Geoscience Reference
In-Depth Information
7.1.1.
Forecasting and predetermination
Depending on the aims of studies carried out in this field, a hydrologist can be
faced with many different problems associated with the forecasting and
predetermination of water levels. More often than not, hydrologic analyses are
carried out with the aim of creating hydraulic structures. The analyses are therefore
carried out and formulated in such a way that their results can predict the probability
of failure of the hydraulic structures. What happens if we want to create a rainwater
drainage network that overflows, on average, 10% of the time? The aim of the
hydrologist here is to determine decennial flood levels. There is one chance in 10
that these levels will be exceeded in any year. As far as the protection of a dam is
concerned, the risk of it exceeding its limits 10% of the time is much too high due to
the disastrous effects this would have on human life. This figure of 10% therefore
needs to be reduced, and in order to do this, it is necessary to study the flood levels
in the area for the past 1,000 to 10,000 years [GUI 67].
Regarding water supply, it is often necessary to work on improving the
structures that are already in place (for example, straightening and strengthening the
dams for storing water) [ALE 96]. Drought or rationing irrigation are then translated
into economic terms to be balanced with necessary investments. Several
comparative studies have been carried out in this field and we ask the hydrologist
volume adjustments for a wide range of failure rates. This type of problem is part of
the category known as predetermination in which a hydrological phenomenon (such
as flood, annual water supply, etc) is associated with a probability level.
The second problem that occurs is the problem of forecasting. Problems
associated with forecasting tend to occur less than the problems associated with
predetermination. As far as forecasting is concerned a particular risk is not
associated with a particular phenomenon, but a particular phenomenon is associated
with a particular date: for example, what will the water level of the River Seine in
Paris be tomorrow? The problem associated with forecasting is a management
problem that cannot be applied when hydraulic structures are being developed.
However, statistics play an important role as far as forecasting is concerned because
it is necessary for a forecast to be associated with a confidence index. The answer to
the question, what will the water level of the River Seine in Paris be tomorrow, is
more than likely to be: 950 m
3
/s and there will be a 70% chance that the water level
will be between 880 and 1,100 m
3
/s.
The aim of this short introduction is to make the reader aware of the different
problems that a hydrologist has to deal with, and to make the reader aware of the
fact that these problems are linked to statistics that provide information on a long
series of water level observations. Such information is not very widely available and
this justifies the fact that the field of hydrology is becoming increasingly closer to
the field of climatology in which temporal distance is much greater.
