Using Science to Evaluate Restoration Efforts and Ecosystem …ibdev.mcb.berkeley.edu/labs/power/publications/Golet... · 2008. 9. 15. · 368 INTRODUCTION In 1988 The Nature Conservancy

368

INTRODUCTION

In 1988 The Nature Conservancy (TNC) and itspartners (including the U.S. Fish and WildlifeService, the CA Department of Water Resources,the CA Department of Fish and Game, and theCA Department of Parks and Recreation)launched the Sacramento River Project. Broadlystated, the goal of the Sacramento River Project(hereafter “the Project”) is to restore the riparianecosystem of the Sacramento River over ~100river miles, from Red Bluff to Colusa. Toachieve this goal the Project is pursuing thefollowing strategies: 1) acquiring flood-pronelands (giving priority to those that contain and/or

border remnant riparian habitats), 2) revegetatingland with native trees, shrubs and understory,and 3) restoring natural river processes. Thegoals and strategies of the Project are consistentwith those described in the Sacramento RiverConservation Area Handbook (CA ResourcesAgency 2000). To maximize the success of theProject we are developing a multifaceted scienceprogram (Figure 1). The purpose of the programis to provide scientific understanding ofecosystem dynamics so that conservation actionscan be critically evaluated and new restorationstrategies can be developed. In this paper weprovide an overview of some of the importantprojects that are currently underway and present

Using Science to Evaluate Restoration Efforts and Ecosystem Health onthe Sacramento River Project, California

G.H. GOLET1, D.L. BROWN2, E.E. CRONE3, G.R. GEUPEL4, S.E. GRECO5, K.D. HOLL6,

D.E. JUKKOLA1, G.M. KONDOLF7, E.W. LARSEN8, F.K. LIGON9,

R.A. LUSTER1, M.P. MARCHETTI10, N. NUR4, B.K. ORR9, D.R. PETERSON1, M.E. POWER11,

W.E. RAINEY11, M.D. ROBERTS1, J.G. SILVEIRA12, S.L. SMALL4, J.C. VICK9,

D.S. WILSON10, AND D.M. WOOD10

1The Nature Conservancy, Chico, CA 959282Department of Geosciences, California State University, Chico, CA 95929

3Wildlife Biology Program, University of Montana, Missoula, MT 598124Point Reyes Bird Observatory, Stinson Beach, CA 94970

5Department of Environmental Design, University of California, Davis CA 956166Department of Environmental Studies, University of California, Santa Cruz CA 95064

7Department of Landscape Architecture and Environmental Planning and Department of

Geography, University of California, Berkeley, CA 947208Department of Geology, University of California, Davis, CA 95616

9Stillwater Sciences, Berkeley, CA 9470410Department of Biology, California State University, Chico, CA 95929

11Department of Integrative Biology, University of California, Berkeley, CA 9472012U.S. Fish and Wildlife Service, Sacramento National Wildlife Refuge Complex, Willows, CA 95988

ABSTRACT. The Nature Conservancy (TNC) and its partners are attempting to restore the riparianecosystem of the Sacramento River over ~100 river miles, from Red Bluff to Colusa. To evaluatebaseline ecosystem conditions, determine how the system is responding to current managementpractices (including restoration efforts) and better define current threats, TNC is collaborating withscientists from academic institutions, state and federal agencies, and other non-profit organizations.Our scientific investigations combine broad-based monitoring and focused research. Monitoring awide range of biological, physical and chemical parameters helps us track system dynamics, andconducting focused research allows us to resolve key uncertainties concerning ecosystem function.In this paper we introduce important studies that are currently underway and identify futureresearch needs. Only through comprehensive scientific investigation and coordinated effort will weachieve our goal of restoring ecosystem health to this complex system.

ggolet

in Faber, P.M. (ed.) 2003. California Riparian Systems: Processes and Floodplain Management, Ecology, and Restoration. 2001 Riparian Habitat and Floodplains Conference Proceedings, Riparian Habitat Joint Venture, Sacramento, CA

369

USING SCIENCE TO EVALUATE RESTORATION EFFORTS AND ECOSYSTEM HEALTH ON THE SACRAMENTO RIVER PROJECT, CALIFORNIA

the direction in which the program is heading. Inaddition to communicating some importantinitial findings, this paper illustrates the processwhereby current scientific investigations arebeing used to inform future research andmonitoring initiatives.

BACKGROUND

The Sacramento River is the largest river inCalifornia. Its watershed comprisesapproximately 67,728 square kilometers, 17% ofthe state’s total land area, but suppliesapproximately 35% of the state’s total watersupply (Buer et al. 1989). Even in its presentdegraded condition the Sacramento River is themost diverse and extensive river ecosystem inCalifornia, composed of a rich mosaic of aquatichabitats, oxbow lakes, sloughs, seasonalwetlands, riparian forests, valley oak woodlands,and grasslands. Riparian habitats in general arerich in biological diversity (Naiman et al. 1993).In the arid and semi-arid portions of the westernUnited States these habitats harbor the highestnumber of bird species (Knopf et al. 1988), andalthough several species have been extirpatedfrom the region (e.g., Least Bell’s Vireo [Vireo

bellii pusillus]) the Sacramento River continuesto provide important nesting habitat for manyspecies including the CA endangered Yellow-billedCuckoo (Coccyzus americanus), and theCA threatened Bank Swallow (Riparia riparia)and Swainson’s Hawk (Buteo swainsoni). Thebreeding distributions of many neotropicalmigratory bird species are restricted relative towhat they were historically, however, suggestingthat current land and water managementpractices threaten the long-term viability of theavifauna in this region. Once common speciesthat are now only patchily distributed as breedersin the northern Sacramento Valley include theYellow Warbler (Dendroica petechia), SongSparrow (Melospiza melodia), Yellow-breastedChat (Icteria virens) and Blue Grosbeak(Guiraca caerulea) (Nur et al. 1997a).

The Sacramento River watershed also provideshabitat for 125 species of fish (Wang 1986), andis the state’s most important resource forsalmonids. Although four distinct runs ofchinook salmon (Oncorhynchus tshawytscha)still ply these waters to reach their spawninggrounds, their overall abundance has declinedprecipitously (by more 75% since the 1950’s,Yoshiyama et al. 1998), and both the winter andspring runs have special status (federallyendangered, and federally threatenedrespectively). Other special-status fishes includethe federally threatened Central Valley steelheadtrout (Oncorhynchus mykiss) and Sacramentospittail (Pogonichthys macrolepidotus). Thusalthough the diversity of fishes in this region isstill great relative to other areas, the future ofmany species that inhabit this watershed isuncertain, and extirpations of some species (e.g.,the thick-tailed chub, Gila craisicauda) havealready been observed.

