4
Applying the Indicators to Example Fans

Not all alluvial fans, or those geologic features that are commonly believed to be alluvial fans, are subject to alluvial fan flooding. To show how dramatically such sites can vary, the committee selected seven sites for in-depth analysis and applied the indicators presented in Chapter 3. The sites represent a wide range of flood processes, from unconfined water flooding and debris flows on untrenched active fans to confined water flooding in fully trenched inactive alluvial fans. Six alluvial fans in the western United States are used to illustrate different flood processes, and a group of fans in Virginia illustrate a particular type of flood hazard in the eastern United States (Figure 4-1). By applying the indicators to each of the example sites, the committee was able to see whether or not the site meets the criteria suggested earlier in the proposed definition of an alluvial fan. In addition, insights are gained about how the definition and the indicators function in the field, and the advantages and disadvantages to those who ultimately will have to apply the guidance in a regulatory context.

Each of the examples also represents a different amount of study (Table 4-1). The Arizona examples show the problems faced in major urbanizing areas where there is intense interest and resources to support detailed investigation. The California examples represent a modest amount of study that included a brief field reconnaissance of each fan and compilation of geologic, topographic, and soil maps and aerial photographs. A similar approach was used to characterize the Utah fan, including examination of many technical reports produced following unusual flooding of 1983 and 1984. An exhaustive study of technical literature was the basis to typify the Virginia fans, which illustrate that alluvial fan flooding is not strictly a western phenomenon. Four of the sites were inspected by the committee, and the other three sites were inspected by at least one committee member. However, the committee wants to emphasize that it did not conduct a thorough field investigation of any site, as would be required for regulatory purposes, and thus these examples are purely illustrative and not intended to influence decision making on these fans.

HENDERSON CANYON, CALIFORNIA

The Henderson Canyon alluvial fan, which is located in eastern San Diego County near Borrego Springs, California, is below a drainage basin of approximately 16.6 km2 (6.40 mi2) that



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 83
4 Applying the Indicators to Example Fans Not all alluvial fans, or those geologic features that are commonly believed to be alluvial fans, are subject to alluvial fan flooding. To show how dramatically such sites can vary, the committee selected seven sites for in-depth analysis and applied the indicators presented in Chapter 3. The sites represent a wide range of flood processes, from unconfined water flooding and debris flows on untrenched active fans to confined water flooding in fully trenched inactive alluvial fans. Six alluvial fans in the western United States are used to illustrate different flood processes, and a group of fans in Virginia illustrate a particular type of flood hazard in the eastern United States (Figure 4-1). By applying the indicators to each of the example sites, the committee was able to see whether or not the site meets the criteria suggested earlier in the proposed definition of an alluvial fan. In addition, insights are gained about how the definition and the indicators function in the field, and the advantages and disadvantages to those who ultimately will have to apply the guidance in a regulatory context. Each of the examples also represents a different amount of study (Table 4-1). The Arizona examples show the problems faced in major urbanizing areas where there is intense interest and resources to support detailed investigation. The California examples represent a modest amount of study that included a brief field reconnaissance of each fan and compilation of geologic, topographic, and soil maps and aerial photographs. A similar approach was used to characterize the Utah fan, including examination of many technical reports produced following unusual flooding of 1983 and 1984. An exhaustive study of technical literature was the basis to typify the Virginia fans, which illustrate that alluvial fan flooding is not strictly a western phenomenon. Four of the sites were inspected by the committee, and the other three sites were inspected by at least one committee member. However, the committee wants to emphasize that it did not conduct a thorough field investigation of any site, as would be required for regulatory purposes, and thus these examples are purely illustrative and not intended to influence decision making on these fans. HENDERSON CANYON, CALIFORNIA The Henderson Canyon alluvial fan, which is located in eastern San Diego County near Borrego Springs, California, is below a drainage basin of approximately 16.6 km2 (6.40 mi2) that

OCR for page 83
FIGURE 4-1 Six fans in the western United States are used to illustrate different flood processes, and a group of fans in Virginia illustrate a particular type of flood hazard in the eastern United States.

OCR for page 83
Table 4-1 Amount of Study, Sedimentation, Major Flood Processes, Flow Path Movement, and Relevant Comments for Example Fans (sites are listed in order from the West Coast of the United States) Site Amount of Study Sedimentation Major Process Subject to Alluvial Fan Flooding Comments Henderson Canyon, California Modest Inactive and active Water flood Yes Flooding confined to large trenches on relict fan. Sheetflooding on active fan. The use of maps, aerial photographs, soil surveys, and field reconnaissance is described in the example. See Appendix A. Thousand Palms, California Modest Active Water flood Yes Sheetflooding on fan. The general alignment of the fan has been altered by faulting. Lytle Creek, California Modest Inactive Water flood No Flooding confined to a single large trenched channel. See Appendix A. Tortolita Mountains, Arizona Extensive Inactive Water flood No Network of flow paths has appearance of active fan, but flow paths were stable during major flood. See Wild Burro alluvial fan in Appendix A. Carefree, Arizona Extensive Inactive Water flood No Flow is confined to network of trenched distributary channels with no evidence of flow path movement. The use of soil surveys by the Natural Resources Conservation Service is described in the example. See Appendix A. Rudd Creek, Utah Average Active Debris flow and water flood Yes Major debris flow in 1983 damaged or destroyed many homes. Episodes of debris flows are on the order of once every 100 to 1,400 years. See Wasatch Front alluvial fans in Appendix A. Nelson County, Virginia Average Active Debris flow and water flood Yes? Episodes of debris flows are on the order of once every 3,000 to 4,000 years.

