NATO Advanced Study Institute – Hungarian Geological Survey (1998 sept. 6-18, Mátraháza): Deposit and Geoenvironmental Models for Resource Exploitation and Environmental Security – Abstract volume

Árpád Ferenc Lorberer:

Investigation of a phenolic groundwater pollution at Nyírlugos village and the preliminary modelling of the remediation

Nyírlugos is a small village in Northeast Hungary. Some years ago groundwater pollution was discovered inside the village.

The pollution was caused by a mill, which existed there between 1930-1949. During the work as a by-product coal-tar was produced. This was pumped into a shallow well in dissolved form. The mill was destroyed in 1950 and the territory was planted with apple trees.

Nyírlugos village lies at the centre of the Nyírség ridge, which is one of the two regional hydrolgeological recharge area of the Great Hungarian Plain. This means, that the base groundwater-flow goes downward with I=0.01 m/m vertical hydraulic gradient. The village’s water supplying wells are pumping from the depth of 120-150 meters. The water of the wells showed high tricium-isotopic concentration, as a sign of descending young freatic water

All our regional examinations showed, that a really sensitive area was polluted in that site.

Several boreholes were drilled at the endangered zone. The geological build-up shows cyclical Quarternary sediments, sand, aleurolite, and clay. This situation is typical for the entire 300 meters of the Pleistocene layers in this area.

Three thin clay layer separate four distinct sandy-aleurolitic aquifers.

The pollution was discovered until the depth of 20 meters in an area of about 120 square meters. The second aquifer showed the largest pollution, the third was nearly clear. A retardation coefficient of 30 could be estimated about the phenolic pollutant in that site.

The hydraulic heads in all the aquifers showed a south-eastward groundwater flow with a gradient of 0.002 m/m. The heads decreased downward between the aquifers, with the same gradient.

Pumping tests showed leakage among the three upper aquifers. They are separated in their hydraulic heads, but the depression of the wells could spread quickly upward across the layers.

The hydraulic parameters of the layers could be determined from the results of the pumping tests and from grainsize analyses. (See the next table.)

Table 1: Hydraulical parameters of the layers of the polluted area

We investigated the possibility of the flow of pollution into the local water supplying wells. To do this, first we made a model about the worse possible situation, where there is only one 150 m. thick layer, the groundwater flows into the direction of the wells, and no retardation exists. Even that model showed, that no pollutant can reach the wells in 50 years time. After that, modelling of the wellhead protection areas also showed that these are horizontally not extended, and not endangered at all by this pollution.

The BIOKÖR Ltd. planned to remediate the whole pollution until the end of 2001. The extraction of the polluted groundwater was planned using two or three wells in two steps. The strangely polluted water of the two upper aquifer must be pumped out in the first step, and than the third aquifer can be cleaned, if it will be necessary.

Using the known hydrogeological parameters, and the results of the pilot tests we made preliminary models to determine the depression cones of the wells. We modelled the same (260x170 m²) area for the same time-period using different programs. We modelled the depression of two pumping wells, with a yield of 25 m³/day in each, both recharging from the third layer (the second aquifer).

In the models we worked with six layers, with leakage between them, using the ARV and the Processing Modflow software. Both can calculate wellhead protection area delineation in stratified three-dimensional hydrogeological situations. This models can produce really good estimations in case when vertical flowage is important across the layers.

We modelled the area with the Flowpath, estimating that there is only one 26 meter thick unconfined aquifer. ( fig 1.)

Fig. 1: Capture zone and the path of the groundwater flow

The results showed, that these are semi-unconfined layers, with strong leakage among them (as it was discovered). ( fig 2.) Pumping from the third layer will act on the upper two as well.

Fig. 2: Modell results using the flowpath

All the models showed, that it is enough to use two wells. They will completely change the water of the known polluted area in one year time. The depression areas of the models were nearly the same, although these programs use really different mathematical estimations.

Abstract

A phenolic groundwater pollution was examined at Nyirlugos village. This village is in a pollution-sensitive territory at the centre of a regional recharge area.

The pollution was found in three aquifers in an area of about 150 m² until the depth of 20 meters. Sand, aleurite, and clay layers were found in that site. Leakage was found among the aquifers The hydraulical parameters and the retardation of the pollutant could be estimated.

Hydrogeological models showed, that the local water-supplying wells are not endangered by that pollution. Three models were used to calculate the depression of the future pumping wells. The results of the different models were nearly the same. They all showed, that the groundwater of the polluted area can be pumped out in a year’s time; and the area can be cleaned till the end of the century