ELMTO ISSUE RECLAMATION z z z z z z z lo lis amnhabitat, salmon plains, Flood oe ntenwWsigo tt fossil, State Washington new 59 the p. on Washington, Notes Sammamish, the Lake from and features Peninsula coseismic Olympic 48 probable p. of Washington, ages southwest Radiocarbon from fossils trace Miocene 43 29 Early p. p. Awards, tailings, Reclamation metal-mine Mine of Surface reclamation in trends and Innovations off-channel as 21 pits p. gravel habitat, and salmon sand flood-plain of Reclamation

NATURAL RESOURCES vlinrejoins avulsion anstem main

abandoned main stem channel n adadgae iig .3 p. mining, gravel and sand and IE VLINIT LO-LI RVLPTMNS1995/96 MINES PIT GRAVEL FLOOD-PLAIN INTO AVULSION RIVER valley wall amtu columbi Mammuthus ,p.68 rvlptlakes pit gravel ETME 1998 SEPTEMBER O.2,N.2/3 NO. 26, VOL. W n rsn path present and 96avulsion 1996

ASHINGTON atFr ei River Lewis Fork East G avulsion on of point 1995 EOLOGY valley wall rvlpit gravel lower lakes upper pit avulsion avulsion on of point 1995 1996 Association of American State WASHINGTON Geologists Present Results of GEOLOGY STATEMAP Program Vol. 26, No. 2/3 September 1998 Raymond Lasmanis, State Geologist Washington Department of Natural Resources Washington Geology (ISSN 1058-2134) is published four times each year by the Washington State Department of Natural Resources, Division of Geology and Earth Resources Division of Geology and Earth Resources. This publication is free upon PO Box 47007, Olympia, WA 98504-7007 request. The Division also publishes bulletins, information circulars, reports of investigations, geologic maps, and open-file reports. A list of these publications will be sent upon request. n March 18, 1998, the Association of American State OGeologists hosted a reception in the foyer of the Rayburn DIVISION OF GEOLOGY AND EARTH RESOURCES House Office Building in Washington, D.C. Members of Con- Raymond Lasmanis, State Geologist gress and their staff, committee staff, and representatives of J. Eric Schuster, Assistant State Geologist agencies that use geologic maps were invited guests. Forty- William S. Lingley, Jr., Assistant State Geologist five state geologists participated, and each state presented a poster displaying a STATEMAP product and derivative infor- Geologists (Olympia) Editor mation that addresses societal needs. Joe D. Dragovich Katherine M. Reed I presented the poster for Washington State. It consisted of Wendy J. Gerstel Senior Cartographer/ two derivative maps derived from a digital version of the Robert L. (Josh) Logan GIS Specialist 1:100,000-scale Seattle quadrangle. Developed using Depart- David K. Norman Carl F. T. Harris Stephen P. Palmer Patrick T. Pringle Cartographers Katherine M. Reed Keith G. Ikerd Henry W. (Hank) Schasse Anne Heinitz Timothy J. Walsh Production Editor/ Weldon W. Rau (volunteer) Designer Geologist (Spokane) Jaretta M. (Jari) Roloff Robert E. Derkey Data Communications Geologists (Regions) Technician Garth Anderson (Northwest) J. Renee Christensen Charles W. (Chuck) Gulick Administrative Assistant (Northeast) Janis G. Allen Rex J. Hapala (Southwest) Regulatory Programs Lorraine Powell (Southeast) Assistant Stephanie Zurenko (Central) Mary Ann Shawver Senior Librarian Clerical Staff Connie J. Manson Philip H. Dobson Library Information Cathrine Kenner Specialist Lee Walkling

MAIN OFFICE FIELD OFFICE State Geologist Ray Lasmanis with Washington STATEMAP poster. Department of Natural Resources Department of Natural Resources Division of Geology Division of Geology and Earth Resources and Earth Resources ment of Natural Resources GIS capabilities, one of the poster PO Box 47007 904 W. Riverside, Room 209 maps showed locations of rock, sand, and gravel resources. Olympia, WA 98504-7007 Spokane, WA 99201-1011 The second map showed earthquake liquefaction hazard areas Phone: (360) 902-1450 Phone: (509) 456-3255 ranked by the potential severity of the hazard. The narrative Fax: (360) 902-1785 Fax: (509) 456-6115 with the poster listed various applications of digital geologic (See map on inside back cover E-mail: [email protected] map data and Division clients, such as: for main office location.) Publications available from the z Forest resource management – Weyerhaeuser Internet Connections: Olympia address only. Library inquiries: Copying is encouraged, but please Corporation, Simpson Timber, Boise Cascade [email protected] acknowledge us as the source. z Fisheries management – Skagit System Cooperative, [email protected] Printed on recycled paper. Washington Association of Federated Tribes Subscriptions/address changes: Printed in the U.S.A. [email protected] z Mineral resource exploration – BHP Minerals, Western URL: http://www.wa.gov/dnr/htdocs/ger/ger.html Mining Corporation z Federal and State agencies – U.S. Forest Service, Cover Photo: The East Fork Lewis River, Clark County, southwest Wash- Washington Department of Ecology ington, in June 1994. Ponds created by gravel mining here cover approxi- mately 200 acres on both sides of the river and are about 30 ft deep. The val- z Local governments – Clark County, Lewis County, ley is approximately 1 mile wide in this area. During 1995 and 1996, the river Tacoma-Pierce County Health Department avulsed through a gravel pond (shown by solid line) and abandoned about 1,700 ft of channel. Later avulsion forced the river into the lower gravel pits z Research – Central Washington and Evergreen State and abandoned an additional 3,200 ft of channel. Small arrows indicate the universities, Pacific Northwest Laboratory 1994 path of the river (today’s abandoned channel). See related article, p. 3. Continued on page 75

2 Washington Geology, vol. 26, no. 2/3, September 1998 Flood Plains, Salmon Habitat, and Sand and Gravel Mining

David K. Norman C. Jeff Cederholm William S. Lingley, Jr. Department of Natural Resources Department of Natural Resources Department of Natural Resources Division of Geology and Earth Resources Resource Planning and Asset Management Division Division of Geology and Earth Resources PO Box 47007, Olympia, WA 98504-7007 PO Box 47014, Olympia, WA 98504-7014 PO Box 47007, Olympia, WA 98504-7007

INTRODUCTION morphic flood plain, or the area where fluvial erosion has cre- ated a flat valley, is generally much larger than the “calculated” People who live or work along rivers and their flood plains 100-yr flood plain and is where mining occurs. know these are hydrologically and biologically dynamic envi- Flood plains support diverse plant and animal populations, ronments. Rivers flood and shift their courses from time to as well as much of our agricultural system (Teskey and Hink- time, resulting in natural cycles of erosion and deposition of ley, 1977). Vegetation along river banks provides habitat for sand and gravel. The river and its banks are also home to many most wildlife. Significantly, the riverine environment is home fish and wildlife species. In this age of rapid land development, to the many species of Pacific salmon (Oncorhynchus sp.) in- however, people have turned to rivers (dredging and gravel bar digenous to Washington (Palmisano and others, 1993; Wash- mining) and flood plains (gravel pit lakes) as sources of sand ington Department of Fish and Wildlife, 1997; Sedell and and gravel for construction aggregates. Luchessa, 1982). The riparian area is where aquatic and terres- Gravel mining practices such as channel dredging are typi- trial ecosystems interact and is essential to both fish and wild- cally done with either a dragline or suction dredge (Fig. 1) and life. Streamside plants shade the water, help moderate water are conducted mostly in the Columbia River, where a shipping temperature, and promote stream-bank stability, as well as pro- channel is maintained, and the Cowlitz River, where large vol- viding the organic nutrient load in the aquatic ecosystem. umes of sediment from the 1980 Mount St. Helens eruption are These plants are the source of large in-stream woody debris still being deposited. Gravel bar mining (scalping) (Fig. 2 ) is that provides refuge and food sources (Washington Depart- performed in many Washington rivers for aggregate and in an ment of Natural Resources, 1996). Salmon use gravel channels attempt at flood control (Collins, 1996). Miners no longer dig for spawning and clay-bottom channels, with their rich into the banks of rivers (Fig. 3). Also, the Army Corps of Engi- subaqueous flora, for overwintering, feeding, and refuge. neers, state agencies, and some local governments have recog- Side-channels (yazoo tributaries) and oxbow lakes are types of nized the environmental effects of in-stream mining. The wallbase channels (Fig. 4A) or off-channel sites that are pri- Grays Harbor County Planning Department, for example, has mary overwintering habitats for juvenile coho salmon and cut- mandated reduced rates of gravel removal on the Satsop, Wy- throat trout (Peterson, 1980; Cederholm and Scarlett, 1981; noochee, and Humptulips Rivers (Collins and Dunne, 1990). (See Fig. 6.) Flood-plain mining, digging gravel pit lakes adjacent to rivers, is a common activity in Washington. “Flood plain” as defined in Washington’s Shoreline Management Act (Revised Code of Washington [RCW] 90.58) means the 100-year flood plain, the area suscepti- ble to flooding by a stream for which there is a one percent chance of inunda- tion in any given year (Washington De- partment of Ecology, 1994). The 100- year flood plain definition must be used with caution because the small flood- frequency data sets result in large po- tential for error in identifying the 100- year event and determining the area af- fected (Mount, 1995). Also, as water- plume sheds become more developed and roofs and roads present large areas of impermeable surfaces, high-intensity peak runoff occurs more frequently and with greater volume (Booth, 1991). The definition of the 100-year flood plain is important in Washington be- Figure 1. A suction gravel dredge and barges at work in the Chehalis River in the mid-1960s. cause most regulation of development This type of mining now occurs chiefly in the Columbia and Cowlitz Rivers to remove sediment de- and mining depends on where this posited after the Mount St. Helens eruption. Note the suspended sediment plume moving down- “line” is drawn. Furthermore, the geo- stream. (Photo by Lloyd Phinny, Washington Department of Fisheries.)

Washington Geology, vol. 26, no. 2/3, September 1998 3 Peterson and Reid, 1984; Cederholm and Scarlett, 1991). Wallbase channels gullies, ridges, commonly look like swamps, ponds, or and scrape marks small tributaries and are connected to the main stem of the river by a small stream called an ingress/egress channel. Rivers in Washington generally have steep gradients and V-shaped cross sections in their upper reaches. Headward erosion occurs where stream point bar gradients rapidly steepen, as in the mining Olympic and Cascade mountains. As the rivers approach the ocean or find their way across the Columbia Basin, gradients decrease, valleys broaden, flow velocity decreases, and gravels are deposited. Deposition is greatest at the gradient changes. The river accom- modates this aggradation by lateral (horizontal) migration. Traces of chan- nel shifts and stream meanders are Figure 2. Scalping a point bar in the Wynoochee River in about 1965. Current rules require that the post-mining ridges and gullies on the gravel bar to be graded to 2 percent toward the river and common flood-plain features; oxbow that no areas where fish could be trapped remain. Gravel bar mining has been restricted in several lakes and side channels are testament to rivers, such as the Wynoochee where over-mining of gravels was proven by gravel bedload trans- the history of this activity (Fig. 4A,B). port studies conducted for Grays Harbor County (Collins and Dunne, 1990). (Photo by Lloyd Flood-plain gravel mining gener- Phinny, Washington Department of Fisheries.) ally occurs where the gravel bedload has been deposited, for example, at gra- dient changes or above or below topo- gravel stockpiles graphic constrictions through which a and processing area river flows (such as Union or Selah gaps on the Yakima River). In Wash- ington, many gravel pits near rivers have been excavated for fill material sediment for major highway projects, a process pond that further complicates flood-plain sediment being functions. Examples are I-90 along the released from pond Yakima River, I-5 where it crosses the Skookumchuck River at Centralia, I-82 near metropolitan Yakima on the Ya- kima and Naches Rivers, and in several places along the East Fork Lewis River sediment near Vancouver. Gravel sources lo- plume cated near the point of use significantly reduce the cost of the aggregate. Seeking the lowest cost material, gravel miners commonly choose to ex- cavate large, deep ponds adjacent to Wynoochee River active river channels (Fig. 4A). These pits have the potential to significantly change the physical and ecological Figure 3. This plume of turbid water (lighter gray area at top) from a gravel mine on the Wy- function of flood plains (Collins, 1996, noochee River was created by bar scalping and shoreline mining, no longer practiced in Washing- 1997). Wherever a channel shifts into a ton. The small pond at the left center of the photo was used as a settling pond. Note turbid water gravel pit or multiple pits that are large discharge from the sediment pond to the river at the center of photo. (Photo taken about 1965 by relative to the scale of the flood plain Lloyd Phinny, Washington Department of Fisheries.) and the river’s sediment transport re- gime, natural recovery of original flood plain environment and similar channel morphology could take millennia (Collins, 1997). The time for development and consequent flood magnitude increase. In the recovery is highly dependent on the availability of sediment, long term, stream capture by gravel pits is a near certainty. Be- particle size, gradient, and the size of excavations to be filled. cause the gravel pits have a lower base elevation, there is risk Regardless of the best planning and intentions, impacts of of rapid channel change into the pits during high flows, a pro- flood-plain mining may simply be delayed until the river is cess termed avulsion. The flooded pits “capture” the stream. captured by the gravel pit. While capture may not occur in the The effects of avulsion are similar to those of in-stream mining next 100-year flood event, it is likely to occur in the future as discussed in Evoy and Holland (1989), Collins and Dunne

4 Washington Geology, vol. 26, no. 2/3, September 1998 geomorphic flood plain

abandoned oxbow channel wallbase wallbase lake channel channel (pond) cutoff ingress/egress channel

dike/levee/berm wallbase point wetlands channel bar

side channel riparian erosion zone point bar wallbase channel (pond) main stem abandoned gravel pit river channel berm dragline 10–15 deep¢ operating tower dike levee/ sand & gravel & stockpiles dike/ pit/lake

t water table tailhold t water table dragline pit/lake 25– bucket 100 deep¢ alluvial gravels (hyporheic zone) bedrock A not to scale

Figure 4. A. Diagram showing a flood plain with an active, meandering river channel and dragline digging a gravel pit. Note the rela- tion of the pit to the water table and main stem river channel with dikes built to protect the pit and channelize the river. Also shown are point bars, wallbase channels (including side channels, ponds, and oxbow lakes) and associated wetlands, all of which are impor- oxbow tant for salmon habitat. B. Oxbow lakes with narrow riparian zones on the broad and flat river Chehalis River flood plain near the Chehalis airport. Oxbow lakes such as these that are oxbow lake oxbow hydraulically connected to the main stem of a river can provide off-channel habitat for salmon. Here, much of the riparian zone has been removed for agriculture. oxbow B river

(1990), Netsch and others (1981), Kondolf and Graham Mat- Flood-plain mine lakes, the loss of natural flood-plain habi- thews (1993), Kondolf (1993, 1994), and Williamson and oth- tat, and the isolation of the flood plain from its river by armor- ers (1995a,b). They may include: ing and dikes that protect against avulsion are semipermanent z lowering the river bed upstream and downstream of consequences of flood-plain mining. Careful reclamation may mining operations, causing river bed erosion and (or) restore some of the previous function of a flood plain; how- channel incision and bank erosion and collapse, ever, it may be impossible to replace lost habitat in the near z term if a substantial amount of a flood plain has been converted eroding of footings for bridges or utility rights-of-way, to pit lakes (Collins, 1996, 1997). z changing aquatic habitat, Careful siting, planning, limiting mining, a thorough z unnaturally simplifying the complex natural stream hydrogeological analysis, use of alternative resources, and in- system, novative reclamation can mitigate and reduce some mining im- z increasing suspended sediment, and pacts. This article describes river processes and mining opera- z tions and discusses some examples of flood-plain mines. Rec- abandoning reaches of spawning gravels or damaging lamation of these mines is the subject of the article by Norman these gravels by channel erosion or deposition of silts (this issue, p. 21). in spawning and rearing reaches.

Washington Geology, vol. 26, no. 2/3, September 1998 5 FLOOD-PLAIN HYDRAULIC PROCESSES during floods combine to create a flood plain. The width of the flood plain is a function of many factors, including the size of Many Washington river valleys reflect millions of years of flu- the stream, relative rates of meander migration and downcut- vial processes and in some places modification by glaciers. ting, and the ease with which the valley walls are eroded. Dur- Most lowland river valleys are filled with alluvial sand, gravel, ing periods between floods, the meandering stream occupies silt, and clay, which have been spread out across the adjacent only a portion of the breadth of the flood plain. Slow lateral vegetated flood plains wherever flood flows have left the main channel migrations over long periods of time tend to maintain a channel. Small irregularities in the channel cause local varia- single flood plain with a fairly level surface. In aggrading tions in flow velocity, which in turn contribute to variations in reaches, where cobbles and boulders are the dominant bedload erosion and sediment deposition. From these variations, the and sediment transport rates are high, the river may take on a river begins slow lateral displacement, depositing on one braided morphology. Braided rivers are characterized by a stream bank while eroding, scraping, and undercutting the channel dividing and reuniting in an intricate interlaced fash- other bank as it gradually migrates from side to side across its ion. Early maps of the state show that many rivers flowing from valley floor. The valley widens over time by erosion, particu- the Cascades had braided channels. larly on the outside of river bends and where the flow impinges If conditions are suddenly and significantly changed, rela- against the valley walls. The river’s course appears snakelike tive rates of meandering and downcutting may accelerate, and when viewed from above (Leopold and others, 1964). the old flood plain may be cut up by development of new, nar- The water strikes with greatest force against the outer (con- rower channels. Incision can result from an increase in the riv- cave), downstream bank of the meander bend, causing erosion. er’s discharge, reduction in the size of its bedload, dredging, Flow is slower on the inside (convex) bank of a meander where avulsion into a gravel pit, or upstream levee construction. deposition occurs. These deposits are called point bars, gravel As a valley widens, the stream’s flood plain also widens. bars, or lateral accretion surfaces. Over time, meanders can en- The flood plain width can be limited, as it is along the en- large sideways and shift downstream by a combination of ero- trenched Yakima River canyon between Ellensburg and Roza sion and sedimentation processes. A meandering river gener- dam (Figs. 5, 6), or very broad, as it is along the Chehalis River ally retains the same approximate channel width, but a given downstream of Oakville, where it is more than 2 mi wide river channel may eventually occupy most parts of a flood plain (Waitt, 1979; U.S. Army Corps of Engineers, 1970). A flood throughout its geologic history (Fig. 4A,B). plain’s width can also be out of proportion to its river channel. Meander bends do not enlarge indefinitely because they For example, the Chehalis River appears to have an undersized become sizable detours for the flowing water. Eventually, the channel in the wide valley along its lower reaches. During the river will cut off a meander and abandon the longer, curved Pleistocene epoch, the Chehalis River valley was the outlet for channel for the shorter, more direct route. The cutoff meander the enormous amounts of glacial outwash and meltwater from sections are termed oxbows. Where they are filled with water, the Puget ice lobe (Bretz, 1913; Eddy, 1966). The modern Che- they are called oxbow lakes or ponds (Fig. 4A,B). Meander halis River drains a much lower and smaller watershed area. cutoffs most commonly occur during floods, when large vol- The upper reaches of rivers such as the Stillaguamish, Cowlitz, umes of water have enough additional momentum to carve the and east fork of the Lewis (Mundorff, 1984) flow in more re- shortcut and bypass the meander bend. stricted glacier-carved valleys; only in the lower parts of these As river profiles or individual reaches become stable over a valleys do they appear underfit in proportion to their overall period of years, they achieve a balance between erosion and flood-plain size. deposition. At equilibrium, the river or a reach is a graded stream, one in which the slope, velocity, and discharge com- bine to transport its sediment load with neither erosion nor sedimentation (Mackin, 1948). That balance is gov- erned by the elevation of its base level (for example, an adjacent ocean or lake level) and the many physical and bio- logical factors that control equilibrium. Base levels and the stream’s equilib- rium profile can be altered by human intervention, such as dam building, ac- tivities that cause river avulsion, or re- moval of gravel bars. Gradually the stream’s profile will reach a new equi- librium, but this may change the distri- bution of erosion and deposition sites and also the shape of the channel. A flood plain is a flat or gently slop- ing region along a stream channel, typi- cally extending laterally to the adjacent slopes, onto which the stream expands during floods (Fig. 4A,B). With time, Figure 5. Yakima Canyon near Umtanum gaging station; view to the north, toward Ellensburg. lateral erosion associated with mean- Note the entrenched river with meander loops cut into the basalt flows. The resistant basalt con- dering, channel sediment deposition, strains the channel, and no flood plain can develop here, and no off-channel habitat can be cre- and additional overbank sedimentation ated as on a flatter flood plain.

6 Washington Geology, vol. 26, no. 2/3, September 1998 Forks Everett 48o

2 Clearwater R. Seattle Leavenworth 2

Bremerton 405 North R. River Wenatchee River Bend 90 Elum

Tacoma Cle 97 28 101 AII OCEANPACIFIC West Fork Moses Humptulips Cle 281 Lake 101 East Fork 5 Elum 90 Wynoochee Satsop River Naches River Ellensburg Chehalis 47o Aberdeen Olympia Yakima Oakville Selah 12 Skookumchuck Gap pits River River Roza Centralia Fork dam Chehalis 101 North 1212 Columbia River NewaukumFork R. Morton Toledo South River Yakima Union Gap pits River Raymond Mayfield Toledo dam Cowlitz Parker IFA Nursery pits 4 504 82 Richland 22 Pasco Kelso Columbia Kennewick Longview 5 97 82

o River 46o 124 123o 14 La Center o East Fork River 119 Ridgefield Salmon Lewis River White Salmon Goldendale pits Columbia 0 20 mi Cr. 120o Vancouver 205 0 30 km 14 OREGON o 122o 121

Figure 6. Rivers, towns, and flood-plain gravel mines in Washington referred to in this article.

BIOLOGICAL PROCESSES Poorly planned mining exposes these salmon habitats to po- IN FLOOD PLAINS tential avulsion. Many gravel pit mines have been established Off-Channel Habitat for Salmon in active wallbase channels, sloughs, and (or) marshes (Col- lins, 1996, 1997), resulting in destruction of salmon habitat. In Rivers are not isolated ribbons of water flowing through a val- rare instances, flood-plain mining has created effective off- ley; they are connected to their valleys and flood plains by both channel habitat (see Norman, this issue, p. 21). surface water and shallow ground-water sources (Collins, 1996). Wallbase channels (Fig. 4A,B) form on flood plains Benthic Insects and the Hyporheic Zone where runoff reoccupies abandoned channels created by the Benthic macroinvertebrates are an important food source for migration of main stem rivers (Peterson and Reid, 1984). Wall- juvenile salmon (Mundie 1969). Many benthic invertebrates, base channels develop along meander scars and oxbows or be- such as caddis flies, mayflies, and stone flies, spend all or part hind gravel bars; in most places they appear as swamps, ponds, of their lives in a clean gravelly substrate that has abundant or small tributaries and are connected to the main stem (Fig. 7). oxygenated water (Pennak 1978; Merritt and Cummins, 1984). During spring and fall, juvenile salmon migrate out of main Healthy streams are characterized by a high species diversity rivers into tributaries, including wallbase channels, where they of benthic macroinvertebrates and a moderate number of indi- seek refuge from high flows and turbidity for varying periods viduals of any given species group (Brookes, 1989). Unhealthy of time (Cederholm and Scarlett, 1981; Peterson and Reid, streams, on the other hand, are characterized by low species di- 1984; Peterson, 1982; Scarlett and Cederholm, 1984; Brown versity and large numbers of individuals in any species group. and Hartman, 1988). In fall, these movements are initiated dur- The hyporheic zone (Fig. 4A), the shallow unconfined aq- ing the first freshet (Peterson, 1980). Increased discharge and uifer under the flood plain that is in hydraulic continuity with turbidity in the main river relative to clear water flowing from the river, may extend miles horizontally across the width of the wallbase channels attracts large numbers of juvenile coho and flood plain and many yards beneath the surface (Stanford and cutthroat. While in these channels, salmon take advantage of a Ward, 1988). This zone provides extensive interstitial (inter- rich food supply of aquatic insects and relatively stable hydro- granular) habitat for benthic macroinvertebrates. The hypo- logical conditions. Juvenile salmon typically feed in the shal- rheic zone serves as a refuge for benthic insects during lows and seek cover from predators in deeper water or in droughts and high-flow events, and it is capable of re- woody debris complexes and emergent vegetation. The growth supplying the population of an area at and immediately below that the juvenile salmon immigrants (Fig. 8) are able to acquire the stream bed once conditions in the stream improve. This in these habitats improves their overall size and survival rates zone plays a major ecological role in streams, especially those (Peterson, 1980; 1982).

Washington Geology, vol. 26, no. 2/3, September 1998 7 location of ingress/ of inset egress channel

wallbase channel pond flow

wallbase channel pond

ingress/ egress channel

Figure 7. A wallbase channel pond on the Clearwater River that serves as off-channel habitat for salmon. This pond covers only 3 acres and has a maximum depth of 12 ft. This remnant of a channel cutoff is partially covered with lily pads and is surrounded with a robust riparian zone. The inset shows a close-up of the pond area. that have coarse substrates where most aquatic insects spend important parts of their life cycle (Ward, 1992). Overly large and deep flood-plain mines may significantly interrupt local hyporheic processes through removal of gravels and (or) by changing water-table eleva- tions.

Effects of Avulsion Figure 8. Coho smolt being measured for growth after spending the winter in the pond shown in Salmon depend on a river system that Figure 7. Scale is in centimeters. Smolts that spend the winter in wallbase channels and ponds are functions well. Salmon lay their eggs in 25 percent longer and weigh more than smolt that migrate immediately to the ocean (Cederholm and Scarlett, 1991). Coho that overwinter in ponds such as this one grow larger and have a better gravels, where cool, oxygenated water chance of survival once they migrate to the ocean. flows freely over the eggs, assuring their normal development. Generally, the most productive river substrates are those that are stable ease outbreaks and may kill significant numbers of adult and and have a wide range of particle sizes. They also produce the juvenile fish (Bjornn and Reiser, 1991). most diverse invertebrate populations (Brookes, 1989). If the Pits that are warmer than the adjacent river may be ideal gravel substrate is disturbed or is filled and covered by silt and habitat for warm-water fish, such as large mouth bass or yellow clay, as can happen during floods and avulsion episodes, perch, which are predators of juvenile salmon (Kondolf and salmon egg nests (redds) can be eroded away or smothered. others, 1996). This may be particularly applicable in the lower Additionally, the benthic invertebrate community is likely to reaches of the Yakima River, where low flows and warming change and become less productive when the stream bed is al- river water during summers is a chronic problem. Additionally, tered. Where velocities are persistently high, many kinds of smolting juvenile salmon may become disoriented in the quiet benthic macroinvertebrates may be absent. waters of a pit that has been captured by a river (Ken Bates, When a river breaches a pit, the river biota can be cata- Washington Department of Fish and Wildlife [WADFW], writ- strophically changed. Water temperatures may rise during ten commun., 1998). summer and early fall because the relatively slack water in the pits is exposed to sunlight for long periods. Water temperatures o RIVERS AND FLOOD PLAINS AFFECTED above 75 F may be fatal to some salmon species, and tempera- BY MINING IN WASHINGTON tures between 60o and 75oF can increase their metabolic rates and stress levels. While moderate increases in water tempera- Flood-plain mining has occurred on many rivers in Washing- ture can increase growth rates, large increases can cause dis- ton State and is highly concentrated along the Yakima River

8 Washington Geology, vol. 26, no. 2/3, September 1998 and its tributaries (Naches and Cle Elum Rivers), the Chehalis Additionally, dikes and levees can be damaged by dewater- River and its tributaries (Wynoochee, Satsop, Newaukum, and ing an adjacent pit by pumping or flow though an outlet chan- Skookumchuck Rivers), and the Cowlitz River and East Fork nel. During floods, when the river’s level is abnormally high, a Lewis Rivers (Fig. 6)(Collins, 1996, 1997). Collins estimates large static head exists between the higher river water surface that about one-sixth of Washington’s gravel production was re- and that of the pit. Here, the potential for dike collapse and moved from riverine sources between 1970 and 1991, and most avulsion is great. The dikes are weakened by the constant head of this mining was located on flood plains and active gravel difference. Floods that overtop dikes can allow a river to rap- bars. However, this paper does not attempt to address all the idly cut through these structures as the flow drops from the flood-plain gravel pits in the state. river surface into the pit, causing headward erosion. In effect, Many currently permitted mines cover several hundred dikes and levees are only short term solutions to long term acres of flood plain that have been excavated or are scheduled natural processes such as channel migration and flooding. for excavation. The depth and size of many of these mines was restricted by economics and the types of machinery used. Consequences of River Avulsion Gravel mining has created pit lakes ranging from a few Typically, avulsion occurs during floods. Avulsion is charac- acres to several hundred acres and with depths averaging about terized by a sudden change in the course of a river that causes it 30 ft. Several mined lakes are as deep as 90 ft, and some are as to break through a low point such as a meander neck (to form an much as five times as deep as the adjacent river. Some of these oxbow lake) or to rush into a gravel pit. Avulsion events occur deep lakes are within 50 ft of the active river channels. This in gravel pit lakes because the pit surface is lower than the head differential makes them highly vulnerable to river avul- river. The old river channel can be partially or completely sion during high flows. Rivers have avulsed at several mines in abandoned in the avulsion process. Washington (see Table 1) in the last 14 years, most of them dur- Pits are typically wider and deeper than the river (Fig. 4A) ing storms in 1995 and 1996. and out of proportion to the overall scale of the river (Collins, 1997; Woodward-Clyde Consultants, 1980a,b). The severity Some Effects of River Diking and Channelization of the consequences of breaching pits depends on the relative Mining in depositional reaches commonly occurs on both sides scale of the mining operation and the river. Significant of a river (for example, mines on the north and south banks of amounts of material must be moved from the river bed and the East Fork Lewis River). The typical method of attempting banks to fill a deep pit. to prevent a river from flooding a pit is armoring the active Once avulsion and breaching begin, the knickpoint, where channel bank with riprap or building dikes and levees. Where the river enters the pit (Fig. 9), immediately moves upstream, bank armoring and dikes tend to minimize lateral erosion, they and the riverbed starts to be scoured (Kondolf, 1993; Kondolf may reduce the supply of sand and gravel to the active channel and others, 1996; Collins and Dunne, 1990; Collins, 1996, and prevent maintenance or creation of habitat diversity in both 1997). This scouring cuts downward and lowers the river ele- the main stem river and side channels. vation, a condition called incision. Avulsion into gravel pits by Channelization using bank armoring and dikes attempts to the Clackamas River at Clackamas, Oregon, in February of confine the river to one main channel and prevent it from mi- 1996 resulted in incision of more than 2 yards over a distance grating across the flood plain. Channelization can smooth and of one-third of a mile upstream of the pit (Kondolf and others, (or) straighten a channel, a process that reduces energy dissipa- 1996). Hazardous consequences of this kind of erosion can be tion and moves channel and bank instability and flooding undercutting of levees, bridge supports, pipelines, and utility problems (as well as increased sediment load) to downstream towers and other structures. For example, when a gravel pit on reaches. When the channel or thalweg (the deepest part of the the alluvial fan of Tujunga Wash near Los Angeles, California, channel) is moved or deflected by levees or dikes, the realign- captured the river, the knickpoint migrated upstream to under- ment may temporarily affect the downstream channel and mine nearby highway bridges (Scott, 1973). banks as the thalweg scours through salmon spawning habitats The negative effects caused by breaching a small (5 acres or or impinges on channel banks (WADFW, 1998). less) or shallow (10 ft) flood-plain pit may be negligible and Channelization can change flow velocity and the substrate, may add channel complexity to a channelized river. If the pit is which may change benthic macroinvertebrate populations on small and shallow relative to the river, it may quickly fill with which fish depend for food (Brookes, 1989). Additionally, sediment. However, even small gravel pit lakes in locations where armoring interferes with natural meander patterns, any that are prone to avulsion or near structures can be problematic. deposition will be confined to the active channel. Sediment For example, in Clark County near Vancouver where Salmon may have to be removed on a regular basis just to maintain Creek crosses I-5, an avulsion in 1996 into small (<5 acres) channel depth and flow capacity (Brookes, 1989). Gravel gravel pits caused 4 ft of knickpoint incision at a county bridge buildup in channelized rivers can result in the river being 1/4 mi upstream and has created a fish passage barrier for steel- perched above adjacent flood plains—and the adjacent gravel head, a threatened species on Salmon Creek (Ken Bates, pits—and making it prone to avulsion. WADFW, oral commun., 1998). Some levees (also called flood berms) that are used to pro- After avulsion has occurred, the temporary fluvial base tect gravel pits from flooding and avulsion are composed of level is the pit bottom. Therefore, for a considerable distance materials originally carried by the river to the pit site. The ma- upstream, the river channel tends to incise and straighten as it terials are, therefore, of a size that the river can easily transport works to establish a new equilibrium and grade. Because of again. These dikes can be washed away during a flood or this, existing channels (and salmon habitat) may be abandoned gradually made porous as fines are washed out. Through this and replaced by a deep channel as wide as the pit. Sediment washing process a levee can become a windrow of unconsoli- eventually fills the pit, and a new stream channel is formed. dated gravel that is highly vulnerable to collapse. Channel mi- While the breached pit is gradually filling with gravel from up- gration and single-event bank failures reduce the width of the stream sources, little gravel will be transported past the pit dike and weaken it. to downstream areas. The downstream channel and bars

Washington Geology, vol. 26, no. 2/3, September 1998 9 consequently will erode if they are not replenished with coarse bed material. The river, in essence, mines its own gravel bars, becomes less stable, and is inhospitable to salmon. A The time required for the river to re-establish equilibrium depends on the size and depth of the pit, the river’s ability to transport sediment, and the availability of sediment (Wood- ward-Clyde Consultants, 1980a,b). It can also depend on the instabilities caused by the headcut upstream or bedload deple- tions downstream. If the river has already eroded down to bed- rock in its upper reaches and sediment is not available, equilib- rium may not be re-established for a very long time. B knickpoint EXAMPLES OF FLOOD-PLAIN SAND AND GRAVEL MINING AND STREAM CAPTURE excavation Recent avulsion and stream capture events at gravel mines in Washington are listed in Table 1. The following paragraphs discuss some of these events. C Cowlitz River Upstream knickpoint of Toledo migration In November 1995 after heavy rains, the Kirkendoll revetment at river mile 38 on the Cowlitz River failed (Fig. 10). About a erosion deposition within pit quarter of the Cowlitz flow rushed through the opening and across the IFA Nursery, then found its way to a gravel pit near Figure 9. A. A profile of a streambed before mining. B. The same the river 3,200 ft downstream between river miles 36 and 37. profile showing a depression excavated in the stream bed when stream The mine pit partially filled with sediment and debris in a mat- flow is not strong enough to move the bedload. The knickpoint is where the stream profile abruptly changes. C. During high flows, or once avul- ter of days. Part of the river severed access to about 13 homes sion and breaching begin, the knickpoint immediately retreats up- until residents were able to construct a temporary dike of loose stream as the sediment at that location is removed by erosion and de- sand and gravel across the breached area as waters receded. posited in the pit or downstream and the river tries to re-establish its During the February 1996 (Fig. 11) event, the Cowlitz River grade and equilibrium. The high flow also attacks the downstream side breached this temporary dike and re-established itself in the of the depression where the slope also changes. (Modified from Kon- 1995 channel. During the November 1995 event, the Cowlitz dolf, 1993.) River flow was greater than 68,400 ft3/sec as measured below Mayfield dam, approximately 18 mi upstream. (This value is Zurenko, WADNR, oral commun., 1997). However, no agen- the maximum flow that can be recorded on that instrument, and cies did a survey of the elevation of the river bed before and af- no estimate was made as to amount of flow above that value.) ter the avulsion to determine if the event had caused incision. During February 1996, the flow at the Mayfield dam was not as When the February 1996 flood receded, water was diverted high, but tributaries downstream of the dam were contributing back into the pre-November 1995 channel by repairing the more water than they had in November 1995 (Stephanie dike. The scouring and downcutting between the IFA Nursery Zurenko, Washington Department of Natural Resources and the gravel pit were then evident. According to the U.S. [WADNR], oral commun., 1997). Army Corps of Engineers, complete abandonment of the 1995 Because the pits behind the Kirkendoll revetment were channel would likely have occurred if the revetment had not only 20 acres in area and less than 20 ft deep, no obvious been replaced and reinforced with riprap in the summer of 1996 changes to the grade of the river occurred. During flooding and because the existing Cowlitz River bed was higher than the filling of the pit, scouring deepened the new channel and a area protected by the revetment (Stephanie Zurenko, WADNR, knickpoint migrated upstream, as evidenced by a moving wave oral commun., 1997). Most of the material deposited in the pit immediately upstream of the pit and at the revetment (Stephanie consisted of fine sediment. The large amount of sand, silt, and soil present was probably derived from the nursery. Stumps and logs were also Table 1. Recent avulsions or stream captures deposited in the pit. Because of the de- Location or operation River Year Location Date mined Acres bris, miners now use an excavator Upstream Ridgefield pits East Fork Lewis 1995 sec. 19, T4N, R2E 1960s 6 rather than the former dragline to exca- vate gravels. The large volume of sand, Ridgefield pits East Fork Lewis 1996 secs. 13, 24, T4N, R2E 1980s-90 70 soil, and debris in the pit likely makes Salmon Creek Park ponds Salmon Creek 1996 sec. 35, T3N, R1E early 1970s 5 this gravel mine less valuable. Pits upstream of Toledo Cowlitz 1995, 1996 sec. 10, T11N, R1W ongoing 20 Gravel pits at Toledo Cowlitz 1995, 1996 secs. 8, 17, T11N, R1W ongoing 108 Cowlitz River at Toledo Mouth of Wynoochee River Wynoochee 1984 sec. 18, T17N, R7W 1960s 20 Extensive gravel mining occurs along Walker pit Yakima 1996 sec. 36, T11N, R20E ongoing 12 the Cowlitz River at Toledo (Fig. 12). Parker pit Yakima 1996 sec. 20, T12N, R19E 1980s 35 Before the mid-1930s, the course of the Selah Gap pits Yakima 1996 sec. 31, T14N, R19E ongoing 250 Cowlitz River near the gravel pits be- Gladmar Park Yakima 1996 sec. 13, T18N, R17E 1960s 30 tween river miles 34 and 35 consisted I-90 pits Yakima 1996 sec. 29, T18N, R18E 1960s 20 of two or more channels (Collins,

10 Washington Geology, vol. 26, no. 2/3, September 1998 olt River Cowlitz /sec measured at the Mayfield 3

IFA ,400 ft

revetment

Kirkendoll

Nursery end to allow fish access. lled with sediment during the November

+ he locations of the November Toledo dike

+

+

+

on at Toledo, so there is no information about

+

+

+ path

+ + scoured

+

+

+

+

+

homes

isolated

+

+

pit

+

gravel

+

+ /sec. Homes labeled are those isolated by the 1995/1996 floods. The river

+ 3

+ f the revetment. The gravel pit had fi

+

+

+

+

+

+

+

+

+

+

+

+ /sec at the dam (Luis Fuste, USGS, oral commun., 1998), where average winter flows are approxi- 3

+

+

o

+ tely 3 mi upriver. Flow is to the left. Arrows show the scoured paths of the No vember 1995 and February

o

+

o /sec (Williams, 1985). The pits at Toledo are predominantly in the 1937 and 1854 channels. Note that prior to

+

o 3

+

o

+

+ the adjacent flood plain downstream o

o

+

+ ilt by the mining company. The pit is still connected to the river at the lower

o

+

+

o

+

o

+

o

+

o

+

o

+

o

+

o

+

+

o

+

o

+

o

+

+

o

+

+

dike

o

+

o

breached

+

gravel

o

+

pit lakes

o

+

+

o

+

o

+

o

+

o

+

o

+

o

+

o

+

+

+

+

+

+

o

1854 1937 1974 1984 avulsion path 1995/6

o /sec. The USGS calculates the 100-year event flow to be 91,448 ft

o

o

o 3

o

o

+++ Vertical air photo (July 1996) of the Cowlitz River from Toledo to approxima

o o o

TOLEDO 1996 floods. The Kirkendoll revetment was designed by the U.S. Army Corps of Engineers to contain a flow of 48,000 ft failure and partial capture offlood the flow Cowlitz. at Dike the failure pit. continued The until water dike receded at after the the upper February end 1996 of event. the There gravel is pit no lake longer was a gaging rebu stati 1937, the Cowlitz River had more than one channel at the Toledo gravel pitsllins, (Co 1996). After 1937, the Cowlitz was channelized. Arrows indicate t Dam station 13 mi upstream of Toledo, but the February 1996 peak flow was less than 44,900 ft Figure 10. mately 20,000 ft 1995 flood, so there was no significant deposition there from the February 1996 flood. Peak flow during the November 1995 flood may have exceeded the 68 broke through the revetment into the 1854 channel because it was higher than

Washington Geology, vol. 26, no. 2/3, September 1998 11 Figure 11. The area of the Kirkendoll re- vetment failure on the Cowlitz River and the Kirkendoll ¬ Cowlitz River revetment breach new channel established during the Febru- ary 1996 flood. The dragline tower (to the left of the mine equipment at the center of the IFA Nursery photo) is about 60 ft high. The houses 1854 channel isolated by flood waters are in the upper left. (avulsion path) During maximum flood flows, the fields in the lower right were inundated. (Photo by Alex isolated Nagygyor, DNR Central Region, 1996). houses

1996). The Cowlitz River, as mapped in 1854 and even in 1941, had a much dragline more complicated course and flood tower screens plain than it does today (Fig. 10) (Col- flooded lins, 1996). The river is now channel- Cowlitz River gravel ized and pinned between the gravel pit pits lakes and the bluffs on which the town gravel of Toledo has been built. Mining occurs stockpiles in the 1937 and 1854 channels, the side channels, sloughs, marshes, and ox- bows along the river. During the November 1995 and February 1996 floods, the gravel pits, including stockpiles, and processing areas were inundated (Fig. 13). The dike (constructed of sand and gravel) area of protecting the gravel pits from the river dike failure failed and allowed some of the water to Cowlitz R. flow into one of the ponds close to the river. The river breached a riparian avulsion path buffer area as it exited the lower pond. left open for After the flood, sand and gravel was fish passage bulldozed by the mine operator to form gravel pit a new dike in the area of the upper breach, forcing the river into its former lakes channel. The lower breach was left open to allow fish access to the ponds, which are no longer mined. The re- mainder of the site is being mined. As at the pits upstream of Toledo, the effects of the avulsion at Toledo on the bed of the Cowlitz River were not measured, and the total volume of material moved TOLEDO was not estimated.

