U.S. DEPAR1MENT OF THE INTERIOR MISCELLANEOUS INVESTIGATIONS SERIES U.S. GEOLOOICAL SURVEY MAP I-2199 PAMPHLET SURFICIAL GEOLOGIC MAP OF THE WEBER SEGMENT, WASATCH FAULT ZONE, WEBER AND DAVIS COUNTIES, UTAH By Alan R. Nelson and Stephen F. Personius Overview Schwartz and Coppersmith (1984), identified six dis­ crete fault segments, but recent work by a consortium of investigators has identified 10 or 11 segments (Ma­ This map, the fourth in a series of 1:50,000-scale chette and others, 1987, 1989, in press; Wheeler and maps of the Wasatch fault zone, depicts the surficial Krystinik, 1988). In northern Utah, the Ogden and deposits and faults along the Weber segment and part Collinston segments of Schwartz and Coppersmith of the adjacent Salt Lake City segment of the Wasatch (1984) have been separated into three new segments fault zone in northern Utah (fig. 1). This is the first map (fig. 1); northward from Salt Lake City, the Weber, that is sufficiently detailed to be useful for interpreting Brigham City, and Collinston segments have been de­ the paleoseismic history of this part of the Wasatch fault lineated (Personius, 1986, 1988a; Machette and others, zone. The map is also a guide to the best sites for future 1987, 1989, 1992). Nelson and Personius (1987) detailed studies of paleoearthquake magnitude and suggested that the Pleasant View salient at the northern recurrence on the segment. These types of geologic end of the Weber segment is a non-conservative barrier, studies will form a quantitative basis for earthquake and Bruhn and others (1987, p. 345) described the Salt hazard assessment along the Wasatch Front. Lake salient at the southern end of the segment as a One of the main objectives of our mapping of the similar barrier. Weber segment was to determine if the distribution of Our mapping of deposits and young fault scarps deposits and fault scarps between North Ogden and shows that the Weber segment is perhaps the most Bountiful indicated that this part of the fault has geologically distinct segment of the Wasatch fault zone. behaved as a single fault segment during the latest Scarp heights in deposits of all ages decrease markedly Pleistocene and Holocene, or whether it consisted of near the segment boundaries and remnants of pre­ two or more segments. Most major normal and strike­ Bonneville-lake-cycle fans are selectively preserved in slip fault zones are thought to be composed of several these areas (fig. 2). Gaps in Holocene faulting and seismically independent segments (Zoback, 1983; significant differences in Holocene slip rates between Schwartz and Coppersmith, 1986; Doser, 1989; Then­ the ends of the Weber segment and the adjacent hans and Barnhard, 1989). Segments are often bounded Brigham City and Salt Lake City segments further by "non-conservative barriers" to fault-rupture propa­ support the mapped boundaries. Thus, scarp distribu­ gation (Crone and others, 1987; Fonseca, 1988; dePolo tion and faulted-surface offset data (fault throw) for the and others, 1989; Thatcher and Bonilla, 1989). Non­ latest Pleistocene and Holocene show that the mapped conservative barriers are regions along a fault where the northern and southern boundaries of the segment are orientation of the slip vector changes between adjacent clearly zones where surface ruptures have tended to parts of a fault; these regions have been observed to change direction and die out. Finally, the lower struc­ mark the location of the initiation and termination of tural relief at the Pleasant View and Salt Lake bedrock fault ruptures (King and Nabelek, 1985). The concept salients, first recognized by Gilbert (1928), shows that of fault segmentation is important to paleoseismic these segment boundaries have probably persisted analysis of active fault zones because during a major throughout the Quaternary (Wheeler and Krystinik, earthquake, surface faulting is commonly restricted to 1988). one or two segments of the fault and the length of The Weber segment is also one of the longest ( 61 surface rupture is roughly proportional to the size of the km) and most active segments of the Wasatch fault earthquake. zone. Although all large Pleistocene and Holocene Initial work on segmentation of the Wasatch fault earthquakes probably did not rupture the entire Weber zone, summarized in Swan and others (1980) and segment, empirical fault rupture length-magnitude and 1 fault displacement-magnitude relationships (Bonilla erosion of fault-scarp crests, landsliding, and lateral and others, 1984; Machette and others, 1989) suggest spreading have produced a landscape in which sites with that the magnitude of earthquakes that did rupture the simple scarps that represent total vertical slip across the entire length of the Weber segment may have been as fault zone are rare. However, many sites have scarps large as M 7¥2. The most recent (probable) event on the that yield surface offsets that can be determined to be segment may have occurred