prepared by
Mark Hill
William S. Platts
Ecosystem Sciences
This technical memorandum (Task II A4 of the workplan) describes alternative methods to manage tules and organic sediments (commonly called "muck") in the Lower Owens River Project. Existing and projected future conditions of tules and organic sediments are discussed, as well as a review of ecological values that are provided by tules in the riverine-riparian environment. Options to prepare the channel for initial flows are discussed and evaluated, and a management approach is recommended.
The lower Owens River supports a high biomass of rushes (Scirpus acutus) and cattails (Typha sp.) collectively known as tules. Tules completely dominate wetted reaches of the channel from just above Mazourka Canyon Road to the Delta. Rush and cattail dominance will continue in certain reaches, and perhaps increase, with future stream flows of 40 cfs base flow and up to 200 cfs annual riparian flow. Tules have both a positive and negative effect on water quality. Prolific tule growth and consequential die-off, could have an ongoing and deleterious effect on dissolved oxygen, BOD and sediment transport and deposition. Channel dominance by tules, as well as the influence of beaver dams and other hydrologic controls, influences stream flow and creates backwater effects.
Excessive tule biomass can be a disadvantage in the development of a flowing and functioning river, but tules can also provide many ecological benefits. Tule growth in the lower Owens River provides bank and channel stability, reduces erosion and adds shade and nutrients. High density tule stands are essential habitat for many bird and animal species and provide winter habitat for waterfowl and shorebirds. Dense vegetation stands also provide valuable refuge and early rearing habitat for both native and introduced fish species. Stands of emergent vegetation also filters sediments from stream flow which improves water quality; tules remove nutrients, organics and suspended solids, and modify low winter and high summer temperatures
The issue, therefore, is not the elimination of tules but the control and management of excessive growth and its influence on river flow and function. We must acknowledge that tule encroachment into the channel will occur in the future no matter what method is used to prepare the channel, but will be influenced by the magnitude and interaction of depth, velocity and shading, as well as competition with other vegetation types.
Deposits of sand, silt and organic sediments (muck) are confined to the wetted reach of the lower Owens River. Low volume and low velocity flows over the years have encouraged the accumulation of organic sediment and sand throughout the lower channel; deposits are several feet thick in some places, while in other areas less accumulation has occurred. Organic sediments can be a source of organic loading if high flows cause scour and mobilize bottom materials. On the other hand, organic sediments that are not mobilized can contribute to anaerobic streambed conditions.
Organic sediments can have a deleterious influence on water quality in the short term, but in the long term organic sediments that have been mobilized by stream flow are an important component of the riparian and stream bank building process. Riparian flows build and irrigate landforms, redistribute sediments, scour pools and undercut banks. Landforms like streambanks, floodplains, terraces and channels in the lower Owens are platforms upon which riparian vegetation (primarily willow and cottonwood) grow. Riparian flows will vortice sediments onto landforms, which not only builds and maintains the landform, but sets the stage for seeding and germination of riparian plants. Organic sediments that are mobilized and deposited on landforms also contribute nutrients by functioning as fertilizer. This vorticing of organic sediments is particularly important in the lower Owens River. Successful establishment of cottonwoods and willows commonly occurs first on point bars created by newly deposited material within the 2- to 10-year floodplain (McBride and Strahan 1984; Bradley and Smith 1986). High flows are necessary to create the energy to mobilize organic sediments and allow lateral deposition.
Saturated, finely textured soils, often associated with low-gradient riparian zones, can cause anaerobic conditions. Such sites can be unsuitable for the establishment of cottonwoods and willows (Kauffman et al. 1997). When these hydrologic and geomorphic conditions exist, natural plant communities can be dominated by tules and hydrophytic grasses. However, tules growing on slightly submerged landforms initially slow riparian flows and cause significant deposition of sediments. As landforms are built-up by tule-induced surface deposition, surface elevations exceed that of adjacent water surface elevations. Willows and cottonwoods can then establish and out-compete tules over time as these vegetated landforms develop. It is recognized that certain non-native, invasion plant species, such as salt cedar and Russian olive, are present in substantial numbers along some reaches of the river. These species will continue to be present when flows are introduced and may increase in some areas of the river as a result of seed distribution with stream flows.
