Fuqin Li

An Evaluation of the Use of Atmospheric and BRDF Correction to Standardize Landsat Data

Normalizing for atmospheric and land surface bidirectional reflectance distribution function (BRDF) effects is essential in satellite data processing. It is important both for a single scene when the combination of land covers, sun, and view angles create anisotropy and for multiple scenes in which the sun angle changes. As a consequence, it is important for inter-sensor calibration and comparison. Procedures based on physics-based models have been applied successfully with the Moderate Resolution Imaging Spectroradiometer (MODIS) data. For Landsat and other higher resolution data, similar options exist. However, the estimation of BRDF models using internal fitting is not available due to the smaller variation of view and solar angles and infrequent revisits. In this paper, we explore the potential for developing operational procedures to correct Landsat data using coupled physics-based atmospheric and BRDF models. The process was realized using BRDF shape functions derived from MODIS with the MODTRAN 4 radiative transfer model. The atmospheric and BRDF correction algorithm was tested for reflectance factor estimation using Landsat data for two sites with different land covers in Australia. The Landsat reflectance values had a good agreement with ground based spectroradiometer measurements. In addition, overlapping images from adjacent paths in Queensland, Australia, were also used to validate the BRDF correction. The results clearly show that the algorithm can remove most of the BRDF effect without empirical adjustment. The comparison between normalized Landsat and MODIS reflectance factor also shows a good relationship, indicating that cross calibration between the two sensors is achievable.

Issues in the application of Digital Surface Model data to correct the terrain illumination effects in Landsat images

The accuracy of topographic correction of Landsat data based on a Digital Surface Model (DSM) depends on the quality, scale and spatial resolution of the DSM data used and the co-registration between the DSM and the satellite image. A physics-based bidirectional reflectance distribution function (BRDF) and atmospheric correction model in conjunction with a 1-second DSM was used to conduct the analysis in this paper. The results show that for the examples used from Australia, the 1-second DSM, can provide an effective product for this task. However, it was found that some remaining artefacts in the DSM data, originally due to radar shadow, can still cause significant local errors in the correction. Where they occur, false shadows and over-corrected surface reflectance factors can be observed. More generally, accurate co-registration between satellite images and DSM data was found to be critical for effective correction. Mis-registration by one or two pixels could lead to large errors of retrieved surface reflectance factors in gully and ridge areas. Using low-resolution DSM data in conjunction with high-resolution satellite images will also fail to correct significant terrain components where they occur at the finer scales of the satellite images. DSM resolution appropriate to the resolution of satellite image and the roughness of the terrain is needed for effective results, and the rougher the terrain, the more critical will be the accurate registration.

Applying machine learning methods and time series analysis to create a National Dynamic Land Cover Dataset for Australia

The National Dynamic Land Cover Dataset (DLCD) classifies Australian land cover into 34 categories, which conform to 2007 International Standards Organisation (ISO) Land Cover Standard (19144-2). The DLCD has been developed by Geoscience Australia and the Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES), aiming to provide nationally consistent land cover information to federal and state governments and general public. This paper describes the modeling procedure to generate the DLCD, including machine learning methodologies and time series analysis techniques involved in the process.

The variability of satellite derived surface BRDF shape over Australia from 2001 to 2011

The intra- and inter-annual variability of surface bidirectional reflectance distribution function (BRDF) in Australia has been analyzed using 11 years (2001-2011) of MODIS BRDF data. A statistic called here Root Mean Square (RMS) was used as a BRDF shape indicator to represent the overall BRDF shape and an Australian vegetation structure map was used to separate the different BRDF shape patterns by structure. The results show that the intra-annual variation of BRDF shape is stronger than the inter-annual variation although it is not clear yet whether the variation is related more to climate patterns or to vegetation structure (height and cover) or landcover class. However, BRDF shape patterns have strong similarity with vegetation structure classes. There is strong correlation between RMS and the Normalized Difference Vegetation Index (NDVI) at annual scale within structural classes indicating good relationship between BRDF and annual changes in cover within the classes.

Creating multi-sensor time series using data from Landsat-5 TM and Landsat-7 ETM+ to characterise vegetation dynamics

The Landsat series of satellites provide the longest contiguous earth observation record of the Earths surface. This provides the unique capacity to track changes in vegetation over multiple decades. This paper illustrates how standardized Landsat data can be combined to create a time series of sensor independent observations. The impact of side-lap and cloud frequency on observation frequency are also examined with reference to two adjacent path/rows of data in southern Australia. The generation of Landsat scale time series provides the opportunity to track both subtle and dramatic changes in vegetation cover in much higher levels of detail than previously possible. However the approach presents new challenges associated with developing time series analysis techniques to characterize time series that have uncertain observation frequencies.

Evaluation of the TanDEM-X intermediate DEM for terrain illumination correction in Landsat data

An appropriate resolution of the Digital Elevation Model (DEM) data with sufficient quality of the gradient field is critical for effective correction of remotely sensed data over mountainous areas. Conversely, using performance of terrain illumination correction and scale-based analysis, such as filter bank analysis, the quality of DEM data can be evaluated. In this study, TanDEM-X Intermediate DEM (IDEM) data at 12 m resolution and the 1-arc second Shuttle Radar Topography Mission (SRTM) data were used independently to evaluate the relative effectiveness of the terrain illumination correction for Landsat 8 optical data over Tasmania. Results from the terrain illumination correction and filter bank analysis show that IDEM 12 m data can resolve finer details of terrain shading than the SRTM based DEM and deliver better results in areas with detail-rich terrain. However, since the data available for this study is an intermediate product, spikes and other noise artefacts were prevalent, especially over areas covered by water. Operational use of the IDEM would require the removal of such noise artefacts.