SUSTAINING WESTERN BY AQUATIC FOOD WEBS MARY E. POWER, WITH SARAH J. KUPFERBERG, G. WAYNE MINSHALL, MANUEL c. MOLLES AND MICHAEL s. PARKER Introduction Citizens of the USA increasingly value the diverse natural landscapes of the American West (henceforth, "the West"), and the natural ecosystems and native species these landscapes support. Biodiversity conservation often focuses on the more conspicuous vertebrates like fishes, birds, or mammals, which are of commercial, recreational, or spiritual value to various constituencies. In order to maintain such species, the food webs that in turn sustain them must also be pre- served. All life in rivers and lakes depends on energy derived from "primary producers" (algae and higher aquatic plants that harvest sunlight and nutrients), or from terrestrial detritus (dead organic material) that washes into the aquatic habitat and is consumed by fungi, bacteria and other "detritivores." Primary produc- ers and microbial detritivores are founda- tions for food chains, in which organisms at higher trophic levels (e.g., predators) consume organisms at lower trophic lev- els (e.g., detritivores and grazers) that in turn eat plants, detritus, and associated microbes. The chain is really better thought of as a web, as feeding relation- ships among organisms are complex. All aquatic life is woven into this web. While we rightfully are concerned with the larg- er, charismatic, commercially important organisms like salmon, we also need to protect and preserve the invertebrates and microorganisms that sustain them. Here we discuss intermediate con- sumers in these food webs as critical resources, not only for their roles in sup- porting organisms of greater public inter- est and recognition, but also for other ecosystem services, their own intrinsic value, and for their potential to serve as "sentinels" indicating when and whcre environments are deteriorating. Organisms in "Healthy" Food Webs as Critical Resources Aquatic Food-web Structure and Function (A Primer) In aquatic food webs, smaller organisms are typically eaten by larg- er ones. This size structure contrasts with common patterns in ter- restrial webs, and arises because of constraints and opportunities of life in water. Aquatic plants float, so there is no need for rigid stems or trunks to reach the light. Further, small plants with high surface-to-volume ratios can more easily acquire dissolved nutri- ents. For both reasons, aquatic primary producers like algae often are tiny and have very high reproductive rates. Aquatic predators typically lack grasping appendages with which to tear prey apart before swallowing it, because such appendages are hydrodynami- cally costly. Therefore, predators are "gape-limited,'' consuming only the prey items they can fit in their mouths. Gape-limited ani- mals in aquatic webs thus tend to increase in size and longevity from lower to higher trophic levels. These features commonly lead to inverted trophic pyramids in aquatic ecosystems, where small biomasses of fast-growing plants and microbes with rapid turnover support larger biomasses of longer-lived, larger animals. The efficiency with which energy and nutrients are routed up through food webs depends largely on characteristics of the con- sumers at intermediate positions between primary producers (or detritus and microbes, the other major energy sources) and top predators such as fishes. If these consumers efficiently harvest plant or microbial production and also are vulnerable to predation themselves, then energy and nutrients pass quickly from rapidly growing plants and microbes through herbivore-detritivore con- sumers (typically invertebrates like aquatic insects), to be stored in the bodies of slower-growing, longer-lived predators (typically ver- tebrates like fishes). Linkages Among Aquatic Ecosystems and Watersheds Surface-water habitats are inextricably linked to their watersheds in both and and humid regions of the West by hydrology, chemistry, sediment, and organic-matter transport. Inputs from watersheds strongly affect aquatic food webs, and therefore the river-watershed exchanges mediated by organisms. Export in the other direction, from aquatic to terrestrial ecosystems. is particularly crucial to ter- restrial consumers in arid environments (57). Two important groups of intermediate consumers, aquatic insects and amphibians, have complex life cycles that link aquatic and terrestrial food webs. Larval aquatic insects support aquatic predators like fishes, then emerge as winged adults to feed terrestrial invertebrates, reptiles, amphibians, birds, and mammals (12,48-49,57, 107, 123, 153- 155). Amphibians reproduce and live as larvae in water, while adults typically spend much of their life on land. Ongoing degradation and loss of surface-water habitats, due primarily to human activity, threaten or damage populations of aquatic organisms throughout all 19 western states. While con- cern for native species is growing, interest in maintaining eco- logical services and economic benefits of "healthy" rivers, lakes, springs, and wetlands is already nearly universal. Here we pre- sent a scenario suggesting how food webs should function in "healthy" aquatic ecosystems. We then review selected cases to illustrate how human impacts on the landscape affect these webs by altering habitats and lives of their constituent organisms and ecosystem services these plants and animals perform. Ecological Services Provided by "Healthy" Aquatic Food Webs Ecosystem "health" is dificult to define across a region as broad and diverse as the West because of variations in biogeography, climate, and geology. One generalization relevant to aquatic ecosystems, however, is that those which are healthy have fea- tures that retain and recycle nutrients within local watersheds. Well-vegetated watersheds, for example, slow or prevent nutri- /!ox 1. The length of functionally significant food chains can be counted as the number of trophic levels ihat are altemaiely released or suppressed following the removal of a top 1 vedatoK Organisms constituting a particular trophic level can, if not suppressed by their own predators, potentially regulate (prevent outbreaks) of their prey or resources at ihe next lower trophic level. Severe environmental degradation ofien shor?ensfinctional food chains to the point where this buffering capacity is lost. A worst-case scenario has occiirred near the outskirts of Phoenix, AZ. where water wells have been shut down due to [ox-ic accuniulations of nitrates. Residents in the region are literallj drinking their own autotnobile exhaust (S. Fisher and N. Gritntn, iri 67). Food webs in the arid, eroded, mvr-grazed waterdieds collecting this well water can be argued to have no functionally I igtiificaiit trophic leselJ-i.e., ril)ariati and aquatic primary producrion is itisiiflcietit to seyuesrcr tiitrogcti inlo plant biomass. If watersheds sipport siiflcieiit plant growh, \carer wells may remain iiticoiitaiiiitiated, but eutrophication nray become a problem if inacrophjte or algal accrual is excessive. Grazers may help to regulate plant accnial, but if uncontrolled, could give rise to nuisance outbreaks of insects. Predatory itivertebrates and wid1 fish (three trophic levels) will improve the situation, arid if rhese und the gm:ersfeed (urger fish. birds, atid wildlye, recreational and aesthetic values of the ecosjsretn arc ctihaticed. Loiigerjood chains are of en positively associated \c*irh enviroriinentul qriulity and biodiversit): uiiless they result from inrroductions of alien .pries whirh can have unanricipated, adverse effects 011 native biota. ent losses from the land (50, I 16). Energy and nutrients that do reach "healthy" aquatic food webs tend to be routed up through consumers to be stored in the bodies of long-lived predators, as just described. These "top-heavy" food webs buffer watersheds by preventing pulses of nutrients that periodically wash into channels (e.g., during rainstorms) from rapidly flushing down drainage net- works, where they accumulate and can contaminate water bodies downstream (Box 1; IS). In addition to buffering ecosystems by storing nutrients, vertebrate predators are active and mobile. They therefore can resist being swept downstream, and retain nutrients locally. Some, in the course of diel, annual, or life-history migrations, translocate nutrients many kilometers upstream (e.g., migrating adult salmon; 5, 16.87) or upslope into the terrestrial watershed (e.g., bats forag- ing on river insects and roosting in caves or trees; 123). In these ways, native fish- es and other vertebrate predators feeding on aquatic production help retain and restore fertility in upper parts of water- sheds, further reducing the potential for eutrophication of downstream wells, lakes, and estuaries. Two factors appear crucial for mainte- nance of "healthy" aquatic food webs. First, rivers in general and Western rivers in particular require quasi-natural hydro- logic regimes with periodic flooding for maintenance of healthy, indigenous ecosystems (33,81,90, 121-122, 133, 151). Western river biota evolved under extreme hydrologic fluctuation, over time scales of millenia (53). Native