466 publications from this institution
No abstract is provided for this article.
No abstract is provided for this article.
Four different approaches to the classification of land cover for whole continents using multitemporal images of the normalized difference vegetation index derived from the Advanced Very High Resolution Radiometer of the NOAA series of satellites are discussed. The first approach uses only two dates from different seasons and classification dependent upon subdivision of the resultant two-dimensional feature space by an analyst using a track ball. The second approach involves a similar method of partitioning the feature space, but with the two dimensions being the first and second principal components derived from 13 four-week composite images. The third approach uses the maximum likelihood rule to derive the classified map. In the fourth approach, the amount of deviation from characteristic curves is used as a basis for classification.
The invertibility of an accurate discrete ordinates canopy reflectance model is investigated through a series of experiments. Effects of different canopy types, noise levels, spectral ranges, and sampling geometries [including the NOAA Advanced Very High Resolution Radiometer (AVHRR) and the proposed Multi-angle Imaging SpectroRadiometer (MISR) satellite sampling schemes] are considered. Both error-free synthetic bidirectional reflectance data and empirical field reflectance data are utilized. Results suggest that the model can retrieve soil and canopy parameters with reasonable accuracy in most cases, and surface state parameters (absorbed radiation, spectral albedo and photosynthetic efficiency) with high accuracy in all cases. The efficiency of several commonly used optimization algorithms is also assessed, and a model sensitivity study is conducted.
Techniques developed for the prediction of winter wheat yields from remotely sensed data indicating crop status over the growing season are tested for their applicability to corn. Ground-based spectral measurements in the Landsat Thematic Mapper bands 3 (0.62-0.69 microns), 4 (0.76-0.90 microns) and 5 (1.55-1.75 microns) were performed at one-week intervals throughout the growing season for 24 plots of corn, and analyzed to derive spectral ratios and normalized spectral differences of the IR and shortwave IR bands with the red. The ratios of the near IR and shortwave IR bands are found to provide the highest and most consistent correlations with corn yield and dry matter accumulation, however the value of band 5 could not be tested due to the absence of water stress conditions. Integration of spectral ratios over several dates improved the correlations over those of any single date by achieving a seasonal, rather than instantaneous, estimate of crop status. Results point to the desirability of further tests under other growth conditions to determine whether satellite-derived data will be useful in providing corn yield information.
Tami Bond , environmental engineer and professor in the Department of Civil and Environmental Engineering at the University of Illinois at Urbana‐Champaign, has been selected as a 2014 MacArthur Fellow.
The NOAA-7 polar orbiting, sun-synchronous, operational satellite carries the 5 channel Advanced Very High Resolution Radiometer (AVHRR). Data are acquired globally at a resolution of 4 km on a daily basis. This data provides the means of frequently monitoring global vegetation on continental scales. Techniques for compositing and cloud screening a vegetation index for Africa are presented. The sample data set covers 9 days beginning August 16, 1983 and is compared with a semi-operational vegetation index product produced by NOAA.
Remotely sensed measurements from NOAA-AVHRR expressed as normalized difference vegetation index (NDVI) have generated a 23-year time series appropriate for long-term studies of Sahel region. The close coupling between Sahelian rainfall and the growth of vegetation has made it possible to utilize NDVI data as proxy for the land surface response to precipitation variability. Examination of this time series reveals two periods; (a) 1982–1993 marked by below average NDVI and persistence of drought with a signature large-scale drought during the 1982–1985 period; and (b) 1994–2003, marked by a trend towards ‘wetter’ conditions with region-wide above normal NDVI conditions with maxima in 1994 and 1999. These patterns agree with recent region-wide trends in Sahel rainfall. However taken in the context of long-term Sahelian climate history, these conditions are still far below the wetter conditions that prevailed in the region from 1930 to 1965. These trend patterns can therefore only be considered to be a gradual recovery from extreme drought conditions that peaked during the 1983–1985 period. Systematic studies of changes on the landscape using high spatial resolution satellite data sets such as those from LANDSAT, SPOT and MODIS will provide a detailed spatial quantification and description of the recovery patterns at local scale.
Analysis of variance methods has been applied to in situ grassland spectral reflectance data in order to determine the classes or levels of total wet biomass that can be resolved spectrally by a single narrow band measurement. Ground-truth clipping of blue grama grass plots was performed immediately following spectral reflectance measurements at 91 wavelength intervals which were 0.005 microns apart over the spectral range from 0.350 to 0.800 microns. It was found that the photographic infrared region of 0.750 to 0.800 microns could be used to distinguish three classes or levels of total wet biomass. Four or five classes, particularly at higher biomass levels, could not be distinguished by this technique.
