SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
The author has identified the following significant results. With rare exceptions, avalanche areas cannot be identified on LANDSAT imagery. Avalanche hazard mapping on a regional scale is best conducted using LANDSAT imagery in conjunction with complementary data sources. Level of detail of such maps will be limited by the amount and completeness of the complementary information used.
A Joint Research Project (JRP) invlving NASA/Goddard Space Flight Center and the Pennsylvania Bureau of Forestry/Division of Forest Pest Management demonstrates the utility of Landsat data for assessing forest insect damage. A major effort within the project has been the creation of map-registered, statewide Landsat digital data base for Pennsylvania. The data base, developed and stored on computers at the Pennsylvania State University Computation Center, contains Landsat imagery, a Landsat-derived forest resource map, and digitized data layers depicting Forest Pest Management District boundaries and county boundaries. A data management front-end system was also developed to provide an interface between the various layers of information within the data base and image analysis software. This front-end system insures than an automated assessment of defoliation damage can be conducted and summarized by geographic area or jurisdiction of interest.
The utility of lineaments identified on LANDSAT imagery as an exploration tool in the search for hydrocarbons within three Appalachian test sites were assessed. The optimum LANDSAT imagery enhancement technique for displaying lineaments was identified. The LANDSAT lineament data were analyzed and results were correlated with oil and gas field information for each of three tests sites. Good correlations were found for several states. Successful techniques can be incorporated into a broader exploration model.
For BOReal Ecosystem-Atmosphere Study (BOREAS), the level-3p Landsat Thematic Mapper (TM) data were used to supplement the level-3s Landsat TM products. Along with the other remotely sensed images, the Landsat TM images were collected in order to provide spatially extensive information over the primary study areas. This information includes radiant energy, detailed land cover, and biophysical parameter maps such as Fraction of Photosynthetically Active Radiation (FPAR) and Leaf Area Index (LAI). Although very similar to the level-3s Landsat TM products, the level-3p images were processed with ground control information, which improved the accuracy of the geographic coordinates provided. Geographically, the level-3p images cover the BOREAS Northern Study Area (NSA) and Southern Study Area (SSA). Temporally, the four images cover the period of 20-Aug-1988 to 07-Jun-1994. Except for the 07-Jun-1994 image, which contains seven bands, the other three contain only three bands.
The BOREAS TE-18 team used a radiometric rectification process to produce standardized DN values for a series of Landsat TM images of the BOREAS SSA and NSA in order to compare images that were collected under different atmospheric conditions. The images for each study area were referenced to an image that had very clear atmospheric qualities. The reference image for the SSA was collected on 02-Sep-1994, while the reference image for the NSA was collected on 21-Jun-1995. the 23 rectified images cover the period of 07-Jul-1985 to 18 Sep-1994 in the SSA and from 22-Jun-1984 to 09-Jun-1994 in the NSA. Each of the reference scenes had coincident atmospheric optical thickness measurements made by RSS-11. The radiometric rectification process is described in more detail by Hall et al. (199 1). The original Landsat TM data were received from CCRS for use in the BOREAS project. The data are stored in binary image-format files. Due to the nature of the radiometric rectification process and copyright issues, these full-resolution images may not be publicly distributed. However, a spatially degraded 60-m resolution version of the images is available on the BOREAS CD-ROM series. See Sections 15 and 16 for information about how to possibly acquire the full resolution data. Information about the full-resolution images is provided in an inventory listing on the CD-ROMs. The data files are available on a CD-ROM (see document number 20010000884), or from the Oak Ridge National Laboratory (ORNL) Distributed Activity Archive Center (DAAC).
The BOREAS TE-18 team used a radiometric rectification process to produce standardized DN values for a series of Landsat TM images of the BOREAS SSA and NSA in order to compare images that were collected under different atmospheric conditions. The images for each study area were referenced to an image that had very clear atmospheric qualities. The reference image for the SSA was collected on 02-Sep-1994, while the reference image for the NSA was collected on 2 1 Jun-1995. The 23 rectified images cover the period of 07-Jul-1985 to 18-Sep-1994 in the SSA and 22-Jun-1984 to 09-Jun-1994 in the NSA. Each of the reference scenes had coincident atmospheric optical thickness measurements made by RSS-11. The radiometric rectification process is described in more detail by Hall et al. (1991). The original Landsat TM data were received from CCRS for use in the BOREAS project. Due to the nature of the radiometric rectification process and copyright issues, the full-resolution (30-m) images may not be publicly distributed. However, this spatially degraded 60-m resolution version of the images may be openly distributed and is available on the BOREAS CD-ROM series. After the radiometric rectification processing, the original data were degraded to a 60-m pixel size from the original 30-m pixel size by averaging the data over a 2- by 2-pixel window. The data are stored in binary image-format files. The data files are available on a CD-ROM (see document number 20010000884), or from the Oak Ridge National Laboratory (ORNL) Distributed Activity Archive Center (DAAC).