species can typically resist or recover from scour- ing floods or dewatering droughts if watersheds contain the second crucial fac- tor: structure that provides refuge in slow water during high flow or any water dur- ing drought (42). Types of refuge vary longitudinally in river networks, among regions, and across habitat types, but include hyporheic habitats (water-filled spaces below the surface of the river bed); woody debrisjams or beaver dams and associated pools; and off-channel habi- tats. Off-channel aquatic habitats were once much more widespread. They include undercuts beneath banks stabi- lized by riparian vegetation; off-channel pools, backwaters, and interconnecting secondary channels; and marshy flood- plains, inundated by high water, which damped discharge peaks and stabilized and retained sediments. Historic and Ongoing Degradation of Habitats and Food Webs Historic Changes. Across the West, a general pattern of deterioration of sur- face-water habitats followed clearing of forests, plowing of grasslands, and intro- duction of livestock. As vegetation that retained sediments and absorbed runoff was lost, floods and flood-borne sedi- ments eroded watersheds and caused widespread gullying and downcutting of rivers. Positive feedback followed as entrenchment (downcutting) of rivers lowered water tables, sometimes several meters, further stressing riparian vegeta- tion. These conditions were exacerbated by roads, mining, agriculture, and other activities that choked rivers with unnatu- rally high sediment loads, and in some cases chemical pollutants. nated critical features of rivers that served as refuge for biota from hydro- logic variation. Marshlands (citnegas) that had moderated fluctuations, retained and recycled nuirients, and served as refuges, nurseries, and rich feeding grounds for aquatic animals, were lost to grazing, and then to desiccation as chan- nel downcutting, flow diversion, or groundwater mining for agriculture low- ered water tables. Flood flows confined in entrenched channels or behind man- made levees focused erosion on strcambeds, dccpening scour (90). Hyporheic habitats were lost as gravel beds were eroded to bedrock or choked with cxccssivc fine sediments. Refuges and pools providcd by beaver or log jams were lost as beaver and trces were unsustainably harvested (79, 129-13 I). These changes simultaneously elimi- Ongoing stresses. Historical degradation of surface-water habi- tats has left their biota even more vulnerable to present-day stresses. Ongoing practices which continue to degrade aquatic ecosystems (102) include: o flow regulation, diversion, and groundwater mining, which distort hydrologic regimes and eliminate, simplify, or frag- ment habitats; o deliberate or inadvertent introductions of alien species; o unregulated mining, agriculture, grazing, and timber harvest; o profligate agricultural irrigation, depleting and polluting sur- face waters; and o urbanization. Unsustainable practices in the industries, along with superfires resulting from fire suppression and spread of introduced plants that act as fuel, accelerate watershed erosion, causing excessive sediment loading of channels. In many cases, stresses interact synergistically (e.g., habitat degradation facilitates invasions by alien species, then alien species exacerbate habitat degradation). After rivers and watersheds have lost the vegetative and geG morphic structures that retained nutrients and sediments, damped hydrologic fluctuations, and provided cover, organisms that retained energy and nutrients and routed them through food webs to higher trophic levels cannot persist in sufficient densi- ties to maintain these services. Eliminating hydrologic fluctua- tions like scouring floods is not a solution, and in fact makes matters worse. Study of artificially regulated rivers has shown periodic floods to be necessary to maintain habitat (2,73-74,90- 91), native species (33,84), and food-web configurations that support fishes and other predators (1 18, 120- 122, 15 1). Flow Regulation and Altered Hydrologic Regimes Artificial, flow-regulating structures (dams, diversions) for agri- culture and hydropower and/or flood control have been installed throughout all large rivers and many smaller ones in the West. Only -70 km of the 2000-km-long Columbia River runs free without the hindrance of dams, which contribute to declines in the region's salmon and steelhead populations to only a few percent of their historic abundance (22, 149). Smaller streams have not escaped. Almost every creek in the Sierra Nevada of CA has been dammed (37). Most of what remains of the CA water system, termed the most massive rearrangement of nature ever attempted (61). is an elaborate network of dams, diversions, canals, and levees where water- discharge regimes are utterly unnatural. Dams drastically alter thermal, geomorphic, and hydrologic characters of rivers. Thermal impacts on invertebrates have been extensively documented (47, 143-144). Deep (hypolim- netic) release reservoirs cause abnormal winter-warm, summer- cool conditions that disrupt seasonal cues necessary for life cycles of aquatic insects. For example, they may be "fooled" by winter-warm water into emerging as adults into lethally cold winter air temperatures (143). Thermal effects attenuate down- stream, but geomorphic impacts (e.g., channel entrenchment when upstream sediment supplies are cut off) extend over much longer reaches, hundreds of kilometers downstream from high dams (1 50). Direct adverse effects on anadromous fish are well known and reviewed elsewhere (e.g., 22, 103). We focus on how altered flow regimes harm fishes indirectly, through impacts on invertebrates and food webs. ferent ways. "Hydropeaking" for electric power generation causes abnormally frequent fluctuations, changing river stage (depth) by meters as often as several times a day. Small fishes and invertebrates are stranded as water lowers suddenly and channels and side pools are drained, then flushed downstream when sudden surges occur (7,88,124). Such flows can have direct, lethal effects on invertebrates and young life-stages of fish, and also harm fishes indirectly by diminishing their inver- tebrate food supply (147). In contrast, dams built to regulate water supplies for agricul- ture or flood control reduce natural flow variation, lowering flood peaks and elevating baseflows during low-water periods. Lack of variation also can harm aquatic species, e.g., water birds who nest on sandbars that emerge from the Missouri River during low flows (22). In northern CA, eliminating the high flows that periodically scour river beds degrades food webs that support fish. In Meditemnean climates such as CA, rivers experience bed-scouring floods in winter months and low, decreasing flows during summer drought. Winter floods reset the ecological community, which recovers afterwards as plant and animal populations build back up during the process of "succession" (41). A few weeks or months after flooding, invertebrate grazers are initially dominated by mobile, unar- mored spccies (e.g., mayfly nymphs) that are good fish food. These early successional insects recover quickly after scour and are vulnerable to predators. Over summer, after months of low flow, they are gradually replaced by slower growing, more heavily armored insects (large caddisflies), or sessile larvae (e.,.., aquatic moths) that attach to the substrate and live in silk or stone cases. Both are relatively invulnerable to fish and other predators. Consequently, the later successional kinds dominate the "grazers" when flood-free periods are longer than a year, such as during prolonged drought or in channels with artificial regulation (118-1 19, 122, 151). Preliminary experimental and survey data show lower salmonid growth and densities under flood-free conditions, supporting the inference that floods bene- fit fishes indirectly through the food web (108, 151). Anothcr wcll-documented ecosystem service of flushing Dams with different purposes distort river discharges in dif- flows is the cleansing and resupply of spawning gravels (e.g., 65-66,73, 101). Natural floods flush tine sediments from stream beds, opening pores in gravels CIU- cia1 for salmon egg incubation and also as habitat and refuge for invertebrates and young life-stages of fish, including salmon. When reservoirs separate rivers from their natural sediment supply, bed materials are not flushed or renewed. Stream beds become armored or embed- ded as clean spawning gravels are choked with fine sediments or exported without replacement (66,73). In addition, flush- ing flows often suppress invading alien species. Today, many non-native species that threaten natives in western rivers come from still water or sluggishly flow- ing aquatic habitats (e.g., bullfrogs; 52). largemouth bass and other piscivorous centrarchids (98-99), and mosquitofish (80). These non-native fishes and frogs move into steeper parts of watersheds dur- ing low flow, but are displaced down- stream to a greater degree than are natives during flood (68.80). Alien (Non-native) Species Of all types of damage to aquatic species and food webs, that most difficult to reverse is the