The UAE plans to launch multiple satellites in the near future for space exploration. The satellite data are expected to bring significant benefits to many industries including civil and infrastructure engineering. Currently, digital elevation models (DEM) and satellite images have been used to obtain geologic and soil information for construction sites when soil boring data are limited. This trend will be further facilitated when the satellite data become available from different agencies. However, the data volume from satellites are very large, and the user needs to process these data automatically in order to be able to extract reliable results.
The aim of this study is to develop data processing technology to characterize the geologic and geotechnical site conditions in Abu Dhabi by using available DEMs and satellite images. These data are used in combination with site exploration data in Abu Dhabi (i.e., 600+ borings and several laboratory and in situ tests) available to the authors from ongoing collaboration with the spatial division of the Municipality of Abu Dhabi City. DEM data will be collected from publicly available data resources, such as United Sates Geology Survey Shuttle Radar Topography Mission (USGS SRTM). In addition, many satellite images are publically available (e.g., USGS Landsat).
The project will make use of these existing data in collaboration with Japanese researchers from Hiroshima University, and analyze and treat them to be used for geotechnical and geological applications. These data are combined with the soil exploration data in Abu Dhabi, to develop prediction models of subsurface soil properties from satellite data. Data analysis uses machine-learning techniques under the open platform of Python and R to develop and determine the best prediction models of subsurface soil properties. This approach will be useful in the city of Abu Dhabi, which is under expansion, in particular the areas where the soil exploration data are limited. Moreover, the technology will be applicable to different locations where only the limited boring data are available but the construction of the infrastructure is required.