately more information (as such provided through satellite data) can better inform relevant stakeholders (internal and external to China and improve outcomes; This has been proven with reporting on China’s air quality; South China Sea military operations (CSIS AMTI), and forced detention centers for Uyghurs in Xinjiang. 

Yun Sun

There are also disagreements over the best way to tackle China’s secrecy, refusal to share information and to engage China in productive conversations over the management of the Lancang-Mekong River moving forward. Fair criticism is warranted, but many experts argue that “naming and shaming” is more of a media strategy instead of a constructive approach to facilitate cooperation. This is particularly true when the naming and shaming seems to be the end rather than the means themselves. In your view, what is the most constructive way forward to help the Mekong countries engage China and forge cooperation?

Brian Eyler

Equipping Mekong countries with applications that report near-real time conditions of both the river’s hydrological cycle and upstream dam operations will provide those countries with data critically important to downstream decision making as well as their own water policy toward China. From an institutional perspective, the Mekong River Commission (MRC) is the best fit organization to house and recognized this data. The MRC should be able to objectively weigh the validity and value or such monitoring applications free from political interference. However, there are tens of millions of people in Mekong Basin that rely on its natural resource base for their livelihoods. The level of involvement of these stakeholders in influencing policy outcomes varies on a country by country basis, but equipping these stakeholders with objective, evidence-based findings can drive citizen science activities that lead to a more sustainable Mekong. Some stakeholders like communities in the Golden Triangle those on the Tonle Sap and in the Mekong Delta which can be impacted on a daily or seasonal basis by upstream dam operations have every right to be equipped with real-time information just like they are able to check the weather forecast on a frequent basis.

I understand how these actions might appear as “naming and shaming” particularly to Chinese stakeholders. However, this is not the goal of the investigation. My full intention is to promote fair, transparent and cooperative arrangements among all sectors of society, who are dependent on the integrity of the Mekong river. Furthermore, and as stated above, this information is critically important to downstream stakeholders throughout the basin who are impacted by the devastating effects of China’s dam operations.

Recent & Related

Chapter
Benjamin Creutzfeldt • Parsifal D’Sola Alvarado
Resource
Brian Eyler • Regan Kwan

Subscription Options

* indicates required

Research Areas

Pivotal Places

Publications & Project Lists

38 North: News and Analysis on North Korea