The goal is to generate a soft ground risk map. To generate this map the following information is required: - Topographic information
- Terrain information
- Lithology, geology and structural properties of the surface
The topographic information from the SRTM DEM provides the input to elevation and slope mapping. The terrain, lithology, geology and structural properties are provided by Landsat 7 ETM. Quickbird data (2006) were used to complement the multispectral images. All EO datasets were clipped to a common area of interest. |
|
The Landsat 7 ETM data were merged to create RGB composite images with different band combinations used to highlight different aspects of the surface. The data is stored in a GIS database (summarised in the diagram above). Utilizing previous studies and gaining knowledge of the area will significantly improve the results. Studying the DEM reveals some slight variations in the topography; inland depressions could mean the occurrence of sabkha, where water could pool (Figure 1). Figure 1: DEM (left) and Landsat 742 RGB (right) for the study areaThe findings from the EO data were calibrated with surface geological probing at selected locations with focus on the edge of the sabkha. Mineralogical indicators (right), like the presence of discoidal gypsum, were used to identify a water table close to the surface. The occurrence of discoidal gypsum may be related to the locations where sabkha is thought to exist. Field verified points are correlated with the Landsat 7 ETM data and a supervised classification forms the basis for the sabkha risk maps. Plotting the locations where vibrator trucks got stuck and interpolating the vibrator sweep performance data to a raster image, such as distortion data, allows us to overlay this information in aGIS.The findings from the remote sensing and surface geological data are calibrated. The correlation between the sabkha risk map (left) and the vibrator total harmonic distortion (THD) attribute map (right) is shown in Figure 2. Figure 2: Sabkha risk map from unsupervised classification of Landsat satellite data (left), Vibrator distortion map (right)The comparison of multispectral Landsat data with known field measurements, such as known locations of sabkha and geotagged images, makes it possible to identify areas of sabkha. Analysis of all bands showed the clarity in a combination of thermal infrared, short-wave infrared and visible data, which enabled the best multiband image to be created. Anything appearing in red or orange in the sabkha risk map (figure 2 left) has a high or medium risk of sabkha. The impact on the data quality is obtained from vibrator QC attributes such as total harmonic distortion. The gridded QC attributes are displayed as a map, which is correlated with the risk map obtained from remote sensing. High distortion is correlated with the areas of high sabkha risk. Hence, the impact of sabkha on vibroseis operation is on both logistics and data quality. This suggests that we might select a different energy source for wet sabkha source points that cannot be accessed by vibrators. |
|
|