The current condition of the Sacramento Rivercan best be understood by considering theanthropogenic alterations of the landscape thathave taken place over the past 150 years. Humanactivities that have reshaped the ecosysteminclude dam, weir and levee construction, bankrevetment, installation of floodplain drainagesystems, gravel mining, hydraulic mining, urbanand agricultural encroachment, pollution,vegetation removal (including logging), and theintroduction of non-indigenous invasive species.Scott and Marquiss (1984) discuss how humanalteration of the Sacramento Valley has affectedthe river, and Patten (1998) provides a generalreview of the risk that such alterations pose toriparian ecosystem health in semi-arid regions.A detailed examination of the effects of an altered flowregime on ecosystem processes within the Project area isprovided by Kondolf et al. (2000).

identify hypothesized functional relationships

review existing data

review research and monitoring programs

with similar objectives

identify specific objectives

calibrate field techniques to the focal ecosystem

define study parameters, sampling frequencies and locations

implement research and monitoring program

analyze and archive data

identify future research and

monitoring needs

communicate findings (ecosystem response, functional relationships)

devise new management

strategies

define programmatic objectives

FIGURE 1. Conceptual model of the Sacramento River Project’sresearch and monitoring program. The overall objective of thisprogram is to improve our understanding of ecosystem dynam-ics so we may better characterize baseline conditions, definecurrent threats, evaluate ecosystem response to managementpractices and prioritize restoration strategies.

370

The Sacramento River Project is focusingconservation efforts on the Red Bluff to Colusastretch of the river because it is here that thepotential for ecosystem restoration is greatest.Prior to European settlement, this area harboreda vast amount of riparian habitat (Figure 2), andalthough degraded, this river stretch still retainssome connection with its floodplain, and thussome semblance of ecosystem function (Goreand Shields 1995). South of the Project areanatural riverine ecosystem processes are muchmore compromised as the river is confined to itsbanks by levees. In general from Red Bluff toColusa the river is sinuous and anabranching,although the northerly stretch tends to be of aslightly steeper gradient than the southerlystretch. Above Chico Landing there are short

stable reaches that are relatively narrow andstraight with low sinuosity and minor bankerosion. Interspersed between these reaches aresinuous unstable sections. Below Chico Landingthere are fewer islands, overall sinuosity isgreater, and natural levees bordering the river aremore well developed (Buer et al. 1989).

Prior to European settlement, the Project areahad an estimated pre-settlement riparian zone ofapproximately 88,376 ha. We determined this byclipping Greco’s (1999) historical riparian zonecoverage (derived from the Holmes et al. [1913]soil map) by the area outlined in Figure 2.Included in this coverage were both mixedriparian forests and valley oak woodlands.Today, only 10% of the historical riparian zone

FIGURE 2. Maps of the A) northernand B) southern areas of TheSacramento River Project. Thehistorical riparian zone wasmapped by Greco (1999), basedon interpretations of the Holmeset al. (1913) soil map. Coverageof remnant riparian habitat isfrom the GeographicalInformation Center at CaliforniaState University, Chico andcoverage of lands in conserva-tion ownership is from the U. S.Fish and Wildlife Service, theCA Department of Parks andRecreation, the CA Departmentof Water Resources, the CADepartment of Fish and Gameand The Nature Conservancy.

G.H. GOLET, ET AL.

371


remains. Furthermore only 43% of the remnanthabitat (approximately half of which is forested,Table 1) in this region is protected.

In 1987 the U.S. Congress passed legislationmandating the establishment of the of a 7,284hectare National Wildlife Refuge on theSacramento River, and as of June 2003, theRefuge consisted of 3,667 flood-prone hectaresdistributed in 22 units. In addition to the U.S.Fish and Wildlife Service, entities that acquireand/or manage conservation lands in this areainclude: two other federal agencies (U.S. Bureauof Land Management, U.S. Forest Service), threeCalifornia state agencies (CA WildlifeConservation Board, CA Department of Fish andGame, CA Department of Parks and Recreation),and several non-profits (e.g., The NatureConservancy, River Partners). The Project areaencompasses seven counties with localjurisdiction for lands and activities along the river.

Significant conservation milestones that havehelped propel conservation efforts of this Projectforward include the passage the Central ValleyImprovement Act, CA Senate Bill 1086, CAState Propositions 12, 13 and 40 Senate

Concurrent Resolution No. 62, and the CALFEDBay Delta Program (see http://www.calfedewp.org/).

CURRENT RESEARCH AND MONITORING

Current research and monitoring efforts arefocusing on gathering baseline data, refiningecosystem function models (e.g., the U. S. ArmyCorps of Engineer’s Ecosystem Function Model[Jones and Stokes 2000]) and adapting andimplementing draft monitoring plans (includingCALFED’s Terrestrial and AmphibiousMonitoring Plan [A. J. Atkinson, personalcommunication]), CALFED’s AquaticMonitoring Plan [R.L. Brown, personalcommunication], and TNC’s guide to watershedmonitoring [D. Braun, the Freshwater Initiative,personal communication]). Appropriate metricsfor tracking ecosystem dynamics are beingidentified from these studies and used topopulate our recently developed monitoringframework (TNC 2003). We are developing andcalibrating field techniques specifically for theProject area. Coordination with ongoing efforts,TNC-funded or otherwise, is playing a majorrole in the process. A sampling of the currentresearch and monitoring studies that are

TABLE 1. Amount of a) remnant habitat, and b) land in conservation ownership within the 2.5-year FEMA

flood zone, and the area depicted in Fig. 1. Habitat type mapping was done by the Geographic

Information Center at California State University, Chico through inspection of high-resolution

(1:7,200) aerial photographs taken in 1999. Areas given in hectares, and percents indicated in

parentheses.

category

within 2.5-year

flood zone

within the area

depicted in Fig. 1

a) Remnant Habitat Type

blackberry scrub 24 (<1) 54 (1)

cottonwood forest 1,449 (21) 1,683 (19)

disturbed 171 (3) 200 (2)

highly disturbed riparian 2 (<1) 43 (1)

gravel bar 1,232 (18) 1,355 (15)

giant reed (Arundo donax) 30 (<1) 48 (1)

herbland cover 861 (13) 951 (10)

marsh 104 (2) 233 (3)

mixed riparian forest 2,124 (31) 3,382 (37)

riparian scrub 760 (11) 997 (11)

valley oak 28 (<1) 170 (2)

Total 6785 9114

b) Land in Conservation Ownership

remnant habitat 2,796 (45) 3,377 (39)

revegetated habitat 621 (10) 1,168 (13)

other (farmland and roads, etc.) 2,767 (45) 4,247 (48)

Total 6184 8792

372

G.H. GOLET, ET AL.

underway is provided below. We selected a fewof these studies (the first three) for detaileddiscussion because they present some interestinginitial findings with respect to ecosystem health,and also because they provide valuable lessonswith regard to how research and monitoringprograms are structured.

