After
hurricane Irene in 2011 and “superstorm” Sandy in 2012, it seems like every
year has storm that causes the “100-year” flood or greater. This term can be
confusing because it refers to a flood that has an average recurrence interval of 100 years, which means that in a given year,
it has a 1% chance of occurring not one that only occurs every 100 years. This
is why hydrologists prefer to talk about the 1% annual exceedance probability
flood.
So,
how do hydrologists determine how large the 1% annual exceedance probability
flood is? In part because our streamflow records are relatively short, it is
impossible to determine what is the “true” 1% annual exceedance probability
flood. But, we can use the available annual maximum streamflow records to fit a
statistical probability distribution. From this frequency curve we can estimate
the magnitude of annual maximum streamflow for a given exceedance probability,
as shown in the example below.
The key assumption of this method is that the historical
data given an accurate representation of the distribution of future floods. Within
the debate over the effects of climate change, this assumption of stationarity
has been questioned, if not outright rejected. Even ignoring the climate
question, this assumption is often problematic because of land use changes or
modifications to stream channels that affect flood flows. For example, flood control structures
such as dams reduce annual peak flows. Conversely, urbanization increases the
amount of impervious surfaces, increasing surface runoff and annual peak flows.
Accurate estimation of the 1% annual exceedance probability
flood is important because it is the standard used in the designation of
high-risk areas for the National Flood Insurance Program. As shown in the
figure from the National Weather Service below, floods cause billions of
dollars of economic damage each year.
The National Flood Insurance Program was established as a
way to help pay for and mitigate these costs. Homeowners living in the 100-year
floodplain are required to purchase flood insurance, which is not covered under
standard homeowner insurance policies. Often municipalities restrict
development in the designated floodplain or have stricter building codes to
make structures more flood-proof. Incorrect designation of these high-risk
areas may mean that many of these areas are inappropriately developed and
people living there are unknowingly at risk. When floods inevitably do occur
the damages are then unnecessarily costly, and taxpayers are often left to
cover the costs through emergency relief funds, Thus, properly quantifying
flood risks is in everyone’s best interest.
References:
Holmes, R.R., Jr.,
Dinicola, K., 2010, 100-Year flood–it's all about chance: U.S. Geological
Survey General Information Product 106, 1 p. http://pubs.usgs.gov/gip/106/
National Flood
Insurance Program: http://www.floodsmart.gov/






