The sedimentary succession cored at Site 1108 penetrated about half of the local thickness of the hanging-wall sequence of a detachment fault adjacent to the Moresby Seamount, as inferred from seismic data (see "Leg 180 Summary" chapter). Structural observations and measurements made at Site 1108 are summarized in Figure F21. In the following text, the sections' depths (in mbsf) refer to their respective tops.
The 485-m-thick section of Holocene-middle Pliocene sediments drilled at Site 1108 (see "Biostratigraphy") can be subdivided into five distinct structural domains on the basis of marked variations in both lithologies and types of deformation.
Domain I is represented by a thin sequence of clay-bearing ooze and recovered pebbles (lithostratigraphic Units I and II; see "Lithostratigraphic Unit I" and "Lithostratigraphic Unit II"), extending from the seafloor to the top of Core 180-1108B-8R (62.7 mbsf). Recovery was very low; therefore, the primary or reworked origin of the pebbles recovered cannot be firmly elucidated.
Domain II consists mainly of unconsolidated sand (lithostratigraphic Unit III) overlying a more lithified sandstone sequence (lithostratigraphic Subunit IVA) that shows some evidence for brittle deformation, including fractures and micro-normal faults. This domain extends from the top of Core 180-1108B-8R down to the top of Core 18R (158.6 mbsf).
Domain III is an interval marked by considerable faulting, extending from Core 180-1108B-18R to the top of Core 22R (197.2 mbsf). Within this interval of semilithified claystones to sandstones (lithostratigraphic Subunit IVB) stress led predominantly to brittle failure.
Domain IV comprises little-deformed turbiditic sandstones (lithostratigraphic Subunit IVA) that are present between the top of Core 180-1108B-22R and the top of Core 38R (350.5 mbsf). Brittle deformation is expressed only by a few minor normal faults.
Domain V encompasses the lower part of lithostratigraphic Unit IV. It is observed from Core 180-1108B-38R to the base of Hole 1108B (i.e., 485 mbsf). The mainly fine grained silty material in this interval shows intense deformation, including scaly fabrics, fracturing, and shear along fault planes.
The majority of the structures observed at Site 1108 are thought to be real tectonic features, although drilling-induced fracturing cannot be ruled out for some. In fact, the RCB drilling process was partly hindered by the well-cemented rock in that the core rotated in the barrel for a considerable length of time. This problem has been pointed out previously (Lundberg and Moore, 1986). The causes of subhorizontal fragmentation is thus unclear. Nevertheless, most of the brecciation and normal faulting cannot be explained as an artifact of drilling; therefore, the features are inferred to relate to tectonics.
In the following section, the deformational structures of the above-mentioned domains are described mainly in terms of fault/fracture frequency, with peculiar attention to the fault zone of structural Domain III.
Within the undeformed, soft sediments at the top of the succession (Domain I; see Fig. F21), no structures other than horizontal bedding planes are observed.
Beneath Domain I and throughout Domain II, the bedding dips vary considerably, ranging from 0º to 75º with an average attitude in the interval 10º-20º (Fig. F22). Some of these bedding planes have been reactivated as slip planes (e.g., Sections 180-1108B-8R-1 [62.7 mbsf] and 9R-1 [72.3 mbsf]). Also, faults intersecting bedding planes at high angles are in evidence, as illustrated in Figure F23 by a steeply inclined dip-slip fault in Section 180-1108B-10R-1 (81.9 mbsf). At Section 180-1108B-17R-1 (149.0 mbsf), the observation of staircase-like morphologies along a system of minor steeply dipping slip planes indicates reverse faulting. Brittle deformation is also illustrated by intense fragmentation of intervals composed of centimeter-sized angular blocks of siltstone (Sections 180-1108B-9R-1, and 10R-1).
The top of Domain III is defined by a sharp change in both lithology and deformational style, as mirrored by intensely fractured clays and claystones. However, the exact nature of the contact between Domains II and III remains uncertain because of poor core recovery. The abrupt increase in fracturing within Domain III is clearly shown on the frequency distribution curve in Figure F21. Domain III typically shows alternations of extensive intervals of unbroken material, up to 1 m long (Sections 180-1108B-18R-1 [158.6 mbsf] and 18R-4 [163.1 mbsf]), and highly strained zones. Within the unbroken intervals, bedding is often preserved and, where measured, is inclined ~10º on average (Fig. F22).
