America’s Reservoirs Are Filling Up — With Dirt
America’s Reservoirs Are Filling Up — With Dirt
Drought is shrinking the water supply. Wildfire is stripping the landscape. Then extreme rain arrives and washes part of the hillside into the reservoirs designed to protect us from drought.
Before worrying about whether a reservoir is half empty or half full, perhaps we should ask another question: how much of it is now filled with mud?
When we think about drought in the American West, the images are familiar: exposed reservoir banks, dry riverbeds, struggling farms, shrinking snowpack and arguments over who should receive the remaining water.
But an article from Bloomberg recently highlighted a much less obvious problem.
Dirt.
Wildfires burn vegetation from hillsides. Intense rainfall then falls onto those damaged watersheds. Instead of being intercepted by leaves, roots and forest litter and gradually soaking into the soil, water can rush downhill carrying ash, sand, soil, stones and even trees.
Some of that material eventually arrives in a reservoir.
The water may later leave.
The sediment often doesn't.
And very slowly — or after a major fire, sometimes surprisingly quickly — part of our water-storage infrastructure becomes soil-storage infrastructure instead.
That matters enormously in a region where every available litre of water is becoming more valuable.
A Reservoir Has Two Ways of Becoming Smaller
There is an important distinction here.
A reservoir can contain less water because its water level has fallen.
But it can also contain less water because its physical capacity has fallen.
Imagine a bathtub.
If you empty half the water, you can refill it tomorrow.
Now imagine gradually tipping buckets of sand into the bathtub.
You can add water again, but you can no longer put as much water into it.
That is essentially what sedimentation does to reservoirs.
And it is not a theoretical problem.
A major 2026 study examining more than 57,000 reservoirs across the continental United States estimated that sediment accumulation had reduced overall designed reservoir storage by about 7.7% by 2025. More strikingly, more than two-fifths of the reservoirs studied were estimated to have lost over a quarter of their original storage capacity.
That is water-storage infrastructure which physically still exists — dams, spillways, roads, pipes and pumping systems — but where part of the space behind the dam is no longer available for water.
Why Wildfire Makes the Problem Worse
Sediment entering rivers is completely natural.
Mountains erode. Rivers carry particles downstream. Sand and gravel move during floods.
A reservoir interrupts that process.
When fast-moving river water reaches the comparatively still water behind a dam, it slows down. Larger particles settle first. Finer silts and clays can travel further before eventually settling.
Over decades, sediment accumulates.
But wildfire can dramatically accelerate the process.
A healthy forest is surprisingly good at protecting the soil beneath it.
There are:
- leaves and branches intercepting rain;
- vegetation slowing surface water;
- roots helping bind soil together;
- leaf litter covering the ground;
- organic material allowing water to infiltrate;
- fallen branches and vegetation disrupting runoff.
A severe wildfire can remove much of that protection.
Fire can also change the characteristics of the soil itself. USGS research describes how fire can reduce infiltration, increase overland flow and leave abundant loose material ready to be mobilised. Post-fire debris flows can consequently be triggered by rainfall that would have produced relatively little runoff before the fire.
Then comes the rain.
And what would normally be welcomed as desperately needed water can become something much more destructive.
The Strange Combination: Drought and Flood
At first sight there appears to be a contradiction.
How can somewhere suffering from drought also experience torrential rain?
Very easily.
Drought is not simply the absence of every individual rainstorm.
A region can receive too little precipitation over months or years and still experience occasional extremely intense storms.
And intense rain falling onto dry, damaged or recently burned terrain is not necessarily particularly useful rain.
Instead of gradually replenishing groundwater or being absorbed by vegetation, much of it can become rapid runoff.
So you can have the extraordinary situation where:
the landscape desperately needs water,
yet
the rain that finally arrives causes flooding, erosion and reservoir sedimentation.
More rain does not automatically mean more usable water.
A Dramatic Example From Colorado
Colorado provides a particularly good demonstration of how large the problem can become.
The 1996 Buffalo Creek Fire burned around 12,000 acres upstream of Denver's Strontia Springs Reservoir.
Only two months later, flash flooding washed an estimated 160,000 cubic yards of debris and sediment into the reservoir — approximately 17,000 dump-truck loads.
Then came the enormous Hayman Fire in 2002.
Following the Buffalo Creek and Hayman fires and subsequent storms, more than one million cubic yards of additional sediment accumulated in Strontia Springs Reservoir. Denver Water has spent tens of millions of dollars dealing with water treatment, sediment removal, restoration and infrastructure consequences.