OCR for page 83
heads on the eastern slopes of rugged mountains at an elevation of 1,420 m (4,659 feet) (Figures 4-2 and 4-3). Most of the mountainous basin above an elevation of 488 m (1,601 feet) is practically barren of vegetation, and runoff from the steep slopes is rapid. The region is tectonically active, but active faulting is generally located to the north and south of the basin and alluvial fan (Sharp, 1972). The alluvial fan is an example of an arid-clime composite fan with both relict debris flow and modern water flow processes where hazards on the relict fan have been significantly altered by geologically recent channel trenching. Recognizing and Characterizing Alluvial Fans Determining whether or not a Landform is an Alluvial Fan This landform, known as the Henderson Canyon alluvial fan, was identified as an alluvial fan using the criteria defined in Chapter 3 for material composition, morphology, and location. Composition The site is identified as "sloping gullied land" of an alluvial fan on National Resource Conservation Service (NRCS, 1973) soil maps on a 7.5-minute series orthophoto base. It consists of alluvial sediments derived from igneous, sedimentary, and metamorphic rocks. Carrizo soil, which is gravelly sand derived from granitic alluvium and is also associated with alluvial fans, is shown on the lower fan near the valley. The type and relative position of the mapped soils suggest entrenched channels in a relict fan with an active fan downslope in the Carrizo soil. Only upon field inspection of the alluvial fan was it clear that modern channels are deeply trenched into relict debris deposits of the sloping gullied land shown on the soil map (NRCS, 1973). Numerous massive mounds of debris flow deposits are composed of many 0.15-to 0.9-m (0.5 to 3.0-foot) boulders, and many of the fines have been washed from the debris matrix, forming sandy interlobe areas (Figure 4-4). Deposits are massive with distinct boundaries readily observed by field inspection; there is inverse grading and there is a concentration of large boulders at the snout of the deposited lobes. Two major entrenched channels combine in the relict material to form a single entrenched channel that leads to modern alluvial deposits of gravelly sand and scattered cobbles. On the modern deposits, there is some stratification of thin beds that appear to have been deposited as large sheets. These loose and friable deposits are mapped as Carrizo-type soil. The fan is thus composed of relict and modern alluvial deposits like those of an alluvial fan. Morphology The site has the general appearance of a sector of a cone with concentric contour lines that are generally convex downslope and laterally confined in an "embayment" within the general alignment of a mountain front. The form and general bounds of the site can be readily identified on the available 7.5-minute series USGS topographic map (Borrego Palm Canyon Quadrangle). The landform is shaped like a partly extended fan that attenuates at the lateral bounds of a large valley to the east. Location The Henderson Canyon fan is located at a topographic break in lateral confinement at the upper end of the embayment. Many such breaks, although subtle, were observed by

OCR for page 83
FIGURE 4-2 Henderson Canyon drainage basin showing relict alluvial fan boundaries, location of active alluvial fan, location of apex of relict fan, and location of apex of active fan. Hjalmarson and Kemna (1991) using channel profiles of the change in channel slope between topographic map contours. Hooke (1967) described this flattening and steepening of channel slope where confinement is lost at the apex (or intersection point). This break is not apparent on the channel profile defined using the USGS topographic map (Figure 4-3) possibly because the profile represents modern drainage effects. Rather, the surface of this relict fan is a few meters above the present trenched stream channel as observed on the USGS topographic map when used in conjunction with color-infrared aerial photographs obtained from the EROS Data Center of the USGS and orthophoto maps. The topographic break is located near the confluence of two major mountain streams at the upslope edge of relict alluvial deposits. There is a significant change in the surface texture at this location as shown on the aerial photographs and orthophoto maps.

OCR for page 83
FIGURE 4-3 Profile of Henderson Canyon relict alluvial fan near Borrego Springs, California. FIGURE 4-4 View looking downslope and across to the south from center of Henderson Canyon relict alluvial fan at typical boulder mound 1.2 to 1.8 m (3.9 to 5.9 feet) high, March 21, 1995. Courtesy of H. W. Hjalmarson.

OCR for page 83
Defining the Lateral Boundaries and Topographic Apex of the Alluvial Fan The lateral bounds are at the toe of steep-bedrock mountain slopes that form the embayment. Below the mountain front the lateral bounds are defined by topographic ridges between fan drainage channels and the adjacent drainage channels. A swale-like drainage that traverses the fan at the western edge of Borrego Valley forms the fan toe. The fan toe closely coincides with the lower limits of the Carrizo soil unit shown on the NRCS soil survey maps. The general bounds of the alluvial fan can be readily identified on the available 7.5-minute series USGS topographic map (Borrego Palm Canyon Quadrangle). Defining the Nature of the Alluvial Fan Environment A few large channels and an area of active sedimentation are readily apparent on the color-infrared aerial photographs. The location of these channels coincided with evidence of trenched channels in the upper- and mid-fan areas, as suggested by the saw-toothed appearance of the contour lines on the USGS topographic map. The few large kinks in the contours that point upslope are typical of a fan surface with large incised channels. The field investigation revealed light grayish colored rock on the bed and banks of trenched channels, which is indicative of recent abrasion during sediment transport. The adjacent boulders in the debris lobes were lightly covered with rock varnish, which is indicative of a stable surface. Field examination was needed to precisely locate the wide, flat hydrographic apex of the active alluvial fan located on the right, or south, side of the relict fan (Figure 4-2). The hydrographic apex is located at a gradual hydraulic expansion at the end of the large trenched channel. Active Fan There is active sedimentation below the hydrographic apex and active erosion above it. The average slope of the active fan is about 0.025 over a length of about 1.2 km (0.7 mi). The poorly defined channels are slightly braided, with large width-to-depth ratios in the upper fan. There are large sheetflood areas in the middle fan. The surface material appears to be very recent and has not developed a soil. The width of the active fan is approximately one-third the total width along the toe of the Henderson Canyon alluvial fan. The lateral boundaries of the active fan were apparent on orthophoto and topographic maps at 1:2400 scale with 5-foot (1.52 m) contour intervals (furnished by the San Diego County Department of Public Works). These maps were available for the area below the major incised channels that included most of the active fan. The boundaries of the active fan area could be defined on these maps, especially when used in conjunction with the color-infrared aerial photographs. A few relict debris lobes also were indicated on the large-scale topographic maps, but most were attenuated and indistinct. Many sheetflood paths are clearly shown on the photographs, but the lateral bounds of sheetflood are indistinct. A field inspection was needed to distinguish between the sheetflood deposits of the active fan and the deposits below the