Many studies have shown the importance of aquatic vegetation in providing food and refuge for the juveniles of a number of fish species (e.g., Savino and Stein 1982; Keast 1984; Rozas and Odum 1988; Schramm and Jirka 1989). Human activities that reduce or eliminate aquatic vegetation such as dredging, herbicide application, or mechanical removal, could have severe impacts on the survival of juvenile fishes and thus on their recruitment to adult populations (Hayse and Wissing 1996). Laboratory studies have demonstrated that juvenile bluegills (Lepomis macrochirus) are highly vulnerable to predation by largemouth bass (Micropterus salmoides) when the stand density of vegetation falls below certain levels (Savino and Stein 1982; Gotceitas and Colgan 1987). These studies also indicate that juvenile bass species discriminate among densities of vegetation and select vegetation densities that are high enough to reduce predation risk (Gotceitas and Colgan 1987).
In addition to reducing predation risk, increases in vegetation density can also decrease the rate at which juvenile bass catch invertebrate prey (Savino et al. 1992). Thus, the selection among densities of vegetation by juvenile bass could depend upon a number of factors, especially the availability of invertebrate prey and the risk of predation in each of the vegetation densities available. Because other studies have suggested that predation risk and food availability may affect the density of artificial vegetation selected by juvenile fish, Hayse and Wissing (1996) conducted experiments to measure how growth rates of age-0 bluegills and predation by largemouth bass were affected by stem density. They found predation rates were significantly lower in medium and high stem densities than in low and zero densities; high-density vegetation offered significantly greater protection than medium density vegetation stands (Hayse and Wissing 1996).
Our snorkeling surveys in the lower Owens River, and other streams in the watershed, substantiate both laboratory and field experiments in that small fish select for dense tule stands for both protection and a food source. Photo number one illustrates Owens tui chub holding close to tule stands for protection. In the event that a predator (typically largemouth bass) invades their space, Owens tui chub quickly move deeper into the tule stand where larger fish cannot follow. We have observed similar behavior for juvenile largemouth bass in beaver ponds throughout the lower Owens River.
Lower Owens River emergent wetlands provide valuable resources for a variety of resident and migrant wildlife. Aerial photo interpretation of riverine/riparian vegetation types mapped 442 acres of existing emergent vegetation (Ecosystem Sciences and White Horse Associates 1997); most of this vegetation type is situated in a relatively narrow band from 2 to 3 m wide along riverbanks (Ecosystem Sciences 1994). Emergent vegetation is also concentrated in other shallow water areas, especially in beaver ponds and other impoundments.
According to the California Department of Fish and Game’s California Wildlife
Habitat Relationships Program (CWHR), the number of wildlife species that are associated
with aquatic emergent vegetation is 140 species with a preferred relationship1, 75 with a secondarily essential relationship, and 34 with
an essential relationship (CDFG 1997).
Owens tui chub and other small fish seeking refuge in tule stands
(photo by D. Deppert)
Typical dominant plant species in the lower Owens River are cattail and bulrush. These perennial species are ubiquitous in the Owens Valley and most temperate wetlands. Both cattail and bulrush provide food values for water birds, although they are far less productive and of much lesser importance than seeds, leaves and stems of many other plants (Kadlec and Smith 1989). Decomposing vegetation indirectly provides material (dead stems and leaves) that feed detrital-based food webs, including invertebrates that are necessary to waterbirds (Drobney 1980; Drobny and Fredrickson 1985; Heitmeyer 1988) and many species of herpetofauna, terrestrial avifauna and mammals (including many species of bats). Herbivores like mule deer and beaver consume the rhizomes, stems and leaves of young cattail and bulrush (Fredrickson and Laubhan 1994; Jenkins and Busher 1979; Hall 1981).
One of the greatest values provided by cattail and bulrush is the tall robust structure that is important horizontal and vertical cover, structural habitat diversity, and micro-sites for other smaller emergent and aquatic plant species. Structure provides cover for nesting, protection from predators, habitat for broods, and attachment of nests, and protection from inclement weather (Fredrickson and Laubhan 1994).
Wintering dabblers feed on residual grain in flooded grain fields or on aquatic vegetation in seasonal marshes. Winter resting cover for mallards (Anas platyrhynchos) consists of permanent marshes that contain at least 5-15% persistent emergent or woody vegetation (U.S. Fish and Wildlife Service 1986a). Wintering Canada geese (Branta canadensis) feed on residual grain, grasses, and forbs in non-flooded grain fields and grasslands and roost during the day and night on islands or shorelines that are barren of trees or other tall vegetation (Springer and Lowe in press).