Yakima River at Parker Figure 12. The gravel pits near Toledo on the Cowlitz River in 1994, prior to partial avulsion into During the February 1996 flood, part of the pits closest to the river. Large arrows indicate the position of the channel during the flooding; the Yakima River avulsed through the small arrows, where the dike failed and the lower end of the pit lake left open for fish passage. gravel pits near Parker between river miles 105 and 106 on the south side of Union Gap (Fig. 14A,B). Ice jams may have played a role in stream diversions in this instance. The river entered sev- eral gravel pit ponds totaling 35 acres and averaging 10 ft deep. Mining stopped at this depth because a 100-ft thick clay layer underlies the gravel in this reach (Len Sali, Columbia Redi- Mix, oral commun., 1998). Because the surfaces of the gravel pit lakes were Figure 13. The flooded Toledo pit, February 1996. Both the 1995 and 1996 floods inundated the lower than the river level and the low- sand and gravel pits, stockpiles, and processing areas across the river from Toledo. Dams up- flow course of the river was around the stream were forced to release water at high rates to prevent the impounded water from overtop- ping, so “flood control” efficiency was lower.

12 Washington Geology, vol. 26, no. 2/3, September 1998 gravel pits, it was inevitable that the river would avulse at flood stage. A The flood breached the dikes in two places at the north end of the ponds. The miner attempted to fill the breaches with rip rap; however, the Washington Department of Fish and Wildlife stopped the operation before filling was complete because the miner had no permits. One breach was partially filled, and the other remains as it was Yakima River after it was breached during the flood. The dike openings where the channel I-82 flows into the ponds are approximately 100 ft in width. On their upstream ends, the pits have begun filling with gravel and are now about 4 ft deep. At the avulsion lower end of the mine, the river exits paths through two outlets that are about 50 ft breached wide (Len Sali, Columbia Redi-Mix, dikes oral commun., 1998). The river currently has a braided, meandering course through the ponds. In this instance, the channel has be- come more complicated, and because re-enters existing channel the ponds were shallow, no negative ef- B fects have been recorded by regulatory agencies. However, there is no moni- toring of the site. No further extraction will occur at the site, but crushing and Yakima River asphalt batching of material brought to breached dike the site is ongoing. I-82 Yakima River at Selah Gap During the flood of February 1996 at Selah Gap, ice jams, high water flow, and gravel mining combined to cause avulsion into Washington’s largest (areally) flood-plain mining area. The Figure 14. A. The Yakima River and gravel pits near Parker in 1994. A portion of the river mined area covered approximately 250 avulsed through these gravel pits between river miles 105 and 106 on the south side of Union Gap. acres, and the maximum depth of exca- Flow is to the right, and the airplane from which this photo was taken is positioned above Union vation was about 25 ft (Fig. 15). When Gap. The point of dike breaching and the new path of river and abandoned channel are indicated the dike upstream of the pit was by the lines. B. In this 1998 photo, a portion of the Yakima River is flowing through the Parker breached, the entire Yakima River flow gravel pits. Photo is taken from above Interstate Highway 82. Points of breaching are indicated by entered the gravel pits between river arrows. (Photo by Mary Ann Shawver, WADNR.) miles 118 and 120, exiting at the down- stream end of the site (Fig. 16). During this flood, the peak flow of the Yakima River was the river bed during the flood and deposited on gravel bars and 27,200 ft3/sec as measured at the Umtanum gaging station private lands just upstream of the pits. 18 mi upstream and above Roza dam. The calculated 100-year Dike building (Fig. 17) forced the Yakima River back into event at Umtanum station is 27,459 ft3/sec (Williams and Pear- its pre-February 1996 channel by September 1996. As a result, son, 1985b). Large ice jams played an undefined but likely sig- river bed disruption and incision were halted or slowed. How- nificant role in the avulsion. Approximately 8,000 ft of channel ever, because of concerns about such floods in the future, the was abandoned when the flood waters receded. mining company decided to rebuild the dikes that tightly con- Results of the avulsion are difficult to quantify. About 6 to strain the river. These dikes increase the erosive power of the 8 ft of incision occurred immediately upstream. There was lo- river further downstream. The mine operator also installed an cal knickpoint migration as evidenced by a migrating standing engineered armored spillway at a low point in the dike that al- wave and increased bank erosion as the river began to re-estab- lows the river to overtop it there and reduce its flow. This sill lish its grade (Lorraine Powell, WADNR, oral commun., keeps the bedload in the main channel and reduces the potential 1996). We estimate that at least 300,000 yd3 of gravel was for incision in this area. scoured from the river bed and deposited as a layer a minimum of 6 ft thick in the excavated pits over a 33-acre area. We also East Fork Lewis River near La Center 3 estimate that more than 100,000 yd of gravel was moved from On the East Fork Lewis River, in Clark County, intensive min- ing has occurred between river miles 8 and 9. The river has not

Washington Geology, vol. 26, no. 2/3, September 1998 13 Figure 15. The Yakima River and gravel Selah Gap pits near Selah Gap in 1994. During the flood of February 1996, breaching of the dike caused the entire flow to enter the gravel pits between river miles 118 and 120 and eventu- ally exit at the downstream end. After the flood, diking and other river training devices built by the mining company diverted the Yakima river back into its original channel. River been dammed and has one of western Washington’s few remaining wild (na- pit tive) steelhead runs. This fish has been listed as threatened on this river. lakes Gravel mine ponds on both sides of the river cover approximately 200 acres, path of average about 30 ft deep, and are sited avulsion in abandoned channels in this formerly braided river system (Fig. 18) where the river issued from a V-shaped valley I-82 Yakima River onto a broad complex flood plain. The channelafter abandoned avulsion

o

o

o

o

x

o

x

o

x

x

o

x

o

x

o

x

o

1866 x

1936 o

x

o

1958 x

o

x xxx 1972 o

o

x

x 1979 o

o

x o o o

o 1991 Yakima River x

o

x

o

x

o

x

o

x

o 1866

x flooded

o pits 1936

x

o

x

o

x

o

x

old meander o

scars x

o

x

o

x

o

x

o

o

1979 x

o

x 1866

o

x 1996 path

o

x

x avulsion I-82 o 1936

x

x

o

x

x

o

x

x

o

x

o

x

o

x

o

x

x

o

x

x

o

x

o

x

x

x

x

x

x

x

Figure 16. Selah Gap pits between Interstate Highway 82 and the Yakima River. Most of the gravel pit lakes are located in the 1866 and 1936 channels. Traces of numerous meander scars and oxbow cutoffs from earlier, natural lateral channel migrations remain, west of the river. Note the path of the February 1996 avulsion. In this September 1996 photo, dikes have returned the Yakima River to its February 1996 channel.

14 Washington Geology, vol. 26, no. 2/3, September 1998 dike

avulsion path

Figure 17. In the spring of 1996, the mining company rebuilt the dike along the Yakima River by dumping oversize rock in the area near the old gravel pits; this dike eventually rechannelized the river.

1854 x o xo o o 1937 x o o 1954 x o o o o 1963 x o xxx 1984 x x o x o avulsion path 1995/6 o o x x o o x o x o x o o o o x o o o o x o o o o x o o o x o x o o x o o o o o o o o x o o o o o x o x o o x o x x o x x o o x o x x o x x x o o oo o o o x o ox o xo x o ox o o o x o o o o o o o o o o o o o o o o o o o o o

Figure 18. Vertical air photo (1996) showing the paths of abandoned river channels from the 1995/1996 avulsions and older channel scars. The braided channel shown in 1854 and 1858 surveys is now occupied by gravel pit lakes. At that time, the river course was much more complicated (Collins, 1996) and exhibited a braided morphology. current pits occupy approximately two-thirds of the mile-wide During November 1995, the river avulsed through a gravel valley and most of the 100-year flood plain in this reach. Sev- pit pond at mile 9 (sec. 19) and abandoned about 1,700 ft of eral dike failures, avulsion events, and diversions have oc- channel (Fig. 19) and spawning gravels. During February curred along the river. 1996, the east fork flooded again. The Heissen gaging station

Washington Geology, vol. 26, no. 2/3, September 1998 15 located approximately 10 mi upstream abandoned point of avulsion of the gravel pits was washed out in this channel upper flood, but the peak flow was indirectly pit estimated at 28,600 ft3/sec by a U.S. Geological Survey (USGS) hydrolo- gist. Earlier calculations (Williams and haul Pearson, 1985a) indicated the 100-year road peak flow event at Heissen to be 20,046 ft3/sec. Since the 1996 flood event, the eventual dike failure 100-year event has been recalculated 3 river and avulsion point by the USGS at 22,200 ft /sec, using seen in Fig. 18 the 28,600 ft3/sec observation as the maximum peak flow (Sumioka and oth- pit ers, 1998). No avulsion of the east fork lakes into the lower gravel pits occurred dur- pit lakes ing this event. However, significant dike bank erosion set the stage for eventual failure path of stream capture. avulsion In November of 1996, the river avulsed through six closely spaced gravel pit ponds on the south side of the river and the river eventually aban- doned about 3,200 ft of channel and spawning gravels. Avulsion began near Figure 19. The East Fork Lewis River at flood stage in February 1996. The river had already an outside bend of the river near the avulsed through an upper gravel pit at the top of photo and later avulsed into the pits at the center haul road from the pits (Fig. 20). Lat- and bottom right of photo. Avulsion into the downstream pits began near an outside bend of the eral channel migration and the series of river located near the severely eroded haul road out of the pits. (Photo by Dan Miller, Friends of the floods severely eroded the road, which East Fork.) had acted as the dike to keep the river out of the pits. The main stem was cap- tured and now flows through the south bank gravel pits ordinary high water mark if the floodway is not mapped. When (Fig. 21). mines are proposed, the local jurisdiction becomes the Lead Some of the results of the avulsions are: Agency under provisions of the State Environmental Policy Act (SEPA). The Washington Department of Ecology (Ecol- z about 10 ft of channel downcutting as the knickpoint ogy) must also give its approval to terms of a Shoreline or Con- migrated upstream, ditional Use Permit issued by a local jurisdiction. Ecology’s z increased erosion along the south bank, 1994 Shoreline Management Guidebook recommends that lo- z abandonment of about 4,900 ft of channel where cal governments encourage miners “to locate activities outside salmon and steelhead had spawned, and the shoreline jurisdiction”— in other words, generally 200 ft from the floodway, or off the 100-year flood plain. z sluggish flow through the gravel pits. Several other agencies of secondary importance have re- The depth of downcutting was estimated as the height dif- sponsibility for regulating mining on flood plains and rivers. ference between the bed of the abandoned channel and bed of Any work, including mining, that uses, diverts, obstructs, or the current channel. The severity of the effects may be related changes natural flow or the bed of any waters of the state re- to the fact that these ponds were much deeper and wider than quires a Hydraulic Project Approval (RCW 75.20-100) from the normal river channel. We estimate that it will require more 3 the Washington Department of Fish and Wildlife (WADFW). than 2 million yd of sand and gravel, which must be derived However, because WADFW jurisdiction is restricted to waters from the channel and banks, to refill the 70-acre pits through of the state, the agency may not have jurisdiction over flood- which the river flows. plain mining if mining does not involve the active channel. Af- ter an avulsion has occurred WADFW would have jurisdiction STATE MINING REGULATIONS over any work occurring in the pits/river. The Department of Natural Resources administers the Surface Mine Reclamation The principle law regulating activity on the shorelines of the Act (RCW 78.44), which generally requires mines to be re- state is the Shoreline Management Act (RCW 90.58), which is claimed immediately after each segment is mined. The 1993 re- administered by cities and counties. Many local jurisdictions vision of this law requires that most mines in flood-plain envi- also regulate flood-plain mining under provisions of condi- ronments be reclaimed as beneficial wetlands. As part of the tional use permits or other land-use ordinances and the Growth reclamation plan for a mine, the act also requires that “where Management Act (RCW 36.70A). Several counties also have mining on flood plains or in river or stream channels is contem- mining ordinances. The Shoreline Management Act generally plated, a thoroughly documented hydrologic evaluation that applies to mining activities on the 100-year flood plain and as- will outline measures that would protect against or would miti- sociated wetlands. Regulation may vary according to the “mas- gate avulsion and erosion as determined by the department” be ter plan” of each local jurisdiction. For instance, in Lewis included. County, the Shoreline Management Act applies only to areas The U.S. Army Corps of Engineers regulates excavation or within 200 ft of the floodway of flooding that occurs with rea- filling of wetlands through section 404 of the Clean Water Act sonable regularity, although not necessarily annually, or the

16 Washington Geology, vol. 26, no. 2/3, September 1998 Figure 20. A. Before avulsion of the East Fork Lewis River. In the center of this 1990 photo are the bank and haul road that were A breached in 1996. The river is at the top right. A gravel pit lake is at the lower left. A pole and portable screen are at the left. B. After avulsion, the haul road/dike has been breached, and the river flows into parked sparpole and gravel pits. The remains of the haul road can portable screen be seen on the left across the river. “A” indi- cates same spot on each photo. The aban- A East Fork Lewis River doned channel of the river is shown in the haul road center right of photo. Approximately 10 ft of point of incision has occurred immediately upstream avulsion of where the river entered the pits. Taken February 1998 from approximately the same position as A. C. This February 1998 view shows the present position of the East Fork Lewis River; the drift boat is entering the pit former pit area. At the top right is the rem- lake nant of the old haul road. The gravels and cobbles in the foreground are the old river bed after avulsion. About 10 ft of incision has occurred here, on the basis of the difference between the height of the abandoned chan- nel thalweg and the present-day channel. B

(Title 33). Most flood-plain mines gravel stockpiles and crusher/screens would occur in wetlands or portions of wetlands and would be required to ap- haul road ply for a section 404 permit. A abandoned main stem The Growth Management Act channel (RCW 36.70A) directs counties to des- ignate “mineral resource areas” and point of “critical areas” such as wetlands or avulsion “fish and wildlife conservation areas” that include water of the state (Lingley and Jazdzewski, 1994). However, the Growth Management Act has not been fully implemented and does not apply to all counties. As wild salmon populations dwin- dle and the fish are listed as threatened and endangered species, the Endan- gered Species Act may play a role in fu- ture siting of riverine mining. It is not yet clear what that role will be. C A complete discussion of regula- tions for mining is given in Norman (1994). old haul road/ Recommendations for dike Planning and Siting gravel delta deposited Mine site selection or plans for mine in old gravel pit expansion should not be made on the basis of a perception that, because gravel mining has occurred historically in an area, it is appropriate to continue mining or that river rock is the only drift boat thalweg of going into pit abandoned channel source of construction aggregate in that market. Potential consequences of min- ® ing can be severe, and the function of the flood plain can be lost, along with fish and wildlife habitat. Wherever East Fork Lewis River possible, large gravel mines should be located in uplands away from the river

Washington Geology, vol. 26, no. 2/3, September 1998 17 valley wall

abandoned river channel F

pits flood-plain gravel pits

E

1996 avulsion path valley wall abandoned river channel D C main stem A river channel B 1995 avulsion path

Figure 21. Vertical air photo of the East Fork Lewis River showing the course of the river in November 1997. At A, the river changed its course into a gravel pit in 1995. The river re-entered its channel at B. The channel between A and B is now abandoned. C indicates where the haul road was breached and the river avulsed into the gravel pits in 1996. D marks the start of the channel abandoned in this event. E is along the avulsion path as the river finds its way through the series of ponds. The river re-enters its old channel at F. The valley walls are approximately 1 mi apart near the gravel pits. valley floors. A poor second choice is to locate mining on ter- sion in the near term. In some places, buffers may be effective races and the inactive flood plain, that is, above the 100-year because the river may not utilize the entire flood plain. How- flood plain. In Washington, upland deposits offer ample rock ever, in most instances, buffers only delay the inevitable. De- supplies. Mining these deposits eliminates potential for stream termining an adequate distance between the flood-plain mine- capture or river avulsion. Furthermore, pits in these locations pit lake and the river will depend on understanding the rate of have a good potential for successful long-term reclamation. river meandering and the risk of avulsion. In parts of Europe and the eastern United States, gravel is If flood-plain mining is approved, the main goals of plan- slowly being replaced by crushed quarry rock for use as con- ning, siting, and reclamation are: struction aggregate. In these areas, the transition is occurring n The mining should not increase the potential for river avul- mainly because limited gravel resources are nearly depleted. sion. However, substitution of crushed quarry rock for gravel has n distinct environmental advantages in that considerably more Fish and wildlife habitat should be protected. rock is produced from quarries for the surface disturbance and n Riparian areas should be protected, both to provide habitat quarries can easily be located away from flood plains and aqui- and to improve flood-plain stability. fers. The relatively small disturbance occurs because most n Reclamation should ultimately enhance salmon habitat quarries contain 100 percent usable rock, whereas gravel de- (Norman, this issue, p. 21). posits have high porosity (meaning less material per volume) and fines that are not economic. n If there is potential for migration of the river into a gravel If mine plans call for sites on flood plains, then wide, topo- pit, the site must be reclaimed in a way that is hydrologi- graphically higher, and thickly vegetated buffers should be cally compatible with the adjacent river. considered as a means of reducing the probability of river avul-

18 Washington Geology, vol. 26, no. 2/3, September 1998 Environmental Analysis Stephanie Zurenko, Rex Hapala, and Larry Dominguez, Before any new mining or expansion is allowed on a flood WADNR; Don Samuelson, Grays Harbor College; Ken Bates, plain, miners must make a rigorous environmental analysis of WADFW; Al Wald, Washington Department of Ecology; Les- the mining plan. This should include, at minimum, a geohydro- lie Lingley, U.S. Forest Service; and Beth Norman, Pierce Col- logical analysis of the affected reaches of the river system. A lege. We thank Jari Roloff for help with editing and graphics thorough plan will also include: and Keith Ikerd for drafting Figure 6. z a topographic map of the existing conditions and surrounding lands at a 2-ft contour interval and at an REFERENCES CITED appropriate scale, as well as flood profiles; Bjornn, T. C.; Reiser, D. W., 1991 Habitat requirements of salmonids z maps and cross sections that show depths and locations in streams. In Meehan, W. R., editor, Influences of forest and of all bodies of water, the stream profile, and the rangeland management on salmonid fishes and their habitats: elevation of the river bed measured from a permanent American Fisheries Society Special Publication 19, p. 83-138. station, such as a nearby bridge, near the gravel mine; Booth, D. B., 1991, Urbanization and the natural drainage system— z Impacts, solutions, and prognoses: Northwest Environmental a geomorphic analysis that identifies historic channels Journal, v. 7, no. 1, p. 93-118. and channel migration trends on the basis of Bretz, J H., 1913, Glaciation of the Puget Sound region: Washington examination of all available data, such as historic air Geological Survey Bulletin 8, 244 p., 3 plates. photos and maps, and that considers geological and Brookes, Andrew, 1989, Channelized rivers—Perspectives for envi- artificial controls of the channel, such as armored ronmental management: Wiley, 326 p. banks, dikes, bridges, dams, and other mine sites; Brown, T. G.; Hartman, G. F., 1988, Contribution of seasonally z a detailed chronology and description of historical flooded lands and minor tributaries to the production of coho precipitation, flooding, discharge, sediment transport, salmon in Carnation Creek, British Columbia: American Fisher- including description of sediment sizes in and adjacent ies Society Transactions, v. 117, no. 6, p. 546-551. to the proposed mine site; Cederholm, C. J.; Scarlett, W. J., 1981, Seasonal immigrations of ju- z maps of vegetation and analysis of its role in flood and venile salmonids into four small tributaries of the Clearwater erosion control, as well as a description of the relation River, Washington, 1977–1981. In Brannon, E. L.; Salo, E. O., between the sediment distribution and the biota, editors, Proceedings of the salmon and trout migratory behavior especially as it applies to bank erosion and avulsion; symposium: University of Washington School of Fisheries, p. 98- 110. z an analysis of avulsion or stream capture potential, Cederholm, C. J.; Scarlett, W. J., 1991, The beaded channel—A low- including the consequences of stream capture, channel cost technique for enhancing winter habitat of coho salmon. In incision, and scouring; Colt, John; White, R. J., editors, Fisheries bioengineering sympo- z an analysis of potential damage to neighboring sium: American Fisheries Society Symposium 10, p. 104-108. properties, fish and wildlife habitat, bridges, and Collins, B. D., 1996, Geomorphology and riverine gravel removal in rights-of-way and the effects of existing or proposed Washington State. In Booth, D. B., Geology and geomorphology dikes and levees and their long-term maintenance; of stream channels—Course manual: University of Washington z an analysis of channel stability, magnitude and Center for Urban Water Resources Management, p. IX-1–IX-45. frequency of the 5-, 10-, 25-, and 100-year floods, Collins, B. D., 1997, Application of geomorphology to planning and channel and flood-plain hydraulics near the proposed assessment of riverine gravel removal in Washington. In Booth, D. B., Geology and geomorphology of stream channels—Course mine site, and any previous stream capture events; flow manual: University of Washington, p. IX-1–IX-46. paths for the 5-, 10-, 25-, and 100-year floods before Collins, B. D.; Dunne, Thomas, 1990, Fluvial geomorphology and and after the mining project; river-gravel mining—A guide for planners, case studies included: z a carefully documented study of potential impacts to California Division of Mines and Geology Special Publication 98, salmon species listed under the Endangered Species 29 p. Act. Eddy, P. A., 1966, Preliminary investigation of the geology and In summary, because flood-plain gravel pits are in the ground-water resources of the lower Chehalis River valley, and dynamic riverine environment, the environmental analysis adjacent areas, Grays Harbor County, Washington: Washington should be undertaken with great care, taking into account long- Division of Water Resources Water-Supply Bulletin 30, 70 p., term stability of the site, and include proposals for enhancing 2 plates. or restoring the site’s fish and wildlife habitat. Evoy, Barbara; Holland, Mel, 1989, Surface and groundwater man- agement in surface mined-land reclamation: California Division ACKNOWLEDGMENTS of Mines and Geology Special Report 163, 39 p. Kondolf, G. M., 1993, The reclamation concept in regulation of Funding for this project was provided by the U.S. Environmen- gravel mining in California: Journal of Environmental Planning tal Protection Agency (USEPA) through the Tri-State Agree- and Management, v. 36, no. 3, p. 395-406 ment of Oregon, Idaho, and Washington to share and develop Kondolf, G. M., 1994, Environmental planning in regulation and man- technical information regarding mining. We thank Nick Ceto agement of instream gravel mining in California: Landscape and and Bill Riley (USEPA) for helpful discussions, Brian Collins Urban Planning, v. 29, no. 2-3, p. 185-199. for discussions regarding riverine mining, and the following Kondolf, G. M.; Graham Matthews, W. V., 1993, Management of for discussions, reviews, and comments: Peter Wampler and coarse sediment on regulated rivers: University of California Wa- Frank Schnitzer, Oregon Department of Geology and Mineral ter Resources Center Report 80, 128 p. Industries; Ray Lasmanis, Matt Brunengo, Lorraine Powell,

Washington Geology, vol. 26, no. 2/3, September 1998 19 Kondolf, G. M.; Vick, J. C.; Ramirez, T. M., 1996, Salmon spawning Scott, K. M., 1973, Scour and fill in Tujunga Wash—A fanhead valley habitat rehabilitations in the Merced, Tuolumne, and Stanislaus in urban southern California—1969: U.S. Geological Survey Pro- Rivers, California—An evaluation of project planning and per- fessional Paper 732-B, 29 p. formance: University of California Water Resources Center Re- Sedell, J. R.; Luchessa, K. J., 1982, Using the historical record as an port 90, 147 p. aid to salmonid habitat enhancement. In Armantrout, N. B., edi- Leopold, L. B.; Wolman, M. G.; Miller, J. P., 1964, Fluvial processes tor, Acquisition and utilization of aquatic habitat inventory infor- in geomorphology: W.H. Freeman and Company, 522 p. mation—Proceedings of a symposium held 28–30 October, 1981, Lingley, W. S., Jr.; Jazdzewski, S. P., 1994, Aspects of growth man- Portland, Oregon: American Fisheries Society, p. 210-223. agement planning for mineral resource lands: Washington Geol- Stanford, J. A.; Ward, J. V., 1988, The hyporheic habitat of river eco- ogy, v. 22, no. 2, p. 36-45. systems: Nature, v. 335, no. 6185, p. 64-66. Mackin, J. H., 1948, Concept of the graded river: Geological Society Sumioka, S. S.; Kresch, D. L.; Kasnick, K. D., 1998, Magnitude and of America Bulletin, v. 59, no. 5, p. 463-512. frequency of floods in Washington: U.S. Geological Survey Merritt, R. W.; Cummins, K. W., editors, 1984, An introduction to the Water-Resources Investigations Report 97-4277, 91 p., 1 plate. aquatic insects of North America; 2nd ed.: Kendall/Hunt Publish- Teskey, R. O.; Hinkley, T. M., 1977, Impact of water level changes on ing Co., 722 p. woody riparian and wetland communities; Volume 1, Plant and Mount, J. F., 1995, California rivers and streams—The conflict be- soil responses: U.S. Fish and Wildlife Service Biological Services tween fluvial process and land use: University of California Press, Program FWS/OBS-77/58, 30 p. 359 p. U.S. Army Corps of Engineers, 1970, Flood plain information—Ya- Mundie, J. H., 1969, Ecological implications of the diet of juvenile kima and Naches Rivers Yakima–Union Gap, Washington: U.S. coho in streams. In Northcote, T. G., editor, Symposium on Army Corps of Engineers, 57 p. salmon and trout in streams: University of British Columbia Insti- Waitt, R. B., 1979, Late Cenozoic deposits, landforms, stratigraphy, tute of Fisheries H. R. MacMillan Lectures in Fisheries, p. 135- and tectonism in Kittitas Valley, Washington: U.S. Geological 152. Survey Professional Paper 1127, 18 p. Mundorff, M. J., 1984, Glaciation in the lower Lewis River basin, Ward, J. V., 1992, Aquatic insect ecology; Volume 1, Biology and southwestern Cascade Range, Washington: Northwest Science, habitat: John Wiley & Sons, Inc., 438 p. v. 58, no. 4, p. 269-281. Washington Department of Ecology, 1994, Shoreline management Netsch, Norval; Rundquist, L. A.; Moulton, L. L. 1981, Effects of guidebook; 2nd ed.: Washington Department of Ecology Shore- floodplain gravel mining in Alaska on physical factors important land and Coastal Management Program, 93-104A, 626 p. to salmon spawning. In Washington Water Research Center, Pro- Washington Department of Fish and Wildlife, 1997, Wild salmonid ceedings from the conference, Salmon-Spawning Gravel—A Re- policy—Draft environmental impact statement, recommended al- newable Resource in the Pacific Northwest?: Washington Water ternative justification statement: Washington Department of Fish Research Center Report 39, p. 154-167. and Wildlife, 1 v. Norman, D. K., 1994, Surface mining in Washington—Regulatory re- Washington Department of Fish and Wildlife, 1998, Integrated bank sponsibilities of federal, state, and local government agencies, protection guidelines—Draft: Washington Department of Fish 1994: Washington Division of Geology and Earth Resources and Wildlife, p. 1-1–1-6. Open File Report 94-4, 26 p. Washington Department of Natural Resources, 1996, Draft habitat Norman, D. K., 1998, Reclamation of flood-plain sand and gravel pits conservation plan, March 1996: Washington Department of Natu- as off-channel salmon habitat: Washington Geology, v. 26, no. 2/ ral Resources, 1 v. 3, p. 21-28. Williams, J. R.; Pearson, H. E., 1985a, Streamflow statistics and Palmisano, J. F.; Ellis, R. H.; Kaczynski, V. W., 1993, The impact of drainage-basin characteristics for the southwestern and eastern environmental and management factors on Washington’s wild regions, Washington; Volume I, Southwestern Washington: U.S. anadromous salmon and trout; Summary and recommendations: Geological Survey Open-File Report 84-145A, 424 p., 1 plate. Washington Forest Protection Association; Washington Depart- Williams, J. R.; Pearson, H. E., 1985b, Streamflow statistics and ment of Natural Resources, 371 p. drainage-basin characteristics for the southwestern and eastern Pennak, R. W., 1978, Fresh-water invertebrates of the United States; regions, Washington; Volume II, Eastern Washington: U.S. Geo- 2nd ed.: John Wiley & Sons, Inc., 803 p. logical Survey Open-File Report 84-145B, 662 p., 1 plate. Peterson, N. P., 1980, The role of spring ponds in the winter ecology Williamson, K. J.; Bella, D. A.; Beschta, R. L.; Grant, Gordon; and natural production of coho salmon (Oncorhynchus kisutch)on Klingeman, P. C.; Li, H. W.; Nelson, P. O., 1995a, Gravel distur- the Olympic Peninsula Washington: University of Washington bance impacts on salmon habitat and stream health; Volume I— Master of Science thesis, 96 p. Summary Report: Oregon Water Resources Research Institute Peterson, N. P., 1982, Population characteristics of juvenile coho [under contract to] the Oregon Division of State Lands, 49 p. salmon (Oncorhynchus kisutch) overwintering in riverine ponds: Williamson, K. J.; Bella, D. A.; Beschta, R. L.; Grant, Gordon; Canadian Journal of Fisheries and Aquatic Sciences, v. 39, no. 9, Klingeman, P. C.; Li, H. W.; Nelson, P. O., 1995b, Gravel distur- p. 1303-1307. bance impacts on salmon habitat and stream health; Volume II— Peterson, N. P.; Reid, L. M., 1984, Wall-base channels—Their evolu- Technical background report: Oregon Water Resources Research tion, distribution, and use by juvenile coho salmon in the Clear- Institute [under contract to] the Oregon Division of State Lands, water River, Washington. In Walton, J. M.; Houston, D. B., edi- 225 p. tors, Proceedings of the Olympic wild fish conference: Peninsula Woodward-Clyde Consultants, 1980a, Gravel removal studies in Arc- College Fisheries Technology Program, p. 215-225. tic and subarctic floodplains in Alaska; Technical report: U.S. Scarlett, W. J.; Cederholm, C. J., 1984, Juvenile coho salmon fall- Fish and Wildlife Service FWS/OBS-80/08, 404 p. winter utilization of two small tributaries of the Clearwater River, Woodward-Clyde Consultants, 1980b, Gravel removal guidelines Jefferson County, Washington. In Walton, J. M.; Houston, D. B., manual for Arctic and Subarctic floodplains: U.S. Fish and Wild- editors, Proceedings of the Olympic wild fish conference: Penin- life Service FWS/OBS-80/09, 171 p. n sula College Fisheries Technology Program, p. 227-242.

20 Washington Geology, vol. 26, no. 2/3, September 1998 Reclamation of Flood-Plain Sand and Gravel Pits as Off-Channel Salmon Habitat

David K. Norman Washington Department of Natural Resources Division of Geology and Earth Resources PO Box 47007, Olympia, WA 98504-7007 e-mail: [email protected]

INTRODUCTION The success of projects that convert gravel pit ponds to off- channel habitat generally depends on the following factors: Many permitted sand and gravel mining sites located on Wash- ington’s 100-year flood plains will require reclamation. How- z good access for fish to leave and enter main river ever, even in the long term, reclamation may not restore the en- channels, tire ecologic function of the flood plain (Collins, 1996, 1997; z low risk of avulsion (stream capture), flooding, or Kondolf, 1993). Nonetheless, opportunities exist in some drought, and places to develop highly productive wetlands and off-channel z adequate cover, food supply, and water quality habitat for salmon, many species of which are struggling to (Samuelson and others, 1997). maintain strong populations. Fairly shallow, small ponds that have complex shapes and Side-channels and oxbow lakes connected to the main stem are in areas that are not frequently flooded or likely to be af- of a river are types of wallbase channels or off-channel sites fected by avulsion are the best candidates for this type of recla- that are primary overwintering habitats for juvenile coho mation. Only the edges of large, deep gravel pit lakes are pro- salmon and cutthroat trout (Cederholm and Scarlett, 1981, ductive habitat for salmon; the lack of cover and aquatic plants 1991; Peterson and Reid, 1984). Enhancing areas of reclaimed elsewhere in the water is not attractive to these fish (Norman habitat is most feasible in small (<5 acres), shallow gravel pit and others, this issue, p. 3). lakes (Fig. 1) where wallbase channels can be mimicked. Carefully prepared site plans and well-executed reclama- Large, deep gravel pit lakes may not be convertible to off- tion can enhance site stability so as to maximize the longevity channel habitat. Furthermore, the consequences to the riverine of the new habitat without the use of dikes. When preparing environment of either avulsion or maintaining isolation from mine plans, miners should consider the location, size, shape, the river with dikes are much greater at deep pits (Norman and and depth of pits, as well as local geomorphology and hydrol- others, this issue, p. 3) than at a small pit, which may be left un- ogy, in order to preserve flood-plain and river dynamics during protected. and after mining. Revegetation of the reclaimed riparian zone Wallbase channels are created on flood plains where runoff should concentrate on establishing dense growths of diverse is channeled through swales created by the migration of the main stem river. These channels can de- velop either along abandoned meanders or between the long ridges (termed scroll bars) in a large meander loop that are built at the inner edge of the low- water channel during bankfull stages. In most places, wallbase channels ap- pear as swamps, ponds, or small tribu- taries and are connected to the main stem of the river. In spring and fall, ju- venile salmon move out of the main river and into tributaries and side chan- nels to spend the following season (Cederholm and Scarlett, 1981; Peter- son and Reid, 1984). Off-channel habi- tats also provide refuge for overwinter- ing coho (Brown, 1985), chinook, and steelhead, and chum salmon spawn in these areas (Fig. 2) (Grays Harbor Col- lege and others, 1990). Efforts to en- hance some natural wallbase channels and overwintering ponds as environ- ments for juvenile coho salmon have met with success (Cederholm and oth- Figure 1. Sand and gravel pit lakes such as this one can be reclaimed as highly beneficial fish ers, 1988; Cederholm and Scarlett, and wildlife habitat with riparian zones. When connected to rivers and streams, these kinds of 1991; Peterson, 1985). ponds can provide essential off-channel habitat for juvenile salmon. Placing abundant large woody debris in these ponds can create the complex structures needed for salmon habitat.

Washington Geology, vol. 26, no. 2/3, September 1998 21 the slow currents. They also eat copepods (small crustaceans) and oligochaetes (such as earthworms) that live on or in the pond floor. Because these juvenile salmon migrants have grown larger in these channels than their contemporaries in other environments, their chances of survival are improved.

Pond Shapes and Depths Gravel pit lakes with a regular shape (chiefly rectangular), steep slopes (1.5 horizontal:1 vertical), and uniform depth of- fer little habitat complexity for fish and wildlife (Kondolf, 1993). A productive gravel pit lake connected to the main stem river can result if the mine operator constructs or reshapes ponds so that they have irregular shorelines and a variety of depths (Fig. 3). An arrangement of gravel pits that benefits salmon is a se- ries of connected irregular ponds that have islands and penin- Figure 2. Chum salmon migrating to spawning areas. Studies of sulas (Fig. 4). The shoreline length is maximized, and the con- ponds such as those at the Weyco-Briscoe facility in Grays Harbor nections between ponds provide channels with riffles, thus in- County have shown that chum spawn in gravel banks where ground wa- creasing the dissolved oxygen content and adding complexity. ter wells up. (Washington Department of Fish and Wildlife photo.) Gradually sloped, complex reclaimed shoreline margins help promote aquatic plant development. If the bottom of the pond is irregular, it will offer a range of habitats for plants, vegetation. These plants will create a zone of soils bound by benthic fauna, and fish. roots that will slow erosion and reduce floodwater velocities. Optimal slopes in shallows intended as fish habitat are flat- To begin rehabilitating the function of the riparian zone and tened to at least 5H:1. Slopes of 15H:1V (3 ft water depth at associated wetlands, revegetation plans should take into ac- 45 ft off shore) would provide a broad rim with the potential for count the potential for attracting a wide variety of aquatic and diverse aquatic plant growth. Where a shoreline slopes steeply, terrestrial wildlife, including amphibians and reptiles. Aquatic marsh conditions and space for aquatic plants are confined to a vegetation as cover and environments that promote insect narrow band (Andrews and Kinsman, 1990). populations for food will benefit the salmon. Approaches to Maintaining approximately 25 percent of the lake as creating or restoring wetlands in gravel pits are reviewed by benches and bars less than 2 ft deep, 25 percent as areas 2 to 6 ft Norman and Lingley (1992), Michalski and others (1987), deep, and half the lake more than 10 ft deep provides the com- Prange (1992), Norman and others (1996), and Stevens and plexity that enhances biotic diversity (Norman and Lingley, Vanbianchi (1993). 1992; Norman and others, 1996). For fish and wildlife habitat, This paper focuses on principles and techniques to be con- the higher the percentage of shallow areas, the better the poten- sidered in establishing salmon habitat in reclaimed gravel pit tial for both food production and shelter in aquatic plants and lakes and describes some successful examples. woody debris that has fallen in the pond. Some of the shallow gravel shelves along a pit lake perime- DESIGN OF RECLAIMED PONDS FOR OFF- ter will be places where ground-water flow is intercepted. Up- CHANNEL SALMON HABITAT welling generally occurs along the hydrologic upgradient of ponds. This water is typically well oxygenated, and chum Benefits salmon are known to choose these ground-water-fed shallows Gravel pit lakes can be reclaimed to imitate natural oxbows and (<2 ft deep) for spawning (Grays Harbor College Research, cutoff channels, commonly already present on a flood plain, 1990). that are connected to the main channel (Collins, 1997; Norman Wallbase channel gravel pits should be designed to mini- and others, this issue, p. 3). Off-channel habitat offers numer- mize chances or consequences of avulsion. This can be ous benefits to salmon. Juvenile salmon will migrate into achieved by keeping wallbase channel ponds shallow, small wallbase channels to seek shelter from strong currents and tur- (<5 acre), longer than wide, parallel to the river, and away from bid water during freshets in April and May and the first freshets in the fall. Young salmon typically feed in the shallows and hide in water deeper than 6 ft or in complexes of woody debris in these channels. A few modifications, egress channel such as adding woody debris or aquatic to river very shallow plants, to a reclaimed pond design will intermittent stream greatly increase the ponds’ utility for salmon. This created habitat would be similar to natural wallbase channels. While in these channels, salmon take advantage of the rich food supply Figure 3. The shorelines of ponds used for wildlife habitat should be irregular and planted for of aquatic insects, for example, chiro- cover with a mixture of open meadows and shrubs in the surrounding area. The shape of the pond nomid larvae, that live on the abundant on the left is better suited to supporting wildlife than that of the pond on the right. (Modified from aquatic and riparian vegetation and in Szafoni, 1982.)

