Print this pageAAPG GEO 2010 Middle East
Geoscience Conference & Exhibition
Innovative Geoscience Solutions – Meeting Hydrocarbon Demand in Changing Times
March 7-10, 2010 – Manama, Bahrain
Integrating Remote Sensing and GIS Technologies for Surface Geotechnical and Geological Analysis
(1) Qatar Petroleum, Doha, Qatar.
Whilst there is no substitute for the field work of experienced geoscientists in undertaking geological and geotechnical surface mapping, increasingly surface mapping can be undertaken using remote sensing technologies, together with GIS techniques, with greater confidence than in the past. Several major issues caused problems with early remote sensing techniques, especially using early Landsat satellites. Firstly, the poor resolution (80 metre pixel size) of the imagery and, secondly, the lack of multi-spectral bands specifically dedicated to the detection of radiation useful to discriminate surface geology.
Image processing techniques to correct and analyse remotely sensed data (such as principal component analysis, multispectral classification, image rectification and image enhancement) have not changed significantly in the last thirty years. However, there have been major advances in remote sensing and geodata information management. These include: improved satellite pixel resolution (from Landsat IV’s 80 metres down to 60 centimetres); the addition of remotely-sensed spectral bands specific to geological classification; and advances in computer processing power, image display and geodata integration through GIS.
In part, these improvements are made possible by advances in GIS. Advanced 2D and 3D display techniques adopted at Qatar Petroleum maximize the information that can be analysed and combined. Use of GIS techniques such as image rectification and transparency allow geological interpretations, soil analysis, remotely-sensed imagery, historical photographs, historical interpretations, multi-spectral pixel classifications and topography to be displayed for analysis simultaneously in any combination. Keys to this are reliable coordinate transformations, thoughtful selection of ground referencing points, the design of the GIS geodatabase and available GIS metadata.
No longer are maps constrained by physical size limitations of an arbitrary paper sheet size. No longer must small but important surface features be either grossly enlarged or completely ignored. The digital GIS world allows all features, no matter how small, to be accurately positioned to their correct size. When required, GIS allows artificial enlargement of these features at the plotting stage - otherwise, they remain in the geodatabase in their correct position and relative size.
Significantly, it is now routinely possible to integrate diverse multi-disciplinary geodata sets in various scales from remotely-sensed satellite imagery to surface geology sample in real-time, even in different coordinate systems. This leads to better and more accurate research and multiple inputs into problem solving.
With over forty years of remotely sensed data cover over Qatar, changes to Qatar’s surface can be readily identified and rates of change determined. Through the available digital elevation data and topography, past and current geological processes can be better understood and modeled, such as in relation to recent Holocene sea-level changes.
GIS allows georeferencing and rectification of digitized distorted historical geological paper maps to stable surveyed surface relief lineaments on the ground. Once combined with available high-resolution imagery at large scales, then mapping of surface features can be undertaken with confirmation in the field.
As a result, Qatar’s surface geology and soils are being readily updated through interpretation of high-resolution remotely sensed imagery. In addition, principal component analysis and classification techniques on multi-spectral SWIR and TIR bands can be used to delineate surface features and mapping units. The major benefit of the higher resolution imagery and classification techniques is that accurate mapping can be performed in conjunction with field ground work by experienced geoscientists. Through GIS, the surface geology and soils of Qatar are being re-drawn in much greater detail, with far greater accuracy and with better differentiation of an increased number of identified mapping units.
Early Mapping
In 1947, the first comprehensive aerial photographic survey of the whole of Qatar was undertaken. These images are still preserved and have recently been digitally scanned to allow image integration, manipulation and analysis through GIS.
Mapping of the surface geology of Qatar was undertaken prior to 1969 using remote sensing techniques. However, in 1970 the first comprehensive surface geology mapping of Qatar was undertaken by Cavelier and Salatt. This was partially through remotely sensed data ,but mainly through extensive on-the-ground field work, which corrected earlier assumptions made using remote sensing techniques. Cavelier and Salatt specifically identified the problem with geological interpretation when based largely on photo interpretation.