Mallard broods use wetlands that have sparse to dense emergent vegetation; wetlands devoid of either emergent vegetation or open water are usually avoided (Berg 1956; Godin and Joyner 1981; Talent et al. 1982; Rumble and Flake 1983). According to the U.S. Fish and Wildlife Service (1986b) habitat value is highest when a minimum of 25% of the wetland has emergent vegetation present (including along the shoreline). Canada geese nest in a variety of sites that include dense marshes, islands, cliffs, elevated platforms in marshes, tundra, mats of bulrush, tops of muskrat houses, tops of haystacks and abandoned heron and osprey nests in trees (Bellrose 1978; pers. obs.).
Many species of amphibians, reptiles, mammals and birds directly and indirectly benefit from cattails, bulrush and other emergent vegetation (Zeiner et al. 1988; Zeiner et al. 1990a; Zeiner et al. 1990b). The value of emergent vegetation is related to a complex of factors that include: the plant species composition and species richness; stem density or cover; size and configuration of the vegetation type; vegetation height; relative amount of open water; and surrounding vegetation types. To most species of wildlife the value of cattail and other emergent vegetation decreases in extensive and dense monotypic stands; waterbird and other wildlife species richness and abundance may decrease in these decadent conditions (Weller and Spatcher 1965; Weller and Fredrickson 1973).
Future riparian vegetation types along eight ecologically different reaches in the lower Owens River, were predicted for a streamflow scenario consisting of 40 cfs base flow and up to 200 cfs annual riparian flow. Predicted future emergent vegetation (tules) was based on: (1) results of HEC-2 hydrologic analysis performed during the 1993 controlled flow study; (2) existing landforms and vegetation types mapped from aerial photos; (3) soil types; and (4) existing vegetation and landform attributes measured along cross-channel transects (see Ecosystem Sciences 1997). Figure 1 shows the planning area and the eight reaches of the lower Owens River. The following 13 maps illustrate the predicted tule stands in each of the eight reaches (Maps 1-5, Maps 6-9, Maps 10-13). [Copies of these maps can be obtained by contacting the Inyo County Water Department.]
Concentrations of tules shown in these maps are based upon limited modeling and very conservative analysis. Thus these maps represent a worst case condition. Other environmental conditions critical to limiting tule growth were not incorporated into the vegetation prediction model. The most important environmental influence on tules (after depth and velocity) is light. Predictions do not take into account the effects of shading at intermediate and climax seral stages of willows and cottonwoods nor were we able to incorporate limits on tule growth imposed by low water transparency that is and will be typical of the lower Owens (due to high primary productivity and total suspended solid load). Nevertheless, the maps indicate that tules, under a worst case scenario, will be confined to the stream margins, oxbows, and side channels.
Tules in the lower Owens River grow on four principle landforms: the river channel; levees or streambanks; adjacent floodplains; and oxbows (old channel cuttoffs). Vegetation modeling predicts a total of about 350 acres of tules, excluding the Delta. Table 1 shows predictions for tule distribution by landform and river reach resulting from a 40 cfs base flow with periodic riparian flows of up to 200cfs.
TABLE 1. Predicted distribution of tules in the lower Owens River by reach and
landform.
(Table 1 is not available on the internet.)
Table 1 shows that the highest density of tules predicted occur in Reach 4 the island reach where, the channel is undefined and a broad wetland has formed. In general modeling allocates 55% of tules on channel landforms, 26% on levees (streambanks), 6% on floodplains, and 13% on oxbow landforms.
It must be emphasized that the model predictions for tule density and distribution is a rough approximation which does not account for the effect of shading from both riparian overstory and poor water clarity, thus model results must be taken as the worst case conditions.
Four channel preparation methods are considered:
(1) natural processes that rely on stream flow processes to create depths, velocities and shading that limit tules
(2) fire treatment to burn off existing tule and cattail stands in portions of the channel
(3) mechanical removal of obstructions and dredging to remove organic sediments deposits and deepen channel reaches
(4) treatment with herbicides or other chemicals to reduce the initial stands of tules and cattails
(These four alternatives could be used in various combinations as well).