22 Washington Geology, vol. 26, no. 2/3, September 1998 actively meandering channels (Fig. 4). Habitat ponds must be A¢ connected with the river on the downstream side (Norman and old gravel pit riffles valley wall others, 1996) for optimal fish access. A channel connected on the upstream end can offer a route for river avulsion. A

Substrates buffer outlet channel Off-channel habitat should be designed for a variety of uses by preserving gravel substrates that could be used as spawning ar- eas and by placing soil to accommodate aquatic plant growth. active flood plain Topsoil in shallow water facilitates revegetation and speeds re- establishment of a food chain. Soils placed in the shallows pro- vide a mud bottom that is generally a nutrient-rich substrate for bacteria, plants, and benthic macroinvertebrates. If these food PLAN VIEW sources and sheltered areas are plentiful, young salmon have a better chance of reaching maturity. A A¢ Choice spawning substrates generally have a combination marsh/lowland of round gravels ranging from 0.25 to 3 in. in diameter and island shallow area moving water. Suitable spawning areas may also be provided by (a) springs welling from the pond bottom or sides where the deep region ground-water table has been intercepted by the excavation or (b) those parts of the channels connecting ponds where water is CROSS SECTION moving across gravels. In addition, coarse gravels and cobbles are home to preferred salmon food sources such as immature Figure 4. Plan view and cross section of a reclaimed gravel pit with a stages of caddis flies and mayflies. Juvenile fish also hide in pond shape that mimics a natural river system. Not to scale. (Modified coarse gravel. Most of the pit lake floors and channel banks from Woodward-Clyde, 1980.) should be covered with soil, which will be the most productive substrate for vegetation and insects. Miners are required by Washington law (RCW 78.44) to should be carefully protected and moved into the pond during save and replace all topsoil and sediments within 4 ft of the sur- the reclamation process. Many aquatic plants grow and propa- face and to practice segmental reclamation. Topsoil should be gate themselves vigorously. The initial introduction need not directly replaced from one area of a mine to another without a be vast, and early-planted material may be thinned for planting period of storage. Wherever possible, 12 to 18 in. of the “A ho- during later reclamation stages. Planting times will vary, de- rizon” topsoil should be placed on banks and islands and in all pending on the requirements for successful establishment of shallow areas of the new pond to the depth that light penetrates each species. through the water. Upland Vegetation Aquatic Vegetation Planting riparian areas with native tree species (for example, A network of aquatic plants shelters fish from predators. The cottonwood, poplar, alder, willow, fir, spruce, cedar, pine, ma- soil placed on the lake shore or in shallow water promotes ple) and native grasses, sedges, legumes, and forbs can acceler- growth of sedges, reeds, lily pads, and other aquatic vegetation. ate the process of providing productive habitat, create diver- Plants also are essential to insect populations. The surface sity, and help stabilize the site. Deciduous trees and shrubs, es- area of plants may be many times greater than that of the pond pecially willow and alder, are valuable sources of food for or lake bed; hence plants may considerably increase the insects because their submerged, fallen leaves rapidly degrade amount of surface area that insects can colonize. Many insect and develop a microfauna of bacteria, fungi, and protozoa (An- species lay their eggs on plants. Dead and decaying plants are drews and Kinsman, 1990). also a food resource for insects (Andrews and Kinsman, 1990). Native plants should always be used for this kind of revege- Juvenile coho salmon feed on insects of the family Chiro- tation, preferably those species that grow close to the site. nomidae (non-biting midges) as the young insects emerge Planting willow, osier dogwood, poplar, and cottonwood cut- through the shallow water column (Peterson, 1982a,b). Chiro- tings is an effective and fairly quick method of building a root nomid species colonize every type of freshwater substrate. matrix that can slow erosion. Planting should take place in the However, population densities in sand are generally very low, spring or fall, preferably while plants are in the dormant stage and gravel and rock usually support high densities only where (Norman and others, 1996). well covered with algae. By contrast, organic sediment and plant surfaces can support high chironomid population densi- Buffers ties. To reclaim a mine site as off-channel habitat, miners should Aquatic plants can help slow or prevent erosion on a shore- plan and maintain areas of undisturbed riparian vegetation as line that is subject to wave erosion by absorbing moderate setbacks from rivers. Wide buffers delay a river’s entry into a wave energy. Exotic plant varieties rarely provide the same pit by giving the river room to migrate and to disperse floods. benefits as natives. Aggressive native species, such as common Buffers of well-rooted vegetation on banks and in the riparian cattail and Douglas’ spiraea, should be used cautiously or zone slow erosion and will extend the time that the gravel pit avoided completely because they tend to become a virtual mo- off-channel habitat exists. Highly vegetated buffers can be noculture. seed sources of plants, which will speed the revegetation pro- If there are wetland areas within the site before gravel pit cess. excavation starts, then the aquatic plants and muck soil layer

Washington Geology, vol. 26, no. 2/3, September 1998 23 Calculating an appropriate buffer width may be difficult. In general, the width selected should be based on stream and valley morphology and the rate at which the river migrates. Buffer widths should be adequate to protect or restore full function to both the river and the pits. The Washington Depart- ment of Fish and Wildlife (WADFW) recommends that the riparian habitat area be at least as wide as the 100-year flood plain or 250 ft, whichever is greater (Knutson and Naef, 1997). When designing buffer revegeta- tion, miners should consider the orien- tation of the area relative to the sun. For instance, a north-facing slope receives less sun than a south-facing slope and is likely to be easier to revegetate because it dries more slowly and remains Figure 5. Logs with roots and crowns attached can be lashed together and placed in deep cooler. (@20 ft) water. They can be held in place by dumping overburden, rocks, or sand bags on top to Livestock should be excluded from form reef habitat for fish and aquatic insects. Reefs like this are better constructed of conifers than the reclaimed ponds and riparian areas of deciduous trees, which may rot too quickly. because they can trample and overgraze vegetation and adversely change the water quality. the bar is stable. Connections to rivers should not be straight or constructed where a stream is depositing sediment. Coarse Woody Debris and Shelter For fish, the ideal entrance to a pond is located adjacent to a Overwintering success for juvenile salmon in off-channel ar- main stem channel irregularity that creates a large to moderate eas can be greatly enhanced by adding large woody debris, eddy; fish can recognize the plume of clear water flowing from which provides cover. Side channels that contain such debris the pond into the main river. Connection gradients should be had more coho than channels with no woody debris (Sedell and gentle (0–1%). Logs can be laid in the connection channel to others, 1984; Tschaplinski and Hartman, 1983). Trees need not provide “steps” to lower the gradient and create pools in which completely surround a flood-plain pond or completely line a fish can rest. channel, but the more enclosed the reclaimed lake, the better Riffles or fast water areas are less desirable outlet sites be- the habitat. Eventually, large woody vegetation along the cause fish do not spend much time there and may not find the shoreline will fall into the pond and provide further habitat for outlet. Very shallow outlets may be left high and dry during fish. If none is present, woody debris should be placed on the low water. The designed ingress/egress channel should be deep shore and at a variety of depths. enough to maintain year-round flow. Connecting the ponds to Logs and stumps can be lashed together to form debris jams the river on the upstream section of the mine should be avoided in the wallbase gravel pits ponds. They may be anchored by because this may encourage the river to avulse into the pit at an rocks, overburden, or soils. In water 20 ft deep, logs and accelerated rate. stumps form reef habitat (Fig. 5) for fish and aquatic insects. Miners need to obtain a Hydraulic Project Approval issued Stumps with their intricate masses of rootlets still attached by the Washington Department of Fish and Wildlife if mining (root wads) provide cover and can be easily and cheaply added plans involve a constructed connection to a river or smaller to ponds. Large boulders can also be used to build reefs (Nor- stream. man and Lingley, 1992; Norman and others, 1996). Submerged and anchored tree crowns provide excellent EXAMPLES OF OFF-CHANNEL HABITAT cover along steep banks (Norman and others, 1996). Trees CREATED BY RECLAMATION most suitable for aquatic habitat restoration of this kind are ce- Weyco-Briscoe Ponds on the Wynoochee River dar and fir. Alder can be used if no other trees are available; however, deciduous trees generally decompose too quickly. Many gravel pits are located along the Wynoochee River, which drains the Olympic Mountains (Fig. 6).The Weyco- Connection to the River Briscoe ponds, in sec. 22, T19N, R8W, river mile 17, are exam- ples of small ponds excavated for gravel that have been con- If the reclaimed flood-plain mine is near the active river chan- verted to off-channel habitat (Fig. 7). Using natural pond char- nel, not deeper than the deepest part of the adjacent river, not acteristics identified by Peterson (1985) as a guide, the com- subject to drying up, and not susceptible to flooding, ponds can pany constructed ponds that provide shallow areas for food be connected to the river at the downstream end. These outlet production, deep water for cover, and few perching areas for connections (ingress/egress channels) allow fish to enter and predatory birds. Because of the location, size, shape, and depth leave the off-channel ponds. Where feasible, connections to a of the ponds, avulsion or stream capture is not likely to cause river should be made by way of natural wallbase channels or drastic environmental changes to the river (Collins, 1997). sloughs that enter the main river and flow near the ponds. Other Six ponds were created over a period of seven years during natural features that mimic the natural riverine system, such as the 1980s. All are generally less than 20 ft deep, approximately side channels along gravel bars, can be effective connections if

24 Washington Geology, vol. 26, no. 2/3, September 1998 U405 u2 0.5 to 1.5 acres in area, and connected Humptulips River North River Bend Wenatchee to each other (Partee and Samuelson, pit U90 28

1993). On the downstream end, the 97 AII OCEANPACIFIC u Tacoma Cle Elum ponds join the river to allow fish pas- Weyco- u101 Humptulips Wynoochee 101 Brisco sage. pit Aberdeen The ponds have been monitored Chehalis Olympia 47o Yakima since 1989 by Grays Harbor College to 90 u12 determine the extent of chum salmon River Centralia spawning and use by other salmonid ju- Chehalis 101 u12 Edler veniles for overwintering. Monitoring U5 Morton Yakima pit River of the site in 1990 indicated that the av- Raymond River erage immigrant coho fingerling was Cowlitz 4 49.6 mm long and weighed 1.20 g and 504 0 20 mi U82 0 30 km 22 the average immigrant chinook was Kelso Columbia River 66.52 mm long and weighed 2.36 g. In Longview u97 124o contrast, the average Wynoochee River o 46o 123 o coho fingerling was 30.38 mm long and 122 121o 120o weighed 0.38 g, while the average chi- nook was 41.25 mm long and weighed Figure 6. Location of gravel pits near rivers discussed in this article that have established off- 1.30 g. Immigrants into the ponds in channel fish habitat. 1990 included 1,249 coho and 146 chi- nook. Only 2 chum were recorded im- migrating into the ponds in 1990. How- ever, there were 73 juvenile chum emi- grants that had spawned in the Weyco- Briscoe ponds (Grays Harbor College and others, 1990). Initially, minor amounts of woody debris were placed in the ponds; how- ever, Don Samuelson of Grays Harbor College considered this sparse cover to 150 x 300 ft be a factor that limited use by and sur- vival of salmon (Samuelson and Willis, 1995). In 1995, tons of woody debris were hauled to the site and placed in the ponds (Fig. 8). The ponds are being monitored to determine the results. connecting From November through December of channel 1997, 31 chum adults and 66 coho juve- (ingress/ niles immigrated into the ponds. Only 8 egress) flow juveniles emigrated during this time period. Seven steelhead juveniles im- migrated, and none left. Fish biologists also counted the carcasses of 69 adult chum in the ponds (Samuelson and oth- ers, 1997). Figure 7. The Weyco-Briscoe ponds on the Wynoochee River (Fig. 6) in 1994. Mining and recla- mation were designed to enhance off-channel habitat. An existing creek (at arrow) connects the ponds to the river. Avulsion is likely in the near future because the river has shifted closer to the Gravel Pits on the ponds. Stream capture will not have major environmental consequences because the river chan- Humptulips River nel and pond depths and widths are similar. During the 1980s and early 1990s, gravel mining took place along the Humptulips River, which spawning and off-channel habitat. No quantitative monitoring drains the Olympic Mountains (secs. 8 and 17, T20N, R10W), of the site has been done, but Jack Ljestfield (WADFW, oral about 1/4 mi upstream of the U.S. Highway 101 bridge (Fig. 6). commun., 1994) believes the ponds support both salmon The gravel pits are no deeper than 30 ft and consist of about 20 spawning and off-channel habitat for juveniles. Adult salmon acres of ponds (Fig. 9). Connections to the river were created are readily observed in the ingress/egress channels. by the mining company after mining had ceased. No woody de- bris was placed in the ponds, so there is minimal cover. The in- Yakima River at Union Gap gress/egress channel to one of the ponds dries up in summer. The Edler gravel ponds operated by Len Sali of Columbia However, a well-vegetated riparian zone is now established Ready Mix, Inc., near the town of Union Gap are currently be- along much of the ponds’ perimeter, and salmon usage is gen- ing mined with the intent of providing fish and wildlife habitat erally high during the winter. No further reclamation is as a subsequent use. The ponds are located approximately planned. 500 ft west of the Yakima River (Fig. 10). The reclamation plan A local salmon club now uses one of the ponds for a rearing is to finish mining with four ponds no deeper than 25 ft and pen. The other ponds are unmanaged and left open for salmon covering an approximate total of 30 acres. Individual ponds

Washington Geology, vol. 26, no. 2/3, September 1998 25 will be connected to each other by sinu- ous channels as mining is completed, mimicking natural conditions. The first pond that was mined has been re- claimed (Fig. 10) and is no deeper than 18 ft. The southernmost pond, the last to be excavated, will be connected at fi- nal reclamation to the Yakima River by a channel that will provide fish access to the pond system. All the ponds will be excavated to final slopes of 5:1 at the water edge. Where natural vegetation (trees and brush) exists adjacent to the excava- tion, Sali has selectively trenched to construct islands with established vegetation, which provides a seed source and increases the speed of final reclamation. There is deeper water in predator trenches developed around is- lands. Slopes above water will be no Figure 8. Rootwads, treetops, and slash supplied by Weyerhauser Company in a Weyco- steeper than 3:1 and have the topsoil Briscoe pond. Grays Harbor College received funds through the Watershed Restoration Project respread and grasses and native plants from Washington Departments of Natural Resources and Fish and Wildlife for the project. Dan re-established. Topsoil stripped from Brock and his crew from Northwest Upland Restoration used an excavator to place large logs in the second pond area was replaced the ponds. Wood pieces less than 10 in. in diameter were bundled, towed with a small motor boat around the first pond mined, as re- or by hand, submerged, and held in place by rocks and sandbags. This debris provides complex quired in the segmental reclamation macro- and microhabitats for cover, feeding, and shade. (Photo by Don Samuelson, Grays Harbor plan. All four ponds will ultimately College.) have irregular shorelines and varied depths. Approximately 25 percent of the ponds’ complex shoreline has water 2 ft deep, following the recommenda- tions of regulatory agencies (Norman and Lingley, 1992; Norman and others, 1996). Avulsion occurred in 1971 at gravel pits immediately downstream of the Edler site (Dunne and others, 1981; Dunne and Leopold, 1978). If avulsion or stream capture occurs at the Edler pits, the off-channel habitat would be lost, but the mining company and regu- latory agencies anticipate few negative impacts on the Yakima River because the pit lakes emulate abandoned chan- nels and are not overly wide or deep relative to the river. While the Washington Department of Fish and Wildlife specialists expect that fish and wildlife use of the ponds will be high, the agency has no plans at this time to monitor the effectiveness of the reclamation or the plan that is being Figure 9. Mining operations in the late 1980s along the Humptulips River. Gravel removal opera- implemented. tions have now ceased. Flow is toward the top of the photo. Turbid water (the white reflection in the photo) is seeping though coarse gravels from the active mining pond into a downstream pond. In RECOMMENDATIONS some ponds upwelling spring water (arrow) provides oxygen, which creates favorable spawning areas along gravel banks. (Photo by Jack Ljestfield, WADFW.) Digging more ponds for off-channel habitat may be counterproductive in some reaches, and the ponds are not likely to outperform the natural system. Creation of off- Natural Resources reclamation specialists recommend that channel habitat for salmon should be firmly coupled with plans government agencies, tribes, academia, and industry collabo- for long-term monitoring to determine its effectiveness. Cur- rate in monitoring existing and future sites. Sites should be rently, the only gravel pit lakes that are monitored for off- evaluated by a multidisciplinary team to determine the success channel habitat are the Weyco-Briscoe ponds. Department of of planned habitat and thereby improve future efforts.

26 Washington Geology, vol. 26, no. 2/3, September 1998 ACKNOWLEDGMENTS Funding for this project was provided by the U.S. Environmental Protection Agency (USEPA) through the Tri-State Agreement of Oregon, Idaho, and Washington to share and develop tech- nical information regarding mining. I thank Nick Ceto and Bill Riley (EPA) and Jack Ljestfield (WADFW) for helpful discussions. I thank the follow- ing for discussions and their reviews and comments: Peter Wampler and Frank Schnitzer, Oregon Department of Geology and Mineral Industries; Ray Lasmanis, Lorraine Powell, Stephanie Zurenko, Rex Hapala, Jeff Cederholm, Larry Dominguez, Wash- ington Department of Natural Re- sources; Don Samuelson, Grays Harbor College, Ken Bates (WADFW); Al Figure 10. A pond in a part of the Edler sand and gravel pit near Union Gap south of Yakima Wald, Washington Department of shortly after reclamation and planting with native species (chiefly willows and cottonwood). The is- Ecology; and Beth Norman, Pierce Col- land in the background is in the center of the lake. Trees in the far background are on a reclaimed lege. I thank Jari Roloff for help with shoreline and are part of an undisturbed riparian zone. Shoreline slopes are less than 5:1. The 2- editing and graphics and Keith Ikerd to 5-acre pits at this site were designed with islands and shallow shorelines to enhance fish and for drafting Figure 6. wildlife habitat. The miner replaced topsoil in and around the ponds to promote revegetation. The last pond to be mined will be connected to the Yakima River by a 20-ft-wide channel. SELECTED REFERENCES Andrews, John; Kinsman, D. J. J., 1990, Gravel pit restoration for Grays Harbor College Research (Samuelson, Don; Phipps, James; wildlife—A practical manual: Royal Society for the Protection of Phillipi, Scott; Wargo, Lorna; Willis, Elizabeth; MacMillan, Nor- Birds, 184 p. bert; Cornell, David; Mikkelsen, Nels), 1990, Physical, chemical, Brown, T. G., 1985, The role of abandoned stream channels as over- and biological characteristics of Weyco-Briscoe gravel pit ponds wintering habitat for juvenile salmonids: University of British (July 1989–June 1990); Annual report: Grays Harbor College [un- Columbia Master of Science thesis, 134 p. der contract to] Grays Harbor Planning and Building Department, Cederholm, C. J.; Scarlett, W. J., 1981, Seasonal immigrations of ju- 1v. venile salmonids into four small tributaries of the Clearwater Knutson, K. L.; Naef, V. L., 1997, Management recommendations for River, Washington, 1977–1981. In Brannon, E. L.; Salo, E. O., Washington’s priority habitats—Riparian: Washington Depart- editors, Proceedings of the salmon and trout migratory behavior ment of Fish and Wildlife, 181 p. symposium: University of Washington School of Fisheries, p. 98- Kondolf, G. M., 1993, The reclamation concept in regulation of 110. gravel mining in California: Journal of Environmental Planning Cederholm, C. J.; Scarlett, W. J., 1991, The beaded channel—A low- and Management, v. 36, no. 3, p. 395-406 cost technique for enhancing winter habitat of coho salmon. In Michalski, M. F. P.; Gregory, D. R.; Usher, A. J., 1987, Rehabilitation Colt, John; White, R. J., editors, Fisheries bioengineering sympo- of pits and quarries for fish and wildlife: Ontario Ministry of sium: American Fisheries Society Symposium 10, p. 104-108. Natural Resources, Aggregate Resources Division, 59 p. Cederholm, C. J.; Scarlett, W. J.; Peterson, N. P., 1988, Low-cost en- Norman, D. K.; Cederholm, C. J.; Lingley, W. S, 1998, Flood plains, hancement technique for winter habitat of juvenile coho salmon: salmon habitat, and sand and gravel mining: Washington Geol- North American Journal of Fisheries Management, v. 8, p. 438- ogy, v. 26, no. 2/3, p. 3-20. 441. Norman, D. K.; Lingley, W. S., Jr., 1992, Reclamation of sand and Collins, B. D., 1996, Geomorphology and riverine gravel removal in gravel mines: Washington Geology, v. 20, no. 3, p. 20-31. Washington State. In Booth, D. B., Geology and geomorphology Norman, D. K.; Wampler, P. J.; Throop, A. H.; Schnitzer, E. F.; of stream channels—Course manual: University of Washington Roloff, J. M., 1996, Best management practices for reclaiming Center for Urban Water Resources Management, p. IX-1–IX-45. surface mines in Washington and Oregon: Washington Division Collins, B. D., 1997, Application of geomorphology to planning and of Geology and Earth Resources Open File Report 96-2, 1 v. assessment or riverine gravel removal in Washington. In Booth, Partee, R. R.; Samuelson, D. F., 1993, Weyco-Briscoe ponds habitat D. B., Geology and geomorphology of stream channels—Course enhancement design criteria: Grays Harbor College, 1 v. manual: University of Washington, p. IX-1–IX-46. Peterson, N. P., 1982a, Immigration of juvenile coho salmon (On- Dunne, Thomas; Dietrich, W. E.; Humphrey, N. F.; Tubbs, D. W., corhynchus kisutch) into riverine ponds: Canadian Journal of 1981, Geologic and geomorphic implications for gravel supply. In Fisheries and Aquatic Sciences, v. 39, no. 9, p. 1308-1310. Cassidy, J. J., prefacer, Proceedings from the conference on Peterson, N. P., 1982b Population characteristics of juvenile coho salmon-spawning gravel—A renewable resource in the Pacific salmon (Oncorhynchus kisutch) overwintering in riverine ponds: Northwest?: Washington Water Research Center Report 39, p. 75- Canadian Journal of Fisheries and Aquatic Sciences, v. 39, no. 9, 100. p. 1303-1307. Dunne, Thomas; Leopold, L. B., 1978, Water in environmental plan- ning: W.H. Freeman and Company, 818 p.

Washington Geology, vol. 26, no. 2/3, September 1998 27 Peterson, N. P., 1985, Riverine pond enhancement project, October Sedell, J. R.; Yuska, J. E.; Speaker, R. W., 1984, Habitats and salmo- 1982–December 1983: Washington Department of Fisheries Pro- nid distribution in pristine, sediment-rich river valley systems— gress report 233; Department of Natural Resources, 48 p. S. Fork Hoh and Queets River, Olympic National Park. In Mee- Peterson, N. P.; Reid, L. M., 1984, Wall-base channels—Their evolu- han, W. R.; Merrell, T. R., Jr.; Hanely, T. A., editors, Fish and tion, distribution, and use by juvenile coho salmon in the Clear- wildlife relationships in old-growth forests: American Institute of water River, Washington. In Walton, J. M.; Houston, D. B., Fishery Research Biologists, p. 33-46. editors, Proceedings of the Olympic wild fish conference: Penin- Stevens, M. L.; Vanbianchi, Ron, 1993, Restoring wetlands in Wash- sula College Fisheries Technology Program, p. 215-225. ington—A guidebook for wetland restoration, planning and im- Prange, B. B., 1992, Guidelines and mitigation potential for gravel-pit plementation: Washington Department of Ecology Publication wetland creation: Evergreen State College Master of Environ- 93-17, 1 v. mental Studies essay [thesis], 82 p. Szafoni, R. E., 1982, Wildlife considerations in the development of ri- Samuelson, Don; Harrison, Jim; Brooks, David; Keiran, T. J.; Craig, parian communities—Vegetation, habitats, ecology, manage- Ben; England, Amber; Ambrose, Doug; Arndt, Wrndt; Brumfield, ment. In Svadarsky proceedings; University of Minnesota Agri- Scott; Evans, Scott, 1997, Weyerhaeuser-Briscoe off-channel cultural Experiment Station Miscellaneous Publication 17, p. 59- overwintering ponds, salmonid habitat restoration project; Final 66. Report: Grays Harbor College [under contract to] U.S. Fish and Tschaplinski, P. J.; Hartman, G. F., 1983, Winter distribution of juve- Wildlife Service, 1 v. nile coho salmon (Oncorhynchus kisutch) before and after log- Samuelson, Don; Willis, Elizabeth, 1995, Weyerhaeuser-Briscoe off- ging in Carnation Creek, British Columbia, and some implica- channel overwintering ponds, salmonid habitat restoration proj- tions for overwinter survival: Canadian Journal of Fisheries and ect, Research report: Grays Harbor College [under contract] to Aquatic Science, v. 40, no. 4, p. 452-461. Washington Department of Natural Resources and Washington Woodward-Clyde Consultants, 1980, Gravel removal guidelines Department of Fish and Wildlife, 1 v. manual for Arctic and Subarctic floodplains: U.S. Fish and Wild- life Service FWS/OBS-80/09, 171 p. n

BOOK REVIEW: Collecting the Natural World—Legal requirements and personal liability for collecting plants, animals, rocks, minerals, and fossils

by Donald Wolbery and Patsy Reinard regulations and contacts. The office urges a conservation Geoscience Press, Inc. and preservation ethic: Take only photos and try not to PO Box 42948, Tucson, AZ 85733 leave footprints. 1997. $24.00 n In Washington state, no permits are required at this time for his book presents summaries of federal and state laws re- fossil or mineral collecting on lands managed by the De- Tgarding collecting archaeological, botanical, biological, partment of Natural Resources (DNR), but as a courtesy, and geological items. Few entries are more than a page long, contact Ellis Vonheeder at DNR’s Resource Planning and but the essence of each regulation is set forth. Federal land Asset Management Division, Materials Management Sec- ownership and management are briefly described. The authors tion (360-902-1618). evidently recommend cautious use of their book. The dis- n The Washington Department of Fish and Wildlife requires claimer in small print on the back of the title page states a permit for collecting on land it administers. “This book is an interpretation of relevant state and federal n statutes, a subject which is a constantly moving target. There- Washington law (RCW 79.01.748) states: “Every person fore, it is entirely possible that a particular law has been missed who willfully commits any trespass upon public lands of or misinterpreted or has changed...the reader is directed to con- the state and...digs, quarries, mines, takes, or removes tact the agencies listed for more detailed information.” therefrom any earth, soil, stone, mineral, clay, sand, gravel, Several appendices take up more than half of this 330-page or any valuable materials, shall be guilty of larceny.” book. These cover state symbols, fossils, flora, and fauna; ad- A general rule then is: To avoid trespassing, get permission dresses for various agencies; lists of national parks and monu- and up-to-date information about regulations from the land- ments and when they were established; and the texts of the An- owner (federal, state, or private) before you set out. The legal tiquities Act of 1906, Historic Sites Act of 1935, Archaeologi- complexities and case law call for keeping current. cal Resource Protection Act of 1979, and Federal Endangered Despite the fact that this is a 1997 book, readers should be Species Act of 1973 (with list of species). One of the appendi- aware that there are typos and slips. Further, addresses given in ces is titled “Using basic instruments [tools], maps, and the appendices may be out of date. For example, Washington’s leases”. In this section, the authors provide a sample lease form Historic Preservation Officer can now be reached at PO Box that might be used to arrange access to private lands. But they 48343, Olympia, WA 98504-8343; the office is in Lacey, also recommend getting thorough legal advice about leases to Wash. There is no longer a DNR Division of Lands and Miner- avoid situations similar to what followed discovery of the infa- als. It has become the Resource Planning and Asset Manage- mous “Sue” tyrannosaurus. Paleontologists should be aware ment Division, PO Box 47014, Olympia, WA 98504-7014. that at least 13 acts now affect collecting on federal lands, and Look in your local phone book for information about locations other legislation is pending. of federal or local government offices as well. When contacted in July 1998, state offices had a somewhat This book is only an introduction to the fuzzy world of le- different message for persons interested in collecting: galities in collecting. It might be a useful (though flawed) ref- n Washington’s State Archaeologist Rob Whitlam encour- erence book for libraries, but it is not a comprehensive guide. ages potential collectors to go to the Internet for current The disclaimer says it all. Caveat emptor. Kitty Reed

28 Washington Geology, vol. 26, no. 2/3, September 1998 Innovations and Trends in Reclamation of Metal- Mine Tailings in Washington

David K. Norman Robert L. Raforth Washington Department of Natural Resources Washington Department of Ecology Division of Geology and Earth Resources Central Region Water Quality Program PO Box 47007, Olympia, WA 98504-7007 15 W. Yakima Ave., Yakima, WA 98902-3387 [email protected] [email protected]

INTRODUCTION ers (1992), Smith and Mudder (1991), and Norman and Raforth (1995). Tailings are mine mill waste products consisting of ground To minimize toxicity of tailings, mine planners must con- rock from which the valuable minerals have been extracted. sider original rock chemistry, including metal and sulfide con- Due to inherent chemical and slope instability and other poten- tent, in reclamation. In addition, revegetation success on tail- tial environmental problems, they can represent long-term ings is influenced by chemical and physical changes intro- public and private land-management challenges. For example, duced by the milling process. precious metal tailings may contain dissolved metals and cya- nide complexes (Miller, 1997). If the tailings impoundment is poorly designed, constructed, or reclaimed or is breached or LOCATION AND CONSTRUCTION OF otherwise loses its integrity, release of contaminants and sedi- TAILINGS IMPOUNDMENTS ments can threaten the environment. In response to these po- Because the ultimate success of tailings containment and isola- tential problems, state and federal agencies and local govern- tion depends on the long-term stability of the impoundment ment have steadily increased environmental scrutiny and regu- dam itself, a description of impoundment types is warranted. lation of metal mine tailings since the early 1970s. Tailings impoundments (dams) can be broadly classified as Modern tailings impoundments that are properly designed, follows: sited, constructed, maintained, and reclaimed are unlikely to z valley containment, where a dam is constructed across cause acute environmental problems. At many metal mine the valley, sites, the single most important environmental issue is the qual- ity of reclamation of all the mine’s components, including z level ground impoundment, with ring dike containment waste rock dumps, the mine itself, and the tailings facility. and tailings deposited in the center, There can be no high-quality reclamation without considering z all aspects of a tailings facility. sidehill impoundment, which relies on a dam constructed with three sides. TAILINGS CHARACTERISTICS The rugged topography in Washington metals mining dis- tricts generally dictates that the impoundment design be either Tailings are generally composed of fine sand- or silt-size parti- valley or sidehill containment (Fig. 1). Currently, mining com- cles, typically deposited as a slurry, and, depending upon the panies tend to use mine waste material for impoundment con- method of deposition, may be graded so that coarser material is struction. Waste rock can be used for the main portion of the nearer the point of discharge. Gradients in grain size occur embankment if the rock is of suitable construction quality and both vertically and horizontally. will not generate acid rock drainage. Clean sand and gravel, if At most mines, some of the metals in the mined material available from the overburden, has also been used. Generally, cannot be recovered in the mill and are discharged to a tailings the embankment is constructed in 2- to 8-ft-thick lifts and com- disposal facility (TDF). In older tailings or where gangue met- pacted. Some mine operations outside of Washington use cy- als are not economic, the metal content can be 10 percent or clones (cone-shaped devices that have no moving parts) to higher due to milling inefficiency. Many modern mining op- separate the coarse fraction from the fines by centrifugal force; erations achieve metal concentrations of less than 1 percent in the sediment is discharged at relatively low moisture content tailings (Williamson and others, 1982). However, at some through the bottom. The coarse sands can then be stacked to mines, it is not the metal with economic value that is problem- form the sand portion of various embankment types. The over- atic, rather it is the associated metals that reach toxic levels. flow, or fines, is discharged to the tailings pool away from the Due to variations in milling, original mineralogy, and depo- sand embankment. sition method, tailings can have high salt contents, be acid or Most tailings dams in Washington are low-permeability alkaline, have macronutrient deficiencies, display particle-size water-retaining structures. Tailings dams that can impound 10 stratification, and lack organic matter (Munshower, 1994). acre-feet or more of water measured from the dam crest must Other sources of toxicity are processing agents, catalysts, rea- meet state dam safety standards. The Washington Department gents, and chemicals that are not recovered in the mill and thus of Ecology’s Dam Safety Section regulates planning and con- are discharged to the TDF. Cyanide is the most toxic reagent struction of dams under Revised Code of Washington (RCW) commonly used in metal recovery, but it degrades quickly 90.03 (The Water Code) and RCW 86.16 (The Flood Control when exposed to air. At most modern mine operations, tailings Act). are routed through a cyanide destruction and detoxification cir- TDF dams are generally constructed using the downstream, cuit before discharge to the TDF. Cyanide destruction meth- centerline, modified centerline, or upstream method (Fig. 2). ods, which can also be toxic, are discussed in Denton and oth- The downstream method requires the most material, covers the

Washington Geology, vol. 26, no. 2/3, September 1998 29 containment system is normally con- fined to one point, and seepage is fairly easy to collect. Reclamation of a valley-contained TDF is simpler than for a ring dike TDF because long-term surface drainage can be more readily re-established (Welch and Firlotte, 1989).

TAILINGS SITE SELECTION Site selection for an impoundment must take into consideration the position of ore bodies, distance and elevation from the mill, watershed characteristics, ex- isting and future land uses, site geol- ogy, environmental concerns (such as water quality, dust control, vegetation, and wildlife), property ownership, pro- posed type of closure and reclamation, Figure 1. Echo Bay Minerals Co. (Republic, Wash.) constructed this sidehill tailings impound- and the overall costs of the project. In ment that includes a geosynthetic liner with an underdrain system. The impoundment footprint is Washington, the siting criteria for tail- approximately 91 acres, and the surface area of the tailings is 61 acres. The maximum height of ings impoundments are set forth in the the embankment is 115 ft. The company used a centerline method for dam construction. Metal Mining and Milling Act (RCW 78.56.090). The mining company may propose several sites, or the Depart- largest area, and is the most expensive to build. Figure 3 shows ment of Ecology may select a site. The Department of Ecology a completed downstream method TDF. The downstream must issue a site selection report for the agency’s preferred lo- method results in TDFs that are generally more stable than cation for the TDF. The report must analyze the feasibility of dams constructed using other methods, and the integrity of any reclaiming and stabilizing the tailings; it must also take into ac- geomembrane (low-permeability synthetic) liner can be easily count the objectives of the mine proponent’s application rela- maintained. Costs for the downstream construction method tive to mining and milling operations. RCW 78.56.090 states may be reduced if suitable material is available on the mine “these objectives shall consist of, but not limited to (a) opera- site. In areas that may be subjected to large seismic events, the tional feasibility, (b) compatibility with optimum tailings downstream method is the most stable. The upstream method placement methods, (c) adequate volume capacity, (d) avail- uses the tailings as its base as each lift is added. It is the least ability of construction material, and (e) optimized embank- expensive to build and the least stable (Gipson, 1998). The cen- ment volume.” Siting criteria are qualitatively based on prox- terline method shares both advantages and disadvantages of imity to the 100-year flood plain and surface and ground water, upstream and downstream methods. topographic setting, geologic hazards such as landslides and Construction methods can affect tailings discharge man- active faults, visual impacts, soil, geotechnical, and hydrologic agement. Valley or side valley containment is more effective characteristics. than stacking tailings on level ground. Discharge from a valley

CL CL CL CL EFFECTIVE INITIAL FINAL FILL LINE FINAL INITIAL

A. DOWNSTREAM C. MODIFIED CENTERLINE

CL CL INITIAL AND FINAL FINAL

CL INITIAL

B. CENTERLINE D. UPSTREAM

Figure 2. Cross sections of four types of tailings dam embankments. CL, centerline.

30 Washington Geology, vol. 26, no. 2/3, September 1998 LINERS A liner system helps minimize the chances of leakage from the tailings into ground water or surface water. Historically, liners (soils, clays, and geomembranes) have failed and leaked, but recent advances have made liners far more reliable. Mine location, envi- ronmental considerations, and state laws or regulations help determine the type of liner to be used in tailings facil- ity construction. Further, liner design must be based on the assumption that leaks will occur and that leak detection and recovery systems are necessary. If a site lacks adequate natural ma- terials such as clay, sand, and gravel to construct liner components, a syn- thetic liner system must be used. Typi- cally, multiple liners are required, and Figure 3. Tailings impoundment and dam at the Cannon mine (Wenatchee, Wash.) during recla- the bottom layer consists of clay or mation in 1996. The impoundment is about 35 acres. The dam face is 250 ft high and 1,000 ft long amended soil material. Nonetheless, and extends completely across the Dry Gulch valley. The downstream dam method was used. the mining company, agencies, and Dam materials are mostly basalt and mine waste; a clay core reduces permeability. third party consultants should regularly conduct careful reviews of construc- tion quality assurance/quality control (QA/QC) (Norman and These liners have been used for the Comprehensive Environ- Raforth, 1995). mental Response, Compensation and Liability Act (CERCLA) Important considerations in choosing material for soil lin- containment systems and covers for uranium mine tailings ers are availability and composition. In general, increasing the since 1988. They have the advantage of being easily placed, clay content of the liner material decreases the permeability. can be less costly than imported clays (if local clays are not The liner material must have an appropriate amount and type of available), and are tolerant of handling during installation. A clay to achieve low permeability, high plasticity, and chemical drawback of GCLs is the loss of shear strength stability when stability when in contact with the tailings, which may contain they become hydrated and the potential for significant increase cyanide, metals, acids, or other reactive solutions. Material for in permeability because of the high cation-exchange capacity soil liners can consist of on-site or local borrow materials (if of the bentonite (Richardson, 1994). they have the correct clay content), bentonite, or mixtures of Important considerations in choosing geomembranes are both. Imperfections such as roots must be removed during con- thickness, strength, durability, cost, cover material needed for struction of the liner system. cushioning, the method of placement, and the construction The thickness and method of liner compaction are engi- method and quality of seams between the sections of the liner. neering considerations. Most clay liners are designed to The suitability of a geotextile material varies with the density achieve a hydraulic conductivity (permeability) in the range of achieved during manufacture (Ellison and others, 1992). In ad- 10-6 to 10-7 cm/sec (1.0 to 0.1 ft/yr). To attain low-permeability dition, geomembranes should not react with the reagents used containment, clay liners must be thoroughly mixed, condi- in the milling process or in the tailings. The long-term life ex- tioned, carefully placed, and properly compacted, as well as pectancy of geomembrane liners currently in use is not known protected against damage to continuity due to cracking from but is shortened by exposure to sunlight. Most geomembranes drying and shrinking (Hutchison and Ellison, 1992). Expand- are now manufactured with ultraviolet inhibitors and are ex- able clays (smectite and illite-smectite group clays) are less pected to last more than 50 years even if exposed to sunlight. permeable than other clays (kaolinite or illite) and are more In the past, tailings from mills using flotation-only circuits commonly used. were deposited in unlined facilities. The mine operators relied Impoundments have used synthetic liners since the 1940s. on the perceived “relatively benign” nature and the low perme- Technological advances in the manufacture of synthetic mate- ability of the tailings to minimize ground-water impacts (van rials during the past decade have resulted in the widespread use Zyl, 1997). However, reliance upon these characteristics alone of geomembranes for all components of the liner system (Elli- is not adequate to meet current water-quality protection re- son and others, 1992). Geomembranes are made of polyvinyl quirements, which is another reason liners are required. chloride (PVC), HypalonTM, high-density polyethylene Water quality can be severely degraded if liners fail. Past (HDPE), very low density polyethylene (VLDPE), chlorinated failures of geomembrane liners have been attributed to poor polyethylene (CPE), asphalt/hydraulic asphaltic concrete welding of the seams or joints in the geomembrane or to punc- (HAC) (Dorey and others, 1988), and XR-5 (chlorosulfinated turing during or after placement. Properly cushioning the geo- polyethylene) (Marcus, 1997). The three most commonly used membrane can prevent punctures. New techniques for welding geomembranes are HDPE, PVC and VLDPE. seams have considerably improved seam reliability. Installing Geosynthetic clay liners (GCLs) are also commonly used. a cover layer to cushion and protect the geomembrane has also GCLs sandwich bentonite (composed mostly of smectite- been key in successful operations. Indeed, most failures can be group clay) between two geotextiles (woven or non-woven prevented by strict adherence to QA/QC during pad construc- permeable synthetic fabrics) that are glued or sewn together. tion (Hutchison and Ellison, 1992).