Later work on mapping the surface geology in 1983 used limited field work and was largely based on remote sensing. However, although topography was added to the surface mapping, it is readily apparent that by using analysis of remotely sensed data, earlier corrections made in 1970 appear to have been reversed and significant detail in the 1970 mapping was lost. The recent re-discovery of the 1970 mapping has revitalized the quest to re-appraise Qatar’s surface accurately.
In 1983, the first Atlas of Qatar from remotely sensed Landsat images was published. The surface geology of Qatar was interpreted using image processing techniques on Landsat images taken in 1973. Using the Landsat infrared (band 7), the predominant Middle Eocene Dammam limestone surface features could be reasonably well differentiated from the lower Dammam Rus formation, Miocene Dam formations, aolien sands, marine calcareous sands and Sabkhas. In addition, two striking lineation trends were clearly shown; the dominant NW-SE direction due to the prevailing NW wind direction and lineaments roughly perpendicular to the wind direction as part of a regional trend. However, detailed geological mapping was not possible due to the limitations of early Landsat imagery.
Current Approaches
GIS developments at Qatar Petroleum illustrates the power of a GIS geodatabase in reconciling between several types of historical data and the added value of integration. Key to this integration has been the design and planning of the geodatabase data server to permit access to GIS data in any format and datum, and to use data from multiple database management systems and file-based data sets concurrently, with confidence, and in real-time. The geodatabase provides high-performance, manages extremely large volumes of data, provides spatial integrity, and has a common interface to all leading database management systems. It supports industry data models and multiple data types such as raster (images) and vectors.
The use of metatdata in the geotechnical geodatabase ensures that all available defining information about the data within the geodatabase is known. Details about who collected the data, how the data was collected, for what purpose, to what required degree of accuracy and precision, and in which georeference datum and projection. Metadata encourages recording of the process the geoscientist or geotechnical engineer took to generate the data. It helps document which techniques were adopted to quality assure the results; for example, what work was undertaken on the ground to confirm results obtained remotely. This metadata remains with the spatial data like a document of authentication. Future researchers and users of the data can satisfy themselves that the data is in the same units, coordinate system to, to their required degree of accuracy, and to confirm that all possible steps were taken to authenticate analysis of remotely-sensed data.
The significance of this is that with the established GIS geoadatabase, it is routinely possible to integrate data in various scales from satellite to surface geology in real-time for geoscientists to undertake better and more accurate research, and to provide multiple inputs into problem solving.
GIS data sets have been invaluable in the comprehensive re-evaluation of the surface geology of Qatar. This drove a need for updated integrated maps. Key to this has been the high-resolution image data integrated with highly accurate digitized lineament and relief features.
Other GIS techniques adopted include:
- Overlay of lineament features to assist highlight mapping unit boundaries;
- Use of lineament features to assist in rectifying and aligning old imagery and maps e.g. 1970 Cavelier-Salatt geology map, which has supported mapping and interpretation on the ground.
- Use of transparency to permit rectification and overlay of historical paper maps and historical aerial photography with current satellite imagery. This assists mapping and determining and quantifying active coastal processes. These, in turn, can be used as an analog for ancient reservoirs.
- Up-to-date high-resolution imagery re-projected from Qatar National Grid for geologists experienced in WGS84 to better pick surface features and assist mapping in the field.
- Creation of a 3D triangulated surface to assist in height determination of points and in the analysis of Holocene coastal features.
- Datum and projection conversion on the fly to allow instant integration of diverse data sets with different datum and projections. This allows mapping to be revealed in a datum and projection of choice.
- Integration of field interpretations drawn on a paper map to update surface geology using re-creation of geology polygons through GIS.
Surface geology and soils are being re-evaluated using high-resoltion remotely-sensed aerial photography and multi-spectral satellite imagery which could not be achieved with early Landsat data.
GIS provides an invaluable tool to integrate diverse geodata sets and better evaluate geological processes. Geographic Information Systems (GIS) provide the integration and window into the data.