Proceedings of the International Conference on Large-Scale Water Resources Development in Developing Countries, Kathmandu, October 1997, pp. CR25-34.

The Yamdrok Tso Hydropower Plant in Tibet:
A Multi-facetted and Highly Controversial Project

Petra Seibert (1) and Lorne Stockman (2)

(1) Institute of Meteorology and Physics, University of Agricultural Sciences,
Tuerkenschanzstr. 18, A-1180 Vienna, Austria; e-mail: seibert = mail boku ac at

(2) Free Tibet Campaign,
9 Islington Green, London N1 2XH; e-mail:

ABSTRACT: The recently completed Yamdrok Tso hydropower plant in Tibet is designed as a pumped storage plant with 90 MW capacity. It uses Tibet's largest freshwater lake (Yamdrok Tso) as a reservoir. Concerns have been raised whether the water will indeed be pumped back or whether the lake will be (partially) drained instead. In the case that water from the Yarlung river is being pumped into the lake to maintain its water balance, the different limnological properties of the two water bodies are likely to cause environmental problems. It is questionable whether Tibet needs a peak power generation facility of this size. Tibetans oppose the project also because Yamdrok Tso is considered a sacred lake according to their religious traditions.

1. Introduction

The 90 MW pumped-storage hydropower plant at Yamdrok Tso (Yangchouyong in Chinese) is the largest power station in Tibet [1]. The plant went into trial operation in July of this year and at time of writing was reported to be operating two of its four turbines.

The project has been subject to controversy from the very beginning, for a number of reasons which will be discussed in this paper. These are specifically hydrological, limnological, ecological, techno-economical and socio-political aspects which are all related to each other.

Yamdrok Tso is the largest freshwater lake in Tibet, situated 120 km south of Lhasa at the elevation of 4440 m a.s.l., with a catchment area of 6100 km2 and a surface of 658 km2 [2], figures vary slightly according to different sources. Its turquoise colour delights everybody travelling between Lhasa and Shigatse. The lake is almost a closed system; there is only a little natural runoff through a small tributary of the Yarlung-Tsangpo river, the inflow being balanced mainly by evaporation. The power plant is situated about 850 m below the lake at the Yarlung-Tsangpo river. Thus the plant can either produce net energy by draining the lake or, if the water is replenished in the pumping mode, it can shift base-load to peak-load. Both operation modes have given rise to environmental concerns, because the limnological characteristics of the lake and the river are very different so that pumping back river water is not a simple solution to all problems.

Besides the environmental impacts of the plant, the economic benefits are dubious. A further aggravation is constituted by the fact that the lake is considered sacred in the Tibetan religious traditions.

In October 1996, the International Union for the Conservation of Nature (IUCN) passed a resolution on the Yamdrok Tso project. The resolution was ratified by China but as yet no sign of implementation has materialised which, as the project is now in operation, is of concern.

2. Technical description of the project

The water is diverted from the lake by a 6 km long tunnel leading to a 3 km long penstock which ends at the turbine house near the Yarlung-Tsangpo (see Figure 1). The net head is 816 m. There are 4 Pelton turbines each capable of 3.16 m3/s throughput and able to produce 23.1 MW. According to the project designer - Chengdu Hydro-electric Investigation & Design Institute - the annual production is expected to be 84 GWh, corresponding to 930 h operation at peak power [3]. During the remaining 90% of the time, water shall be pumped back from the river. In the pumping mode, each of the 6 pumps will be able to pump 2 m3/s of water back into the lake. Because there is no substantial natural run-off which can be diverted, operation in this mode means that the plant will not have any net production of electric energy and can serve only for shifting base-load to peak power or as an emergency backup. The construction was plagued by a number of difficulties, including a tunnel collapse after the first trial operation, so that operation started in summer 1997 after 6 years of construction work; the planning dates back to the early 1980s.

Figure 1: Map showing the location of Yamdrok Tso (lake) and Yarlung Tsangpo (river) with the power plant site.

Figure 2: Yamdrok Tso as seen from the Space Shuttle.

3. Problems and corncerns

3.1 Hydrological dimension

The main hydrological concern is that the power station will lead to a drop in water level and consequently further ecological problems. This concern is based on the fact that the lake has no substantial natural runoff which can be diverted through the tunnels to the power plant. Thus, there are basically two possible operation modes:

a) Pumping back the same amount of water which is being drained during the operation periods. This implies that the plant does not produce any net electricity; on the contrary, due to the inherent losses it would consume net energy (maybe 20% of the generation). In this mode limnological impacts would occur which are discussed in section 3.2.

b) Draining the natural storage of water in the lake.