Washington Geology, vol. 26, no. 2/3, September 1998 31 Geotextiles are used above and below the geomembrane permeability tailings to a collection pond for treatment, if layer to protect against penetrations by underlying rocks or needed. Several inches of free draining gravel and coarse sand large particles due to loads from construction activities or the are adequate in most places to rapidly remove the small vol- weight of the waste material. Protecting and cushioning the umes of fluid. However, mining companies generally place liner can be accomplished with clay- to sand-size material. thicker (8–18 in.) layers of sand and gravel for the drain layer. Small rounded gravel has also been used successfully to pro- The sand drainage layer is generally placed between an upper tect geomembranes from puncture. The protective soil must be geomembrane layer and a lower geotextile layer that is in- relatively free of large rocks and roots that could concentrate tended to minimize clogging of the sand by the fine tailings. stress on the liner. The drainage layer commonly includes a network of closely The preceding discussion indicates that liners must be de- spaced perforated pipes that rapidly drain collected fluids and signed on a site-specific basis. Common designs in use today minimize hydraulic head above the liner. consist of: Geonets (a net-like synthetic) have also been used as drain- z two layers of synthetic material separated by a age layers within liner systems and allow effluent to be col- permeable leak-detection layer (sand or a permeable lected at a single point. Geonets are formed with a minimum of synthetic net); two layers of ribs oriented to enhance planar flow. Geonets are 1/ z approximately 4 in. thick, and the structure of the net can pro- a lower layer of clay or clay-amended soil, a middle vide the same planar flow capacity as a 12-in. layer of sand leak-detection layer of sand, and a capping synthetic having a permeability of 10-2 cm/sec (Richardson, 1994). layer; and Geonets have less hydraulic storage than conventional sand or z a composite liner composed of a synthetic layer gravel drains; this allows draining fluid to reach a monitoring immediately overlying a clay or clay-amended soil and recovery point faster. Sand and gravel drainage layers can layer. store a significant volume of water in their pore spaces. The thin geonets have almost no storage capacity, and liquid must REQUIREMENTS FOR TAILINGS FACILITIES be removed continuously, typically by a gravity drain. IN WASHINGTON Leak Detection In Washington, RCW 78.56 (Metals Mining and Milling Act) requires that tailings facilities be designed and operated to pre- Leak detection systems for tailings impoundments have be- vent the release of pollutants. Mine operators must apply “all come commonplace at most metal mines and are required in known available and reasonable technology” to limit the con- Washington by RCW 78.56.100. Detection methods vary centration of potentially toxic materials in a tailings facility widely and may include monitoring wells, lysimeters, pie- and to assure protection of wildlife and human health. zometers, neutron probes, dielectric probes, tracers, and un- Tailings facilities must have a containment system that in- derdrains that route leaking solution to a single point. cludes an engineered liner system, leak detection and collec- One of the newer methods of leak detection uses direct- tion elements, and a seepage collection impoundment to assure current electrical resistivity. To locate leaks, an electrical cur- that a leak of any substance regulated under RCW 90.48 (Wa- rent is generated and sent to the ground between two elec- ter Pollution Control Act) will be detected before escaping to trodes, one above the liner and the other in the soil or material ground or surface water. The design and management of the fa- below the liner. (A grid of electrodes is also effective.) If the cility must ensure that any leaks are detected in a manner that geomembrane is intact, very little electrical current will flow allows for remediation pursuant to RCW 90.48. through the highly resistive plastic liner. If a hole exists in the Also according to RCW 78.56, applicants for metal mines liner, there will be an increase in the electrical current flow at and tailings facilities permits must submit a detailed engineer- the point of leakage, which will create an anomaly in the elec- ing report about the facility design and construction to the De- tric potential measurements (Bishop, 1997) partment of Ecology for review. If a dam is included, the De- partment of Ecology’s Dam Safety Section approves the design WATER QUALITY AND GEOCHEMISTRY and construction, while the Water Quality Section approves the The primary water-quality issue associated with tailings is the waste treatment system. Tailings facility design must take into potential for toxicity to humans, wildlife, and vegetation account natural conditions, such as precipitation and depth to through degradation of surface water and ground water by so- ground water, but not as a replacement for the protection re- lutions draining from the tailings. One source of toxicity is quired by the engineered liner system. processing chemicals introduced during milling. Many mills The goal is to reduce the toxicity of mine or mill tailings employ cyanide leach processing in metal recovery. Although and the potential for long-term release of regulated substances cyanide is toxic, it degrades quickly under oxidizing condi- to the greatest extent practicable through stabilizing, remov- tions. Some degradation byproducts, such as ammonia, are also ing, or reusing the substances. When the tailings facility is potentially toxic, depending on the concentration and dose. closed, isolation and containment of potentially toxic materials Most operations employ an inline cyanide destruction circuit is assured. before the tailings are discharged to the impoundment facility. In many instances, the level of toxicity introduced by mill pro- Drainage Layers cess circuits could be largely managed in the mill; this would Drainage layers must be included as part of the engineered cut costs and contribute to the goal of economic recovery of liner system required by RCW 78.56. Drainage layers are in- minerals. Pollution prevention involves using substitutes for tended to maintain a low hydraulic head above the liner acting toxic chemicals, reducing the amount or concentration of toxic as the containment barrier. Traditionally, high-permeability chemicals used in processing, modifying processes, and im- sand or rounded-gravel drain layers on top of the liner drain proving housekeeping and maintenance (USEPA, 1995). fluid transmitted through or expelled from the low-

32 Washington Geology, vol. 26, no. 2/3, September 1998 Another source of waterborne tox- icity is the generation and release of acid rock drainage (ARD) from the tail- ings. Acid drainage results from the ex- posure of pyrite (FeS2) and other metal sulfides to air, which causes oxidation and generation of acids and dissolution of metals mobilized by percolating wa- ter. Tailings that contain sulfides and insufficient buffering minerals are par- ticularly susceptible to acid generation because finely ground material has a large available reactive surface area. Commonly, older mines deposited tail- ings in valleys (Fig. 4) or on slopes without designed embankments; at times tailings were dumped directly into streams, rivers, or lakes. Some modern tailings impoundment failures have resulted in uncontrolled dis- charges and released ARD and tailings materials (Steffen, Robertson, and Kirsten, 1989). Figure 4. Copper mine tailings in the valley of Railroad Creek above Lake Chelan, Wash. Tail- Mitigation of ARD is difficult and ings from the Holden mine were deposited in the valley bottom in the 1940s and 1950s. The tail- expensive. The greatest concern of ings have generated acid rock drainage and have also been eroded by the stream and interacted regulatory agencies is that they will dis- with ground water beneath the tailings. The Washington Department of Ecology, U.S. Forest Serv- ice, U.S. Environmental Protection Agency, and Alumet, the parent company of Howe Sound Min- cover ARD after mine closure and ing Company that originally mined at Holden, are now cleaning up the site. Because no vegetation abandonment when financial resources was established, blowing dust was a chronic problem. A gravel cover has now stabilized the sur- to fund mitigation are no longer avail- face, and revegetation may be easier. Test plots showed that revegetation is possible on these able. At some mines, the buffering po- acidic tailings; lupine and Sitka alder, plants nativetotheareaandnitrogenfixers,werethemost tential of tailings is chemically “used successful plants in the test plots monitored by the Forest Service (Scherer and others, 1996). up” years after the tailings have been (Photo by Eric Schuster, Washington Department of Natural Resources.) deposited, resulting in ARD. For exam- ple, at the Thompson Creek mine in Idaho, it took about 10 years after deposition for the tailings to quantities of potentially acid-generating and acid-neutralizing generate ARD (Jerry West, Idaho Dept. of Environmental rock. However, the duration of some tests may limit their accu- Quality, oral commun., 1998). Thorough study and laboratory racy in predicting the potential for ARD using various geo- testing of the acid generation and neutralization potential of the chemical tests. Short-term tests are likely to be poor predictors tailings are imperative if planning and design are to anticipate of long-term geochemistry and can hamper attempts to quanti- the ARD issue and avoid buffer depletion. tatively forecast impacts and costs. The test results resemble Current mining and regulatory objectives are to perform meteorological predictions in that, given certain conditions, some level of investigation of the potential for acid generation one can only say that there is a higher or lower probability of before, during, and after mining. Pre-mining investigations acidic or alkaline water (Kleinmann, 1997). should include multiple geochemical tests of the material ex- Mine operators routinely analyze rock samples from the pected to be placed in the tailings disposal facility. If these con- mine/prospect site to determine their acid- and alkaline- servative tests suggest potential for ARD generation in the tail- producing potential on the basis of percent sulfur or pyrite ings, mitigating measures must be included in the milling cir- (Kleinmann, 1997). Valuable information can be gained by in- cuit (for example, a system that removes sulfides) or the TDF vestigating nearby operations or abandoned mines if they are design. If there is an ARD potential, regulators should antici- reasonable analogs of the proposed mine site. If nearby sites pate the need for an engineered design that assures that long- are acid-producing, the mining company should conduct inten- term water quality will not be adversely impacted by dis- sive geochemical investigations during pre-mine planning as charges from the TDF. The engineered design should include well as during mining and reclamation of the new site. The ana- not only containment and isolation of the tailings, but also a re- lytical methods commonly employed to predict ARD fall into liable method of treating post-closure water discharges. The the category of static or kinetic tests; those recommended by mining company should estimate the costs involved in treat- Price (1997) and Price and others (1997b) for a metal leaching/ ment facility design, construction, and maintenance. Innova- ARD prediction program are: tive reclamation might include designs that will minimize oxy- Static tests gen diffusion and percolating water in the tailings. y Adequate baseline water-quality monitoring, including trace-element content (total and soluble sampling of adits and pits near the site, is a critical element in concentration) y mine planning. This information can supplement geological acid-base accounting studies and predictive geochemical results. The mining com- y total-sulfur and amount of sulfate and sulfide pany might construct a geological model of the area to explain y bulk neutralization potential observed variations in water chemistry and to estimate relative y carbonate neutralization potential

Washington Geology, vol. 26, no. 2/3, September 1998 33 y pH Specific needs in designing tailings disposal are: y mineralogy and other geological properties z a milling and metallurgical study of how to remove y petrographic examination sulfides and add alkalinity and of the effects of adding Kinetic tests of reaction rates and drainage chemistry or removing metals to determine the effect of these tactics on the metal leaching and ARD potential, and y pre- and post-weathering characterization z regular operational sampling and analysis of tailings to y humidity cell confirm pre-mining material characterization. y on-site pilot tests of tailings The study should cover the range in ore composition and iden- y mine wall washing stations (performed in-place tify the milling process, pertinent mineralogy, and particle size on the rock face) that most effectively eliminates or minimizes ARD. y site drainage monitoring In tailings, which tend to be more homogenous than waste- The recommended kinetic test is the humidity cell, used for rock dumps, predictions of water quality tend to be more reli- predicting primary reaction rates under aerobic weathering able. At 16 sites studied by Environment Canada, no site with a conditions (Price and others, 1997a). In theory, kinetic tests re- calculated excess neutralization potential produced acidic flect the difference in rates of sulfide oxidation and of calcium drainage (Kleinmann, 1997). carbonate dissolution reactions. Whether this makes kinetic tests more realistic is difficult to say, due to general lack of DISPOSAL METHODS field verification. In contrast, static tests only indicate relative chemical activity among minerals but are not time dependent. Tailings are conventionally deposited as a slurry moved in a An additional problem shared by static and kinetic tests is that pipeline (by gravity or pumping) and delivered to a fixed or their relevance is limited by the degree to which the samples moveable single point in an engineered surface impoundment being tested are truly representative of the mine or waste com- (Fig. 5). This process can place the tailings in fairly thick lay- positions (Kleinmann, 1997). In the case of tailings, represen- ers and, while excess water is typically decanted from the sur- tative samples are generally easier to obtain because ores are face, tailings can remain saturated for years if not dried before blended during grinding and milling. new layers are deposited. Reclamation must wait for the sur- However, Price and others (1997b) suggest the following face layer to dry enough to safely support earthmoving equip- factors should be considered when predicting the future drain- ment (Fig 6). Lower layers either remain saturated or eventu- age chemistry of tailings: ally drain out the bottom of the impoundment. n The composition of tailings may change considerably over While deposition as a slurry directly to an impoundment is time. Milling procedures used by metal mines can remove still the predominant method of tailings disposal, modified sulfides and add alkalinity, both of which reduce the metal slurry methods or several other techniques are gaining accep- leaching/ARD potential. However, other milling practices tance and becoming more widely used. These include dewater- ing, paste, and air-dried deposition. Under some circum- can increase the potential for metal leaching/ARD by add- stances, these alternative methods result in higher quality and ing metal-bearing reagents like copper sulfate. The metal less costly reclamation as well as greater mine efficiency. leaching potential will depend on the mill process, perti- nent mineralogy of the ore, and par- ticle size. n Tailings have fine grain sizes and thus have a large surface area; all mineral components are readily available to weathering processes. n Fine tailings have a reduced pore size and thus lower permeability for both air and water. Restricted air movement may limit the rate of oxy- gen replenishment, thereby reduc- ing acid generation and metal leaching and perhaps changing the balance between acid generation and neutralization. n Significant mineralogical and particle-size segregation may occur when tailings are deposited. Selec- tive deposition of heavy minerals may create zones that have higher metal-leaching potential and are close to the deposition point. Sepa- ration of sulfides or carbonates on Figure 5. Discharge pipes at the Chimney Creek gold mine, Nev., deposit a slurry of saturated tailings beaches may result in local- tailings. This mine air dries tailings before additional tailings are deposited. The pipe is approxi- ized ARD. mately 6 in. in diameter. (Photo by Allen Throop, Oregon Department of Geology and Mineral In- dustries).

34 Washington Geology, vol. 26, no. 2/3, September 1998 High-quality reclamation depends on key engineering tech- Submarine Tailings Disposal niques such as good dam integrity, covers that are nontoxic, Tailings have been placed via discharge of a slurry at a sub- and permanent vegetation, all of which result in water quality merged outfall in lakes and marine water typically more than that is appropriate to beneficial post-mining uses. Options 150 ft deep. One objective of this disposal method is to place other than conventional slurry deposition that allow for earlier tailings where ambient dissolved oxygen concentrations are and better reclamation must be considered, especially if they minimal, thus reducing the potential for metals to be oxidized, result in less impact to the environment by lowering the poten- mobilized into the water column and made biologically avail- tial for release of metals to ground or surface waters. Today, al- able, or transported away from the site of deposition. Other ob- ternative disposal methods are drained and air-dried, subma- jectives are to place tailings in a stable environment and to pre- rine, thickened, and paste. vent fines from entering the shallow, biologically productive euphotic zone (Rankin and others, 1997). Drained and Air-Dried Method Examples of submarine tailings disposal operations are the Mining companies use drained and air-dried (subaerial) meth- BHP Island Copper Mine at Rupert Inlet, B.C.; Kitsaoult Mo- ods extensively in arid environments such as Nevada. The lybdenum Mine at Alice Arm, B.C.; Black Angel lead-zinc method involves sequencing the deposition of thin (<4 in.) lifts mine at Maarmolik, Greenland; and Misima gold-silver mine in of tailings in segments of the impoundment so as to allow pre- Papua New Guinea (Rankin and others, 1997). Mine operators viously deposited lifts to dry and consolidate (Knight-Piesold believe that submarine disposal is an ecologically acceptable Co, oral commun., 1998). These impoundments must have alternative to land-based tailings disposal. Environment Can- enough surface area to allow time for drying in one or more ada and Canada’s Department of Fisheries and Oceans have parts while deposition continues in other parts; tailings must be initiated a review of the ecological issues (Rankin and others, deposited over at least two segments for this method to be suc- 1997). cessful, and as many as 28 segments have been used. The selec- However, it is not likely that submarine tailings disposal tion of number of segments is based on the climate, tailing pro- would be used in Washington as this is not allowed by either duction rate, tailing drying characteristics, and facility shape federal or state law. Thresholds for metal content would proba- (Gipson, 1998). bly exceed those allowed by current regulations. Additionally, In arid climates, desiccation proceeds under high evapora- Department of Natural Resources Aquatic Resources Division tion rates. This reduces the moisture content of the material, regulations (Washington Administrative Code 332-30-166 while increasing the density and shear strength. Drying allows (1)) regarding submarine disposal state “Open water disposal a fairly strong crust to form, but this crust may crack as a result sites are established primarily for the disposal of dredged mate- of shrinking (Newson and others, 1997). rial obtained from marine or fresh waters. These sites are gen- The drained and air-dried method offers the advantages of erally not available for disposal of material derived from up- improving control of the amount and chemistry of the solution land or dryland excavation except when such material would within the facility and developing a stable, denser tailings de- enhance the aquatic habitat.” posit that has a lower moisture content. The tailings are easier to reclaim because they are not saturated and loose. The in- Thickened Tailings crease in density resulting from drained and air-dried deposi- Dry tailings, also called dewatered tailings, may not require a tion permits nearly twice the amount of tailings to be stored in tailings impoundment. This method uses either a filter or a the same volume of tailings facility as an undried slurry (Gip- high-density thickener that relies on flocculation. Typical op- son, 1998). Some companies using this method are Battle erations use a thickener and filter (such as a belt press) in com- Mountain Gold Corp. (San Luis mine, Colo.) and American bination or, increasingly, a thickener alone to produce a tail- Barrick Resources Corp. (North Block tailings facility, Twin ings consistency similar to a paste. Some mines use cyclones, Creekmine,Nev.).(SeealsoFig.5.) belt filters, or other mechanical filter equipment alone to re- duce moisture content in the tailings. Solids contents as high as 85 percent can be obtained from such devices (Johnson, 1997). Each 6-in. layer of processed tail- ings is allowed to dry further prior to adding the next layer. Typically, the layers crack during drying, and the next layer fills the cracks and adds to the pile. Currently, dry stacking operations road create high-slump1, low-viscosity tail- ings that will flow easily as a thin layer over a wide area (Schoenbrunn and La- ros, 1997).

mudcracks 1 The standard slump test is described in ASTM C143. Slump is defined as the height lost when a filled cone 12 in. tall, 8 in. wide at the base, and 4 in. wide at the top is slowly Figure 6. The tailings impoundment at Cannon mine near Wenatchee, Wash., in 1994. Tempo- lifted off a plate and the cone contents are al- rary roads on a geotextile base and sand over saturated tailings (center left and center right of the lowed to “slump”. A 7-in. slump = the initial 12 photo) allow access to drilling sites to help evaluate tailings chemistry. in. height – 5 in. height lost.

Washington Geology, vol. 26, no. 2/3, September 1998 35 reclaimed area where paste was deposited deposition of paste ~4-ft-thick terraces with 7 in. of slump constructed of paste

mill

pipeline

Figure 7. Sketch of a proposed method of paste disposal at ASARCO’s Rock Creek project in Noxon, Mont., that will use a paste stacker to con- struct semicircular windrows (left side of the figure). Each lift (terrace) would be about 4 ft thick and allow for concurrent reclamation shown at the right of the figure. (Modified with permission from Golder Associates, 1996.)

Examples of thickened tailings are at alumina plants such concurrently with mining (Young, 1997). Among the cited ad- as Alunorte in Brazil, Alcoa of Australia’s plants at Pinjarra, vantages of using paste for this project would be limiting the Kwinana, and Wagerup, and Alcan’s plants in Jamaica, at Vau- active areas of disturbance and eliminating the need for a sur- dreuil in Canada, and at Aughinish, Ireland. Placer Dome’s La face water-impoundment, and thereby the potential for seepage Coipa mine in Chile uses horizontal belt filters to create thick- through the deposit. The potential for earthquake-induced liq- ened tailings. After their tailings pond filled up, the Sunshine uefaction would be low due to the low moisture content of the gold mine in Nevada used a beltpress in conjunction with a deposit. Because of the high density achievable with pastes, thickener to create space for tailings from remaining ore re- both the height and footprint of the TDF can be reduced for the serves rather than add a lift to the dam (Schoenbrunn and La- same tonnage of tailings. Most significant, however, would be ros, 1997) that final closure and return of the property to appropriate post- mining land use could occur shortly after paste production Paste ceases (Golder Associates, 1996). Recent developments in dewatering equipment design have Pastes have been used extensively as backfill for under- made practical the consistent production of tailings that have a ground mines. There is now interest in the potential of paste as low moisture content and distinct flow and water retention an alternative to more traditional slurry tailings disposal meth- properties; these are commonly referred to as pastes. Pastes are ods (Cincilla and others, 1997; Golder Associates, 1996). Most dense, viscous mixtures of tailings and water that, unlike slur- rock types are amenable to paste production for surface dis- ries, do not segregate when allowed to rest. Generally speak- posal but may not be suitable for underground backfill. For ex- ing, pastes resemble wet stiff concrete, and an optimum consis- ample, high-sulfide content tailings with portland cement tency for a typical paste would be in the range of a 7-in. slump. added have been used for underground backfill in Quebec to Terminology for concrete tests is used in the evaluation of reduce the volume of surface tailings, strengthen underground pastes. Pastes consist of enough fine particles (at least 15% by workings, and minimize subsidence. But reaction between py- weight passing the 20-micron size sieve) to allow them to flow rite and free calcium ions produced by the dissolution of unsta- through a pipe, yet have enough retained water to create a non- ble portlandite hydrate (Ca(OH)2) can result in the precipita- segregating mixture. However, the most important aspect of tion of swelling secondary gypsum (CaSO4·2H2O) and very ex- paste is that the largest portion of the entrained moisture is held pansive ettringite (Ca6Al2[(OH12)](SO4)3 (OH)12·26H2O), by surface tension in the fine particle matrix. This phenomenon which produces a weak mortar not suitable for backfill (Ouellet produces a material that has an initial moisture content in the and others, 1998). However, acid generation is probably less range of 20 percent (by weight) and, like concrete, a noticeable likely to occur in paste tailings because permeability is de- lack of free-draining water (Cincilla and others, 1997; Golder creased and less oxygen can reach the minerals. Associates, 1996) Pastes have distinct environmental advantages. First, very RECLAMATION AND CLOSURE little free water is available to generate a leachate that might Dewatering and Shaping have detrimental effects on ground water. At some sites, elimi- nating free water during deposition could remove the need for Before beginning final closure and reclamation of a tailings fa- an engineered water-retaining structure. Second, because it cility where the slurry method was used, mine operators com- does not segregate, it can be pumped to a placement site. In ad- monly must remove the water standing on the tailings (Fig. 8). dition, a few percent of portland cement or fly ash can be added Reclamation of saturated tailings can be difficult and costly. to paste; this significantly increases its dry strength, durability, Final reclamation may have to be delayed until tailings have and ability to buffer acid tailings and can encapsulate particles adequately dried (Fig. 9). An alternative method to surface air that are potentially acid generating. drying is the removal of pore water through vertical wick Surface paste disposal at the proposed Rock Creek copper drains. These wick drains give pore water a fast path to the sur- mine project (currently in environmental review) at Noxon, face. Once drying has proceeded enough for equipment to Mont. (Fig. 7), would allow ASARCO to reclaim the tailings work on the tailings surface, reshaping develops the planned

36 Washington Geology, vol. 26, no. 2/3, September 1998 drainage. Generally, drainage of the tailings should be coupled with control of storm water.

Covers Cover designs depend on site-specific evaluations. Designing one cover that would be suitable for all tailings is not possible because of the wide range of climate, soil, and topographic condi- tions at mines (Ritcey, 1989). Covers for tailings are generally ei- ther vegetation or a layer of soil, rock, or water. Under some circumstances abandoned tailings can be revegetated without applying topsoil. Abandoned mine tailings at Telluride, Colo., were revegetated by simply tilling organic Figure 8. Hecla’s Republic Unit tailings impoundment at Republic, Wash., where water is evapo- amendments (manure and hay) and in- rating from the tailings surface prior to reclamation. The company pumped water ponded on the organic fertilizer into the surface (Re- surface though sprinklers to speed evaporation, thereby drying the tailings. White piles in the cen- dente and Baker, 1996). In virtually all ter of the photo are ice formed around the sprinklers during the winter of 1995 operation. In the instances, salvaging and replacing soil foreground, the tailings have begun to dry, and mudcracks have formed. at operating mines would be less costly than amending tailings. For some aban- doned tailings, the simplest covers used to stabilize the surface have been single or multiple layers of gravel. Where wa- ter layers have been used, no vegetation was established. Miners in British Co- lumbia have used a layer of water to re- duce the amount of oxygen entering highly acidic tailings. Other cover designs have incorpo- rated a gravel layer for drainage or a capillary break below the topsoil. Mul- tilayer covers might consist of a top-to- bottom sequence of soil and gravel, in- filtration barriers such as a geomem- brane or a clay layer, and a geotextile (if required for support during con- struction). The U.S. Environmental Protection Agency (1989) recommends a multiple layer cover for hazardous waste such as uranium tailings. Some tailings covers incorporate a barrier to prevent plant roots or animals Figure 9. Cannon mine tailings undergoing final reclamation in 1996 after 2 years of drying. Here from disrupting the integrity of the a woven polypropylene fabric geotextile layer was laid down first as a supportive base on which to drainage and infiltration layers and place sand and soil. A lower cover layer of 2.5 ft of gravelly sand was covered by an upper layer burrowing animals from bringing con- consisting of 2 ft of sandy silt pushed out onto the surface by small bulldozers. The surface was taminants to the surface. Typical bioin- then prepared for revegetation by scarifying. Reseeding occurred the following fall and winter, and trusion barriers are layers of cobbles or grasses are thriving. (See also Figs. 3 and 6.) (Photo courtesy of Asamera Minerals.) coarse gravel placed between the top layer and the waste (and infiltration barrier if present). This biointrusion layer can also be separated erately damp to arid environments. Harvester ants also man- from soil by a geosynthetic filter layer (Henderson and others, aged to invade waste through gravel barriers. Mice, kangaroo 1992). rats, pocket gophers, and prairie dogs also breached the integ- Mining companies have tried biointrusion barriers through- rity of protective barriers. out the world for many types of hazardous wastes, mostly those At their Midnite uranium mine near Ford in northeastern associated with radioactive material. Bowerman and Redente Washington, Dawn Mining Company proposes a 15.3-ft-thick (in press) evaluated studies documenting root intrusion where homogenous cover of sandy soil to protect water quality and barriers had been emplaced. They conclude the ability of many control radon emanation. The site is characterized by sandy kinds of barriers to protect against intrusion is questionable. and sandy loam soil. Ponderosa pine woodland is the local late- For example, several plant species (such as crested wheatgrass, successional native vegetation. The pines are 40 to 60 ft tall at big sagebrush, and saltcedar) penetrated these barriers in mod- the Midnite mine and can produce roots that reach depths as

Washington Geology, vol. 26, no. 2/3, September 1998 37 great as 75 ft, although much shallower rooting is common; ring upper part of a soil profile, including the soil horizon that company consultants found roots below 8 ft at the mine and be- is rich in humus and capable of supporting vegetation together low 10 ft at a nearby site. Other plants native to the area are with other sediments within four vertical feet of the ground sur- bluebunch wheatgrass and needle-and-thread, both perennial face.” This definition indicates that mine operators have to sal- grass. Bluebunch wheatgrass rooting to 4.5 ft has been re- vage more than just the “A” horizon for reclamation. If topsoil ported, and needle-and-thread roots reach deeper than 6 ft or amendments such as hay, biosolids, paper residues, or com- (McLendon and others, 1997). The tailings cover design is in- post are available, revegetation of mine tailings is generally not tended to accommodate the expected final native plant commu- a problem. Permanent stabilization of acid-generating tailings nity. The mining company and their consultants believe that that are covered with a cap to establish a nontoxic rootzone can the noncohesive sandy soils will allow for self-healing of root be accomplished with normal agricultural practices of seedbed holes (left after roots have rotted), thereby addressing biointru- preparation, fertilization, and seeding. sion concerns (McLendon and others, 1997). With a proper seed mix and soils amended with paper waste, which is widely available in the Pacific Northwest and Capillary Barriers other parts of the country, a lasting vegetative cover can be Capillary barriers of fine soil over coarse soil can be a simple, quickly established with little or no additional fertilizer appli- low-cost method of effectively limiting water and oxygen cation. Paper mill waste can act as both an amendment and a movement through the tailings cover (Stormont and others, capillary barrier. Some of the advantages of paper mill residu- 1996). In many settings, the capillary barrier functions as the als are low permeability, high water-holding capacity, low ero- drainage layer as well (Henderson and others, 1992). Capillary dibility due to its fibrous nature, structure and stability suitable forces hold the water in the fine layer until it is removed by for root development, and neutral to basic pH to buffer infil- evapotranspiration or drains where the fine-coarse interface is trating water before it reaches acid tailings. Canadian mining sloped. A capillary barrier is effective if the combined effects companies have successfully used paper pulp waste, a spongy, of evaporation, transpiration, and lateral diversion exceed the partially saturated material that absorbs large quantities of wa- infiltration from precipitation. However, if the fine layer be- ter, as a soil amendment for reclamation (Cabral and others, comes saturated, capillary forces will decrease and water will 1997, 1998). Paper mill sludges are composed of 40 to 60 per- drain quickly through the fine layer into the coarse layer. De- cent organic material and have been “recycled” for successful sign considerations are unique to each site and must account use as landfill caps at several sites (Moo-Young and Zimmie, for seasonality of precipitation and evapotranspiration. 1995; Maltby and Eppstein, 1996; McGee and others, 1996). At the East Sullivan gold-bearing massive sulfide mine in Val d’Or, Quebec, a 6.4-ft-thick cover of softwood and hard- REVEGETATION wood bark over acid-generating tailings effectively prevents Revegetation success generally depends on the cover design oxygen from reaching the tailings. Biosolids from a municipal and tailings chemistry. For vegetation to be effective and water treatment plant tilled into the first foot of the bark add healthy, the roots should be able to penetrate the soil and reach nutrients (Tremblay, 1994). This site now supports a dense enough water that contains adequate nutrients and no toxic grass cover. components. Metalliferous mine tailings commonly contain Re-establishing native plants at mine sites has become an low concentrations of essential plant nutrients; nitrogen levels important objective for reclamation. In many places, desired are invariably inadequate for plant growth (Williamson and post-reclamation plant communities composed of native plant others, 1982). Mining companies must conduct tests prior to species are preferred because most are adapted to the climate and during mining to investigate the physical and chemical na- and elevation of the site. Some native plants can out-compete ture of the tailings and to determine if vegetation can be re- some introduced (exotic) species over time and are more useful established. Important considerations are temperature, precipi- to wildlife. The vegetation at or surrounding the mine site can tation, wind, and aspect, as well as texture of the tailings. be used as a guide to selecting native species. Using native seed Direct revegetation of tailings without soils can be ex- mixes or plants produced from locally collected seeds and cut- tremely difficult. An improved seedbed encourages germina- tings and locally transplanted plants can greatly accelerate re- tion and establishment of vegetation. Establishing vegetation vegetation. If preplanning is sufficient and the appropriate tail- on neutral tailings is less difficult than on nutrient-poor, salt- ings deposition methods selected, soil and native vegetation rich or alkaline or acid tailings. Over any acid-generating tail- can be transferred directly from areas being stripped for mining ings, the soil layer, which is in many places topsoil and a mix- to the TDF. This approach is less expensive and typically more ture of subsoils, must be thick enough to prevent salt or acid successful than long-term soil storage. Soil hauled directly migration to the surface. Acid tailings without soil coverings from a newly stripped area to a reclamation area carries with it must be leached before seeding (Munshower, 1994). It may viable seeds of native vegetation that can rapidly establish on also be necessary to apply lime or some other neutralizing ma- the reclaimed area. This typically reduces the need for added terials, as well as amendments such as manure, wood chips, seed and plant material (Norman and others, 1996). This compost, or biosolids. method would be most appropriate at mines using paste deposi- Incorporating compost or biosolids and industrial wastes, tion and concurrent reclamation. such as fly ash, into acid-generating (or potentially acid- generating) tailings and mine waste has been effective. In Using Cattle as a Technique for Pennsylvania, for example, miners used coal fly ash to neutral- Establishing Revegetation ize acid mine wastes and allow vegetation to be established At a mine at Miami, Ariz., miners have tried an innovative (Scheetz and others, 1998). method of establishing vegetation on tailings (A. Throop, Ore- In Washington, RCW 78.44 has required topsoil salvage gon Department of Geology and Mineral Industries, written only since 1993. It states “‘Topsoil’ means the naturally occur- commun., 1996) (Fig. 10A,B). Where importing topsoil, seed-

38 Washington Geology, vol. 26, no. 2/3, September 1998 ing, fertilizing, and irrigating had failed, penning cattle on the tailings A piles was successful in encouraging vegetation. Generally, cattle trample and kill vegetation. In this case, the cat- tle trampled hay mulch, urine, and ma- nure deep into the tailings and created terraces by their network of trails on the steep slopes. When compared to cattle not on tailings, the blood and tissue of these animals showed no toxic or un- usual concentrations of trace elements (Dagget, 1997).

Coarse Woody Debris Mining companies and regulatory agencies recognize the importance of replacing coarse woody debris on re- claimed areas (Harmon and others, 1986). At mine sites in the Pacific Northwest coarse woody debris is now incorporated in final reclamation. For example, a recently proposed mine will re-place stockpiled woody debris at ap- proximately 7 tons per acre to provide B a substrate for essential microorgan- isms. The company will salvage logs ranging from 10 to 24 in. diameter at their small end and at a variety of lengths longer than 6 ft. Furthermore, salvaged lichen-encrusted rock ran- domly distributed across the reclaimed areas will promote colonization by li- chen (Battle Mountain Gold Co. and Golder Associates, 1998).

CONSTRUCTED WETLANDS Wetlands are an effective means of im- proving water quality; coal mines in the eastern U.S. were the first to use wet- lands in this context. Passive water treatment using constructed wetlands essentially simulates natural processes Figure 10. A. Penned cattle being fed at tailings piles at the Miami, Ariz., copper mine. The tram- pling helped to establish vegetation where several other methods of reclamation had failed. and is being used more commonly to B. Lower benches of Miami copper mine tailings were reclaimed using penned cattle. Upper treat mine water (Fig. 11). Bacterial re- benches have not been reclaimed. The face shown is about 3/4 mi long and 600 ft high and has a duction of sulfate and iron and precipi- slope of 2 horizontal to 1 vertical. The footprint of this tailings impoundment is about 1,100 acres. tation of metal sulfides are important The benches are 50 ft apart vertically, and interbench slopes are 1.3 horizontal to 1 vertical. (Pho- chemical processes in passive anaero- tos by Allen Throop, Oregon Department of Geology and Mineral Industries.) bic treatment. Sulfide precipitates ac- cumulate in the anaerobic zone and are not carried in the effluent (Filipek, 1997; Filipek and others, 1992; Sengupta, 1993). In a surface- TRENDS flow aerobic wetland used to treat acid drainage, the dominant Because mine operators and agencies now have a better under- process is oxidation of iron and precipitation of iron hydrox- standing of the physical function of tailings and have improved ides. Water in aerobic wetlands must be sufficiently alkaline to techniques to reclaim them, tailings have become less problem- keep pH from falling as a result of the hydrolysis of iron. An atic. However, tailings impoundments will require long-term anoxic limestone drain can be installed before water enters the monitoring to verify that they are not a source of contaminants. wetland to add alkalinity and raise pH if no iron is present in the initial drainage (Filipek, 1997; Brodie, 1991). This type of Among recent trends in dealing with tailings are: wetland treatment works best if heavy metal concentrations are z Regulatory controls are becoming more stringent, and low. mine operators are expected to rigidly adhere to them. Example of such control are Washington’s Metal

Washington Geology, vol. 26, no. 2/3, September 1998 39 Mining and Milling Act (RCW 78.56) and Oregon’s Chemical Process Mining Act (ORS 517.952). z Dry tailings deposition systems, such as dewatered, paste, or air- dried tailings, are reducing management problems. z There is much wider recognition that intense geochemical testing early in mine planning is necessary to characterize tailings chemistry and to plan environmentally sound deposition. z To achieve long-term physical stability of the impoundment, proper designs for closure more commonly consider long-term storm-water retention and seismic stability. z An increased use of non-acid- Figure 11. Wetlands being constructed at the base of the Cannon mine (Wenatchee, Wash.) generating waste materials for tailings impoundment. Although water discharged from the underdrain beneath the tailings is not acidic, these aerobic wetlands are being used to improve and maintain water quality. They were construction of mine facilities is built as a hedge against future water degradation predicted on the basis of tailings pore-water reducing both disturbed areas studies conducted during closure. The inlet for the wetland is the lower left of the photo near the and costs. building. After water leaves the wetland, it infiltrates into the ground. z Mine companies attempt to reduce cyanide levels to the lowest concentration consistent with technical Bishop, Ian, 1997, Geophysical approach finds leaks in containment feasibility and reliability to protect human health and system liners: http://news.poweronline.com/feature-articles/fa/ wildlife. 70604.html z Contaminant monitoring programs and response plans Bowerman,A.G.;Redente,E.F.,inpress, Biointrusion of protective barriers at hazardous waste sites: Journal of Environmental Qual- are more widely required. ity. z Geosynthetic and clay liners have become widely Brodie, G. A., 1991, Achieving compliance with staged, aerobic, con- accepted and are commonly used. structed wetlands to treat acid drainage. In Oaks, Wendall; Bow- z Topsoil salvage and revegetation of tailings has den, Joe, editors, Proceedings, Reclamation 2000—Technologies become the norm. for success; American Society of Surface Mining and Reclama- tion, 8th National Meeting, p. 151-174. ACKNOWLEDGMENTS Cabral, A.; Lefebvre, G.; Burnotte, F.; Proulx, M.-F.; Racine, I.; Audet, C., 1998, Developments in the use of deinking residues as Funding for this project was provided by the U.S. Environmen- cover systems for acid generating mine tailings. In Tailings and tal Protection Agency (EPA) through the Tri-State Agreement mine waste, ‘98: A. A. Balkema, p. 379-388. of Oregon, Idaho, and Washington to share and develop techni- Cabral, A.; Lefebvre, G.; Proulx, M. F.; Audet, C.; Labbe, M.; Mi- cal information regarding mining. We thank Nick Ceto and Bill chaud, C., 1997, Use of deinking residues as cover material in the Riley (EPA) for helpful discussions regarding mining. We prevention of AMD generation at an abandoned mine site. In Tail- thank the following for discussions and their reviews: Allen ings and mine waste, ‘97: A. A. Balkema, p. 257-266. Throop, Oregon Department of Geology and Mineral Indus- Cincilla, W. A.; Landriault, D. A.; Verburg, Rens, 1997, Application tries; Jerry West, Idaho Department of Environmental Quality; of paste technology to surface disposal of mineral wastes. In Tail- Ray Lasmanis, Washington Department of Natural Resources; ings and mine waste, ‘97: A. A. Balkema, p. 343-356. Rod Lentz, Okanogan National Forest; Brent Cunderla, Bureau Dagget, Dan, 1997, Convincing evidence: Range Magazine, Summer of Land Management; Jeff White, Battle Mountain Gold Co.; 1997, p. 22-25. Steve McIntosh, Echo Bay Minerals Co.; and Beth Norman, Denton, D. K., Jr.; Iverson, S. R.; Gosling, B. B., 1992, HEAPREC— Pierce College. We thank Reese Hastings of Hart Crowser for A methodology for determining cyanide heap leach reclamation discussions and supplying photographs of the Cannon Mine. performance bonds: U.S. Bureau of Mines Information Circular We thank Jari Roloff for help with editing and graphics. 9328, 79 p. Dorey, Robert; van Zyl, D. J. A.; Kiel, J. E., 1988, Overview of heap REFERENCES CITED leaching technology. In van Zyl, D. J. A.; Hutchison, I. P. G.; Kiel, J. E., editors, Introduction to evaluation, design and opera- Battle Mountain Gold Company; Golder Associates, Inc., 1997, rev. tion of precious metal heap leaching projects: Society of Mining 1998, Crown Jewel mine, Battle Mountain Gold Company, Engineers, Inc., p. 3-22. Okanogan County, Washington, reclamation plan: Battle Moun- tain Gold Company, 1 v.