Today, Denver Water estimates that sediment occupies roughly 13% of the reservoir's storage space.
That is an extraordinary legacy.
The flames disappeared years ago.
The dirt they helped mobilise remains.
The Problem Doesn't End With Lost Storage
Losing reservoir capacity would be serious enough.
Unfortunately, sediment creates several additional problems.
1. Drinking water becomes harder to treat
Water containing large quantities of fine sediment becomes turbid.
Ash and soil can also carry nutrients, organic compounds, metals and other substances into the water.
USGS research has found that post-fire runoff can increase treatment requirements and increase the volume of sludge that treatment plants must dispose of.
So wildfire can eventually appear on somebody's water bill many miles away.
2. Dam equipment can suffer
Sediment is abrasive.
Sand and silt moving through infrastructure can increase wear on:
- pumps;
- valves;
- pipes;
- turbine components;
- hydroelectric equipment.
Accumulated material can also interfere with low-level dam outlets and water intakes.
The US Bureau of Reclamation describes reservoir sedimentation as capable of reducing storage, impairing outlets and intakes, increasing upstream flood levels and affecting recreation infrastructure.
3. Flood storage can disappear
Reservoirs are not always built simply to supply water.
Some provide flood control.
If part of their capacity becomes occupied by sediment, there is less space available to temporarily accommodate floodwater.
This produces another uncomfortable irony.
The sediment delivered by extreme rainfall can reduce the reservoir's ability to cope with future extreme rainfall.
4. Hydroelectric generation can be affected
Water stored at height represents potential energy.
Hydroelectric systems depend upon both water availability and functioning intakes and turbines.
Sediment therefore creates operational problems as well as reducing usable reservoir volume.
Why Don't We Simply Dredge It Out?
That sounds like the obvious answer.
If dirt has filled the reservoir, dig it out.
Unfortunately, reservoirs are enormous.
The quantities involved can be staggering.
Denver Water attempted a major sediment-removal operation at Strontia Springs in 2010. About 228,000 cubic yards were removed, but this was less than half the amount originally targeted and the work cost around $18.5 million.
Dredging presents several questions.
Where do we put the sediment?
Can machinery reach it?
Is the material contaminated?
Can it be pumped somewhere?
What happens to wildlife?
How much fuel and energy will removal consume?
And perhaps most importantly:
how quickly will the reservoir begin filling with sediment again?
Dredging can certainly form part of the solution.
It isn't a magic reset button.
Perhaps We Need to Stop the Dirt Before It Reaches the Reservoir
This may be where the most interesting lesson lies.
Protecting a city's water supply doesn't necessarily begin at the dam.
It may begin many miles upstream in a forest.
Denver Water learned this after its major wildfires.
Following the Hayman Fire, crews constructed sediment traps and debris racks in drainage channels. More permanent structures were later installed to intercept ash, sand and debris before they reached reservoirs.
There are many possible interventions.
Better forest management
Selective thinning, prescribed burning where appropriate and removal or management of excessive fuel can reduce the likelihood that every fire develops into an extremely severe landscape-scale event.
This does not mean attempting to prevent all forest fires.
Fire is a natural and necessary part of many ecosystems.
The challenge is preventing decades of accumulated fuel from producing unnaturally destructive fires in particularly important watersheds.
Restore vegetation quickly
After a severe fire, protecting exposed soil becomes urgent.
Depending upon the terrain and ecosystem, measures may include:
- mulching;
- erosion-control barriers;
- reseeding;
- planting;
- stabilising riverbanks;
- placing logs or structures across vulnerable channels.
Vegetation gradually performs the job for free.
Roots stabilise the ground.
Stems slow runoff.
Leaves intercept rainfall.
Healthy soil absorbs water.
A Forest Can Be Water Infrastructure
This is perhaps the biggest change in thinking required.
We normally imagine water infrastructure as concrete.
Dams.
Pipes.
Pumps.
Treatment plants.
Canals.
But a forested watershed is also part of the water system.
It captures rainfall.
It slows runoff.
It helps water penetrate the ground.
It reduces erosion.
It protects streams.
And it can reduce the amount of sediment arriving at the reservoir downstream.
That means expenditure on healthy forests can sometimes be considered investment in water infrastructure, not simply environmental spending.
Denver Water's experience eventually led to its From Forests to Faucets partnerships, aimed at improving forest and watershed resilience before the next catastrophic fire occurs.
That strikes me as one of the most important lessons from this story.
Sometimes the cheapest way to protect a water-treatment plant is to protect the mountain above it.
But Sediment Isn't Simply "Waste"
There is another complication.
