Hydro · Small impoundments
A pond behind a small dam turns a stream's steady trickle into power you can call on when you want it. It also creates a structure that has to hold, in every storm, for as long as it stands, and a legal obligation that transfers with the deed.
The case for it
Run-of-river gives you whatever the stream is carrying at that moment. An impoundment lets you decide. That difference is genuine, and three situations make it valuable enough to consider seriously.
Water held overnight can be released through the turbine during the evening peak. A site producing 800 watts continuously can instead produce 2,400 watts for eight hours, which is a different kind of usefulness.
On flat ground a dam creates the vertical drop the terrain refuses to provide. This is the one case where a site with no usable head can become a site with some, and it is why flat properties keep asking the question.
Department of Energy guidance says to design against the lowest average flow of the year unless you are building storage. Storage is the exception that lets a site be sized against something better than its worst week.
The other side
The United States has roughly 92,000 dams listed in the National Inventory of Dams, and state governments regulate about 70 percent of themFEMA, National Dam Safety Program Resources for States">[1]. That regulatory apparatus exists because an impounded volume of water is a stored quantity of energy sitting above whatever lies downstream.
FEMA states the ownership position without softening it. Owners can be fiscally and criminally liable for any failure of a dam and all damages resulting from its failure, and any uncontrolled release of the reservoir, whether intentional or from failure, can have devastating effects on persons and property downstream[2].
That responsibility does not lapse. It runs with the property, outlives the person who built the thing, and applies in the years when nobody is thinking about it. A dam is the only component in this entire section that can harm someone who has never heard of your project.
Federal guidelines define a three-level system of low, significant and high hazard potential, based on the probable loss of life, economic loss and environmental damage that a hypothetical failure would cause[2]. It says nothing about how well the dam is built.
The consequence that follows surprises people. A classification can change without the dam changing at all. Build above an empty valley and the rating is low. When somebody puts a house in that valley twenty years later, the same structure becomes a significant or high hazard dam, and the design standards, inspection schedule and emergency planning obligations that apply to it change with it.
Failure modes
Dam engineering is organized around a small number of well-understood failure mechanisms. Knowing them is what separates an owner who notices a problem early from one who does not, and it explains why the design requirements look the way they do.
National statistics attribute roughly 34 percent of U.S. dam failures to overtopping, arising from inadequate spillway design, debris blocking the spillway, or settlement of the dam crest[3].
Water flowing over the crest of an earth embankment erodes it, and the erosion accelerates as the breach deepens. The mechanism is ordinary flowing water doing what flowing water does to soil.
About 30 percent of failures trace to the ground under the dam rather than the dam itself, including settlement and slope instability[3]. This is why geotechnical investigation precedes design, and why a dam cannot be sited by eye. What the soil and rock beneath the footprint will do under load is not visible from the surface.
Roughly 20 percent of U.S. dam failures come from piping, where seepage through the embankment is not properly filtered and carries soil particles away with it, forming voids and sinkholes inside the structure[4].
The Association of State Dam Safety Officials lists where seepage tends to start: around hydraulic structures such as pipes and spillways, through animal burrows, around the roots of woody vegetation, and through cracks in the dam, its appurtenances and its foundation[4].
Their assessment of the endpoint is blunt. Fully developed piping is virtually impossible to control and will likely cause failure[5]. The remedy is preventing it through filter design and vigilance, because there is no good way to stop it once it is running.
Two of the three leading causes are invisible from the outside. Foundation behavior and internal erosion both develop inside or beneath the structure, and ASDSO notes that saturation and internal pressure within a dam are difficult to determine without proper instrumentation[5].
A dam that looks fine may not be fine. That is the specific reason this subject sits outside what a website can responsibly teach, and inside what a qualified engineer is for.
The spillway
Every impoundment needs a controlled route for water it cannot hold. That route is the spillway, and its capacity is the single most consequential number in the design. The Association of State Dam Safety Officials attributes more than 40 percent of recent dam failures to inadequate conveyance capacity[6].
A spillway is not sized for the rain you have seen. It is sized for an inflow design flood, a statistical event far larger than ordinary experience, selected according to the dam's hazard classification. The higher the consequences of failure, the larger the storm the structure has to pass without being overtopped.