TURTLES (PRINCIPLE INVESTIGATOR [PI]: D. WILSON)

Introduction and Methods

The western pond turtle (Clemmys marmorata) isa long-lived vertebrate that is dependent uponbackwater habitats and adjacent terrestrialuplands (Gibbons et al. 1990). Female freshwaterturtles may travel on land over considerabledistances (up to 500 m in some species) to findappropriate nesting sites (Wilson 1998,Goodman et al. 2000). Juvenile turtles needshallow, vegetated backwaters to avoid largepredators. Studies of this species thus present aunique window on the riparian systems itinhabits. Here we report initial findings of astudy of the western pond turtle that wasdesigned to gather and compare basicdemographic data between turtles inhabiting astretch of the Sacramento River and thoseinhabiting a relatively pristine tributary in BigChico Creek Ecological Reserve. Prior to thisstudy little to no data had been collected on thisCA species of special concern in this region.This study demonstrates that multiple benefitscan come from collecting basic monitoring data.Not only do the turtles provide us with acharacterization of ecosystem functionality, butthey also identify existing knowledge gaps,thereby identifying future research needs.

Field work was conducted from April toSeptember, 2000. Study sites included an ~ 1 kmstretch of creek within the Big Chico CreekEcological Reserve (BCCER) (39˚50'47" N,121˚42'40" W) and three backwaters (SamSlough, Jenny Lind Bend, Kaiser Slough) withina few km of one another along the west side ofthe Sacramento River (collectively referred to asthe SR site, 39˚41'58" N, 121˚57'38" W). TheSR and BCCER sites are ~ 27 km apart. Turtleswere captured at both sites using wire funneltraps baited with dead fish. At BCCER turtleswere also captured by hand while snorkeling. Allcaptured turtles were individually marked byfiling triangular notches in the marginal scutes oftheir carapace. Several morphological measurements,mass, and sex were recorded for eachturtle. We tested for a difference in the sex ratiobetween the two sites with a log liklihood ratiotest (G-test) (Fienberg 1970), and used a 2-wayANOVA with sex and site entered as independentvariables to test for differences in turtle size. Wereport means ± standard deviation.

Results and Discussion

The sex ratios of the captured turtles differedsignificantly between sites, with fewer femaleturtles being captured at the Sacramento Riversite than at BCCER, (SR: 26 ± 6 % female, n =43 turtles; BCCER: 60 ± 7 % female, n = 58turtles, G = 12.4, p < 0.001). Furthermore, turtlescaptured at SR were significantly larger (asassessed by carapace length) than turtlescaptured at BCCER (SR: 178 ± 20 mm, n = 43turtles; BCCER: 149 ± 19 mm, n = 58 turtles,F[1,97] = 41.3, p < 0.001, Figure 3). The capturemethod did not appear to affect either the size(F[1,54] = 0.25, p = 0.62) or the sex (p = 0.20,Fisher exact test) of the turtles captured.

These results suggest a potential problem forwestern pond turtles at the SR, as the sex ratiowas highly skewed toward males and there wasno evidence of juveniles recruiting into thepopulation. There are several plausibleexplanations for the observed patterns, the mostlikely perhaps being that female turtles arefacing high levels of mortality when they leavethe aquatic habitat and venture into the uplandsfor nesting. The upland areas of the SacramentoRiver have been heavily altered by agriculturaldevelopment, and contrast greatly with thepristine uplands that surround BCCER. Anotherfactor that may be affecting the SR population isthe presence of a non-indigenous turtle species,the red-eared slider (Trachemys scripta), whichis not found at BCCER. Although it is not knownto what extent these species may compete forresources, the introduced turtle grows to a larger

0

10

20

30

40

50

60

70

80

50-75 76-100 101-125 126-150 151-175 176-200 >200

Carapace length (mm)

Pe

rce

nt

of

turt

les

bccr

sac

BCCER

SR

FIGURE 3. Frequency histogram of western pond turtle carapacelength at Big Chico Creek Ecological Reserve (BCCER) andthe Sacramento River (SR). Field work was conducted fromApril - September, 2000.

373


body size and produces twice as many eggs asthe native turtle (Ernst et al. 1994).

This study demonstrates some of the benefitsthat can come from basic monitoringinvestigations, particularly when they arematched by data collection at appropriatereference sites. This research has identifiedimportant knowledge gaps and identified criticalresearch needs. Future work will determine howwestern pond turtles use floodplain habitats onboth a daily and seasonal basis and attempt toidentify causes of mortality. This informationwill help us better evaluate how current management practices are affecting this special-status species.

SONGBIRDS (PIS: S. SMALL, G. GEUPEL, AND N. NUR)


Because birds have high levels of energyexpenditure and are at the top of the food chainthey can tell us much about ecosystem function.Birds typically have fairly specific habitatrequirements; thus avian research andmonitoring programs often yield valuableinformation about the condition of differenthabitat types within an ecosystem. The Projectarea is used by many species of birds during allseasons of the year, as it provides importanthabitat for breeding, dispersal, migration, andoverwintering (Gaines 1977). Indeed, riparianareas are considered to be the most criticalhabitats for landbirds in all of California(Manley and Davidson 1993, DeSante andGeorge 1994).

Systematic studies of landbirds in the Projectarea were initiated by the Point Reyes BirdObservatory (PRBO) in 1993. These studieshave yielded a wealth of biological informationregarding the ecological condition of bothrestoration sites and remnant habitats.Furthermore, these studies provide a strikingexample of the increased understanding ofecosystem dynamics that can come fromgathering data on parameters additional tospecies distribution and abundance.

We used multiple monitoring methods to assessthe status of bird populations along theSacramento River. This is one of the longestrunning bird population studies in California’sCentral Valley, complementing other avianmonitoring projects on the Cosumnes and SanJoaquin Rivers (also being conducted by PRBOscientists). Because these studies use nationallystandardized monitoring methods (BBIRD,Ralph et al. 1995) results can readily becompared across study sites and bioregions.

