Skip to main content
Category: Effects
Type: Lower vigilance after a structure blocks ordinary floods, while more is built in the protected zone
Origin: Gilbert F. White, Human Adjustment to Floods (University of Chicago, 1945), as cited by Di Baldassarre and colleagues, Hydrology and Earth System Sciences, 30 October 2018. White’s book was not re-read
Also known as: levee effect, safe development paradox, safety dilemma. The journals read here do not print Dam Effect
Quick Answer — The Dam Effect, on this page, is a drop in vigilance after a structure blocks ordinary floods, so a long quiet stretch is treated as safety and people prepare less. Journals call the counted pattern the levee effect, not Dam Effect. In the contiguous United States, levee construction was associated with a 62 percent acceleration in floodplain urban expansion, against 29 percent in the county. That pair is land cover. It is not a dam, and it is not a count of people who forgot a flood.

​
What is the Dam Effect?

The Dam Effect is a drop in crisis awareness among people a protective structure was built to shield.
“increasing levels of flood protection can generate a false sense of security and attract more people to the risky areas.”
That line is the short summary on the journal page for Giuliano Di Baldassarre and colleagues, Hydrology and Earth System Sciences 22, pages 5629–5637, published 30 October 2018. The sentence before it names levees. The abstract’s own words are a sense of complacency that can reduce preparedness. The abstract’s structures are levees or flood-control reservoirs. The names it prints are levee effect, safe development paradox, and safety dilemma. It does not print Dam Effect. This page uses Dam Effect for that pattern. The picture starts before the structure. People who still meet ordinary floods have a reason to stay alert. After the structure is in place, medium and short floods are less likely, and a long stretch without a disaster can be read as an umbrella that cannot fail. The papers read here do not score the same households before and after a dam. They start once protection is there. Normalcy Bias is a separate name for treating a long calm as the way things are. These papers do not score it. Peltzman Effect is a different move: a person takes more risk because a safety rule lowers the cost of a mistake. No belt-law number belongs on a levee or a dam.

​
The Dam Effect in 3 Depths

  • Beginner: Years without a flood, behind a wall or a reservoir, can feel like a promise. Write the flood the structure was not built to stop. The quiet years do not replace that sentence.
  • Practitioner: If a map drops a neighborhood from the official floodplain because a levee is certified for the 100-year flood, write the remaining risk in the same note. That gap is what Jessica Ludy and G. Matt Kondolf name in their 2012 abstract.
  • Advanced: The United States pair is a 62 percent acceleration of floodplain urban expansion against 29 percent in the county, for levees. The positive pattern weakened and reversed after the late 1970s. Italian means of 4.49 and 3.92 are agreement among informed respondents, not a before-and-after of one town.

​
Origin

The journals’ name is older than this page’s title. Di Baldassarre and colleagues trace the literature to Gilbert F. White, Human Adjustment to Floods, Research Paper No. 29, Department of Geography, University of Chicago, 1945. That book was not re-read. No sentence from White is on this page. The 2018 paper is an opinion article. Structural measures such as levees or flood-control reservoirs substantially reduce the probability of flooding at the time they are built. Raising protection can draw settlements and high-value assets into the protected area. The abstract says other studies find complacency that can reduce preparedness. The paper then uses levee effect for a further tendency: protection changes where new settlements go, and exposure can rise. Safe development paradox and safety dilemma are the other labels it records. A flood-control reservoir sits in that class of structures. The land acceleration below was estimated for levees.

​
Key Points

Four readings keep the quiet years, the new buildings, and the survey answers apart. A decade without a flood is not a memory test.
1

No flood is not a zero

At the moment of construction, the chance of flooding falls. It does not become zero above the design. In the Netherlands, Di Baldassarre cites a levee system able to withstand floods with return periods from 500 to 10,000 years. That range is his citation of De Moel and colleagues, 2011. Their paper was not re-read. Availability Heuristic would be a rare flood that is hard to call to mind. These papers did not give that recall test. Ludy and Kondolf’s abstract, Natural Hazards 61, pages 829–842, 4 January 2012, says land behind a levee certified against the 100-year flood is officially outside the floodplain, so residual risk is not disclosed as floodplain. In the Sacramento–San Joaquin Delta, land below sea level can be open to building because levees “protect” it. The PDF was not read. The abstract does not print a response count.
2