OCR for page 83
attenuated debris lobes of the relict fan. Both of these deposits are mapped as Carrizo-type soil (NRCS, 1973). The upper (hydrographic apex) and lower limits (fan toe) of the active fan are approximately defined by the mapped limits of the Carrizo-type soil. However, based on a field inspection of recent sheetflood deposits, much of the toe is below the Carrizo soil at or slightly upslope of a transverse swale-like drainage at the western edge of Borrego Valley. The toe is indistinct, partly because the urban land has been altered by construction or obscured by structures, pavement, and golf courses. Relict Fan The mid-fan is composed of limey debris flow deposits that have been exposed along the entrenched channels. The exposed carbonates of the B horizon extend to depths of nearly 2 m (7.6 feet) along the steep banks of the entrenched channels and are indicative of soils that are more than 10,000 years old in this arid region (Machette, 1985). There is no channel formation or other evidence of flow along the downfan side of the entrenched channel. Evidence of channel formation between the debris lobes is apparent only several hundred meters down the relict fan from the transverse channel. The few small channels in this area are formed by local runoff. The attenuated debris lobes in this area become undefined in the vicinity of the Carrizo soil downslope. Mountainous Drainage Basin Field observations of debris sources on the mountainsides revealed that most of the upper slopes are bare rock and apparently too steep for debris accumulation. The soil in the mountains is loamy coarse sand in texture and is sparse and shallow. The lower slopes of the mountains are covered with boulder debris that appears to be stable because the rock is covered with dark desert varnish and the slopes are less than the angle of repose of the rocks. No slumping was observed. There is one site of a geologically recent small debris slide on the southern side of the drainage basin where the hillslope is at least 38 degrees. This slide is apparent because the scar appears geologically fresh among the darkly varnished surrounding bedrock. Deposited rock from this slide is far from active channels and is not a significant source of material for debris flows down the fan. Little, if any, debris along the stream channels can be seen on aerial photographs. Because there is no known history of recent debris flows in the area and there is little evidence of geologically recent debris flow potential, large debris flows of the size that produced the many mounds are considered unlikely. Storm of August 15–17, 1977 Tropical Storm Doreen produced from 7.6 to 12.7 cm (3.0 to 5 in) of rainfall in the vicinity of the Henderson Canyon alluvial fan. Most of the rain fell during a few hours on the evening of August 16, 1977, producing a peak discharge of 90.6 m3/s (3,200 feet3/s) at the apex

OCR for page 83
of the active alluvial fan about 0.8 km (0.5 mi) upslope of the De Anza Desert Country Club (San Diego County, 1977). This peak discharge was nearly equal to the 100-year flood based on methods described by Thomas et al. (1994). The flow split into two distinct paths upstream of the community. A short distance below the hydrographic apex, where the flow split the floodwaters became less confined and apparently coalesced as sheetflooding. The distribution of floodwater across the active fan at any particular time is unknown, but nearly all of the active fan was inundated at one time or another during the flood. Aerial photographs of flood remnants and hydraulic computations of peak discharge amounts suggest that floodwater covered most of the active fan at the time of the peak discharge. Large amounts of sand with gravel and a few small boulders were deposited throughout the community, and some floodflow passed through the Country Club, inundating farmland to the east. About 100 homes were damaged as previously effective drainage ditches and debris dams were overwhelmed by the floodwater and debris. Floodwater from the drainage basin was conveyed in the incised channels on the relict fan to the hydrographic apex. Most of the basin flow was in the center channel and all downfan flow from the drainage basin was intersected by the channel, crossing the fan from the north. Changes in Flow Path A comparison of entrenched stream channels depicted on aerial photographs showed no discernible channel movement, enlargement, or formation on the relict fan. Three sets of aerial photographs obtained from the EROS Data Center of the USGS were used for the comparison. The photographs were good-quality black-and-white for 1954, poor-quality color-infrared for 1971, and excellent quality color-infrared for 1990. The large-scale orthophoto maps mentioned previously also were used for the assessment of changes in flow path. Significant flow path change was not apparent on the active fan, but minor change of the sheetflood paths is suggested on the photographs and orthophoto maps. Because the paths are obscured by vegetation and possibly by eolian effects, the amount of movement is uncertain. Characterizing Alluvial Fan Flooding Processes Floods have eroded and apparently will continue to erode relict fan material and deposit it on the active fan. All the evidence points to the conclusion that the De Anza Desert Country Club, which is located on the lower portion of the active fan, is in the direct path of future sediment-laden water floods emanating from the 16.6-km2 (6.4-mi2) basin (Figure 4-2). The flood risk on the active fan is much greater than would be predicted by application of the FEMA procedure without recognition of the distinction between the active and the relict portions of the fan. Defining Areas of Active Alluvial Fan Flood Hazard Floodwater leaves the confines of the trenched channel at the hydrologic apex and spreads in two swales as sheetflood. The entire area of the active alluvial fan (Figure 4-2) is subject to alluvial fan flooding.