To determine whether bird use of sites along theriver varied among habitat types, and to assesswhether bird diversity increased as revegatationsites matured, point count surveys wereconducted three times at each site during thebreeding season, following the protocol of Ralphet al. (1995). Species diversity was calculated foreach point count station based on a set of 56known riparian breeders from the SacramentoRiver region (see Appendix 3 in Small et al.2000). Species diversity provides a measure ofthe number of species detected weighted by thenumber of individuals of each species observed(Krebs 1994).

Although studies that only measure numbers ofbirds (abundance, diversity, density, etc.) arevaluable, they may be misleading (Van Horne1983). Long-term demographic data arenecessary to evaluate whether populations areself-sustaining over time. To gain a betterunderstanding of the viability of breeding birdpopulations along the Sacramento River, weestimated nest success of three common opencupnesting species, the Black-headed Grosbeak(Pheucticus melanocephalus), the LazuliBunting (Passerina amoena), and the SpottedTowhee (Pipilo maculatus). These species wereselected for study because they are numerous,and because they breed in both remnant foresthabitats and in areas that have been revegetatedwith native plantings. These species differsomewhat in the habitats they select for nesting.Black-headed Grosbeaks place their nests in themid-canopy shrub layer, Lazuli Buntings nestlow in herbaceous and shrubby vegetation, andSpotted Towhees nest on or very near theground, typically concealing their nests withinground litter or herbaceous and shrubby vegetation.

We conducted analyses of nest survivorshipusing nest data collected from1993 -1999. Nestsuccess was estimated using the Mayfieldmethod (Mayfield 1975), a technique thatcalculates the probability of nest failure (“nestmortality”) on a daily basis by dividing thenumber of failed nests by the total number ofdays nests were observed active. Daily nestsurvival equals one minus daily nest mortality.Standard errors associated with this value werecalculated following Johnson (1979). Wecalculate “total nest survivorship”, theprobability of a nest surviving the entire nestingcycle (laying, incubation, and nestling periods),by raising the daily nest survivorship value to thepower of the total number of days in the nest cycle.


Riparian bird diversity increased significantlyover time as revegetation sites matured (p <

374

G.H. GOLET, ET AL.

0.0001, r2 = 0.26; Figure 4). Furthermore, atrevegetation sites with native plantings greaterthan five years old, bird diversity approachedwhat was observed in remnant riparian areasalong the river (Small et al. 2000). Thesefindings suggest that planting restoration siteswith native trees and shrubs is beneficial toriparian songbirds and that basic monitoring ofthe distribution and abundance of species can beinformative in assessing ecosystem response tomanagement actions.

However, additional data from the bird studiesdo not paint as optimistic a picture for thesongbird community along the SacramentoRiver. In riparian forests and revegetation sitescombined (no difference was detected betweenthese habitat types), total estimated nestsurvivorship was extremely poor for bothunderstory nesting Lazuli Buntings (6%), andSpotted Towhees (11%, Figure 5). Withsource/sink population modeling techniques Nuret al. (1997b) demonstrated that buntingreproductive success would have to at least tripleover observed levels for the Sacramento Riverpopulation to be self-sustaining. Althoughsimilar detailed analyses have not yet been donefor the spotted towhee it is likely that present

levels of nests survivorship are also well belowwhat is needed for the population to be self-sustaining.On the San Joaquin River, anotherunderstory-nesting species, the Song Sparrow,has also experienced extremely low nestsurvivorship (Ballard et al. 1999), suggestingthat there may be widespread problems forground nesting species in the Great Valley.

The primary cause of nest failure was predationfor the Spotted Towhee and the Black-headedGrosbeak. The Lazuli Bunting also suffered highpredation rates, although nest parasitism byBrown-headed Cowbirds (Molothrus ater) wasequally responsible for nest failure (Small et al.2000). While none of these species are classifiedas threatened or endangered, it is likely that thesepopulations are only being sustained byemigration from other “source” populations.Efforts to model population dynamics ofsongbirds breeding on the Sacramento River,using local productivity and annual survival data,are now underway (see Small et al., in review).

Aside from the intrinsic value of the informationthat these avian studies provide, they are usefulin what they reveal about how we prioritize ourdata gathering efforts when monitoring

Rip

ari

an

Bir

d D

ivers

ity

Age of Restoration (years)

1 2 3 4 5 6 7 8

0

2

4

6

8

10p < 0.0001

r2 = 0.26

��

��

��

��

��

��

��

��

��0.00

0.10

0.20

0.30

0.40

Black-headed

GrosbeakLazuli Bunting Spotted Towhee

�� Revegetation

��

Riparian

Forest�� All Habitats

n=21 n=26 n=47

n=10n=23 n=33

n=18

n=29

n=47

To

tal

nest

su

rviv

ors

hip

(Mayfi

eld

esti

mate

s)

FIGURE 4. Overall riparian bird diversity (Shannon-Wiener index)and age (in years) on four TNC Sacramento River restorationsites: Phelan Island, Rio Vista, Flynn, and Ryan. Species diver-sity was measured using a transformation of the standardShannon-Wiener function (H, defined on p.704 in Krebs1994). We used the transformed index (N1), equal to 2 H,because it yields a value that can readily be compared tospecies richness. Data points represent mean annual diversityof individual point count stations. Data are jittered to allowindividual points to be seen. Least squares line of best fit isshown. Data have been standardized to eliminate year effect.This was done by fitting a model which simultaneouslyaccounted for calendar year effects and restoration age, andthen subtracting the estimated year effect for each year ofobservation. Based on point count data from 1994-1999.

FIGURE 5. Mayfield estimates of total nestsurvivorship, by habitat type, of threeopen-cup nesting species on theSacramento River, 1993-1999.

375


populations. The bird studies clearly demonstratethe importance of assessing reproductiveperformance of populations where possible, andfurther they show that simple characterizationsof distribution and abundance, when consideredby themselves, may provide misleadinginformation regarding the health of localpopulations. Like the turtle studies, theseinvestigations also demonstrate that monitoringcan be an effective way to identify knowledgegaps and prioritize future research. Future avianstudies will investigate causes of nest predation,and attempt to identify factors (e.g. flow regime,landscape effects) that affect reproductivesuccess and predation rates.