deliberate or inadvertent introduction of non-native species (75). Declines and disappearances of native frogs and toads have been documented all over the West (6, 14,52, 141), and. alien species are an important factor. Introduced predatory fishes have caused amphibian declines in many Western lakes (97, 100). For example, alpine lakes in the high Sierra Nevada were historically fishless until trout, including European brown trout, were stocked, diminishing or extirpating populations of both native invertebrates and amphibians (8, 9,32, 63-U). Alien bullfrogs, stocked for food in the late 18OOs, also threaten native frogs and other biota throughout CA and neighboring states (19,52,58,69). Bullfrog invasion also coincided with declines of aquatic reptiles such as Mexican garter snakes in AZ (128) and western pond turtle hatchlings in OR (83). Predatory bullfrogs and introduced cray- fish are thought to have been primary causes for Ash Meadows poolfish extinc- tions in 1950s (149). Because introduc- tions of non-native species co-occurred with habitat loss and hydrologic alter- ations, their direct impacts are difficult to tease out. Experimental manipulations in large enclosures have nonetheless con- firmed that bullfrogs, both as adults (62) and tadpoles (69), decrease growth and survival of native frogs. Some introductions have caused food webs to collapse with significant, adverse ecological and economic consequences (1 35). The opossum shrimp was intro- duced into Flathead Lake and River, MT, between 1968 and 1975 by biologists intending it as forage for kokanee salmon. The salmon supported recreational angling and tourism by bird watchers vis- iting to see bald eagles feeding on spent carcasses of salmon following their spawning migrations. After shrimp intro- duction, kokanee declined. The shrimp migrated to great depths by day, so were unavailable as food for the visually feed- (Box 2) Opossum shrimp were also implicated in degradation of water qualify in Luke Tahoe, CA, where its stocking along with exotic trout was followed by periodic decreases and disappearances of native zooplankton, which in turn caused food shortages for fish, and algae to increase (94-95, 125- 126). These ecological changes reduced water clarity of the lake, diminishitig the aesthetic value of the area and its economic value as a resort arid place to live. After estublishment, alien species are dificult or impossible to eradicate. 111 sotlie cases, however; restontig more tiatural hydrologic atid geomorphic corlditioris tips competitive doininatice back it1 favor of native species so they persist in inore natural parts of the habitat (S4). ing kokanee. By night, shrimp moved up to feed heavily on zooplankton, outcompeting young life stages of fish for that resource. With collapse of the salmon, eagles disappeared from their former foraging places along Flathead River and tourism based on fish and eagles withered, with severe economic impacts on local communities (I 35) (Box 2). Loss of Floodplain Habitats Early accounts by the first European explorers of the Rio Grande Valley in NM described a vastly different ecosystem than today. Historically, the river meandered freely within a 2- to 6-km wide floodplain, alternately destroying and promoting regrowth of riparian cottonwood forests. Floodplain habitats were top graphically complex, with numerous sloughs and wetlands. Today, side-to-side migration of the river is constrained by lev- ees throughout nearly the entire 200 km of middle Rio Grande Valley. These levees, along with controlled releases from upstream dams, have disconnected the river from most of its floodplain. A system of drainage ditches and agricultural devel- opment eliminated more 90% of the wetlands (25). In 1918, the valley included over 21,060 ha of wetlands, reduced to 3888 ha Belen Reach (Figure 1) Changes in wetland area, 1835-89, portions of Belen Reach, Middle Rio Grande (25). Q sea11 I:IDI.OO Iuw[Inl IO!I?( '--$- I I I INS by 1935, and to 1620 ha in 1989 (Fig. 1). Today the dominant invertebrates feeding on detritus in riparian forest along the middle Rio Grande are terrestrial isopods, introduced from Europe. Flood prevention had favored dominance of the forest-floor community by these exotics (33). Spring flooding was reintroduced to a riparian grove not flooded for more than 50 years, which significantly reduced abundance of the introduced isopods while increasing abundance of a native floodplain cricket. These native detriti- vores are also abundant in the few riparian forests that continue to experience natural, annual flooding. Thus flood control may not only favor exotic plants and fishes (84), but also exotic invertebrates at the expense of natives. Experimental floods also shed light on how control has reduced ecosystem services of both animal and fungal detritivores. After flooding, abun- dance and activity of fungal decomposers greatly increased in the riparian forest (91). Unflooded sites, in contrast, had great- ly reduced decomposition rates, and so accumulated large fuel loads to create substantial fire hazards. In 1996. one of the (Box 3). Aiwther factor increasing fire frequency and intensity is the invasion tliroughorit the intermountain West of highly flammable European and African grasses (27). Impacts of these grasses on regions like the Great Basin and Sonoran Desert, which had not historically supported continuous carpets of vegetation, are particularly severe. For example, the European annual cheatgrass has spread throughout the Great Basin Desert and increased fire frequency from once evety 60-IIO years to once every 3-5 years (148).Anotlier European annual, red brome, has spread through the Souormi Desert and fiieled fires which killed the diverse native desert vegetation, inclriding saguaro cacti (26~). More recetitl): peretinial bunchgrasses froin Afn'ca are invading from hrlexico where they have been pluirted extensively. The African grasses produce even greater fuel biorriass arid have quicker posrfire recover); so are likely tofrrfl eveti inow darnaging fires (26b). largest wildfires to date consumed 2430 ha of riparian forest in Bosque del Apache NWR (1 37). Disconnecting the Rio Grande from its floodplain has shift- ed the riparian ecosystem from flood- controlled to fire-controlled (89). Sediment Loading Fire suppression on forested uplands throughout the West has led to abnormal fuel accumulation. As a result, wildfires are larger and more intense than before, and consequently more damaging to watershed and riparian vegetation. In addition to threats to life and property, abnormally intense fires due to accumu- lated fuels can greatly increase erosion and sediment yields to streams (Box 3). Increased runoff/erosion following severe fires also may be exacerbated by postfire salvage logging operations. Following most natural wildfires, abun- dant woody debris remains and riparian vegetation regenerates from surviving rootstocks. Streadwatershed ecosys- tems thus recover rapidly, in some cases within -10 years (86). Productivity in intermediate stages of successional recovery (after 10-25 years) may exceed that prior to a fire, perhaps because of terrestrial responses to disturbance analo- gous to those allowing scouring floods to rejuvenate riverine food webs. Sediment loading to channels is not only accelerated because of superfires following fire suppression and increased fuel accumulated from introduced grass- es, but also by mining, grazing, and tim- bering practices, in particular from road construction and use in forested lands. Sedimentation from placer gold mining in the Sierra Nevada has been so exten- sive that surface flow disappeared; stream reaches that were perennial are now seasonally dry (60). Grazing impacts can have similar effects. The John Day River, OR, the longest free- flowing river remaining in the Columbia basin and one of the few salmon-produc- ing rivers in the Northwest still free of hatcheries, is severely degraded by care- less cattle grazing, logging, and irrigation diversion that consume 76% of its total discharge (22.72). (Box 4) South Fork of Salmon River is within the Idaho Batholith, an area of granitic bedrock with steep topography and highly erodible soils. Logging and road construction begun in 1950, coupled with severe storms in 1962. 1964, and 1965, increased sediment loads by 350% overpre-I950 levels. A moratorium on logging in 1965, with natural recovery and watershed rehabilitation, resulted in substantial improvement (1 13). Degradation of rivers by excessive sedimentation is widespread (Box 4). Sediment release from major clearcutting of the Taghee National Forest, ID, caused decline of a blue-ribbon trout fish- ery in Henry's fork of the Snake River. Massive sediment releases were triggered by heavy rains throughout the Pacific Northwest in 1996 as a result of road fail- ures, debris avalanches, and other ere sional events. For more than a century (43, 11 1 - 112), streams throughout the West have been strong,y influenced by open-range grazing (Box 5), and in arid areas, live- stock tends to concentrate near water. Devegetated stream banks contribute silt that fills pore spaces in gravels with fine sediments (24, 145). Such infilling degrades streambed habitat for inverte- brates (10, 11, 13. 