OCR for page 83
Where does Flow Depart from Confined Channels? Below the hydrologic apex the flow paths are difficult to predict because flow is shallow and unconfined. Where does Sheet Flow Deposition Occur? Sheet flow deposition of sediment over most of the active fan, like that for the flood of August 16, 1977, can be expected during a single flood. Small amounts of sediment deposition on the upper active fan may cause changes in the paths of flow because the flood depths are small and flow is unconfined. Where does Debris Flow Deposition Occur? There is no evidence of recent debris flows. Where are there Structures or Obstructions that Might Aggravate or Cause Alluvial Fan Flooding? Nearly all of the area to the west of the hydrologic apex is within the Anza-Borrego Desert State Park and is not subject to development. The urban development of the De Anza Desert Country Club located on the lower portion of the active fan may alter flow paths and concentrate floodflow in streets and other open areas. Where can the Flood Hazard be Mitigated by Means Other than Major Structural Flood Control Measures? Only major structural controls will be effective because development is along the entire lower portion of the active fan. Defining Areas of Nonalluvial Fan Flooding Hazard along Stable Channels Much of the surface of the relict fan is above the level of flooding in the trenched channels and is not subject to alluvial fan flooding. Floodflow from the surrounding mountains is confined to trenched channels that traverse the relict fan. Fundamental hydraulic computations of channel capacity confirm this conclusion. Crude estimates of channel roughness, size, and slope using a hand level and surveying rod show that the channel capacity of the major channels is several times that needed to convey the 100-year flood, estimated using methods by Thomas et al. (1994), across the surface to the apex of the active fan. Most of the runoff below the transverse channel has been from the relict fan itself, with a small amount from a small mountain basin to the north. Floodflow is less confined downfan in this area where the debris mounds are small and intermound areas have filled with sediment. Some sheetflooding and sedimentation along the toe of the relict fan are expected. Determining the Type of Processes Occurring on the Active Parts of the Alluvial Fan A brief field inspection of the deposited material of the active fan suggested the action of water flood processes in the following ways: Sheetflooding is indicated in the mid-fan area by the thin sorted beds of sand with silt and some gravel, which were loose and friable, and appeared continuous over large areas. The channels in the upper fan had very large width-to-depth ratios, indicating water flow. The deposits were permeable. There are no massive and unstratified deposits and no channels or debris mounds in the middle and lower fan that indicated debris flows. No indicators of debris flows (see Table 3-3 and Figure 3-8) were observed on the active fan.

OCR for page 83
THOUSAND PALMS WASH, CALIFORNIA Thousand Palms fan is located in Riverside County, California. Runoff from the Little San Bernardino Mountains to the northeast is collected along the Indio Hills, mostly by Deception wash, a tributary to Thousand Palms wash, at the Mission Creek fault. The drainage area for Thousand Palms wash is 217 km2 (84 mi2). As the wash passes through the Indio Hills and crosses the San Andreas fault zone into the Coachella Valley, it flows onto a broad alluvial fan (Figure 4-5). There is a general lack of soil development and vegetation on the fan. Furthermore, the main channel loses definition shortly after passing through the apex. Recognizing and Characterizing Alluvial Fans The landform is identified on NRCS (1980) soil maps as an alluvial fan. The revised definition was applied to this example, and the landform was found to be an alluvial fan. Determining whether or not a Landform is an Alluvial Fan The fan-shape is apparent from topographic maps. Some of the soil material is from upstream alluvial fan deposits that have been removed by headcutting in response to strike-slip movement at the Mission Creek fault. The loose and friable deposits are in sheets or beds of sand and silt. The fan has classic concentric contours, but the center of the fan bends gradually to the left or east. Thus, the upper part of the fan faces southwest, while the lower part faces more to the south and east. This shape is related to the general morphology of the Coachella Valley, which lies to the south, and probably to slip faulting across the middle of the fan. There is considerable channel widening as Thousand Palms wash leaves the confines of the Indio Hills and crosses the San Andreas fault, where there has been vertical and lateral displacement. The landform has the composition, morphology, and location to meet the committee's criteria for an alluvial fan. Defining the Lateral Boundaries and Topographic Apex of the Alluvial Fan The lateral bounds of the upper part of the fan are at the toes of steep slopes of older alluvial deposits that can be readily identified on the available 7.5-minute series USGS topographic map. Beyond the general alignment of the Indio Hills, the lateral bounds are the topographic trough lines on each side of the fan. These boundaries are swales and appear slightly concave downfan on the topographic map. The bounds generally correspond to the Carsitas and Myoma soils that are associated with alluvial fans (NRCS, 1980a). The western boundary is indistinct in places because of wind-blown deposits of sand and silt. The fan coalesces with a small fan to the west and with two small fans to the east.