40 Washington Geology, vol. 26, no. 2/3, September 1998 Ellison, R. D.; Vincent, I.; Parkinson, Kerry; Shaw, Donald, 1992, mining on the environment and American environmental controls Liner system design. In Hutchison, I. P. G.; Ellison, R. D., editors, on mining: Imperial College Press, p. 725-726. Mine waste management—A resource for mining industry profes- Moo-Young, H. K.; Zimmie, T. F., 1995, Design of landfill covers us- sionals, regulators and consulting engineers: Lewis Publishers, ing paper mill sludges. In Colville, James; Amde, A. M., editors, p. 327-388. Research transformed into practice implementation of NSF re- Erickson, P. M.; Hedin, R. S., 1988, Evaluation of overburden analyti- search—Proceedings of the conference sponsored by the National cal methods as means to predict post-mining coal mine drainage- Science Foundation: American Society of Civil Engineers, p. 14- quality. In Mine drainage and surface mine reclamation—Pro- 28. ceedings of a conference: U.S. Bureau of Mines Information Cir- Munshower, F. F., 1994, Practical handbook of disturbed land revege- cular 9183, v. 1, p. 11-19. tation: Lewis Publishers, 265 p. Filipek, L. H., 1997, Waste. In Marcus, J. J., editor, Mining environ- Newson, Tim; Fujiyasu, Y.; Fahey, M., 1997, The evaporation behav- mental handbook—Effects of mining on the environment and ior of tailings. In Tailings and mine waste, ‘97: A. A. Balkema, American environmental controls on mining: Imperial College p. 137-146. Press, p. 583-591. Norman, D. K.; Raforth, R. L., 1995, Cyanide heap leaching—The Filipek, L. H.; Gusek, James; Wildeman, Thomas, 1992, Wetland de- process, environmental problems, and regulation in Washington: sign for hazardous waste/mining operations [course notes]. In Washington Geology, v. 23, no. 1, p. 30-41. Mining, exploration and the environment: Society of Mining En- Norman, D. K.; Wampler, P. J.; Throop, A. H.; Schnitzer, E. F.; gineers North Pacific; Association of Engineering Geologists, 1 Roloff, J. M., 1996, rev. 1997, Best management practices for re- v. claiming surface mines in Washington and Oregon: Washington Gipson, A. H., Jr., 1998, Tailings disposal—The last 10 years and fu- Division of Geology and Earth Resources Open File Report 96-2, ture trends. In Tailings and mine waste, ‘98: A. A. Balkema, 1v. p. 127-135 Ouellet, J.; Benzaazoua, M.; Servant, S., 1998, Mechanical, minera- Golder Associates, 1996, Evaluation of surface paste placement as an logical and chemical characterization of a paste backfill. In Tail- alternative tailings management technology: Golder Associates ings and mine waste ‘98: A. A. Balkema, p. 139-146. Ltd. Sudbury [under contract to] ASARCO, Inc., Rock Creek Price, W. A., 1997, Draft guidelines and recommended method for the Unit, Noxon, Montana, 1 v. prediction of metal leaching and acid rock drainage at mine sites Harmon, M. E.; Franklin, J. F.; Swanson, F. J.; Sollins, P.; Gregory, S. in British Columbia: British Columbia Ministry of Employment V.; Lattin, J. D.; Anderson, N. H.; Cline, S. P.; Aumen, N. G.; Se- and Investment, Energy and Minerals Division, 159 p. dell, J. R.; Lienkaemper, G. W.; Cromack, K., Jr.; Cummins, K. Price, W. A.; Errington, John; Koyanagi, Victor, 1997, Guidelines for W., 1986, Ecology of coarse woody debris in temperate ecosys- the prediction of acid rock drainage and metal leaching for mines tems. In Macfadyen, Amyan; Ford, E. D., editors, Advances in in British Columbia—Part I. General procedures and information ecological research, v. 15, p. 133-276. requirements. In Proceedings for the Fourth International Confer- Henderson, M. E.; Hutchison, I. P. G.; Poulter, Don; van Zyl, D. J. A., ence on Acid Rock Drainage, Vancouver, B.C., Canada, p. 1-14. 1992 Closure requirements. In Hutchison, I. P. G.; Ellison, R. D., Price, W. A.; Morin, Kevin; Hutt, Nora, 1997, Guidelines for the pre- editors, Mine waste management—A resource for mining indus- diction of acid rock drainage and metal leaching for mines in Brit- try professionals, regulators and consulting engineers: Lewis ish Columbia—Part II. Recommended procedures for static and Publishers, p. 393-451. kinetic testing. In Proceedings for the Fourth International Con- Hutchison, I. P. G.; Ellison, R. D., 1992, editors, Mine waste manage- ference on Acid Rock Drainage, Vancouver, B.C., Canada, p. 15- ment—A resource for mining industry professionals, regulators 30. and consulting engineers: Lewis Publishers, 654 p. Rankin, M.; Miller, T.; Petrovic, S.; Zapf-Gilje, R.; Davidson, S.; Johnson, J. M., 1997, Tailings disposal design. In Marcus,J.J.,editor, Drysdale, K.; van Zyl, D. J. A., 1997, Submarine tailings dis- Mining environmental handbook—Effects of mining on the envi- charge—Optimizing the evaluation and monitoring process. In ronment and American environmental controls on mining: Impe- Tailings and mine waste, ‘97: A. A. Balkema, p. 303-316. rial College Press, p. 428-444. Redente, E. F.; Baker, D. A., 1996, Direct revegetation of mine tail- Kleinmann, R. L. P., 1997, Mine drainage systems. In Marcus,J.J., ings—A case study in Colorado. In Planning, Rehabilitation and editor, Mining environmental handbook—Effects of mining on Treatment of Disturbed Lands, Billings Symposium, 1996, the environment and American environmental controls on min- p. 183-191. ing: Imperial College Press, p. 237-244. Richardson, G. N., 1994, Geosynthetics in waste containment sys- Maltby, Van; Eppstein, L. K., 1996, Conclusion of long-term field tems: Mining Design, Operations, and Closure Conference, 1994, scale study of the use of paper industry residuals as hydraulic bar- [9 p., unpaginated]. rier material in landfill cover systems. In 1996 International Envi- Ritcey, G. M., 1989, Tailings management—Problems and solutions ronmental Conference and Exhibits, [8 p. unpaginated]. in the mining industry: Elsevier, 970 p. Marcus, J. J., editor, 1997, Mining environmental handbook—Effects Scheetz, B. E.; Menghini, M. J.; Hornberger, R. J.; Owen, T. D.; of mining on the environment and American environmental con- Schueck, Joseph; Giovannitti, Ernest, 1998, Beneficial use of coal trols on mining: Imperial College Press, 785 p. ash in mine reclamation and mine drainage pollution abatement in McGee, Sandra; Elliott, David; Taylor, Mark; Nilsson, Kent; Blair, Pennsylvania. In Tailings and mine waste, ‘98: A. A. Balkema, Ted, 1996, Landfill closure by capping with dewatered waste- p. 795-805. water treatment solids. In 1996 International Environmental Con- Schoenbrunn, F.; Laros, T. J., 1997, Dry tailings disposal using Hi- ference and Exhibits, p. 87-91. Density thickening. In Tailings and mine waste, ‘97: A. A. McLendon, Terry; Coleman, Jeffrey; Shepherd, T. A.; Nelson, R. E., Balkema, p. 357-363. 1997, The inclusion of biointrusion considerations in the design Sengupta, M., 1993, Environmental impacts of mining—Monitoring, of the reclamation cover for the DMC tailings impoundments. In restoration, and control: Lewis Publishers, 494 p. Tailings and mine waste, ‘97: A. A. Balkema, p. 267-281. Miller, G. C., 1997, The environmental issues in mining. In Marcus, J. J., editor, 1997, Mining environmental handbook—Effects of

Washington Geology, vol. 26, no. 2/3, September 1998 41 Scherer, George; Zabowski, Darlene; Everett, Richard, 1996, Nutri- ference and the Third International Conference on the Abatement ent content and survival of three native species on ameliorated of Acidic Drainage, Pittsburgh, PA, p. 122-127. abandoned copper mine tailings at Holden, WA: Association of U.S. Environmental Protection Agency, 1989, Technical guidance Abandoned Mine Land Programs, 18th Annual Conference, p. 35- document—Final covers on hazardous waste landfills and surface 43. impoundments: U.S. Environmental Protection Agency Office of Smith, Adrian; Mudder, Terry, 1991, The chemistry and treatment of Solid Waste and Emergency Response, EPA/530-SW-89-047, cyanidation wastes: Mining Journal Books, 345 p. 39 p. Smith, Adrian; Struhsacker, D. W., 1988, Cyanide geochemistry and U.S. Environmental Protection Agency Office of Compliance, 1995, detoxification regulations. In van Zyl, D. J. A.; Hutchison, I. P. Profile of the metal mining industry: U.S. Environmental Protec- G.; Kiel, J. E., editors, Introduction to evaluation, design and op- tion Agency, 123 p. eration of precious metal heap leaching projects: Society of Min- van Zyl, D. J. A., 1997, Liner design principles and practice. In Mar- ing Engineers, Inc., p. 275-292. cus, J. J., editor, 1997, Mining environmental handbook—Effects Steffen, Robertson, and Kirsten, 1989, Draft acid rock drainage tech- of mining on the environment and American environmental con- nical guide; Volume I—British Columbia Acid Mine Drainage trols on mining: Imperial College Press, p. 421-428. Task Force Report: BiTech Publishers Ltd., 1 v. Welch, D. E.; Firlotte, F. W., 1989, Tailings management in the gold Strachan, Clint; van Zyl, D. J. A., 1988, Leach pads and liners. In van mining industry. In Chalkley, M. E.; Conard, B. R.; Lakshmanan, Zyl, D. J. A.; Hutchison, I. P. G.; Kiel, J. E., editors, 1988, Intro- V. I.; Wheeland, K. G., editors, Tailings and effluent manage- duction to evaluation, design and operation of precious metal ment; Proceedings of the International Symposium on Tailings heap leaching projects: Society of Mining Engineers, Inc., p. 176- and Effluent Management, Halifax, August, 1989: Pergamon 202. Press, p. 7-22. Stormont, J. C.; Morris, C. E.; Finley, R. E., 1996, Capillary barriers Williamson, N. A.; Johnson, M. S.; Bradshaw, A. D., 1982, Mine for covering mine wastes. In Tailings and mine waste, ‘96: A. A. wastes reclamation—The establishment of vegetation on metal Balkema, p. 201-210. mine wastes: Mining Journal Books, 103 p. Tremblay, R. L., 1994, Controlling acid mine drainage using an or- Young, D. K., 1997, Feasibility analysis of paste landfill for the Rock ganic cover—The case of the East Sullivan Mine, Abitibi, Que- Creek Project, Noxon, Montana, ASARCO Inc.: Mine Opera- bec. In International Land Reclamation and Mine Drainage Con- tions, Design and Closure Conference ‘97, 19 p. n

BOOK REVIEW: West Coast Fossils—A guide to the ancient life of Vancouver Island

by Rolf Ludvigsen and Graham Beard its educational text as well as inspiration for a trip to Vancou- Harbour Publishing ver Island. The generalized geologic maps for the several peri- Box 219, Madeira Park, BC V0N 2H0 ods represented on the island should make it possible to de- 1997 (revised edition), 216 p. velop a rewarding itinerary. Paperback, $18.95 U.S. and Canadian Users of this thoroughly portable and well-designed guide will need little more than patience and sharp eyes to enjoy the he many fans of the first edition of this fine guide will be hunt. Considering the sizes of some of the new finds, we need Tglad to see the additions of major taxa to the record of fos- to travel with hand lenses at the ready. While the geology of sils in an area just to the north of Washington. The preface to Vancouver Island is not very like that of adjacent Washington, this edition indicates that there are now Carboniferous trilo- the message is clear: Go look. There is much yet to learn about bites, a pterosaur, Cretaceous beetles, articulated fish (Trias- former life nearly everywhere you go. sic), and dinosaurs (a tooth), not to mention Middle Jurassic Kitty Reed fossils and significant new finds of Mesozoic plant remains, all identifications vetted by current experts. Interestingly, credit for most of these discoveries goes to amateur paleontologists. And amateurs get thanks for helping save material at a Creta- SUMMER INTERNS ceous plant locality. To cover the additions, the book has been expanded to show We are fortunate to have two students from Community Youth examples of the new taxa. The many new (and high quality) Services working with us this summer. They have each done an photos augment the book’s clear and helpful information about excellent job for us, and we will be sad when their program is identification as well as histories of the taxa. The remaining finished. text is verbatim from the earlier edition, just slightly rear- Myhanh Tran is working as a clerical assistant this summer. ranged. She was born in Viet Nam, but has lived here for six years. Readers who are not familiar with formation or geographic Myhanh will be a senior at North Thurston High School and is names can check out the helpful stratigraphic section on p. 43. interested in the computer field. She plans to attend South That will make it clear that it is, for example, Protection Forma- Puget Sound Community College and the University of Wash- tion at Nanaimo, and Lambert Formation at Collinshaw Point; ington. commas in the captions might help. MiMi (Melissa) Beach is working part-time in the Geology This revised edition appears only three years after the first. library and the rest of the time as a clerical assistant. She will The authors deserve our thanks for bringing out the important be a sophomore at New Century High School. She is very in- new material so promptly. This book would be an excellent ad- terested in environmental science and hopes to make it a career dition to the region’s libraries and to personal collections for someday.

42 Washington Geology, vol. 26, no. 2/3, September 1998 Surface Mine Reclamation Awards

David K. Norman Washington Department of Natural Resources Echo Bay Division of Geology and Earth Resources Republic Minerals PO Box 47007, Olympia, WA 98504-7007

n 1996, the Department of Natural Resources (DNR) estab- Ilished annual awards to recognize outstanding reclamation Spokane of surface mines. These awards honor permit holders who re- Seattle Quincy claim mines in an exemplary manner. Awards also recognize Aberdeen Celite reclamation efforts on sites mined prior to 1971, before the diatomite mine Northwest Surface Mine Reclamation Act [RCW 78.44] became law. Rock, Inc. The mines to which awards have been given demonstrate one or more of the following : z Innovation in reclamation, such as creating unique wetlands, new approaches to enhancing wildlife and fish habitat, or imaginative shaping of topographic elements. Figure 1. Location of mine sites in relation to the nearest towns. z Voluntary reclamation of mined land that is exempt from reclamation under the Act. z Carefully planned and executed use of native plant species Wayne Pederson, hydrogeologist, in revegetation. Washington Department of Ecology, z Innovative research and approaches to reclamation that can Brad Biggerstaff, engineering geologist, GeoResources, be applied at other mines. Dave Mann, president, z Attention to preserving water quality and preventing Washington Environmental Council, erosion. Ray Lasmanis, manager, DNR z Orderly segmental mine development that results in high- Division of Geology and Earth Resources, and quality reclamation. Scott Nicolai, fisheries biologist, z A consistent, long-term commitment to reclamation. Yakama Indian Nation. z Methods that enhance the environment and reduce Division of Geology and Earth Resources reclamation liability, such as mining to a final slope. Reclamation Award Judges for the Division of Geology and Earth Resources Celite Corporation is the winner of the Division of Geology Reclamation Awards for 1998 are: and Earth Resources Best Reclamation Award for their efforts at their diatomite pit in sec. 17, T18N, R23E, near Quincy

Figure 2. State Geologist Ray Lasmanis presents the DNR Reclamation Award to Bob Katsiouleris, Celite Corporation plant and mine manager, on May 21 at the site of the reclaimed diatomite mine. The crew responsible for this work looks on. From left to right are Aron Bennett, plant safety manager; Mike Wells, quarry maintenance; Bob Katsiouleris, plant and mine manager; Lane Wyman, quarry operator; Ray Lasmanis, State Geolo- gist; Dave Norman, DNR Chief Reclamationist; Steve Vreeman, quarry operator; Gary Collins, quarry operator; Howard Vincent, quarry operator; Arlen Solders, truck driver; Gary Moore, quarry operator; Randy Rutherford, quarry operator; Louie Hoglund, quarry lead operator; Ron Massey, quarry operator. This auger scraper is used in reclamation of the pits; this machine can remove thinner layers of soils than a typical scraper without mixing or degrading poor-quality subsoils.

Washington Geology, vol. 26, no. 2/3, September 1998 43 Figure 3. (Top photo) The Quincy operation before reclamation in 1971. These are the diatomite pits in sec. 17, T18N, R23E near Quincy (Fig. 1). The road in the foreground is also shown in the bottom photo. (DNR photo.) (Bottom photo) Most of the diatomite pits shown in this 1998 photo have now been reclaimed. Previously reclaimed areas have returned to sagebrush steppe through a natural floral succession. In recently reclaimed areas, the company planted grasses to stabilize the soils. (Celite Corporation photo.)

44 Washington Geology, vol. 26, no. 2/3, September 1998 MINING RECLAMATION

TOPSOIL

OVERBURDEN TOPSOIL

MINERAL DEPOSIT DIATOMITE BACKFILL

Ore Planning Topsoil Overburden Ore stockpile Backfill Topsoil Revegetation

Detailed grid Removed topsoil Overburden Ore removal is Ore stockpiles are Backfilling occurs A specialized Crested wheat grass drilling identifies is stockpiled to a materials (loess- scheduled and built and main- year round as auger-scraper is or other suitable the limits of the backfilled area based soils, basalt, planned to meet tained inside or production dedicated for vegetation is mineral so that scheduled for and caliche are plant production immediately schedules dictate topsoil planted. Various reclamation can reclamation. The removed and requirements. adjacent to open and as new replacement. seed application begin with the removal effort stockpiled with Continuous pits. This mining areas are Discing may be techniques are used mine planning concentrates on dozers and push quality sampling minimizes wind opened. Back- necessary to during the optimum process. Drill data the top 12 inches scrapers. Piles are throughout the erosion and filled areas are prepare the soil planting season, is collected and independently, capped with mining process potential impact to capped with basalt for seeding. just prior to used for modeling preserving crushed basalt to categorizes ore surrounding lands. to prevent wind Compaction by precipitation. and computer nutrients and reduce wind types and reduces erosion. heavy equipment scenarios. microbial activity. erosion. waste. following application is minimized.

Figure 4. Mining and reclamation sequence at the Celite mine. Topsoil, overburden, and diatomite are removed in separate steps so as to ensure reclamation success. After the diatomite is removed, pits are backfilled to approximately their original contours, then the topsoil is replaced and re- vegetated. This sequence of processes has led to high-quality reclamation. (Redrawn from Celite Corporation illustration.)

(Figs. 1, 2) in Grant County. Celite Corporation was nominated by Lorraine Powell, DNR Southeast Region Reclamation In- spector. The company has reclaimed nearly 58 acres of this arid post-mining terrain to a high standard (Fig. 3). Segmental rec- lamation and separation and replacement of topsoil and over- burden after the diatomite has been removed are the keys to their success. The work included voluntary reclamation of 13 acres exempt from state reclamation requirements because ex- cavation there preceded enactment of the Surface Mine Recla- mation Act. Celite’s open-pit mining practices concurrent reclamation that involves salvaging topsoil and stripping caliche and basalt from above the diatomite deposits (Fig. 4). All salvaged over- burden is segregated and preserved for backfilling and recla- mation. Topsoil is re-spread over the mined area, fertilized, and seeded. The re-established topography approximates the Figure 5. Scanning electron micrograph of diatoms from the Celite original contours and blends well with the surrounding area. pits. These algae have many more shapes than the “pill box” type This level of reclamation is well beyond requirements of RCW shown here. Magnification is about 500X. The diameter of one of the 78.44. diatoms shown in the center is 70 microns. (Celite Corporation photo.) Successful revegetation has stabilized the mined areas. In many places, one cannot tell the difference between undis- turbed land and reclaimed land. All of the judges were particu- of mining here to reflect Celite’s policy of corporate responsi- larly impressed with this outstanding effort. bility in mine planning, operations, and restoration. Today, Corporate Responsibility Celite begins every mine project with careful reclamation plan- ning that is part of the strategy for opening a new pit, mining, Celite Corporation is a worldwide mining and minerals com- and backfilling. Reclamation continues with monitoring and pany and a leading producer of filter material and mineral fill- maintenance long after topsoil has been replaced and seeded. ers. The corporate headquarters are in Lompoc, California. Concerted reclamation of the depleted diatomite mine areas be- Diatomite gan with Celite’s acquisition of this mine in 1991. The man- Diatomite is made up of the empty shells or frustules of dia- aged transition from mining to reclamation modified all phases toms (Fig. 5), now the dominant photosynthetic organism in

Washington Geology, vol. 26, no. 2/3, September 1998 45 Figure 6. Fourth-grade students learn about tree planting and recla- mation as part of Arbor Day activities. Here students plant evergreen trees after the site was graded. This annual event at Echo Bay’s Kettle River operation at Republic is attended by hundreds of area students, where they learn about the environment, guided by Steve McIntosh. Figure 7. Steve McIntosh (left, in baseball cap), environmental man- ager of Echo Bay Minerals Kettle River Operation, instructs students in most of the world’s aquatic environments. More than 5,600 reseeding of slopes with grass and legumes to prevent erosion. modern species of these single-celled golden brown algae, a type of phytoplankton, are known. Individuals of the larger species are barely visible without a magnifying glass. Most numerous local events and groups. This company has donated frustules are round, but some are elongated or branched or tri- money to Republic and Curlew high schools and the local li- angular. brary and has supported and made funds available to many Diatoms first entered the geologic record in marine upper community functions. For several years, the company has Lower Jurassic rocks (Harwood and Nikolaev, 1995); they made Arbor Day a special event. Federal, state, and local gov- very rapidly increased in both diversity and abundance and are ernment agencies and Echo Bay Minerals have joined forces to common in some mid-Cretaceous rocks. The oldest known created a unique learning experience for hundreds of local continental diatoms are middle Eocene (Bradbury and Krebs, fourth graders. The children learn about erosion, tree planting 1995). By the Miocene, diatomites had become common in the (Fig 6), water pollution, impacts of trails and roads, fire fight- western United States. Local development of species is a com- ing, reclamation (Fig. 7), fish and wildlife habitat, and other as- mon feature in large Tertiary lakes. pects of taking care of the environment. In Washington, diatom remains slowly settled to create Corporate Responsibility thick layers on the bottoms of lakes that repeatedly developed on the surface of the Miocene basalt flows. These deposits are Echo Bay Minerals Co. has maintained good community rela- brilliant white and friable. The diatoms were able to obtain the tions over the years. The company has continued to perform silica needed for their shells from these volcanic rocks. Geo- outstanding reclamation and remained a loyal, involved neigh- logic evidence shows that some of these lakes dried up prior to bor despite a severe down-turn in the price of gold. another outpouring of basalt during the Miocene; in other lakes, the advancing lava flows cooled as basalt pillows. Division of Geology and Earth Resources Diatomite is used for filtration in food processing, render- Reclamationist of the Year Award ing, and in making wine, beer and juices and as mineral fillers, Harlan “Bummie” Stoken (Fig. 8) is the winner of the Reclama- coating agents, and carriers. tionist of the Year Award. Mr. Stoken is a heavy equipment op- erator for Northwest Rock, Inc., of Aberdeen in Grays Harbor Division of Geology and Earth Resources County. He is in large part responsible for the high quality of Good Neighbor Award reclamation at several of the company’s sand and gravel pits. Echo Bay Mineral Co. is the winner of the Good Neighbor Mr. Stoken was nominated by Stephanie Zurenko, DNR Cen- Award. Echo Bay Minerals has made tangible contributions to tral Region Reclamation Inspector.

46 Washington Geology, vol. 26, no. 2/3, September 1998 Figure 8. Harlan “Bummie” Stoken next to the bulldozer he uses to reclaim gravel pits as a heavy equipment operator for Northwest Rock, Inc., of Aberdeen in Grays Harbor County. Bummie’s attention to sloping and contouring help in achieving high-quality rec- lamation.

Mr. Stoken has been operating heavy equipment most of his adult life, and as a consequence of his grading skills and an outstanding “eye” for drainage and topography, he has made landforms that look natural from the otherwise linear features of the mines. At the Brooks (Fig. 9) and Sands mines in Grays Harbor County, Bummie has created wetlands that have sinuous shorelines and islands that will be bene- ficial and productive sites for water- fowl and other wildlife. The judges chose to recognize Mr. Stoken for his sense of what he can do to work with nature in mines. Individual Responsibility A Mr. Stoken takes great care and pride in his reclamation efforts. His ability to imagine the landform and then create it is the reason why Northwest Rock chooses Bummie from among their many equipment operators to do recla- mation work.

References Cited Bradbury, J. P.; Krebs, W. N., 1995, Fossil continental diatoms—Paleolimnology, evolution, and biochronology. In Blome, C. D.; Whelan, P. M.; Reed, K. M., editors, Siliceous microfossils: The Paleontological Society Short Courses in Paleontology Number 8, p. 119-138. Harwood, D. M.; Nikolaev, V. A., 1995, Cretaceous diatoms—Morphology, tax- onomy, biostratigraphy. In Blome, C. D.; Whelan, P. M.; Reed, K. M., editors, Siliceous microfossils: The Paleonto- B logical Society Short Courses in Pale- ontology Number 8, p. 81-106. n

Figure 9. A. The northwest shore of the Brooks mine in Grays Harbor County before reclamation. Steep straight banks provided little fish and wildlife habitat in this gravel pit lake. B. The same bank after grading but before revegetation. Bummie Stoken re- duced the slopes on this now-sinuous shore- line. This will promote a diverse flora and im- prove both aquatic and upland habitats. (Photos by Stephanie Zurenko, DNR Central

Washington Geology, vol. 26, no. 2/3, September 1998 47 Early Miocene Trace Fossils from Southwest Washington

Keith L. Kaler 1322 S. Cherry; Olympia, WA 98501 [email protected]; [email protected]

he area surrounding Mount St. Helens has received much has been possibly reset by the Spirit Lake pluton and may be a Tscientific attention since the eruption of 1980. The Wash- minimum age. Therefore, it appears that the tracks were made ington State Department of Natural Resources (DNR) and the by early Miocene animals (late Arikareean mammal stage). U.S. Geological Survey (USGS) are two major contributors to this research. In 1985, while employed with DNR’s Division of Geology and Earth Resources, I attended a field trip sponsored by the USGS to study the geology around Mount St. Helens and Mount Adams. During the trip, I recovered a slab of rock bear- WASHINGTON ing many impressions made by vertebrates (Figs. 1, 2). On sub- sequent trips to the site, I collected more trackways, isolated Seattle footprints, and other ichnofossils (trace fossils). 0 40 80 mi Trace fossils from this region of Washington are worth studying for several reasons. The geology of the Miocene vol- 0 60 120 km caniclastic rocks between Mount St. Helens and Mount Adams W is complex, and few fossils have been found. These trace fos- F sils and their abundance apparently indicate a varied vertebrate Mount fauna. Vertebrate ichnofossils may give geologists who study St. Helens these rocks better time and stratigraphic controls. Evidence of a well-established animal community may improve estimates of the deposition rates of these sediments through comparisons OREGON to the rates of animal repopulation in the area affected by the J eruption of Mount St. Helens. At the very least, these traces of Figure 1. General location of the ichnofossils. F, ichnites; W, Wildcat Miocene vertebrates may help fill gaps in our knowledge of pa- Creek local fauna, Oligocene (early Arikareean; Lander and Swanson, leontology and paleoecology in the western United States. This 1989); J, location of the nearest known Miocene vertebrate fauna (John report is a preliminary attempt to identify the makers of the Day Formation; late Hemingfordian) near Warm Springs, Oregon. ichnofossils. Almost all material was recovered from talus at the base of o a 15-m-high cliff of volcaniclastic rock, composed of beds of 40 fine- to coarse-grained, angular to subangular fragments of volcanic detritus. Weathered and altered pumice is a common component. The beds range from less than 1 to more than 30 cm in thickness and are well lithified. In some beds, there is a 3- to 10-fold change in thickness within a distance of less than 5 m. Cross bedding has not been observed. A thin bed of tuff(?) was recently discovered approximately 2 m above the trace fossil interval (Fig. 3). The entire sequence continues downslope (east) from the cliff. A rough measurement indicates that this sedimentary unit is more than 35 m thick. The rocks exposed along the fossil-bearing cliff are moderately folded but not overturned. Rocks downslope and to the north of the outcrop are not folded and dip less than 10 degrees. A creek divides more horizontal rocks on the north from the fossiliferous and folded sediments to the south and may indicate the presence of a fault. The fossil-bearing stratum can be traced for more than 10 m before it disappears under talus and vegetation. R. C. Evarts and R. P. Ashley mapped the geology of the area that contains the track-bearing rocks (Evarts and Ashley, 1993; Evarts and others, 1987); they include the rocks in their widespread Tvs unit (“volcaniclastic sedimentary rocks of in- ferred epiclastic origin”). (See Walsh and others, 1987.) A ra- diometric age estimate of 22.6 ±0.6 Ma was obtained from a pyroxene andesite interbedded with these sediments about 4 to 5 km from the site. Evarts and Ashley further state that this date

48 Washington Geology, vol. 26, no. 2/3, September 1998 TRACKWAY 1-2 F B TRACKWAY 1-3 G E A D C D

C A B F E H G 5cm

exposed stratum J below the subtraces I L K

M N SUBTRACES

K impressed (exogenic) I surface J L

G

E H F

SUBTRACES C A D B

TRACKWAY 1-4

TRACKWAY 1-1

strata containing subtraces

TRACKWAY 1-5

Figure 2. Sketch and photo (facing page) of slab 1. The pitted area is the original impressed surface; the smooth area on the bottom of slab is the stratum bearing the subtraces. Note the discontinuity of craters on the top surface. Outlines of the carnivore(?) subtraces have been exaggerated to show the author’s interpretation of the footprint morphology. Arrows indicate the direction the trackway maker was traveling. Arrow in the photo with the degree measurement indicates the approximate declination of light from a 100-watt bulb relative to the slab surface; the bar is 10 cm long.

Washington Geology, vol. 26, no. 2/3, September 1998 49 DESCRIPTION OF MATERIAL Almost all examined ichnofossils were formed at the surface/ air boundary (ex- ogenic) and are molds. Rare isolated casts and counter slabs of rock have been found. This scarcity of natural casts is probably due to the fragility of the 5-mm layer of sediment that filled the impressions. More than 75 percent of all the ver- ? tebrate impressions belong to fewer than ten trackways from which the gait and footprint size can be measured with B reliability. Preservation of the foot- C prints is good, although the larger carnivore-like traces are the least reli- A able for identification. Nonetheless, D enough material is present to indicate the relative size of the animals, their gait, and the basic morphology of their Figure 3. Photo of the outcrop taken in 1988. A, location of slab 2 (Fig. 9); B (mold),C (cast), lo- cation of bird tracks (Fig. 13); D, location of slab 1 (Fig. 2). White line indicates the location of the feet. fossil-bearing stratum. View is to the west. All impressions can be grouped into five categories distinguished by shape and size—two are mammalian in origin, one is from a bird, and has not been preserved in any impression. Splashed sediment another may represent the markings made by a lizard-like ani- and water thrown up by the activity of animals is a possible ex- mal. The fifth group was produced by water. planation, but I have found no supporting evidence. Available material suggests that the impressions were pro- Impressions Made by Water duced by falling water. A raindrop origin appears to be un- The slab in Figure 2 shows numerous round to subround im- likely because there is an uneven distribution of the impres- pressions. They range from 0.3 to 1.3 cm in diameter and are sions (Fig. 2). Raindrops can be expected to fall at random and 0.1 to 0.5 cm deep. There is an approximate alignment of simi- to evenly cover a surface unless blocked by an obstruction. I larly sized craters. The impressions were probably created by have observed similar configurations elsewhere on other slabs water falling from directly above the site or from splashed ma- representing the same stratum. A likely origin appears to be terial (sediment and water) thrown up by passing animals. Fal- water falling from a branch of an overhanging tree or bush. Fig- ling water is suspected because of the random pattern of the im- ure 4 shows the pattern made on fine beach sand from water fal- pressions and the nearly circular outline. I considered impact ling off a deciduous shrub in late fall after a heavy fog; note the by falling lithic detritus such as pumice, but foreign material definite alignment of craters of similar size. Their size appears to be related to the diameter of the limbs on which the water droplets formed. Larger limbs and branches were observed to produce larger drops, smaller branches produced smaller drops. The complexity of the drip pattern on the beach surface is related to the complex intergrowth of the limbs of the shrub. This theory is made more plausible by the presence of fossil woody debris and leaf fossils in the talus. Plant remains tend to be poorly preserved, but I have found an outline of a sheathed bundle of (five?) pine needles.

Impressions Produced by Mammals The majority of the trackways were made by mammals. Verte- brate classes such as the reptiles and amphibians were ruled out; the exception is the single trackway diagrammed in Figure 14. This classification is based on the measured pace angula- tion and footprint morphology. Digits of amphibians and rep- tiles tend to be of disproportionate length (Brown and others, 1984; Murie, 1974) and may be webbed with prominent claws. These characteristics are in contrast to the track producers, which were animals with an efficient style of locomotion. The limbs must have been placed well under the body, the straddle width was small (see Fig. 8), and the digits in all tracks are of Figure 4. Impressions produced by water dropping off a leafless similar size. Webbing is absent, and claw traces, where pres- shrub onto beach sand. Photo was taken in late afternoon after a heavy ent, are small. fog lifted. Linear pattern of the impressions reflects the size and loca- Pace angulation is a measure of locomotion efficiency and o tion of the overhanging limbs. North is to the top; light incidence is 35 represents the placement of the legs and feet under the body from the west. Quarter for scale. during locomotion. The method for measuring this angle is dia-

50 Washington Geology, vol. 26, no. 2/3, September 1998 grammed in Figure 5. This can be understood by comparing the midline placement of the limbs of a turtle and lizard with those of a dog. According to Peabody (1959), a turtle has a pace angle of about H 65o and a lizard about 85o. These animals walk in a sprawled 10 cm manner, with the limbs spread out from the body and the feet G far from the midline of the trackway (Fig. 5). On the other hand, all fossil mammal trackways have pace angles greater than 150o. The pace angle of a modern dog is about 150o,re- external flecting the placement of the limbs and feet close to the midline trackway of the trackway. There can be little argument that the dog walks width in a more graceful and efficient manner than a lizard or turtle. F Figure 6 shows a scheme for connecting the ichnites (fossil tracks) to their probable mammalian creators. The distinguish- pes ing features for tentatively identifying the makers of the larger manus/pes pair ichnites as carnivores are the apparent presence of digitigrady, TRACKWAY 2-1 E inferred size of the track producer, and the number of digits in manus the manus (front feet) and pedes (hind feet), respectively. Digitigrady is a term used to refer to an animal’s ability to walk on its toes, or digits, like a cat, instead of on the sole of its foot, like a human or a bear (plantigrade). In the large ichnites, digit- igrady is suggested by the impression of a plantar pad (the soft, triangular pad on the bottom of the cat’s or dog’s foot behind the digits) and the lack of a posterior metatarsal or metacarpal D pad (the long and broad “sole” pattern of the human foot). The smaller ichnites on the slab were likely made by ro- dents on the basis of the inferred size of the animal, its track C pace angulation size, stride length, and gait. Lagomorphs (rabbits and their relatives) were ruled out because of their large size and special- ized hind feet. The only known Miocene insectivore families stride plantar that could be candidates are the Erinaceidae (hedgehogs) and pad oblique pace the Soricidae (shrews). The erinaceids probably can be ex- cluded from consideration because of their large size and be- cause modern representatives do not make the track patterns observed in the ichnites. The soricids are, in general, too small B to have produced the stride and straddle lengths observed pace (Murie, 1974; Brown and others, 1984). Thus, the insectivores may be, with some confidence, excluded from consideration. A Smaller (Rodent?) Mammal Tracks Tracks made by rodent-like animals make up the majority of the ichnites. I measured five exogenic trackways in which gait Figure 5. Diagram of trackway 2-1 (Fig. 8) showing the methods used and descriptive measurements could be determined. I assume for measuring the pertinent characteristics of the larger trace-fossil pro- these trackways were formed by animals of similar build, per- ducers. The shapes of the footprints have been exaggerated to show haps members of the same taxon, on the basis of similarities in the author’s interpretation of the footprint morphology. Note lack of claw markings, a felid attribute. their construction. More than one individual may have been re- sponsible for track creation, however. Nearly all rodent-like trackways exhibit the gait pattern responds to an increase in stride length and may show great that Hildebrand (1974) defines as the half-bound. In this gait, variation within a trackway. In trackway 2-2 (Fig. 9), the varia- either the right or left manus takes the lead and strikes the sur- tion between two consecutive measurements is nearly 51 per- face before the other three feet. The pedes appear to strike the cent (3.4 cm). The range of the intergroup distance is 7.0 to ground in unison. This is common in present-day rodent loco- 12.3 cm. motion. Trackway 1-1 shows a different gait pattern, called the Straddle width variation amounts to less than 0.2 cm (6 per- bound, where hind- and forefeet strike the ground in unison, cent) in a single trackway. When all trackways are considered, and there is no leading manus (Fig. 7). It is also a gait com- the straddle width lies within a 3.2- to 3.8-cm range. monly observed in many types of small, fast-moving rodents. The five trackways are composed of 47 complete foot- The stride, measured from the extreme anterior of the left prints—23 pedes and 24 manus. The length of the pes (hind and right pedes, ranges from 8.5 (trackway 1-2) to 15.5 cm foot) ranges from 1.0 (trackway 2-3) to 1.9 cm (trackway 1-1) (trackway 1-1) (Fig. 2; refer to Fig. 8 for definitions). The aver- and may vary within a single trackway by as much as 40 per- age length is 11.7 cm. Stride length can vary by as much as 60 cent (trackway 2-2). The manus length ranges from 1.2 (track- percent within the same trackway. ways 2-2 and 1-3) to 1.8 cm (trackway 1-3). Variation in the The intergroup distance corresponds to the length of ex- length of the manus may be as much as 50 percent within a sin- tended flight (Gambaryan, 1974). It is the measure of distance gle trackway (1-3). This measurement is affected by the partial that a saltating (jumping) animal is airborne, with all four feet overlapping of adjacent footprints of different trackways. simultaneously off the ground. An increase of this length cor- There is no apparent correlation between the length and width

Washington Geology, vol. 26, no. 2/3, September 1998 51 of manus or pes; the width does not get 1 8 2 5 3 6 larger with an increase in the impression’s 9 7 length. There is no evidence of backfilling 4 of sediment from the track margins into the

impressions, so this is not likely a reason for Perissodactyla Artiodactyla Marsupalia Primates Proboscidea Insectivora Rodentia Lagomorpha Carnivora Small ichnites Large ichnites any discrepancy in the measurements. lll Except for trackway 2-3, all pedes show digitigrade distinct outward divarication (that is, they plantigrade ll llll l are directed away) from the main axis of unguligrade ll motion. The left pes (eight measurements) 10 divaricates 11°, and the right pes (four 1 or 3 major digits in manus and pes l measurements) divaricates 8°. I saw no cor- 2 or 4 major digits in manus and pes llll relation between the amount of the divarica- lllllll l tion and the length of the stride or extended 5 major digits in manus and pes flight. The lack of a correlation may be due body length usually less than 20 cm lll l to the small number of measurements; a body length rarely, if ever, less than 20 cm lllll l l more accurate representation would be at- tained by a longer continuous trackway. stride usually less than 6 cm l Many tracks have digit and claw impres- stride always greater than 8 cm lllll lll ll sions. Footprints in trackway 2-3 indicate lll l that both right and left manus have five dig- asymmetrical gait (ricochetal to bounding) its. A right pes in trackway 2-2 also has five symmetrical walking gait lllll l l digits. Individual casts of the rodent-like manus or pes may be less than 25 mm long ll l tracks that could not be matched to any trackway also indicate a similar phalangeal manus or pes greater than 25 mm long lllll l l (toe or digit) count. Assuming that the ani- 1 Odd-toed hoofed quadrupeds such as horses, tapirs, and rhinoceroses mals that made these impressions were all 2 Even-toed hoofed quadrupeds such as cows, pigs, and hippopotamuses of the same type, then the phalangeal for- 3 mula of the animal was 5/5—five digits on Mammals that rear young in a pouch on the mother’s abdomen, such as opossums and kangaroos 4 both pedes and manus. Humans, apes, and monkeys Differences in the widths of the straddle 5 Large tusked mammals with trunks and columnar legs, such as elephants, mammoths, and mastodons and the length of the pedes suggests that 6 Insect-eating mammals, such as moles, shrews, and hedgehogs more than one individual was responsible 7 Gnawing and nibbling mammals characterized by four continually growing incisors, such as mice, for all the rodent-like impressions. The dif- squirrels, chipmunks, and rats ference in length between the longest and 8 Mammals resembling rodents but having two pairs of upper incisors, such as rabbits, hares, and pikas shortest pes measurements of all examined 9 Meat-eating mammals such as dogs, cats, bears, seals, and weasels footprints amounts to 0.9 cm, a discrepancy 10 Manus, singular and plural, front foot; pes, singular of pedes, back foot/feet of 90 percent. Within any single trackway this discrepancy is less than 40 percent. Figure 6. Selected diagnostic trackway characteristics of the ichnites and the mammalian Straddle variation is as much as 19 percent orders that were present in North America in ArikareeantoBarstoviantime(lateOligoceneto between the smallest and largest measure- middle Miocene). Characteristics are based on modern and fossil animals. (Data from Sav- ments (3.2 to 3.8 cm) when all the track- age and Russell, 1983.) ways are compared. The straddle variation is not more than 6 percent within any single trackway. An animal’s body length can be estimated from the length size of the trackway producer. There is little doubt that the ani- of the pes impressions. Figure 10 is a graph of the pes length of mals were digitigrade quadrupeds. modern-day rodents. Even with an allowance for error in meas- In Figure 12 (center trackway, 2-1), the record of motion urements because of the nature of the sediment, the Miocene begins with the impression of a right manus and is followed by ichnite producers appear to have been about 10.0 to 15.0 cm seven more footprints; the last, a left pes, is incomplete. Each long, excluding the tail. This graph is not meant to imply any of the four feet is represented by two impressions. The foot- affinity of the track producers to recent animals but is simply to prints are grouped into pairs with the manus in front of the pes be used as a tool to determine body size. The important meas- of the same side of the body. These groups are spaced about urements are summarized in Figure 11. 8cmapart. On portions of footprints 2-1B and 2-1C (Figs. 5, 9), the Large Mammal (Carnivore?) Tracks topmost impressed surface is missing. Subtraces of approxi- The larger tracks are more questionable because their impres- mately 1 cm depth are exposed in an apparently very fine sions are not as distinct as the others. Important features such grained, dark-gray sandstone. Between this sandstone and the as the digit and pad impressions are vaguely defined when upper layer, where present, small cavities have formed; this viewed by direct, overhead light. Ridges separating these re- makes the top layer subject to breakage. The depth of the sub- gions are subtle but can be seen in low-angled (to the slab) traces is evidence that the sediment below this surface was still light. Also, these ridges can be easily traced by running the fin- wet when the tracks were formed. gers over them. By these means, sufficient information was at- Figure 5 gives track-measurement definitions for the larger tained to determine the existence of a footpad (plantar), the footprints. Six measurements of stride length could be made on style of gait, general size of the foot, and even the approximate trackway 2-1. The range is 32.1 to 34.1 cm, and the average is

52 Washington Geology, vol. 26, no. 2/3, September 1998 33.0 cm. Eleven oblique pace measurements TRACKWAY TRACKWAY TRACKWAY TRACKWAY TRACKWAY range from 15.7 to 18.4 cm, averaging 1-1 1-2 1-3 2-2 2-3 16.9 cm in length. Three separate track- breadth measurements fall within a range of O 7.2 ±0.2 cm. The step length, as measured from the most posterior point of track 2-1A to a similar position on 2-1F, is 54.1 cm. The M pace (step) angle was measured using six sets of points. This angle ranges from 152° to 165°, averaging 157°. L The two right manus impressions are K 5.3 cm wide and 4.3 cm long. This compares J well with the best preserved left manus (D), I which is 5.2 cm wide and 4.4 cm long. The L maximum length of all footprints, manus and N K pedes, corresponds to the long axis of what is I presumed to be the fourth digit. M I The two right pedes average 4.9 cm wide J and 3.3 cm long. Left pedes are 4.4 cm wide and 3.3 cm long. This discrepancy between 5cm break the length of right and left pedes is within the L in slab observed range of modern domestic carni- K vore (dog and cat) tracks. The manus are impressed nearly twice as I J deeply as the pedes (5 mm compared to 2 to 3 mm, respectively), consistent with observa- tions of many modern carnivores where the G H center of gravity is closer to the front limbs G G H than to the hind limbs. More weight is placed E F H E on the front limbs, which results in a deeper G impression. Both manus and pedes appear to F have four depressions, which were probably G H F F E E made by four digits. This is a common pha- E langeal count in many living carnivores. No F claw markings are visible on any of the dig- its, evidence that the carnivore-like track maker may have had retractable claws. According to Casamiquela and others (1987) and Mossman and Sarjeant (1983), C the gleno-acetabular distance is a measure- D C D C D C D ment that gives the approximate trunk length D A C of the animal (the length between the fore and A A hind limb attachments). This measurement is B A B B B B made from the distance between the mid- A points of two consecutive pedes and manus Figure 7. Comparison of selected rodent-like trackways drawn to the same scale. Arrow impressions. Two measurements were made points in direction of movement. Letters refer to track designations in Figures 2 and 9. Gaits on trackway 2-1. The distance between the of all trackways (except 1-1) are what Hildebrand (1974) calls the half-bound. Footprints in midpoint of manus D and F (most posterior trackways 1-1, 1-2, and 1-3 (solid outlines) occur on the original impressed surface. All foot- part of the plantar pad) to the midpoint of pe- prints on slab 2 (2-2, 2-3) are subtraces. des A and C is 21.7 cm. The other measure- 5 ment between manus F/G and pedes C/E is 6 20.0 cm, but it may be in error due to the relatively poor condi- 4 tion of tracks G and H. If these measurements are valid, then 2 the animal that produced these tracks would have been about 1 8 the size of a house cat. I observed gaits of living dogs and cats by dipping each of the animals’ paws into a different food coloring and having it 3 9 move over a long sheet of paper and board (Fig. 12, the outside trackways). It became apparent that the ichnite-producer was most likely employing a leisurely walking gait at the time of the 7 imprinting. This conclusion is supported by the works of Murie Figure 8. Diagram showing methods for measuring rodent-like track- (1974) and Halfpenney and Biesiot (1986) on recent quadru- ways (modified from Gambaryan, 1974). Large circles, pedes (hind peds. feet); small circles, manus (forefeet); 1, width of base of pedes; 2, width of base of manus; 3, length of extended flight; 4, fore pace; 5, hind pace; 6, crossed flight; 7, stride length; 8, straddle width; 9, grouping length.