18,23,77, 142). In addition to causing habitat loss and destabilization, fine sediments obstruct respiration, interfere with feeding, and may diminish quality or production of foods (59). A study of 60 grazed and ungrazed streams in northern Basin and Range and Snake River Plain (1 27) found grazed habitats substantially degraded, with dras- tically reduced riparian cover, raw banks, and elevated sediment, water tempera- tures, and nutrients. Grazed sites also had reduced numbers and diversity of inverte- brates that prefer cool water and coarse stony substrates. Stress-tolerant inverte- brate species dominated. The base of food webs appeared to shift from terres- trially derived leaf litter, with inconspicu- ous microbes, to algal production in the channel, with visible accrual of filamen- tous algae (85). Increased algae in streams exposed by livestock may reflect a number of factors. First, destruction of terrestrial, particularly riparian, vegeta- tion and streambank erosion accelerates nutrient and solar flux beyond levels that a pre-impact food web can absorb. Second, loss of woody debris and sedi- ment choking of stream beds degrade habitat, lowering invertebrate densities thereby diminishing their capacity to remove algae and transfer it up the food chain. Both events suggest that func- tionally significant food chains that had routed energy to fishes and terrestri- al consumers have been weakened and shortened by livestock impacts. Some Impacts From Mining Mining operations often yicld metals and other pollutants to streams that have clearly detrimental impacts on residcnt biotas. These can enter from the watcr- shed and be transported in the dissolvcd state or as scdiment (82), and may pass quickly through the syskm or remain for a variable period of time as sediment, adsorbed to various particles, or accumu- lated in plant and animal tissues (143). Metals and metaloids may be taken up directly from the water or through inges- tion by organisms at various points in food webs, then muted upward to con- centrate and sometimes accumulate at higher trophic levels (138). In fact, uptake is often so responsive to these contaminants that analyses of biological material provideg im accurate means of monitoring their presence and concentra- tions (105). Many of these elements and compounds are toxic, and as might be expected, their influences on species and populations are negative (1 17, 134) so when pollutants are reduced the commu- nities recover at variable rates (17). (Box 5) Major increases in livestock ociated with mining in fhe 1850s and 1860s, war efforts in the 1910s and 1940s, and with present human population growth (54, 152). Current livestock densities are near art all time high, ad grazing policy still represents special interest groups. Grazing is permitted on 91% of Federal lands, which conprises 48% of the total landscape in I! western states (39). Open-range grazing occurs on 69% of western ranges, covering 260 million ha with about 40% in a state of degradation (112). Salinization and Pollution From Agriculture In addition to removing up to 100% of instream flow of rivers (e.g. reaches of the Rio Grande; 22). irrigated agriculture in the arid West also causes unnatural accumulations of salts and met- alloids, such as selenium (78), in effluents. Selenium, boron, arsenic, and molybdenum, occurring naturally in soils, are con- centrated at unnatural levels in irrigation return flows (21). The famous case of Kesterson Reservoir, administered by BOR and FWS in the Central Valley of CA, illustrates the threat this poses to aquatic food webs and wildlife depending on them. This large, shallow, saline marsh consisting of 12 ponds sepa- rated by emergent vegetation was originally designed as part of a drainage system to deliver agricultural return water to the sea via San Francisco Bay. Partially because of concern over potential release of pesticides into the bay, a drainage system was never completed. In 1972, the marsh became a terminal storage-evaporation-percolation facility, draining 32 km2 of irrigated farmland. In 1983, biologists were alarmed by embryonic deaths and deformities in chicks of coots, grebes, stilts, and ducks nesting around Kesterson; 20% of nests had deformed chicks and 40% had dead embryos (106). Selenium toxicity rather than pesticide contamination was identified as the cause of deaths and deformi- ties (21,55). Selenium was bioaccumulated by organisms of the aquatic food web (Box 6), which comprised species that with- stood harsh summer conditions that included partial drying, high salinity and temperature, and low oxygen (55). Pollution from agrochemical runoff and spraying has also caused plant and animal biodiversity loss in the prairie potholes (45). This area accounts for more than 50% of North American watcrfowl production (5 I). Ncsting success appears to have (Box 6) Primary producers included a large alga, widgeongrass, and smaller algae wld other microorganisms that grew on the larger plants; the primary consumers (who fed mostly on smaller algae and other microorganisms) were mostly midge larvae and other larval flies; predatory invertebrates included dragonjly and damselfly larvae and non-native nwsquitofsh. The top predators were shorebirds, waterfowl, and nta species of blackbirds, redwings arid tricolored, the latter a candidate for listing as endangered in CA. Managers, under public pressure to act quickly, opted to remove toxic sediments to a disposal site, the most expensive remediation option, which destroyed the tricolored blackbird breeding colony (55-56). declined, however, at -0.5% per year from 1935 to 1992 (3). Several possible alternatives were examined, including loss of some wetlands to drainage (1 39, 146), alteration of hydrologic regime (70) including increased sedimentation, and eutrophication from fertilizer in agricul- tural runoff (104). None seemed the explanation. Loss of nests to mammalian predators, e.g. red foxes that had increased since settlement, was another possible reason. But nesting success had declined at similar rates where predators were managed (i.e.. trapped or fenced) and unmanaged (4). This left agrochemi- cals, e.g., insecticides already implicated in declines of small predators such as smooth green snakes and pygmy shrews (43, as important in ecosystem changes reflected in waterfowl declines. Most potholes are small (less than 0.4 ha) and dense (up to 40/km2), with only small margins of wetland vegetation left by cultivation of adjacent row crops. Because they are embedded within agri- cultural landscapes, it is almost impossible to avoid direct input of aerially sprayed ... pesticides, even under ideal conditions i5 1). Direct input comes mostly from over-spray and aerial drift. Experiments showed organophosphates (Box 7) killed mallard ducklings as well as aquatic macroinvertebrates (29); and that organophosphates per- sisted in wetland soils (30). Management recommendations therefore included farming practices which decrease needs for chemical controls: biological controls and increased buffer zones that are either uncultivated or remain unsprayed when cul- tivated (28). Groundwater Extractiodlrrigation: Lowered Water Tables/Salinization Groundwater mining (pumping that exceeds natural recharge) and diversion of surface waters have both lowered water tables throughout the arid West, threatening intermediate consumers in aquatic food webs and thus the species depending on them. In Mono Lake, CA, brine shrimp and alkali flies were eaten by thousands of waterfowl migrating between North and South America. This highly productive saline lake is a critical "pit stop" for waterfowl on their intercontinental fly-way (92, 140). From 194 1 to 198 1, diversions to Los Angeles, CA, lowered lake lcvels by 14 m. Salinity incrcased towards levels intolera- ble for both invertebratcs, Lhreatening watcrfowl food supply. Sornc waterfowl also were put at risk when lowering water lev- els thrcatencd to give tcrrcstrial prcdators access to the island (Box 7) Use of pesticides in ND (51, and ed considerably from rhe mid-1970s to the mid-1980s (herbicides by S3%, insecticide by 745%, otganophosphates (63%), carbamates (15%). and organochlorines (15%). Synthetic pyrethroids which are toxic specifically to invertebrates rather than birds, are also the most expensive options, and were used ody 7% of the time. Use of the less toxic alternatives wouMnot protect waterfowl from the indirect impacts of decreased aquatic invertebrate abundance. where they nested. Restoration of inflows to Mono Lake in 1993 resolved these threats (93). In many other aquatic ecosys- tems, however, water allocations remain unresolved. Another case involving an endangered snail illustrates the value of a species as an ecosystem sentinel. The Bruneau Hot Springs snail is endemic to a complex of thermal springs adjacent to Bruneau River, south of Mountain Home, ID. A major threat to its existence is drastic, ongoing reduction in springflows due to groundwater mining (44). It violates the law in ID to pump more from an aquifer than is replenished by