Of course, any mix of both modes is also possible. The official statement of the China Electricity Council is that operation will be in mode (a) [3]. However, there are concerns that this promise will not be kept. One indication for this is a paper by Shi Weiguang from the Sichuan Institute of Environmental Protection [2] which investigated operation mode (b) and concludes that adverse impacts will be negligible. Another basis for these concerns is the lack of large base-load generation facilities in Tibet which are typically operated in combination with pumped-storage facilities, providing the pumping energy during the low-demand time.

Let us now investigate the hydrological consequences of operation mode (b). It will cause the water level of the lake to fall with an initial rate w = Qout /A, with A being the lake's surface (620x106 m2) and Qout the artificial outflow. Qout is 43.16 m3/s with full operation of all turbines, according to [3]. Thus the upper limit of w would be 0.64 m/a. If, as a more realistic assumption, Qout would be reduced according to the average generation of 84 GWh/a (10% of the theoretical maximum generation capacity), the value expected according to [3], the water level would drop by 0.064 m/a (about 6 cm per year). This drop then leads to a reduction in size of the lake and accordingly to reduced evaporation. As the lake's water balance is presently maintained basically as an equilibrium of inflow (including precipitation directly onto the lake) and evaporation, the reduced evaporation will reduce the drop rate. Without detailed knowledge of the geometry of the lake's bed and the evaporation rate, the effect of this cannot be quantified. In [2], it has been calculated that a steady state would be reached with the water level lowered by 10.5 m, the area reduced to 82% of its original value and the volume to 64%, respectively. This state would be reached after 50-70 years under the assumption of the mean Qout being about 3 m/s. Due to lack of independent data, these figures cannot be confirmed or refuted, but they indicate that the Chinese authorities are very well considering to operate the power plant in mode (b), especially as this investigation concludes that this would not cause any real harm to the environment. The latter conclusion, however, is overly optimistic, as will be argued in the following sections.

3.2 Environmental impact concerns

There has only been one assessment of possible environmental impact on the Yamdrok Tso lake performed, and that is the paper by Shi Weiguang [2] mentioned above. The assessment addresses the issue of water level maintenance and concludes that a balance will be reached in 50-70 years. It also addresses the maintenance of water quality with regards to the heavy mineralisation level of the lake and its possible impact on the Yarlung-Tsangpo river and concludes that impact "is expected to be negligible even for drinking and irrigation purposes." [2]

The paper fails to address the impacts of pumping water from the Yarlung-Tsangpo river back into the lake (operation mode (a)). It would seem that at the time of writing this paper, this was not a consideration. But since then many statements have been made to suggest that it is in fact the case. On July 24, 1997, International Campaign for Tibet reported that a new hydro-power station was under construction about 100 km west of Yamdrok Tso on the Nyang river. Locals had reported that the purpose of this hydro station was to provide the power for the pumps to return water to the Yamdrok Tso. This project is expected to be completed within 2-3 years [4]. This indicates that an infrastructure is being built with the intention of pumping the water back from the Yarlung-Tsangpo and leaves the issue of the impact of the river water on the limnological make-up of the lake not addressed. It is of concern that introducing water from Tibet's highly silted rivers into an oligotrophic lake would produce eutrophic conditions in the lake over a period of time. Shi Weiguang [2] presented data on the compositions of the water in the lake and the Yarlung-Tsangpo river which illustrate the enormous differences. The pH in the lake is 9.11 while in the river it is 8.13; the total mineralisation in the lake is 1942 mg/l as compared to only 174 mg/l in the river. The oligotrophic character of the lake is shown for example by the nitrate concentration of only 0.16 mg/l whereas it was found to be 0.65 mg/l in the river [2]. The latter figure however, may already have substantially increased and will increase further with the current developments increasing population in towns like Shigatse and the intensification of agriculture in the Yarlung valley. The question of the impact of artificial turbulence caused by the pumping back of water into the lake is also an issue which is left not addressed. These questions should be studied in more detail before a verdict of negligible impact can be accepted.

If we consider operation mode (b), i.e. draining the lake, Weiguang's surmise that a drop in water level of 10.5 m would take 67 years to occur, is based on an outflow rate of 3 m3/s corresponding to 24% utilisation as mentioned above. The statement "the water surface will lower very slowly after the power station running, and the new beach abandoned by water will be covered in vegetation year after year, it is predicted that wind and sand problems will not appear." [2] may be accurate based on the figures at the time. It cannot be ruled out that utilisation will never exceed 24% in the next 70 years, therefore the rate of water level drop may increase and the impact on the shoreline environment may be greater than expected by Weiguang's analyses.

It is subsequently dubious that the author then finds it necessary to state, "The author considers that the core of the environmental problem is the water amount ... Fortunately there is a great amount of fresh water in Bomo lake ... supplying water from Bomo lake is not a difficult problem on the engineering viewpoint." [2]. This suggests reservations that the water level can be maintained. Together with the new plans for building a secondary hydropower plant to provide power for pumping, this statement suggests extending the environmental impacts of running this plant onto two other currently pristine environments. This extension of environmental impact onto secondary locations should be included in an environmental impact assessment of the Yamdrok Tso plant.