OCR for page 83
Location To meet the criteria in the committee's definition of an alluvial fan, the landform of interest must be located at a topographic break where long-term channel migration and sediment accumulation become markedly less confined than upstream of the break. The Valentino fan, like other small debris fans in Nelson County, is located at a topographic break in slope along the eastern flank of the Blue Ridge Mountains. Defining the Toe, Lateral Boundaries, and Topographic Apex of an Alluvial Fan The gradients of the lower parts of the fans are gentler than those at the fan apexes, as can be seen from the greater spacing of contour lines in Figure 4-17a. The small debris fans in Nelson County (Figure 4-18) typically are encircled by slightly incised streams, many of which wrap around the toes of the fans (see also the shape of the ephemeral stream on the right in Figure 4-17a). These encircling streams, many of which are ephemeral, form the lateral boundaries and toes of the fans in Nelson County and can be identified on contour maps. The topographic apex of the Valentino fan occurs approximately at the boundary between grass-covered slopes and tree-covered, steep hillsides. This is also the point where flow in the channel becomes unconfined and more uncertain and thus is coincident with the hydrologic apex. Defining the Nature of the Alluvial Fan Environment and Identifying the Loci of Active Sedimentation Defining Active The committee recommends that the term active be used to refer to that time period during which sedimentation and flooding are possible in the current regime of climate and watershed conditions. In Nelson County, evidence is available to document when and how often episodic debris flow flooding and deposition have occurred. Furthermore, because it is clear that debris flows are associated with evacuation of hollows that must be full or close to full with colluvium before failure, it is possible for the investigator to have some understanding of the likelihood of activity on a given slope and its downstream depositional fan (Reneau et al., 1986). In Nelson County, debris flow deposition and flooding have occurred about three times in the past 11,000 years, the Holocene (Kochel, 1987). Most fans have been constructed from deposits of three different ages, the oldest of which rests on late Pleistocene (˜13,000 years B.P.) solifluction deposits overlying bedrock and has been radiocarbon dated at about 11,000 years at two sites. A younger deposit that has a mid-Holocene age (6,340 years B.P.; Figure 4-17b) is sandwiched between the basal unit and the historical deposit left by Hurricane Camille in 1969. From these data, geologists estimate that recurrence intervals for episodic debris flow deposition in the area are on the order of >3,000 to 4,000 years (Kochel, 1987), although from a stochastic hydrologist's perspective it is important to note that these events are not random in time. Kochel (1987) provides evidence in support of the hypothesis that debris flow activity in the area was initiated by the Pleistocene-Holocene climatic transition, and in particular by the

OCR for page 83
FIGURE 4-18 A debris flow fan and recent drastically deposited rubble from the headwaters of the North Prong of Davis Creek that formed during Hurricane Camille in 1969. SOURCE: Williams and Guy (1973). Original photograph courtesy of the Virginia Division of Mineral Resources (photographer T. M. Gathright II).

OCR for page 83
onset of incursion of tropical air masses and moisture into the central Appalachians. Incursion of tropical moisture was concurrent with retreat of the polar front as Pleistocene glacial conditions waned and ice masses withdrew northward. These moist air masses can become locked in the steep, rugged terrain of the central Appalachians, unleashing intense rainfall events over short time periods. Many workers have documented that some colluvial hollows throughout the region are primed and prone to failure during such storms (Hack and Goodlett, 1960; Clark, 1987; Kochel, 1987; Jacobson, 1993). For all the reasons summarized here, it seems evident that the past 11,000 years has been a time of active albeit sporadic debris fan growth and development in west-central Virginia. If no historical flooding had occurred, and the youngest deposits throughout the region were mid-Holocene in age, an investigator might be tempted to consider only the past few thousand years as critical to assessing the potential for flooding. However, historical flooding and fan sedimentation have occurred, and all deposits indicate long recurrence intervals that probably reflect the amount of time necessary to replenish the sediment supply in colluvial hollows. As a consequence, in this case a time period of 11,000 years is chosen as the best representation of whether or not a fan is active. (One must keep in mind, however, that numerous alluvial fans on the western flanks of the Blue Ridge have no historical or Holocene deposition and thus would be mapped as completely inactive according to this choice of time unit.) An additional reason for this choice of time unit is the possibility that human activities now increase the potential of evacuation of colluvial hollows. As both tourism and urbanization are increasing in the region, it is probable that the potential for activity on a debris fan has increased. Identifying Areas of Flooding and Deposition for the Time Period Chosen to Represent the Active Part of an Alluvial Fan The alluvial fans in Nelson County are very small, and mapping by Kochel (1987) has indicated that single deposits can be traced across the entire fanhead area (Figure 4-17b). In addition, photographs of flooding during Hurricane Camille indicate that areas comprising up to 50 percent of the total fan area were flooded. Finally, some photographs indicate that human-engineered structures and developments affected the paths of flow on mid- and lower fan areas. Therefore, it is prudent to map the entire fan as active, unless it can be demonstrated that part of a fan is of such high relief and resistance relative to the channelways that it is unlikely to be affected. If a fan did not have flooding and sedimentation during Hurricane Camille, it may be more likely to be flooded during the next large storm because upslope colluvial hollows have not been evacuated in several thousand years. Defining and Characterizing Areas of Alluvial Fan Flooding on Active Parts of Alluvial Fans Defining Areas of Alluvial Fan Flooding Hazard Identifying Areas where Flow Departs from Confined Channels (i.e., where Flow Paths are Uncertain) For the same reasons as stated earlier, all parts of the small fans in Nelson County

