You are here

Warning message

Attention! This event has already passed.

Integrated analysis of sub-tropical mountain karst geohydrology in NW Vietnam by field and multisource remotely sensed data

Friday, 20 April, 2007 - 14:00
Campus: Brussels Humanities, Sciences & Engineering campus
Quoc Hung Le
phd defence

Remotely sensed data has become an important source for geoscientists to understand the earth. With the rapidly growing availability of remotely sensed data and field surveys, there is an urgent demand to interpret and analyze these data. Remote sensing is a promising technology to support geohydrological studies, mainly because of the potential to observe areas and entire river basins rather than single points. With the successful development of computer science, geographic information systems (GIS) and image processing systems, it has become feasible to digitally interpret multiple types of remotely sensed and thematic map data. In spite of this promise, the application of remote sensing in geohydrological studies has been slow, especially in the developing countries.

This work aims to develop and apply integrated methods combining the information obtained by geohydrological field mapping and those obtained by analyzing multi-source remotely sensed data in a GIS environment for better understanding the geohydrological characteristics of the sub-tropical vegetated karst region in Vietnam. The study focuses on two important mountainous areas: Suoimuoi catchment and Tamduong valley. Remote sensing analyses are applied, which include lineament analysis and its application for fractured zone studies, fusion method and geological mapping.

An effective approach for the delineation of fracture zones is based on lineament indices extracted from satellite imagery. The complete procedure of lineament analysis consists of four steps: (1) extraction, (2) correction, (3) indices and lineament indices densities, (4) fractured zone definition. The accuracy of extracted lineaments depends strongly on the spatial resolution of the imagery. The most suitable band for lineament extraction in the Suoimuoi catchment is the NIR band. It is tested and shown here that extracted lineaments from the NIR ASTER imagery are considerably less noisy and show a higher accuracy than lineaments extracted from other imagery. Three lineament indices are defined as: (1) lineament frequency, (2) length, and (3) intersection).

The relationship between fractured rock and groundwater is investigated. However, a high level of fractured rock does not guarantee the presence of groundwater in the Suoimuoi area. Therefore, additional information from karst depression analysis is needed and it is concluded that there is a high potential for groundwater in the highly fractured zones, which are close to karstic depressions or rivers. The general direction of groundwater flow can be predicted by the direction of the faults system and depressions in the Suoimuoi catchment where regional tectonic structures control karst groundwater flow. The ASTER image is used, after enhancement, to digitally extract lineaments. Based on indices of lineament densities, a fractured zone is defined in which cave development is expected. Areas with karstic depressions are extracted by combining the result of DEM analysis and ASTER image ration 3/4. The methodology set up for cave analysis in the sub-tropical karst of the Suoimuoi catchment, proves to be a promising method for predicting groundwater flow.

The integration of Landsat ETM+ images with high resolution PAN and aerial photos provide complementary information with respect to the discrimination of major geological features and allows lineament extraction in detail. With the high-resolution images, manmade and natural features are easy to recognize, especially in combination with aerial photos. The Intensity-Hue-Saturation (IHS), Principal Component Analysis (PCA) and a Wavelet-based method are used for image fusion. Quantitative metrics for evaluating the quality of fused images are carried out. It is found that IHS is the most optimal method for image fusion in the Tamduong area, because it has a high potential for visual interpretation. A combination of remotely sensed data and fieldwork is used for geological mapping. The research methodology is applied to explore the sub-tropical vegetated karst area of Tamduong. Based on the remote sensing analysis and fieldwork, a geological map is established. Fractured rock zones are mapped with the help of lineament indices densities. This research clearly shows that the study area is characterized by strong tectonic activity and faulted zones, whereby the gea-structure is controlled by NW-SE folding and fracture systems.

This work not only gives details and insight in the understanding of karst geohydrological characteristics of Suoimuoi and Tamduong, but also provides methods of integrating remotely sensed data for geohydrological studies. The methods applied in this thesis prove to be very useful for geohydrological studies in vegetated mountainous sub-tropical karst areas of developing countries, like Vietnam.