natural recharge, yet local fanners maintain the water is essential to their livelihoods, and with- drawals have produced documented declines in water levels of the springs since 1983. Others, including the Federal government, are concerned that the snail will disappear with the springs, which it will. Because of political controversy over water allocation for habitat vs. farm- ing, the snail has been Federally listed, de- listed, then re-listed as endangered. Some local residents recognize, however, that saving the snail might also save the future of farming in the area. Consequences of Changes: Why We Should be Concerned Increasingly, people of the USA share the conviction that we should preserve natur- al biota and landscapes in the West. Intact aquatic ecosystems are obviously crucial to this conservation effort. In addition to ethical or aesthetic motiva- tions, there are strong economic and pub lic health consequences of landuse policy choices that affect the future of western aquatic ecosystems. Communities that can maintain "healthy" local ecosystems that support fisheries and wildlife will benefit eco- nomically from commercial or recreation- al fisheries, and increasingly from tourism. As states convert from resource extraction to service-based economies, natural ecosystems will enhance the Local and Regional "quality of life" important to choice of location by future businesses, light industries, and highly trained people. If ongoing landuse pnc- tices, like agriculture near Mountain Home, ID, are to be sustained, species like the Hot Springs snail should be con- served as sentinel species whose popula- tion trends signal when water extraction is excessive. Unsustainable timber harvest, grazing, mining or agricultural practices degrade watersheds, causing them to release sedi- ments, nutrients, and sometimes pollu- tants to rivers or other surface waters too rapidly to be assimilated. Both local and downstream ecosystems are damaged. The unlicensed Cushman Dam, in addi- tion to devastating the salmon and steel- head runs on the North Fork Skokomish River, WA, degraded health of the estu- ary and once-productive shellfish beds in the Hood Canal (22). Eutrophication of estuaries raises public health concerns, as it can lead to red tides or blooms of other toxic algae (I). Untreated sewage and toxic chemicals discharged into the lower Rio Grande lead to transboundary health problems (hepatitis, diarrheal diseases) between Mexico and [he USA. Cholera bacteria persist for long periods on or in marine phytoplankton (20,34,36), so eu- trophication of estuaries and coastal la- goons takes on even more public health significance. While cholera is not endem- ic in western USA (Box 8). it is so in the southeast, and was epidemic in temper- ate South America (Peru) in 1991 when more than 300,000 people were infected and more than 3500 died (46). In general, there is increasing evidence that human population growth, rapid glob al mixing of humans and other biota, and environmental disruption are increasing our vulnerability to infectious disease (35, 46,96). The enormous potential econom- ic and personal costs of this threat further motivate efforts to restore and maintain healthy, sustainable aquatic ecosystems. Issues and Recommendations ISSUE - Enormous benefits would come from changing landuse and devel- opment so rivers and streams could once ajain periodically inundate large portions of their natural floodplains. Pathogens, pollutants, and excessive nutrients would be filtered by floodplain soils and vegetation and kept out of water supplies. Off-river aquatic habitats evolved (35), further motivating attention to landuse practices and exposure to the pathogen. One case of the 0139 Bengal strain of cliolera, which shows alarming virulence and transmission rate, was reported in 1993 in CA, although water-supply sanitation is generally sufficient to prevent itfrom becoming endemic except at the USA-Mexican border (35). Cryptosporidiwn and Giardia are protozoan pathogetls of ittunediate concern in the USA, where a 1992 survey showed that nearly 40% of treated drinking water supplies contained one or the othex They are difficult to control because both . organism are resistant to levels of disinfectants used in drinking water; and it takes only a few organistiis to produce infection in humans (110). would again be available during high flows to nurture fish, bird, and other wildlife populations. Fertility of agricultural lands would be naturally and periodically restored. Flood waters would dissipate over large storage areas before rising to damaging stages. As this is being written, estimates of damages from the 1997 New Year's flooding in CA alone are climbing between $1-2 billion. The costs of damage are a simple function of recently expanded construction on floodplains, as well as the weakening of aging levees. Repair of levees and damaged structures will be followed by future flood damage, which will worsen as development and further attempts to regulate the rivers proceed. A practical alternative has been proposed (71): if building on floodplains is permitted, structures should be on stilts. If function requires structures to be low (as for sewage-treatment plants), they could be surrounded by ring levees. While we are literally in the wake of the 1997 flood, the time is right for State or Federal governments to pursue acquisition of flood-prone lands from willing sellers. But if the opportunity is missed this year, it will certainly arise again in the near future. o RECOMMENDATION - Apply purchases, easements, or other means to exclude or regulate certain kinds of develop rnent on naturaljloodplains, with emphasis on restoration and sustainable uses offroodplaidriver corridor ecosystems ISSUE - Water use in excess of sustainable supplies, salin- ization of soils and ground water, and pollution, primarily from agrochemicals and mining, are major problems throughout the West. K,I.Wk.s For example, more rivei(greater than 19,000 km) have been devastated by acid mine pollution than are presently protected by the National Wild and Scenic Rivers Act (22). Many of these problems could be addressed by rethinking social and economic policies and encouraging or rewarding the application of new technologies. Technological methods that could reduce human impacts on aquatic ecosystems are available and await the politi- cal climate and economic conditions that will foster their imple- mentation. They are measures that could buy us time as we cope with the more fundamental problem of how to limit human population densities on fragile westem landscapes. o RECOMMENDATION - Use of all available information and technology to reduce impacts and increase sustainabili- ty of water resources must be implemented. Examples for which follow: a) Timber companies should use recently available Regional digital elevation models (e.g., 31) to choose areas to cut or not to cut, based on slope stability, proximity to stream channels, and other factors that predict landslides or other risks. Wood should not be undervalued, as it is today in part through Federal subsidies. Value-added industries, in which local resi- dents manufacture products like furniture or musical instru- ments from the wood they harvest should be fostered. Alternative biomass sources for paper pulp should be sought from rapidly renewing, high cellulose plants (e.g., hemp or -s a green alga). b) Application of advanced, available technologies also should be required for agriculture. Retooling to use of drip or trickle micro-irrigation will reduce water needs for crop production and prevent rising salt concentrations in soils (1 15). Federal water subsidies should be phased out so that crops like cotton and rice are not grown in inappropriate arid landscapes. Several practices may be used to reduce pesticide flux to aquatic surface waters. Conservation tillage, i.e., leaving surface crop residue on the soil as opposed to conventional plow-disk-plant tillage system, would rcduce fluxes of biocides borne on sediments. For pcsticidcs originating aerially or in runoff, techniques like subsurface injection are promising. Timing and rate of chcmical application are the most important factors which can be manipulated (and regulated) to decrease the magnitude and impact of pesticide fluxes to natural ecosystems ( 1 w. c) Autoclave technologies for metal extraction from ores (132) are presently being implemented at commercially successful mines (e.g., McLaughlin Mine of Homestake Corporation in CA) and can eliminate the risk (or inevitabil- ity) of toxic seepage and acid pollution from heap leaching. ISSUE - The relatively poor knowl- edge of taxonomy and present (let alone historic) patterns of distribution and abundance in aquatic invertebrates severely limits their use as indicator or sentinel species (37-38). Aquatic insects do not appear to have suffered high rates of species extinctions as have other aquatic groups, despite the extensive destruction or modification of their habitats (1 14). This impression, however, could derive in part from igno- rance. Our knowledge of distribution, abundance, and change in invertebrate populations, particularly