Portal de Periódicos da CAPES

Connecting the Yakima fold and thrust belt to active faults in the Puget Lowland, Washington

2011; American Geophysical Union; Volume: 116; Issue: B7 Linguagem: Inglês

10.1029/2010jb008091

2156-2202

Richard J. Blakely, Brian L. Sherrod, Craig S. Weaver, Ray E. Wells, Alan C. Rohay, Elizabeth A. Barnett, Nichole E. Knepprath,

Geological and Geochemical Analysis

Journal of Geophysical Research: Solid EarthVolume 116, Issue B7 Geomagnetism and Paleomagnetism/Marine Geology and GeophysicsFree Access Connecting the Yakima fold and thrust belt to active faults in the Puget Lowland, Washington Richard J. Blakely, Richard J. Blakely [email protected] U.S. Geological Survey, Menlo Park, California, USASearch for more papers by this authorBrian L. Sherrod, Brian L. Sherrod U.S. Geological Survey at Department of Earth and Space Sciences, University of Washington, Seattle, Washington, USASearch for more papers by this authorCraig S. Weaver, Craig S. Weaver U.S. Geological Survey at Department of Earth and Space Sciences, University of Washington, Seattle, Washington, USASearch for more papers by this authorRay E. Wells, Ray E. Wells U.S. Geological Survey, Menlo Park, California, USASearch for more papers by this authorAlan C. Rohay, Alan C. Rohay Environmental Characterization and Risk Assessment Group, Pacific Northwest National Laboratory, Richland, Washington, USASearch for more papers by this authorElizabeth A. Barnett, Elizabeth A. Barnett U.S. Geological Survey at Department of Earth and Space Sciences, University of Washington, Seattle, Washington, USASearch for more papers by this authorNichole E. Knepprath, Nichole E. Knepprath U.S. Geological Survey, Menlo Park, California, USASearch for more papers by this author Richard J. Blakely, Richard J. Blakely [email protected] U.S. Geological Survey, Menlo Park, California, USASearch for more papers by this authorBrian L. Sherrod, Brian L. Sherrod U.S. Geological Survey at Department of Earth and Space Sciences, University of Washington, Seattle, Washington, USASearch for more papers by this authorCraig S. Weaver, Craig S. Weaver U.S. Geological Survey at Department of Earth and Space Sciences, University of Washington, Seattle, Washington, USASearch for more papers by this authorRay E. Wells, Ray E. Wells U.S. Geological Survey, Menlo Park, California, USASearch for more papers by this authorAlan C. Rohay, Alan C. Rohay Environmental Characterization and Risk Assessment Group, Pacific Northwest National Laboratory, Richland, Washington, USASearch for more papers by this authorElizabeth A. Barnett, Elizabeth A. Barnett U.S. Geological Survey at Department of Earth and Space Sciences, University of Washington, Seattle, Washington, USASearch for more papers by this authorNichole E. Knepprath, Nichole E. Knepprath U.S. Geological Survey, Menlo Park, California, USASearch for more papers by this author First published: 28 July 2011 https://doi.org/10.1029/2010JB008091Citations: 32AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Abstract [1] High-resolution aeromagnetic surveys of the Cascade Range and Yakima fold and thrust belt (YFTB), Washington, provide insights on tectonic connections between forearc and back-arc regions of the Cascadia convergent margin. Magnetic surveys were measured at a nominal altitude of 250 m above terrain and along flight lines spaced 400 m apart. Upper crustal rocks in this region have diverse magnetic properties, ranging from highly magnetic rocks of the Miocene Columbia River Basalt Group to weakly magnetic sedimentary rocks of various ages. These distinctive magnetic properties permit mapping of important faults and folds from exposures to covered areas. Magnetic lineaments correspond with mapped Quaternary faults and with scarps identified in lidar (light detection and ranging) topographic data and aerial photography. A two-dimensional model of the northwest striking Umtanum Ridge fault zone, based on magnetic and gravity data and constrained by geologic mapping and three deep wells, suggests that thrust faults extend through the Tertiary section and into underlying pre-Tertiary basement. Excavation of two trenches across a prominent scarp at the base of Umtanum Ridge uncovered