OCR for page 83
appear to be susceptible to alluvial fan flooding. As can be seen in Figures 4-13b, 4-15, 4-16 and 4-18, flow paths are highly unpredictable and prone to expansion and shifting. Because of the small size and low relief of the fan surfaces, channel migration is possible on any part of the fan. Identifying Areas where Sheetflood Deposition Occurs Sheetflood deposition can be seen commonly at the toes of the fans in Nelson County (see Figures 4-13b and 4-16), especially where the fans grade into floodplains of larger streams, and thus have very low slopes. Identifying Areas where Debris Flow Deposition Occurs In Nelson County, all deposition except the sheetflood deposits on the lower parts of the fans is the result of debris flows, as demonstrated by analyses of stratigraphy in the fans (Williams and Guy, 1973; Kochel, 1987). Identifying Areas where Structures or Obstructions Might Aggravate or Cause Alluvial Fan Flooding Evidence of structural controls that aggravated or caused alluvial fan flooding in Nelson County can be seen in Figure 4-19, where an access road with fill built across a colluvial hollow might have triggered collapse of sediment in the hollow because of water collection and diversion. Outside of Nelson County, along the Potomac River where other examples of humid region alluvial fans occur, a good example of the migration of channel flow along a highway can be seen (Figure 4-20). Defining Areas of Nonalluvial Fan Flooding Hazard along Stable Channels None of the channels on the alluvial fans in Nelson County appear to be stable; thus all flooding is deemed to be alluvial fan flooding. SUMMARY Judging whether alluvial fans are subject to flooding is not necessarily simple. The intent of these examples is to illustrate and inform, not to second guess past decisions. Four out of the seven examples fit the revised definition of alluvial fan flooding, (i.e., Henderson, Thousand Palms, Rudd Creek, Nelson County). However, although Lytle Creek, Tortolita, and Carefree are alluvial fans, they do not meet the active fan criteria and are therefore not subject to alluvial fan flooding even though they exhibit some characteristics that distinguish them from riverine flooding. Lytle Creek illustrates a site where a large, wide trench cuts the fan from the topographic apex to Cajon Creek along the distal east side of the fan, thereby isolating adjacent parts of the fan from active sedimentation and alluvial fan flooding. Carefree illustrates a site where the network of incised distributary channels transport sediment through the fan and the fan presently is inactive. The dissected, coalesced fan remnants along the west side of the Tortolita Mountains function to transport sediment, are not aggrading, and do not have the shape of an alluvial fan. The seven sites illustrate the complex nature of flooding on alluvial fans and piedmonts and the advantage of using a systematic approach to flood hazard assessment such as suggested by this committee.

OCR for page 83
FIGURE 4-19 Human activities can increase the potential for release of materials, as shown by this mass movement initiated along an access road in an area damaged during the Blue Ridge, Tennessee, storm of 1973. SOURCE: Reprinted with permission from Clark (1987). Applying the committee's definition to field examples reveals that there is a choice to be made between having a very inclusive definition that consists of cases such as alluvial fans, deltas, and braided alluvial washes; or a somewhat exclusive definition that leaves out all nonalluvial fan cases along with those that display certain characteristics but not others. The first alternative would include sites like the Tijuana River (Box 4-1 and Figure 4-21) and involves changing the name alluvial fan flooding in the current regulations to uncertain flowpath flooding and adopting the committee's definition thereof. If the definition is to include cases that fit only one or two criteria, the distinction between alluvial fan and nonalluvial fan must be dropped because it would be inconsistent to include (1) less severe situations where fans are inactive and not subject to the committee's definition of alluvial fan flooding (i.e., and thus not include alluvial fans such as Lytle Creek and Carefree) and (2) nonalluvial fans with stable but distributary paths of flow like much of the western slope of the Tortolita Mountains. This will result in a more inclusive definition. The second alternative is to keep the term alluvial fan flooding in the regulations, clarify that it applies only to alluvial fans, and adopt the committee's definition thereof. This will result in a more exclusive definition.

OCR for page 83
FIGURE 4-20 Debris fan at mouth of Nelson Run, tributary to North Fork South Branch Potomac River, Virginia. Flow along North Fork is from right to left; field of view is 750 m (2,461 feet) wide. Note diversion of tributary flow along road (to the right along base of mountain slope), which then moved back across the fan to reach the North Fork. SOURCE: Jacobson (1993). This choice is to be made by FEMA based on policy, the tolerance for uncertainty in NFIP mapping procedures, and the resources available to restudy those areas that might fit a more inclusive definition. Regardless, the selection of either alternative will result in a better, more precise definition that can be directly applied to the physical processes associated with a given flooding source.

OCR for page 83
BOX 4-1 WHEN IT IS NOT A FAN BUT IT ACTS LIKE ONE: TIJUANA RIVER, CALIFORNIA The Tijuana River flows to the Pacific Ocean in San Diego County, California. Most of the 1,700-square-mile drainage area is in Mexico. Figure 4-21 was taken shortly after a large flood in 1993 and illustrates the effects of sediment deposition during a flood, which caused the river to split into more than one channel. This process resulted in an uneven distribution of flow across the floodplain and significant deviations from predicted base flood elevations. The Tijuana is a coastal stream and clearly not an alluvial fan. However, this flooding source exhibits the characteristics of the existing NFIP definition of alluvial fan flooding: uncertain flow paths, high-velocity flows that create new flow paths through erosion, shallow flow depths, and high sediment transport rates, which cause significant damage by depositing sand in and around structures. Applying the committee's proposed definition and field indicators to this case yields the following findings Is it an alluvial fan? No. The Tijuana River is a coastal stream grading into a coastal delta. Nature of the fan environment: There is a perennial low-flow channel that presently appears to be the main channel. Riparian vegetation is present both in the channel and on other parts of the floodplain. Sediment transport: Coarse sediment yielded by watershed erosion processes has deposited long before it reaches this part of the river, which has a slope of approximately 0.2 percent. Topographic confinement: The floodplain is topographically bounded on the north and south. Due to natural flooding processes, however, there is considerable uncertainty associated with how flow paths are laterally distributed during a flood. Characterizing flooding processes: Review of historical floods since the mid-19th century indicates that the low-flow channel has shifted its location numerous times. Early maps also indicate that there used to be several low-flow channels along the floodplain. Flows do not spread evenly across the floodplain but rather form a number of concentrated conveyance regions. Application of the traditional flood paradigm (using HEC-2) has been somewhat successful even though it fails to account for bank erosion, avulsion, scour, and other aspects of the actual flooding processes. Structural mitigation required: Floodway setbacks, elevation on fill, or similar measures may not necessarily be adequate to mitigate the flood hazard. Although this flooding source is topographically confined, natural sediment transport processes introduce considerable uncertainty into the prediction of stage-discharge relationships for the Tijuana River. The presence of multiple flow paths and the uncertainty they introduce does not, by itself, mean that this area is subject to alluvial fan flooding because the area is not an alluvial fan. The Tijuana River floodplain does, however, fall into the category of uncertain flow path flooding. It would seem reasonable that more stringent rules might apply to the mitigation of flood hazards for this case.