The natural approach is to simply let nature take her course. After base and riparian flows are initiated, no active intervention is taken to prepare the channel- tules and beaver dams are left in place. However, manmade flow obstructions, such as remnant dams and unused irrigation diversions, will be removed. This approach will require time to show tangible results, perhaps three to five years. Relying upon natural forces of stream flows to limit tules by flooding, scouring and the redistribution of sediments and organic sediments in order to establish and grow woody riparian vegetation (willow, cottonwood, etc.) will require patience. In the short term the river could experience poor water quality from decomposing vegetation and suspended solids, some adverse affects on the health and size of fish populations, localized but non-destructive flooding, localized bank erosion in existing, unvegetated dry reaches, and slow development of open-water areas. The natural approach to tule control is based upon the proposition that humans cannot physically construct a river and do a better job than nature. Given adequate time, the river will build a better ecosystem without human intervention, which almost always entails significant and often irretrievable environmental impacts.
Fire intervention could temporarily open the channel by burning off the exposed stems and leaves of tules and cattails. This is an initial and one time only treatment to jump start the recovery process. The idea behind fire treatment is that natural forces of flowing water will have a more open channel to shape and form without the interference of dense vegetation, thus erosive powers within stream flows will be more efficiently utilized to limit future tule encroachment, redistribute sediments and organic sediments, scour substrate, build stream banks and create fish habitat. On the other hand, by preparing the channel by burning emergent vegetation we will cause a severe negative impact on wildlife. Birds and mammals will not have sufficient time to adjust to the sudden change in habitat, or even to escape the initial fire destruction. Risk associated with fire containment is also high. Fire will also destroy surrounding established riparian vegetation, killing existing willows and cottonwoods that are the seed source for riparian vegetation development, and fire may also give salt cedar a competitive edge over willows. Fire treatment will remove vegetative biomass that, under the natural alternative, would decrease dissolved oxygen. However, fire treatment will release nutrients that are bound in plant tissue. An adverse affect on air quality during burning would also be experienced. Costs associated with fire treatment include personnel time to set and manage fires, obtain necessary permits, and fire fighting and safety equipment. However, the single most important deleterious environmental impact from burning tules would be the irretrievable loss of seed sources critical for re-establishing riparian vegetation.
Based on recent experience, fire management may not be an effective tool. Photo number two shows the effects of a wildfire in the Lone Pine Pond area in 1996. While this fire did remove the standing crop of tules from the channel and surrounding landforms, it appears that the fire actually stimulated the regrowth of tules by burning off decadent vegetation that had occupied growing space. New tule rhizomes were quickly cloned to fill the vacant space. This wildfire also burnt most of the adjacent mature cottonwood and willow trees.
Mechanical and dredging alternatives focus upon heavy equipment to remove significant tule stands from channel reaches, as well as deepening the channel at selected locations to limit future tule encroachment. This alternative is not intended to build fish habitat (i.e., pools) and is a costly alternative that would require substantial manpower and equipment. While vegetation biomass would be removed from the system, activity associated with dredging would increase the amount of material carried in suspension by stream flows. Most importantly, extensive in-channel work could cause fish kills larger than the one in 1993. In the historically watered reaches of the lower Owens River organic sediments would be dredged; this material, along with the vegetation, must be hauled out of the watershed to a safe containment area.
It is our opinion that the deleterious effects of organic sediments have been over emphasized. Instead of removing organic sediments, the preferred objective should be to hold and incorporate organic sediments into the river system. It is true that organic sediment deposits are a source of growth media for tules, and mobilized organic sediments can deplete dissolved oxygen and release ammonia and hydrogen sulfide. However, it is also true that valuable nutrients and soil particles are bound in organic sediment deposits. Our suggestion is to allow stream flows, particularly high, and out-of-channel flows, to scour, suspend and redistribute organic sediments over a period of time. Energy associated with high stream flows creates a vorticing effect that lifts sediments and other organic particles up and onto the floodplain where it is deposited. Organic sediments are essential ingredients in stream bank building and riparian habitat development. Redistribution and deposition of organic sediments will take time, however, the benefits to the riverine ecosystem outweigh the disadvantages of the time required to redistribute the material.
Fire effect on tules at Lone Pine Ponds 1996. Top hoto shows channel conditions immediately after the fire and the bottom photo shows the tule response some weeks later (photo by M. Hill).