Washington Geology, vol. 26, no. 2/3, September 1998 53 o 40 H TRACKWAY 2-3

A G B 10 cm D C

F I E G J L F E G F E H C H A D B

TRACKWAY 2-2 ? o D ? 35 I

K ? C

M TRACKWAY 2-1

O B

A

Figure 9. Photo and sketch of slab 2. (See Fig. 3 for its location on the talus pile.) Arrows indicate direction of movement of track producers. All footprints on the slab are subtraces. Figures 5 and 8 show the methods used for measuring the trackways in this study. Bar on photo is 10 cm long. Inset is a closeup of a right? pes showing the 4 digit component of a carnivore footprint. Bar is 5 cm long. This footprint is not from slab 2. Arrows with degree measurements indicate the approximate declination of the incoming light from a 100-watt bulb relative to the slab surface.

600

Figure 10. Graph showing body length (excluding tail) 500 ranges for extant North American rodent families when the pes (hind foot) length is known. Numbers in parentheses refer to the number of species per family considered. The relation of body length to pes length tends to be a linear 400 plot by family group. The patterned area is the approximate plot of the trace fossils and is not meant to imply taxonomic affiliation. (Data from Hall, 1981.)

300 GEOMYIDAE (25) SCIURIDAE (85)

Body length (mm) 200 CRICETIDAE (133)

HETEROMYIDAE (50) 100 ZAPODIDAE (5)

0 0 10 20 30 40 50 60 70 80 90 100 110 120 Pes (hind foot) length (mm)

54 Washington Geology, vol. 26, no. 2/3, September 1998 Bird Tracks TRACKWAY 1- TRACKWAY 1- TRACKWAY 1- TRACKWAY 2- TRACKWAY 2- The best bird trackway is shown 1 2 3 2 3 in Figure 13. The diagram is a gait style bound half bound half bound half bound half bound composite of two counter slabs. stride length 11.6 cm 12.8 cm 8.7 cm 10.4 cm Only the subtrace (cast) and the (each cycle— 10.8 cm 12.0 cm 15.6 cm 10.0 cm 14.4 cm bottom of the mold is visible. The pes to pes) 12.0 cm imprinted surface could not be ob- 3.6 cm (C–D) 3.3 cm (C–D) straddle 3.4 cm (C–D) 3.4 cm (C–D) served without damage to the 3.6 cm (G–H) 3.3 cm (G–H) 3.8 cm (C–D) (per grouping) 3.6 cm (G–H) 3.5 cm (G–H) trackway. Other bird footprints, 3.6 cm (K–L) 3.2 cm (K–L) on other slabs, have been discov- hind foot size ered but do not form any discerni- (average) 1.6 cm/0.8 cm 1.4 cm/0.8 cm 1.4 cm/0.9 cm 1.5 cm/1.0 cm 1.2 cm/0.9 cm ble trackway and could not be length/width matched to those in Figure 13. length of 11.0 cm (C–E) 7.0 cm (D–F) 6.7 cm (C–F) All impressions indicate an extended flight 7.8 cm (D–E) 8.2 cm (C–E) 12.3 cm (G–I) 7.6 cm (H–J) animal with three digits, presuma- (intergroup) 10.1 cm (G–J) bly digits 2, 3, and 4. I saw no 2(C) 13 (D) clear indication of digit 1, the hal- diverticulation of 15 (G) 13 (G) 20 (C) less than 2 hind foot from 12 (C) lux (“thumb”), in any footprint. 7(H) 7(H) 12 (G) (G and H) axis of motion Claw marks are intriguingly ab- 6 (I) 6(K) sent; most recent and fossil bird 4.4 cm (A–D) 3.8 cm (A–D) 6.2 cm (A–D) 4.5 cm (A–D) footprints show indentations grouping length 3.5 cm (E–H) 3.5 cm (E–H) 4.3 cm (A–D) 4.9 cm (E–H) 5.0 cm (E–H) made by claws (Brown and others, 4.0 cm (I–L) 4.0 cm (I–L) 1984). In contrast to many of the living non-wading birds, the dig- Figure 11. Summary of measurements of the rodent trackways. (Refer to Figs. 2, 8, 9, and 11.) its in the trackway are relatively short and broad. The width may DOMESTIC CAT TRACKWAY 2-1 DOMESTIC CAT reflect what Brown and others (1984) call smooth, lobate, digi- (pine board) (“wet” sand) (paper) tal webbing found in living grebes (Wallace, 1955). It is possi- ble that other types of webbed feet were responsible for the im- RP pressions. My observations of a domestic goose showed that RP RM indentations similar to the ichnites could also be produced by a LP H distally webbed foot impressed into soft mud. The bird’s weight and the soft sediment substrate allowed the centers of LM G RM the webbing to “bow up” between the digits, and a false “lo- bate” digit impression was formed. In many footprints the out- LP F RP line of this bowed-up webbing was not observed. LP Only three well-preserved fossil footprints allowed reliable E RM measurements. The digital length was measured from the rear of the imprint, in the approximate area of the metatarsal im- LM LM pression, to the tip of the digit. Only digit 3 showed a consistent length for all three tracks and ranged from 4.7 to 4.9 cm. The LP D lengths of digit 2 ranged from 3.2 to 3.7 cm, and digit 4 varied LM C from3.4to4.4cm.Divaricationbetweendigits2and4was RP very close in all tracks (106° to 109°). RP

Two measurements of stride and pace angulation measured RM from the most posterior point on the impression, show that pace RM B RP angulation was 160° and 169° and stride length ranged from 18.6 to 19.4 cm. A RM It is possible to calculate the approximate height of the ichnite-producing bird to its hip joint (acetabulum) using trigo- LP nometry (Fig. 13 inset) if its step angle is known (Thulborn, LP 1989; Casamiquela and others, 1987). The step angle is formed between the lines drawn from the acetabulum (origin) along the LM legs to the feet (in a bipedal animal) when both feet are simulta- LM neously placed on the ground while the animal is moving. Esti- 10 cm mates of this measurement must be made on living animals that exhibit similar traits of footprint morphology, gait (walking), and ecology as the ichnite-maker. Using suitable photographs Figure 12. Diagram comparing the gaits of domestic cats with fossil in Thompson and Ely (1989) and Udvardy (1977), I estimated trackway 2-1 (center). Each cat was walking moderately fast, which is the step angle to be from 27° to 35° for wading birds such as why the manus and pes are farther apart than is observed in the, possi- bly, slower walking ichnite-producer.RM,rightmanus;RP,rightpes; herons, gulls, and geese. From the formula given in Figure 13 LM, left manus; LP, left pes. and the average pace length of 9.7 cm for the ichnite-maker, the height could have been from 15.4 to 20.2 cm.

Washington Geology, vol. 26, no. 2/3, September 1998 55 left foot

right foot 5cm

40o

acetabulum “hip joint” f y x = tan f

step angle

= height

x y right foot left foot impression impression 35o pace distance (2y )

Figure 13. Sketch and photos of the bird trackway. The solid line is the outline of the slab that bears the subtraces (casts), the dashed out- line is of the slab bearing the molds. Dashed portions of the bird foot- prints are reconstructions of their borders drawn from the complete footprints. Arrow points in the direction of movement. The smaller foot- prints were made by the rodent-like track makers. The inset shows the method used to calculate the height of the track maker to the acetabu- lum (hip joint) when the step angle and pace length are known. In the Probable Lizard or Salamander Impressions photo, the bars are 10 cm long. Arrows with degree measurements indi- The subtrace of a trackway of a small quadruped is diagramed cate the declination of light from a 100-watt bulb relative to the surface in Figure 14. Individual footprints appear to be represented by of each slab. (Formula inset modified from Thulborn, 1989.) groups of one to three indentations that were probably made by claws. Because this is a subtrace, it is impossible to be certain of the phalangeal formula for the trackway maker; even the dif- length (distance between the attachments of the limbs to the ferentiation between manus and pes is questionable. It is possi- trunk) was approximately 4.2 cm. More finds of these trace fos- ble, however, to get some idea as to the gait, the efficiency of sils will help refine these measurements and speculations. locomotion, size, and other criteria from the trackway meas- urements. CONCLUSIONS AND SPECULATIONS Six measurements of stride length ranged from 4.0 to 4.4 cm, with an average of 4.2 cm. Two pace measurements are From available evidence it appears that four types of verte- 1.2 and 1.3 cm, respectively. The three reliable pace angulation brates left impressions on a wet, muddy, surface in early Mio- measurements are 89°. A single interpes distance is 0.7 cm. cene time. The traces of what appear to be a wading bird, The width of the pace angulation pattern is 1.7 cm. rodent-like animals, spatially associated large quadrupeds with According to Peabody (1959), the ratio of the stride length carnivore affinities, and a probable lizard or salamander give to the width of the pace angulation pattern is correlated with tantalizing clues to an ancient ecology. The area, possibly a the pace angulation measurement for salamanders and other small lake or playa, was probably being visited by animals in vertebrates. In the fossil trackway this ratio is approximately search of food or drink. There is evidence of vegetation from 2.5/1 and matches the expected pace angulation (about 90°) for the water drop impressions and twig and leaf fossils found in animals with a salamander-like style of locomotion, body plan, other strata. The existence of what are apparently small desic- and gait (slow walking). It therefore appears that the animal cation cracks in some of the ichnite-bearing rocks indicates that created the footprints was likely built like a lizard or sala- that the body of water began to evaporate almost immediately mander. A “quick and dirty” but reliable estimate of the body after the animals’ visit.

56 Washington Geology, vol. 26, no. 2/3, September 1998 40o

rodent-like impressions ?

rodent-like impressions ?

5cm

Figure 14. Diagram of the lizard- or salamander-like subtraces. The dashed circles are the subtraces of the rodent-like impressions. Arrows point in the direction of movement. No impression of a tail can be seen on this slab because the thin layer of mud in which it might have left a trace was bro- ken away and only the subtraces remain. Arrow in the photo with the degree measurement indicates the approximate declination of light from a 100- watt bulb relative to the slab surface; the bar is 5 cm long.

Only a small part of the surface of the ichnofossil-bearing good source of information on specimen photography. Last, interval, perhaps less than 3 m², has been recovered. Almost but not least, I thank my late pet goose, Brünnhilde, and her every collected fragment of this surface bears some sort of neighborhood canine and feline friends for helping me better trace fossil. Even with the available evidence it is apparent that understand vertebrate locomotion and footprint morphology. a rich and fairly diverse animal community was thriving in this area around 23 million years ago. Volcanic eruptions and sub- REFERENCES CITED sequent deposition of volcanic detritus were evidently spo- radic enough to allow the formation or recovery of moderately Brown, R. W.; Lawrence, M. J.; Pope, J., 1984, The Larousse guide to complex animal communities. More work is needed on the animal tracks, trails, and signs: Larousse and Company, Inc., sediments and taphonomy to better understand the environ- 320 p. ment represented by this thick rock unit. Casamiquela, R. M.; Deamthieu, G. R.; Haubold, H.; Leonardi, G.; No definite conclusions can be reached about the identity Sarjeant, W. A. S., 1987, Glossary and manual of tetrapod foot- of the footprint makers, but there are many clues. Even with print palaeoichnology: Brasil Departamento Nacional da Produ- further trackway discoveries, classification will be tentative cao Mineral, 75 p., 20 pl. unless associated skeletal material is recovered. Evarts, R. C.; Ashley, R. P., 1993, Geologic map of the Cowlitz Falls quadrangle, Lewis and Skamania counties, Washington: U.S. Acknowledgments Geological Survey Geologic Quadrangle Map GQ-1682, 1 sheet, scale 1:24,000, with 10-p. text. I thank my mother, Flora Z. Kaler, for her help on nearly all of Evarts, R. C.; Ashley, R. P, 1993, Geologic map of the Vanson Peak the numerous collecting trips to the site. W. A. S. Sarjeant, quadrangle, Lewis, Cowlitz, and Skamania counties, Washing- University of Saskatchewan, read an early draft of this study, ton: U.S. Geological Survey Geologic Quadrangle Map GQ-1680, supplied many excellent suggestions and comments, and intro- 1 sheet, scale 1:24,000, with 12-p. text. duced me to useful reference literature. Eric Schuster was a

Washington Geology, vol. 26, no. 2/3, September 1998 57 Evarts, R. C.; Ashley, R. P.; Smith, J. G., 1987, Geology of the Mount Murie, O. J., 1974, A field guide to animal tracks, 2d ed.: Houghton Saint Helens area—Record of discontinuous volcanic and plu- Mifflin Company, 375 p. tonic activity in the Cascade arc of southern Washington: Journal Peabody, F. E., 1959, Trackways of living and fossil salamanders: of Geophysical Research., v. 92, no. B1, p. 10,155-10,169. University of California Publications in Zoology, v. 63, no. 1, Gambaryan, P. P., 1974, How mammals run—Anatomical adapta- 72 p. tions [Beg mlekopitayushchikh—Prisposobitel’nye, trans. by Savage, D. E.; Russell, D. E., 1983, Mammalian paleofaunas of the Hilary Hardin]: John Wiley & Sons; Israel Program for Scientific world: Addison-Wesley Publishing Company, 432 p. Translations, 367 p. Wallace, G. T., 1955, An introduction to ornithology: The Macmillan Halfpenney, J.; Biesiot, E., 1986, A field guide to mammal tracking in Company, 441 p. western America: Johnson Books [Boulder, Colo.] 164 p. Thompson, M. C.; Ely, C., 1989, Birds in Kansas, Volume 1: Univer- Hall, E. R., 1981, The mammals of North America, 2 v., 2nd ed.: John sity of Kansas Museum of Natural History, 404 p. Wiley & Sons, 1181 p. Thulborn, R. A., 1989, The gaits of dinosaurs. Chapter 5 in Gillette, Hildebrand, Milton, 1974, Analysis of vertebrate structure: John D. O.; Lockley, M. G., editors, Dinosaur tracks and traces: Cam- Wiley & Sons, 710 p. bridge University Press, p. 39-50. Lander, E. B.; Swanson, D. A., 1989, Chadronian (early Oligocene) Udvardy, M. D. F., 1977, The Audubon Society field guide of North and early Arikareean (late Oligocene) land mammal assemblages American birds—Western region: Alfred A. Knopf, 854 p. from the Ohanapecosh Formation, Tieton basin area, central Cas- Wallace, G. T., 1955, An introduction to ornithology: The Macmillan cade Range, south-central Washington: Geological Society of Company, 441 p. America Abstracts with Programs, v. 21, no. 5, p. 104. Walsh, T. J.; Korosec, M. A.; Phillips, W. M.; Logan, R. L.; Schasse, Mossman, D. J.; Sarjeant, W. A. S., 1983, The footprints of extinct H. W., 1987, Geologic map of Washington—Southwest quadrant: animals: Scientific American, v. 248, p. 74-84. Washington Division of Geology and Earth Resources Geologic Map GM-34, 2 sheets, scale 1:250,000, with 28-p. text. n

Digital Geologic Maps Available Arc/Info versions are available for the following 1:100,000 geologic maps: Astoria, Centralia, Chehalis River, Chelan, Chewelah, Colville, Hood River, Ilwaco, Mount Adams, Mount Baker, Mount Rainier, Mount St. Helens, Nespelem, Omak, Port Townsend, Priest Rapids, Republic, Richland, Robinson Mountain, Sauk River, Seattle, Skykomish River, Snoqualmie Pass, Spokane, Tacoma (south half), Twisp, Vancouver, Wenatchee, and Westport as shown below.

o o o 117o 124 123 122o 121o 120o CANADA 119o 118 49o 49o USA Roche Mount Harbor Bellingham Robinson Republic Colville Baker Mountain Oroville 125o 48o 30¢ o ¢ C 48 30 AN AD US A Cape A Port Port Sauk Nespelem Chewelah Flattery Angeles Townsend River Twisp Omak

48o 48o

IDAHO Mount Coulee Forks Seattle Skykomish Banks Spokane Olympus River Chelan Lake Dam o ¢ 47 30 47o 30¢ Copalis Snoqualmie Beach Shelton Tacoma Wenatchee Moses Ritzville Rosalia Pass Lake o 47 47o

Westport Chehalis Centralia Mount Priest Connell Pullman River Rainier Yakima Rapids

o ¢ 46 30 46o 30¢

Clarkston Ilwaco Astoria Mount Mount Toppenish Walla Walla St. Helens Adams Richland (and Orofino)

o 46 46o 125o o 124o OREGON 118o 117 Hood Vancouver Goldendale EXPLANATION 020304010 mi River Hermiston

0 10 204030 50 60 km 45o 30¢ Completed 7/1/96–6/30/97 o 123 122o 121o 120o 119o We are distributing copies free on 8-mm tape. We provide the tape and ask that you send us a blank Completed 7/1/97–6/30/98 tape in return. Contact Carl Harris at (360) 902-1453 for details. We’d appreciate your giving us credit as the source of the data. This work was supported by the U.S. In progress 7/1/98–9/30/98 Geological Survey STATEMAP program, agreements 1434-HQ-96-AG-01523, 1434-HQ-97-AG- 01809, and 1434-HQ-98-AG-2062. Under contract for 10/1/98–6/30/99

58 Washington Geology, vol. 26, no. 2/3, September 1998 Radiocarbon Ages of Probable Coseismic Features from the Olympic Peninsula and Lake Sammamish, Washington

Robert L. Logan1, Robert L. Schuster2,PatrickT.Pringle1, Timothy J. Walsh1, and Stephen P. Palmer1 1Washington Department of Natural Resources 2 U.S. Geological Survey Division of Geology and Earth Resources Box 25046, MS 966 PO Box 47007, Olympia, WA 98504-7007 Denver, CO 80225

uring the past decade, we have searched for clues to the Sammamish shoreline near the Seattle fault, the Lake Crescent Dseismic history of western Washington. Our research has landslide complex in the northern Olympic Peninsula, numer- concentrated on sites on the Olympic Peninsula and in Lake ous large rock avalanches in the southern peninsula that origi- Sammamish. Features we have examined include faults, sub- nated on rugged mountainsides composed of Crescent Forma- marine landslides, and landslide-dammed lakes. This article tion basalt, and two young, nearby faults are natural features describes these investigations and discusses the implications that can provide new insight and raise new questions about past of our radiocarbon ages. seismic events in western Washington (Fig. 1). Despite the considerable geographic separation of the In the interest of brevity and simplicity, we use the same Olympic Peninsula sites from Lake Sammamish, temporal rela- terminology as in our article in the previous issue of Washing- tions of most sites we investigated are intriguing, and several ton Geology (Pringle and others, 1998). We use Varnes’ (1978) may be closely related to other seismically induced features in classification of landslides, and we report uncorrected 14C the region (Atwater and Moore, 1992; Bucknam and others, ages. Both calibrated and uncalibrated ages are shown in Table 1992; Jacoby and others, 1992; Karlin and Abella, 1992; 1. Years before present (yr B.P.) are years before 1950. Schuster and others, 1992). A collapse of a portion of the Lake

124o 123o 122o 0 5 10 mi Harrigan Pt. Lyre River 750 ±50 Saratoga Pt. 0 5 10 15 km Thompson Pt. Lake 280 ±60Sutherland 300 ±60 Port Angeles x landslide Fairholm Mt. Indian Creek figure Lake Storm complex location 48o Crescent King (>600)

Elwha River

Everett 101 Hoh River

5 Kirkland Queets River ³1160±35 West Point Lake Lena Lake Winslow 1108 ±16 Washington 1170 ±50 >850 ±45 Hamma R. <1670 ±60 405 Lake Hamma SEATTLE Sammamish Jefferson Lake Hamma FAULT D Quinault River 1080 ±50 Hamma R. Bremerton Restoration Point W. Mercer U ? 47o 1080 ±60 >2900 ±60 Island Greenwood Point 30¢ >620 ±90 Seattle Vasa Lake Lake Cushman Mt. 1155 ±40 Park 1050 ±60 Campbell Tree Grove Price Lake <1560 ±90 Puget Sound Quinault Campground 400 ±50 xRose 1080 ±50 S. Mercer Pine Lake Lynch Cove Island 90 >710 ±60 Lake * >450 >540 >1170 ±90 1121 ±15 Cushman Canal SMEF £1420 ±70 Burley Wynoochee Spider Lake >1130 ±50 Lake 1060 ±60 Hood 1090 ±50 CRF±70 <1880 1160 ±50 “Upper” Dry EXPLANATION OFSITE SYMBOLS Bed Lake River Tree-ring counts that establish minimum ages for landslides “Lower” Dry >200 Green Oakland Radiocarbon dates from this article Bed Lake Bay KING West Fork Humptulips River 1010 ±50 >1190 ±60 SubmarinePIERCE landslides in Lake WashingtonRiver 1030 ±50 Tidal marsh with no abrupt sea level change Canyon River Shelton Areas of abrupt uplift Subsided tidal marsh

Wynoochee River

West Fork Satsop Tree ring and radiocarbon dates

Figure 1. Locations of radiocarbon-dated material and other selected neotectonic features in western Washington. CRF, Canyon River fault; SMEF, Saddle Mountain East fault. All dates in years B.P.

Washington Geology, vol. 26, no. 2/3, September 1998 59 Table 1. Radiocarbon age data for selected landslide deposits in western Washington. * denotes influence of bomb 14C. Radiocarbon ages given in yr B.P.

Beta 13C Calibrated and Lab 14C 13C/ adjusted corrected age Sample location no. age1 12C age (yr B.C.orA.D.)2 Legal location Comments

2 B.C. Detrital charcoal from silty sand layer that was either 1880 1880 sec. 19, Canyon River fault 109347 -25.0 (A.D. 130) offset by fault or deposited shortly after last movement on ±70 ±70 T22N R6W A.D. 329 the fault.

A.D. 1217 Wood from vertical snag at surface of lake. Unknowns Lake Crescent 750 sec. 14, 96390 -25.1 750 ±50 (A.D. 1282) include amount of outer surface erosion of the snag and if (Harrigan Point) ±50 T30N R9W A.D. 1373 snag was rooted in lake bottom.

A.D. 1516 Wood from branch of tilted (30o from vertical), rooted Lake Crescent 300 sec. 30, 75774 -29.8 230 ±60 (A.D. 1663) deciduous tree probably carried into the lake by a (Thompson Point) ±60 T30N R9W A.D. 1954 landslide; sampled at depth of 66 ft.

A.D. 1472 Outer rings from rooted, tilted snag probably carried into Lake Crescent 280 sec. 18, 39669 (A.D. 1647) the lake by a landslide. Amount of outer surface erosion (Saratoga Point) ±60 T30N R8W A.D. 1954 unknown.

922 B.C. Detrital branch-size wood that protruded into the water Hamma Hamma 2640 2680 sec. 2, 50545 -22.4 (818 B.C.) from the submerged portion of the lacustrine silts in the River ±60 ±60 T24N R4W 784 B.C. streambank at 4.4 ft below the river surface.

1035 B.C. Detrital branch-size wood that protruded into the water Hamma Hamma 2750 2770 sec. 2, 50546 -24.0 (906 B.C.) from the submerged portion of the lacustrine silts in the River ±60 ±60 T24N R4W 807 B.C. streambank at 5.0 ft below the river surface.

843 B.C. Detrital branch-size wood that protruded into the water Hamma Hamma 2660 2590 sec. 2, 50547 -29.0 (795 B.C.) from the submerged portion of the lacustrine silts in the River ±70 ±70 T24N R4W 518 B.C. streambank at 5.5 ft below the river surface.

1262 B.C. Detrital branch-size wood that protruded into the water Hamma Hamma 2950 2900 sec. 2, 50598 -28.1 (1045 B.C.) from the submerged portion of the lacustrine silts in the River ±60 ±60 T24N R4W 910 B.C. streambank at 6.8 ft below the river surface.

Hamma Hamma 2960 1401 B.C. (1158, 1145, sec. 2, Detrital charcoal from lake deposits slightly above 39798 River (charcoal) ±80 1134 B.C.) 921 B.C. T24N R4W boulders of landslide dam.

1340 A.D. 634 (A.D. 671) sec. 35, Wood from inner part of drowned snag taken at high-water Lena Lake 32671 ±50 A.D. 783 T25N R4W stand. Amount of snag surface erosion unknown.

1300 A.D. 654 (A.D. 690) sec. 35, Wood from inner part of drowned snag taken at high-water Lena Lake 32672 ±50 A.D. 872 T25N R4W stand. Amount of snag surface erosion unknown.

1470 1440 A.D. 538 (A.D. 635) sec. 35, Wood from outer 15 rings of drowned snag sampled below Lena Lake 58575 -26.8 ±60 ±60 A.D. 681 T25N R4W the surface of lake-bottom silt.

1170 1150 A.D. 779 (A.D. 891) sec. 35, Wood from outer rings that had grown over charred older Lena Lake 58576 -25.8 ±50 ±50 A.D. 1005 T25N R4W wood of drowned snag.

1150 A.D. 685 (A.D. 821, NE1/4 sec. 21, Wood from inner part of drowned snag taken at high-water Jefferson Lake 42123 ±50 840, 860) A.D. 967 T24N R4W stand. Amount of snag surface erosion unknown.

1210 A.D. 1678 (A.D. NE1/4 sec. 21, Wood from inner part of drowned snag taken at high-water Jefferson Lake 42124 ±50 1955*) A.D. 1955* T24N R4W stand. Amount of snag surface erosion unknown.

1080 1050 A.D. 1710 (A.D. NE1/4 sec. 21, Jefferson Lake 58573 -26.8 Outer rings adjacent to bark of drowned snag. ±60 ±60 1955*) A.D. 1955* T24N R4W

1080 1070 A.D. 890 (A.D. 1005) NE1/4 sec. 21, Jefferson Lake 58574 -25.8 Outer rings adjacent to bark of drowned snag. ±50 ±50 A.D. 1150 T24N R4W Wood adjacent to bark of snag that was drowned when the 1080 1050 A.D. 665 (A.D. 776) secs. 22 & 23, Price Lake 74570 -27.0 Saddle Mountain East fault scarp dammed Lilliwaup ±50 ±50 A.D. 888 T23N R4W Creek.

1290 1260 A.D. 689 (A.D. 883) sec. 10, Wood from inner rings of eroded snag collected above the Spider Lake 50550 ±50 ±50 A.D. 997 T22N R6W surface of the lake.

1190 1180 A.D. 779 (A.D. 891) sec. 10, Wood from inner rings of eroded snag collected above the Spider Lake 50602 ±60 ±60 A.D. 1005 T22N R6W surface of the lake.

1060 1050 A.D. 871 (A.D. 973) sec. 10, Wood adjacent to the bark of drowned snag collected Spider Lake 58580 -25.5 ±60 ±60 A.D. 1023 T22N R6W below surface of lake-bottom silt.

1160 1150 A.D. 719 (A.D. 883) sec. 10, Wood adjacent to the bark of drowned snag collected Spider Lake 58581 -25.6 ±50 ±50 A.D. 984 T22N R6W below surface of lake-bottom silt.

1090 1100 A.D. 779 (A.D. 891) sec. 10, Rings 73–87 in from bark, centered on ring 80. From root Spider Lake 58582 -24.4 ±50 ±50 A.D. 1005 T22N R6W of Douglas-fir snag.

60 Washington Geology, vol. 26, no. 2/3, September 1998 Beta 13C Calibrated and Lab 14C 13C/ adjusted corrected age Sample location no. age1 12C age (yr B.C.orA.D.)2 Legal location Comments

Lower 1160 1180 A.D. 951 (A.D. 1053) S1/2 sec. 12, Wood from inner part of eroded snag, broken from top of 50544 -23.9 Dry Bed Lake ±50 ±50 A.D. 1103 [+80] T21N R6W snag during high-water stand.

Lower 930 A.D. 719 (A.D. 883) S1/2 sec. 12, Rings 68–92 in from bark centered on ring 80. From stem 58570 -26.2 920 ±50 Dry Bed Lake ±50 A.D. 984 T21N R6W of Douglas-fir snag.

A.D. 1097 (A.D. 1144, Lower 1010 1020 S1/2 sec. 12, Rings 68–92 in from bark centered on ring 80. From stem 58571 -24.3 1155, 1207, 1213, Dry Bed Lake ±50 ±50 T21N R6W of Douglas-fir snag (same tree as above). 1239) A.D. 1309 [+80]

Lower 1030 1010 A.D. 978 (A.D. 1097) S1/2 sec. 12, Rings 72–86 in from bark, centered on ring 80. From root 58572 -26.1 Dry Bed Lake ±50 ±50 A.D. 1239 [+80] T21N R6W of Douglas-fir snag.

Small 3/8 in. diameter twig-size pieces of wood recovered Campbell Tree 710 A.D. 1047 (A.D. 1100) sec. 15, 50538 -25.3 710 ±60 from drill hole at 16 ft below the ground surface, Grove Campground ±60 A.D. 1242 [+80] T23N R8W significance of date questionable.

360 A.D. 1225 (A.D. 1290) sec. 10, Wood adjacent to the bark of a vertical snag buried by Lake Cushman 50539 -22.5 400 ±50 ±50 A.D. 1398 T23N R5W landslide debris.

390 A.D. 1430 (A.D. 1473) sec. 10, Wood from adjacent to the bark of a vertical snag buried Lake Cushman 50540 -23.1 420 ±50 ±50 A.D. 1641 T23N R5W by landslide debris.

Lake Sammamish 1450 A.D. 547 (A.D. 628) sec. 18, Sample of inner wood from the top of a snag protruding 80713 (Greenwood Point) ±40 A.D. 663 T24N R6E from the lake surface.

Lake Sammamish 1330 A.D. 640 (A.D. 676) sec. 18, Sample of inner wood from the top of a snag protruding 80917 (Greenwood Point) ±50 A.D. 786 T24N R6E from the lake surface. Sample of upper part of root, 15 rings in from bark of a Lake Sammamish 1050 1000 A.D. 905 (A.D. 1029) sec. 18, 117413 -28.4 drowned tree that was rooted in a landslide deposit on the (Greenwood Point) ±60 ±60 A.D. 1183 [+7] T24N R6E bottom of the lake.

1 13 Radiocarbon ages are normalized to d C = –25.0o/oo unless no value is shown in column 4. Error terms are 1s with a lab error multiplier of 2 unless noted. 2 Radiocarbon ages are calibrated in sidereal (calendar) years using tree ring data in the CALIB 3.0.3 program of Stuiver and Reimer (1993) and reported as [–1s, (calculated intercept age), +1s] using a lab error multiplier of 2. Commonly, fluctuations in the calibration curve may yield more than one intercept or calibrated age. B.P. (before present) ages are reported with respect to A.D. 1950. Where shown in brackets, a correction factor was applied to adjust lab data for approximate number of growth rings of the sample from the bark. See Colman and others (1987), for a discussion of 14C terminologies.

Lake Crescent abundance of drowned trees to a depth of about 80 ft, unless the Lakes Crescent and Sutherland (Fig. 1) are located in a valley landslides occurred soon after the glaciers melted, before for- that was gouged out by the Juan de Fuca lobe of the continental ests were established. ice sheet during the late Pleistocene. Steeply dipping, gener- Using a glass-bottomed viewer to see to depths of nearly ally east-west striking, Tertiary sedimentary and volcanic 100 ft, we searched the entire shoreline of Lake Crescent in an rocks bound these lakes. The two lakes were formerly one large effort to find the drowned forest of submerged trees that theo- retically should be rooted in the lake bottom. In several trips to lake that drained eastward through Indian Creek and eventu- 14 ally the Elwha River, as Lake Sutherland does today. The lakes the area, we have collected only four samples for C dating. are now separated by a landslide complex that was emplaced Two of the samples are discussed in Logan and Schuster during several catastrophic episodes of landslide activity from (1991). The other two samples were taken from a snag off Har- both sides of the valley (Reagan, 1909; Brown and others, rigan Point and from a submerged tree rooted at a depth of 1960; Weaver, 1937; Fiksdal and Brunengo, 1980; Tabor, about 66 ft slightly east of Thompson Point near the western 1987; Logan and Schuster, 1991). The resulting rock-debris end of the lake (Fig. 1). The snag at Harrigan Point was sam- dam raised the surface of Lake Crescent about 80 ft, so that it pled at the surface of the lake from a boat and had almost cer- now empties into the Lyre River, which drains northward into tainly been eroded by wave action and decay. Although we do the Strait of Juan de Fuca. not know how much erosion has taken place, previous experi- The ages of emplacement of the landslide complex are dif- ence (Logan and Walsh, 1995; Pringle and others, 1998) has ficult to determine with certainty. The landslide deposit of the shown that loss of 200 to 300 years (or several inches) of outer growth rings because of rot, exfoliation, and (or) other erosive Mount Storm King episode from the south valley wall, which is 14 mentioned in local tribal legend (Reagan, 1909), is clearly processes is not uncommon. This snag yielded a corrected C older than debris originating from the north. It is not only geo- date of 750 ±50 yr B.P. (Table 1). Allowing for erosion, the tree morphically more subdued, but its toe underlies the landslide would be about 600 years old (or younger). This age is similar debris from the north valley wall. A count of the annual growth to the minimum age of the landslide complex as indicated by rings of a stump rooted on the youngest part of the slide com- the age of the rooted stump mentioned above. The snag is a plex indicates that the deposit is at least 600 years old. When conifer and appeared to be vertical, as if in growth position. the surface of Lake Crescent was raised to its present level, for- This would suggest that it was drowned by an abrupt rise in the ested shoreline slopes should have been inundated, leaving an level of Lake Crescent. However, because we sampled it from the surface without help from a diver, we do not know if the

Washington Geology, vol. 26, no. 2/3, September 1998 61 Figure 2. Submerged snag apparently in growth position about 80 ft below the lake surface near the west end of Lake Crescent. The photographer, Rick Manning, reported a beach-like sandy area, not visible in the pho- tograph, slightly deeper than the tree. tree is rooted. We are left with the ques- tion—Why aren’t there more sub- merged trees in Lake Crescent? A photograph (Fig. 2) taken by a diveratadepthof60to80ftinLake Crescent about 4.5 mi east of Fairholm (Fig. 1) depicts a tree in apparent growth position. We hired another diver to sample the tree, but he was un- able to find it or any other drowned trees along a 300-ft section of the shoreline in that area. However, he did report seeing an apparent ancient shoreline or beach at about 80 ft below the lake surface. Farther west, at the site near Thompson Point (Fig. 1), a diver from the National Park Serv- ice (NPS) recovered a sample from a drowned deciduous snag. The snag was rooted but was tilted about 30o from vertical. We could not determine from its orientation if the tree was in its original growth position. Furthermore, the morphology of the hillside above the adjacent shoreline strongly suggests that the submerged tree rode a landslide to its present location. The 14 sample yielded a C age of 300 ±60 yr B.P., supporting the like- lihood of landslide emplacement after Lake Crescent rose to its current level at least 600 years ago. The age is close to the 280 ±60 yr B.P. age (from an eroded snag; Logan and Schuster, 1991; also Table 1) from Saratoga Point. Could it be that both landslides were caused by an earthquake at about that time? Despite being stymied by the lack of drowned trees and other evidence, such as organic deposits in depressions on the surface of the landslide complex that separates the two lakes, we are continuing our efforts to determine the age of the land- slide that raised Lake Crescent to its present level. We believe that such a determination is important to understanding the seismic history of western Washington. Our next step may be to core sediment in the bottom of Lake Crescent for stra- tigraphic or radiocarbon evidence.