for insects, is limited by lack of taxonomic expertise and lack of past or present survey and inventory data (37), even in National Parks (1 36). toring is to assess whether specific restoration and mitigation projects (for example for wetlands) are functioning ecologically, in other words, if functional food webs are establishing. Monitoring of invertebrates will help determine whether newly created habitats can deliver the ecosystem services we desire and require, or are just providing aesthetic value as open space. We need more and repeated inventories both generally throughout the West and at key sites where environmental trends are monitored. We also need more inverte- brate taxonomists. Habitat requirements or trophic roles can differ among species within a genus, even when congeners are difficult to distinguish morphologically. Another reason for invertebrate moni- In some situations, however, simple abundances of three easily distinguished insect orders, mayflies, stoneflies, and caddisflies can serve as useful if coarse indicators for water quality monitoring (40, 142). For example, concentrations of arsenic, cadmium, copper, lead and total metals detected in river invertebrates cor- relate highly with the abundances of those three groups. Organized volunteers, for example school groups, could be trained to quantify easily recognized and sur- veyed taxonomic groups, and would make widespread monitoring more exten- sive and affoxtable (1 09). o RECOMMENDATION - Emphasize support for training in systematics and taxonomy at both professional and non-professional lev- els by increasing support for muse- ums, research centers, and general educational facilities. Perform inventories and maintain data bases on organisms to monitor and detect significant trends or changes in ecosystems, and to test models applied towards predicting future trajectories under various management and ecologi- cal scenarios (76). Conclusions Organisms at higher trophic levels (e.g., predators) consume organisms at lower levels (e.g., detritivores and grazers) that in turn eat plants, detritus, and microbes, creating a complex web of feeding relationships. All aquatic life is woven into this web, and perturbations created by human intcrvention disrupts it. These disruptions are reflected throughout the food web, reduc- ing its efficiency at energy retention, cycling, and transport, and ultimately breaking linkages among subsystems which results in ecosystem collapse. Western aquatic habitats have suffered severe impacts from myriad human sources. Although consider- able knowledge is available and new techniques exist to amelio- rate dangerous situations, their existence is only now being rec- ognized, acknowledged, and applied. Society must move swiftly to assure sustainability of water resources for human use, and even more swiftly if native biotas and naturalness in aquatic systems of the West are to be maintained into the future. References 1. Anderson, D.M. 1994. Red tides. Sci. Am. 271: 62-68. 2. Barinaga, M. 1996. A recipe for river recovery? Science 273: 1648-1650. 3. Beauchamp. W.D.. R.R. Koford, T.D. Nudds, R.G. Clark & D.H. Johnson. 1996a. Long-term declines in nest success of prairie duch. J. Wildl. Mgmt. 60: 247-257. 4. Beauchamp. W. D., T. D. Nudds & R. G. Clark. 1996b. Duck nest success declines with and without predator management. J. Wildl. Mgmt. 60: 258-264. 5. Bilby, R.E.. B.R. Fransen & P.A. Bisson. 1996. Incorporation of nitrogen and carbon from spawning coho salmon into the trophic system of s:nall streams: Evidence from stable isotopes. Canad. J. Fish. Aq. Sci. 53: 164-173. 6. Blaustein, A.R., D.G. Hokit & R.K. O'Hara. 1994. Pathogenic fungus contributes to amphibian losses in the Pacific Northwest. Biol. Conserv. 67: 251-254. 7. Bovee, K.D. 1985. Evaluation of the effects of hydropeaking on aquatic macroinvertebrates using PHABSIM. Pp. 23b.241, in E Olson, R. White & R. Hamre (eds.), Proceedings of a Symposium on Small Hydropower and Fisheries. Am. Fish. Soc., Bethesda. MD. 8. Bradford, D.F. 1989. Allotopic distribution of native frogs and introduced fishes in high Sierra Nevada lakes of California: Implications of the negative effect of fish introductions. Copeia 1989: 775-778. 9. Bradford, D.F.. F. Tabatabai & D.M. Graber. 1993. Isolation of remaining population of the native frog, J&m~~uscosa. by intro- duced fishes in Sequoia and Kings Canyon National parks, California. Conserv. Biol. 7 882-888. IO. Brusven, M.A. & K.V. Prather. 1974. Influence of stream sediments on distribution of macrobenthos. J. Entomol. Soc. Brit. Columb. 71: 25-32. 11. Brusven, M.A. & S.T. Rose. 1981, Influence of substrate composition and suspended sediment on insect predation by the toment sculpin, Cottus rhotheus, Can. J. Fish. Aq. Sci. 38: 1444-1448. 12. Busby, D.G. & S.G. Sealy. 1979. Feeding ecology ofa population ofnesting yellow warblers. Canad. J. Zool. 57: 1670-1681. 13. Buscemi. P.A. 1966. The importance of sedimentary organics in the distribution of benthic organisms. Pp. 79-86, in K. Cummins, C. Tryon & R. Hanman (eds.), Organism-substrate Relationships in Streams. Pymatuning Lab. Ecol., Univ. Pittsburgh, PA. 14. Carey, C. 1993. Hypothesis concerning the causes of disappearance of boreal toads from the mountains of Colorado. Conserv. Biol. 7: 355-361. 15. Carpenter, S.R. & K.L. Cottingham. Unpublished manuscript. 16. Cederholm, CJ., D.B. Houston, D.L. Cole & W.J. Scarlett. 1989. Fate of coho salmon (Oncorhvnchus kisutcWcarcasses in spawning streams. Canad. J. Fish. Aq. Sci. 46: 1347-1355. 17. Chadwick, J.W., S.P. Canton & R.L. Dent. 1986. Recovery of benthic invertebrate communities in Silver Bow Creek, Montana, following improved metal mine wastewater treatment. Wat. Air Soil Poll. 28: 427-438. 18. Chutter, EM. 1969. The effects of silt and sand on the invertebrate fauna of streams and rivers. Hydrobiologia 34: 57-76. 19. Clarkson, R.W. & J.C. Ronbaugh. 1989. Status of leopard frogs (Rana Dipiens complex: Ranidae) in Arizona and southeastern California. SW Nat. 34: 531-538. 20. Colwell. R.R. 1996. Global climate and infectious disease: The cholera pandigm. Science 274: 2025. 21. Congdon, C. 1993. Tackling nonpoiot source pollulion problems through tradeable discharge permits: The San Joaquin Valley experience. Pp. 183-200. in D. Wilcove and M. Bean (eds.). The Big Kill: Declining Biodiversity in America's Lakes and Rivers. Environ. Def. Fund., Wash., DC. 22. Conn, D. 1996. North America's hlost Endangered and Threatened Rivers of 1996. Am. Rivers, \'ash., DC. 23. Cordone, A.J. & D.W. Kelley. 1961. The influence of in organic sediment on the aquatic life of streams. CA Fish Game 47: 189- 228. 24. Corn. P.S. & R.B. Bury. 1989. Logging in western Oregon: Responses of headwater habitats and sueam amphibians. For. Ecol. Mgrnt. 29: 39-57. 25. Cnwford. C.S.. A.C. Cully, R. Leuthcuser, M.S. Sifuentes, L.H. White & J.P. Wilber. 1993. Middle Rio Gnnde Ecosystem: Bosque Biological Managcrncnt Plan. RVS, Albuquerque, NM. 26. D'Antonio, C.M. a. pcrs. comm. b. D'Antonio pcrs. comm. from A. Burqucz. 27. D'Antonio, C.M. & P.M. Vitousek. 1992. Biological invasions by exotic grasses, the grasdfire cycle, and global change. An Rev. Ecol. Systemat. 23: 63-87. 28. Delphey, P.J. & J.J. Dinsmore. 1993. Breeding bird communities of recently restored and natural prairie potholes. Wetlands 13: 200-206. 29. Dieter, C.D., W.G. Duffy & L.D. Flake. 1995a. Environmental fate of phorate and its metabolites in northern prairie wet- lands. J. Freshwat Ecol. IO 103-109. 30. Dieter C.D., L.D. Flake & W.G. Duffy. 3995b. Effects of phorate on ducklings in northern prairie wetlands. J. Wildl. Mgmt. 59: 498-505. 31. Dietrich, W.E.. R. Reiss, M.-L. Hsu & D.R. Montgomery. 1995. A process-based model for colluvial soil depth and shal- low landsliding using digital elevation data. Hydrol. Proc. 9: 383-400 32. Drost, C.A. & G.M. Fellers. 1996. Collapse of a regional frog fauna in the Yosemite Area of the California Sierra Nevada, USA. Conserv. Biol. 10: 414-425. 33. Ellis, L.M.. M.C. Molles, Jr. & C.S. Crawford. 1996. Seasonal flooding and riparian forest restoration in the middle Rio Gmde Valley. Final Rept., FWS, Albuquerque, NM. 34. Epstein, P.R. 1993. Algal blooms in the spread and persistence of cholera. BioSystems 31. 35. Epstein, P.R. 1995. Emerging diseases and ecosystem instability: New threats to public health. Am. J. Publ. Health 85: 36. Epstein, P.R.. T.E. Ford & R.R. Colwell. 1993. Marine ecosystems. The Lancet 13 Nov 1993. 37. Erman, N.A. 1996. Status of aquatic invertebrates. In Sierra Nevada Ecosystem Project: Final Report to Congress. Univ. CA Davis Cent. Wat. Wildl. Resour., Davis. 38. Erman. N.A. & C.D. Nagano. 1992. A review of California caddisflies (Trichoptera) listed as candidate species on the 1989 federal "Endangered and threatened wildlife and plants: Animal notice of review." CA Fish Game 78: 45-56. 39. Ferguson, D. & N. Ferguson. 1983. Sacred Cows at the Public Trough. Maverick Publ., Bend, OR. 40. Firehock, K. & J. West. 1995. A brief history of volunteer biological water monitoring using macroinvertebrates. J. N Am. Benthol. SOC. 14: 197-202. 