Tules can initially be controlled with a herbicide, or some other chemical approach, in a one time application intended to jump start recovery of the river. The drawbacks to this alternative are numerous, but the most important considerations are toxicity effects in both the acute and chronic time frames on wildlife and other organisms. Herbicides would also have residual effects. Decaying vegetation would cause severe oxygen depletion at the start up of instream flows and BOD loading could last for several years.
Table 2 summarizes primary advantages and disadvantages of each channel preparation method. The alternatives described have not been developed in any detail, but are presented with the most significant advantages and disadvantages of each. To implement any but the natural channel preparation method will require preparation of detailed procedures and protocols to protect other environmental features.
TABLE 2. Contingency Summary of Channel Preparation Methods
|
EFFECTIVE TIME FRAME |
|
|
|
|
|
|
| NATURAL | LONG | MODERATE |
|
NONE | LOW | LOW | LOW |
| FIRE TREATMENT | SHORT | MODERATE | LOW | NONE TO MODERATE | HIGH | HIGH | MODERATE |
| MECHANICAL AND/OR DREDGING | SHORT | HIGH | HIGH | MODERATE | HIGH |
|
HIGH |
| HERBICIDE AND OTHER CHEMICAL TREATMENT | MODERATE | HIGH | LOW | HIGH | HIGH | HIGH | MODERATE |
Cattails are generally restricted to relatively calm water and bulrushes commonly grow in channel locations where water velocity is relatively fast. Rhizomes are the overwintering organ which perpetuates the clone each growing season; the survival and vigor of rhizomes will determine the survival of the clone. Cattail leaves and bulrush stems provide photosynthate and a free path for diffusion of oxygen to the rhizomes that remain buried in anoxic sediments; maintenance of emergent organs are necessary to maintain the rhizome.
Maintaining sufficient current velocity and depth is a potential natural strategy to control emergent plants in the Lower Owens River. Studies performed in 1993 showed that depth and velocity control tules by: (1) prevention of encroachment of tules into existing channels; and (2) removal of tule clones once they have become established in the channel; for both bulrushes and cattails a process of stem lodging may accomplish the former, and erosion of rhizomes from the substrate may serve the latter.
Stem lodging is the breaking of stems due to drag from current that deprives the rhizome of support from the stem (i.e., oxygen supply and photosynthate) and thus limits clonal expansion and rhizome vigor. The depth/velocity relation between lodged and unlodged stems is:
|
U x D = 12.8 Where: U is velocity measured in m/s (measurements in calm, D is depth in m |
The above equation establishes a mathematical envelop that describes lodging in tule stems. Where current or depth exceed this relationship tule stems will lodge; clone expansion can be prevented if current velocities are greater than this limit. Results of measurements along tule trimlines during a peak flow event show that the current velocity necessary to move bed material may also eventually remove rhizomes.
Tules are ultimately controlled by the interaction of light, depth, velocity, as well as competition. Maximum depth limiting tule growth is a function of light penetration that permits photosynthesis by inundated portions of tule stems. By reducing photon flux, partial or complete shading of tules may greatly reduce the maximum depth where emergent plants can grow. As will be seen in the following analysis, light penetration (shading) of the Lower Owens River is a critical component for tule control in stream reaches where depth and velocity are inadequate.
We applied the depth-velocity relationship to the various flow scenarios modeled in the Lower Owens River. Analysis indicates that tule trimlines are clustered below about 50 cm/s, and at high flows in most reaches of the river both depth and velocity must approach or exceed 200 cfs to control tules.
As overhead canopy expands to shade the river light penetration into the water column becomes a significant tule control mechanism. As the river recovers to a healthy, functioning riverine ecosystem, shading by willows, cottonwoods and other trees will increase over time. A dense canopy will develop over much of the river within the first decade; this will create a very significant shading effect. The four-way interaction of depth, velocity, light and competition limiting tule growth should result in a natural tule control mechanism at flows also compatible with fish and wildlife. Strong evidence of this interaction occurs in many reaches of the lower Owens River below Keeler Bridge. A dense cottonwood/willow canopy causes limited tule growth even though the water is shallow and has very little velocity.