evidence for bending moment faulting possibly caused by a blind thrust. Using aeromagnetic, gravity, and paleoseismic evidence, we postulate possible tectonic connections between the YFTB in eastern Washington and active faults of the Puget Lowland. We suggest that faults and folds of Umtanum Ridge extend northwestward through the Cascade Range and merge with the Southern Whidbey Island and Seattle faults near Snoqualmie Pass 35 km east of Seattle. Recent earthquakes (MW ≤ 5.3) suggest that this confluence of faults may be seismically active today. Key Points Tectonic connections between Cascadia forearc and backarc regions in Washington New aeromagnetic data illuminate Quaternary faults in central Washington Implications for seismic hazards in Pacific Northwest urban areas 1. Introduction [2] The Cascadia convergent margin (northern California, Oregon, Washington, and British Columbia) is a region of profound tectonism and magmatism, ultimately caused by oblique subduction of the Juan de Fuca plate beneath North America (Figure 1). The Pacific plate, moving northwestward ∼50 mm/y relative to North America, establishes dextral shear across the Juan de Fuca plate and western North America [Atwater, 1970; DeMets et al., 1994]. Most of the Pacific-North America relative motion is accommodated by the Juan de Fuca spreading center and Cascadia subduction zone, but the remaining 20 to 25 percent is distributed across Oregon and Washington [McCaffrey et al., 2007]. Oregon is rotating clockwise at about 1°/Ma, squeezing Washington against a slower-moving Canadian buttress [Wells et al., 1998; McCaffrey et al., 2007]. Most people in this region live in the seismically active forearc lowland, consisting of the Puget Lowland in Washington, Willamette Valley in Oregon, and Fraser Lowland in British Columbia. Figure 1Open in figure viewerPowerPoint Tectonic and magmatic setting of the Pacific Northwest. Red lines are faults from U.S. Geological Survey Quaternary fault database (http://earthquake.usgs.gov/hazards/qfaults). Quaternary volcanoes from Guffanti and Weaver [1988]. Earthquake locations and magnitudes from the Pacific Northwest Seismic Network and historic records, as compiled by Katie Keranen (written communication, 2009). Olympic Wallowa lineament as described by Raisz [1945]. Green dashed rectangle shows area of Figure 2. F, Fraser Lowland; P, Puget Lowland; W, Willamette Valley. Pole of rotation is for Oregon Coast Range domain relative to North America, 1.02° Myr−1 [McCaffrey et al., 2007]. [3] In the Puget Lowland, a complex system of east-west and northwest trending faults (Figure 2) accommodates 4.4 ± 0.3 mm/yr of permanent north-south shortening [Wells et al., 1998; Mazzotti et al., 2002; McCaffrey et al., 2007]. Numerous paleoseismic studies [Nelson et al., 2003; Johnson et al., 2004a; Sherrod et al., 2004; Kelsey et al., 2004; Sherrod et al., 2008] demonstrate that Puget Lowland faults produced earthquakes as large as MW 7 in Holocene time, and a diffuse pattern of modern-day earthquakes (Figure 2) shows that many of these faults likely remain active today. Puget Lowland faults are largely concealed by young glacial deposits, water, and urban development, and much of what we know about their mapped location and three-dimensional framework has come from geophysical investigations [e.g., Johnson et al., 1994, 1996; Brocher et al., 2001; Blakely et al., 2002; Pratt et al., 1997]. Figure 2Open in figure viewerPowerPoint (a) Earthquakes and Quaternary faults of western and central Washington and northern Oregon. Earthquake locations and magnitudes provided by Katie Keranen (written communication, 2009). YFTB, Yakima fold and thrust belt. Red lines are Quaternary faults from U.S. Geological Survey Quaternary fault database (http://earthquake.usgs.gov/hazards/qfaults). DM, Devils Mt.; SW, Southern Whidbey Island; SF, Seattle; TF, Tacoma; OF, Olympia; DF, Doty; SC, Straight Creek; FH, Frenchman Hills; SM, Saddle Mountains; UR, Umtanum Ridge; AR, Ahtanum Ridge; RM, Rattlesnake Mountain; TR, Toppenish Ridge; HH, Horse Heaven Hills; CH, Columbia Hills; GM, Gable Mountain. Cities and towns indicated by white boxes: E, Everett; S, Seattle; T, Tacoma; O, Olympia; P, Portland; D, The Dalles; PA, Pasco; Y, Yakima; EL, Ellensburg; W, Wenatchee; H, Hyak (Snoqualmie Pass). Yellow triangles are major volcanoes: GP, Glacier Peak; Mount Rainier; MSH, Mount St. Helens; MA, Mount Adams. Black dashed rectangle is location of Figures 4, 6–10. (b) Magnetic anomalies of central Washington and northern Oregon [Finn et al., 1989; Roberts et al., 1997]. Dotted lines indicate boundaries of high-resolution aeromagnetic data used in subsequent figures: PL, Puget Lowland magnetic survey [Blakely et al., 1999]; CE, Cle Elum survey; HA, Hanford survey. Label D indicates north-northwest striking magnetic anomalies interpreted as dikes by Swanson et al. [1979] and discussed in text. [4] As its name implies, the Yakima fold and thrust belt (YFTB), situated on the east side of the Cascade Range, is also a region of profound deformation [e.g., Reidel et al., 1989a, 1989b]. Columbia River Basalt Group (CRBG; 17.5–6.0 Ma) and suprabasalt sedimentary deposits record Late Miocene and Pliocene faulting and folding in the YFTB. Paleoseismic investigations show that the YFTB continued to evolve in Quaternary time [West et al., 1996; Campbell and Bentley, 1981; Repasky et al., 2009], and moderate-sized earthquakes in the region suggest that some of these structures remain active today (Figure 2). [5] If extended southeastward along strike, important Holocene faults in the Puget Lowland coincide with Quaternary faults of the YFTB. Structures linking the two regions would cross the Cascade Range along the Olympic-Wallowa lineament (OWL, Figure 1), a topographic and structural lineament extending from the Olympic Peninsula in Washington to the Wallowa Mountains in Oregon [Raisz, 1945; Hooper and Conrey, 1989]. McCaffrey et al. [2000, 2007] argued from GPS measurements that the OWL is the primary locus of deformation resulting from clockwise rotation of Oregon and Washington relative to stationary North America. In this paper, we use (1) new aeromagnetic and paleoseismic data and (2) existing gravity and lidar (light detection and ranging) data to investigate the possibility that Puget Lowland and YFTB structures are kinematically linked, and we attempt to map the linking structures through the intervening Cascade Range of Washington. 2. Geologic and Structural Setting [6] Paleomagnetic [e.g., Simpson and Cox, 1977; Wells, 1990; Wells et al., 1998] and global positioning system (GPS) measurements [e.g., Mazzotti et al., 2002; McCaffrey et al., 2000, 2007] show that western Oregon and Washington are rotating clockwise with respect to stable North America at a rate of 0.4 to 1.0° Myr−1 and have been doing so at approximately steady rates for the last 10–15 Ma. This broad rotation relative to stable North America within British Columbia produces horizontal strain that varies in direction and magnitude throughout Cascadia. The region of Washington immediately west of the Cascade Range, for example, is translating northward at a rate of 4.2 to 6.2 mm/yr relative to stable North America, whereas regions immediately east of the Cascade Range are translating northeastward at ∼1/3 that rate [McCaffrey et al., 2007]. McCaffrey et al. [2007] modeled Pacific Northwest GPS velocities with strain accommodated between rigid blocks. The YFTB constitutes two block boundaries in their model, together accommodating ∼3 mm/yr of northeast shortening. [7] It should be noted that the various maps employ two published fault databases. When discussing mapped geology, we show all faults regardless of age, as mapped at 1:100,000 scale by the Department of Natural Resources, Washington State. When the purpose of the map is to describe geologic hazards, we only show faults currently considered to have been active in Quaternary time (http://earthquake.usgs.gov/hazards/qfaults). The latter database is essentially a subset of the former, and captions indicate which database was employed. Puget Lowland [8] North-south compression in westernmost Washington is accommodated in part by a system of east-west and northwest striking crustal faults crossing the Puget Lowland (Figure 2). Many of these faults have been active in Holocene time and are spatially and structurally associated with large structural basins and uplifts observable in gravity and seismic data. Three Holocene faults are particularly important to subsequent discussion: (1) The east striking Seattle fault is a north verging thrust fault that, over the course of the last 40 m.y., has lifted its hanging wall up and over the Seattle basin to the north, now a 9 to 10 km deep basin filled with Oligocene and younger sedimentary rocks and glacial deposits [Johnson et al., 1994; Pratt et al., 1997; Blakely et al., 2002; Brocher et al., 2001, 2004; ten Brink et al., 2002]. (2) The Tacoma fault, along the southern margin of the Seattle uplift, is a south verging thrust fault serving as the structural contact between the uplift and the Tacoma basin to the south [Johnson et al., 2004b; Sherrod et al., 2004]. (3) The northwest striking Southern Whidbey Island fault, extending from near Vancouver Island to east of Seattle, accommodates right-lateral oblique displacement and forms the southwestern structural margin of the Everett basin [Johnson et al., 1996; Sherrod et al., 2008]. The Seattle and Southern Whidbey Island faults, if continued eastward along strike, merge near Snoqualmie Pass (Figure 2, label H) 35 km east of Seattle. [9] Holocene fault scarps of the Puget Lowland are each associated with linear magnetic anomalies that reflect the juxtaposition of lithologic units of differing magnetic properties. This association has proved useful in mapping Puget Lowland faults where concealed beneath Pleistocene glacial deposits, water, and urban development [e.g., Blakely et al., 2002, 2009; Sherrod et al., 2008]. With the advent of airborne lidar methods during the past decade, numerous topographic scarps were discovered throughout the Puget Lowland. These scarps often offset Pleistocene and younger surfaces and lie parallel and very near to linear magnetic anomalies [e.g., Sherrod et al., 2008]. Paleoseismic excavations, targeted at lidar-identified scarps, show a rich history of Holocene earthquake activity throughout the Puget Lowland [e.g., Nelson et al., 2003; Sherrod et al., 2008; Johnson et al., 2004a]. In particular, the Seattle, Tacoma, and Southern Whidbey Island faults each produced multiple MW 6.5–7.0 earthquakes in Holocene time. Yakima Fold and Thrust Belt [10] Flows of the CRBG (Figure 3) dominate the geologic landscape of the YFTB. CRBG flood basalts erupted from vents and fissures in southeastern Washington and northeastern Oregon 17.5 Ma to 6.0 Ma (Figure 4). Together they filled a volume of 174,000 km3, covered an area of about 163,700 km2 [Tolan et al., 1989, 2009], and in some cases flowed all the way to the Pacific Ocean [Beeson et al., 1989; Wells et al., 2009]. By far the greatest pulse of CRBG extrusion occurred 17.0 to 15.6 Ma, with eruption of the Grande Ronde Basalt: approximately 120 individual flows that together comprise 148,600 km3 of basaltic lava [Reidel et al., 1989b]. The Grande Ronde Basalt is the most prevalent CRBG formation in our study area. Eruptions of Grande Ronde Basalt spanned four polarity intervals (Figure 3), at least two of which (R2 and N2) are exposed in our study area (Figure 4). Figure 3Open in figure viewerPowerPoint Stratigraphic elements of the CRBG [Reidel et al., 1989b]. N, R, T, E refer to normal, reversed, transitional, and east directed magnetizations, respectively. Figure 4Open in figure viewerPowerPoint Generalized geology of the YFTB and surrounding regions, simplified from Walsh et al. [1987], Schuster et al. [1997], Stoffel et al. [1991], and Dragovich et al. [2002]. Black solid lines are faults of all types and ages, modified from the above references. Note that these faults differ from those shown on Figures 1, 2, 6, 7, 9, 10, and 20, which reflect Quaternary faults only (http://earthquake.usgs.gov/hazards/qfaults). Black dotted line indicates western limit of exposed Columbia River Basalt Group. Black dashed line shows extent of high-resolution aeromagnetic surveys discussed in text. Red line is location of gravity and magnetic model discussed in text and shown in Figure 20. Magenta symbols are deep exploratory boreholes [Reidel et al., 1989b] discussed in text. UR, Umtanum Ridge. [11] North-south compression of the YFTB has formed a series of east-west anticlines, synclines, and associated faults more or less evenly spaced across the landscape (Figures 2 and 4). The mapped distribution of YFTB structures essentially fans out in the westward direction (Figure 2), such that the northern anticlines (Frenchman Hills, Saddle Mountains and Umtanum Ridge) are directed northwestward toward the Puget Lowland, and southern anticlines (Toppenish Ridge, Horse Heaven Hills, and Columbia Hills) are directed southwestward toward Portland, Oregon. This distribution suggests complexities in the north-south compressive strain over the course of YFTB evolution. The region now occupied by the YFTB was a focus of back-arc subsidence before, during, and after CRBG emplacement [Reidel, 1984, Reidel et al., 1989a, 1994], and YFTB anticlines continued to evolve during this time so that the degree of folding increases with depth into CRBG stratigraphy [Reidel, 1984]. Deformation of CRBG also postdates its eruption and emplacement [e.g., Mitchell and Montgomery, 2006; West et al., 1996; Campbell and Bentley, 1981], as discussed in the following sections. [12] CRBG flood basalts naturally filled whatever terrain existed at the time of extrusion, but in general, each newly erupted flow formed a quasi-horizontal layer that recorded subsequent folding and faulting. This obvious point has important implications for geophysical analysis: Basalts are strongly magnetic, having induced magnetizations on the order of 0.1 to 0.5 A/m and natural remanent magnetizations of 1 to 10 A/m. Deformed CRBG flood basalts, therefore, produce distinctive magnetic anomalies that facilitate mapping and characterization of these structures. The CRBG includes both normal and reversely magnetized flows, as well as flows with transitional remanent magnetizations. CRBG flows have Koenigsberger ratios generally >10 (Figure 5), and the juxtaposition of normal and reversed flows adds complexity to the analysis of magnetic anomalies. Figure 5Open in figure viewerPowerPoint Koenigsberger ratios (Q) for a set of CRBG rocks. Data provided by Jon Hagstrum (written communication, 2009). [13] Pre-Tertiary rocks exposed in the northern part of our study area (Figure 4) likely extend in the subsurface beneath parts of the CRBG [Campbell, 1989]. These basement rocks include Jurassic metamorphic and ophiolitic rocks of the Ingalls Tectonic Complex [Dragovich et al., 2002] and Cretaceous granitic rocks of the Mount Stuart batholith [Smith, 1904; Dragovich et al., 2002]. As discussed subsequently, gravity anomalies indicate the subsurface distribution of these lithologies. [14] Continental sedimentary and volcaniclastic rocks comprise the Tertiary stratigraphy above pre-Tertiary basement and below CRBG flows, and variations in their stratigraphic thickness have important implications for both hydrocarbon exploration and earthquake hazards. Geophysical studies indicate that basement relief exceeds relief on the base of CRBG [Saltus, 1993; Jarchow, 1991], implying that YFTB deformation was underway before and continued during CRBG emplacement. If it can be shown that topography on the basement interface is spatially associated with CRBG folds and faults, this would provide a strong case that faults seen at the surface extend into the basement rather than shoaling into detachment surfaces within or at the base of CRBG. In the study area, exposures of pre-CRBG Tertiary rocks include Eocene continental sedimentary rocks of the Swauk, Manastash, Roslyn, and Chumstick Formations; and volcanic rocks of the Oligocene Ohanepecosh Formation and Miocene Fifes Peak Formation [Tabor et al., 2000; Walsh et al., 1987; Stoffel et al., 1991; Schuster et al., 1997; Dragovich et al., 2002]. [15] CRBG flows are overlain and intercalated by Miocene continental sedimentary and volcaniclastic rocks of the Ellensburg Formation, which in turn is overlain by Pliocene-Miocene Ringold Formation, Pliocene Thorp Gravel, and Pleistocene outburst flood deposits [Walsh et al., 1987; Schuster et al., 1997; Dragovich et al., 2002]. All of these post-CRBG units are deformed, including the 1 Ma to 12 ka Hanford Formation [Repasky et al., 2009]. Olympic-Wallowa Lineament [16] A regional-scale topographic lineament, extending from the Olympic Peninsula to the Wallowa Mountains, is superimposed on the tectonic magmatic framework of the Pacific Northwest (Figure 1). The origin of the Olympic-Wallowa lineament, first recognized 65 years ago by Raisz [1945], is still a matter of discussion. Kienle et al. [1977] referred to that part that crosses the central Columbia Basin as the Cle Elum-Wallula (CLEW) lineament. Within the YFTB, the OWL corresponds with the Rattlesnake Mountain and Umtanum Ridge anticlines (Figure 2). Beyond the Columbia basin, the OWL aligns with linear topographic features in pre-CRBG rocks both northwest and southeast of these anticlines [Campbell, 1989]. Hooper and Conrey [1989] envisioned the OWL as a giant megashear accommodating differential extension rates, with opening rates to the south approximately 20 percent greater than regions to the north. A differential opening rate implies shear strain along the OWL, but direct evidence of strike-slip displacement has been elusive [Tabor et al., 1984; Reidel et al., 1989a; Price and Watkinson, 1989; Saltus, 1993]. Moreover, dextral shear on the OWL is not evident in GPS measurements. McCaffrey et al. [2007] document northeast directed shortening across the OWL, decreasing in magnitude to the southeast. In their microplate model, the OWL forms a boundary between YFTB blocks, with poles of rotation in Idaho near the southeastward projection of the OWL. 3. Geophysical Analysis Aeromagnetic Anomalies [17] Two high-resolution airborne magnetic surveys acquired by the U.S. Geological Survey in 2008 and 2009 (Figure 6) help improve our understanding of the YFTB and possible links to neighboring tectonic structures. The surveys were acquired by separate geophysical companies working under contract to the USGS. Both surveys measured the total magnetic field at a nominal altitude of 250 m above ground or as near to the ground as safely possible. Topographic relief in some parts of the study area necessitated higher flight altitudes; nevertheless, 87 percent of the area was flown at altitudes less than 500 m above ground, and 99.4 percent at altitudes less than 1000 m. Flight lines were directed east-west in the western survey and north-south in the eastern survey. In each survey, flight lines and perpendicular tie lines were spaced 400 m and 4 km apart, respectively. Stationary magnetometers were operated continuously during data acquisition in order to monitor and subsequently correct for transient magnetic fields. Total field measurements were reduced to total field anomaly values by subtraction of the International Geomagnetic Reference Field, updated to the date of flying. The two new surveys were gridded at 100 m spacing, then merged with each other and with a third high-resolution magnetic survey flown over the Puget Lowland in 1997 [Blakely et al., 1999]. Each of the three surveys overlaps its neighbors by small amounts, which facilitated the merging procedure. Merging was accomplished by determining a suture path within the overlap regions. Mismatches at each point of the suture path were then corrected within circular regions surrounding each point. Figure 6 shows merged total field magnetic anomalies reduced to the pole. Together, the three aeromagnetic surveys extend from the YFTB to the Puget Lowland and include the intervening region where the OWL crosses the Cascade Range. Figure 6Open in figure viewerPowerPoint Total field magnetic anomalies of the YFTB and surrounding regions. Brightly colored region shows magnetic anomalies from a merge of three high-resolution surveys (see Figure 2b for survey identification). Anomalies transformed to the north magnetic pole. Subdued background colors show magnetic anomalies from a statewide compilation [Finn et al., 1989] reduced to pole. White lines are Quaternary faults from the U.S. Geological Survey Quaternary fault database (http://earthquake.usgs.gov/hazards/qfaults). Note that these faults differ from those shown in Figure 4, which show faults of all ages. Dotted line shows mapped western extent of CRBG. Red line is location of magnetic and gravity profile (Figure 20). Label D indicates north-northwest striking magnetic anomalies interpreted as dikes [Swanson et al., 1979]. See Figure 2 for description of other labels. [18] The new airborne magnetic surveys cover the entire northern parts