41. Fisher, S.G. 1983. Succession in streams. Pp. 7-27, in J. Barns & G. Minshall (eds.), Stream Ecology: Application and Testing of General Ecological 'Iheory. Plenum Press, NY. 42. Fisher, S.G.. L.J. Gray, N.B. Grimm & D.E. Busch. 1982. Temporal succession in a desert stream ecosystem following flash flooding. Ecol. Monogr. 52: 93-1 IO. 43. Fleishner, T.L. 1994. Ecological costs of livestock grazing in western North America. Conserv. Biol. 8: 629-644. 44. FWS (Fish Wildl. Ser.). 1992. Fed. Regist. 57(1930): 457624764. Cited from: Berenbrock, C. 1993. Effects of well dis- charges on hydraulic heads and spring discharges from the geothermal aquifer system in the Bruneau area, Owyhee County, southwestern Idaho. GS Wat.-Res Invest. Rept. 93-4001:1-58. 45. Galatowitsch, S.M. & A.G. van der Valk. 1994. Restoring prairie wetlands. IA St. Univ. Press, Ames. 46. Garret, L. 1994. The Coming Plague. Newly Emerging Diseases in a World Out of Balance. Fanar, Strous & Giroux Publ., NY. 47. Gislason. J.C. 1985. Aquatic insect abundance in a regulated stream under fluctuating and stable diel flow patterns. N Am. J. Fish. Mgmt. 5: 39-46. 48. Gray, L.J. 1989. Emergence production and export of aquatic insects from a tallgm prairie system. SW Nat. 34: 31 3-3 18. 49. Gray, L.J. 1993. Response of insectivorous birds to emerging aquatic insects in riparian habitats of a tallgrass prairie stream. Am. Midl. Nat. 129: 288-300. 50. Gregory, SV. F.J. Swanson, W.A. McKee & K.W. Cummins. 1991. An ecosystem perspective of riparian zones. Bioscience 41: 540-551. 5 I. Grue, C. E., M. W. Tome, G. A. Swanson, S. M. Borthwick 8: L. R. DeWeese. 1989. Agricultural chemicals and the quali- ty of prairie pothole wetlands for adult and juvenile waterfowl-What are the concerns? In National Symposium on Protection of Wetlands from Agricultural Impacts. FWS Biol. Rept. 88(16): 55-64. 52. Hayes. M. P. & M. R. Jennings. 1986. Decline of ranid frog species in western North America: Are bullfrogs /Ram cntes- hcinnnl responsible? J. Herpetol. 20: 490-509. 53. Hirschboeck, K.K. 1988. Flood hydroclimatology. Pp. 27-49, in V. Baker, R. Kochel & P. Patton, eds. Flood Gcornorphology. John Wiley B: Sons, NY. 168- 172. 54. Holechek, J.L. 1993. Policy changes on federal rangelands: A perspective. J. Soil Wat. Conserv. 48: 166. 55, Home, A.J. 1988. A research scientist's perspective on the management of Kesterson Reservoir: A marsh contaminated with selenium-rich agricultural drain water. Lk. Reser. Mgmt. 4: 187-198. 56. Home, A.J. 1991. Selenium detoxification in wetlands by permanent flooding. 1. Effects on a macroalga, an epiphytic herbivore, and an invertebrate predator in the long-term mesccosm experiment at Keserson Reservoir, California. Wat. Air Soil Poll. 57-58: 57. Jackson, J.K. & S.G. Fisher. 1986. Secondary production, emergence, and export of aquatic insects of a Sonoran Desert stream. 58. Jennings, M.R. 1996. Status of amphibians. In Sierra Nevada Ecosystem Project: Final Report to Congress. Univ. CA Davis, Cent Wat. Wildl. Resour., Davis. 59. Johnson, R.K.. T. Wiederholm & D.V. Rosenberg. 1993. Freshwater biomonitoring using individual organisms, populations, and species assemblages of benthic macroinvertebrates. Pp. 40-158, in D. Rosenberg & V. Resh (eds.). Freshwater Biomonitodng and Benthic Macroinvertebrates. Chapman & Hall, NY. 60. Jones, R.C. & C.C. Clark. 1987. Impact of watershed urbanization on stream insect communities. Wat. Res. Bull. 23: 1017- 1055. 61. Kahrl. W.L. 1979. The California Water Atlas. Kaufmann Inc., Los Altos, CA. 62. Kiesecker, J. Personal communication. 63. Knapp, R. 1996. Non-native trout in natural lakes of the Sierra Nevada: An analysis of their distribution and impacts on native aquatic biota. In Sierra Nevada Ecosystem Project: Final Report to Congress. Univ. CA Davis, Cent Wat. Wildl. Resour., Davis. 64. Knapp, R., W. Matthews & V. Vredenburg. Unpublished data. 65. Kondolf, G.M., C.F. Cada, MJ. Sale & T. Felando. 1991. Distribution and stability of potential salmonid spawning gravels in steep boulder-bed streams of the eastern Sierra Nevada. Trans. Am. Fish. Scc. 120: 177-186. 66. Kondolf, G.M., M.J. Sale & M.G. Wolman. 1993. Modification of fluvial gravel size by spawning salmonids. Wat. Res. Res. 29: 67. Koppes, S. 1990. Delving into desert streams. AZ St. Univ. Res 5: 16-19. 68. Kupferberg, S.J. 1996. Hydrologic and geomorphic factors affecting conservation of a river-breeding frog lRana bovliiL Ecol. 69. Kupferberg, S.J. 1997. Bullfrog catesbeiana) invasion of a California River: The role of larval competition. Ecology: in press. 70. hnoo. M.J.. K. Lang, T. Waltz & G.S. Phillips. 1994. An altered amphibian assemblage: Dickinson Counly, Iowa, 70 years after Frank Blanchard's survey. Am. Midl. Nat. 131: 311-319. 71. hopold, L.B. & T. Maddock. 1954. The Hood Control Controversy. Ronald Press, NY. 72. Li, H., G.A. Lamberti, T.N. Pearsons, C.K. Tait, J. L. & J. Buckhouse. 1994. Cumulative effects of riparian disturbances along high desert trout streams of the John Day basin, Oregon. Trans. Am. Fish Soc, 123: 627-640. 73. Ligon, F.K., W.E. Dietrich & W.J. Trush. 1995. Downstream ecological effects of darns. Bioscience 45: 183-192. 74. Lind, A.J., H.H. Welsh, Jr. & R.A. Wilson. 1996. The effects of a dam on breeding habitat and survival of the foothill yellow- legged frog (Rnna bovlii) in northwestern California. Herpetol. Rev. 27: 62-67. 75. Lodge. D.M. 1993. Biological invasions: Lessons for ecology, Trends Ecol. Evol. 8: 133-137. 76. Ludwig, D., B. Walker & C.S. Hollings. 1996. Sustainability, stability. and resilience. Conserv. Ecol., in press. 77. Luedike, R.J., M.A.Brusven & EJ. W3tI.s. 1976. Benthic insect community changes in relation to instream alterations of a sediment-polluted stream. Melanderia 23: 21-39. 78. hlarshall, E. 1986. High selenium levels confirmed in six states. Science 231: 11 1. 79. Maser, C. & J.R. Sedell. 1994. From he Forest to the Sea: The Ecology of Wood in Streams. Rivers, Estuaries and Oceans. SI. Lucie Press, Delray Beach, FL. 80. hleffe, G.K. 1984. Effects of abiotic disturbance on coexistence of predator-prey fish species. Ecology 65: 1525-1534. 81. Meffe, G.K. & W.L. Minckley. 1987. Persistence and stability of fish and invertebrate assemblages in a repeatedly disturbed Sonoran Desert stream. Am. Mid. Nat. 117:177-191. 82. Miller, J.R., J. Rowland, P.J. Lechlcr, hl. Desilets & L. Hsu. 1995. Dispersal of mercury-contaminated sediments by geomorphic processes, Sixmile Canyon, Nevada. USA: Implications to site characterization and remediation of fluvial environmcnts. 43-52. Ecol~gy 67: 629-638. 2265-2274. Appl. 6: 1332-1344. 83. Milner, R. L. 1986. Status of the western pond turtle (ClemvS marmonta) in northwestern Washington. WA Dept. Game, Nongame Div.. Olympia. 84. Minckiey, W.L. & G.K. Meffe. 1987. Differential selection by flooding in stream fish communities of the arid American south- west. Pp. 93-104. in W. Matthews & D. Heins (eds.), Community and Evolutionary Ecology of North American Stream Fishes. Univ. OK Press, Norman. 85. Minshall. G.W. Unpublished data. 86. Minshall, G.W., D.A. Andrews, J.T. Brock, C.T. Robinson & D.E. Lawrence. 1990. Changes in wild trout habitat following forest fire. Pp. 11 1-1 19, in F. Richardson & R. Hamre (eds.), Wild Trout IV: Proceedings of the Symposium. U.S. Gov. Print. Off., Wash., DC. 87. Minshall. G.W.. E. Hitchcock & J.R. Barnes. 1991. Decomposition of rainbow trout (Oncorhvnchus~YkisBcarcasses in a for- est stream ecosystem inhabited only by nonanadromous fish populations. Canad. J. Fish. Aq. Sci. 48: 191-195. 88. Minshall, G.W. & P.V. Winger. 1968, The effect of reduction of streandlow on invertebrate drift. Ecology 49: 380-382. 89. Molles, M.C., Jr. Unpublished data. 90. Molles, M.C.. Jr., C.S. Crawford & L.M. Ellis. 1995. The effects of an experimental flood on litter dynamics in the middle Rio Grande riparian ecosystem. Regul. Riv. 11: 275-281. 91. hlolles, M.C.. Jr., C.S. Crawford, L.M. Ellis & J.R. Thibault. 1996. Influences of flooding on dynamics of coarse woody debris and forest floor organic matter in riparian forests. J. N Am. Benthol. SOC. 13: 154-155. 92. MLBC (Mono Lake Biol. Comm.). 1985. Cited in 149. 93. MBESC (Mono Basin Ecosys. Study Comm.). 1987. The Mono Basin ecosystem: Effects of changing lake level. NRC, Natl. Acad. Press, Wash., DC. 94. Morgan, M.D. 1980. Life history characteristics of two introduced populations of Mvsis relict& Ecology 61: 551-561. 95. Morgan, M.D., S.T. Threlkeld & C.R. Goldman. 1978. Impact of the introduction of kokanee [Oncorhvnchus sum shrimp (Mvsis relicta) on a subalpine lake. J. Fish. Res. Bd. Canad. 35: 1572-1579. 96. Morse, S. 1991. Emerging viruses: Defining the rules for viral traffic. Persp. Biol. Med. 34: 387-409. 97. Moyle, P. B. 1973. Effects of introduced bullfrogs, Copeia 1973: 18-22. 98. Moyle, P.B. 1976. Inland Fishes of California. Univ. CA Press, Berkeley. 99. Moyle. P.B., H.W. Li & B.A. Barton. 1986. The Frankenstein effect: Impact of introduced fishes on native fishes in North America. Pp. 415-426, in R. Stroud (ed.), Fish Culture in Fisheries Management. Am. Fish. SOC.. Bethesda, MD. 100. Moyle, P.B., R.M. Yohiyama & R.A. Knapp. 