DIRECTING FLOWS (PIS: E. LARSEN AND S. GRECO)


Empirical modeling of hydraulic factors thatcontrol meander migration can improve ourability to evaluate long-term effects of floodplainmanagement on ecosystem dynamics. Here we

demonstrate how the mechanics of flow andsediment transport in a curved river channel canbe used to simulate channel migration patternspredicted to result from alternative managementscenarios. The results of these simulations haveimplications for riparian forest regeneration, asmeander migration is a fundamental drivingforce in riparian vegetation succession (Scott etal. 1997, Shafroth et al. 1998, Greco 1999). Inaddition to predicting channel movement, themodel generates estimates of the amount offloodplain bottomland that is reworked, which isindicative of the amount of substrate area that issuitable for riparian vegetation colonization.Regeneration of early successional vegetationcommunities is an important process inmaintaining the habitat heterogeneity requiredfor healthy riparian systems (Naiman, et al. 1993).

California’s Department of Water Resourcesdevised a number of alternative bankstabilization scenarios for river miles 216 - 226of the Sacramento River, near Woodson Bridge

FIGURE 6. Simulations of channel meander following different management scenarios: A) simulates river behavior if freefrom anthropogenic constraints (riprap removed); B) simulates river behavior if existing riprap is maintained; C) simulates river behavior if current riprap is maintained and additional riprap is added near the bridge; D) simulatesriver behavior if the channel is redirected from its present location to Kopta Slough (this has been proposed as a method of reducing erosion in the vicinity of the bridge and the state recreation area).

376

G.H. GOLET, ET AL.

State Recreation Area. These scenarios weredefined in part by the extent of channelmigration projected to take place over the nexthalf-century (CA Department of WaterResources 1998). Planning scenarios aredetailed in the legend of Figure 6, but includeallowing unrestrained meandering to occur byremoving existing rock revetment (riprap),adding additional riprap, and redirecting the flowof the river through an abandoned river channel(backwater slough).

To evaluate the effects of channel constraints onchannel migration, we applied a meandermigration model developed by Johannesson andParker (1989). This model assumes that the localbank erosion rate is proportional to a localvelocity factor times a bank erosion coefficient.These two components of the model simulate theerosive force of flowing water (represented bythe velocity component) and the resisting forceof the banks (represented by the erosioncoefficient). The model requires an estimation ofa characteristic discharge that mimics theintegrated effect of the variable flow regime. Therationale is the same as that used in traditionalgeomorphic analyses where channel form andprocesses are scaled by the “bankfull” or“dominant” discharge (Wolman and Miller1960). Accordingly, the model does not try tosimulate the effects of particular flow events, butproduces estimates of long-term rates of erosionor channel migration. The velocity field iscalculated analytically and the bank erodabilityis represented by an empirically estimatedcoefficient (Larsen 1995).


The 16-km reach of river channel was found torework between 8.5 and 48.5 hectares of land in50 years (Table 2), depending on themanagement scenario selected. Clear differencesin channel meander patterns are predicted to

result under varying channel stabilizationscenarios. The channel sinuosity decreases withtime in the cases where the channel isconstrained (Table 2). Decreasing sinuositymeans that the channel straightens and thereforeincreases in longitudinal slope. Importantly, thesimulations suggest that the riprap currently inplace upstream from the Woodson Bridge StateRecreation Area causes more erosion of therecreation area than if no riprap were presentupstream. The model suggests that relocating thechannel to the west, and allowing subsequentunconstrained migration, could relieve theerosion pressure in the recreation area. Thisunconstrained migration also appears to rework asubstantial area of land (26.5 hectares) during 50years of projected migration, thus conferringsignificant ecosystem benefits. This case studydemonstrates how sophisticated modeling toolsare being applied to evaluate the effects ofalternative river management options. Additionaldetails of this study are provided in Larsen andGreco (2002).

ADDITIONAL PROJECTS UNDERWAY

Groundwater, Nutrient Cycling, and Soil Development

PIs: D. Brown and D. Wood. Rivers interactwith surrounding terrestrial systems through acomplex array of physical, chemical, andbiological processes. The types and magnitudesof these interactions vary both spatially andtemporally, often spanning several orders ofmagnitude (e.g., millimeters to kilometers, andminutes to centuries). Selected interactionsbetween the Sacramento River and its floodplainare being investigated at the Rio Vista (formerly“River Vista”) restoration unit, located on theeast bank of the river south of Woodson Bridge.For example, we are studying the evolution ofsoil nutrient (carbon and nitrogen) cycles,nitrogen mineralization rates, and plant litter andsoil organic carbon dynamics. To monitor these

TABLE 2. Amount of land projected as being reworked and channel sinuosity according to four

different management scenarios. Values were predicted from simulations of a meander

migration model.

simulation

management action

area

reworked

(ha)

Sinuosity

pre/postmigration

A No riprap in the study reach 48.5 1.4/1.6

B Current riprap maintained 12.1 1.4/1.3

C Riprap extended from park bend to bridge 8.5 1.4/1.3

D River redirected through Kopta Slough 26.5 1.1/1.1

377


parameters we placed sample grids in threeriparian vegetation sites: Rio Vista revegetationsite VII (planted in 1999); revegetation site II(planted in 1994); and an adjacent riparian forestremnant (to provide a reference condition).

We are also investigating the role local ripariansystems play in mediating floodplain solutetransport to the Sacramento River. Fullyfunctioning riparian ecosystems have repeatedlybeen shown to improve groundwater and streamquality by removing undesirable constituentssuch as nutrients and pesticides (Lowrance et al.1985). To study these processes at revegetationsites and in a remnant riparian forest ninegroundwater-monitoring wells were installedalong three transects spanning the sites describedabove. The wells extend from nearby agricultural fields to the edge of the Sacramento River. Quarterly groundwater samples are being analyzed for nitrates, andwater levels are being monitored on a weekly to monthlybasis, depending on the season. Reconnaissancemonitoring for organophosphate pesticides isalso planned. Preliminary results of these studiesare detailed in Brown and Wood (2002).

Hydraulic/Geomorphic Evolution of Oxbow Lakes

PIs: M. Kondolf, and I. Morken (student).Oxbow lakes and other floodplain water bodiescontribute enormous biodiversity to lowlandriver ecosystems (Bornette et al. 1998, Ward andStanford 1995). Their unusual ecological valuestems in part from their evolution over time fromfully aquatic to terrestrial ecosystems as they fillin with nutrient-rich silt and sand during floods.Along the Sacramento River, most of thesefeatures have been filled, dyked, or otherwisedisconnected from the riverine ecosystem.However, on the reach between Red Bluff andColusa, a number of oxbow lakes remain. We arecollaborating with colleagues from theUniversity of Lyon, France to document thephysical and ecological evolution of a set of sixoxbow lakes that cut off from the mainstem ofthe Sacramento River in the twentieth century.We are mapping historical planform fromsequential maps and aerial photographs;measuring sedimentation rates from cores; anddeveloping stage-discharge relations from longtermgauging records and observations of stageover a range of flows. Data derived from theseinitiatives will be used to reconstruct inundationhistories to test oxbow evolution modelsdeveloped in France. Preliminary results fromthis study are dtailed in Morken and Kondolf (2003).

