We have used remote sensing as an additional method in 1:200 000-scale reconnaissance mapping of the Ivory Coast. Landsat imagery was chosen for its low cost, and its interest for relatively small-scale work and its synthetic and multispectral properties. This proved perfectly satisfactory, especially in the bush savanna to the north of latitude 7 deg 30'. The imagery was also compared with aeromagnetic survey results. The lithostructural features revealed by MSS can be directly correlated with field observations. 1) Certain clear facies variations (amphibolites or gabbros among acidic rocks, for example) are spectrally well expressed. Conglomerates are commonly distinctive (on the Katiola sheet for example), when they are sufficiently extensive and they form ridges that can be followed several tens of kilometres. 2) The traces of planar structures can, at least locally, be followed and correlated with a regional schistosity. Certain features mappable on images confirm offset across transcurrent structures identified on the ground (N-S transcurrent fault zones, for example). Our experience in Ivory Coast shows that the use of Landsat MSS imagery should systematically be considered for any small- scale studies in which only a small part of the budget can be attributed to remote sensing.

The article shows some basic approaches to discriminate ignimbrite type pyroclastic flow deposits in Landsat Thematic Mapper imagery of semi-arid areas. Beside certain topographical and tectonical fea- tures which already describe ignimbrites and associated pyroclastic flows in those environments, our interest is focused on the influence of specific mineralogical and depositional characteristics on the spectral reflectance response. Spectral signatures in the visible and near infrared region of various fresh and weathered samples as well as desert varnish and soil samples were recorded using a GER SIRIS laboratory spectroscope to determine the factors controlling their proper response in all Thematic Mapper bands. Whole rock geochemistry data, X-ray powder diffraction analysis and microscopical studies as well as surface interpretations of the samples documented certain properties which might cause different spectral signatures also of geochemically mostly equivalent rocks. According to the semi-arid conditions of the South Patagonian Massif which are the most important constituents with respect to rock weathering and mineral alteration a more effective discrimination of the mostly leucocratic rocks was obtained using TM bands 7-4-1 and 7-5-2 as RGB false colour com- posites. Three image processing techniques, contrast stretched ratio composites, decorrelation stretched false colour composites and relative channel colour composites were examined to distinguish the variable ignimbrite outcrops in the chosen test site.

More than seventy reflectance spectra measured on mainly carbonate and evaporitic rocks outcropping in an area of northern Somalia have been analysed and classified through the recognition of specifically defined spectrofacies. The spectrofacies are associations of absorption features produced by specific electronic and molecular processes or a combination of them. The spectrofacies provide synthetic and exhaustive descriptions of the physical-chemical properties of the rocks. In the study area, the stratigraphic distribution of the spectrofacies resulted to be significantly consistent with the lithostratigraphic units known in the literature. The spectral curves of the weathered surfaces of the same samples have shown, in some cases, a specific relationship with the spectra of the fresh rocks, thus adding complementary information on the nature of the rocks (e.g.., specific rock texture appar- ently associated with lichen cover). The spectra of the weathered surfaces have also been used for the simulation of the Thematic Mapper responses by convolution with the sensitivity curves. The comparison between simulated and original TM values provided the key for the lithologic interpretation of TM multiband images.

The Big Chino Valley is a 50 km-long tectonic depression of the Basin and Range province of the South- western United States. It is bordered on the NE side by an important normal fault, the Big Chino Fault. The activity of the latter has been hypothesised on the basis of the presence of a 20 m-high fault scarp and on local geomorphological studies. Moreover, a magnitude 4.9 earthquake occurred in southern Arizona in 1976 has been attributed to this fault. The climate in the Big Chino Valley is semi-arid with average rainfall of about 400 mm per year; a very sparse vegetation cover is present, yielding a good possibility for the geo-lithologic application of remote sensing data. The analysis of the TM spectral bands shows, in the short wave infrared, a clear variation in the reflected radiance across the fault scarp. Also the available radar (SLAR) images show a marked difference in response between the two sides of the fault. An explanation of this phenomena has been found in the interaction between the geomorphic evolution, the pedological composition, and the periodic occurrence of coseismic deformation along the fault. Other effects of the latter process have been investigated on colour D- stretched images whose interpretation allowed to detect two paleoseismic events of the Big Chino Fault. This work demonstrates that important information on the seismological parameters of active faults in arid and semiarid climates can be extracted from the analysis of satellite spectral data in the visible and near -infrared.

