Search NASA⌕ Search

Engineering topics

Ranson, K. J.

Publications and source records attributed to Ranson, K. J..

40 records · Page 3

Interpreting vegetation reflectance measurements as a function of solar zenith angle

Spectral hemispherical-conical reflectances of a nadir looking sensor were taken throughout the day for a lodgepole pine and two grass canopies. Mathematical simulations of both spectral hemispherical-conical and bi-hemispherical reflectances were performed for two theoretical canopies of contrasting geometric structure. These results and comparisons with literature studies showed a great amount of variability of vegetation canopy reflectances as a function of solar zenith angle. Explanations for this variability are discussed and recommendations for further measurements are proposed.

Kimes, D. S.↗

The mixture problem in computer mapping of terrain: Improved techniques for establishing spectral signature, atmospheric path radiance, and transmittance

The results of LANDSAT and Skylab research programs on the effects of the atmosphere on computer mapping of terrain include: (1) the concept of a ground truth map needs to be drastically revised; (2) the concept of training areas and test areas is not as simple as generally thought because of the problem of pixels that represent a mixture of terrain classes; (3) this mixture problem needs to be more widely recognized and dealt with by techniques of calculating spectral signatures of mixed classes, or by other methods; (4) atmospheric effects should be considered in computer mapping of terrain and in monitoring changes; and (5) terrain features may be used as calibration panels on the ground, from which atmospheric conditions can be determined and monitored. Results are presented of a test area in mountainous terrain of south-central Colorado for which an initial classification was made using simulated mixture-class spectral signatures and actual LANDSAT-1-MSS data.

Smedes, H. W.↗

Skylab-EREP studies in computer mapping of terrain in the Cripple Creek-Canon City area of Colorado

Multispectral-scanner data from satellites are used as input to computers for automatically mapping terrain classes of ground cover. Some major problems faced in this remote-sensing task include: (1) the effect of mixtures of classes and, primarily because of mixtures, the problem of what constitutes accurate control data, and (2) effects of the atmosphere on spectral responses. The fundamental principles of these problems are presented along with results of studies of them for a test site of Colorado, using LANDSAT-1 data.

Smedes, H. W.↗