Identifying a “Naturalized” Flow Regime

PI: M. Roberts. Riparian vegetation providescritical terrestrial and aquatic habitat for wildlife,

improves water quality, and yields aesthetic andrecreational values (Patten 1998). Cottonwoods(Populus spp.) are a keystone pioneer species ofriparian habitats, but land-use practices and riverregulation have caused widespread reduction intheir distribution and regeneration (Rood andMahoney 1990, Braatne et al. 1996). Within theProject area some of the point bars have feweryoung cottonwood trees than would be expectedon an unregulated river (S. Rood, personal observation).

To understand how the flow regime may bebetter managed to support the Sacramento Riverriparian ecosystem, we are studying thegeomorphic and hydrologic factors that governFremont cottonwood (Populus fremontii)recruitment. By identifying the conditions thatcottonwoods need to successfully recruit, wehope to provide information that land and watermanagement agencies can use to revitalize theriparian forests of this important ecosystem.

More specifically, our study aims to calibratescientifically rigorous cottonwood recruitmentmodels that have been developed in otherwatersheds (e.g., Mahoney and Rood 1998).Toward this end we have collecteddendrochronological, hydrologic and topographicsurvey data, and have conducted a fieldreconnaissance of recent cottonwood recruitmentsites in the Project area. We used the Indicatorsof Hydrologic Alteration (IHA) software(Richter et al. 1996) to contrast current andhistorical hydrographs and to identify factors thatmay limit recruitment. Our preliminary results(detailed in Roberts et al. 2002) indicate thatcottonwoods recruit at higher elevations on thepoint bars of the Sacramento River than has beenobserved at other rivers where similar studieshave been conducted. This information caninform flow regulation decisions that attempt tomaximize both human and ecosystem benefits.Similar definitions of flow requirements areneeded for other species.

Riparian Landscape Change Analysis and Geospatial Modeling

PI: S. Greco. We are developing a spatialmodeling tool for assessing the impacts of rivermanagement actions (e.g., modifications of flowregime, bank stabilization, levee configuration,etc.) on wildlife habitats over time spans ofdecades to centuries. The empirically-basedmodel seeks to link wildlife habitats to landscapefeatures predicted to arise under different river-managementscenarios (see Greco et al. 2002). Wildlife habitats are being defined for a set of riparian-obligate indicator species based upon habitat suitability indices (HSI). To build and calibrate the model we are mapping vegetation from aerial photographs, analyzing historical

378

G.H. GOLET, ET AL.

flow records, and developing spatially-explicitHSI models.

As a first step in this process, baseline land coverconditions are being mapped from 1997 aerialcolor photographs (1:12,000) using astereoscope. Estimates are made of vegetationheight and canopy cover according a modifiedversion of the CA Wildlife Habitat RelationshipSystem (Mayer and Laudenslayer 1988). We arecombining vegetation data with 1997topographic and bathymetric data (collected bythe U.S. Army Corps of Engineers) to defineinitial conditions for landscape state-and-transitionmodeling being conducted with ageographic information system (GIS) followingtechniques outlined by Plant et al. (1999).Historical photographs are similarly be analyzedto define how the vegetation communitiesrespond to patterns of channel meander andfloodplain inundation (see Greco 1999).

We are also developing a multi-mediavisualization tool to communicate anddisseminate to a wide audience the inherentlydynamic nature the riparian landscape of theSacramento River. This tool will effectivelydemonstrate the dependence of riparianecosystem function on landscape change.Animations of both historical and forecastedlandscape change will be produced.

Pattern and Process in Riparian Vegetation Succession

PI: D. Wood. Riparian vegetation along theSacramento River has been classified intocommunity types but the transition rates betweencommunity types during vegetation successionhave received little attention. Understanding therate and frequency of transition from onecommunity type to another is key tounderstanding the biotic complexity andmanagement of riparian ecosystems. Forexample, the rate at which a mixed ripariancommunity becomes a valley oak community isnot known, yet valley oak woodland is a valuedcommunity type in this ecosystem. Managersneed to know how often, or even whether newcommunities are being created as a result ofnatural succession.

We are focusing on three transitions: (1) betweengravel or point bars and cottonwood-willow; (2)between mature cottonwood-willow forest andmixed riparian forest, and (3) between mixedriparian forest and valley oak woodland. Tostudy cottonwood-willow establishment, a seriesof permanently marked belt transects have beenestablished running perpendicular from the waterto mature forest on active gravel bars.Vegetation, substrate, and elevation are being

recorded in a continuous series of quadrats.Together with hydrological data, this study willestablish links between substrate type,hydrology, and survival rates of cottonwoodsand willows. To study the latter two transitionrates, a series of plots have been established indifferent-aged stands of all three communitytypes (cottonwood-willow, mixed riparian, andvalley oak woodland). Stand age is beingdetermined using a combination of aerial photosand tree ages. In each plot we are determiningthe density and dominance of all woody species,and estimating the percent cover of herbaceousspecies. Succession trajectories of different-agedstands (the chronosequence) will be analyzedgraphically using standard ordination techniques.Preliminary results from these studies aredetailed in Wood (2003a, 2003b).

Understory Vegetation Communities

PIs: K. Holl and E. Crone. Most riparian habitatrestoration efforts in the Project area thus farhave focused on planting and monitoring woodyspecies. We are surveying the understoryvegetation communities in sites planted withwoody riparian species at least five yearspreviously and in remnant riparian forests todetermine success in restoring understory plantcommunities. We are combining field data withGIS analyses of surrounding habitat types to testwhether understory community composition ismore strongly influenced by site-specific factors(e.g., depth to water table, soil texture, previousland use) or landscape-level factors (e.g.,distance to river channel, distance to remnantriparian forest). Identifying factors influencingthe distribution and abundance of plant specieswithin riparian understory areas will help usbetter understand (and model) the dynamics ofspecies that depend upon these habitats (e.g.,songbirds, see the Biocomplexity Project below).Preliminary results from this study are detailedin Holl and Crone (in review).

