Data Series 830
Data Processing Used for This Data SeriesIn processing and assessing the magnetic and gravity data in the near and intermediate vicinity of Newberry Volcano, several approaches were employed. Among those included here are Centre for Exploration Targeting (CET) analysis and analytic signal (both described below); because of the disparate quality of the gravity and magnetic data, standard band-pass filtering is not shown, as it only emphasized irregular data density and dataset merges. For the same reasons 2-D modeling is not realistically possible with the public-domain data. No processing beyond gridding and projection were used for the NURE radiometric data, as the data were unusually noisy and had gaps of as much as 10 km between flight lines. Only processing that gave images found to correlate with known features is discussed below. CET Grid Analysis CET grid analysis was developed by (and named for) the Centre for Exploration Targeting, University of Western Australia. The approach uses phase congruency—discontinuities and linear features all have components that are maximally in phase after a 2-D Fourier transformation, whether one is looking for symmetric (line or cylinder) or asymmetric (step-offset) features. The CET approach has the advantage of providing a dimensionless value (phase congruency in the Fourier-transformed frequency domain) that is independent of amplitudes or contrast in the original data (Kovesi, 1997; Holden and others, 2008). CET grid analysis provides trend detection that begins with texture analysis. This is essentially image enhancement suitable for analyzing regions of subdued magnetic or gravity responses and is particularly useful at the large scales shown here. Texture analysis can first enhance the local data contrast, and this is followed by structure detection (the phase congruency), useful for identifying linear discontinuities (edge detection). The discontinuity structure detection takes the texture analysis output and builds a skeletal structure of the regions of discontinuity. The output is a set of a binary line segments separating each of the discontinuity regions, isolating orientation changes and offsets caused by discrete magnetic susceptibility or density structures. This process emulates the traditional manual drawing of interpretive lines but is done automatically over large regions and uses subtle information in the data often not apparent to the human eye. CET grid analysis is also a form of spatial autocorrelation. Regions of magnetic and gravity discontinuity correspond with, and can reveal, lithological boundaries, faults, and dikes helpful to understand local and regional geology. Analytic SignalAnalytic Signal (Nabighian, 1972; Roest and others, 1992; Blakely, 1996) is one of several methods for calculating depth to source from gravity and magnetic data (discussed further below). Because a magnetic field is dipolar, a magnetic “high” rarely lies over its source, and geophysicists have long searched for ways to bypass the heavy processing previously necessary to deal with this issue. Nabigian (1972) first developed the concept of Analytic Signal, or energy envelope, of magnetic anomalies. It was used in the current effort not to model for depth to source but instead to assess the suitability of the data for 2-D and two-and-a-half dimensional (2.5-D) modeling. Roest and others (1992) showed that the amplitude (absolute value) of the three dimensional (3-D) analytic signal at any point can be easily derived from the square root of the summed squares of the three orthogonal gradients of the total magnetic field. The x and y derivatives can be calculated directly from a total magnetic field grid using a simple 3×3 digital filter, and the vertical gradient is routinely calculated using fast Fourier transform (FFT) techniques. The advantage of the Analytic Signal is that it is independent of the vector or dipolar nature of the Earth’s magnetic field, as well as the asymmetric shape of most magnetic anomalies. It is also independent of the nature (remnant versus induced magnetism) of the source of the anomalous magnetic field of interest. The Analytic-Signal shape can be readily used to determine the depth to the magnetic sources—depth is equal to the anomaly width at half amplitude. Because the Analytic-Signal algorithm works intimately with gradients, it is very sensitive to certain types of noise, especially dataset splices. This will be shown and discussed further below. |
|
|
Suggested citation: Wynn, Jeff, 2014, Gravity, magnetic, and radiometric data for Newberry Volcano, Oregon, and vicinity: U.S. Geological Survey Data Series 830, https://dx.doi.org/10.3133/ds830. U.S. Department of the Interior U.S. Geological Survey |