Hamma Hamma River The Hamma Hamma River flows from the east flank of the Olympic Range into Hood Canal in the Puget Lowland (Fig. 1). At its confluence with Cabin Creek, the river was dammed by a large rock avalanche that originated in the basalt cliffs above the south bank. Fine lake sediment containing organic material filled the valley bottom upstream of the landslide dam, forming the large flat area on which the Lena Creek Campground was built. We have obtained five radiocarbon dates from wood and detrital charcoal in these sediments. The samples yielded ra- diocarbon ages of about 2,900 yr B.P. (Table 1) and younger. The wood samples were taken from the submerged portion of a cut bank at depths ranging from 4.4 to 6.8 ft below the riv- er’s surface (Fig. 3). By using a long-handled tree pruning de- Figure 3. Detrital wood exposed in the submerged part of a stream vice, we were able to reach down into the river to saw off and bank in the Hamma Hamma River. Radiocarbon dating of the wood from recover wood from small-diameter logs and branches that pro- this and a nearby site established a minimum age of about 2,900 yr B.P. truded from the submerged bank into the water. The oldest for the sediment deposited upstream of the landslide that dammed the wood was taken 6.8 ft below the surface of the river at the base river.

62 Washington Geology, vol. 26, no. 2/3, September 1998 of the exposure (Fig. 3). With one exception (see Table 1), the high. We cut the samples from the upper, eroded parts of snags. upper samples are generally younger than the lower samples. From these samples we expected and got older ages (1,150 ±50 This exception can be attributed to the chaotic nature of deposi- and1,210±50yrB.P.). Subsequent samples, taken during a tion of wood and sediment in the aggrading shallow lake bot- much lower stand of the lake, were of wood adjacent to the bark tom or to the expected standard deviations in radiocarbon ages. on preserved stump roots and yielded ages of 1,080 ±50 to 60 Because the total depth of sediment impounded by the Hamma yr B.P. These probably closely represent the age of the Jeffer- Hamma River rock avalanche is unknown, the 2,900 yr B.P. son Lake landslide dam. The upper part of Jefferson Lake is date represents only a minimum age for the slide. A similar es- separated from Jefferson Lake by an alluvial fan and small timate of the minimum age of the landslide is provided by an- landslide and does not contain snags. We suspect that if any other radiocarbon date of 2,960 ±80 yr B.P. from detrital char- snags existed, they have been buried by sediments. coal in silt slightly above basalt boulders that make up the land- slide deposit. Pine Lake The dam that impounded the sediments in the Hamma Pine Lake (Figs. 1, 4) is located in the headwaters of Pine Hamma River valley must have been breached between about Creek, a tributary of the Skokomish River. It is about 4 mi east 2,500 and 600 years ago. About 4 ft of sediment was deposited of the Wynoochee Reservoir, 2.5 mi northwest of Spider Lake during the 400 or so years for which we have some age control. (Schuster and others, 1992), and 3 mi north of the Canyon The top of the impounded sediment is at least 10 ft above the River fault (Walsh and others, 1997). Pine Creek is dammed by surface of the river at the sample location, so considerably large, locally derived, angular basalt boulders similar to those more sedimentation took place after the youngest (2,500 yr that dammed the Hamma Hamma River and other landslide- B.P.) wood was deposited. If we assume that the sedimentation dammed lakes in the southeastern Olympics. The boulders rate behind the dam was on the average about 1 ft per 100 years, originated from a bowl-shaped depression in the ridge that it is possible that the dam could have been breached at about bounds the north side of the lake. The antiquity of the landslide 1,000 to 1,500 yr B.P. is confirmed only by the conifer trees growing on both the land- Dendrochronologic observations suggest that the youngest slide deposit and the landslide scarp. Cores from one of these sediment behind the landslide dam is at least 600 years old. We trees yielded a minimum of 541 annual growth rings. The age estimated this age from annual growth rings on the stump of a of the stand of old-growth trees in the general vicinity of the cedar that had grown on the impounded sediments. (We lake is 690 years, on the basis of the last known local wildfire counted 429 rings on the outer part of the hollow tree and esti- (Richard Carlson, U.S. Forest Service [USFS], oral commun., mated the number of missing rings to be 100 to 200.) Could the 1998). The lack of an ancient drowned forest in the lake could landslide dam have been breached during an extremely large be due to the lake’s small size and rapid sedimentation rate. A earthquake at about 1,100 years ago, about the same time as the delta of fine sediment occupies a portion of the upstream part other landslide dammed lakes formed nearby? of the lake. Also, large pulses of sediment could have been Lena Lake Located only 2 mi northwest of the Hamma Hamma River rock avalanche (Fig. 1), Lena Lake is also impounded by a rock ava- lanche, this one derived from a basalt ridge slightly east of the lake. Interpretation of the ages of wood retrieved from snags in Lena Lake is complicated by the fact that the surfaces of most of the snags consist of charred wood, indicating that a fire swept the area at one time. The fire may have killed some trees and charred others without killing them before the lake formed. scarp New growth over fire-blackened wood on sample Beta 58576 boulder-rich supports this interpretation. Sample Beta 58575, for example, landslide was composed of the first 15 rings adjacent to bark and was deposit sampled below the lake bottom surface. Its age of 1,470 ±60 yr ? ? B.P. (probable date of the fire) is anomalously older than Beta Pine 58576 (1,170 ±50 yr B.P.), which is also outer wood taken Lake slightly below the surface of the lake-bottom mud from outer wood that had grown over a fire scar. Beta 58575 is also older than two other samples of wood closer to the center of the snag; these were taken at higher water stand, above the lake bottom probably from an eroded surface, and yielded uncorrected ages of 1,340 and 1,300 ±50 yr B.P. We infer that the 1,170 ±50 age is closest to the actual age of the landslide that impounded the lake.

Jefferson Lake Situated about 3.5 mi southwest of the Hamma Hamma River Figure 4. Numerous open-canopy channels leading into Pine Lake site, Jefferson Lake is studded with drowned conifer snags in are likely sources of sediment that could have buried drowned tree vertical growth position. Two sets of two radiocarbon ages snags. Alternatively, the lake may have formed on the landslide, and from Jefferson Lake are listed in Table 1. The first two ages are thus would not contain drowned snags. Tree ring counts on cores from a not corrected for 13C, and the samples were taken during recon- tree growing on the slide set the minimum age of the slide at about 540 naissance stages when the water level of the lake was relatively years.

Washington Geology, vol. 26, no. 2/3, September 1998 63 delivered to the lake through several steep ravines that feed into the lake. In aerial photographs, these ravines ap- pear as openings in the tree canopy and are likely channels for debris torrents. It is also possible that the landslide that impounds Pine Lake could have com- pletely buried the forest that presum- ably covered the ancient valley floor.

Spider Lake A drainage divide between the Sko- komish and Satsop River basins was moved westward about one mile when the large rock avalanche that currently impounds Spider Lake thundered into the headwaters of the middle fork of the Satsop River. Spider Lake now drains eastward into Cedar Creek, a tributary of the Skokomish River. The lake con- tains an abundance of snags in growth position that are easily reached when the water level is low (Fig. 5). Previ- Figure 5. Snags in Spider Lake. Radiocarbon dating has established the time of the landslide- ously reported radiocarbon ages of damming event. 1,290 ±50 and 1,190 ±60 yr B.P. (Schuster and others, 1992) were from samples obtained dur- ing a high-water stand and from inner rings of snags. Again, as expected, our later samples, taken from outer wood adjacent to bark during a lower water stand, resulted in three younger ages: 1,060 ±60; 1,090 ±50; and 1,160 ±50 yr B.P., similar to ages from other nearby faults and landslide-dammed lakes.

Dry Bed Lakes “Lower” (LDBL) (Fig. 6) and “upper” Dry Bed Lakes (UDBL) (Fig. 7), located about 6.5 mi south-southeast of Spider Lake (Fig. 1), are separated by a large landslide deposit that is proba- bly much younger than the landslide that impounds LDBL. The age of LDBL is well established at about 1,100 yr B.P.assug- gested by radiocarbon ages for outer wood beneath the bark of drowned trees in LDBL (Table 1) and those ages previously re- ported by Schuster and others (1992). There are no drowned trees in UDBL; however, cores from a living hemlock and a Douglas fir on the upper landslide divide each contained about 200 rings and provide a minimum age for the slide. Average re- currence for wildfires in the southern Olympic Peninsula is about 200 years (Richard Carlson, USFS, oral commun., 1998), so the slide could be much older than 200 years, or pos- sibly the same age as the Lake Cushman (Table 1) and Lake Crescent landslides.

Campbell Tree Grove Campground Located near the headwaters of the West Fork Humptulips River, the Campbell Tree Grove Campground is situated in a valley carved by an alpine glacier. Sometime after the glacier Figure 6. “Lower” Dry Bed Lake and partially submerged tree snags. retreated, large pieces of rock debris slid or fell from the ridge The water level in these lakes drops much lower during extended dry periods, which allows sampling of outer tree rings preserved by sedi- on the north side of the valley. In an unpublished 1978 hydro- ment surrounding the bases of the trees. geologic project report for a water well for the campground, USFS personnel identified the rock debris as a landslide de- posit. bris. Radiocarbon analysis of the sample resulted in an uncali- 14 During the drilling of the well through the rock debris into brated C age of 710 ±60 yr B.P.(Table1). underlying alluvium, the drill penetrated wood at 16 ft below Estimating the age of the slide from this sample, however, the surface. USFS geologists suggested that the wood was is problematic. First, we do not know what part of the tree the probably part of a tree that was killed and buried by the rock de- wood is from, so a radiocarbon date would be a poor estimate

64 Washington Geology, vol. 26, no. 2/3, September 1998 of the age of burial. The sample appears to be the remains of small limbs about 3/ 8 in. in diameter, which suggests that the wood could be from young parts of a tree, and therefore give a more accu- rate estimate of age of death and burial. Second, after we had received the radiocarbon date, we determined that the rock debris in the campground area probably has accumulated periodically by relatively small episodes of rockfall, not as one large rockslide. Unlike the other rockslide sites, this one lacks a large bowl-shaped source area on the upland slopes and is, instead, situated below a narrow, steep-sided ravine that appears more likely to occasionally calve blocks of rock. The oldest trees on the surface of the landslide deposit (Fig. 8) are as much as 7.5 ft in diameter and are about 450 years old (deter- mined by counting rings on recently cut Douglas fir stumps). Several other populations of younger trees also grow on the rock debris. Many of the larger trees have large boulders leaning Figure 7. “Upper” Dry Bed Lake has no standing tree snags because the snags have been buried against them and could have survived by the landslide that separates the two lakes. Note the boulders protruding from the bank on the occasional rockfalls but not a major left side of the photograph. The shoreline with the boulders is part of the landslide that separates rockslide. Openings in the canopy cre- the two lakes. The shoreline without boulders, on the right, is the valley wall. ated by large rocks crashing through the forest would allow growth of the smaller trees. Windstorms or wildfires could also cause canopy openings, but evidence for such events tends to be short lived.

Lake Cushman A large rockslide from Mount Rose, above the north shore of Lake Cush- man, buried a rooted tree from which we were able to sample wood (Fig. 9). The samples were taken from wood ad- jacent to the bark, and thus represent the youngest wood in the tree. We do not know if the tree was alive when it was buried; that is, the sample may have come from a snag that had been dead for more than 100 years. It follows that our approximately 400 yr B.P.age (Table 1) from the snag could be older than the actual age of burial by the land- slide. Therefore, the landslide may have occurred as recently as 300 yr B.P., possibly about the same time as the upper of the Dry Bed Lakes formed Figure 8. The trees in the Campbell Tree Grove Campground are of various ages. Here, the old- or landslides were activated at Lake est trees are about 450 years old. Some of the older trees have boulders leaning against them, Crescent. whereas some of the younger trees grow on boulders.

Price Lake Two active shallow crustal faults have been identified in the outer wood of a snag that was rooted in the bottom of the lake. southeast Olympic Mountains. The Saddle Mountain East fault Thetreefromwhichthewoodsamplecamemusthavediedby (SMEF) has produced a scarp that dams Lilliwaup Creek to drowning within the year (Dan Omdal, Washington Depart- form Price Lake (Fig. 1) (Carson, 1973; Wilson, 1975) located ment of Natural Resources [DNR], oral commun., 1998) fol- 2 mi east of Lake Cushman. We obtained a sample from the lowing the latest seismic episode on the SMEF. The sample

Washington Geology, vol. 26, no. 2/3, September 1998 65 (Beta 74570) yielded a radiocarbon age of 1,050 ±50 yr B.P., similar to several of the landslide ages reported in Table 1. The Canyon River fault (CRF, Fig. 1) is the other local ac- tive fault. This fault also formed a scarp that dammed local drainages. Walsh and others (1997) reported an age of 1,880 ±70 yr B.P. from detrital charcoal that probably represents a maximum age for the latest movement on the fault. On the basis of their ages and location, these young struc- tures could be the source for moderate earthquakes responsible for formation of the nearby landslide lakes. However, they are also similar in age to other seismically suspect features in west- ern Washington (Fig. 1).

Lake Sammamish Greenwood Point is located at the southern end of Lake Sam- mamish less than a mile south of the projected trace of the Seat- tle fault (Bucknam and others, 1992) and associated nearby structures (Fig. 10). Two embayments along the north shore- line of the point, clusters of tilted tree snags offshore, and depth-finder sonar profiles of the lake bottom provided evi- dence that the shoreline collapsed and slid into the lake, carry- ing an ancient forest with it. If a great earthquake occurred about 1,100 years ago, as suggested by other studies referenced above, then it is reasonable to speculate that the shoreline could have collapsed as part of a seismically induced landslide. Logan and Walsh (1995) reported radiocarbon ages of 1,450 ±40 and 1,330 ±50 yr B.P. (Beta 80713 and 80917, re- spectively, in Table 1) for wood recovered from snags that pro- trude from the surface of Lake Sammamish near Greenwood Point. The ages are probably about 200 years too old because they came from the inner (older) parts of the trees (Logan and Walsh, 1995). In a cooperative effort with Gordon Jacoby of Lamont-Doherty Earth Observatory, Columbia University, we employed divers to recover wood from trees that were rooted on the surface of the submarine landslide. Radiocarbon analy- sis revealed that one of the trees rooted in the landslide surface Figure 9. Snag exposed in an excavation of Lake Cushman rockslide drowned about 1,050 ±60 years ago (Table 1). The sample was debris from Mount Rose. Wood in the outer rings in contact with bark from a root of the tree about 15 rings from the bark, and the ra- was sampled here. diocarbon age should be very close to the time of the tree’s death.

Discussion To date, we have found evidence of fault activity or possible seismically induced landsliding in three general age clusters: about 2,900, 1,100, and 300 yr B.P. The ages apparently fit the timing of other seismic or tectonic features that have been iden- tified in the Puget Lowland (see, for example, Fig. 1) and sub- sided marshes along Pacific Ocean coastline (Atwater,1996). However, our data lack the resolution to distinguish between locally induced or regionally induced features. In one sense, the most common and most tightly spatially related radiocarbon ages cluster around 1,100 yr B.P.inthe southeast Olympic Range. Does this mean that the large rock avalanches found there were caused by shallow nearby earth- quakes that occurred on the Canyon River fault and (or) the Saddle Mountain East fault? On the other hand, 1,100 yr B.P. matches the timing of movement on the Seattle fault and Pa- cific coast tidal marsh subsidence linked to a large subduction zone earthquake. Figure 10. Fault (shown by arrow) near south shore of Lake Sam- Although the evidence permits the 300 yr B.P. features to be mamish at Vasa Park. This fault displaces upper Tertiary sedimentary correlated to the A.D. 1700 event (Atwater, 1996; Atwater and rock and is thought to be associated with the Seattle fault, which proj- Hemphill-Haley, 1997; Yamaguchi and others, 1997; Jacoby ects across the lake slightly north of the radiocarbon sample site and and others, 1997), the events are not well enough constrained may have been the source of the earthquake that triggered the shore- in time or space to demonstrate that they had a single cause. line failure at Greenwood Point.

66 Washington Geology, vol. 26, no. 2/3, September 1998 Another possibility is that they were local events closely Colman, S. M.; Pierce, K. L.; Birkeland, P. W., 1987, Suggested ter- spaced in time. minology for Quaternary dating methods: Quaternary Research, The 2,900 yr B.P. age on the Hamma Hamma River site is v. 28, p. 314-319. within the 2,800 to 3,300 yr B.P. range of Atwater’s (1996) Fiksdal, A. J.; Brunengo, M. J., 1980, Forest Slope Stability Project, tidal marsh subsidence event. Perhaps the Hamma Hamma Phase I: Washington Department of Ecology Technical Report River landslide was triggered by a subduction zone earth- 80-2a, 18 p., 7 plates. quake, too. Jacoby, G. C.; Williams, P. L.; Buckley, B. M., 1992, Tree ring corre- lation between prehistoric landslides and abrupt tectonic events in Acknowledgments Seattle, Washington: Science, v. 258, no. 5088, p. 1621-1623. We thank Ken Neal (retired), Peter Erban, and Dennis Schnei- Jacoby, G. C.; Bunker, D. E.; Benson, B. E., 1997, Tree-ring evidence der of the USFS for providing information on the Campbell for an A.D. 1700 Cascadia earthquake in Washington and northern Tree Grove Campground area; Richard Carlson (USFS) for Oregon: Geology, v. 25, no. 11, p. 999-1002. providing forest history information; Brian Atwater and Fitz- Karlin, R. E.; Abella, S. E. B., 1992, Paleoearthquakes in the Puget hugh Lee (retired) of the USGS for field advice, and Boyd Ben- Sound region recorded in sediments from Lake Washington, son for help in retrieving samples; Doug Williams, Mike U.S.A.: Science, v. 258, no. 5088, p. 1617-1620. Chevalier, Ken Dean, and Craig Mason of DNR Aquatic Re- Logan, R. L.; Schuster, R. L., 1991, Lakes divided—The origin of sources Division for retrieving wood from Lake Sammamish Lake Crescent and Lake Sutherland, Clallam County, Washing- and searching Lake Crescent for submerged trees; Rick and ton: Washington Geology, v. 19, no. 1, p. 38-42. Terry Manning for information about Lake Crescent; Gordon Logan, R. L.; Walsh, T. J., 1995, Evidence for a large prehistoric seis- Jacoby of Lamont-Doherty Earth Observatory of Columbia mically induced landslide into Lake Sammamish: Washington University for funding retrieval of wood from Lake Sam- Geology, v. 23, no. 4, p. 3-5. mamish; Mike Butler, Larry Lang, and the NPS dive team for Pringle, P. T.; Schuster, R. L.; Logan, R. L., 1998, New radiocarbon the search for and recovery of wood from Lake Crescent; ages of major landslides in the Cascade Range, Washington: Cathrine Kenner DNR Division of Geology and Earth Re- Washington Geology, v. 26, no. 1, p. 31-39; errata, Washington sources (DGER) and Leslie Pringle for their assistance in our Geology, v. 26, no. 2/3, p. 70. dendrochronology research; and Eric Schuster (DGER) for his Reagan, A. B., 1909, Some notes on the Olympic Peninsula, Washing- technical review. ton: Kansas Academy of Science Transactions, v. 22, p. 131-238. Schuster, R. L.; Logan, R. L.; Pringle, P. T., 1992, Prehistoric rock References Cited avalanches in the Olympic Mountains, Washington: Science, v. 258, no. 5088, p. 1620-1621. Atwater, B. F., 1996, Coastal evidence for great earthquakes in west- Stuiver, Minze; Reimer, P. J., 1993, Extended 14C data base and re- ern Washington. In Rogers, A. M.; Walsh, T. J.; Kockelman, W. 14 J.; Priest, G. R., editors, Assessing earthquake hazards and reduc- vised CALIB 3.0 C age calibration program: Radiocarbon, v. 35, ing risk in the Pacific Northwest: U.S. Geological Survey Profes- no. 1, p. 215-230. sional Paper 1560, p. 77-90. Tabor, R. W., 1987, Geology of Olympic National Park: Pacific Atwater, B. F.; Hemphill-Haley, Eileen, 1997, Recurrence intervals Northwest National Parks & Forests Association [Seattle, Wash.], for great earthquakes of the past 3,500 years at northeastern Wil- 144 p., 2 plates. lapa Bay, Washington: U.S. Geological Survey Professional Pa- Varnes, D. J., 1978, Slope movement types and processes. In Schus- per 1576, 108 p. ter, R. L.; Krizek, R. J., editors, Landslides—Analysis and con- Atwater, B. F.; Moore, A. L., 1992, A tsunami about 1000 years ago in trol: National Research Council Transportation Research Board Puget Sound, Washington: Science, v. 258, no. 5088, p. 1614- Special Report 176, p. 11-33. 1617. Walsh, T. J.; Logan, R. L.; Neal, K. G., 1997, The Canyon River fault, Brown, R. D., Jr.; Gower, H. D.; Snavely, P. D., Jr., 1960, Geology of an active fault in the southern Olympic Range, Washington: the Port Angeles–Lake Crescent area, Clallam County, Washing- Washington Geology, v. 25, no. 4, p. 21-24. ton: U.S. Geological Survey Oil and Gas Investigations Map OM- Weaver, C. E., 1937, Tertiary stratigraphy of western Washington and 203, 1 sheet, scale 1:62,500. northwestern Oregon: University of Washington Publications in Bucknam, R. C.; Hemphill-Haley, Eileen; Leopold, E. B., 1992, Geology, v. 4, 266 p. Abrupt uplift within the past 1700 years at southern Puget Sound, Wilson, J. R., 1975, Geology of the Price Lake area, Mason County, Washington: Science, v. 258, no. 5088, p. 1611-1614. Washington: North Carolina State University Master of Science Carson, R. J., 1973, First known active fault in Washington: Washing- thesis, 79 p., 2 plates. ton Geologic Newsletter, v. 1, no. 3, p. 1-2. Yamaguchi, D. K.; Atwater, B. F.; Bunker, D. E.; Benson, B. E.; Reid, M. S.; 1997, Tree-ring dating the 1700 Cascadia earthquake: Na- ture, v. 389, p. 922. n “Mt. Rainier’s Deadly Potential” In 1997, Oregon Public Broadcasting released this half- Northwest Paleontological hour video on the geologic hazards associated with Mount Rainier, primarily lahars, mudflows, and debris flows, but Association Fall Events not excluding an eruption. You can order this film ($25) November 7: Jody Bourgeois will speak on paleotsunami from Oregon Public Broadcasting, 7140 SW Macadam, deposits, Burke Museum, 1:00 pm. Portland, OR 97219, or by calling (503) 293-1982. Please specify Oregon Field Guide episode #909 and the film title. December 5: Invertebrate and plant fossil identification (The film contains footage of Division geologist Pat Pringle workshop at the Burke Museum. at work.) Oregon Public Broadcasting has a web page: For details contact the NPA president, Mike Sternberg (360) http://education.opb.org/learning/ofg/rainier/order.html. 293-2405

Washington Geology, vol. 26, no. 2/3, September 1998 67 Notes on the new Washington State fossil, Mammuthus columbi

Bax R. Barton Box 278 Seahurst, WA 98062-0278

he appearance of a note regarding the new Washington mammoth finds were made on the Copelin Ranch along Hang- TState fossil species in Washington Geology (v. 26, no. 1, man Creek, Spokane County, in 1876. These finds, later identi- 1998) leads me to offer some additional observations concern- fied as from M. columbi, were reportedly shipped to New York ing this species and its discovery in Washington. and Chicago, where they became part of the Cope Collection of Unlike mastodons, which were not elephants, mammoths the American Museum of Natural History and the Chicago (genus Mammuthus) were large, specialized elephants com- Academy of Science Collections in the Field Museum of Natu- mon to the Pleistocene epoch. This genus first evolved in the ral History (Hay, 1927). early Pliocene (4.0 to 5.0 Ma) of Africa, and by the early Pleis- What little we know of the diet of mammoths in general and tocene (ca. 1.7 Ma), mammoths had spread throughout Asia Columbian mammoths in particular is derived from three and into North America (Shoshani and Tassy, 1996; Webb and sources. As mammoths were elephants, it is assumed that the others, 1989). dietary preferences of the two remaining genera of elephants, The Columbian mammoth was first recognized as a distinct Elephas and Loxodonta, give some indication of the diet of species by H. Falconer in 1857 from subfossil specimens re- mammoths (Eltringham, 1982). Secondly, our Russian col- covered near Darien, Georgia, and housed in the collections of leagues have analyzed the pollen and spore contents of the gas- the British Museum of Natural History. Falconer argued that trointestinal tracts from several of the frozen M. primigenius these new finds from North America were morphologically carcasses from Siberia (Ukraintseva, 1993). Finally, pollen distinct from the two previously known species of mammoth and initial (presence/absence only) macrobotanic analysis has (M. meridionalis, the “southern mammoth”, and M. primigen- been carried out on desiccated dung boluses recovered from ius, the “woolly mammoth”) and constituted a new species, some of the dry caves of the Colorado Plateau—boluses most which he named Elephas [M.] columbi. Since then, there have likely from Columbian mammoths (Hansen, 1980). All of these been a great many more finds of Falconer’s ‘new’ mammoth, analyses suggest that mammoths were obligate herbivores with including many from Washington State. We know much more a dietary preference for graminoids (grasses and sedges), Ar- about this species than Falconer could learn from the few temisia and chenopods (herbs), and meadow-bog mosses, specimens available to him at the time (Falconer, 1857). ferns, and aquatic plants. In some instances, these data suggest Because of their noteworthy size and taphonomic rigor, a diet of as much as 95 percent grasses and sedges. mammoth remains (mostly molars) are probably the most com- Evidence for the presence of pine (either cones or needles) monly reported vertebrate subfossils in this state. Of the two in their diet is minimal, and it is unlikely that pine ever consti- species of mammoth found in Washington—M. imperator and tuted more than 1 percent of a Columbia mammoth’s chosen M. columbi—Columbian mammoths are by far the most com- diet. At such low levels of consumption, it is more likely that mon. Many finds from this state have previously been referred any pine in their diet was accidentally ingested while they for- to as woolly mammoths, but thus far none has been substanti- aged for grasses and sedges across meadows and around the ated as M. primigenius when analyzed by modern methods. In edges of bogs and ponds. the Puget Lowland, of 31 previously reported finds that could be analyzed to the species level, 27 (or 87 percent) proved to be References Cited from Columbian mammoths (Barton, 1992). Barton, B. R., 1992, Late-glacial mammoths of the Georgia/Puget In western Washington, mammoth finds are heavily con- Lowlands [abstract]: Chronology and paleoenvironments of the centrated in the central and northern Puget Lowland, though a western and southern margins of the Cordilleran ice sheet during few are reported from the outer Pacific coastal zone and from the last glaciation (25,000–10,000 years ago): Quaternary Re- the lower Columbia River subprovince. The earliest mammoth search Center/University of Washington [Seattle],1 p. finds recovered from western Washington were discovered at Eltringham, S. K., 1982, Elephants: Blandford, Poole, 262 p. Scatchet Head on Whidbey Island around 1860 (Lawson, Falconer, H., 1857, On the species of mastodon and elephant occur- 1874), but these are thought to have been destroyed in the San ring in the fossil state in Great Britain, Part 1, Mastodon: Quar- Francisco earthquake and firestorm of 1906 before they could terly Journal of the Geological Society of London, v. 13, p. 307- be satisfactorily referred to species. Another specimen from 360. the same locality, apparently recovered some time in the 1880s, Hansen, R. M., 1980, Appendix III—Late Pleistocene plant fragments is currently part of the University of California, Berkeley, pale- in the dungs of herbivores at Cowboy Cave. In Jennings, J. D.; ontology collections, and this specimen is clearly from a Co- Holmer, R. N., editors, Cowboy Cave: University of Utah Anthro- lumbian mammoth. pological Papers, no. 104, p. 179-189. In eastern Washington, virtually all of the mammoth finds Hay, O. P., 1927, The Pleistocene of the western region of North are from the Columbian Basin. Here the earliest find of a Co- America and its vertebrated animals: Carnegie Institution of lumbian mammoth may be that of a partial mandible (jaw) and Washington Publication No. 332B, 346 p. molar recovered along the Walla Walla River in 1870 and now part of the collections of the University of Oregon. Other early

68 Washington Geology, vol. 26, no. 2/3, September 1998 Lawson, J. S., 1874, Letter accompanying donation to the museum: Proceedings of the California Academy of Science, v. V (1873-4), Earth Science Week Set for October p. 379-380. “The goal for Earth Science Week,” says American Geo- Shoshani, Jeheskel; Tassy, Pascal, editors, 1996, The Proboscidea— logical Institute (AGI) President Susan Landon, “is to have Evolution and palaeoecology of elephants and their relatives: Ox- every geoscientist in the country do something in his or her ford University Press, 472 p. community to promote the earth sciences.” As sponsor of Ukraintseve, V. V., 1993, Vegetation cover and environment of the Earth Science Week, AGI is a clearinghouse for ideas, ac- “Mammoth Epoch” in Siberia: Mammoth Site of Hot Springs, tivities, and special events, and provides support materials South Dakota, 309 p. that make it easy for volunteers to participate. Schools, uni- Washington Geology, 1998, Mammoth is now state fossil: Washing- versities, museums, state geological surveys, and AGI mem- ton Geology, v. 26, no. 1, p. 42. ber societies are planning Earth Science Week events. Webb, S. D.; Morgan, G. S.; Hulbert, R. C., Jr.; Jones, D. S.; MacFad- The governors of 14 states have issued Earth Science den, B. J.; Mueller, P. A., 1989, Geochronology of a rich early Week proclamations, and more are expected. The founda- Pleistocene vertebrate fauna, Leisey Shell Pit, Tampa Bay, Flor- tion of each proclamation is that geology and the earth sci- ida: Quaternary Research, v. 32, p. 96-110. n ences are fundamental to society and to our quality of life. Citizens who understand geology and the earth sciences can make wise decisions for land use and management. The ERRATA FROM THE PREVIOUS ISSUE earth sciences are crucial for addressing environmental and We had a major typesetting error in the article “New radiocar- ecological issues and provide the basis for preparing for and bon ages of major landslides in the Cascade Range, Washing- mitigating natural hazards. ton” in Washington Geology, v. 26, no. 1, p. 32. Part of the page Earth Science Week gained Congressional recognition should be replaced. Simply copy this page and cut along the on July 15 when Sen. Ron Wyden (D-Ore.) entered the Earth dashed line. Paste this over the text on p. 32 and all will be well. Science Week resolution into the Congressional Record. When citing the article in a list of references, please cite as The resolution, issued by the Association of American State follows: Geologists, designates the second week of October as Earth Pringle, P. T.; Schuster, R. L.; Logan, R. L., 1998, New radiocarbon Science Week. ages of major landslides in the Cascade Range, Washington: Information about Earth Science Week, as well as a copy Washington Geology, v. 26, no. 1, p. 31-39; errata, Washington of the Association of American State Geologists resolution, Geology, v. 26, no. 2/3, p. 69. is available from the American Geological Institute and on the World Wide Web at http://www.earthsciweek.org. (From an AGI news release, July 15, 1998.) Age estimates that have been derived from radiocarbon 14 ( C) dating methods are given as “yr B.P.”, meaning radiocar- bon years before present, where the “present” is A.D. 1950. Ra- Figure 3. Protruding logs exposed along Glacier Creek in Whatcom diocarbon years can differ slightly from calendar years because County were buried by the Church Mountain rock slide–debris ava- of variations in the carbon-isotope content of atmospheric car- lanche deposit about 2,340 yr B.P. The rock slide–debris avalanche originated at Church Mountain (partially cloud covered in the back- bon dioxide through time. For simplicity, raw radiocarbon ages ground) about 6.5 km to the northeast across the North Fork Nooksack are used in this text. Table 1 on p. 34 shows both raw ages and River valley. The lower log (in shadow) is western redcedar (sample calibrated ages. CMA-92-1), and the upper Douglas-fir log is the source of samples Previous researchers have suggested that rock slide and de- CMA-92-2 and mbr of Cary and others (1992) mentioned in Table 1. bris-avalanche deposits record prehistoric seismic shaking in Washington (Schuster and others, 1992, 1994, 1995; Engebret- son and others, 1995, 1996), neighboring British Columbia that drastically slows the rate of decomposition of drowned or (Clague and Shilts, 1993; Evans and Savigny, 1994), and else- buried snags (dead trees), including Douglas fir and western where (Keefer, 1984). The landslides mentioned herein were hemlock, which are susceptible to relatively rapid deteriora- likely to have been triggered by strong shaking for reasons tion in air (Cline and others, 1980; Harmon and others, 1986; similar to those noted by Schuster and others (1992) in their Fig. 2). By studying and radiocarbon dating the remains of study of rock avalanches in the eastern Olympic Mountains: plants, such as rooted trees that are preserved in the lakes (sub- z fossil trees), we can estimate the age of the landslide and of its The rocks that slid or avalanched have not failed at possible triggering earthquake. such scales historically during storms. At five of the sites mentioned in the introduction, landslide- z Worldwide, 29 of 71 rock avalanches that are included dammed lakes that were impounded at the time of the landslide in an inventory of landslide dams (Costa and Schuster, are still in existence. The drowned forests preserved in these 1991) were triggered by earthquakes having lakes, the snags or carbon preserved in deposits at the other lo- magnitudes of 6.0 or greater (Keefer, 1984). cations, and the landslide deposits are the subject of this paper. z In New Zealand, the distribution of lakes dammed by During the study, we also visited Tomyhoi Lake, a landslides approximates the locations of shallow landslide-dammed lake in the North Cascades (T40N, R5E); earthquakes of magnitude 6.5 or greater (Perrin and landslides along the Cascade River at Kindy Creek (T34N, Hancox, 1992). R12E), at Newaukum Lake (T14N, R3E), and at Hager Lake (T13N, R9E). In addition we visited three landslides along the Evidence of these earthquakes can be preserved when the upper Cispus River: west of Blue Lake (T11N, R9E) and at landslides bury forests and (or) create natural dams that cause a Johnson Creek and Wobbly Lake (both T11N, R10E). We lake to form and thereby a forest to drown. The lakes and land- could not find subfossil trees or other organic material suitable slide deposits commonly provide a fairly anoxic environment for dating at any of these sites. A carbon sample found beneath

Washington Geology, vol. 26, no. 2/3, September 1998 69 Selected Additions to the Library of the Division of Geology and Earth Resources March 1998 through July 1998

THESES Reasenberg, P. A., editor, 1998, The Loma Prieta, California, earth- quake of October 17, 1989—Aftershocks and postseismic effects: Brandes, J. A. G., 1996, Isotopic effects of denitrification in the ma- U.S. Geological Survey Professional Paper 1550-D, 312 p. rine environment: University of Washington Doctor of Philoso- phy thesis, 188 p. Switzer, J. C.; Jungblut, W. L.; Porcella, R. L., compilers, 1997, Cata- logue of U.S. Geological Survey strong-motion records, 1994: Chandler, R. D., 1995, Improving urban stormwater runoff monitor- U.S. Geological Survey Circular 1152, 34 p. ing practices: University of Washington Doctor of Philosophy thesis, 243 p. Vaccaro, J. J.; Hansen, A. J., Jr.; Jones, M. A., 1998, Hydrogeologic framework of the Puget Sound aquifer system, Washington and Doughty, P. T. B., 1995, Tectonic evolution of the Priest River com- British Columbia: U.S. Geological Survey Professional Paper plex and the age of basement gneisses—Constraints from geo- 1424-D, 77 p., 1 plate. chronology and metamorphic thermobarometry: Queen’s Univer- sity Doctor of Philosophy thesis, 372 p. Wiggins, W. D.; Ruppert, G. P.; Smith, R. R.; Reed, L. L.; Courts, M. L., 1998, Water resources data, Washington water year 1997: U.S. Haug, B. J., 1998, High resolution seismic reflection interpretations Geological Survey Water-Data Report WA-97-1, 528 p. of the Hood Canal–Discovery Bay fault zone; Puget Sound, Washington: Portland State University Master of Science thesis, Williamson, A. K.; Munn, M. D.; Ryker, S. J.; Wagner, R. J.; Ebbert, 1v. J. C.; Vanderpool, A. M., 1998, Water quality in the central Co- lumbia Plateau, Washington and Idaho, 1992–95: U.S. Geological Humphreys, P. W. D., 1998, Ground and surface water interaction Survey Circular 1144, 35 p. near a plywood manufacturing facility on the Lake Roosevelt shoreline, Kettle Falls, Washington: Western Washington Uni- Yashinsky, Mark, 1998, The Loma Prieta, California, earthquake of versity Master of Science thesis, 207 p. October 17, 1989—Highway systems: U.S. Geological Survey Professional Paper 1552-B, 191 p. Pearce, W. W., 1943, The Holden mine: University of Utah Bachelor of Science thesis, 32 p., 3 plates. Open-File and Water-Resources Investigations Reports Plimpton, C. L., 1984, An ethnoarchaeological study of Honeymoon Bacon, C. R.; Mastin, L. G.; Scott, K. M.; Nathenson, Manuel, 1997, Heights—The original camp of the Holden mine, Holden, Wash- Volcano and earthquake hazards in the Crater Lake region, Ore- ington: Washington State University Master of Arts thesis, 213 p. gon: U.S. Geological Survey Open-File Report 97-487, 32 p., Vaugeois, L. M., 1998, Stratigraphic, petrographic, and paleogeo- 1 plate. graphic analysis of the upper Eocene Hoko River Formation, Baum, R. L.; Chleborad, A. F.; Schuster, R. L., 1998, Landslides trig- Olympic Peninsula, Washington: Western Washington Univer- gered by the winter 1996–97 storms in the Puget Lowland, Wash- sity Master of Science thesis, 98 p., 2 plates, 1 CD-ROM. ington: U.S. Geological Survey Open-File Report 98-239, 16 p., 1 plate. U.S. GEOLOGICAL SURVEY Black, R. W.; Silkey, Mariabeth, 1998, Water-quality assessment of Published Reports the Puget Sound basin, Washington—Summary of stream biologi- Atwater, B. F.; Hemphill-Haley, Eileen, 1997, Recurrence intervals cal data through 1995: U.S. Geological Survey Water-Resources for great earthquakes of the past 3,500 years at northeastern Wil- Investigations Report 97-4164, 78 p. lapa Bay, Washington: U.S. Geological Survey Professional Pa- Brocher, T. M.; Ruebel, A. L., 1998, Compilation of 29 sonic and den- per 1576, 108 p. sity logs from 23 oil test wells in western Washington State: U.S. Bohlen, S. R.; Halley, R. B.; Hickman, S. H.; Johnson, S. Y.; Lowen- Geological Survey Open-File Report 98-249, 60 p. stern, J. B.; Muhs, D. R.; Plumlee, G. S.; Thompson, G. A.; Embrey, S. S.; Inkpen, E. L., 1998, Water-quality assessment of the Trauger, D. L.; Zoback, M. L., 1998, Geology for a changing Puget Sound basin, Washington, nutrient transport in rivers, world—A science strategy for the Geologic Division of the U.S. 1980–93: U.S. Geological Survey Water-Resources Investiga- Geological Survey, 2000–2010: U.S. Geological Survey Circular tions Report 97-4270, 30 p. 1172, 59 p. Embrey, S. S.; Jones, J. L., 1998, Reconnaissance hydrogeology Dinehart, R. L., 1998, Sediment transport at gaging stations near and water quality of the Swinomish Indian Reservation, Skagit Mount St. Helens, Washington, 1980–90—Data collection and County, Washington: U.S. Geological Survey Water-Resources analysis: U.S. Geological Survey Professional Paper 1573, 105 p. Investigations Report 96-4031, 28 p., 1 plate. Holzer, T. L., editor, 1998, The Loma Prieta, California, earthquake Finn, C. A.; Brenner, K. C.; McCafferty, Anne; Kucks, R. P., 1998, of October 17, 1989—Liquefaction: U.S. Geological Survey Pro- Merged aeromagnetic data for Washington—A web site for distri- fessional Paper 1551-B, 314 p., in folder with 3 plates. bution of gridded data and plot files: U.S. Geological Survey Johnson, S. Y.; Tennyson, M. E.; Lingley, W. S., Jr.; Law, B. E., Open-File Report 98-241, 1 v. 1997, Petroleum geology of the State of Washington: U.S. Geo- Frost, T. P.; Raines, G. L.; Almquist, C. L.; Johnson, B. R., 1996, logical Survey Professional Paper 1582, 40 p. Digital map of possible bat habitats for the Pacific Northwest—A Keefer, D. K., editor, 1998, The Loma Prieta, California earthquake contribution to the Interior Columbia River Basin Ecosystem of October 17, 1989—Landslides: U. S. Geological Survey Pro- Management Project: U.S. Geological Survey Open-File Report fessional Paper 1551-C, 185 p., in folder with 5 plates. 95-683, 34 p. McFarland, W. D.; Morgan, D. S., 1996, Description of the ground- Johnson, B. R.; Derkey, P. D., 1998, Digital geologic map of the Spo- water flow system in the Portland basin, Oregon and Washington: kane 1:100,000 quadrangle, Washington and Idaho—A digital U.S. Geological Survey Water-Supply Paper 2470-A, 58 p., 7 database for the 1990 N. L. Joseph map: U.S. Geological Survey plates. Open-File Report 98-115, 13 p.