At the World Conservation Congress 1st session in Montreal, Canada, 14-23 October 1996, China signed a resolution entitled, `Co-operation between IUCN and the People's Republic of China on protection of the environment and biodiversity.' In the resolution it is requested of China to `strengthen their effort of co-operation with the international community in exchange of information, including that related to the local environment at Yamdrok Tso,… consider establishing a nature reserve at Yamdrok Tso, …(and) …calls upon the IUCN commissions to work with China in identifying areas of collaboration on maintaining the ecological health of Yamdrok Tso.'[5]

So far there has been no indication that these points are being followed up and it is of concern that the plant is now operating while these agreements are yet to be executed. Those concerned with this project including the Tibetan people, call for a new environment impact assessment to be carried out in collaboration with the IUCN, the findings be made available to the international community and that problems be identified and addressed so that environmental impact can be minimised.

3.3 Techno-economical dimension

Besides the immediate environmental consequences of a hydropower plant, the questions of whether there is really a need for new generation capacity equal to that of the planned facility and whether it is the most cost-effective solution have to be investigated in an integrated environmental impact assessment. This requires detailed data about the electricity system and the anticipated or desired economical developments. For Tibet, such data is not available outside of the PRC. In the following we present a preliminary assessment, based on information contained in [6] and [7].

These two publications differ slightly in their figures; Information Research Institute [6] say that the data is for 1994 while Zha Keming [7] does not indicate the year. Therefore we rely on [6]. Accordingly, the installed capacity in "Xizang Autonomous Region" (= Tibet Autonomous Region, TAR) was 166.7 MW of which 72.5% was hydropower. The electricity generation amounted to 357 GWh with a hydro share of 73.4%. According to [7], there are no high voltage transmission lines (110 kV or more) in the TAR and the TAR grid is not connected with other grids. This data means that only 24% of the generation capacity was utilised, corresponding to 2140 full load hours. This is a very low value (for comparison, in the biggest Chinese network - Central China Power Network - the respective number is 59%), indicating a strong reserve capacity [6]. Construction of new (net) generation facilities would appear necessary only if an enormous growth in the electricity consumption is anticipated. It is more difficult to judge the specific need for a pumped-storage plant, i.e. the need for a facility to provide additional peak load. Given the lack of large industries in the TAR, the load curve is probably more uneven than in an average network. Nevertheless, with only 24% average utilisation of the installed generation capacity, such a need is not easily intelligible. If there is one, load management possibilities should be investigated, and if the need for more capacity is confirmed, environmentally less harmful swell plants on rivers could be considered as an alternative.

The investment costs for the Yamdrok Tso power plants are quoted as US$ 200 mill., plus an additional US$ 36 mill. for steel lining of the tunnel which had collapsed in 1996 [8]. This is a high sum even for western standards. Consequently, development strategies relying to a higher degree on local resources would appear to be much more effective.

According to [9] only 0.82% of the potential of small hydropower in Tibet has been used so far. Though it is certainly not desirable to bring this figure close to 100%, it is obvious that there are many possibilities. Tibet is also abundant in geothermal, solar and wind energy. Guidelines for development projects in Tibet issued by the Central Tibetan Administration in exile explicitly discourage large scale projects, and encourage small scale hydropower projects for Tibetan villages [10].

3.4 Social, cultural and religious dimension

The history of the project indicates that it has been initiated and pushed through by the Chinese, inside and outside the TAR, while the Tibetan people - including those holding positions in administration and politics - were opposed to the project. In addition to the scientifically based concerns which were discussed in the previous sections, the outstanding cultural and spiritual significance of Yamdrok Tso as a sacred lake is the basis of the Tibetan opposition. Furthermore, it is obvious that such massive investments in the electricity infrastructure of Central Tibet go along with the current Chinese development policy in Tibet, which aims at modernisation through the attraction of Chinese colonists taking over the more qualified jobs in commerce, tourism and the growing industrial facilities. This development has caused great concern among Tibetans, openly voiced by the many individuals and organisations in exile. On the other hand, remote villages where only Tibetans live will not obtain access to electricity if all the money is poured into the large-scale projects.

These concerns have to be viewed in the light of the political situation in which the project has been carried out. It is characterised on one hand by the general difficulties to voice opinions different from those of the authorities in the political system of the PRC, and on the other hand by the tensions between China who regards Tibet as one of its provinces and Tibetans regarding China as a colonial power.