Protected land can fill

The 2018 paper says building levees, or other structural protection such as flood-control reservoirs, is often tied to heavier urbanization behind the protection. More people and assets are then exposed to floods that are less frequent and can be catastrophic. Its summary of De Moel and colleagues says the urban area in the Netherlands that can potentially be flooded increased six-fold in the 20th century. After the catastrophic flood of 1953, the share of urban area in flood-prone areas dropped, then grew in recent decades from about 27 percent to about 31 percent as works such as the Delta Works raised protection. That growth brought economic benefits and offset part of the risk decline from stronger protection. De Moel was not re-read. Do not treat the six-fold figure and the 27-to-31 share as one statistic.
3

Some answers treat the works as enough

Di Baldassarre prints four means from De Marchi and colleagues, 2007, a report this page did not re-read. Informed respondents in northeastern Italy scored statements from 1 to 5. A 5 is strong agreement. The means are 4.49 (the works give a feeling of safety), 3.92 (they eliminate the possibility of serious damage), 3.48 (they help the local economy), and 1.76 (they cost too much for the benefits). The 2018 paper does not print how many people answered. It reads 3.92 as very little awareness of residual risk, and says people behind levees there were not well motivated to take private precautions. Zero-Risk Bias is a preference for driving one risk to zero on a different task. The 3.92 is agreement with a sentence. It is not that ranking.
4

Precautions do not fall everywhere

The same section cites Germany and France, where households in protected areas took more private mitigation than households outside them. In the German insurance study, they were also more likely to hold flood insurance, which the 2018 paper ties to that insurance system. Those papers were not re-read. No percentage from them is on this page. A Dutch study cited beside them finds that people are mostly unaware how high the levees are rated, even though the rating is extremely high. Unaware of the level is not the answer “serious damage has been eliminated.” In the United States, the positive levee effect on urban expansion weakened and then reversed after the late 1970s. The 2023 abstract attributes that turn to rules that suppress floodplain building. It does not print the size of the reversal.

​
Applications

Four checks keep a decade without a flood from standing in for the flood the structure was not built to stop.

Write the design flood in the notice

When a levee or a flood-control reservoir opens, put the design event and what remains above it in the same announcement. Ludy and Kondolf’s abstract says a 100-year certification can take the land off the official floodplain. The remaining risk should not live only in an engineering file.

Compare the floodplain with the county

If you manage land, compare the pace of building on the floodplain with the county around it. The United States abstract’s pair is 62 percent and 29 percent, for levees. Those two numbers are not a target for a dam, a firm, or a city you have not measured.

Ask which private step stopped

Agreement that works “eliminate the possibility of serious damage” is not a count of unused insurance. In the German study the 2018 paper cites, protected households were more likely to hold flood insurance. Write the question, the rule system, and what was counted.

Mark the plan apart from the surprise

Di Baldassarre’s report of Hallegatte, 2017, says one goal of protection is economic activity on valuable flood-prone land. That book was not re-read. Write whether the new buildings were inside the plan. The 2018 paper says growth past the plan depends on communication and politics, not on height alone.