We studied the kinematics and the tectonics setting of the Guatemalan part of the Jocotan fault system (JFS), the southernmost of the three left-lateral fault systems considered to form the North America- Caribbean plate boundary. Remote sensing studies were carried out on standard false colour composite 1:250,000 and band 3 (near infrared) 1:250,000 Landsat MSS scenes and include a supervised morpho-structural classification and a statistical analysis of both volcano-tectonic and tectonic lineaments. The interpretation of these data revealed a clear left-lateral deformation pattern in the form of Riedel shear fractures with a dominant orientation 60 degree(s)N-75 degree(s)E at any length class of lineaments. Two NNE trending extensional structures, the Jocotan and the Chiquimula half-grabens, are located along the central part of the JFS and are interpreted as releasing bend structures associated with a change in the fault trend from ENE to E-W. Although the JFS appears a very marked feature, no definite evidences of recent active tectonic motions were detected. The field study confirmed that the JFS is a first order left-lateral strike-slip structure, which accommodated also a certain amount of southward thrusts. Field inspections along land fissures opened during the 1976 Guatemala earthquake indicated that they were likely due to gravitational failure triggered by the ground motion.

The collision between India and Eurasia at the southern margin of Tibet appears to be responsible for the active tectonics throughout much of eastern Asia. The study area, which corresponds to the Phung-Chu/Arun region in the Tibetan Himalayas, is geologically related to the brittle deformations that affected the High Himalayan Crystalline (HHC) unit and the Tethyan Himalayan Zone (THZ) in the region between the great thrusts in Nepal and the Yarlung-Tsangbo Suture Zone (YTSZ) in Tibet, characterised by right-lateral shear tectonics. The tectonic setting is mainly represented by two lineament trends: the pattern E-W (longitudinal) and N-S (transversal); the former being driven by the dextral shear displacement style of the YTSZ, according to the wrench tectonic concepts, and the latter by the tensile stress field of Quaternary age, as derived from en-echelon style. In these large and very hardly accessible areas where geographic information are missing or poor, the space imagery can be useful as a cartographic support and offer promising results in the field of geostructural analysis. Image analysis was achieved by means of a conventional Landsat-TM photoproduct (RGB=432), frame 140- 040/Q4 and several Space Shuttle-LFC stereo images, enlarged at a scale of 1:100,000.

The experimental data of the spaceborne Modular Optoelectronic Multispectral Scanner MOMS-02 acquired on the German Spacelab D2 mission will be used in the near future for a broad range of geo- scientific objectives. The simultaneous acquisition of high spatial resolution multispectral and stereo data by one system enables a real combination of thematic and topographic information, thus consid- erably improving the verification and interpretation of dynamic changes of the Earth's surface by means of remote sensing. The sensor was successfully launched on board of the Second German Spacelab Mission D2 on Space Shuttle flight STS-55 from 26 April to May 6 1993. The MOMS-02 system itself is a combination of two modules, a panchromatic, high-resolution stereo module with one nadir-looking (resolution 4.2 m x 4.2 m) and two tilted channels and a 4-channel multispectral module covering the visible (VIS) and near infrared (NIR) range of the electromagnetic spectrum at 12.8 m x 12.8 m ground resolution. The width and centre wavelength of the multispectral spectral bands are optimised for the detection of the spectral response of vegetation and for the discrimination of Fe-bearing rock and soil surfaces due to significant absorption in the VIS/NIR spectral range. The paper gives an overview on geoscientific applications in Australia, Egypt and South America. First images of MOMS-02 data clearly show the ability of the multispectral, as well as the combination of multispectral and high resolution data, to improve geoscientific studies that were done with Landsat TM data in the past. In the near future the MOMS-02 science team will produce high accuracy digital elevation models derived from the stereo data which will be used in combination with MOMS-02 or other sensors multispectral datasets for a wide range of geoscientific and environmental applications (e.g. landuse, erosion risk, natural hazards), as well as for radiometric correction of satellite images. After the first experimental mission on Spacelab D2, MOMS-02 is planned to be deployed in early 1995 on the Russian space platform MIR incorporated in the environmental PRIRODA module. This mission, offering the opportunity of multisensor data registration with a repetition rate of 2 to 7 days, is considered an indispensable step to achieve high spatial, spectral and temporal resolution environmental monitoring at global scales.