Coherent pieces of silty material are consistently cut by various sets of planar-type fractures, including both open fractures with no indication of movement, and minor fault surfaces marked by slickenlines. The entire fracture population identified throughout Domain III dips at moderate angles, clustered around 45º (Fig. F24). However, both shallow- and high-angle normal faults (Section 180-1108B-20R-1 [178.0 mbsf]) are also well documented. Analysis of the plunge of the slickenline data set indicates a dominantly dip-slip direction of displacement (~75% of the data collected), but some oblique-slip and pure strike-slip faults are also documented (Fig. F25). Very few indications of sense of movement have been recorded, and most of the few apparent millimeter-scale offsets directly measured on the core face indicate normal sense of displacement (Section 180-1108B-19R-1 [168.3 mbsf]).
The general arrangement of the deformed zones is dominated by brecciated zones that are largely developed throughout Domain III. They comprise centimeter to decimeter-sized angular fragments of fine-grained siltstone and clay with polished surfaces, lying in a random orientation on the observed core surfaces.
A gradual vertical transition from coherent, or slightly fractured, core pieces to progressively broken intervals was observed at Sections 180-1108B-18R-1 (158.6 mbsf), 18R-3 (161.6 mbsf), and 19R-4 (171.3 mbsf); these, in turn, pass into individual rock fragments whose size diminishes gradually <1 cm (Fig. F26). Because no significant variation in lithology is observed, the brecciation process can be assigned to disaggregation of an original coherent material in response to increasing brittle fracturing. As stated before, the inferred primary fault-related brecciation may have been later emphasized during drilling operations.
The boundary between Domains III and IV also corresponds to a well-marked lithologic and deformational limit. In Domain IV, the coarser grained and more competent turbiditic sandstones of lithostratigraphic Subunit IVA (Fig. F21) exhibit less intense fracturing and dip at shallower angles (<10º) (Fig. F22). As a result, Cores 180-1108B-22R through 37R (197.2 to 350.5 mbsf) show intervals of mostly coherent and intact sandstones as long as 50 cm. Along this zone the only significant deformation consists of a system of horizontal joints that are likely to represent drilling-induced disturbances, probably reactivating primary sedimentary anisotropies.
From Section 180-1108B-36R-1 (331.3 mbsf) to the top of Section 38R-1 (350.5 mbsf), the basal part of Domain IV is characterized by an increase in minor brittle fracture deformation. In some places, the core surface splits into coherent pieces, a few decimeters in length, along striated vertical fault planes displaying either dip-slip or strike-slip displacements (Sections 180-1108B-36R-2 [332.3 mbsf], 37R-2 [342.1 mbsf], and 37R-3 [343.5 mbsf]). The limit between Domains IV and V is placed on top of Section 180-1108B-38R-1 (350.5 mbsf), below which we note a significant change in the density of the brittle structures, as well as a pronounced lithologic variation, passing downward from coarse-grained sandstones into fine-grained silt/clay material (see "Lithostratigraphy").
Domain V can be subdivided into three structural subdomains according to lithologic and tectonic criteria.
Subdomain Va, restricted to Section 180-1108B-38R-1, corresponds to a highly fragmented zone that contains randomly arranged decimeter- to centimeter-sized angular blocks of fine-grained silt/clay material. This brecciated zone comprises two decimeter-thick horizontal bands showing a strongly preferred orientation of fissility or fracture planes. These pervasive flat-lying surfaces enclose centimeter- to millimeter-thick pieces of clay with polished and highly striated surfaces. This fabric is similar to scaly fabrics that might result from a horizontal layer-parallel shearing, coeval with brittle normal faulting.
Subdomain Vb extends from Section 180-1108B-39R-1 (360.1 mbsf) to Section 44R-1 (408.2 mbsf) and comprises horizontally bedded, dominantly coarse-grained sandstones exhibiting little tectonic deformation. A few examples of normal faults were observed, such as a minor steeply dipping fault set, as illustrated in Figure F27 (Section 180-1108B-44R-1). In fact, the most prominent structures of Subdomain Vb correspond to curve-shaped fractures, largely developed at Sections 180-1108B-39R-1, 40R-1 and 40R-2 (370.7 mbsf), 42R-4 (391.0 mbsf), and 44R-1 (408.2 mbsf); these are likely to be drilling induced.