For BOReal Ecosystem-Atmosphere Study (BOREAS),the level-3s Landsat Thematic Mapper (TM) data, along with the other remotely sensed images,were collected in order to provide spatially extensive information over the primary study areas. This information includes radiant energy,detailed land cover, and biophysical parameter maps such as Fraction of Photosynthetically Active Radiation (FPAR) and Leaf area Index (LAI). CCRS collected and supplied the level-3s images to BOREAS for use in the remote sensing research activities. Geographically,the bulk of the level-3s images cover the BOREAS Northern Study Area (NSA) and Southern Study Area (SSA) with a few images covering the area between the NSA and SSA. Temporally,the images cover the period of 22-Jun-1984 to 30-Jul-1996. The images are available in binary,image-format files.
The procedure developed uses two-phase sampling, stratification and multidate Landsat imagery to produce the estimate. Maximizing the advantages of both spectral data and field patterns available from Landsat, three dates of Landsat imagery are interpreted to provide county-wide estimates of the proportion of land irrigated. Ground data, collected on a subset of the area interpreted on Landsat, are used to calibrate the satellite estimate. The satellite and ground measurements are reduced to proportion data and linked using a regression estimator to produce the estimate of irrigated land. It is estimated that 3.99 million hectares were irrigated with a relative standard error of + or - 1.18% at the 99% confidence level. Acreage tabulations provided by the Department of Water Resources show that the Landsat-based estimate differed by less than 0.4% at the state level.
The applicability of LANDSAT MSS imagery for discriminating geobotanical associations observed in zones of cassiterite-rich metasomatic alterations in the granitic body of Serra da Pedra Branca was investigated. Computer compatible tapes of dry and rainy season imagery were analyzed. Image enlargement, corrections, linear contrast stretch, and ratioing of noncorrelated spectral bands were performed using the Image 100 with a grey scale of 256 levels between zero and 255. Only bands 5 and 7 were considered. Band ratioing of noncorrelated channels (5 and 7) of rainy season imagery permits distinction of areas with different vegetation coverage percentage, which corresponds to geobotanial associations in the area studied. The linear contrast stretch of channel 5, especially of the dry season image is very unsatisfactory in this area.
The author has identified the following significant results. Ten EREP data passes over the Wyoming test site provided excellent S190A and S190B coverage and some useful S192 imagery. These data were employed in an evaluation of the EREP imaging sensors in several earth resources applications. Boysen Reservoir and Hyattsville were test areas for band to band comparison of the S190 and S192 sensors and for evaluation of the image data for geologic mapping. Contrast measurements were made from the S192 image data for typical sequence of sedimentary rocks. Histograms compiled from these measurements show that near infrared S192 bands provide the greatest amount of contrast between geologic units. Comparison was also made between LANDSAT imagery and S190B and aerial photography for regional land use mapping. The S190B photography was found far superior to the color composite LANDSAT imagery and was almost as effective as the 1:120,000 scale aerial photography. A map of linear elements prepared from LANDSAT and EREP imagery of the southwestern Bighorn Mountains provided an important aid in defining the relationship between fracture and ground water movement through the Madison aquifer.
Plant residue on the surface of cultivated soils in Miami County, Indiana is analyzed in terms of quantity and type with Landsat TM data to generate information for a conservation program for agricultural soil. The Landsat data are enhanced with land-ownership data in a geographic information system to facilitate classification with maximum-likelihood, minimum-distance, and neural-network classifiers. The most effective classifications resulted from the use of the neural network on the enhanced TM data.
