Quarterly Newsletter of the Finnish Highway Transportation Technology Transfer Center, FinnT2
Address: Finnish National Road Administration, FinnT2, P.O. Box 33, 00521 Helsinki, FINLAND
Fax Int. 358 204 44 2675. E-mail: [email protected] Editor: Arto Tevajarvi, Tel. Int 358 204 44 2032
Editor-in-Chief: Jarmo Ikonen, Tel. Int. 358 204 44 2118
MATHEMATICAL MODEL FOR GROUNDWATER CONTAMINATION
DUE TO SALT
In the late eighties and early nineties, the
potential contamination of groundwater due to
salt from highway de-icing became an issue of
concern in Finland. To obtain quantitative
estimales about the extent of this potential
problem at present and in the future, a joint
research programme between The Technical
Research Centre of Finland and The Nationai
Board of Waters and Environment was
launched in 1992, with Finnra as the main
funding agency. The objective was to obtain
order-of-magnitude information that can be
used to assist decision making when looking
for socio-economically optimal approaches to
the winter maintenance of Finnish highways.
FORMING MATHEMATICAL MODELS ON SALT SPREADING
IN THE GROUND
Mathematical modeling along with extensive
field measurements and analysis of existing
field data are used as main methods of
investigation.The investigations include both
detailed modeling of two well-investigated
sites with chloride contamination from the
adjacent highway and more general modeling
studies representing typical groundwater
conditions in Finland. The latter investigations
are supported by statistically analyzed data
from several hundreds ot groundwater formations
in Central and Southern Finland.
In modeling the spreading of salt in representative
average conditions, example groundwater
formations are studied. The formations,
even though synthetic, are chosen to represent
conditions typical of the Finnish groundwater
formations vulnerable to this type of
pollution. A nation wide risk-assessment
study, in which several hundreds of groundwater
formations were classified and analysed
for their risk of pollution from highway salting,
was used as a basis when determining these
example formations.
First, the history of salt spreading from its
beginning to the present is modeled. The
results from 1992 show the anticipated behaviour,
with high concentrations near the road,
decreasing towards the sides of the formation
and towards the water intake plant. An exception
is the formation which collects salt-rich
water from its surroundings due to pumping
exceeding the natural yield of the formation.
The water intake plant has a diluting effect on
chloride concentrations, and the concentrations
observed at the plants are considerably
lower than those in the vicinity of the roads.
The size of the formation is a significant factor
affecting chloride concentrations; in the example
case, the chloride concentrations at the
water intake plant were about 100 mg/l in a
minimum size formation, whereas in a similar
medium size formation they were about 20
mg/l.
EVALUATION OF MODELING
The reliability of the 1992 results is evaluated
through comparison with data from 300 existing
formations from which the different factors
affecting the spread of the chioride are analysed
statistically. Both the general patterns of
chloride distribution as well as the actual
concentrations agree well. Simulated chloride
concentrations in medium size formations at
the water intake plant were usually about 20
mg/l, varying between 16-34 mg/l, the size of
the formation being the most significant factor
affecting the concentration. In the statistical
data the concentrations were of the same order
of magnitude, the average concentrations
varying between 16-26 mg/l and the median
values between 10-23 mg/l. Taking into
account the approximations involived in both
the modeling and the statistical study, the
agreement is good and the modeling results
can therefore be considered as being of the
correct order of magnitude. The models can
then also be used for making order-of-magnitude
predictions of future developments.
IMPLEMENTATION OF MODELING IN SIMULATIONS
In prediction simulations from 1992 to 2022
three different scenarios are studied: 1) saltinq
will be continued at the high level of the early
nineties (9.6 t/kma), 2) no salt will be used,
and 3) the amount of salt used will be reduced to
about half of the high values (5.0 t/kma). If the
application of salt continues at
the highest level, the concentrations at water
intake plants in medium size formations will
increase to about 30-50 mg/l in 30 years, with
much higher concentrations being observed in
the vicinity of the roads. During the same
period, the values at the water intake plant in
a minimum size formation will increase up to
170 mg/l and will be highest near the vicinity
of the road, at around 240 mg/l. lf no salt is
used, concentrations usually fall below 10
mg/l in 30 years. In the case studied, decreasing
the amount of sait used to about half
causes either an increase of a few units or, in
the case of the minimum size formation, a
decrease of a few units.
Based on the results of the simulations,
order-of-magnitude estimates can be made about
average spread and development of chloride
contamination in groundwater formations of
different type and size. The results are approximate
and the trends observed in the
results can be used as supplementary information
when making estimates about the
current status and future development of
chloride concentrations.
The site-specific modeling studies also included
in the collaboration project in
question are still going on. The results show
good agreement with the aforementioned
results concerning the average trends and
spread of chloride. Valuable additionai information
is, however, expected from the site-specific
studies concerning the effect of local
geological heterogeneities and seasonal
variations.
AULI NIEMI, Technical Research Centre of Finland
Contents of FinnContact 4/1995:
Top
[Paluu etusivulle]
[Back to main page]