Dam removal is a relatively new area of science. It wasn’t until the mid-1990s that tearing down a dam to restore a river became a regular occurrence across the United States.
Most dam removals have been studied and monitored closely to ensure that results meet expectations and to avoid harmful consequences. But there haven’t been many comprehensive reviews of the field of dam removal to answer broader questions, such as: How many dam removals actually helped fish recolonize rivers? How long did that take? What were the effects of releasing sediment stored behind dams?
Now we can have a comprehensive look at those questions, and many others. In a new study overseen by the U.S. Geological Survey John Wesley Powell Center for Analysis and Synthesis in Fort Collins, Colorado, a team of 22 scientists reviewed more than 200 studies on dam removals around the country.
Their conclusions show that changes can occur much faster after dam removal than anticipated. Fish often recolonize historic upstream habitat within weeks. And sediment trapped behind dams can flush out and then reach a new stable condition in a downstream river channel within months.
These are generalizations, and every dam removal is unique. So results will vary depending on reservoir size, sediment conditions, streamflow behavior and many other variables.
Another important caveat from the study is that most dam removals have been small. There still haven’t been many results to indicate what happens when a large dam is removed that had significantly altered natural river flows. About 1,300 dams have been removed in the U.S., according to the study, and around 1,000 of these have been removed just since 1996. But only 83 were more than 32ft tall.
To explain the results further, Water Deeply recently interviewed several authors of the study via email: Melissa Foley, research ecologist, and Amy East, research geologist, both at the USGS Pacific Coastal & Marine Science Center; Gordon Grant, research hydrologist at the U.S. Forest Service Pacific Northwest Research Station; and J. Ryan Bellmore, research fish biologist at the USFS Pacific Northwest Research Station.
Water Deeply: What’s unique about this study?
Melissa Foley: In this review, we bridged multiple disciplines to tell the broader story of how rivers respond to dam removal, emphasizing the physical or geomorphic response, as well as the ecological response.
For the physical response, in particular, we compared how rivers responded to dam removals in different geographic settings to theories of how rivers “should” respond that have been developed over the last 30 years. That comparison revealed that rivers are responding to dam removal faster than predicted.
Understanding the ecological response to dam removal can be more challenging. But in many cases, it is closely tied to the physical response. The amount and size of sediment released from storage behind a removed dam and how it moves downstream is a key factor here.
As the volume of sediment being carried by the river decreases downstream of the former dam, the ecosystem begins to resemble upstream sites that were not affected by dam removal. In the former reservoir, a similar transformation occurs as the system changes from a lake-like system to a river system.
For fish species, movement upstream tends to happen quickly after the barrier is removed.
Water Deeply: Your results seem to suggest that important changes occur much faster than we thought. Is that a fair assessment? What changes in particular?
Foley: We see that the speed at which the river responds to dam removal is largely dependent on the speed of dam removal and the type of sediment that has accumulated behind the dam. In general, sediment movement downstream occurs most quickly when dams are removed quickly and the reservoir sediment is unconsolidated. In many cases, high river flows have not been necessary to move a majority of sediment downstream after dam removal.
Amy East: Knowing that this type of rapid “flushing” of sediment out of rivers can occur after dam removal – even without large floods – is useful for helping us understand how landscapes will change after other natural disturbances that release a similar sediment pulse, such as large landslides or volcanic eruptions. Even when digesting a fairly large sediment pulse, the system settles down substantially after three to five years.
Water Deeply: How might results differ between small and large dams?
Foley: The main difference between small and large dams is likely to be the amount of sediment that is eroded after dam removal.
In either case, however, the initial ecological response to dam removal is generally the same: The number and types of critters found downstream of the dam decrease after dam removal, while the types of critters found in the former reservoir change from lake-adapted to flowing water-adapted species. The length of time it takes for downstream critter populations, in particular, to rebound after dam removal depends on how much sediment was deposited in the river channel and if the grain size of that sediment changed.
Gordon Grant: Also, it’s hard to take a large dam out all at once, so they tend to be taken out in stages. This introduces a different dynamic, because the sediment released from a large dam removal leaves the former reservoir as a series of pulses, whereas the instantaneous removal of a small dam results in the system being flooded with sediment (albeit a smaller volume) all at once.
Water Deeply: Your results seem to show that the problem of trapped sediment sorts itself out within a few weeks or months after a dam removal. So is the sediment problem over-rated?
East: Rather than saying weeks to months, it’s better to characterize the time frame of major sediment response as months to a few years. After three to five years, the sediment situation stabilized a lot after even the largest dam-removals we’ve studied.
Foley: There are a number of factors that determine whether the sediment is a “problem.” First, the sediment that accumulates behind the dam may or may not be contaminated by organics, heavy metals, agricultural products or other material, which, if present, introduces additional logistics, expense and risk to a dam removal.
Second, downstream infrastructure may be affected by a sudden increase in the amount of sediment being transported down a river. Dam removal on the Elwha River was stopped for many months because a downstream water treatment facility could not handle the increase in sediment in the river.
However, the restarting of the sediment conveyor belt, in many instances, is beneficial to the river. Downstream sediment deposition can create new habitat for vegetation, restore viable habitat for aquatic species such as salmon, and rebuild eroded coastal shorelines that were sediment-starved while the dam was in place. So sediment release can be a double-edged sword.
Water Deeply: You make a distinction between “response” and “recovery” of species and habitats after dam removal. Why is this important?
Foley: “Response” is a generic term that means change; “recovery” is a very specific term in the ecological sense that means a return to a specific state or condition.
In many cases, we define “recovery” using a historical state. However, we often do not know what an ecosystem looked like before dam emplacement, so it is difficult to talk about recovery in the true sense. In addition, in many watersheds, there have been other changes to the landscape – including land use, water availability and climate – that could preclude “recovery” after a dam is removed.
J. Ryan Bellmore: Many rivers may never “recover” following dam removal. Remnants of the dam may continue to exist in the shape of the river channel and also in the community of organisms that occupy the river.
For example, in the past, non-native and invasive species were often introduced to reservoirs to provide recreational fishing opportunities. Once dams are removed, however, many of these species remain, forever changing the structure of the ecological community, and perhaps obstructing the recovery of desired native species.
Water Deeply: What are the key unanswered questions that remain?
Foley: First, there are few long-term studies that characterize the trajectory of ecosystem response to dam removal. Most studies are conducted for less than two years, which limits our ability to understand physical and ecological progress and how they are linked.
Second, the dam removals that have occurred in the U.S. have come from a relatively limited range of geographic conditions, which limits our scope of inference. For example, dam removals in arid climates have been limited. Expanding the context of removals would also allow scientists and managers to evaluate which dam removals are likely to result in desired outcomes.
Third, we need more holistic studies that integrate physical, water quality and biological responses to dam removal. Implementing new technologies can help expand dam removal monitoring. Finally, ready access to up-to-date information and data will help expand the scope of dam removal science and feed into future management decisions around dam removal.