The synoptic view and the repetitive acquisition of LANDSAT imagery provide precise information, in real-time, for monitoring preserved areas based on spectral, temporal and spatial properties. The purpose of this study was to monitor, with the use of multispectral imagery, the systematic annual burning, which causes the degradation of ecosystems in the National Park of Araguaia. LANDSAT imagery of channel 5 (0.6 a 0.7 microns) and 7 (0.8 a 1.1 microns), at the scale of 1:250.000, were used to identify and delimit vegetation units and burned area, based on photointerpretation parameter of tonality. The results show that the gallery forest can be discriminated from the seasonally flooded 'campo cerrado', and that 4,14% of the study area was burned. Conclusions point out that the LANDSAT images can be used for the implementation of environmental protection in national parks.
Efforts to develop a technology for signature extension during LACIE phases 1 and 2 are described. A number of haze and Sun angle correction procedures were developed and tested. These included the ROOSTER and OSCAR cluster-matching algorithms and their modifications, the MLEST and UHMLE maximum likelihood estimation procedures, and the ATCOR procedure. All these algorithms were tested on simulated data and consecutive-day LANDSAT imagery. The ATCOR, OSCAR, and MLEST algorithms were also tested for their capability to geographically extend signatures using LANDSAT imagery.
The author has identified the following significant results. An effort to relate water parameters of the lower Chesapeake Bay area to multispectral scanner images of LANDSAT 1 has shown that some spectral bands can be correlated to water parameters, and has demonstrated the feasibility of synoptic mapping of estuaries by satellite. Bands 5 and 6 were shown to be useful for monitoring total particles. Band 5 showed high correlation with suspended sediment concentration. Attenuation coefficients monitored continuously by ship along three baselines were cross correlated with radiance values on three days. Improved correlations resulted when tidal conditions were taken into consideration. A contouring program was developed to display sediment variation in the lower Chesapeake Bay from the MSS bands.
The synergistic relationships among LANDSAT imagery, Skylab photographs, and aerial photographs were useful for establishing areas of near surface bedrock. Lineaments were located on LANDSAT imagery and aerial photographs during 1978 and near surface water tables were to be located during 1980. Both of these subjects can be identified by remote sensing methods more reliably than individual outcrops, which are small and occur in a wide variety of environments with a wide range of responses. Bedrock outcrops themselves could not be resolved by any of the data sources used, nor did any combination of data sources specifically identify rock at the ground surface. The data sources could not simply be combined mathematically to produce a visual image of probable areas of near surface bedrock. Outcrops and near surface bedrock had to be verified visually at the site. Despite these drawbacks, a procedure for locating areas of near surface bedrock within which actual surface outcrops may occur was developed.
The paper shows that accuracies of computer classification of species-specific forest cover types from Landsat imagery can be improved by 27% or more through the incorporation of topographic information from digital terrain tapes registered to multidate Landsat imagery. The topographic information improves classification accuracies because many common forest tree species have preferred elevation ranges and slope aspects. These preferences allow the separation of forest cover types which have similar spectral signatures but different species compositions. It is noted that the development of a classification system which uses prior probabilities and sets of prior probabilities conditioned by one or two external variables represents a significant increase in classification power.
The author has identified the following significant results. Both LANDSAT imagery and digital data were studied for usefulness in surveying water conditions of Minnesota lakes. Initial consideration was given to analysis of LANDSAT image densities because of the low technologic and cost requirements. The techniques employed, however, yield inconsistent and unreliable results. A set of criteria are given for using LANDSAT data in identification of three categories of particulate contaminants in Lake Superior. A linear transformation giving the relationship between the residual LANDSAT intensities and concentrations of three contaminants was obtained from correlation of remote sensing data with insitu measurements. LANDSAT imagery was found useful in placing peat bogs and fens in their respective geologic settings. Artificial disturbances and drainageways in peatlands could be recognized and classified.
The present study describes the validation of the the NASA team algorithm for the determination of sea ice concentrations from the Defense Meteorological Satellite Program special sensor microwave imager (SSM/I). A total of 28 cloud-free Landsat scenes were selected to permit validation of the passive microwave ice concentration algorithm for a range of ice concentrations and ice types. The sensitivity of the NASA team algorithm to the selection of locally and seasonally adjusted algorithm parameters is discussed. Mean absolute differences between SSM/I and Landsat ice concentrations are within 1 percent during fall using local and global tie points (standard deviations of the difference are +/-3.1 and +/-6.2 percent, respectively). In areas with greater amounts of nilas and young ice, the NASA team algorithm was found to underestimate ice concentrations by as much as 9 percent. It is inferred that the standard deviation between SSM/I and Landsat ice concentrations decreases from +/-7 to +/-5 percent with local tie points compared to the global ones for spring and fall.