This message, by Alan Potkin, appeared on one of the Lao email lists. It is a useful critical follow-up to my earlier post on the Mekong and an associated press report in Al-Jazeera.

In an Al-Jazeera article published on 8 March 2010, Dr. Andrew Walker, an anthropologist at the Australian National University, who also personally publishes the New Mandala blog, was quoted as believing that “the Chinese dams have little impact on the current low flow levels of the Mekong”.

I wouldn’t have necessarily disagreed at first, but I have since come to realize that the new Xiaowan dam, the fourth (of the eventual eight existing and proposed) comprising the Lan├зang-Jiang cascade has been almost certainly filling its reservoir for the first time. (The Lan├зang-Jiang is what the Chinese call the upper Mekong inside Yunnan.)

Quick and dirty online research indicates that such filling has apparently been underway since no earlier than the 2008 rainy season, when the Xiaowan dam itself was finished, and possibly didn’t really begin until a year later. Bringing that reservoir to Full Pool Elevation is “expected to take ten years”, according to the Chinese. Unlike the other three already completed projects in Yunnan, which can be more plausibly described as “run-of-river” (e.g., by comparison, the capacity of the Manwan reservoir is only about 0.9 km3), Xiaowan has a designed reservoir volume of about 15 km3 (cubic kilometers), of which nearly 14 km3 would be dead storage, so until that level is reached, presumably, no hydropower can be generated and wheeled to wherever it’s going.

While my tendency would have been to pooh-pooh the recent China-bashing, especially by the Thais, but from investigating the Xiaowan project considerably further these past few days, and seeing how the Chinese are extremely sensitive about acknowledging even the possibility of the Lan├зang cascade exacerbating record downstream drought low flows (no question that the present drought is for real, but that’s another issue), I’m beginning to see the matter quite differently.

In principle, of course, once the Lan├зang Jiang cascade has been completed and the reservoir staircase fully filled, there should be at least the possibility of dry season flow augmentation and wet season flood amelioration. But that speaks not at all to the present situation with Xiaowan, which is the most upstream of the four dams now completed.

Interestingly, the (undated?) image of the Xiaowan project area now up on Google Earth evidently predates any significant construction there, much less showing the status of the impoundment which is supposed to encompass a surface area on the order of 180 km2. Check it out yourself: approx. coordinates for the dam site are 24┬║ 41′ 08″ N., 100┬║ 05′ 59″ E.

(Much newer high-resolution satellite imagery is available from sources such as Geo Eye, which yesterday quoted me a price of US$12/km2, with a 50 km2 minimum. The entire Xiaowan project area is probably something like 350 km2, from the head of the reservoir footprint to the toe of the bank stabilization works extending maybe 10 km below the dam, spillways, and turbinated outfalls. So I won’t be posting the Geo Eye pix into the LaoFAB repository just yet.)

Given the enormous volume required to fill Xiaowan, the most critical gaging stations to evaluate a deleterious Chinese contribution to extreme low flows now evident as far downstream, certainly, as Luang Prabang, would be those two just above the Xiaowan pool (ca. 190 km in length, when totally full!), and just below the 292 meter high Xiaowan dam. The difference in discharge data between these two sites would indicate exactly how much of the Lan├зang Jiang is being retained now to fill Xiaowan.

The Chinese agreement to provide streamflow data from Manowan (which is just downstream of Xiaowan); and from Jinghong (in Sipsong Banna), which also has relatively minor storage volume, is welcome and is a positive step. But it isn’t rocket science to realize that those latter two gaging stations alone provide essentially zero information as to how much water is presently being retained much further upstream to fill the new Xiaowan reservoir. Why is the MRC being so coy over this?

The Xiaowan project operator, a private sector PRC company, is likely bound by a very onerous late-delivery penalty clause in their Power Purchase Agreement (PPA). If I have it right, the non-delivery or substantive delay penalty clause for Nam Thuen 2 in their PPA with EGAT was something like US $1M/day, which may partly explain why NTEC has just started wheeling power to Thailand on an “experimental” or “non-commercial” basis, to finesse nominally binding requirements from NT2’s multilateral development bank funders and guarantors to have certain mitigation projects absolutely in place before full commercial operation. (At about 1,050 MW, NT2 has only a fraction of the eventual installed capacity of Xiaowan, which will be 4,200 MW when all six turbines there are operational. So the PPA penalty for delaying Xiaowan may well be four times that for NT2.)

Thanks for all due consideration.


Alan Potkin, Ph. D.
Adjunct Consultant
Center for Southeast Asian Studies
Northern Illinois University