THE 100 YEAR FLOODPLAIN
Misunderstandings, the Facts and Actions Needed
THE SITUATION
The term, "the 100 year floodplain" may be the most mis-understood concept ever developed. The common interpretation is that this is the area where a flood is not expected to occur within the next 100 years. This is completely and even dangerously wrong. This view is responsible for much development in areas which should not be so used, and, in consequence, flooding problems. In turn, corrections to flooding leads to development of increased drainage, to unnecessary expenses, to waste of water resources, and even to aquifer damage.
Corrective action is needed. The first part of this must be education
of decision makers on the true meaning of the term, and of it's proper
use. Probably some education on the resulting costs of improper use are
also needed. Action against applications involving the 100 year flood plane
in particular will be needed, since several laws use this concept . And
finally, changes in current law and interpretations will be needed to stop
future damage.
MEANING OF "THE 100 YEAR FLOOD"
The water level in interior waters such as rivers and lakes, and their smaller cousins, streams, rivulets and ponds normally changes with time, largely as a result of rainfall variation. If the water level has been measured over a period of time, statistical analysis can be used to describe the way the level changes. Then, with the assumption that there have been no appreciable changes in influencing factors, the statistical results can be used as an generalized prediction of what can happen in the future.
Of the measurements of water level, the highest level in a given year is called "the flood". The record of floods over a period of time is the flood distribution, which is analyzed by some selected statistical method. Now for interior water bodies there is a minimum level, the stream, lake, etc bed. Thus the statistics used must be based on the fact that the distribution is limited on one end by zero: negative levels do not occur. There are several statistical approaches which can be used. One found to be suitable uses a special graph paper, called "extreme value paper". Its major horizontal axis is non-linear, being calibrated by the mathematical expression, x=exp(-exp(-exp (y))), where x is a probability and y is derived from the number of measurements. A second horizontal axis is derived from x and is called the return period: a probability of 50 percent is a return period of 2 years, a probability of 10 percent is a return period of 10 years, 1% 100 years. and so on. The vertical axis is linear, and is the flood height. There are equivalent computer programs, but they tend to hide the process.
To use the paper, the flood record is ordered, from the smallest value to the largest. Each is assigned a probability: if there are N records, the first probability is equal to1/(N+1), the second to 2/(N+1) and so on to N/(N+1). Each pair of points, flood height and probability can now be plotted. It will be found that that the points make a definite almost straight line, although sometimes one point will be off this. (The reason for this is that more data was really needed, to fully define the line). A statistical method called "Least Squares Best Fit" can now be used to give the best possible straight line describing the data.
Strictly, this line should stop at the last point plotted. However, in many to most situations, the data period will not be long enough to reach a desired point, such as the 1% or 100 year line. The practice is to extend the calculated best fit line, usually to the edge of the page, beyond the 0.1% or 1000 year point. This is an extrapolation, and these always reduce the accuracy.
There is another factor: There is a 50 percent chance that the true
return period is less than indicated by the fit line, and equally a 50%
chance that it is longer. For extrapolated lines, two points that are easy
to remember are that there is a 16% chance that the indicated 100 year
flood height could occur within the next 31.6 years, and equally, a 16%
chance that it will be delayed for 316 years. the spread is less if the
data duration exceeds the time projected. The 100 year flood could even
occur this or next year.
DEVELOPMENT OF FLOOD PLAIN DATA
The flood height data developed from the analysis needs to be transferred to the local terrain geography. Using the 100 year estimated flood height as an an example, this height is marked on a geological survey map, on all of the contour points of that height. For a lake or pond, the result is a closed shape enclosing the water body. For a stream, there will be a line on either side of the water. The area within the contour, and the area between the lines which includes the stream water is the flood plain, in this example the 100 year flood plain. The 100 year flood plain and other contours are assembled by the Federal Emergency Management Agency into "Flood Hazard Maps", which are really intended to establish insurance rates. They are also used by local governments in decision making (too many times inaccurately): the maps are available from FEMA in paper and CD form, and in the planning departments of cities and counties.
There are factors which need to be considered in using these maps. Interpolation is used in their preparation as necessary to estimate the exact positions: surveys may have contour intervals such as 2 or 5 feet in flat country, more in mountains. This can be a problem in central Volusia County, where the land height may change only one foot in a mile of distance. And note that a 100 year flood contour may not enclose an identified water body: a low area with poor natural drainage can fill with water during wet periods. Again, this can be a problem in central Volusia County, which is one reason for the 100 or so miles of drainage canals in the County.
There is another important factor here. The flood level reached in the 50 year return time is lower than the 100 year flood, and still lower for the 25 and 10 year floods. As a result, the the contour lines for the 100 year flood encloses all areas 'more frequently flooded, and usually encloses a stream or lake. Unless additional statements are included with "the site is in the 100 year flood plane", there is no statement of the frequency of flooding expected: it may be as often as each year. Omitting these additional statements, such as "The site is at the 25 year flood contour" is the source of much of the flooding problems encountered.
Decision bodies and their staffs have a record of ignoring this fact,
never asking what the return period is at the site being considered. The
result is flooding at the next period of heavy rain, loud complaints, and
action to "improve drainage". Almost always this is to move the water as
rapidly as possible to a lower point, a large lake, a stream or a canal.
An existing stream may be straightened, a canal widened or deepened, or
even new canals dug. The consequence is loss of water to the system, often
to the aquifer. visit the Tiger Bay canal during a rainy season to see
this.
ADDITIONAL FACTORS
Current law allows construction in the 100 year flood plane if the fill
required does not increase the flood height by more than one foot. Data
and calculations to support this are rare, or even nonexistent in some
administrations. Further, the cumulative effect of several such construction
projects is always ignored. A result is flooding of adjacent or nearby
properties, and even extension of the 100 year contour towards the 500
year line.
QUESTIONS TO ASK REGARDING DEVELOPMENT
The first question is: have there been changes which render the "flood map" inaccurate? Such actions as construction of a road, clearing a large area of forest, construction of a major shopping center, or paving a large area can change flood levels. The effect should be determined.
The second question is: what is the cumulative effect of previous developments on increasing flood levels? This is probably difficult to determine, since flood height change seems to be rarely determined for any development, road, etc..
Usually, the final question is: what is the flood return period and projected flood height at the site being considered, taking into account the results of the earlier questions?
But if it appears that a flooding problem could arise, is seems necessary to raise the question of costs: It does not seem proper for the public to pay for the necessary corrective action, so some agreement or even a bond seems to be indicated.
Unless these questions are answered fully and show that there will be no flooding problems and therefor no need for increased drainage, or that the public will not be saddled with flood costs, the position should be that any decision should be postponed until the answers are available.
And if there are flood problems which clearly require increase drainage
and consequent water loss, the project should be opposed, with careful
statement of the reasons for the opposition.