The sulfur dioxide molecule shows absorption features in UV, thermal IR and microwave regions. These three portions of the electromagnetic spectrum can be investigated with remote sensing systems. The first non-imaging remote sensing instrument used to make quantitative measures of SO2 in volcanic plumes was the COSPEC. The resultant measurements have allowed volcanologist to recognise that a rapid increase in the sulfur dioxide flux from a volcano can signal an impeding eruption. This technique although producing good results can require a considerable logistical planning and in some cases personal risk. Spaceborne imaging sensors have been used to detect stratospheric sulfur dioxide plumes during volcanic eruptions. Nevertheless the atmospheric effects and coarse spatial resolution (between 5 and 50 Km) of the acquired images prevent the detection of tropospheric volcanic plumes formed by the continuous degassing from summit craters. In this study an analysis of image data acquired from NASA's Airborne Thermal Infrared Multispectral Scanner (TIMS) to estimate sulfur dioxide content of a tropospheric volcanic plume is presented. The estimation procedure is based on LOWTRAN 7 radiative transfer code, which is used to model the radiance perceived by TIMS when it looks at ground through a volcanic plume. The described technique was developed on the TIMS data acquired over Mt. Etna, Sicily on July 29, 1986.

SAR interferometry is the ideal tool for monitoring deformation fields, provided that they show some kind of spatial continuity and that the surface involved show a sufficient degree of geometric and electromagnetic stability. The technique has been successfully used to measure landslides, with radar image acquisitions separated by up to one and a half years. We how some results obtained by the French Space Agency in these domains. Our procedure makes use of the digital terrain model (DEM) of the area combined with two radar acquisitions. We believe this technique is the most appropriate for small moves detection since there always exists some kind of DEM in a given area. At worse, one could build the DEM using other data. Radar interferometry itself may provide the DEM if a 'motion free' time interval is available. Our studies indicate that the 'phase unwrapping,' often considered as a requirement of the interferometric technique, can be bypassed most of the time. This contributes to making this technique very user-friendly and allows us to hope it will spread widely very soon.

Tigre Province is in northern Ethiopia between Eritrea on the north and east, and Begemdir and Wollo on the south. The area is comprised of five main physiographic units: Danakil Depression, Escarpment, Tigre and Eritrea Plateau, Semien Mountains and the Western Lowlands. The first two drain into the closed drainage basin of Dallol whilst the other three drain into the Nilo Through the Tezeke, Mareb and Barka River. Three main rock complexes build up the prospect area. The metamorphic and intrusive rocks of the Precambrian basement, the Paleozoic-Mesozoic Sediments and Tertiary Trap Volcanics, and the Young Sedimentary and Volcanic Rocks of the Danakil Depression. The only existing geological data on this area is represented by maps assembled from different sources on very large scale. Thanks to interactive computer processing, it is becoming possible to trace, on the relief as it appears on the display unit, lithological boundaries and the faults which affect them, measuring each of their geometric characteristics. Image processing coupled with pattern recognition programs has made it possible to take into consideration landscape units combining morphological, botanical and geological signatures, and more generally to quantify the morphological elements. Satellite images (Landsat TM and Spot XS) have been processed to produce thematic maps.

Directional and gradient filters have been applied in Landsat TM image to help rock discrimination in a study area of SW Angola, where Precambrian basement rocks are well exposed. A well defined limit has been drawn on filtered images, primarily on the basis of lineation pattern, between the two main rock groups outcropping in the area, namely the Schistose Mainly Metasedimentary Complex and the Gneissic-migmatitic-granitoid Complex. Various lithological units have been mapped within the two complexes on the basis of other features still related to texture, such as lineation density which, at this level, can be associated to bedding planes, foliation and/or schistosity. Discrimination of the recent continental cover took benefit from this technique as the lineation-free lithologies stand out as smooth patches in all the products. By displaying at least 5 major textural patterns in the whole area, directional and gradient filters highlight rock properties and structure which must be related not only to lithology but also to geological history. Although less powerful than a multispectral approach, the present one can support traditional analysis and integrate a spectrally-based study when, as in the present case, texture enters as Intensity in an IHS transformation.