OCR for page 83
FIGURE 4-21 Tijuana River north of the Mexico border (1993). Courtesy of Aerial Fotobank, Inc. REFERENCES Bryant, B. 1988. Geology of the Farmington Canyon complex, Wasatch Mountains, Utah. U.S. Geological Survey Professional Paper 1476. Washington, D.C.: Department of the Interior. Cain, J. M., and M. T. Beatty. 1968. The use of soil maps in the delineation of floodplains. Water Resources Research 4(1): 173–182.

OCR for page 83
Camp, P. D. 1986. Soil Survey of Aguila-Carefree Area, Parts of Maricopa and Pinal Counties, Arizona. U.S. Soil Conservation Service Report. Washington, D.C: Department of the Interior. Chin, E. H., B. A. Aldridge, and R. J. Longfield. 1991. Floods of February 1980 in Southern California and Central Arizona. U. S. Geological Survey Professional Paper 1494. Washington, D.C.: Department of the Interior. Clark, G. M. 1987. Debris slide and debris flow historical events in the Appalachians south of the glacial border. Pp. 139–155 in Reviews in Engineering Geology, Volume VII. Boulder, Colo.: The Geological Society of America. Costa, J., and G. Williams. 1984. Debris flow dynamics. U. S. Geological Survey Open-File Report 84-606, 22-1/2 minutes (videotape). Reston, Va.: U.S. Geological Survey. DMA Consulting Engineers. 1985. Alluvial Fan Flooding Methodology—An Analysis. Federal Emergency Management Agency Contract EMW-84-C-1488. Washington, D.C.: FEMA. Eckis, R. 1928. Alluvial fans of the Cucamonga district, Southern California. Journal of Geology 36:224–247. Federal Emergency Management Agency (FEMA). 1990. Fan, An Alluvial Fan Flooding Computer Program . Washington, D.C.: FEMA. Fields, J. J. 1994. Surficial Process; Channel Change, and Geological Methods of Flood Hazard Assessment on Fluvially Dominated Alluvial Fans in Arizona. Ph.D. dissertation, University of Arizona. Fuller, J. E. 1990. Misapplication of the FEMA alluvial fan model: a case history. Pp. 367–377 in Hydraulics and Hydrology of Arid Lands, R. H. French, ed. New York: American Society of Civil Engineers. Hack, J. T., and J. C. Goodlett. 1960. Geomorphology and Forest Ecology of a Mountain Region in the Central Appalachians: U.S. Geological Survey Professional Paper 347. Reston, Va.: U.S. Geological Survey. Hereford, R., K. S. Thompson, K. J. Burke, and H. C. Fairley. 1995. Late Holocene Debris Fans and Alluvial Chronology of the Colorado River, Eastern Grand Canyon, Arizona: U.S. Geological Survey Open-File Report 95-97. Reston, Va.: U.S. Geological Survey. Hjalmarson, H. W., and S. P Kemna. 1991. Flood Hazards of Distributary-Flow Areas in Southwestern Arizona. U.S. Geological Survey Water Resources Investigations Report 91-4171. Reston, Va.: U.S. Geological Survey. Hooke, R. L. 1967. Processes on arid-region alluvial fans. J. Geol. 75:438–460. Jacobson, R. B., ed. 1993. Geomorphic studies of the storm and flood of November 3–5, in the Upper Potomac and Cheat River basins in West Virginia and Virginia: U.S. Geological Survey Bulletin 1981 . Reston, Va.: U.S. Geological Survey. Keaton, J. R. 1995. Dilemmas in regulating debris flow hazards in Davis county, Utah. Pp. 185–192 in Environmental and Engineering Geology of the Wasatch Front Region, W. R. Lund, ed. Publication 24. Salt Lake City: Utah Geological Association. Keaton, J. R., L. R. Anderson, and C. C. Mathewson. 1988. Assessing Debris Flow Hazards on Alluvial Fans in Davis County, Utah. Final report for U.S. Geological Survey Landslide Hazard Reduction Program, Agreement No. 14-08-0001-A0507. Reston, Va.: U.S. Geological Survey.