Based upon our research and experience in the Owens River Gorge, we anticipate that tules will occupy channel landforms when the following environmental conditions occur: (1) riparian overstory (particularly tree willows and cottonwoods) does not develop; (2) water depth is less than six feet; (3) light penetration is greater than three feet; (4) and stream velocities associated with high flows are too low to prevent rhizome cloning. All four environmental conditions must be present to encourage significant tule stand density. Consequently, the management method we recommend for tule control is the natural approach. In most reaches of the lower Owens River under the stream flow management program of 40 cfs base flow and up to 200 cfs annual riparian flows the four environmental conditions necessary for tule growth should not occur as the riverine-riparian system recovers.
We also recommend that management of organic sediments be based upon the natural consequences of stream flow. Mechanical removal of organic sediment deposits represents additional environmental impact and degradation in the wetted reaches that could retard river recovery. Redistribution of bottom sediments, including organic sediments, onto landforms is an essential stream bank and riparian vegetation building and maintenance processes. In the short term organic sediments may exacerbate poor water quality conditions, but in the long run as the ecosystem moves from a dysfunctional to a functional state, water quality will approach stasis and sediment redox potential will change as anaerobic conditions decline. In short, leaving organic sediments in place poses less risk to ecosystem recovery than mechanical removal.
At the same time, we are concerned and cautious that a fish kill not be repeated in the river when flows are reintroduced. To prevent a fish kill and to minimize stress on existing fish populations from rapidly deteriorating water quality conditions, flow introduction must be gradual and carefully monitored.
At this stage of planning we have only imperfect models and professional judgement to rely upon. The ultimate determinate of tule stand density and organic sediment influences throughout the lower Owens River is empirical measurement and observation. Our recommendation for initiating natural river recovery is not intended to be absolute. It may be necessary to intervene and mechanically remove some tule stands to create channels and prevent excessive backwater effects. Adaptive management provides us with the flexibility to respond to real time conditions and revise management methods to fit actual conditions. For example, monitoring of tule beds may indicate that flow would be enhanced in the short term by removing tules in a specific problem area to create an open channel.
However, the keys to successful tule and organic sediment management with stream flow are time and patience; it will take time for riparian vegetation to develop and patience to recognize that tules and organic sediments are critical and important components of the lower Owens River ecosystem. Tules provide extremely valuable fish and wildlife habitat, add to the overall diversity and to the ultimate stability of the ecosystem.
1. California Department of Fish and Game Wildlife Habitat Relationships habitat element definitions:
preferred -- the element is used by the species to a greater degree than what would be expected from its abundance, the element enhances the value of the habitat, but is not essential for the species presence;
secondarilarily essential -- an element must be present within the home range of the species for the species to be present unless it is compensated by the presence of another secondarily essential element that serves the same function to the species;
essential -- the element must be present within the home range of a species for the species to be present.
2. Groeneveld, D.P. 1994. Hydrodynamic control of emergent aquatic plants in the Owens River Valley, California. Report to Inyo County and LADWP, Bishop, CA.
Bellrose, F.C. 1978. Ducks, Geese and Swans of North America. Stackpole Books, Harrisburg, PA. 540 pp.
Berg, P.F. 1956. A study of waterfowl broods in eastern Montana with special reference to movements and the relationship of reservoir fencing to production. J. Wildl. Manage. 20:253-262.
Bradley, C. E. and D. G. Smith. 1986. Plains cottonwood recruitment and survival on a prairie meandering river floodplain, Milk River, southern Alberta and northern Montana. Can. J. Bot. 64:1433-1442.
California Department of Fish and Game (CDFG). 1997. California Wildlife Habitats Relationships Program (CWHR). Version 5.3 Database and Users Manual. California Department of Fish and Game, Wildlife Management Branch. Sacramento CA.
Drobney, R. D and L.H. Fredrickson. 1985. Protein acquisition: a possible proximate factor limiting clutch size in wood ducks. Wildfowl 36: 122-128.
Drobney, R. D. 1980. Reproductive bioenergetics of wood ducks. Auk 973480-490.
Ecosystem Sciences 1994. Evaluation of Flows on Wildlife Habitat Quality and Quantity for the Lower Owens River Using Habitat Suitability Index (HSI) Models.
Ecosystem Sciences and White Horse Associates 1997. Predicted Future Vegetation Types Lower Owens River.