Yamdrok Tso, whose form is said to have the shape of a great scorpion, is sacred to the Tibetan people as one of the four "Great Wrathful Lakes" and abode of the guardian Goddess Dorje Gegkyi Tso. The lake, the islands and all the adjacent hills and mountains are closely associated with the deeds of Padmasambhava who established Buddhism in Tibet in the Eighth Century. A number of important monasteries are situated around the lake, among them Samding, the seat of Dorje Phagmo. Tibetan Buddhists as well as Bonpos hold that our environment is populated by spirits, such as the Naga in the waters and the Sadak in the ground. Polluting the water or digging in the ground will upset these spirits and they are provoked to harm human beings, for example through diseases. In addition, it is believed that Yamdrok Tso is connected with the `life power' (Tibetan bLa) of the Tibetan people. According to Nebsky-Wojkowitz [11], "Very frequently a lake is regarded as a bla gnas on which the life of a man or even of a whole nation depends. Thus for example, a legend claims that the Yamdrok Lake is the `life-power lake' (bla mtsho) of the Tibetan nation, and should it dry up then the whole population of the Land of Snows will meet its death." (quoted in [1]). Thus, it is understandable why Tibetans have so strongly opposed the construction of the power plant. The late Panchen Lama was able to halt the construction works; just months after his death in 1989 it was announced that the construction would proceed as planned.

4. Conclusions

The Yamdrok Tso pumped-storage hydropower plant in Tibet, which has recently been put into operation, has been a controversial project from the planning until now. There are a number of valid concerns related to the project. The plant might be operated in a unsustainable manner with respect to the water storage in the lake, causing a part of the lake's bed to fall dry with associated impact on the wildlife in this sensitive area. On the other hand, if water is pumped back from the river, the limnological condition of the lake (high in solved minerals but low in nutrients) will be changed, and again there will be a massive ecological impact. Thus, this is a no-win situation. Furthermore, it was shown that the technical need and appropriateness of this power plant is very doubtful.

Conclusions which can be drawn from these findings are twofold, relating on one hand to other projects to come and on the other to the future development at Yamdrok Tso itself. With respect to other development projects in Tibet, we strongly recommend that careful environmental impact assessments, discussed in a public participation process, should be carried out. Though certainly difficult under the present political circumstances, we think that this could lead to an alternative path in the development of the infrastructure in Tibet which would cause less harm and suffering to nature and people than the current direction which is exemplified in the Yamdrok Tso power plant case.

As operation has been started in the Yamdrok Tso power plant, the most important next steps would be to carefully monitor the limnological and ecological developments in the lake. The authorities should endeavour to implement points of the IUCN resolution mentioned in section 3.2 and protect the lake's environment. The release of detailed data to the international community on the electricity consumption and production in Tibet, the operation mode of the Yamdrok Tso plant, and any results of monitoring programmes would be an important step towards augmenting mutual confidence and co-operation in the spirit of the IUCN resolution.

5. References

1.Tibet Support Group U.K. (1996), Death of a Sacred Lake. The Yamdrok Tso Hydro-Electric Power Generation Project of Tibet (London: Free Tibet Campaign), 40 pp.

2. Weiguang, Shi. (1991), Hydropower development of plateau closed lake in Tibet: Environmental problem and protection. In: Tsakiris, G. (ed.) Advances in Water Resources Technology (Rotterdam: Balkema, 1991), 349-356.

3. Reynolds, Patrick. (1995), Tibet hydro set to go. International Water Power & Dam Construction May 1995, 44-45.

4. International Campaign for Tibet. (1997), New dam said to make electricity for pump-storage scheme. World Tibet Network News, 27 July 1997, The Canada-Tibet Committee. (On-line at

5. International Union for Conservation of Nature. (1996), Co-operation between IUCN and the People's Republic of China on protection of the environment and biodiversity. World Conservation Congress, Montreal Canada, 23 October 1996. (Geneva: IUCN Doc. CGR1.115)

6. Information Research Institute, Ministry of Electric Power (PRC) (1995), Electric Power Industry in China. (Epoch Printing Co., Ltd.), 49 pp.

7. Zha Keming. (1995), Energie-Entwicklungspolitik in China unter besonderer Beruecksichtigung der Elektizitaetswirtschaft. Elektrizitaetswirtschaft, 94(19), 1170-1179.

8. TIN News Update. (9 August 1996) Hydro-Electric Project: Tunnel Collapse, (London: Tibet Information Network).

9. Dansie, Janet. (1995), Orient Express or slow boat? International Water Power & Dam Construction, March 1995, 18-20.

10. Environment and Development Desk, DIIR, Central Tibetan Administration, Dharamsala (25 August 1995), Guidelines for International Development Projects in Tibet. (On-line) Available:

11. Nebesky-Wojkowitz, Rene de (1956), Oracles and Demons of Tibet: The Cult and Iconography of the Tibetan Protective Deities. (Kathmandu: Pilgrims Publishing House)

-> Table of Contents (Petra Seibert's Home page)

Responsible for the contents: Petra Seibert. last change: 14th day, 11th month, earth-tiger year. URL of this site: