Chapter 3

Structural features in the study area include a zone of northwest trending near vertical faults coincident with the Olympic-Wallowa Lineament (OWL-zone), north to northeast trending near vertical faults of the Hite Fault System (HFS) and associated segment of the Blue Mountains Uplift, north to northwest trending near vertical faults associated with the La Grande Graben, shallowly dipping faults, and Wanapum basalt feeder dikes (Maps 1-4; Figure14). Many of these structures have been mapped and described by earlier workers (i.e. Kienle and others, 1979; Kendall, 1981; Swanson and others, 1981; Sandness and others, 1982). Faults and lineaments of the OWL-zone are relatively more abundant to the west while those associated with the Hite Fault System are more abundant to the east (Map 1; Figure14). Faults associated with the La Grande Graben are mostly located in the southeast corner of the study area, southeast of the Hite Fault (Map 1; Figure14). Detailed outcrop descriptions of all studied faults are provided in Appendices E and F.

Faults Coincident with the Olympic-Wallowa-Lineament

The greatest number of faults and lineaments present in the study area have a northwesterly trend and are apparently part of a broad structural zone coincident with the west-northwest trending OWL (Map 1 and Figure14 and Figure15). Slickensides associated with observed faults are dominantly sub-horizontal (Figures 16, 17, 18, and 19). Steeply plunging slickensides or grooves are only observed on four northwest trending faults zones (locations 7, 8, 14, and 17 on Map 2). At location 7, there is a cluster of normal faults (Figure 20 and Figure 21). A minor normal fault with poorly developed steeply plunging slickensides is present at location 8 along Tiger Creek Road (Figure 22). The fault zones at locations 14 and 17 have both sub-horizontal and steeply plunging slickensides ( Figure 23 and Figure 24). At location 14, vertical offsets are only about 2 meters (Figure 24). Offset on the same fault is even less at location 13 on Map 2 (Figure 13).

Where observed, indicators of the direction of strike-slip on northwest trending faults in the form of steps on slickenside surfaces or plunge of slickensides combined with apparent vertical offset consistently suggest right-lateral movement (for example, Figure 26; Appendices E and F). Documenting the degree of strike-slip displacement is difficult due to the nearly horizontal disposition of the basalt flows and lack of good markers such as offset vertical dikes. Where strike-slip displacements along northwest trending faults can be estimated, they are small and right-lateral, ranging between 5 and 50 meters (Appendices E and F). For example, the right-lateral strike-slip component calculated from plunge of slickensides combined with apparent vertical offset is between 23 and 38 meters at location 11 on Map 2 and between 8 and 47 meters at location 27 on Map 3.

The Forks Fault and the Lincton Mountain Faults are the only named OWL-zone faults in the mapped area. The N45^oW trending Forks Fault forms a greater than 10 meter wide breccia zone in the valley of the North Fork Walla Walla River along the county road in section 23 of township 5 north and range 36 east (location 26 on Map 3; Figure 27). Sub-horizontal slickensides with steps compatible with right-lateral displacement are present in the fault zone. Along Powerline Road (Map 3) the Forks Fault puts the lowermost Sand Hollow type flow and uppermost Grande Ronde unit flows side-by-side. This requires a down to the southwest displacement component of 35 to 45 meters. The Forks Fault does not crop out along Powerline Road, only a small sub-parallel fault is observed (Figure 28; location 25 on Map 3). The horizontal component of displacement is undetermined.

The steeply dipping N45^oW to N55^oW trending Lincton Mountain Faults are visible as topographic and aerial photographic lineaments to the south of the South Fork Walla Walla River on Lincton Mountain (Map 4). From Blalock Mountain, on the order of 10 meters of down to the north offset of one of the Lincton Mountain Faults located in section 9 of township 4 north, range 37 east is clearly visible (location 31 on Map 4). Kendall (1981) reports vertical offsets of as much as 30 meters on these faults.

The nearly linear valley of the South Fork Walla Walla River, while part of the original definition of the OWL by Raisz (1945), does not contain any faults with significant post-CRBG displacement. There is no noticeable vertical displacement of basalt flows across the valley nor is there visible horizontal or vertical displacement of feeder dikes. However, the valley still follows an unusually strong lineament. It also parallels a number of faults including the Lincton Mountain Faults (Maps 1, 3-4). The presence of a series of minor faults cannot be completely ruled out.

The Hite Fault System

North-south to northeast-southwest trending faults of the Hite Fault System (HFS) form the major through going faults in the study area. Detailed outcrop descriptions of studied faults are provided in Appendices E and F. The number of faults and lineaments of the HFS are fewer than those associated with the OWL (Figure 14 and Figure 15; Map 1). Apparent vertical displacements on faults of the HFS range from none to approximately 310 meters in a down to the west sense. Nearly all of these faults show evidence of strike-slip. Where slickensides are observed they are dominantly sub-horizontal (for example, locations 7, 10-11, 12, and 42 in Appendices E and F; Figure16). Where demonstrable, the sense of slip is left-lateral (Maps 2, 3, and 4). Steps compatible with left-lateral strike-slip are observed on slickenside surfaces at locations 5, 7, and 18 on Map 2. At location 12, down to the west apparent vertical offset and south plunging slickensides indicate a 2 to 9 meters left-lateral strike-slip component (Appendix E).

The western most fault of the Hite Fault System examined is the approximately 25 kilometer long Peterson Ridge Fault which follows a distinct north to northeast trending topographic lineament (Maps 1, 3, and 4). The most accessible out crop of the Peterson Ridge Fault occurs at an approximate elevation of 2300 feet in a flat spot on the ridge in the far western part of section 4 in township 4 north, range 37 east (location 32 on Maps 3 and 4. At this location, it is visible as a N16oE trending standing wall of well-lithified breccia without observed slickensides (Figure 29). A line of vegetation extends northward from the breccia wall. Flows were sampled to the east and west of the fault near this location. Correlating these flows to the defined stratigraphy indicates approximately 60 meters of apparent down to the west offset across the Peterson Ridge Fault. To the northwest in the southeast quarter of section 28 of township 3 north, range 37 east (near the northern limit of the mapped area on Map 3), Kendall (1981) reports finding both sub-horizontal and down dip slickensides. Also, based on differential apparent vertical displacement along the Peterson Ridge Fault between Blalock and Lincton Mountains and the strike and dip of flows, Kendall (1981) estimates that the observed displacements could be accounted for by 60 to 70 meters of down to the west dip-slip or 200 to 250 meters of left-lateral strike-slip.

The north to northeast trending Blalock Mountain Fault meets the South Fork Fault at the bottom of the valley in the southeast quarter of section 10 of township 4 north, range 37 east (location 35 on Map 4). It is not certain whether the Blalock Mountain Fault extends to the south of the South Fork Fault. The Blalock Mountain Fault does not crop out in the South Fork Walla Walla River valley, but it is recognizable from the 9 to 12 meters of down to the west vertical offset of basalt flows mapped across it on Blalock Mountain (Figure 30). A slight steepening of the dip of basalt flows immediately to the east of the Blalock Mountain Fault (location 36 on Map 4) is compatible with drag folding associated with down to the west displacement. While slickensides are not observed because of the lack of fault outcrop, the Blalock Mountain Fault is sub-parallel to other faults of the Hite Fault System which have a demonstrated component of left-lateral strike-slip. Both Swanson (1981) and Kendall (1981) interpret the Blalock Mountain Fault as having undergone at least a small degree of left-lateral strike-slip.

East of the Hite Fault, a N35^oE trending fault crosses the eastern South Fork stratigraphic section at about 3300 feet elevation near the center of the southern half of section 7 of township 4 north, range 38 east (location 42 on Map 4). It crops out as a 12 to 15 meter wide saddle which is bounded on both sides by highly fractured rock containing nearly horizontal slickensides (Figure10). There is no measurable vertical displacement across this Fault. To the south at about 3640 and 3740 feet elevation on this same ridge, two small faults with sub-horizontal slicks and insignificant vertical offset also crop out.

In the vicinity of the South Fork Walla Walla River, the Hite Fault has a 1 kilometer right step (mapped by Kendall, 1981) and appears to break into two distinct fault traces (Maps 1 and 4). Based on the dominantly northwest trending geometry of minor faults within the right step of the Hite Fault, Kendall (1981) concluded that strike-slip on the Hite Fault is left-lateral. The presence of this step also greatly limits the degree of syn- and post-CRBG strike-slip movement along the Hite Fault proper. To the north along Tiger Creek Road, horizontal slickensides with steps compatible with left-lateral displacement are observed near the center of an 80 to 90 meter wide zone of brecciated and highly fractured rock of the Hite Fault with multiple discrete sub-vertical slip surfaces (Figure32 and Figure33; location 18 on Map 2). The surface with steps does not parallel the N20^oE trend of the Hite Fault. Instead, it strikes N40^oW which suggests counter-clockwise rotation through an acute angle. This rotation is compatible with left-lateral strike-slip.

The magnitude, sense, and timing of vertical offset across the Hite Fault have been constrained by correlations between two stratigraphic sections located on opposite sides of the Hite Fault on Blalock Mountain (ridge east of location 36 on Map 4; Figure 1) and the first ridge to the west of Rimrock Ridge (location 42 on Map 4). There is some error in both the flow thickness and elevations which follow. The data is the best combined estimate from map locations and elevations, altimeter measurements, and eye-height thickness measurements. Thicknesses of individual flows are good to about one meter for the thinnest flows to three meters for the thickest flows. Error in elevations of flow contacts is probably less than half of the 40 foot (12 meters) contour interval. Variations in actual flow thickness of plus or minus up to one meter may be observed over distances of a few tens of meters. Strike-slip may juxtapose different parts of the same flow with different thicknesses.

Vertical displacement of the lowermost correlated flow (unit 10 in Table 1) is about 295 meters while the displacement of the stratigraphic level of the magnetic reversal is about 310 meters (Table 1, Appendix A, and Figure 34). The 15 meter difference in displacements corresponds with 15 meters of apparent thinning of this interval of the stratigraphy in the western section relative to the eastern section implying a small degree of down to the east displacement. The lack of a few reversed polarity flows on the east side of the Hite that are present on the west side and vice versa suggests that both limited down to the west and down to the east apparent vertical displacement might have occurred. Such apparent displacements would be compatible with a dominantly strike-slip sense of movement. At Tiger Creek, the offset of the magnetic reversal across the Hite Fault is similar to that observed farther south, about 300 meters measured between a point on the west side of the Hite Fault about 0.4 kilometers southwest of location 18 on Map 2 and location 19 on the east side.

Significant thinning of the Grande Ronde N2 part of the section is documented across the Hite Fault. The stratigraphic level of the R2-N2 magnetic reversal is about 310 meters higher near Rimrock Ridge on the east side of the Hite Fault relative to on Blalock Mountain on the west side while the top of the Grande Ronde is only about 250 meters higher to the east (Map 4; Figure 33). The differential offset requires approximately 60 meters of thinning of the Grande Ronde N2 part of the stratigraphy which represents a time period of 100,000 to 200,000 years (Baksi, 1989). This thinning is documented by the two stratigraphic sections on opposite sides of the Hite Fault. The Grande Ronde N2 part of the stratigraphy is about 290 meters and 230 meters thick to the west and east of the Hite Fault respectively, a difference of 60 meters. This appears to have occurred mostly through thinning of individual flows. The absence of some flows to the east may mark episodes of more rapid displacement. The absence of some normal polarity flows on the west side of the Hite Fault which are present on the east side might indicate limited episodes of down to the east offset instead of the dominant down to the west offset. Such displacements would be compatible with a dominantly strike-slip sense of movement during Grande Ronde N2 time.

Down to the west offset after the eruption of the Umatilla flow, the uppermost flow on the table lands on top of Blalock Mountain and at Table Glade (Map 4), may be estimated from the different elevations of the two table lands as there is only a very small dip between them. The elevation difference of the estimated top of the Umatilla flow is about 220 meters and suggests about 30 meters of thinning on the east side of the Hite Fault of the post-Grande Ronde portion of the stratigraphy which represents a time interval of 1.1 to 2.6 million years (Map 4; Figure 34; Baksi, 1989). This cannot be confirmed by the stratigraphy as the location of the contact between the Sentinel Gap and Umatilla portions of the stratigraphy is not exposed in the two stratigraphic sections being compared and because the location of the top of the Umatilla flow is approximate because of subsequent erosion. Because of displacements on the north to north-northwest trending horst and graben structures present on the southeast side of the Hite Fault, the offset of the top of the Umatilla flow between Blalock Mountain and Target Meadows is less, only about 100 meters higher to the east. The 220 meter (between Blalock Mountain and Table Glade) and 100 meter (between Blalock Mountain and Target Meadows) displacements of the top of the Umatilla flow occurred during the last 13 to 14 million years (Figure 5; Baksi, 1989).

Blue Mountains Uplift

The study area includes part of the western flank of the segment of the Blue Mountains uplift intimately associated with the Hite Fault System. Here, the Blue Mountains are characterized by generally shallowly northwest dipping basalt flows in the west (Map 3) and virtually horizontal flows in the east (Map 4) which are displaced in a down to the west sense by north-northeast trending faults of the Hite Fault System (Figure 34). On Lincton Mountain, Basket Mountain, and Blalock Mountain, for example, the topography tends to form a dissected dip slope (Maps 3 and 4; Figure 35). Where the flows are horizontal, in the Table Glade-Target Meadows area and on the upper portion of Blalock Mountain, the topography is a dissected tableland (Map 4; Figure 30).

Within the study area, the total uplift of the Blue Mountains by flexure and faulting is significant. Uplift occurred during the eruptive period of the Columbia River Basalt Group (CRBG) and continued after the eruption of the Umatilla flow, the youngest CRBG flow in the field area. At least one Grande Ronde N2 flow apparently thins and disappears from west to east. This flow, the lowermost flow of the Western Composite Section (unit 40 in Table 1), cannot be correlated with any flow in either of the two stratigraphic sections to the southeast. The Sand Hollow part of the section (Tfsh; unit 47) also thins to the east. At least three Sand Hollow flows are present in the western most part of the field area (Map 3; Western Composite Section of Table 1) while only one thin flow is present to the east on Blalock Mountain (location 37 on Map 4). Further east, no Sand Hollow type flows are present (Map 4; Figure 34). Uplift by faulting is present as a down to the west component of vertical movement on faults of the Hite Fault System. Systematic down to the west movement across the Hite Fault began at least as early as Grand Ronde N2 time. Displacement of the top of the Lookingglass flow, which is present near both the northwestern and southeastern limits of the mapped area, records much uplift by combined flexure and faulting. Older units are not exposed in the western part of the field area so combined uplift of the R2-N2 contact, for example, cannot be determined. In the eastern part of the field area near Table Glade (Map 4), the top of the Lookingglass flow is approximately 1100 meters higher than to the west where the flow top is exposed about one meter above the Walla Walla river (1.5 km west of location 22 on Map 3; Figure 34). Taking into account the thinning of N2 and later flows on the east side of the Hite Fault increases the total to at least 1200 meters.

Kendall (1981) suggested that the uplift of the Blue Mountains is related to a core of relatively low density crust similar to the Wallowa Mountains. Gravity work by Sobczyk (1994; see Sobczyk's Figure17) does reveal Bouguer gravity lows (relatively low density or thickened crust) under the segment of the Blue Mountains containing the study area.

Graben Related Faults

In the southeastern portion of the study area (Map 4; Figure 14) is a series of north to north-northwest trending faults which appear to be an extension of the La Grande Fault System which is associated with the La Grande Graben (Kienle and others, 1979; Kendall, 1981). Gehrels and others (1980) interpret the strike-slip component of movement on the La Grande Fault System as right-lateral. Included in this group of faults are the South Fork Faults, the Ninemile Fault, and the Bald Mountain Fault, and several unnamed faults (Map 4). Kendall (1981) also prefers to associate the Lincton Mountain Faults with the La Grande Fault System in spite of the north-northwest, OWL parallel, trend of the Lincton Mountain Faults. With the exception of the South Fork Faults, the La Grande Fault System appears to die out across the Hite Fault proper (Maps 1 and 4; Figure 14).

Swanson and others (1981) interpret the N15oW to N20oW trending South Fork faults (Map 4) as having right-lateral strike slip and down to the west vertical offset. Near a trail in the northwest quarter of the southeast quarter of section 10 of township 4 north, range 37 east (location 35 on Map 4), however, vertical offset of a distinctive oxidized flow top with sediment across the South Fork Fault is 2 to 3 meters in a down to the east sense. Kendall (1981) reports differential down to the east offset of the Grande Ronde R2-N2 magnetic reversal and of the most westerly of the low-angle dikes (north of location 35 on Map 4) from which he calculated that at least 55 meters of sub-horizontal right-lateral strike slip must have occurred. This calculation assumes that the dike is planar, which it is not, so the calculation must be considered a very crude estimate. To the east, vertical offsets across the graben faults are much greater. Kendall (1981) reports 97 meters of down to the west offset across the Ninemile Fault to the south of the study area. Offset between Table Glade and Target Meadows is about 120 meters down to the northeast distrubuted between several graben faults.

Shallowly Dipping Faults

Shallowly dipping faults are exposed in at least five locations in the study area. A stereo plot of planes and slickensides associated with some of these faults is provided in Figure 36. In the breccia zone associated with the Forks Fault (location 26 on Map 3) there is a series of shallowly east-dipping planes (Figure 27). Poorly developed north-south trending and down-dip trending grooves are present on some of these surfaces. At location 28 on Map 3, a N55^oE striking and 27oW dipping normal fault zone is present (Figure 37). Two antithetic fault planes are present at this location with strikes and dips of N52^oE, 54^oE and N42^oE, 37^oE. This normal fault accommodates the same down to the west offset observed on the vertical faults of the Hite Fault System. At location 29 a small fault, possibly another normal fault, striking N15^oE and dipping 40^oW is present.

On Lincton Mountain two faults are exposed in a road cut on the north side of Lincton Road in the southeast quarter of the southwest quarter of section 6 of township 4 north, range 37 east (location 30 on Maps 3 and 4). The westerly fault zone strikes to the northeast and dips shallowly to the southeast with approximately dip-slip slickensides. The easterly fault zone also strikes to the northeast but dips to the northwest with dip-slip slickensides (Figure 38). If this easterly fault is an incipient thrust fault, it would be compatible with the north-northwest regional compression direction with the westerly fault being an antithetic fault. Precise strikes and dips were not obtained because of limited outcrop of these faults.

Another low-angle fault is exposed just to the east of the Peterson Ridge Fault at an approximate elevation of 2200 feet in the western half of section 4 in township 4 north, range 37 east (location 32 on Maps and 4). It appears as a bedding plane fault with slickensides trending N5^oE to N20oE and plunging between 7^o and 15^o to the north. Kendall (1981) reports finding a low angle fault near the mouth of Flume Canyon in the southwest quarter of section 32 of township 5 north, range 37 east. These latter two faults may also be incipient thrust faults.

Wanapum Basalt Feeder Dikes

At least nine dikes of the Frenchman Springs Member of the Wanapum Basalt Formation are present in the study area. The specific chemical types are tentatively identified as Ginkgo, Sand Hollow, Sentinel Gap, and Palouse Falls (Figure 13). Corresponding flows were sampled for the Ginkgo, Sand Hollow, and Sentinel Gap types. No flows corresponding to the Palouse Falls type dikes were located. In the field, dikes are distinguishable from Grande Ronde flows by generally fresher appearance, horizontal columns in the case of vertical dikes and inclined columns in the case of low angle dikes, visible discordant margins, glassy chilled margins, and sometimes by the presence of large plagioclase phenocrysts. Chemical analyses of all observed dikes were used to identify their chemical type and confirm their distinction from the Grande Ronde flows which they cut (Figure 12). Only three of the sampled dikes are vertical. The remaining dikes dip shallowly to the west. Thicknesses of the shallowly dipping dikes vary from about 0.5 meters to greater than 15 meters.

The three vertical dikes include two north to N5^oE trending Palouse Falls dikes exposed along Mill Creek Road (Figure 39; labeled dikes on Map 2) and one N25^oW trending Sand Hollow dike exposed along the county road west of the forks of the Walla Walla River (location 27 on Map 3). These dikes fit the regional pattern in which CRBG dikes tend to be essentially vertical and trend north to north-northwest having opened at right angles to the horizontally directed regional maximum extensional strain direction (Hooper and Conrey, 1989).

At least six north-northwest striking and shallowly westward dipping dikes are exposed in the valley of the South Fork Walla Walla River (Figure 40, Figure 41, and Figure 42; Map 4). Four of these dikes were mapped previously by Swanson and others (1981). Sampling of the six dikes identifies them as Sentinel Gap and Ginkgo chemical types. The average strike and dip of the entire eastern most mapped Sentinel Gap type feeder dike (location 39 on Map 4; Figure 42) calculated from three points without intervening faults is N15^oW, 12^oW. In this area, the flows penetrated by the dike dip, at most, 1o to the northwest. Kienle and others (1979), Kendall (1981), and Swanson and others (1981) record similar west dipping dikes in the valley of the North Fork Walla Walla River which were not sampled in this study. The dikes exposed in the North Fork valley are nearly on strike with and may be extensions of those in the South Fork valley (Kendall, 1981). Swanson and others (1981) report additional dikes outside of the area of this study south of Milton-Freewater including two dikes which dip about 20^o to the east along Pine Creek in section 1 of township 3 north and range 35 east. The strikes of the low-angle dikes are similar to the regional strikes of vertical dikes, but their shallow dips are unusual for both the Columbia Plateau and for dikes in general (Spencer, 1985; Walker, 1986; Swanson and others, 1989; Hooper and Conrey, 1989; Figure 2 and Figure 43). The west-dipping low-angle dikes require an explanation which differs from that of the vertical dikes, and their grossly parallel orientation indicates a common control of their propagation trajectories (Map 4; Figure 43).

The formation of dikes has been modelled using fracture mechanics. In this method, dikes are viewed as fluid-filled mode-I (pure extension) fractures formed perpendicular to the minimum principal stress direction, s3, and inflated by internal magma pressure which produces tensional stresses at the fracture tip (for example, Delaney and others, 1981; Rodgers and Bird, 1987; Lister, 1991; Lister and Kerr, 1991). Evidence that s3 is usually horizontal is seen in the vertical plane of most dikes.

Average dips of the observed low-angle dikes become shallower as the dikes approach the earth's surface. From the three point method, the average strike and dip of the lower half of the easternmost Sand Hollow type dike is N18oW and 17oW while the upper half (location 39 on Map 4; Figure 42) strikes N20^oW and dips only 8^oW. Continuation of this trend implies that at depth, away from surface stresses, the dikes may steepen to near vertical and have a more typical orientation. Strikes and dips of dikes are plotted in Figure 43. The constant average strike suggests that s1 remained horizontal and was oriented approximately N20^oW. Rotation of s3 about s1 from horizontal at depth sub-horizontal near the surface could explain the change changing dip of the dikes. Assuming that the dikes propogated at purely mode-I fractures implies that the dike propagated perpendicular to s3 following a curved path. A changing s3 orientation would also cause a shear stress to be resolved onto the dike plane producing mixed-mode fracturing (Pollard, 1987). Mixed mode fracturing would cause dikes to propagate along a curved path (Figure 44). Perhaps stresses resulting from transition between principally strike-slip and principally dip-slip movement could rotate s3 producing low-angle dikes. Stress resulting from the observed anticlinal flexure would produce steeply east-dipping dikes not shallowly west-dipping dikes.

Free surface interaction is also capable of producing non-vertical dikes (Pollard, 1987). As magma rises toward the earth's surface, it may reach a point where its density is equivalent to that if its host rocks called the level of neutral buoyancy. At this point, the magma will tend to spread laterally rather than ascend, possibly forming sills (for example, Pollard and Holzhausen, 1979; Lister and Kerr, 1991). Near the free surface marked by the surface of the earth, non-vertical fractures such as sills will tend to propagate out of their original plane and curve toward the free surface producing inclined dikes or "climbing sills" (Pollard and Holzhausen, 1979). Inclined dikes produced according the this model would tend to steepen as they approach the free surface. The dikes in the field area, however, are observed to dip less steeply as they approach the present day surface.

Another potential explanation for the low-angle dikes is that they are influenced by pre-existing shallowly west-dipping faults. Two such faults have been located in the study area at locations 28 and 29 on Map 3, but they strike to the northeast rather than to the northwest (Figure 36). Slickensides have been located along the margin of one low-angle dike indicating some association between the dikes and faults (Figure 40).

The low-angle dikes mapped in this study are not planar. They often flatten out between flows and steeply ramp upward through them. At other locations, they cut with a shallow dip through the nearly horizontal flows (Figure 40 and Figure 42). Local strikes vary between northeast and northwest and dips vary between 9^o and 40^o to the northwest and southwest. Anisotropy in the host medium could perturb dike propagation producing localized variations in dike orientation which deviate from being perpendicular to s3. Such anisotropy is present in the basalts in the form of horizontal planes, contacte between flows, and vertical fractures, cooling joints, in the basalts. Sub-horizontal and sub-vertical dike segments may be explained by perturbation of the propagating dike tip by the horizontal planes of weakness and abundant vertical fractures.

Interaction between the major fault systems

Three major fault systems which include faults associated with the Olympic-Wallowa-Lineament (OWL), Hite Fault System, and La Grande Fault System intersect in the study area (Map 1; Figure 14). The interaction between the many associated faults has implications for the relationship between the larger fault zones. A cluster of graben related faults of the La Grande Fault System including the Ninemile Fault largely terminates against the Hite Fault in the southeast corner of the study area (Maps 1 and 4; Figure 14). The only graben related faults present to west of the Hite Fault are the South Fork Faults. Since the main Hite Fault splays and terminates a few kilometers to the south of the study area the termination of graben faults against the Hite Fault implies a small component of northeast increasing left-lateral strike-slip on the Hite Fault (Figure 14). Limitation of the amount of slip imposed by the right step of the Hite Fault (location 38 on Map 4) implies that most of the strike-slip developed occurs to the northeast of the step.

The South Fork Fault is apparently offset in a left-lateral sense from the cluster of graben faults on the east side of the Hite Fault (Map 4). This apparent offset might be the result of strike-slip along the Hite Fault. If strike-slip is the case, it would be unreasonable to expect more post exposed CRBG displacement than the approximately 3.5 kilometers of apparent offset between South Fork and Ninemile Faults. Because of the limitation imposed by the right-step of the Hite Fault seen in the exposed basalts, any greater displacement (such as that which would be implied by correlating graben faults to the northeast of the Ninemile Fault with the South Fork Fault) would require movement prior to the eruption of exposed CRBG flows. The much greater vertical offset across the Ninemile Fault (97 meters down to the west; Kendall, 1981) and nearby graben faults than on the South Fork Fault (2 meters down to the east at location 35 on Map 4) suggests that any strike-slip displacement of a formerly continuous South Fork-Ninemile fault would have taken place before significant vertical displacement occurred on the graben faults. An alternate hypothesis which does not require but does permit strike-slip is that the South Fork Fault formed independently from the Ninemile Fault and other graben faults.

It has been suggested previously by a number of authors (for example, Sandness and others, 1982; Hooper and Conrey, 1989) that the OWL-zone and Hite Fault System represent conjugate megashears. The overall pattern of faults and lineaments in the area (Map 1; Figures 14 and 15) does give this impression (Sandness and others, 1982). Such a conjugate relationship with right-lateral strike-slip on the west-northwest trending OWL and left-lateral strike-slip on the north-northeast trending Hite Fault System is compatible with the regional strain pattern. This pattern is one of north-northwest directed compression and east-northeast directed extension deduced from east-west trending anticlines and synclines of the Yakima fold belt and Lewiston structure and north to northwest trending CRBG dikes (Figure 2 and Figure 4).

If a conjugate relationship is the case, one would expect a complicated pattern of faulting in which faults of one set displace faults of the other set and vice versa. Apparently offset faults and lineaments are present in the study area. At location 28 on Map 3, an outcrop scale right-lateral offset of an approximately north-south trending fault by an east-west trending fault is visible (Figure 45). At locations 15 on Map 2 and locations 34 and 38 on Map 4 OWL parallel faults and lineaments are offset in a left-lateral sense across faults of the Hite Fault System. On the trend of the Lincton Mountain Faults, there is the 1 kilometer right step of the Hite Fault (location 38 on Map 4). Here a pre-exposed basalt displacement could have occurred with post-basalt reactivation and later faulting producing the Lincton Mountain Faults and undisplaced Peterson Ridge Fault. On a much larger scale, two groups of faults of the Hite Fault System, one including the Peterson Ridge Fault and Hite Fault and the other including the Saddle Hollow Fault, appear to be offset 10 to 15 kilometers in a right lateral sense along a projection of the Wallula Fault zone (Figure14). Graben extension may have contributed to this apparent displacement. If this offset is the result of right-lateral strike-slip, the proposed cratonic margin of Reidel and others (1994) would imply that cratonic crust displaced southward along the Hite Fault System is also displaced in a right-lateral sense along the OWL-zone.

In a conjugate system of faults, one might expect to find evidence of limited simultaneous movement at fault intersections. Possible evidence for limited simultaneous movement on faults of the two conjugate systems is found along Tiger Creek Road (location 2 on Map 2). Here there is a crush zone at the intersection of a N75^oW trending fault zone and less well developed north-south fault. In the crush zone there are abundant small planes with slickensides with trends intermediate between those of the two intersecting faults.

One must be cautious when interpreting apparent offsets as displacement of one fault by another. Studies of well exposed strike-slip faults in the Entrada Sandstone of Arches National Park, Utah (Zhao and Johnson, 1991), and experimental studies with clay (for example, Oertel, 1965; Reches, 1988) reveal that while faults often do displace one another, faults with apparent offset may also form independently. One needs to determine the true age relationships between faults to be certain that displacement has occurred. Unfortunately, in the field area of this study, this is not usually possible. Reches (1988) and Zhao and Johnson (1991) also note that in conjugate systems, segments of left lateral faults tend to step to the right while segments of right-lateral faults tend to step to the left producing potential apparent displacements. One conjugate step without a crossing fault of the opposing system is observed in the Lincton Mountain Faults (Map 4).

Discussion

The lack of significant, continuous west to northwest trending faults in the study area and to the southeast makes any argument for large amounts (tens of kilometers) of post-CRBG strike-slip along the structural zone coincident with the OWL difficult to sustain. A few meters to a few tens of meters of displacement on a number of small faults may add up to a few kilometers of total displacement but not to tens of kilometers. Faults which could have produced the apparent right-lateral offset of the Hite Fault System across an extension of the Wallula Fault zone have not been found at the surface. Any displacement, if the offset is the result of displacement, must be older than the CRBG. Similarly, it is unlikely that the apparent 20 to 30 kilometers of right-lateral offset of the Cornucopia and Grande Ronde dike swarms could be the result of post CRBG displacement (Figure 2). Significant movement could have occurred prior to the eruption of the CRBG, and the two dike swarms could represent older displacement of a pre-existing zone of crustal weakness which was intruded by the feeder dikes. Alternately, the dike swarms could have formed in an en echelon pattern which would reflect the same pattern of strain. The presence of an older OWL zone would also explain the alignment of many small structures for 100's of kilometers along a consistent trend across the region (Figure 2). The difficulty in tracing continuous structures coincident with the OWL to the east of the Blue Mountains is probably the result of attenuation of the associated fault zone (OWL-zone) with displacement shifting to the south, perhaps to the Vale Zone producing a large extensional step in which the La Grande Graben, Baker graben, and other extensional valleys occur. The trend of the OWL leads into the Idaho Batholith which may be more resistant to strike-slip movement than the crust underlying the Snake River Plain which the southeast end of the Vale Zone penetrates. This may force such a shift in displacement from the OWL-zone to the Vale Zone. Many faults of the Wallula Fault Zone appear to trend toward the grabens and may be involved in extension and transfer of displacement (Figure 2). The name Wallula-Vale Transfer Zone is suggested for this transfer system which includes the OWL-zone, Vale Zone, and intervening extensional faults (Figure 46). The presence of extensional and graben forming faults on the south side of the OWL-zone without equivalent extensional structures to on the north side and the termination of the OWL-zone against the Idaho Batholith to the east require some right-lateral strike-slip along the OWL-zone (Figure 4 and Figure 46).

Reidel and others (1994) state that the rate of deformation apparent in ridge growth in the Yakima Fold Belt and subsidence in the Columbia River basin has decreased since the middle-Miocene when the bulk of the CRBG was erupted. A similar decrease in the deformation rate throughout the region is compatible with the OWL-zone and HFS being largely pre-basalt structures with some reactivated faulting penetrating the basalts.

Return to . . . Chapter 2

Continue to . . . Chapter 4

Table of Contents