Calibration of a Terrestrial Laser Scanner - Institute of Geodesy and ...

112 6. Applications of Terrestrial Laser ScanningFigure 6.4: Inlet gutter and vegetated road shoulder (left), watershed and initially assumed catchment area (right).Flow direction due to preferential flowbeneath the road is indicated.In the pilot plant, the road runoff was collected in a storage tank using a 0 5 m long gutter located directlybeneath the road. The volume was read periodically, cf. Figure 6.4, left.The rainfall height was measuredat a distance of 6m from the road. The first calculations for a catchment area of 15m2, 0 5m gutter lengthtimes 3 m road width, were not meaningful because too much water was collected in the sampling tankrelative to the road surface of 15 m2. The catchment area must be much largerthan assumed. This wouldbe the case if the road would have a slight slope in the longitudinal direction or other effects would occur,such as preferential flow.In order to assess the latter, simple tracer experimentswere conducted. Theserevealed significant water flow from the area adjacent to the road into the inlet gutter of the water storagetank. The length of this section is 2 5 m to 3 5 m and the watershed can be located within a range of 1 m,cf. Figure 6.4, right. Although this tracer experiment answered the question that there is some preferentialflow, the questionof the size of the catchment area is still unsolved.Calculation of Catchment AreasThe catchment area was calculated by using the noise-reduced and geo-referenced laser scanning data.Based on these data, a grid with a sampling interval of 8 cm, the streams leaving the road, cf. Figure 6.5,as well as the different catchment areas for the inlet gutter, cf. Figure 6.6, were derived. These calculationsshowed that water flow is almost perpendicular to the road alignment. In addition, sensitivity analysesshowed the point density for deriving the grid is sufficient and a decreased point spacing of the grid doesnot significantlyaffect the size of the different catchment areas.As measured by the tracer experiments, the assumed location of the watershed lies within a range of about1 m. Based on the coloured tracer experiment it was not possible to estimate the watershed more accurately.For this reason, it was assumed that the watershed is located in the middle of the range at 0 5 m. Conse¬quently,in Figure 6.6.the two different catchment areas number 1and 2 lead to the catchment area of 14 1 m2, as shownBased on the catchment area of 14 1 m2, the maximal runoff was calculated and was compared with the col¬lected road runoff. As a result, 36 % of the maximal runoff drains directly into the storagetank and thereforewould infiltrate into the vegetated road shoulder. The other 64 % of the road runoff are dispersed mostlythrough diffusion within a distance of the road of 25 m [Steiner et al., 2005]. As the heavybe assumed to be nearly proportional to the water flow, about 36 % of the pollutantmetal load canload in the road runoffinfiltrate the vegetated road shoulder directly below the road. During infiltration, many pollutantssuch asheavy metals and organic substances are accumulated. Thus, accumulation results in a high concentration