Dams, dikes and dredging: Can we 'fix' our rivers?

'We need to fix this river" is the sentiment that seems to work its way to center stage at most gatherings of recently flooded residents.

It may not be the first statement, but it seeps out, usually after some official presents revised federal flood maps, or tries explaining why bigger dams, higher levees or more dredging will not reduce their risks. These highly charged statements are often followed by equally passionate cries to keep the river natural — whatever that means in 2010.

But is there a reasonable role for dams, dikes and dredging?

When Captain George Vancouver visited our shores in the late 1700s, rainwater was either absorbed into the forest or ran a hydrologic gantlet through soil, roots and vegetation as it worked its way to the ocean through networks of beaver ponds and meandering rivers choked with large woody debris. When there was a flood, water spread out and built floodplains, enriching soils and improving water quality.

We changed this landscape. There was no overall plan. As problems arose, they were addressed. At first, people adapted to the landscape. You just have to drive North along Interstate 5 to the Skagit Valley to see the oldest homes built comfortably above and away from the river.

Over time, we began to rely on flood-control efforts. You just have to drive along the Green or Puyallup rivers to see newer homes built at grade on reclaimed lands "protected" by levees towering above them.

This year, concerns associated with risks from the Green River and Howard Hansen Dam reminded us that we have placed buildings and infrastructure on lands that flood. We have built dams to control flows, straightened channels to provide more buildable flat land, sandwiched river corridors between dikes and in the process not only greatly reduced flood-storage areas, but degraded water quality and aquatic life.

And we continue to cover our watersheds with an impermeable built environment creating more flood damage and reduced water quality. Now we face a changing climate predicted to increase flooding in Western Washington.

What can we do to "fix our rivers"?

Do we need more dams?

Big dams can control flows, prevent flooding and store water. Big dams also trap sediment and can cause downstream rivers to incise, destabilizing valley bottoms. And let's not forget we live in earthquake country; big dams present a risk to downstream residents should they fail.

But our foothills, plateaus and floodplains provide extensive small-scale detention possibilities that offer an abundance of surface and aquifer storage opportunities. And, thanks to past glaciers, our valleys are filled with unconsolidated sediment that provides excellent storage.

Beavers helped in the past; now we must follow their lead. We could benefit from more dams, but not big ones; small dams scalable by salmon each offering a few acres of detention, scattered throughout the watershed.

Do we need more dikes?

Where we have created a dike-dependent built environment, removing entire systems is not a viable option. But do levees really protect? Herein lies a contradiction. Levees seldom reduce risk. They reduce the frequency of flooding at the cost of increasing the magnitude.

Again, think of the Green River. Before the Howard Hansen Dam, the valley flooded a lot — only a few feet in most places, but often. After the dam was built, new development moved into these "protected" lands but with the dam compromised we now face the possibility of a multibillion-dollar disaster if too much rain falls too fast. The dam and dikes have increased risk. Frequencies are lower, but the associated impacts have greatly increased.

We need to rethink levees as they are becoming prohibitively expensive to build, maintain, and cannot assure safety. Instead of building them up, we need to build lower ones that are not continuous on both sides of the channel and set farther back from the river.

We need to think of flooding as inevitable on floodplains and need to build our communities to accommodate shallow flooding that allows water to spread over the land. This approach would store water on the land, aquifers, and allow natural processes to improve water quality.

Should we dredge rivers?

Dredging can reduce risk but it is very expensive. When we dig a hole in the channel it just fills up again. Removing too much sediment can result in extensive channel cutting downstream. Instead of dredging we need better soil management and harvesting practices, and allowing rivers to meander and lengthen within wider corridors.

Here are some rules of thumb:

Dams are OK — just not big ones, and we should seriously begin reclaiming habitat and reintroducing beaver communities.

Dikes are fine — just not as big, close to the river, or as lengthy or continuous.

And we should let rivers dredge themselves.

David R. Montgomery, author of "King of Fish: The Thousand-Year Run of Salmon" and "Dirt: The Erosion of Civilizations," is professor of geomorphology at the University of Washington and a MacArthur Fellow.