of the YFTB, including the Frenchman Hills, Saddle Mountains, Umtanum Ridge, Ahtanum Ridge, and Toppenish Ridge anticlines, and their possible extensions into the Cascade Range (Figure 6). The surveys are underlain by rocks with diverse magnetic properties, ranging from highly magnetic CRBG, with both normal and reversed remanent magnetization, to essentially nonmagnetic sedimentary rocks. Flows of the CRBG produce a distinctive, short-wavelength pattern of magnetic anomalies, making it possible to trace the western extent of CRBG volcanic units (Figure 6, dotted line). Each faulted anticline within the CRBG produces a clear aeromagnetic lineament, which is expected considering the high magnetizations of CRBG rocks. Of particular note in Figure 6 is the close alignment of distinct magnetic gradients with each Quaternary fault cataloged by the U.S. Geological Survey (http://earthquake.usgs.gov/hazards/qfaults). Additional linear anomalies lie west of CRBG exposures with trends similar to Quaternary faults mapped within the CRBG. [19] Figure 2a shows significant topographic relief associated with YFTB anticlines and synclines, and we should consider the possibility that linear magnetic anomalies in Figure 6 are caused simply by topographic effects in this highly magnetic terrain. To investigate this possibility, we calculated the magnetic anomalies that would be observed on a horizontal surface immediately above the highest topography assuming uniformly magnetized crust and using the method of Parker [1972]. These calculated anomalies showed little resemblance to observed anomalies when continued upward to the same level. We conclude that long-wavelength topographic anomalies are not significant in this area, and that observed anomalies instead are caused mostly by subsurface magnetic sources (e.g., faulted contacts, folded layers, changes in magnetic polarity, etc.). [20] Figure 7 shows magnetic anomalies filtered in order to emphasize shallow magnetic sources. These anomalies were calculated by analytically continuing the original magnetic field (Figure 6) 50 m upward, then subtracting that result from the original field. This two-step procedure is equivalent to a discrete vertical derivative, which emphasizes magnetic anomalies caused by shallow sources at the expense of anomalies originating from deeper sources [Blakely, 1995]. Regions of distinctly different magnetic character are apparent in Figure 7. For example, it is easy to distinguish near-surface CRBG from less magnetic lithologies, and it is apparent from this pattern that CRBG does not extend west of its geologically mapped surface exposures (Figure 7, dotted line). Figure 7Open in figure viewerPowerPoint Residual magnetic anomalies of the YFTB and surrounding regions. Magnetic anomalies of Figure 6 filtered in order to emphasize shallow magnetic sources, as described in text. White lines are Quaternary faults (see Figure 6). See Figure 6 for description of other items. [21] Figure 7 dramatically illuminates the major folds and thrusts of the YFTB, with distinct linear anomalies closely aligned along each mapped Quaternary fault (http://earthquake.usgs.gov/hazards/qfaults). A number of linear anomalies with similar trend lie northwest of the mapped faults and beyond exposed CRBG. In some cases, these linear anomalies can be explained by mapped lithologic contacts. For example, the northwest striking magnetic lineament immediately west of Cle Elum (Figures 6 and 7) overlies a fault-bounded sliver of Eocene volcanic rocks [Walsh et al., 1987; Dragovich et al., 2002]. Other northwest striking lineaments, however, are not associated with mapped features and apparently originate from lithologic contacts concealed by younger rocks and surficial deposits. [22] We employed a method described by Phillips et al. [2007] to assist in mapping aeromagnetic lineaments. Black lines (made up of intersecting black dots) in Figure 8 indicate the locations of magnetic contacts calculated directly from the mathematical curvature of total field magnetic anomalies. Although this method is entirely objective, it does involve several simplifying assumptions, the most significant being the assumption that faults dip vertically. Violations of this assumpti