This forum is about wrong numbers in science, politics and the media. It respects good science and good English.
In the case of man-made dams they might work as Frank says, because the dam keepers might maintain the water level below the maximum, thus creating a temporary storage capacity for flood waters that might be released more slowly. If the flood is such that it reaches maximum level, then the delay gives time for people downstream to evacuate.
Not the case with beaver dams, though.
An interesting thought. JEB's point that the dam is already full is perhaps questionable because the dam is leaky. Were it a concrete arch dam that was full, obviously John would be right and any extra water would flow over, under or round it and continue on its merry way, but a beaver dam lets most of the water through its mesh of logs and so the impounded lake is seldom up to the top.
Most floods that concern us last a matter of hours and are the result of cloudbursts or rapid thaws. Such a pulse of water, assuming it doesn't simply destroy the dam, will be impounded and raise the water level in the pond, much as in a balance pond alongside a motorway. From there the water is steadily leaked out into the watercourse.
I see a beaver dam as the hydraulic equivalent of an inductor. Any further thoughts?
"Rodents Of Unusual Size? I don't think they exist."
After the flood, Mrs Sizzlepot of Upper Gumtry found a dead beaver in her kitchen...... "We had it stuffed and mounted" she said. "Now all the visitors like to stroke it. We call it Gordon because it was all washed up."
I am a professional civil engineer, now retired, but I have done a good deal of hydraulic engineering, and hydrology studies. I feel I have the expertise to comment.
Beavers construct their dams for a purpose, and that purpose is to impound water to a sufficient level so that they can build shelters above the water level that are basically "dry" with air to breath, but with underwater entrances, that predators can't easily use.
Three significant factors for determining a quantity (Q) at a point downstream are: Intensity of a storm over time, runoff area and time of concentration. The last is the time it takes for a drop of water from the furthest distant point of the watershed area to reach the point downstream when it is desired to know the "Q". Longer times of concentration result in lower "Q"s.
When the beaver dam is being constructed, water is impounded, and the beavers keep making the dam higher until the desired depth is reached. While the water is being impounded, the downstream flows will be less. Once the water level is at the desired(by the beavers) level, additional water will not be impounded, and what goes in come out.
Some additional water may be impounded in high intensity storms, lessening the downstream flow. Unfortunately, dams constructed by beavers may well be damaged by additional flow in and higher ponded levels, and the outflow then is higher than would be the case without any dam at all. Outflow = inflow + flow of previously impounded water. This would result in higher flows downstream of the dam than there would be with no dam at all.
Hydrology is complex, usually with a good many assumptions or guesses. Beaver dams may, depending on the length and intensity of a given storm, lessen flooding downstream in low intensity storms and increase flooding downstream significantly in high intensity and longer storms.