This is the point where amateur reasoning most often goes wrong. The sentence that precedes a great many failures is some version of the creek never gets that high. The spillway exists precisely for the year it does, and the size of that year is a hydrology question with a calculable answer rather than a matter of local memory.
A spillway partly blocked by branches is a smaller spillway. The blockage arrives with exactly the storm that needs the full capacity, which is why debris blockage sits inside the overtopping statistic rather than beside it.
ASDSO documents serious incidents where flows leaving the spillway eroded the downstream slope of the embankment, in several cases at peak flows well below the design flood[7]. Where the water lands is part of the design.
An embankment that settles a foot over two decades has given a foot of margin back to the reservoir. Crest settlement appears in the overtopping cause list for that reason, and it is found by survey rather than by looking.
Where this guide stops
This page covers what an impoundment is, what it offers, what it demands, and how the structures fail. It does not provide embankment design, spillway sizing, compaction specifications, or construction sequence, and no responsible source will provide them to a general audience.
The reason is in the failure statistics above. Two of the three leading causes are conditions you cannot see, and the third turns on a flood magnitude that has to be calculated rather than remembered. Those are not gaps a careful amateur closes with effort.
Federal standards for dams receiving federal assistance describe the arrangement plainly, and it is a reasonable benchmark for any impoundment regardless of funding[8]:
Note the fourth item. Money set aside for the decades after construction is part of the standard, not an afterthought, because the failures in the statistics above are overwhelmingly failures of aging structures rather than new ones.
Start with your state dam safety office. Every state has one, they are accustomed to talking with landowners, and a conversation before design costs nothing. They will tell you what your state regulates, at what size, and what your particular site would be classified as.
The permitting stack
A run-of-river diversion generally needs a water right and a state approval. An impoundment adds layers, and each one has its own timeline and its own ability to stop the project.
States regulate about 70 percent of the dams in the national inventory[1]. Thresholds for what counts as a regulated dam vary considerably by state, usually keyed to height and storage volume, so the first question is whether your proposal is regulated at all.
Storing water is legally distinct from diverting it. A right to divert does not automatically include a right to impound, and in prior appropriation states the storage of water carries its own priority date and its own scrutiny.
Placing fill into waters of the United States requires a permit from the Army Corps of Engineers[9]. Building an embankment across a stream is that. The program requires showing impacts were avoided and minimized, and that no less damaging practicable alternative exists.
Generating power from an impoundment can bring the Federal Energy Regulatory Commission in. There are exemption pathways for small projects, and notably a simpler route for projects at an existing dam. The permits guide covers this in full.
Section 404 requires that no discharge be permitted if a practicable alternative exists with less adverse impact on the aquatic ecosystem[9]. Where a run-of-river diversion could produce comparable power on the same property, that is an obvious less damaging alternative, and the applicant has to answer for it.
The honest recommendation
An existing impoundment, an old mill pond, a farm pond built decades ago, changes the question entirely. The structure exists, the impact has already happened, and adding generation is a far smaller proposition than creating a reservoir. Federal exemption pathways specifically accommodate small projects at existing dams. If you own one, have it assessed, because its condition is now your responsibility whether you generate power or not.
On genuinely flat ground with substantial flow, an impoundment is the only way to create usable drop. This is a real case, and it is also the case most likely to produce a large structure holding a large volume, so it is where engineering matters most.
Ponds get built for irrigation, livestock, fire protection and fish. Where one is being engineered and permitted for those reasons anyway, adding a turbine to the outlet is a modest addition to a project already receiving professional attention. The economics change completely when the dam is not on the power project's budget.
If your property has usable head and a stream that keeps running in August, build run-of-river. It produces power around the clock, it asks for no impoundment, it carries no dam liability, and the previous guide covers it end to end. The overwhelming majority of households that arrive at this page wanting a dam are better served by a pipe.
Next
Run-of-river and impoundments are the two arrangements that account for most working micro-hydro. The waterwheel guide covers the third, which trades efficiency for something a person can build with ordinary tools and understand by looking at it. After that the section turns to machinery, civil works, control and law.
Sources
Failure statistics are national aggregates across dams of all sizes and are used here to show which mechanisms dominate, not to predict any individual structure. Regulatory thresholds, hazard classification criteria and permitting requirements vary by state. Contact your state dam safety office before planning any impoundment.