2-1: The Niobrara Fm is bright, having the highest average reflectances through the whole spectral interval. In the visible, it has the highest reflectance in the red but reflectances in the green and blue are not much lower, so there is a roughly even mix of the three primary colors, yield dominantly a white with a weak yellowish red overtone. And that is how it looks in the field. The Niobrara typically is a chalky limestone which everywhere, when freshly exposed, is whitish to gray. At the other extreme, the Thermopolis Fm is a black shale and this is indicated by its low reflectances in the visible, having about the same value from blue to red (i.e., a low reflectance, spectrally uniform material is black to dark gray). The Frontier Fm is commonly brownish, but has a greenish member (from the mineral Glauconite); its spectral curve peaks in the green in the visible. The Chugwater Fm has a spectral curve in the visible that rises from low to high reflectance from blue to red - it is thus reddish. In the field, it is a distinctive unit visible for miles away by its medium-red color. BACK
2-2: As said in the previous answer, the Thermopolis is a blackish shale. The spectral curve for theMowry Fm is similar in shape to the Thermopolis but has a much higher average reflectance. It is medium-gray in hand specimen and in the field (often with a grayish-yellow weathered surface). The Jelm Sandstone has a similar spectral curve in the visible to that of the White River Conglomerate. Thus, they would not be readily separable in a visible image. There is about a 5% difference in reflectance in the Near-IR segment (around 0.9 mm) sampled by MSS 7 and TM 4. The main distinction is at longer wavelengths in the Near-IR. The Jelm has absorption bands at 1.9 and 2.3 mm; the White River doesn't have these bands. BACK
2-3: You probably feel confident in separating 4 to 5 or 6 at most. Actually, there are 14 separable formations (or, in one case, two members within a formation) present in the scene. Only one unit, which is quite light toned, really stands in sharp contrast to all others. Other bands do a better job at separation, as we will see later in this section. BACK
2-4: An anticline is an upfold, the rocks curving upwards in an arch, with the top being a crest and the sides being inclined limbs moving downward in opposite directions. A syncline is the reverse, a downfold, with its lowest point in the center or trough. These folds usually are layered and are three dimensional. They tend to die out or plunge in either direction perpendicular to the place where they are maximally folded. Anticlines grade into synclines into anticlines into synclines into ŠŠ in a fold belt (in the Harrisburg scene we studied at the end of Section 1, the ridges were mostly anticlines that were eroded such that rock units resistant to erosion remained topographically above weaker units). If you are familiar with the plot of the mathematical "sine" curve, its sequence of up and down curves is analogous to anticlines and synclines. Up and down water waves are also comparable. We will see images of these folds in other sections of the Tutorial. BACK
2-5: At least four major units are distinguishable. Actually, there are several more formation present but these are either thin and hard to see or occur at the base of the monocline. The black tone in the other aerial photo is actually vegetation, as seen in the foreground of the first color photo. BACK
2-6: Band 7 is, overall, darker. The Navajo Fm, bright with high contrast in Band 1, shows only slight contrast in Band 7. BACK
2-7: They both do a good job. But the Color Hybrid appears to better define and distinguish units in the left 2/3rd of the image and the PCA the remaining upper right 1/3rd. Choice of color assignment also helps to discriminate differences, as for example, the color contrast between Upper and Lower Moenkopi members. BACK
2-8: Some of the black is probably shadows related to cuts in the ridges, mesa walls, etc; the Navajo false alarms may be an artifact of topography - certain slopes cut into rocks with normally different signatures may be facing the incoming sun, an effect we noted in the Morro Bay scene; the alluvium is probably a mixture of talus (rock debris that falls from cliffs) at the base of steep hills (note where some of the white is located), slopewash from the higher ground on either side, and floodplain deposits from the small, usually dry creek. The Mancos unit crops out upstream, and contains less of the alluvial cover, so that the Mancos signature dominates. BACK
2-9: The first IDIMS classification, while not as colorful or bright, seems superior to the Idrisi one. The precise choice of training sites may have been a factor; scene quality (relative reflectances) may also have been a factor. Different maximum likelihood classifiers were used in Idrisi and IDIMS - perhaps the one applied in the IDIMS system was more sensitive. The winter IDIMS image suffers in quality (hence, in accuracy) to the summer one, largely because of low Sun angle, which did not illuminate the scene as efficiently. The black patterns are mostly shadows in the erosional recesses in the hogback. BACK
