Strömsbergs bruk is the location of an old ironworks in Uppland, polluted by many of the metals often connected to iron ores and the waste products of iron production. One of the pollutants found in high concentrations at several samplingpoints at the site is Arsenic, wich will be the focus for this Bioscreen modelling. An old slag heap in the area has been identified as the likely pointsorce for the arsenic and the Bioscreen modelling is run to study the potential spread from that point, to the nearby stream Tämnarån. The most valuable fieldmeasurement used is of the groundwater As concetration in a groundwater pipe, between the slag heap and the stream.
The groundwaterlevel in the area around the site is found at between 1,7 and 2,5 meters depth with a hydraulic gradient of between 0,1 and 0,5 m/m. The soils consist of some sections of glacial mud but is dominated by fill material in the fraction sandy gravel, this is also partly made up of brick fragments. The coarseness of the fill material in large parts of the area makes it a permeable ground and therefore should have the risk of fast spreading pollutants. According to the Bioscreen manual a clean sand has a conductivity of between 0.01 - 1 cm/s, and a gravel a conductivity of over 1 cm/s. The consultants on the site estimated the hydraulic conductivity in the ground at Strömsbergs bruk to 0.01 cm/s despite the fact that the material was described as gravel. Both the estimated values from the field, and the standard sandy gravel value from the manual, are run to see how sensitive the model is to changes in this parameter. To see how the result might vary with changes in porosity and hydraulic gradient, scenario 1 is also run for a top- and a bottom tablevalue for these parameters.
Discussion:
The different scenarios show very little variation in concentration. This is somewhat strange considering especially the hydraulic gradient that changed considerably between scenario 1 and 2. This could lead us to believe that the groundwater flow is not greatly affected by this, maybe because of a high hydraulic gradient or the fact that even the low value used in scenario 2 still does not affect groundwater flow since it maintains the same value for porosity. But both these theories can be excluded by the sensitivity tests that show changes to these parameters alone only marginally changes the result.
Neither of the scenarios seem to match the field measurement of 0,31 mg/L 60 meters from the source. This can have several explanations, many having to do with lacking proper values to put into the model and the suitability of the site for this type of modelling focusing on a point source. What is meant by the latter is the fact that the field measurements are gathered long after the actual ore pile was first dumped in the location, making it very likely that the leaching of As from the pile has been extensive and ongoing for a long time. This leads to a lot of the arsenic, that originate from the point source, has now infiltrated the ground with leach waterwhere it has been adsorbed. This complicates the situation since the point source is now no longer a single point source, but leaching from the ground itself is now the main contributor to the concentrations in groundwater. Also the levels of arsenic found in the groundwater pipe are very low, making it more difficult to model.
To improve these results more field measurements of the old ore dump is needed. The arsenic content of the actual slag pile and leaching tests is needed to have accurate values to put into the model, as well as values from the surrounding ground. This would make it possible to study whether the old pile is in fact a point source any longer, and if not a different type of modelling is needed to evaluate the potential spread of arsenic in the area.
Image: The area around Strömsbergs bruk with the old slag heap circled in red and the location of the groundwater pipe measurements circeled in blue.
Scenario 1: from field estimations (see figure 1)
Hydraulic conductivity: 0.01
Porosity: 0.15
Gradient: 0.03
Dispersion: Estimated plume length: 100m / 328 feet
Scenario 2: from Bioscreen manual table (see figure 2)
Hydraulic conductivity: 1.0
Porosity: 0.35
Gradient: 0.03
Dispersion: Estimated plume length: 100m / 328 feet
