Scientific Investigations Report 2006–5218
U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2006–5218
The individual interferograms represent vertical displacement that occurred between SAR data acquisitions (table 1) and the stacked interferogram represents vertical displacement that occurred from September 8, 1992, to December 10, 1999. There is a three-color and a multicolor interferogram for each individual period, whereas there is only a three-color stacked interferogram. The deformation represented in the interferograms depicts altitude changes relative to a point of assumed zero vertical displacement; zero horizontal displacement is assumed.
The Landsat images (fig. 4), acquired in 1993 (Landsat 5) and 2000 (Landsat 7), provide a spatial reference for the interferograms, indicate land use, and, when compared, they show urban development between 1993 and 2000. The Landsat images are presented with the near-infrared, red, and green bands (4, 3, and 2, respectively) colored red, green, and blue, respectively. With this coloration, red is indicative of healthy growing vegetation, black indicates open water, and gray or white indicates bare soil.
To determine areas of deformation and amounts of vertical displacement for a given period, it is advantageous to examine both types of individual interferograms. Three-color interferograms typically are better for identifying areas of deformation, delineating large deformation patterns, and estimating the amount of displacement. Multicolor interferograms typically are better for estimating relative displacement, identifying decorrelated areas, and delineating localized deformation patterns. Displacement measurements and deformation patterns cannot be estimated for decorrelated areas and should be estimated with caution where atmospheric phase noise is suspected (see “Limitations” section).
Temporal decorrelation may result from the relative movement of reflectors within a radar-resolution cell and often is associated with the movement of vegetation parts (leaves, twigs), vegetation growth or removal, changes in soil moisture, erosion, and, commonly in Las Vegas Valley, urban development. Decorrelated areas are pixilated in interferograms, appearing as a range of stippled colors, and more easily identified in multicolor interferograms than in three-color interferograms (fig. 5). Temporal decorrelation in the Las Vegas Valley interferograms typically is limited to small areas, thus the remainder of the interferogram can be evaluated. Golf courses, easily identified as red areas in the Landsat images (fig. 5B), commonly are associated with temporal decorrelation, likely resulting from plant growth, landscaping changes, or watering practices.
Atmospheric phase noise results from changes in atmospheric conditions between data acquisitions and, therefore, occurs to varying degrees in all interferograms. Thus, all interferograms should be interpreted with caution because atmospheric phase noise can modulate displacement signals. Atmospheric effects can be difficult to identify because they can occur across an entire interferogram and can resemble deformation patterns. The location, shape, and magnitude of the noise is assumed unique to specific acquisition dates, thus if atmospheric effects are suspected in an interferogram, other interferograms with a common acquisition date should have similar noise patterns. Because land-subsidence and uplift trends in Las Vegas Valley typically have annual cycles (Pavelko, 2000; Hoffmann and others, 2001) but atmospheric phase noise usually is associated with specific acquisition dates, careful examination and comparison of multiple interferograms can assist in determining whether a pattern is the result of deformation or other phase noise. As a rule of thumb for these Las Vegas Valley interferograms, patterns that annually repeat typically are the result of land deformation and patterns that appear only in interferograms with common acquisition dates typically are the result of phase noise. For example, interferograms with an acquisition date of May 29, 1998, display phase noise likely related to atmospheric moisture changes (fig. 6; table 1).
The individual three-color interferograms depict 224 mm of vertical displacement with a gradational color scheme that ranges from purple (subsidence) to yellow (minimal or no displacement) to green (uplift; fig. 2) and the stacked interferogram depicts 448 mm of vertical displacement that ranges from red (subsidence) to yellow (minimal or no displacement) to blue (uplift; fig. 3). Deforming areas and deformation patterns are relatively easy to identify by noting the location and shapes of darker purple and darker green areas for individual interferograms and darker red and darker blue areas for stacked interferograms. The color-scale bars show the magnitude of displacement.
Multicolor interferograms depict displacement with a repeating color sequence (fig. 2). One continuous sequence of colors represents 20 mm of displacement. Within a color sequence, any single color with a continuous and arcuate or linear pattern is analogous to a displacement contour. The order of colors indicates whether deformation is subsidence or uplift, as shown in the color-scale bar.
Displacement should be estimated only for areas with well-defined partial, full, or multiple sets of colors that are surrounded by or adjacent to areas with no displacement. Generally, broad areas of one color are areas with no displacement. The estimated displacement is relative only to the adjacent area and is not necessarily the total amount of displacement. Relative displacement is estimated by noting the colors between the area of interest and the adjacent area. If the colors are not a full sequence, use the color-scale bar to estimate the relative displacement. If the colors include one or more sequences, assign 20 mm of displacement to each full sequence, treat any remaining partial sequence as above, and add the estimated partial-sequence amount to the full-sequence amount.
Multicolor interferograms typically are better than three-color interferograms for identifying decorrelated areas because the color contrast between decorrelated and correlated areas is higher in multicolor interferograms. Localized deformation patterns typically are easier to delineate in multicolor interferograms because 20 mm of displacement is represented by multiple colors, whereas for three-color interferograms, 224 mm of displacement is represented by only three colors (fig. 2).
< Back to Table of Contents
< Previous Section |
Next Section >
| AccessibilityFOIAPrivacyPolicies and Notices | |
  |
|