2-10: Most of the eastern U.S. is covered by forests, grasses, crops, or urban areas; in fact, it has been estimated that less than 2% of the eastern surface consists of exposed bedrock and that is strongly weathered (abnormal) so that it doesn't show fresh rock. In the deserts, weathering can produce a thin weathered surface, usually rich in iron, that coats the rock (in places more than 40% of the ground surface is outcrop) so that those of different colors and other properties are masked by this uniform-appearing coating, making discrimination of rock type difficult. The Waterpocket Fold area shows minimal coatings. BACK
2-11: The anticlines are both structurally and topographically high - making up the anticlinal mountains. The synclines are downcurves and here they occupy the valleys which are in process of being filled with erosional debris (alluvium) washed in from mountain erosion. BACK
2-12: There is a belt of folding running diagonally across the image that strongly resembles the folds we saw in central Pennsylvania in the Section 1 exam. In Morocco, these folds are likewise ridges that stand out because their strata are more erosionally-resistant than weaker strata in-between. In a sense, this arid Atlas scene is like a Pennsylvania scene without the masking effect of vegetation. BACK
2-13: In the northern half of the mountains that run left-right across the central part of the scene, there is a second strike-slip or wrench fault (part of it is occupied by a dry stream [thin] valley). It, too, is left-lateral, that is, the northern half has moved westward. BACK
2-14: The Orthris Zone is hard to separate from the juxtaposed Pindus Zone. The Pelagonian Zone shows more topographic variability than the others and has both a valley and a mountain component. This would happen if that Zone were a thrust block that had internal stratigraphic continuity but its lower units were more easily eroded than the upper ones (the mountains to the east). Both zones were likely identified as separate primarily from field evidence, mainly as discontinuities in rock ages (i.e., juxtaposed rocks whose ages indicate some age intervals are missing). BACK
2-15: A fracture (or joint) is just a crack or break in the rock in which the rock on either side springs apart some small distance. A fault is a break in which the rock on one side slides or slips against the rock on the other side so that each side is displaced some distance from the other. As seen from the air or space, in a photo/image, a fracture is just a linear mark in which the tone of the rocks is the same on both sides. Most faults cause enough movement for individual layers or even formations to be displaced, so that there may be a sharp discontinuity in tonal pattern, in which one type of rock is brought against another. Or, in the China image, topographic parts of a mountain systems are visibly offset by the faulting. BACK
2-16: This is the way it is done professionally: Place a tissue overlay on the image and trace the fractures as a map. Now, start at any one fracture. Use a protractor to measure the angle it makes with the horizontal, from 0 to 180°. Record that angle. Mark the fracture line with a small cross-mark to indicate you have completed its measurement. Do the same for all other fractures. Place the angles you measure in a table of ranges - thus set up bins like 0 - 5°, 6 - 10°, 176 - 180°. Now make a plot of narrow wedges, each with a 5° angular width, for all of the above intervals from 0 to 180°. Fill in each width to a length set by the number of individual fractures in that angular interval (adopt some unit of length). You will end up with what is known as a "Rose Diagram". To see what this looks like, simply go back to the page you left and scroll down to the bottom of the second figure down. It has two such diagrams. Look at the fractures map and try to correlate their orientation frequencies with the Rose Diagram. BACK
2-17: You could use a geostationary satellite - one whose orbit is far out and has the satellite's velocity the same as the rotating Earth below. This see's the Earth at all times of day and night, so the angular illumination effect progressively enhances fracture/faults at different Sun azimuths. Trouble is, at that distance one would need a powerful telescope to get adequate resolution. Or, you could launch an afternoon counterpart to a morning overpass satellite like Landsat, with both travelling in the same orbital sequences but time staggered. However, nobody in NASA (or Congress) would buy this idea unless more uses than just fracture detection can be found to justify the huge expense. BACK
2-18: In the rose diagram for the West part of the scene, there is a notable trend running north-northwest that isn't picked up by the satellites. This is probably a Sun angle effect - this trend is real but is largely missed owing to illumination bias. BACK
2-19: The number of fractures (or, more properly, the density, or number per unit area) is less in the Superior province then in the Grenville Province. BACK
2-20: The Landsat images, even when enlarged, did not clearly demarcate or otherwise bring to view the cross-fracture system. The computer-based edge enhancement technique exposed the presence of these fractures which might have been missed otherwise. Examination of the photos from the aircraft flight could likely have done the same thing but that flight was expensive and really was done ex post facto to corroborate the Landsat evidence. BACK
Collaborators: Code
935 NASA GSFC, GST,
USAF Academy,
Contributor Information
Last Updated: July '99
Webmaster: Bill Dickinson Jr.
Site Curator: Nannette Fekete
Please direct any comments to the Site
Curator eerstweb@gst.com.