Effect of the inlet-outlet positions on the hydraulic performance of horizontal subsurface-flow wetlands constructed with heterogeneous porous media.

Effects of the inlet and outlet position on flow patterns of saturated fluids in a horizontal subsurface-flow constructed wetland were investigated experimentally using a quasi two-dimensional flow cell representing a vertical cross-section in the longitudinal direction of the wetland. The filter medium consisted of glass beads that were either uniformly sized or a mixture of sizes with a distribution corresponding to wetland filter media. Flow through the filter bed was visualized by injecting dyed fluid into the water-saturated model. Next, breakthrough curves were obtained using chloride tracer. Flow through the homogeneous filter formed by uniformly sized beads displayed a clear density-driven component. Using mixed sizes, finer and coarser beads tended to separate into alternating and incomplete layers. Flow occurred preferentially along roughly horizontal high-conductivity paths of coarser filter material. Density-driven vertical flow was much slower than the horizontal flow. Nevertheless, appropriate positioning of the inlet and less importantly the outlet could to some extent mitigate the dominant effect of the medium stratification on the flow patterns. Using inlet-outlet configurations that forced the flow through larger portions of the filter bed by injecting into low-conductivity layers and opposing the gravity-driven flow increased the treatment efficiency.

The effect of the scale of horizontal subsurface flow constructed wetlands on flow and transport parameters.

Horizontal subsurface flow constructed wetlands have proven their efficiency in treating wastewater and removing the pollutants of concern. Treatment efficiency depends on the wastewater residence time, which is a function of the hydraulic loading and the physical conditions of the constructed filter system, which can be described with effective parameters such as: hydraulic conductivity, porosity, dispersivity etc. Because spatial variability is often scale dependent, these effective parameters may be affected by the scale of the system being studied. In this paper the results of tracer experiments in constructed filters using saturated horizontal flow at three scales (small and medium lab scales and full-scale system) using the same filter media is reported. Light-weight aggregate (filter media termed Filtralite-P) was used at all scales. Increasing the scale was associated with increasing dispersivity, meanwhile hydraulic conductivity experienced dramatic reduction and variation by increasing the examined scale. Observed changes in the hydraulic parameters indicate that heterogeneity at different scales should be taken into account when the performance of LWA filters are evaluated from small-scale experiments.

Loss of pesticides from agricultural fields in SE Norway--runoff through surface and drainage water.

Loss of two pesticides with different mobility characteristics, bentazone (Koc approximately 34) and propiconazole (Koc approximately 1800), were studied at three agricultural fields (Askim, Bjørnebekk and Syverud) in SE Norway. A conservative tracer (Br) was used to follow the flow of water. The loss of pesticides varied among the fields, depending on hydrological characteristics and soil properties. The loss of pesticides was higher from two artificially levelled silty clay loam soils with poor aggregate stability (Askim and Bjørnebekk) compared to a loam/silt loam soil with increased content of organic carbon and better aggregate stability (Syverud). The total accumulated loss was <0.5% from all fields. The highest pesticide concentrations were measured at the first runoff episode after application for both the mobile (bentazone) and less mobile pesticide (propiconazole) in the surface runoff. In the drainage water, the peak for the less mobile pesticide coincided with the Br tracer, while the peak for mobile pesticide appeared earlier than the Br tracer. Rapid movement of water, particles and pesticides through soils indicate flow through macropores. Larger proportions (in percent of total applied) of both the mobile and the strongly sorbed pesticides were lost through the drainage as compared to the loss through surface runoff at Askim. Here, it is suggested that macropore flow contribute to the increased loss of pesticides through the drainage. At Syverud, high infiltration capacity reduces the amount of water available for surface runoff, and somewhat higher loss of the mobile pesticide was registered in the drainage compared to the surface runoff. For the strongly sorbed pesticide, however, propiconazole was neither detected in surface nor in drainage water at Syverud. Generally, there was a higher percentage loss of the mobile compared to the strongly sorbed pesticide in both surface and drainage water, which is in agreement with the pesticides mobility characteristics in soil. An exception was, however, the erodible soil Bjørnebekk, where a higher fraction of propiconazole was lost in the surface runoff compared to bentazone. Large amounts of sediment transport from the Bjørnebekk field probably contributed to enhanced transport of the strongly sorbed pesticide.

Influences of suspended particles on the runoff of pesticides from an agricultural field at Askim, SE-Norway.

Field and laboratory experiments were conducted to study the loss of particles from agricultural fields, and the role of suspended particles in carrying pesticides in surface runoff and drainage water. Propiconazole, a widely used fungicide was applied to experimental fields located at Askim, SE-Norway. Samples from surface runoff and drainage water were collected and analyzed for sediment mass, pesticides, particulate and dissolved organic carbon through a whole year. The surface soil and the runoff material were characterized by its particle size distribution, organic carbon content in size fractions and its ability to bind propiconazole. The results show that (1) particle runoff mostly occurred during the rainfall event shortly after harrowing in autumn. The highest particle concentration observed in the surface runoff water was 4600 mg l(-1), and in the drainage water 1130 mg l(-1); (2) the erosion of surface soil is size selective. The runoff sediment contained finer particle/aggregates rich in organic matter compared to its original surface soil; (3) the distribution coefficient (Kd) of propiconazole was significantly higher in the runoff sediment than in the parent soil. According to our calculation, particle-bound propiconazole can represent up to 23% of the total amount of propiconazole in a water sample with a sediment concentration of 7600 mg l(-1), which will significantly influence the transport behavior of the pesticide.

Small-scale variation in deposition of radiocaesium from the Chernobyl fallout on cultivated grasslands in Norway.

Soil samples from nine field experiments on cultivated grassland showed a variation in deposition of radiocaesium, expressed as the relative standard deviation, from 11 to 111% within an area of 16.5 x 10 m2. At the site with the highest variability, situated in a mountainous region of south eastern Norway, a further study was performed. Double sampling including surface measurements with in situ gamma-spectrometry using a portable multi-channel analyser (3 x 3 in Nal detector) and limited soil sampling gave the same average deposition and standard error as an extensive soil sampling. There were no trends in the deposition within the experimental field and most of the variation was found within plots (1.5 x 2.5 m2). 'Hot spots' were found in connection with small depressions (2-5 cm) in the terrain, and the area of a hot spot was estimated to be 100 cm2 from field measurements with a Geiger-Müller counter.

Different approaches for estimating the deposition of radiocaesium on mountain pasture in southern Norway.

Different approaches to the estimation of radiocaesium deposition on different types of indigenous mountain pastures have been examined. In situ gamma spectrometric measurements performed using a portable multi-channel analyser (3 x 3 in Nal detector) with the detector placed on the ground, on a tripod at a height of 100 cm above the ground or fastened to a rucksack frame 90 cm above the ground were compared with conventional soil sampling. Field gamma-ray measurements in connection with soil sampling appear to be a promising tool for estimates of deposition on mineral soils in areas with different topography and vegetation.