1996. Status of fish and fisheries. In Sierra Nevada Ecosystem Project: Final Report to Congress. Univ. CA Davis Cent. Wat. Wildl. Resour., Davis. 101. Mundie, J.H. 1974. Optimization of the salmonid nursery stream. J. Fish. Res. Bd. Canad. 31: 1827-1836. 102. Naiman, R.J., J.J. Magnuson, D.M. McKnight & J.A. Stanford. 1995. The Freshwater Imperative. Island Press, Wash., DC. 103. NAS. 1996. Upstream: Salmon and Society in the Pacific Northwest. NRC, Natl. Acad. Press., Wash., DC. IN. Neely, R.K. & J.L. Baker. 1989. Nitrogen and phosphorous dynamics and the fate of agricultural runoff. Pp. 92-131, in A. Van der Valk (ed.) Northern Prairie Wetlands. IH St. Univ. Press, Ames. 105. Nelson, S.M. & S.G. Campbell. 1995. Integrated assessment of metals contamination in a lotic system using water chemistry. transplanted bryophytes. and macroinvertebrates. J. Freshwat. Ecol. IO: 409-420. 106. Ohlcndorf, J.M., D.J. Hoffman, M.K. Saiki & T.W. Aldrich. 1987. Embryonic mortalities and abnormalitis of aquatic birds: Apparent impacts by selenium from irrigation drainwater. Sci. Tot. Environ. 48. 107. Parker, M.S. & M.E. Power. 1993. Algal-mediated differences in aquatic insect eme9ence and the effect on a terrestrial preda- tor. J. N Am. Benthol. Soc. IO: 171. 108. Parker, M.S. & M.E. Power. 1997. Unpublished data. 109. Penrose, D. & S.M. Call. 1995. Volunteer monitoring of benthic macroinvertebrates: A regulalory biologists' perspective. J. N Am. Benthol. Soc. 14: 203-209. 110. Plait. A.E. 1996. Infeciing oursclves: How environmental and social distruptions trigger discase. World Watch Pap. 129. I 11. Platts, W.S. 1978. Livestock interactions with fish and aquatic environments: Problems in evaluation. Trans. N Am. Wildl. Nat. Resour. Conf. 43: 488-503. and opos- gatesbeiana. on the native frogs of the San Joaquin Valley, California. 112. Plans, W.S. 1991. Livestock grazing. In Influences of Forest and Rangeland Management on Salmonid Fishes and Their Habitat. Am. Fish. Soc. Spec. Publ. 19. 113. Platts, W.S. & W.E Megahan. 1975. lime trends in riverbed sediment composition in salmon and steelhead spwaning areas: South Fork Salmon River, Idaho. Trans N Am. Wildl. Nat. Res. Conf. 40: 229-239. 114. Polhemus, D.A. 1993. Conservation of aquatic insects: Worldwide crisis or localized threats? Am. Zool. 33: 588-598. 115. Postel, S. 1993. Water and agriculture. Pp. 5666, in P. Bleick (ed.). Water in Crisis. Oxford Univ. Press, NY. 116. Postel, S. & S.R. Carpenter. 1997. Freshwater ecosystem services. In press, in G. Daily (ed.), Nature's Services. Island Press, Wash., DC. 117. Poulton. B.C., D.P. Monda & D.F. Woodward. 1995. Relations between benthic community structure and metals concentntion in aquatic macroinvertebrates: Clark Fork River, Montana. J. Freshwater Ecol. 10 277-293. 118. Power, M.E. 1992. Hydrologic and trophic controls of seasonal algal blooms in northern California rivers. Arch. Hydrobiol. 125: 385-410. 119. Power, M.E. 1995. Floods, food chains and ecosystem processes in rivers. Pp. 52-60, in C. Jones and J. Lawton (eds.). Linking Species and Ecosystems. Chapman & Hall, NY. 120. Power, M.E., W.E. Dietrich & J.C. Finlay. 1996. Dams and downstream aquatic biodiversity: Potential food web consequences of hydrologic and geomorphic change. Environ. Mgrnt. 20 887-895. 121. Power, M.E., G. Parker, W.E. Dietrich & A. Sun. 1995a. How does floodplain width affect floodplain river ecology? A prelimi- nary exploration using simulations. Geomorph. 13: 301-317. 122. Power, M.E., M.S. Parker & J.T. Wootton. 1995b. Disturbance and food chain length in rivers. Pp. 286-297, in G. Polis and K. Winemiller (eds.), Food Webs: Integration of Patterns and Dynamics. Chapman & Hall, NY. 123. Rainey, W.E. & E.D. Pierson. 1996. Cantara spill effects on bat populations of the Upper Sacramento River, 1991-1995. CA Fish Game Cantara Prog. Final Rept. FG2099R1, Sacramento. 124. Reiser, D.W., M.P. Ramey & T.A. Wesche. 1989. Flushing flows. Pp. 91-135, in J. Gore and G. Petts (eds.). Alternatives in Regulated River Management. CRC Press, Boca Raton, FL. 125. Richards. R., C.R. Goldman, E. Byron & C. Leviton. 1991. The mysids and lake trout of Lake Tahoe: A 25-year history of changes in the fertility. plankton, and fishery of an alpine lake. Am. Fish. Soc. Symp. 9: 30-38. 126. Richards, R.C.. C.R. Goldman, T.C. Frantz & R. Wickwire. 1975. Where have all the Danhnia gone: The decline of a major cladoceran in Lake Tahoe, California-Nevada. Verhand, Int. Verein. Limnol, 19: 835-842. 127. Robinson, C.T. & G.W. Minshall. 1995. Effects of open-range livestock grazing on stream communities. Pp. 145-154, in K. Steele (ed.), Animal Waste and the Land-water Interface. CRC Press, Boca Raton, FL. 128. Schwalbe, C.R. & P.C. Rosen. 1988. Preliminary report on the effect of bullfrogs on wetland herpetofaunas in southeastern Arizona. In Management of Amphibians, Reptiles, and Small Mammals in North America, Proceedings of the Symposium. FS Ser. Cen. Tech. Rcpt. RM-166: 166-173. 129. Sedell, J.R. & J.L. Froggatt. 1984. Importance of streamside forests to large rivers: The isolation of the Willamette River, Oregon USA, from its floodplain by snagging and streamside forest removal. Int. Ver. Theroret. Ang. Limnol. 22: 1828-1834. 130. Sedell, J.R. & K. J. Luchessa. 1982. Using the historical record as an aid to salmonid habitat enhancement. Pp. 210-223, in N. Armanrout (cd.), Acquisition and Utilization of Habitat Information in Streams. Proc. Symp. W Div. Am. Fish. Soc., Portland, OR. 131. Sedell. J.R., G.H. Reeves. F.R. Hauer, J.A. Stanford & C.P. Haakins. 1990. Role of refugia in recovery from disturbances: hlodcrn fragmcntcd and disconnected river systems. Environ. hlgrnt. I?: 71 1-721. 132. Silva, hl.A. 1988. Cyanide heap leaching in California. CA Geol. July. 133. Sparks. R.E. 1995. Nccd for ecosyslcm management of large rivers and their floodplains. BioScience 45: 168-1 82. 134. Spehar, R.L.. R.L. Anderson & J.T. Fiandt. 1978. Toxicity and bioaccumulation of cadmium and lead in aquatic macroinverte- bratcs. Environ. Poll. 15: 195-208. 135. Spenccr. C.N.. B.R. McClclland 8: J.A. Stanford. 1991. Shrimp stocking, salmon collapse, and eagle displacement. Bioscience 41: 14-21. 136. Stohlgren, T.J. 6( J.F. Quinn. 1992. An asscssment of biotic inventories in western U.S. National Parks. J. Nat. Areas 12: 135- 154. 137. Stuevcr. hl.C.. C.S. Crawford, M.C. hlolles, Jr.. C.S. White 8: E. hluldavin. 1997. Initial Assessment of the Role of Fire in the Middle Rio Grande Bosque. Intcrnat. Assoc. Wildl. Fire, Coeur d'Alene, ID. 138. Tmmernians. K.R.. E. Spijkcrman, hl. Tonkcs & H. Govcrs. 1993. Cadmium and zinc uptake by two spccies of aquatic invcr- tebrate predators from dietary and aqueous sources. Canad. J. Fish. Aq. Sci. 43: 655-662. 139. liner, R.W. 1984. Wetlands of the United States: Current status and recent trends. FWS, Natl. Wetlands Inv., Wash., DC. 140. Vale, T.R. 1980. Mono Lake, California: Saving a lake or serving a city? Environ. Conserv. 7: 190-192. 141. Vial, J.L. & L. Saylor. 1993. The status of amphibian populations: a compilation and analysis. Working document I. IUCNlSSC declining amphibian task force. World Conser. Union, Gland, Switzerland. 142. Wallace, J.B. & M.E. Gurtz. 1986. Response of && mayflies (Ephemeroptera) to catchment logging. Am. Midl. Nat. 115: 143. Ward, J.V. & J.A. Stanford. 1979. Ecological factors controlling stream zoobenthos with emphasis on thermal modifications of regulated streams. Pp. 35-55, in J. Ward & J. Stanford (eds.), The Ecology of Regulated Streams. Plenum Press, NY. 144. Warner, G. 1991. Remember the San Joaquin. In A. Lufkin (ed.). California's Salmon and Steelhead. The Struggle to Restore an Impeded Resource. Univ. CA Press, Berkeley. 145. Waters, T.F. 1995. Sediment in Streams: Sources, Biological Effects, and Control. Am. Fish. Soc. Monogr., Bethesda, MD. 146. Weinhold. C.E. & A.G. Van der Valk. 1992. The impact of duration of drainage on the seed banks of northern prairie wetlands. Canad. J. Bot. 67: 1878-1884. 147. Weisberg, S.B. & W.H. Burton. 1993. Enhancement of fish feeding and growth after an increase in minimum flow below the Conowingo Dam. N Am. J. Fish. Mgmt.13: 103-109. 148. Whisenant. S. 1980. Changing fire frequencies on Idaho's Snake River plains: Ecological and management implications. In Proceedings of the Symposium on Cheatgrass Invasion, Shrub Dieoff. and Other Aspects of Shrub Biology and Management. FS Gen. Tech. Rept. INT-276: 4-10. 149. Wilcove, D.S. & W.J. Bean. The Big Kill. Declining Biodiversity in America's Lakes and Streams. Environ. Def. Fund, Wash.. DC. 150. Williams, C.P. & M.G. Wolman. 1984. Downstream effects of dams on alluvial rivers. GS Prof. Pap. 1286 151. Woouon, J.T., M.S. Parker & M.E. Powers. 1996. The effect of dis ubance on river food webs. Science 273: 1558-1560. 152. Young. J.A. & B.A. Sparks. 1985. Cattle in the Cold Deserts. UT St. Univ. Press, Logan. 153. Zeiner, D.C., W.F. Laudenslayer, Jr., K.E. Mayer & M. White. 1988. Amphibians and Reptiles. CA Dept. Fish Game, Sacramento. 154. Zeiner. D.C., W.F. Laudenslayer, Jr.. K.E. Mayer & M. White. 1990a. Birds. CA Dept. Fish Game, Sacramento. 155. Zeiner, D.C., W.F. Laudenslayer, Jr., K.E. Mayer & M. White. 1990b. Mammals. CA Dept. Fish Game, Sacramento. 25-41.