From the reservoir manager's perspective, sediment is something occupying valuable storage.
From nature's perspective, sediment is part of the river.
Before dams were built, rivers continually moved sand, silt and gravel downstream.
That sediment helped create:
- riverbanks;
- wetlands;
- floodplains;
- beaches;
- estuaries;
- coastal deltas.
Trap too much sediment behind dams and downstream ecosystems may become sediment-starved.
The same 2026 national study that examined reservoir storage loss also highlighted the enormous quantity of material now trapped behind American dams and the implications for sediment-starved river deltas.
So we cannot simply say:
sediment = bad.
The real challenge is managing sediment rather than pretending it doesn't exist.
Can Reservoirs Be Designed to Pass Sediment Through?
Increasingly, water engineers are looking beyond continuously dredging material from reservoirs.
Depending upon the geography and dam design, possible techniques include:
Sediment bypass tunnels — diverting sediment-rich floodwater around a reservoir.
Sluicing — allowing high flows to transport sediment through dam outlets.
Flushing — lowering water levels and using flowing water to remobilise accumulated deposits.
Mechanical dredging — physically excavating sediment.
Sediment traps upstream — capturing material somewhere easier and cheaper to clean.
Watershed restoration — reducing abnormal erosion before material enters rivers.
No single technique suits every reservoir.
But there is an important change in philosophy here.
For much of the twentieth century we often built reservoirs and effectively accepted that sediment would eventually accumulate.
The twenty-first-century question may increasingly become:
How do we operate reservoirs so that they can continue functioning for centuries rather than merely decades?
We Also Need to Know How Much Storage We Really Have
There is another surprisingly basic problem.
You cannot manage lost capacity properly if you don't know how much sediment is sitting underwater.
Reservoir managers therefore increasingly need good bathymetric surveys — essentially mapping the underwater landscape.
Sonar can measure the reservoir floor.
Repeated surveys reveal where sediment is accumulating and how quickly storage volume is disappearing.
This is much more important than simply measuring the height of the water.
A gauge might tell you that a reservoir has reached a particular level.
But if the bottom of that reservoir has risen because of decades of sediment deposition, the same water level does not necessarily represent the same quantity of stored water.
A 2026 US reservoir sedimentation project highlighted just how incomplete the national picture remains: fewer than 1% of US reservoirs have apparently been surveyed sufficiently to assess storage loss from sediment accumulation.
You cannot protect storage capacity you haven't measured.
What Does This Have to Do With Going Green?
Quite a lot.
Going green is sometimes reduced to consumer decisions:
buy an electric car;
install solar panels;
change a light bulb;
recycle more packaging.
Those things have their place.
But climate resilience also means looking after the natural systems upon which our engineered systems depend.
A reservoir does not begin at the dam.
Its system extends backwards through the:
river → streams → valleys → soils → forests → mountains.
Damage the watershed badly enough and eventually the consequences appear at the reservoir.
This is why environmental management and infrastructure management increasingly need to become the same conversation.
The Lesson Isn't Limited to America
The scale and combination of wildfire, drought and huge western reservoirs makes the American example particularly dramatic.
But the principle applies almost everywhere.
Whenever we remove vegetation, compact soil, build over drainage areas or allow erosion to accelerate, rainfall moves faster across the landscape.
And wherever rivers enter reservoirs, lakes, ponds or drainage systems, sediment can accumulate.
There is even a small-scale lesson for our own gardens.
Bare soil washes away more readily than planted soil.
Roots stabilise banks.
Mulches protect surfaces.
Hedges intercept water.
Rain gardens slow runoff.
Leaves and vegetation are not merely decoration.
They are part of a remarkably effective natural water-management system.
Perhaps the Next Water Revolution Happens Upstream
For decades, the obvious response to increasing demand for water was to build another dam.
There may still be places where additional storage is needed.
But we also need to protect the storage we already have.
That means thinking about reservoirs not as isolated concrete structures but as the final component of enormous living watersheds.
It means monitoring sediment.
Managing forests.
Restoring burned landscapes.
Controlling erosion.
Rethinking dam operations.
Using water more efficiently.
And designing infrastructure around the reality that both drought and intense rainfall may challenge the same landscape.
I began looking at this as another story about drought.
I ended up thinking it was really a story about soil.
Because in a dry region, losing water is bad enough. Losing the space in which you could have stored the next rainfall may be even harder to reverse.
The next time we see photographs of a depleted reservoir and ask, "Where has all the water gone?", perhaps we should also ask:
Comments
Post a Comment