70 Washington Geology, vol. 26, no. 2/3, September 1998 Jones, J. L.; Roberts, L. M., 1998, Shallow ground-water quality be- Peterson, John, 1998, A city’s perspective to the 1997 landslides. neath row crops and orchards in the Columbia Basin Irrigation [47 p., unpaginated.] Project area, Washington: U.S. Geological Survey Water- Walker, Jim, 1998, City of Edmonds landslide policy and Mead- Resources Investigations Report 97-4238, 29 p. owdale area case history. [46 p., unpaginated.] Kiilsgaard, T. H., 1998, Mining properties in Washington that were Ballard, Sanford, 1995, Ground water flow velocity in the bank of the involved in the DMA, DMEA, or OME mineral exploration pro- Columbia River, Hanford, Washington: Sandia National Labora- grams, 1950–1974: U.S. Geological Survey Open-File Report 93- tories SAND95-2187, 18 p. 232, 32 p., 1 plate. Buchanan, J. P.; Olness, I. A., 1993, Ground water flow model of the Knepper, D. H., Jr.; Langer, W. H.; Miller, S. H., 1994, Remote sens- Spokane Valley portion of the Spokane Valley–Rathdrum Prairie ing and airborne geophysics in the assessment of natural aggre- aquifer system: Spokane County Engineers contract report, 1 v. gate resources; U.S. Geological Survey Open-File Report 94-158, Includes: 1v. Olness, I. A., 1993, Formulation of a finite-difference ground- Krimmel, R. M., 1998, Water, ice, and meteorological measurements water flow model for the Spokane Valley aquifer, Washing- at South Cascade Glacier, Washington, 1997 balance year: U.S. ton. 1 v. Geological Survey Water-Resources Investigations Report 98- 4090, 30 p. Bucy, L. K.; Funk, W. H., compilers, 1996, Lake Roosevelt manage- ment plan: Washington Water Research Center Report 93, 127 p. Munn, M. D.; McHenry, M. L.; Sampson, V., 1996, Benthic macroin- vertebrate communities in the Elwha River basin, 1994–95: U.S. Campbell, N. P., 1998, Geology of the Yakima area; rev. ed.: [Pri- Geological Survey Open-File Report 96-588, 27 p. vately published by the author], 68 p. Snyder, D. T.; Wilkinson, J. M.; Orzol, L. L., 1996, Use of a ground- Dames & Moore, 1998, Proposal—Construction review of Ahtanum water flow model with particle tracking to evaluate ground-water reservoir project, Ahtanum Irrigation District: Dames & Moore, vulnerability, Clark County, Washington: U.S. Geological Sur- 1v. vey Open-File Report 96-328, 71 p. Dean, N. E.; Boldt, C. M. K.; Schultze, L. E.; Nilsen, D. N.; Isaacson, Sumioka, S. S.; Kresch, D. L.; Kasnick, K. D., 1998, Magnitude and A. E.; Williams, B. C.; Kirschner, F. E.; Moore, B. W.; Stroud, W. frequency of floods in Washington: U.S. Geological Survey P.; Peters, D. C., 1996, Midnite mine summary report: U.S. Water-Resources Investigations Report 97-4277, 91 p., 1 plate. Bureau of Mines Report of Investigations 9604, 18 p. Thomas, B. E.; Cox, S. E., 1998, Ground-water age, flow, and quality Erickson, Denis, 1998, Sumas–Blaine surficial aquifer nitrate charac- near a landfill, and changes in ground-water conditions from 1976 terization: Washington Department of Ecology Publication 98- to 1996 in the Swinomish Indian Reservation, northwestern 310, 1 v. Washington: U.S. Geological Survey Water-Resources Investiga- Ganz, Steven; Traynor, Theresa, compilers, 1998, Tsunami hazard tions Report 98-4014, 58 p. mitigation symposium—Proceedings volume: Western States Seismic Policy Council, 78 p. U.S. Geological Survey Contract Reports Includes: Carpenter, J. E., 1996, Workshops to explain the changes resulting Bernard, E. N., 1998, Session—Opening session; Topic—The from seismic rezone from UBC Zone 2B to Zone 3 in southwest National Tsunami Hazard Mitigation Program. p. 5-8. Washington State: U.S. Geological Survey contract report, 7 p. Blackford, M. E., 1998, Session—Warning guidance; Topic—Pa- Hyndman, R. D.; Wang, Kelin; Dragert, Herb, 1997, Cascadia mega- cific Tsunami Warning Center operations. p. 9-22. thrust earthquake potential—Constraints from current deforma- Darienzo, M. E., 1998, Session—Efforts in mitigation, public tion and thermal regime; Final report: U.S. Geological Survey awareness and emergency response; Topic—Efforts in miti- contract report, 22 p. gation, public awareness, and emergency response. p. 37. Trehu, A. M., 1997, Collaborative research (USGS/OSU/UTEP)—A Dengler, L. A., 1998, Session—Efforts in mitigation, public high-resolution seismic reflection and refraction survey of the awareness and emergency response; Topic—Tsunami mitiga- Washington study corridor; Final technical report: U.S. Geologi- tion efforts on California’s North Coast. p. 39-43. cal Survey contract report, 15 p. Hegemeyer, Richard, 1998, Session—National Tsunami Hazard OTHER REPORTS ON WASHINGTON GEOLOGY Mitigation Program; Topic—Tsunami hazard mitigation. p. 27-33. American Society of Civil Engineers, Seattle Section; University of Washington Department of Civil Engineering; U.S. Geological Hull, D. A.; Karel, Angie, 1998, Session—Efforts in mitigation, Survey, 1998, Landslides in the Puget Sound region—Seminar: public awareness and emergency response; Topic—Oregon American Society of Civil Engineers, Seattle Section, 1 v. strategy for mitigation and public awareness. p. 45-63. Includes: Oppenheimer, D. H., 1998, Session—Efforts in mitigation, public awareness and emergency response; Topic—Consolidated re- Gilbert, Wade; Laprade, W. T., 1998, Woodway landslide—A re- porting of earthquakes and tsunamis. p. 65-67. minder and an opportunity. 16 p. Sokolowski, T. J., 1998, Session—Warning guidance; The West Horvitz, G. E.; Beaver, J. R.; Whelan, M. P., 1998, Cost and miti- Coast/Alaska Tsunami Warning Center. p. 23-25. gation of large-scale landslides. [73 p., unpaginated.] Walsh, T. J., 1988, Session—National Tsunami Hazard Mitiga- Koloski, J. W., 1998, Humans as a geologic agent. [5 p., unpagi- tion Program; Topic—Tsunami hazards in Washington. p. 35. nated.] Yanagi, Brian, 1998, Session—Efforts in mitigation, public Mann, Glen, 1998, Costs and challenges of mitigating small-scale awareness and emergency response; Topic—Internet tsunami landslides. 31 p. resources. p. 69. Morlan, L. E., 1998, Conclusions and recommendations followed Goodrich, Randi; Goodrich, Michael, 1996, A rock grows up—The by questions and answers to the speakers. [28 p., unpagi- Pacific Northwest, up close and personal: GeoQuest Publications nated.] [Lake Oswego, Ore.], 52 p. Palmer, L. A., 1998, Coastal bluff landslides—Magnitude, mechanism, and management. 13 p.

Washington Geology, vol. 26, no. 2/3, September 1998 71 Huckell/Weinman Associates, Inc.; and others, 1995, Cadman Black Booth, D. B., 1991, Urbanization and the natural drainage system— Diamond mine pit expansion—Draft environmental impact state- Impacts, solutions, and prognoses: Northwest Environmental ment: King County Department of Development and Environ- Journal, v. 7, no. 1, p. 93-118. mental Services, 1 v. Brandon, M. T.; Ague, J. J., 1998, Regional tilt of the Mount Stuart Johnson, Art; White, Jody; Huntamer, Dickey, 1997, Effects of batholith, Washington, determined using aluminum-in- Holden mine on the water, sediments, and benthic invertebrates of hornblende barometry—Implications for northward translation of Railroad Creek (Lake Chelan): Washington Department of Ecol- Baja British Columbia—Discussion and reply; Reply: Geological ogy Publication 97-330, 1 v. Society of America Bulletin, v. 110, no. 5, p. 687-690. Kirschner, F. E., 1996, Groundwater flow model (GWFM) develop- Burges, S. J.; Wigmosta, M. S.; Meena, J. M., 1998, Hydrological ef- ment, Midnite mine, Wellpinit, WA: U.S. Bureau of Mines Report fects of land-use change in a zero-order catchment: Journal of Hy- of Investigations 9608, 6 p. drologic Engineering, v. 3, no. 2, p. 86-97. Lingley, W. S., Jr.; Logan, R. L.; Walsh, T. J.; Gerstel, W. J.; Schasse, Carey, S. N., 1996, Modeling of tephra fallout from explosive erup- H. W., 1996?, Reconnaissance geology of the Matheny Ridge– tions. In Scarpa, Roberto; Tilling, R. I., editors, Monitoring and Higley Peak areas, Olympic Peninsula, Washington: Washington mitigation of volcano hazards: Springer-Verlag, p. 429-461. Division of Geology and Earth Resources [contract report], 31 p., Chesley, J. T.; Ruiz, Joaquin, 1998, Crust-mantle interaction in large 1 plate. igneous provinces—Implications from the Re-Os isotope syste- Northwest Mining Association, 1998, Service directory: Northwest matics of the Columbia River flood basalts: Earth and Planetary Mining Association, 324 p. Science Letters, v. 154, p. 1-11. Orsen, M. J., 1998, Ginkgo petrified forest: Ginkgo Gem Shop [Van- Clague, J. J.; Turner, Robert; Groulx, Bertrand, 1998, Initiatives in tage, Wash.], 24 p. geoliteracy—Making earth science accessible to Vancouver resi- Peters, D. C.; Smith, M. A.; Ferderer, D. A., 1996, Data dictionary and dents: Innovation, v. 2, no. 4, p. 12-16. discussion for the Midnite mine GIS database: U.S. Bureau of Cornish, John, 1998, Minerals from the Weyerhaeuser Lincoln Creek Mines Report of Investigations 9611, 2 v. quarry, Lewis County, Washington: Rocks and Minerals, v. 73, Schultze, L. E.; Nilsen, D. N.; Isaacson, A. E.; Lahoda, E. J., 1996, no. 4, p. 268-275. U.S. Bureau of Mines final report—Midnite mine water treatment Cowie, G. L.; Hedges, J. I., 1992, Sources and reactivities of amino studies: U.S. Bureau of Mines Report of Investigations 9605, 52 p. acids in a coastal marine environment: Limnology and Oceanog- Shipman, Hugh, 1998, Shoreline changes at North Beach, Samish raphy, v. 37, no. 4, p. 703-724. Island: Washington Department of Ecology Publication 98-101, 1 v. Dahlgren, R. A.; Dragoo, J. P.; Ugolini, F. C., 1997, Weathering of U.S. Environmental Protection Agency, 1995, Declaration of the rec- Mt. St. Helens tephra under a cryic-udic climatic regime: Soil Sci- ord of decision—USDOE Hanford environmental restoration dis- ence Society of America Journal, v. 61, no. 5, p. 1519-1525. posal facility, Hanford site, Benton County, Washington: U.S. Danner, W. R., 1997, Fusulinids and other Paleozoic foraminifera of Environmental Protection Agency, 1 v. accreted terranes, southwestern British Columbia and northwest- Washington Department of Natural Resources, 1996, Hazel water- ern Washington. In Ross, C. A.; Ross, J. R. P.; Brenckle, P. L., shed analysis: Washington Department of Natural Resources, 1 editors, Late Paleozoic foraminifera—Their biostratigraphy, evo- v., 19 plates. lution, and paleoecology and the Mid-Carboniferous boundary: Washington State Emergency Management Council, Seismic Safety Cushman Foundation for Foraminiferal Research Special Publi- Subcommittee, 1998, Earthquake safety in Washington State—A cation 36, p. 21-26. progress report on activities for the period 1992–1997: Washing- del Moral, Roger, 1998, Early succession on lahars spawned by ton State Emergency Management Council, 1 v. Mount St. Helens: American Journal of Botany, v. 85, no. 6, p. Wildrick, L. L.; Culhane, Tom; Davidson, Don; Sinclair, K. A., 1998, 820-828. Watershed assessment, water resource inventory area 28, Dodson, Allen; Kennedy, B. M.; DePaolo, D. J., 1997, Helium and Salmon–Washougal: Washington Department of Ecology Open- neon isotopes in the Imnaha Basalt, Columbia River Basalt File Technical Report 98-02, 1 v. Group—Evidence for a Yellowstone plume source: Earth and Williams, B. C.; Riley, J. A., 1996, Hydraulic characterization of Planetary Science Letters, v. 150, no. 3-4, p. 443-451. Midnite mine, Wellpinit, WA—Summary of 1994 field season: Doughty, P. T.; Price, R. A.; Parrish, R. R., 1998, Geology and U-Pb U.S. Bureau of Mines Report of Investigations 9606, 33 p. geochronology of Archean basement and Proterozoic cover in the Williams, B. C.; Riley, J. A.; Montgomery, J. R.; Robinson, J. A., Priest River complex, northwestern United States, and their impli- 1996, Hydrologic and geophysical studies at Midnite mine, cations for Cordilleran structure and Precambrian continent re- Wellpinit, WA—Summary of 1995 field season: U.S. Bureau of constructions: Canadian Journal of Earth Sciences, v. 35, no. 1, Mines Report of Investigations 9607, 34 p. p. 39-54. Drake, Bob, 1998, Plant plans pay off: Rock Products, v. 101, no. 4, PAPERS ON WASHINGTON GEOLOGY p. 37-42. Benito, Gerardo, 1997, Energy expenditure and geomorphic work of Drake, Deanne; Ewing, Kern; Dunn, Patrick, 1998, Techniques to the cataclysmic Missoula flooding in the Columbia River Gorge, promote germination of seed from Puget Sound prairies: Restora- USA: Earth Surface Processes and Landforms, v. 22, no. 5, p. 457- tion & Management Notes, v. 16, no. 1, p. 33-40. 472. Garvin, J. B., 1996, Topographic characterization and monitoring of Bernard, E. N., 1998, Program aims to reduce impact of tsunamis on volcanoes via airborne laser altimetry. In McGuire, W. J.; Jones, Pacific states: Eos (American Geophysical Union Transactions), A. P.; Neuberg, J., editors, Volcano instability on the earth and v. 79, no. 22, p. 258, 262-263. other planets: Geological Society Special Publication 110, p. 137- Bohrmann, Gerhard; Greinert, Jens; Suess, Erwin; Torres, Marta, 152. 1998, Authigenic carbonates from the Cascadia subduction zone Geiger, Beth, 1998, Heaps of confusion: Earth, v. 7, no. 4, p. 34-37. and their relation to gas hydrate stability: Geology, v. 26, no. 7, Geist, D. R.; Joy, M. C.; Lee, D. R.; Gonser, Tom, 1998, A method for p. 647-650. installing piezometers in large cobble bed rivers: Ground Water Monitoring & Remediation, v. 18, no. 1, p. 78-82.

72 Washington Geology, vol. 26, no. 2/3, September 1998 Guilbault, J.-P.; Clague, J. J.; Lapointe, Martine, 1995, Amount of shoreline, Commencement Bay Superfund Site, Tacoma, Wash- subsidence during a late Holocene earthquake—Evidence from ington: Environmental Science & Technology, v. 30, no. 5, fossil tidal marsh foraminifera at Vancouver Island, west coast of p. 1645-1651. Canada: Palaeogeography, Palaeoclimatology, Palaeoecology, Martin, Ken, 1998, Nineteenth Century gold rushes—The stampede v. 118, no. 12, p. 49-71. north: International California Mining Journal, v. 67, no. 9, p. 40- Hammond, P. E., 1998, Fifes Peaks volcano, southern Washington 42. Cascade Range—A rock climber’s challenge: Mazama, v. 79, McCabe, G. T., Jr.; Hinton, S. A.; Emmett, R. L., 1998, Benthic inver- no. 13, p. 30-32, 34. tebrates and sediment characteristics in a shallow navigation Hart, B. S.; Hamilton, T. S.; Barrie, J. V., 1998, Sedimentation rates channel of the lower Columbia River, before and after dredging: and patterns on a deep-water delta (Fraser delta, Canada)—Inte- Northwest Science, v. 72, no. 2, p. 116-126. gration of high-resolution seismic stratigraphy, core lithofacies, McGown, Robert, 1998, North Cascades glacier climate project: and 137Cs fallout stratigraphy: Journal of Sedimentary Research, Mazama, v. 79, no. 13, p. 26, 28. v. 68, no. 4, p. 556-568. McKinley, J. P.; Stevens, T. O.; Fredrickson, J. K.; Zachara, J. M.; Heylmun, E. B., 1998, Gold at Monte Cristo, Washington: Interna- Colwell, F. S.; Wagnon, K. B.; Smith, S. C.; Rawson, S. A.; Bjorn- tional California Mining Journal, v. 67, no. 7, p. 15-16, 19-20. stad, B. N., 1998, Biogeochemistry of anaerobic lacustrine and Iriyama, Jun, 1996, An observation report on the Newberry and Paleosol sediments within an aerobic unconfined aquifer: Geomi- Mount St. Helens volcanoes—Participation in the 1996 GRC re- crobiology Journal, v. 14, no. 1, p. 23-39. search field trip: Chinetsu (Journal of the Japan Geothermal En- McKnight, D. M.; Thorn, K. A.; Wershaw, R. L., 1987, Changes in ergy Association), v. 33, no. 4, p. 64-69. dissolved organic material in Spirit Lake, Washington, USA: Sci- Jacobel, R. W.; Anderson, S. K.; Rioux, D. F., 1992, A microproces- ence of the Total Environment, v. 62, [Special Issue], p. 187-192. sor based ice-radar system for surface profiling. In Pilon, J. A., McNutt, S. R., 1996, Seismic monitoring and eruption forecasting of editor, Ground penetrating radar: Geological Survey of Canada volcanoes—A review of the state-of-the-art and case histories. In Paper 90-4, p. 101-105. Scarpa, Roberto; Tilling, R. I., editors, Monitoring and mitigation Jennings, D. A.; Petersen, J. N.; Skeen, R. S.; Peyton, B. M.; Hooker, of volcano hazards: Springer-Verlag, p. 99-146. B. S.; Johnstone, D. L.; Yonge, D. R., 1995, An experimental Medici, John, 1996, USA—Die abenteuerliche Bergung der schoenen study of microbial transport in porous media. In Hinchee, R. E.; Pyrite von Washington [USA—The adventure of saving beautiful Fredrickson, Jim; Alleman, B. C., editors, Bioaugmentation for Washington pyrite]. In Weise, Christian, editor, Pyrit und Marka- site remediation: Battelle Press, Bioremediation 3, v. 3, p. 97-103. sit—Das eiserne Ueberall-Mineral [Pyrite and marcasite—The Jones, T. L.; Serne, R. J., 1995, Contaminant release from solidified widespread iron mineral: ExtraLapis, v. 11, p. 66-71. radioactive wastes buried in unsaturated sediments—Lysimeter Meier, M. F.; Armstrong, R. L.; Dyurgerov, M. B., 1997, Comments study: Journal of Environmental Quality, v. 24, no. 6, p. 1063- on “Annual net balance of North Cascades glaciers, 1984–94,” by 1073. M. S. Pelto: Journal of Glaciology, v. 43, no. 143, p. 192-193. Lane, Jeff, 1998, Chehalis Red: North American Prospectors Crack Miller, M. M.; Dragert, Herb; Endo, E. T.; Freymueller, J. T.; Gold- and Crevice, April 1998, p. 9. finger, Chris; Kelsey, H. M.; Humphreys, E. D.; Johnson, D. J.; Laprade, W. T., 1993, The geology of Queen Anne Hill. In Reinartz, McCaffrey, Robert; Oldow, J. S.; Qamar, A. I.; Rubin, C. M., K. F., Queen Anne—Community on the hill: Queen Anne Histori- 1998, Precise measurements help gauge Pacific Northwest’s cal Society, p. 1-6. earthquake potential: Eos (American Geophysical Union Trans- Lasmanis, Raymond; Norman, D. K.; Cannon, Bart, 1998, Prelimi- actions), v. 79, no. 23, p. 269, 275. nary study of minerals in Tacoma smelter slags: International As- Mofjeld, H. O.; Foreman, M. G. G.; Ruffman, Alan, 1997, West Coast sociation of Collectors of Slag Minerals Newsletter, v. 8, no. 2, p. tides during Cascadia subduction zone tsunamis: Geophysical Re- 2-8. search Letters, v. 24, no. 17, p. 2215-2218. Lee, J. M.; Nahajski, Anthony; Miller, Sky, 1997, Riverbank stabili- Morell, Virginia, 1998, Kennewick man’s trials continue: Science, zation program: Journal of Water Resources Planning and Man- v. 280, no. 5361, p. 190-192. agement, v. 123, no. 5, p. 292-294. Newcomer, D. R.; Hall, S. H.; Vermeul, V. R., 1996, Use of improved Lescinsky, D. T.; Sisson, T. W., 1998, Ridge-forming, ice-bounded hydrologic testing and borehole geophysical logging methods for lava flows at Mount Rainier, Washington: Geology, v. 26, no. 4, aquifer characterization: Ground Water Monitoring & Remedia- p. 351-354. tion, v. 16, no. 1, p. 67-72. Lindberg, D. R.; Hedegaard, Claus, 1996, A deep water patellogastro- Newhall, C. G.; Punongbayan, R. S., 1996, The narrow margin of suc- pod from Oligocene water-logged wood of Washington State, cessful volcanic risk mitigation. In Scarpa, Roberto; Tilling, R. I., USA (Acmaeoidea: Pectinodonta): Journal of Molluscan Studies, editors, Monitoring and mitigation of volcano hazards: Springer- v. 58, no. 4, p. 433-441. Verlag, p. 807-838. Liszak, J. L.; Ross, C. A., 1997, Foraminifera and associated faunas, Obermeier, S. F., 1996, Use of liquefaction-induced features for pa- Lower Carboniferous–Lower Permian Chilliwack Group, Black leoseismic analysis—An overview of how seismic liquefaction Mountain, northwest Washington. In Ross, C. A.; Ross, J. R. P.; features can be distinguished from other features and how their re- Brenckle, P. L., editors, Late Paleozoic foraminifera—Their gional distribution and properties of source sediment can be used biostratigraphy, evolution, and paleoecology and the Mid- to infer the location and strength of Holocene paleo-earthquakes: Carboniferous boundary: Cushman Foundation for Foraminifera Engineering Geology, v. 44, no. 1-4, p. 1-76. Research Special Publication 36, p. 89-93. Oil and Gas Journal, 1998, Resource estimate climbs on Pacific OCS: Liu, Mian; Shen, Yunqing, 1998, Crustal collapse, mantle upwelling, Oil & Gas Journal, v. 96, no. 11, p. 93-96. and Cenozoic extension in the North American Cordillera: Tec- Parsons, Tom; Trehû, A. M.; Luetgert, J. H.; Miller, K. C.; Kilbride, tonics, v. 17, no. 2, p. 311-321. Fiona; Wells, R. E.; Fisher, M. A.; Flueh, E. R.; ten Brink, U. S.; Mariner, P. E.; Holzmer, F. J.; Jackson, R. E.; Meinardus, H. W.; Christensen, N. I., 1998, A new view into the Cascadia subduction Wolf, F. G., 1996, Effects of high pH on arsenic mobility in a shal- zone and volcanic arc—Implications for earthquake hazards low sandy aquifer and on aquifer permeability along the adjacent along the Washington margin: Geology, v. 26, no. 3, p. 199-202.

Washington Geology, vol. 26, no. 2/3, September 1998 73 Paterson, S. R.; Miller, R. B., 1998, Regional tilt of the Mount Stuart Yuan, Jun; Murray, H. H., 1997, The importance of crystal morphol- batholith, Washington, determined using aluminum-in-horn- ogy on the viscosity of concentrated suspensions of kaolins: Ap- blende barometry—Implications for northward translation of plied Clay Science, v. 12, no. 3, p. 209-219. Baja British Columbia—Discussion and reply; Discussion: Geo- Zierenberg, R. A.; Fouquet, Yves; Miller, D. J.; Bahr, J. M.; Baker, P. logical Society of America Bulletin, v. 110, no. 5, p. 685-690. A.; Bjerkgard, T.; Brunner, C. A.; Duckworth, R. C.; Gable, R.; Pelto, M. S., 1996, Annual net balance of North Cascade glaciers, and others, 1998, The deep structure of a sea-floor hydrothermal 1984–94: Journal of Glaciology, v. 42, no. 140, p. 3-9. deposit: Nature, v. 392, no. 6675, p. 485-488. Pelto, M. S., 1997, Reply to the comments by Meier and others on Zobin, V. M., 1996, The epicentral areas of volcanic earthquakes and “Annual net balance of North Cascades glaciers, 1984–94,” by M. their volcanotectonic significance: Geofisica Internacional, v. 35, S. Pelto: Journal of Glaciology, v. 43, no. 143, p. 193-196. no. 2, p. 95-103. Pelto, M. S., 1998, Changes in glaciers on Glacier Peak in the last 100 OTHER REPORTS OF INTEREST years: Mazama, v. 79, no. 13, p. 29, 39 Pickett, P. J., 1997, Pollutant loading capacity for the Black River, American Geological Institute, compiler, 1997, Dictionary of mining, Chehalis River system, Washington: Journal of the American Wa- mineral, and related terms; 2nd ed.: American Geological Insti- ter Resources Association, v. 33, no. 2, p. 465-480. tute, 646 p. Piedmouguet, S.; Cailletaud, J.-P., translators, 1988, Mesurer jusque Bernard, E. N., editor, 1991, Tsunami hazard—A practical guide for dans le futur [Measuring into the future]: Geometre, v. 132, no. 3, tsunami hazard reduction: Kluwer Academic Publishers, 1 v. p. 60-63. Bobrowsky, P. T., editor, 1998, Aggregate resources—A global per- Poeter, Eileen; Townsend, Peter, 1991, Application of multiple indi- spective: A. A. Balkema, 470 p. cator conditional stochastic simulation: GeoTech/Geochautauqua Borchers, J. W., editor, 1998, Land subsidence case studies and cur- ‘91, p. 41-49. rent research—Proceedings of the Dr. Joseph F. Poland Sympo- Rasmussen, M. G.; Evans, B. W.; Kuehner, S. M., 1998, Low- sium on Land Subsidence: Star Publishing Company (Association temperature fayalite, greenalite, and minnesotaite from the Over- of Engineering Geologists Special Publication 8), 576 p. look gold deposit, Washington—Phase relations in the system California Seismic Safety Commission, 1997, California earthquake FeO-SiO2-H20: Canadian Mineralogist, v. 36, Part 1, p. 147-162. loss reduction plan, 1997–2001: California Seismic Safety Com- Reeves, G. H.; Benda, L. E.; Burnett, K. M.; Bisson, P. A.; Sedell, J. mission, 49 p. R., 1995, A disturbance-based ecosystem approach to maintain- Geological Survey of Canada, 1998, Cordillera and Pacific Margin: ing and restoring freshwater habitats of evolutionarily significant Geological Survey of Canada Current Research 1998-A, 1 v. units of anadromous salmonids in the Pacific Northwest. In Niel- Includes: sen, J. L.; Powers, D. A., editors, Evolution and the aquatic Clague, J. J.; Bobrowsky, P. T.; Hutchinson, I.; Mathewes, R. W., ecosystem—Defining unique units in population conservation: 1998, Geological evidence for past large earthquakes in American Fisheries Society Symposium 17, p. 334-349. southwest British Columbia. p. 217-224. Scherer, George; Zabowski, Darlene; Everett, Richard, 1996, Nutri- Claudy, N. H., editor, 1997, National directory of geoscience data re- ent content and survival of three native species on ameliorated positories: American Geological Institute, 91 p. abandoned copper mine tailings at Holden, WA: Association of Abandoned Mine Land Programs, 18th Annual Conference, p. 35- Fisher, R. V.; Heiken, Grant; Hulen, J. B., 1997, Volcanoes—Cruci- 43. bles of change: Princeton University Press, 317 p. Stasiuk, M. V.; Jaupart, Claude, 1997, Lava flow shapes and dimen- Francis, Peter, 1997?, Volcanoes—A planetary perspective; Slide set: sions as reflections of magma system conditions: Journal of Vol- [Privately printed by the author], 8 p., 40 slides. canology and Geothermal Research, v. 78, no. 1-2, p. 31-50. Gusiakov, V. K., editor, 1997, Tsunami mitigation and risk assess- Su, Chunming; Harsh, J. B., 1996, Alteration of imogolite, allophane, ment—Report of the International Workshop, Petropavlovsk- and acidic soil clays by chemical extractants: Soil Science Soci- Kamchatskiy, Russia, August 21–24, 1996: Russian Academy of ety of America Journal, v. 60, no. 1, p. 77-85. Sciences, 68 p. Svedela Industri AB, 1998, New sand and gravel plant maintains qual- Includes: ity of life: Coal Age, v. 103, no. 3, Supplement, p. 32, 35. Preuss, Jane; Priest, G. R., 1997, Tsunami hazard mitigation and Symons, N. P.; Crosson, R. S., 1997, Seismic velocity structure of the counter measures with an example from Oregon. p. 43-46. Puget Sound region from 3-D non-linear tomography: Geophysi- Giancola, Diane, editor, 1997, Canadian mines handbook, 1997–98: cal Research Letters, v. 24, no. 21, p. 2593-2596. Southam Mining Publications Group, 684 p. Tschernich, R. W., 1998, Extraordinary natolite-lined pockets at the Ham, S. H., 1992, Environmental interpretation—A practical guide Weyerhaeuser Lincoln Creek quarry, Doty Hills, Lewis County, for people with big ideas and small budgets: North American Washington: Micro Probe, v. 8, no. 7, 7 p. Press, 456 p. Ugolini, F. C.; Minden, R.; Dawson, H. J.; Zachara, J. M., 1977, An Hambrey, Michael; Alean, Jurg, 1992, Glaciers: Cambridge Univer- example of soil processes in the Abies amabilis zone of central sity Press, 207 p. Cascades, Washington: Soil Science, v. 124, no. 5, p. 291-302. Jackson, J. A., editor, 1997, Glossary of geology; 4th ed.: American Ward, P. D., 1998, Coils of time: Discover, March 1998, p. 100-106. Geological Institute, 769 p. Watanabe, Yasuchi; Kubota, Yoshihiro; Hayashi, Toshihiko; Yoshi- Kelsey, H. M.; Witter, R. C.; Hemphill-Haley, Eileen, 1998, Re- nari, Akio, 1996, Outline of the epithermal gold deposits in the sponse of a small Oregon estuary to coseismic subsidence and western U.S.: Chishitsu News, v. 3, no. 499, p. 39-52. postseismic uplift in the past 300 years: Geology, v. 26, no. 3, p. Wilson, M. A.; Burt, R.; Sobecki, T. M.; Engel, R. J.; Hipple, K., 231-234. 1996, Soil properties and genesis of pans in till-derived andisols, Knutson, K. L.; Naef, V. L., 1997, Management recommendations for Olympic Peninsla, Washington: Soil Science Society of America Washington’s priority habitats—Riparian: Washington Depart- Journal, v. 60, no. 1, p. 206-218. ment of Fish and Wildlife, 181 p.

74 Washington Geology, vol. 26, no. 2/3, September 1998 Kuroiwa, Julio; Zupka, Dusan, reviser; HOW TO FIND OUR MAIN OFFICE Schneider, Christine V., translator,

Washington St. 1995, Tsunamis—Population evacua- Capitol Way tion and land use planning for disaster Franklin St. mitigation; Localities studied in Peru Plum St. (1981–1994): UNDHA, 1 v. Union Ave. Luckman, B. H.; Briffa, K. R.; Jones, P. D.; Schweingruber, F. H., 1997, Tree- 11th Ave. EXIT ring based reconstruction of summer VP temperatures at the Columbia Ice- H Natural 105B field, Alberta, Canada, AD 1073– Resources 1983: Holocene, v. 7, no. 4, p. 375- Building 389. 14th To Menzies, John, editor, 1995, Modern gla- Ave. EXIT Seattle State 105A cial environments—Processes, dy- Capitol

namics and sediments: Butterworth Jefferson St. Heinemann Glacial Environments, Capitol Lake 5 Maple Park Ave. v. 1, 621 p.

Capitol Way Menzies, John, editor, 1996, Past glacial environments—Sediments, forms and Division of Geology and Earth Resources techniques: Butterworth Heinemann Natural Resources Bldg., Room 148 Glacial Environments, v. 2, 598 p. 1111 Washington St. S.E. Olympia, WA 98501 EXIT Meyer, H. W.; Manchester, S. R., 1997, 105 The Oligocene Bridge Creek Flora of (See p. 2 for our mailing address.) Visitor parking (VP) is available on the John Day Formation, Oregon: Level P1 at $.50/hour. Use the Henderson Blvd. University of California Publications Washington St. entrance. Geological Sciences, v. 141, 195 p.,

75 photo plates. To Portland Pojar, Jim; MacKinnon, Andy, compil- ers and editors, 1994, Plants of the Pacific Northwest coast—Washington, Oregon, British Colum- U.S. Federal Emergency Management Agency, 1998, Multi-hazard bia, and Alaska: British Columbia Ministry of Forests, 527 p. identification and risk assessment—The cornerstone of the na- Priest, G. R.; Myers, E. P., III; Baptista, A. M.; Kamphaus, R. A.; Fie- tional mitigation strategy: U.S. Federal Emergency Management dorowicz, B. K.; Peterson, C. D.; Horning, T. S., 1998, Tsunami Agency, 369 p. hazard map of the Seaside–Gearhart area, Clatsop County, Ore- Washington Department of Natural Resources, 1997, Final habitat gon: Oregon Department of Geology and Mineral Industries Inter- conservation plan: Washington Department of Natural Resources, pretive Map Series IMS-3, 1 sheet, scale 1:12,000. 1v. Schultz, R. A.; Siddharthan, R. V., editors, 1998, Engineering geol- Washington Department of Transportation, 1998, Standard specifica- ogy and geotechnical engineering; Proceedings of the 33rd sym- tions for road, bridge, and municipal construction, 1998—Eng- posium: University of Nevada, Reno Department of Geological lish: Washington Department of Transportation, 1 v. Sciences, 267 p. Washington Department of Transportation, 1998, Standard specifica- Scott, M. W., 1997, Named awards in the geosciences, United States tions for road, bridge, and municipal construction, 1998—Metric: and Canada, 1996: American Geological Institute, 196 p. Washington Department of Transportation, 1 v. Society for Mining, Metallurgy, and Exploration, Inc.; American Wolberg, Donald; Reinard, Patsy, 1997, Collecting the natural Geological Institute, 1998, Remediation of historical mine sites— world—Legal requirements and personal liability for collecting Technical summaries and bibliography: Society for Mining, Met- plants, animals, rocks, minerals, and fossils: Geoscience Press, allurgy, and Exploration, Inc., 118 p. Inc., 329 p. n Stark, P. L., compiler; Geitgey, R. P.; Neuendorf, K. K. E., revisers, 1997, rev. 1998, Index to geologic maps of Oregon by U.S. Geo- logical Survey topographic quadrangle name, 1883–1997: Ore- Association of American State gon Department of Geology and Mineral Industries Open-File Re- Geologists Present Results of port O-97-33, 66 p. Tsuchiya, Yoshito; Shuto, Nobuo, editors, 1995, Tsunami—Progress STATEMAP Program in prediction, disaster prevention and warning: Kluwer Academic Continued from page 2 Publishers Advances in Natural and Technological Hazards Re- search, v. 4, 336 p. The STATEMAP program is part of the National Coopera- tive Geologic Mapping Program, for which the U.S. Geologi- cal Survey is lead agency. Washington’s state geological sur- REVISED USGS REPORT RELEASED vey is now in its third year of project work funded in part by the Water Resources Investigations Report 92-4109, “Hydrol- program. (See p. 58 for information about digital products.) ogy and quality of ground water in Northern Thurston The reception was well attended. Many comments about County, Washington” has been revised and reissued. Copies the high quality of the work and interest in the program came of the original report should be destroyed. For further infor- from U.S. Geological Survey. This event certainly was a high- mation contact Brian Drost, (253) 428-3600 x2642, or Gary light in the history of the Association of American State Geolo- Turney, x2626, at the USGS district office in Tacoma. gists.

Washington Geology, vol. 26, no. 2/3, September 1998 75 Washington Geology, Vol. 26, No. 2/3, September 1998 - - - from an BULK RATE Washington State U.S. POSTAGE PAID Department of Printing compiled and edited Washington Geology Digital Index to the Geology and Mineral Resources of If you move and do not notify us, you will have to wait If you supply your +4 digit, zip extension with your always-tight budget. Help us use ourting resources well us by know let- ifceive you this have “journal”. We moved will oryour do no name an off address longer change the wish or list to take immediately. re- until the undeliverable copy comesfice back with from a the new post address, of- office and for we this service have and to re-mail pay150 your copies $1 copy. of We to the got the last about post issuewhat returned it at $1 costs each—five to times mail the copy initially. new address, it saves ourjob staff of a lot maintaining of anare time several accurate and ways makes mailing to the list contactthe easier. us; left look column There on under p. Main 2. Office in CHANGED YOUROR ADDRESS CHANGED YOUR MIND? The Division pays for Washington, 1798 through Julyby 1998, Connie J. Manson,file contains is the citations now and indexing available foritems on more and than includes CD-ROM. 32,000 both The theographies items listed and in those our non-Washington printedbrary. bibli items The held disk inWindows contains our 3.1 the li or searchWashington higher. software residents It and only) sells runs =postage $3.50. for on and (Please $3.22 handling include for + $1.00 each .28 order.) tax (for Washington Bibliography Available on CD-ROM The No more current map is available. Weyou offer may this report request free, multiple and copies. (Orders must be prepaid. Maketo check the or Department money of order Natural payable Resources.ington Taxes residents apply to only. Wash Pleasehandling of include orders $1.00 to be for sent postage by mail.) and ------(please note: Availability of federal se lands since then, although , Open File Report 98-6, by R. E. , Open File Report 98-7, by H. W. Fossils in Washington , Open File Report 98-5, by J. D. Drago These maps show the various categories of ADDRESS SERVICE REQUESTED Department of Natural Resources Division of Geology andPO Earth Box Resources 47007 Olympia, WA 98504-7007 Federal Lands for Mineraland Exploration Development In the process of moving our publicationswe to a found new storage several area, boxes of GM-30— ever, we are making$1 corner-stapled each. photocopies available for lands for mineral exploration and developmentof in Washington. the State federal land in Washingtongovernment as has they not were given in upsome 1984. trades the The have taken federal place between DNRService and on the the U.S. Olympic Forest Peninsula. Ining addition, Range the (U.S. Yakima Army) Fir has expandedland north to parcels I-90. that The are federal shown onavailable GM-30 remain maps so; as restricted the or trend un for is endangered to species withdraw or more various ecological federal and land other reasons. written in 1959, reprinted in 1983) is now out of print. How Fossils in Washington Information Circular 33, Schasse and R. L.cross Logan. sections, 16 correlation diagram). p., Price 4Washington $2.32 appendices, residents + only) 2 .18 = tax plates $2.50. (for (map, vich,D.K.Norman,C.L.Grisamer,R.L.Logan,andG.An derson. Three platesgram), (maps, 80 cross p. text sections, (with appendices.33 correlation of tax basic dia (for data). Washington Price $4.17 residents + only) =Geologic $4.50). map of the 7.5-minutekane Dartford quadrangle, County, Spo Washington lam County, Washington Derkey, W. J. Gerstel, and R. L.(map, Logan. 7 cross p., 1 sections, appendix, 1 correlation plate tax diagram). (for Price Washington residents $1.85 only) + = $2.00. .15 Geologic map of the 7.5-minute Sequim quadrangle, Clal DIVISION PUBLICATIONS New Releases Geologic map and interpretedand geologic history Alger of 7.5-minute the quadrangle,County, Bow Washington western Whatcom