Factors Affecting Planted Species

PIs: D. Wood, R. Luster. We are monitoringshort- and long-term plant community dynamicsat horticultural restoration sites. To date TNChas planted over 1,168 hectares with native trees,shrubs and more recently, herbacecous species.Alpert et al. (1999) reported on factors affectingplanted species at a subset of these sites duringthe maintenance phase (the first three yearsfollowing planting, when irrigation and weedcontrol are practiced), and more recently Griggsand Golet (2002) provided an assessment of thesurvival and growth of valley oaks (Quercuslobata) following the cessation of maintenanceactivities. We are furthering these efforts by

379


collecting data at long-term plots (n = 106) thatwere established during the first year of planting.Data collected at these sites will be analyzed todetermine the relative importance of physicaland biological factors in dictating 1) the survivaland growth of planted species; 2) the structuralcomplexity of the vegetation; 3) what speciescolonize the sites (in collaboration with K. Holland E. Crone, see above); and 4) the successionaltrajectories of the different community types.Also, we have recently initiated a project with H.Cushman and J. Coleman (student) at SonomaState University to identify appropriate restoration techniques and post-restoration monitoring protocols for riparian-associated grasslands.

Bank Swallow Population and Ecology Studies

PIs: E. Crone and K. Moffatt (student). InCalifornia, bank swallow (Riparia riparia)populations have been extirpated fromapproximately half of their historic range, andcurrently over 70% of the remaining populationis found in the Sacramento Valley (Schlorff1997). Bank Swallows nest colonially in rivercut banks, and require active channel meanderingto create habitat suitable for nesting. Along theSacramento River bank swallows have beenadversely affected by widespread bankstabilization projects. In addition, because theyforage for insects over lands adjacent to the river,their population dynamics may be affected byland use practices (e.g., pesticide use) whichvary widely on the floodplain. To betterunderstand the relative importance ofhydrological and land-use patterns for bankswallow population viability, we are collaborating with R. Schlorff and B. Garrison to analyze a long-runningdataset of colony size and location.

Salmonid Use of the Floodplain

PIs: M. Marchetti and M. Limm (student).Sacramento River floodplains provide beneficialrearing habitat for multiple species of native fish(e.g., chinook salmon Sacramento splittail)(Sommer et al. 2001). Physical factorsinfluencing the quality of habitat for these fishesinclude water depth, flow rate, area, duration ofinundation, temperature, and the amount ofvegetation/cover. Riparian restoration efforts inthe floodplain along the Sacramento River havebeen successful in replacing agricultural fieldswith native vegetation. Yet during flood events,the aquatic ecosystem in the heavily vegetatedrestoration areas likely experiences loweramounts of solar energy input and thereforelower primary productivity. With a decrease inprimary production, one would expect decreasedenergy transfer to higher trophic-level organisms(including fishes). To determine whether

floodplain habitat type affects juvenile salmongrowth rates we are comparingmacroinvertebrate drift density, juvenile chinooksalmon diet and growth rates (through otolithmicrostructure examination techniques) atvegetated restoration areas, agricultural fields,backwater habitats, and the main river channel.This work may help us better understand to howfloodplain management decisions affect salmonand their prey. Preliminary results of this studyare presented in Limm and Marchetti (2003).

Sacramento River National Wildlife Refuge Wildlife Surveys

PI: J. Silveira. Table 3 presents a partial list ofspecies that are surveyed in conjunction with theUSFWS in the Project area. Several of thesesurveys are cooperative, with Refuge personnelassisting PIs from other organizations. Datacollected in these surveys provide importantbaseline data and help us assess how wildlifeoccurrence patterns change through time and inresponse to management actions. For example,Bank Swallow survey data suggest a favorableresponse to a levee removal project implementedat river mile 233 in late fall 1999. Following theremoval of the levee the river further eroded andenlarged an existing cutbank. The subsequentspring, a nearby swallow colony expanded intothis newly eroded area, forming the largestcolony on the river that year (2,770 burrows).The year previous (1999) there were only 930burrows at this site. These results support thenotion that bank swallow population declines inthis region are at least in part attributable todecades of habitat loss that began in 1960 whenthe Sacramento Bank Protection Project wasauthorized by the U.S. Congress (Schlorff 1997).

A LOOK TO THE FUTURE: LONG-TERM RESEARCH

AND MONITORING

A long-term research and monitoring program isbeing instituted to evaluate baseline ecosystemconditions, determine how the system isresponding to current management practices(including restoration efforts), better definecurrent threats, and prioritize restorationstrategies (for details see TNC [2003]). Resultsof the current research and monitoring projectsare guiding the program’s development.However, prioritizing future efforts remainschallenging. Because funding is limited it isimperative that our perceptions of relativeimportance of scientific investigations areaccurate, as these are what dictate the extent towhich individual elements of the program aredeveloped. Fortunately, a number of projectscurrently underway are focusing on this verydifficult question of how to prioritize future

380

G.H. GOLET, ET AL.

research and monitoring efforts in the Projectarea. These include the Biocomplexity andBiological Indicators Projects, each of which isintroduced below.

BIOCOMPLEXITY PROJECT

PIs: E. Crone, K. Holl, M. Kondolf, and N. Nur.The vast majority of restoration efforts,including those for Sacramento River riparianforests, are based on the assumption that, ifappropriate physical conditions are created andwoody plants are successfully established at asite, that site will be colonized by nativeherbaceous plants and animals. Alternatively,establishment of native plant and animal speciesmay be limited by biological processes thatoperate at larger spatial scales than individualsites, such as dispersal limitation of desirednative species and movement of undesirablespecies (such as weeds, pathogens, andpredators) across the landscape. To designefficient restoration strategies, we must

understand the relative importance of physicaland biological processes in determining longtermcommunity dynamics. This requiresworking across traditional disciplinaryboundaries to link physical and biologicalprocesses at site and landscape scales.

