George Annandale has worked all over the world, studying, constructing and retrofitting dams and reservoirs to manage the sediment they accumulate. A native South African, Annandale, 59, is a water resources program leader for Golder and Associates, an international engineering and consulting firm. High Country News Executive Director Paul Larmer caught up with him in his office in Lakewood, Colo.

High Country News What attracted you to dams and water management?
George Annandale South Africa has a harsh hydrological cycle and droughts that typically last seven years. As a young child, I used to go to my grandparents' farm near the Orange River. It carries a lot of sediment -- a load three or four times higher than what we have here in Colorado. One of the largest dams was built there, and there have been many studies done of its sedimentation problems. When I earned my Ph.D. in 1984, with a thesis on sedimentation and dams, I thought I'd have everyone wanting my services. But it wasn't until 2000 that sedimentation was recognized as a huge problem.

HCN Why hasn't it been taken more seriously?

Annandale Well, usually you can't see it because it is underwater, so people ignore it. Also, the benefits from the facility (the dam and reservoir) do not decrease immediately. It's usually when the dam is significantly filled with sediment and there is a sudden drop in the reliability of water supply that everyone starts to panic. And then it is too late.

HCN Are all dams doomed by sedimentation buildup?

Annandale A dam can live in perpetuity, but you have to manage for sediment regularly and from the beginning. Most dams were built in the 1960s and 1970s with a philosophy that they had a limited life, perhaps 100 years. That's OK for roads, because you can just repave them, but not for dams (because) it means that three or four generations enjoy the benefits, but the fifth generation has to pay for it all. At the least, each generation should be investing in a fund for decommissioning dams. But I have found that it is much more economical to actively manage dams to extend their lives.

HCN How do you remove sediment?

Annandale You can move it around the dam by bypassing the reservoir during high flows when the river is carrying the most sediment. Bypass tunnels are typically used in short reservoirs; there are a lot of these projects in Japan. You have a tunnel with a diversion dam upstream, and when the flood comes, you divert it into the tunnel and release it below the dam. This has ecological benefits to the river. Another approach is to carry as much sediment through the reservoir as possible. There are density currents in many reservoirs that carry very high sediment concentrations along the floor; you can install low-level gates on the dam and open them when these currents occur, so the sediment goes through. To remove sediment that already exists, one option is to dredge. Denver Water is currently dredging Strontia Springs Reservoir and told me that it is costing $30 a cubic yard; this is very expensive, so usually you just dredge close to intakes to keep them operational. If you have the right geometry, the most economical method is flushing. You need a reservoir in a fairly narrow valley with a steep river and a dam with low gates. Then you open the gates and draw all the water down to create river-like flow conditions in the reservoir that take out the sediment.

HCN Could we flush out Lake Powell?

Annandale The problem is that if you flush it, there is no backup system to provide water. A couple of years ago, I did a study of the Tarbela Dam in Pakistan. The reservoir is 100 kilometers long and rapidly silting up. It provides 30 percent of the power and 50 percent of the irrigation water for 160 million people. Flushing is feasible there, but it would take the system out of commission for about three months, and then you need something else to fill in during that time. They don't have that. In many places around the world, we need more dams.

HCN Does the West have enough dams?

Annandale Let's take Denver as an example. I did some calculations and found that Denver Water's current supply will fail once every 46 years. That's not good enough. Cities need 99 percent reliability. California is also short on storage.

HCN What about groundwater?

Annandale It is not a sustainable alternative to river water. To replenish a shallow aquifer can take 100 to 200 years; to replenish deep fossil water can take 10,000 years. On the other hand, if you removed all of the water from rivers, it would take just 16 days to have flows back. But rivers vary seasonally and from year to year; if you use more than 30 percent of the mean annual flow of a river, you need storage to ensure a reliable water supply.

HCN How does climate change play into this?

Annandale Climate change stresses the importance of dams -- and large dams -- even more. Most of the investigators of climate change indicate that hydrological variability will increase. I'm working on a project in Kenya where we're looking at the sensitivity of supply based on variable flows, and it shows that you need larger reservoirs to withstand these conditions. I don't think the dam-building era should ever be over. If we don't increase storage, we will have severe shortages.