​
Case Study

In the contiguous United States, a paper in Nature Sustainability (21 August 2023, volume 6, pages 1578–1586) asks how floodplain urban expansion lines up with levee construction. The text read here is the abstract. The article body is behind a subscription and was not read. The authors use a national levee database and annual land-cover maps. The data note says those historical maps run from 1938 to 2005. The author list on the journal page was collapsed and was not extracted. The abstract reports an association. Levee construction is associated with a 62 percent acceleration in floodplain urban expansion, outpacing the county’s 29 percent. The authors call that gap a clear change in risk perception after levees are built. The 62 percent is an acceleration of the urban-expansion rate. It is not a head count. It is not a damage rate. It is not a result for dams. The same page lists a national dam inventory among its data sources. That inventory is not the 62 percent. Over time, the positive levee effect weakened and then reversed since the late 1970s. The abstract attributes the turn to legislative regulations that suppress floodplain urban expansion. It does not print how large the reversal was, and it does not name the statutes. It also says regions that lacked strong momentum for population growth, and that had frequent floods, tend to rely more strongly on levees. It suggests a more diversified portfolio there. No region name is in the abstract. The lesson stays inside those sentences. A structure that cuts ordinary flooding can sit beside faster building on the floodplain. The authors read the faster building as a change in perceived risk. They do not report a survey that scored the same households before the levee and after it. A rule aimed at the building can weaken the pattern. The reversal is the boundary on treating 62 percent as permanent. A separate Stockton subdivision, in the 2012 abstract, did not understand its flood risk. That sentence has no sample size and is not this 62 percent.

​
Boundaries and Failure Modes

The pattern fails where protected households prepare more, not less. Di Baldassarre’s German and French citations are that boundary. The late-1970s reversal is another: the positive land pattern is not a law. Hallegatte, as the 2018 paper reports him, is a third. Extra losses from a rare flood can sit beside investment benefits, and those benefits can be the reason the structure was built. Planned use of protected land is not, by itself, the failure. A common misuse is to move the numbers. The 62 percent and the 29 percent belong to levee-linked floodplain urban expansion in the contiguous United States. A flood-control reservoir is in the 2018 class of structures. It is not the object of that acceleration. The four Italian means have no respondent count on this page. Stockton’s abstract gives a median household income of $80,000, says 70 percent of respondents had a four-year degree or higher, and says the subdivision was ethnically mixed. It then says that, despite education and income, they did not understand the risk of being flooded. It does not say what share misunderstood, and it does not say how many people answered. Do not paste 62 percent, 4.49, or 70 percent onto a dam, a firm, a shop, or a family.

​
Common Misconceptions

Three retellings travel with the wall. One renames the journals. One turns a land rate into a memory rate. One treats every protected household as unprepared.
The 2018 abstract prints levee effect, safe development paradox, and safety dilemma. The 2023 title uses levee effect. Neither text prints Dam Effect. This page uses Dam Effect for the drop in vigilance after protection. The journal word stays attached to the papers that measured land and, in the Italian means, agreement about safety.
The 62 percent is an acceleration of floodplain urban expansion associated with levee construction. The county figure beside it is 29 percent. The authors read the gap as a change in risk perception. That reading is theirs. The abstract is not a survey of residents who had forgotten a flood, and it is not a damage rate. A national dam inventory is listed in the data note. It is not the 62 percent.
The Italian means include 4.49 for a feeling of safety and 3.92 for eliminating serious damage, among informed respondents, with no printed sample size. Di Baldassarre reads 3.92 as little awareness of residual risk. The same section says German and French studies found that protected households can take more private mitigation, and that the German insurance study found they were more likely to hold flood insurance. Stockton’s abstract says one new subdivision did not understand the risk. It does not print a percent.
The Dam Effect sits beside a safety rule that changes behavior, beside a wish to wipe one risk out, beside a calm treated as normal, beside a rare event that is hard to recall, and beside a rosy personal forecast. None of those pages owns the 62 percent.

Peltzman Effect

Offsetting behavior that shrinks a mandated safety gain because the protected person takes more risk. The Dam Effect is lower preparedness and more building behind a structure. No belt elasticity belongs here.

Zero-Risk Bias

A preference for driving one risk to zero. The Italian 3.92 is agreement that protection works eliminate serious damage. It is not that ranking task.

Normalcy Bias

Treating a long calm as the way things are. Years without a flood behind a levee can look like that. These papers do not score the bias.

Availability Heuristic

Judging by what comes to mind easily. A rare flood is a candidate. The 2023 abstract counted land cover, not recall.

Optimism Bias

Expecting a better personal outcome than the odds support. Stockton’s abstract says residents did not understand the flood risk. It does not print an optimism score.

​
One-Line Takeaway

When a structure keeps ordinary floods out, write the flood still above the design. The quiet years do not replace that sentence.