Fredrickson L.H. and M.K. Laubhan. 1994. Managing Wetlands for Wildlife. Pages 623-647. in T.A. Bookhout ed. Research and Management Techniques for Wildlife and Habitat. The Wildlife Society. Bethesda, MD. 740pp.
Godin, P.R. and D.E. Joyner. 1981. Pond ecology and its influence on mallard use in Ontario, Canada. Wildfowl 32:28-34.
Gotceitas, V. and P. Colgan 1987. Selection between densities of artificial vegetation by young bluegills avoiding predation. TAFS 116:40-49.
Hall, E.R. 1981. THE MAMMALS OF NORTH AMERICA. John Wiley & Sons,
Hayse, J. W. and T. E. Wissing 1996. Effects of stem density of artificial vegetation on abundance and growth of age-0 bluegills and predation by largemouth bass. TAFS 125:422-433.
Heitmeyer, M. E. 1988. Protein costs of the prebasic molt of female mallards. Condor 90:263-266.
Heuner, J.D. 1984. Macroinvertebrate response to marsh management. M.S. Thesis, Utah State Univ., Logan. 85pp.
Jenkins, S.H., and P.E. Busher. 1979. CASTOR CANADENSIS. MAMMALIAN SPECIES, 120 1-8
Keast, A. 1984. The introduced aquatic macrophyte, Myriophyllum spicatum, as habitat for fish and their invertebrate prey. Canadian J. Zoology 62:1289-1303.
Kadlec, J.A. and L.M. Smith. 1989. The Great Basin marshes. Pages 451-474 in L.M. Smith, R.L. Pederson, and R.M. Kaminski, eds. Habitat Management for Migrating and Wintering Waterfowl in North America. Texas Tech University Press, Lubbock. 560 pp.
Kauffman, J. B., R. L. Beschta, N. Otting, and D. Lytjen. 1997. An ecological perspective of riparian and stream restoration in the Western United States. Fisheries 22:12-24.
McBride, J. R. and J. Strahan. 1984. Establishment and survival of woody riparian species on gravel bars of an intermittent stream. A. Midl. Nat. 112:235-245.
Rozas, L. P. and W. E. Odum 1988. Occupation of submerged aquatic vegetation by fishes: testing the roles of food and refuge. Oecologia 77:101-106.
Rumble, M.A. and L.D. Flake. 1983. Management considerations to enhance use of stock ponds by waterfowl broods. J. Range Manage. 36:691-694.
Savino, J. F. and R. A. Stein 1982. Predator-prey interaction between largemouth bass and bluegills as influenced by simulated, submerged vegetation. TAFS 111:255-266.
Savino, J. F., E. A. Marschall, and R. A. Stein. 1992. Bluegill growth as modified by plant density: an exploration of underlying mechanisms. Oceologia 89:153-160.
Schramm, H. L., Jr. and K. J. Kirka 1989. Epiphytic macroinvertebrates as a food source for bluegills in Florida lakes. TAFS 118:416-426.
Springer and Lowe. In press. Population, distribution, and ecology of migrating and wintering Aleutian Canada geese. Proc. Int. Canada Goose Symposium.
Talent, L.G., G.L. Krapu, and R.L. Jarvis. 1982. Habitat use by mallard broods in south central North Dakota. J. Wildl. Manage. 46:629-635.
U.S. Fish and Wildlife Service. 1986a. Draft Habitat Suitability Index Model: Mallard (Wintering). U.S. Fish and Wildlife Service, Division of Ecological Services, Sacramento, CA.
U.S. Fish and Wildlife Service. 1986b. Draft Habitat Suitability Index Model: Mallard (Breeding). U.S. Fish and Wildlife Service, Division of Ecological Services, Sacramento, CA.
Zeiner, D.C., W.F. Laudenslayer, Jr., and K.E. Mayer (eds). 1988. California's Wildlife: Volume I Amphibians and Reptiles. California Department of Fish and Game, Sacramento, California. 272 pp.
Zeiner, D. C., W. F. Laudenslayer, Jr., K. E. Mayer, and M. White (eds.), 1990a. California's wildlife. Vol. II. Birds. California Department of Fish and Game. Sacramento. 732 pp.
Zeiner, D. C., W. F. Laudenslayer, Jr., K. E. Mayer, and M. White (eds.), 1990b. California's wildlife. Vol. III. Mammals. California Department of Fish and Game. Sacramento. 407 pp.