Subdomain Vc extends from Core 180-1108B-45R (417.8 mbsf) to the base of the cored section. It is composed of coarse-grained sandstones, locally highly fragmented, with an average dip of ~15º (Fig. F22). The nature of the transition from this inclined sequence to the almost horizontal overlying series is not clear, but it may represent a progressively fanning dip pattern, or an abrupt fault-induced feature.
We observed only a few examples (~10) of filled vein/fracture structures cutting through the section drilled at Site 1108. These structures are expressed by either millimeter-thick open fractures or normal and strike-slip fault planes commonly filled with carbonates in association with epidote. Figure F21 shows that most of the filled vein/fracture pattern occurs in the basal half of the unfaulted Domain IV. So far, no particular spatial or time relationships have been observed between the different vein/fracture-filled systems.
The 485-m-long vertical section drilled at Site 1108 penetrated two structural domains that exhibit typical characteristics of fault zones, namely the Domains III and V. According to the kinematics of most of the core-scale structures observed within these two fault zones, they are both confidently interpreted as dominantly extensional. A schematic structural diagram of the entire drilled section is shown in Figure F28A . Because of poor core recovery along the faulted contacts, the exact nature and geometry of these fault zones remain conjectural.
Domain III forms the shallowest fault zone (labeled FZ1), is ~40 m thick, and is developed throughout a fine-grained, clay-rich silty sequence separating two thick packages of less deformed and more competent sandstones. At a first approximation, its specific location with respect to the general lithostratigraphy illustrates that, as usually stated for any given deformed rocks, grain size represents an important factor in the distribution of brittle strain, the finer grained material being the locus of major deformation. Postcruise physical properties measurements should help to determine the extent to which additional factors such as the porosity (function of the ratio clast/matrix) have played a key role in the deformation process. Within the fault zone itself, deformation is mainly accommodated by brittle failure along discrete moderately dipping normal faults. The close association of this extensive fault/fracture network with widely distributed brecciated facies, of supposed tectonic origin, suggests a heterogeneous vertical distribution of brittle strain toward narrow, higher strained bands.
Quite similar conclusions can be reached about the second fault zone (FZ2), the base of which should occur below the end of the drilled section (Fig. F28A). First, its upper limit is also seen to put into contact the little-deformed sandstones series of its hanging wall with highly deformed clayey silts lying in the footwall. There, the occurrence of two thin flat-lying bands with more pervasive "scaly" fabric that pass vertically into highly brecciated levels, and then into fragmented material, leads us to suggest a model of brittle deformation involving a concentration of strain along discrete zones, probably reactivating primary sedimentary planar anisotropies. The development of these flat-lying deformed bands within the FZ2 extensional fault zone remains to be explained.
For both FZ1 and FZ2, the top of the fault zone coincides with a pronounced change in some physical properties measurements, as indicated, for example, by a decrease of the bulk density (see "Physical Properties"). On the other hand, the importance of FZ1 is emphasized (1) by the anomalous trend of the depth/porosity curve, indicating that part of the upper stratigraphic units at Hole 1108B has probably been removed by both erosion and faulting (see "Physical Properties"); (2) by an inflection of the in situ temperature curve in the depth interval 100-164 mbsf (see "In Situ Temperature Measurements"); and (3) by a 0.6-Ma age offset in the magnetostratigraphy between Cores 180-1108B-18R and 19R (see "Magnetostratigraphy"), which have the greatest number of fractures and microfaults (Fig. F21).
On the interpretative model of Figure F28B, the highest dips measured in the hanging wall of FZ1 are tentatively interpreted in terms of tilt-related rotations, although the mutual relationships between the boundary fault plane and the hanging-wall inclined strata are difficult to determine because of poor recovery.
Interpretation of paleomagnetic measurements might supply reliable constraints for reorienting the core structures to geographical coordinates. These constraints could more specifically determine the synthetic or antithetic position of the faults identified at Site 1108 with respect to the Moresby Seamount detachment system (Fig. F28C).