only 500 years ago; average maximum or minimum values; such sites help constrain event recurrence is about 1,200 years, but recurrence vertical slip values for uppermost Pleistocene and Holo­ intervals may vary from 300 to 2,200 years. Although cene deposits along much of the fault zone. another large-magnitude earthquake may not rupture As noted by Gilbert (1928), scarps of differing the Weber segment for many hundreds of years, the heights in deposits of different ages clearly indicate vertical component of slip on the fault during the most many surface-rupture events on the Weber segment recent event may have been smaller (probably < 0.6 m) since the fall of take level from the Bonneville shoreline. than the slip typical of most events (1-3m), suggesting In a few areas, systematic differences in scarps heights that the recurrence of large earthquakes may be mark­ on upper Holocene fans show that there have been at edly nonuniform; there may be sufficient pres.ent stress least three faulting events during the late Holocene. on the segment to produce a large event in the near Recent studies of trenches and exposures across the future. Detailed site-specific trenching investigations at fault at three sites along the segment confirm that there several more sites on the segment are needed to better have been at least three and probably four events on at define Holocene recurrence and the potential hazard of least the northern half of the segment in the last 4,000 a M 6Y:z-7lh event. years (Nelson and others, 1987; Machette and others, The pattern of fault scarps along the Weber segment 1992). Typical vertical displacements for these events is highly variable, but we can make some generaliza­ (1.5-2.5 m) and surface offsets (0.5-1.5 m) calculated tions about the character of the scarps. In some areas it from scarps on the youngest fans (probably < 2 ka) is difficult to distinguish secondary fault traces from indicate that most scarps are the product of two or more landslide scarps, lateral-spread headscarps, fluvial events. Total post-Bonneville offset along much of the scarps, and lacustrine shorelines, but the main trace of Weber segment is about 25-28 m. Thus, there have the fault is usually obvious. The complexity of scarp probably been at least 10 and perhaps 15 surface patterns is partly a reflection of the age of the deposits faulting events in the past 15,000 years. on which the scarps are formed: more scarps tend to be Slip rates show the same changes along the segment preserved in older deposits or where thin young depos­ as the surface-offset values from which they are derived its are draped over preexisting scarps. Multiple parallel (fig. 2). The uncertainties in the ages of Holocene traces of the fault are usually best developed where the deposits along the fault far exceed the uncertainties in trend of the main scarp changes abruptly, such as at the determining offsets; uppermost Pleistocene deposits mouths of larger canyons and at some small reent~ants are relatively well dated at 13-16 ka. Thus, even if we ( embayments) in the mountain front. Many of the were able to greatly improve the accuracy of our offset reentrants probably mark the location of preexisting values, the errors of our Holocene slip rates would structures that locally control the main trace of the fault remain large. Although post-latest Pleistocene slip rates (Gilbert, 1928). Over larger areas, trends of small are fairly uniform in the center of the segment, they may scarps become more diverse and the mountain front decrease more rapidly to the south than to the north. bends abruptly westward near the Pleasant View and There may be significant changes in middle to late Salt Lake salients, which bound the segment on its Holocene rates between the northern and southern north and south ends. halves of the segment that we cannot resolve because of Scarp heights and surface offsets across the scarps dating uncertainties and the scarcity of Holocene de­ are also highly variable along the segment, but they posits that cross the fault in the south. Similarly, there unquestionably decrease by at least a factor of two for are suggestions that late Holocene slip rates north of all ages of deposits near both ends of the segment (fig. Kaysville are greater than early and middle Holocene 2). The catastrophic fall of the level of Lake Bonneville slip rates, but our present data are insufficient to confirm from the Bonneville shoreline, the rapid fall in lake level this preliminary interpretation. from the Provo shoreline, the poorly consolidated de­ Most other maps of this part of the Wasatch fault posits exposed along the mountain front, and the rapid zone are reconnaissance maps showing a generalized relative rates of uplift of the footwall of the fault have main trace of the fault (Marsell, 1964; Morisawa, 1972; resulted in a highly active geologic environment along Van Horn, 1975; Miller, 1980; Davis, 1983, 1985; the fault zone.