OCR for page 83
Kochel, R. C., and Johnson, R. A. 1984. Geomorphology and sedimentology of humid-temperate alluvial fans, central Virginia. Pp. 109–122 in Gravels and Conglomerates, E. Koster and R. Steel. Canadian Society of Petroleum Geologists Memoir 10. Kochel, R. C. 1987. Holocene debris flows in central Virginia: Pp. 139–155 in Reviews in Engineering Geology, Volume VII. Boulder, Colo.: The Geological Society of America. Lowe, M. 1993. Debris Flow Hazards: A Guide for Land Use Planning, Davis County, Utah. U.S. Geological Survey Professional Paper 1519. Reston, Va., U.S. Geological Survey. MacArthur, R. C. and D. L. Hamilton. 1986. Corps of Engineers efforts in modeling mudflows. Pp. 549–554 in Proceedings of a Western State High Risk Flood Areas Symposium, Improving the Effectiveness of Floodplain Management in Arid and Semi-Arid Regions. Madison, Wisc.: Association of State Floodplain Managers. Machette, M. N. 1985. Calcic soils of the southwestern United States. Geological Society of America Special Paper 203. Boulder, Colo.: The Geological Society of America. Mathewson, C. C., and J. R. Keaton. 1990. Multiple phenomena of debris flow processes: A challenge for hazard assessments. Pp. 549–554 in Proceedings of the International Symposium, Hydraulics/Hydrology of Arid Lands. New York: Hydraulics Division of American Society of Civil Engineers. McGlashan, H. D., and F. C. Ebert. 1918. Southern California floods of January 1916. U.S. Geological Survey Water-Supply Paper 426. Reston, Va.: U.S. Geological Survey. Mills, H. H. 1983 Piedmont evolution at Roan Mountain, North Carolina. Geografisika Annaler 65A:111–126. Mills, H. H., and J. B. Allison. 1994. Controls on the variations of fan-surface age in the Blue Ridge Mountains of Haywood County, North Carolina. Physical Geography 156:465–480. Natural Resources Conservation Service (NRCS). 1968. Soil Survey of Davis-Weber Area, Utah . U.S. Department of Agriculture. Washington, D.C.: U.S. Department of Agriculture. Natural Resources Conservation Service (NRCS). 1973. Soil Survey of San Diego County, California. Washington, D.C.: U.S. Department of Agriculture. Natural Resources Conservation Service (NRCS). 1980a. Soil Survey of Riverside County, California, Coachella Valley Area. Washington, D.C.: U.S. Department of Agriculture. Natural Resources Conservation Service (NRCS). 1980b. Soil Survey of San Bernardino County—Southwestern Part, California. Washington, D.C.: U.S. Department of Agriculture. Patton, P. C., and V. R. Baker. 1976. Morphology and floods in small drainage basins subject to diverse hydrogeomorphic controls. Water Resources Research 12(5):941–952. Pearthree, P. A., K. A. Demsey, J. Onten, K. R. Vincent, and P. K. House. 1992. Geomorphic Assessment of Fluvial Behavior and Flood-Prone Areas on the Tortolita Piedmont, Pima County, Arizona. Arizona Geological Survey Open-file Report 91-11. Tucson, Ariz.: Arizona Geological Survey. Reneau, S. L., W. E. Dietrich, D. J. Donahue, A. J. T. Juli, and M. Rubin. 1986. Late Quaternary History of Colluvial Deposition and Erosion in Hollows, Central California Coast ranges. Geological Society of America Bulletin 102(7):969–982. Rhoads, B. L. 1986. Flood hazard assessment for land use planning near desert mountains. Environmental Management 10(1):97–106. Ritter, D. F., R. C. Kochel, and J. Miller. 1995. Process Geomorphology, 3rd ed. Dubuque, Iowa, Times Mirror Higher Education Group.

OCR for page 83
Ritter, J. B., J. R. Miller, Y. Enzel, S. D. Howes, G. Nadon, M. D. Grubb, K. A. Hoover, T. Olsen, S. L. Reneau, D. Sacks, C. L. Summa, I. Taylor, K. C. N. Touysinhthiphonexay, E. G. Yodis, N. P. Schneider, D. F. Ritter, and S. G. Wells. 1993. Quaternary evolution of Cedar Creek alluvial fan, Montana. Geomorphology 8:287–304. San Diego County. 1977. Storm Report for Tropical Storm Doreen, August 15–17, 1977. San Diego: County of San Diego Department of Sanitation and Flood Control. Shanmugan, G. 1996. High-density turbidity currents: Are they sandy debris flows? Journal of Sedimentary Research 66(1):2–10. Sharp, R. V. 1972. The Borrego Mountain Earthquake of April 9, 1968. U.S. Geological Survey Professional Paper 787. Reston, Va.: U.S. Geological Survey. Thomas, B. E., H. W. Hjalmarson, and S. D. Waltemeyer. 1994. Methods for Estimating Magnitude and Frequency of Floods in the Southwestern United States. U.S. Geological Survey Open-File Report 93-419. Reston, Va.: U.S. Geological Survey. Troxell, H. P. 1942. Floods of March 1938 in Southern California. U.S. Geological Survey Water-Supply Paper 844. Reston, Va.: U.S. Geological Survey. Waananen, A. O. 1969. Floods of January and February 1969 in Central and Southern California. U.S. Geological Survey Open-File Report. Reston, Va.: U.S. Geological Survey. Wells, S. G., and A. M. Harvey. 1987. Sedimentologic and geomorphic variations in storm-generated alluvial fans, Howgill Fells, northwest England. Geological Society of America Bulletin 98:182–198. Williams, G. P., and H. P. Guy. 1973. Erosional and depositional aspects of Hurricane Camille in Virginia, 1969. U.S. Geological Survey Professional Paper 804. Reston, Va.: U.S. Geological Survey. Williams, S. R., and M. Lowe. 1990. Process-based debris flow prediction method. Pp. 66–71 in Hydraulics/Hydrology of Arid Lands, R. H. French, ed. New York: American Society of Civil Engineers.