ACRES Update, Issue 20, 20th Anniversary Edition, December 1999 ACRES releases SPOT 4 products Celebrating 20 years Satellie imagery used in projects to reduce global warming Alice Springs Landsat 7 poster

ACRES Update, Issue 18, June 1999 Online ordering of satellite imagery ACRES to offer Landsat 7 imagery SAR data detects oil slicks New $2 million optiacal data processor

The product SAR.SLC is a single look complex digital image generated from raw SAR data using up-to-date auxiliary parameters. The image, projected on slant range, referred to as 'quarter scene' or quadrant corresponds to approximately 75 km wide and at least 75 km long. The JERS SAR.SLC format is based on the general definition of the SAR CEOS format (ref. ER-IS-EPS-GS-5902).

ACRES Update, Issue 21, July 2000 Terra Oberving the Earth New Product Catalogue Remote Sensing and emergency management

Legacy product - no abstract available

Archived imagery

This study assessed the geochemical indicators for carbonatite-associated and alkaline igneous REE mineral systems based on state-scale ASTER data for Western Australia and the National Geochemical Survey of Australia. In the latter we applied discrete field models obtained by attributing catchment outlet sediment geochemical data to their catchment basins. The cerium data was found to be a suitable proxy for the main features of REE distribution across the continent. In general, this provided a well-expressed positive correlation between identified catchments with anomalous (elevated) Ce and the mineral occurrences and deposits that they contained but not all REE deposits were identified by these methods. The study also included more detailed mineral mapping using the recently released ASTER geoscience map of Western Australia (WA). A study of the relatively well exposed Yangibana 'ironstone' dykes in the Gascoyne Province of Western Australia showed relatively good correlations with known REE occurrences for the AlOH Group Composition, MgOH Group Composition and Ferrous Iron Content in MgOH/carbonate ASTER products. However, no well defined correlations were observed for Mt Weld, Ponton and Cummins Range all of which occur under cover. The ASTER SWIR products have difficulty detecting the types of minerals present in alteration zones surrounding these carbonatites. Some of these minerals may be detected with the ASTER thermal infrared geoscience products which were not available at the time of this study. These products will include a silica index (e.g. for quartz, feldspars, Al-clays), a carbonate index (e.g. for calcite, dolomite, magnesite, siderite, ankerite) and a mafic group index (e.g. for pyroxenes, garnets, olivine, epidote, chlorite, calcite, magnesite, dolomite). However, the lower resolution of the thermal infrared bands (~90 m) may still make it difficult to detect carbonatite-associated mineralisation. Finally, the ASTER data for the Brockman (Hastings) alkaline igneous REE deposit was also examined. The mineralised unit of this deposit only reaches a maximum width of 35 m and so, once again, proved difficult to detect with ASTER. However, the ASTER maps readily highlighted the contact between the Olympio Formation to the east with the Biscay Formation to the west and demonstrated the mapping capabilities of the ASTER products where the lithologies are relatively well exposed. In order to overcome the problem of only sampling the surface, we combined the ASTER data with the regional magnetic data because the magnetic signal is derived from a greater depth. As many carbonatite complexes are surrounded by mafic alkaline rocks, they often show up as a magnetic bull's eye combined with a gravity low and ringed by a gravity high.

Normalising for atmospheric and land surface bidirectional reflectance distribution function (BRDF) effects is important in satellite data processing. It is particularly important for standardising time series data and for inter-sensor calibration and comparison. Procedures based on physical models have been applied successfully with the Moderate Resolution Imaging Spectroradiometer (MODIS) data products at global scales. For Landsat and other higher resolution data, similar options exist except that the estimation of BRDF using internal fitting, as used for MODIS, is not available due to the smaller variation of view and solar angles, sun-synchronous view and infrequent revisits. Despite this, the use of physical models for atmospheric correction and BRDF normalising can still be appropriate. In this study, we explore the potential for developing operational procedures to correct higher resolution sensor data based on combined atmospheric and BRDF models. The process was realised using BRDF parameters (shape functions) derived from MODIS and using the MODTRAN 4 radiative transfer model. The approach was tested using Landsat data for two sites with different land covers in Australia. The retrieved Landsat reflectance values have good agreement with ground based spectroradiometer measurements with the root mean square difference (RMSD) for both sites being less than 3%. The comparison between normalised Landsat and MODIS reflectance shows a strong relationship, indicating that cross-calibration between the two sensors is achievable. Strategies which may provide effective BRDF parameters before MODIS was available (and after its mission is complete) are also discussed as well as the options that exist for an operational system in the context of monitoring land cover change.

Legacy product - no abstract available

ACRES Tehnical Document - updated 4 September 2000. Dynamic range values for ACRES TM data products.