Preliminary lithological investigation of a poorly known area can draw substantial help from remotely- sensed data, particularly in arid environments. A reconnaissance mapping of the Nama Group sedi- mentary succession in north-western South Africa (Richtersveld area) was performed with two strategies: visual interpretation and automatic classification. Both strategies share the same purpose: the identification and labelling/coding of the various surface types that are present in the image under study. Although often used alternatively, they can be successfully integrated. The first of such strategies is highly dependent on the interpreter's skill and intuition and therefore rather subjective; automatic classification, on the other hand, is rigidly parametric, although influenced by the creation of the spectral signatures for the training sets. The present paper offers some preliminary evaluations of the results deriving from the two approaches and contributes to the knowledge of an area that has not yet been mapped in detail.

The Aniene River rises in the Simbruini Mountains, runs for about 100 km and empties into the Tiber River within Rome's urban area. In its last stretch, it flows from Tivoli to the Tiber river through the travertine deposits of the Tivoli and Bagni Albule area and the volcanic and alluvial sediments of the roman countryside. The quality of the water found in the rivers drainage basin, suffers enormously from human activities, especially in the lower course. Our main goal was to demonstrate that remote sensing may be used to detect and compare natural conditions and situations resulting from human activities as regards the hydrogeological hazard of the Aniene River, thus directing future interventions aimed at mitigating human impact and restoring previous environmental balance. This work describes a procedure that, starting from cartography data acquisition and satellite image processing, permitted us to produce an hydrogeological vulnerability map of the lower course of the Aniene River using an option of the Xpace module multilayer modelling (software Easy Pace by PCI). In the future we propose to extend this procedure to the entire Aniene basin and to relate the hydrogeological vulnerability map with human activities that impact on the area.

Present work aims at defining, by means of lineament analysis from satellite imagery, the geo- struc- tural scenario of the upper Inland Niger Delta (Mali Republic), trying to match main results of previous investigations. The study region has been for a long time examined for geomorphologic and hydrological studies. These researches have highlighted major steps in the quaternary evolution of the Middle Niger valley. In 1929, Furon provided a preliminary description of the ancient hydrographic pattern of Niger Delta. In 1965, Tricart outlined on the basis of geological field suney that, only hypothesising post- Cretaceous tectonic events, it could be possible to give insight into some meaningful regional drainage changes. In 1982, Simon et al. compared geophysical and classical geological documents with digital satellite data, pointing out the existence of linear features related to deep structures, never revealed by classical photointerpretation works. In 1983, McIntosh presented a reconstruction of the geomorphological evolution in the last 20,000 years, mostly related to palaeoclimatic oscillations. In 1987, Guerre and Sow have performed an hydrogeological research in the area, by interpreting geological, hydrological and geophysical data collected during field surveys. In 1988, Jacobberger, by analysing satellite data, confirmed, within the Azauoad dunes field north of Tombouctou, the oc- currence of abandoned river channels, already pointed out by Palausi(1955). Present research has concerned a 70,000 sq. km wide area, Iying between the city of Mopti (to the south), the Faguibine lake (to the north), the city of Tomboutctou (to the west) and the town of Lere' (to the east), including most of the Upper Inland Niger river Delta. The analysis has been carried out by processing the MSS bands of the two Landsat subscenes covering the study area in different dates, corresponding to dry and wet seasons. The Landsat MSS bands have been geometrically corrected and georeferenced with respect to a UTM grid, locating Ground Control Points in the available topographic maps. These images have been then processed by convolving them with suitable spatial low-pass and directional filters, enhancing structures with peculiar wavelengths and trends. Particular attention has been driven on removing the imprints of surface features such as sand dunes which may be considered noise in this study. The obtained lineament spatial distribution has been checked with already mentioned literature information and, where possible, geophysical and topographic data, with the aim of providing an improved structural interpretation of recent hydrographic network changes.

The importance and the limits of the study of thermal inertia using Landsat TM data for lithological discrimination are here described. A new algorithm was developed to compute thermal inertia using monotemporal data, by combining values of the absorbed radiance (co- albedo) with those of the emitted radiance (thermal band), considering also bandwidths and calibration coefficients of the TM sensor. this algorithm was tested on a 1024 X 1024 pixel Landsat TM image concerning an area located in southwestern Angola where different rock types outcrop.