In the initial phase of our research, we arecombining data from ongoing mapping andmonitoring programs (described earlier) toformulate simple empirical models linkinghydrology, vegetation, and bird communities inSacramento River riparian habitats. Currentmonitoring efforts are not sufficient toparameterize predictive mechanistic models, butmay allow us to quantify how certain we areabout the relative importance of possiblemanagement actions. Using empirical modelsand traditional sensitivity analysis, we will testwhich suites of processes (restoring floodregimes, restoring channel meandering,establishing appropriate woody plant species,establishment of connectedness among riparian

TABLE 3. Taxa surveyed at the Sacramento River National Wildlife Refuge.

taxa special-status designation

Plants

California hibiscus (Hibiscus lasiocarpus) CNPS1 List 2

Ferris' milk-vetch (Astragalus tener var. ferrisiae) CNPS List 1B

fox sedge (Carex vulpinoidea) CNPS List 2

square-stemmed spike-rush (Eleocharis quadrangulata) CNPS List 2

Birds

Bank Swallow (Riparia riparia) CA Threatened

Black-crowned Night Heron (Nycticorax nycticorax) None

Double-creasted Cormorant (Phalacrocorax auritus) CA Species of Special Concern

Great Blue Heron (Ardea herodias) None

Great Egret (Ardea alba) None

Greater Sandhill Crane (Grus canadensis) CA Threatened Species

Osprey (Pandion haliaetus) CA Species of Special Concern

Snowy Egret (Egretta thula) None

Tricolored Blackbird (Agelaius tricolor) Federal Species of Special Concern

White-faced Ibis (Plegadis chihi) Federal Species of Special Concern

Yellow-billed Cuckoo (Coccyzus americanus) CA Threatened

Reptiles

giant garter snake (Thamnophis couchi gigas) Federally Threatened

Vernal pool crustaceans

vernal pool fairy shrimp (Branchinecta lynchi) Federally Threatened

vernal pool tadpole shrimp (Lepidurus parckardi) Federally Endangered

California linderiella (Linderiella occidentalis) CA Special Animal2

1California Native Plant Society

2World Conservation Union Red List threatened species

381


sites, or human uses of floodplain habitats) mostaffect plants and songbirds, based on our currentknowledge of ecosystem dynamics. We will alsouse an uncertainty analysis (that takes intoaccount both sensitivity and estimation error,Schultz and Crone 1998) to ask which areas offuture research would most increase our abilityto identify correct management actions for restoration.

Data compilation to date has identified broadareas that are not well monitored under existingprograms, but are key steps linking restoration ofhydrological processes, plant communities, andsongbird communities. These include: (1) stage-dischargerelationships at restored (planted with woody species) sites, (2) dynamics of herbaceous plants at restored sites, and (3) effects of human land use on insect and mammalcommunities. We have also identified a set ofempirical relationships that link what is knownabout hydrology, geomorphology, plantcommunities, and songbird population dynamics.Ongoing modeling will identify which of these,or other areas, are most important to study todirect future restoration efforts. In the future, wehope to develop an increasingly mechanistic,predictive understanding of riparian forestecosystem dynamics using an iterative process ofmodeling, manipulative experiments,monitoring, and adaptive management.Preliminary results from the BiocomplexityProject are detailed in Crone and Holl (2002).

BIOLOGICAL INDICATORS PROJECT

PIs: F. Ligon, B. Orr, M. Power, W. Rainey andJ. Vick. Operational definitions and measurableindicators of ecosystem health are needed toprioritize restoration efforts, and steer systemstoward desired endpoints. The restoration goalfor an ecosystem is not a single state, but a rangeof ecosystem states and processes that occur overtime in an ecosystem that supports native speciesand responds to dynamic environmental regimesrepresentative of natural conditions that appliedbefore systems were massively altered byhumans. Increasing our understanding of thelinkages between physical structure andprocesses and biotic responses in Central Valleyriver systems is critical to improving our abilityto predict the effects of various restoration orrehabilitation actions on biological populationsof interest (as discussed above for theBiocomplexity Project) and in developingeffective indicators to monitor and test theresults of implementing such actions (the focusof this study).

Our underlying conceptual model is shown inFigure 7. In this model, natural watershed inputs(e.g., water, sediment, nutrients) andanthropogenic alterations to these inputs

influence important physical processes (e.g.,sediment transport, channel migration, streamheating). These processes determine the physicaland chemical attributes (e.g., hydroperiod, bedsubstrate composition, nutrient availability) thataffect habitat structure and connectivity in riverriparian-floodplain systems. Species abundanceand distribution and trophic structure are directlyaffected by these habitat attributes. For instance,the species composition and age-class-structureof riparian vegetation are affected by inundationregimes and sediment deposition patterns.Attempts to restore these systems will generateprocesses and feedback loops that will haveshort-term (days to months) and long-term (yearsto decades) effects on aquatic, riparian, andupland terrestrial ecosystems. For example, inthe short-term, breaching levees would alterspatial patterns of sediment deposition andtexture and the depth, timing, and duration ofinundation. These changes would affect survival,recruitment, and hence spatial and taxonomicstructure of riparian vegetation and aquaticmacrophytes, and the microbial films associatedwith their submerged surfaces. Changes inlarger, rooted vegetation would also generatelonger-term feedback in spatial patterns ofsediment deposition and texture, both of whichwould reconfigure habitat structure and alter thefood production base for invertebrates, fish,amphibians, reptiles, mammals, and birds.

Watershed Inputs

• water • sediment • nutrients

• energy • large woody debris • chemical pollutants

Fluvial Geomorphic Processes

• sediment transport/deposition/scour • channel migration and bank erosion • floodplain construction and inundation• surface and groundwater interactions

Geomorphic Attributes

• channel morphology (size, slope, shape, bed and bank composition) • floodplain morphology • water turbidity and temperature

Habitat Structure, Complexity, and Connectivity

• instream aquatic habitat • shaded riparian aquatic habitat • riparian woodlands • seasonally inundated floodplain wetlands

Biotic Responses (Aquatic, Riparian, and Terrestrial Plants and Animals)

• abundance and distribution of native and exotic species • community composition and structure • food web structure

Human Land Use and Flow Regulation

Natural Disturbance

FIGURE 7. A simplified conceptual model of the physical and ecological linkages used in developing biotic response indices of river ecosystem health.

382

G.H. GOLET, ET AL.

Determining whether management is movingsystems toward more desirable, healthier states isa research question, as emphasized above for theBiocomplexity Project. In particular, there is aneed to develop indicators of ecosystemprocesses, such as nutrient cycling and food webinteractions, that can complement morecommonly used indicators based on ecologicalpatterns (e.g., patterns of distribution andabundance of individual species or assemblagesof species) (Bunn et al. 1999, TNC 2003). Tomeet this need, we are conducting a pilot studyto examine, at a range of sites along theSacramento River, several potential indicatorsthat integrate over various spatial and temporal scales.

A small-scale indicator, but one that might bemeaningful for understanding energy flowthrough food webs, is the accrual of organisms atvarious trophic levels on substrate particles inriver channels. Different disturbance,sedimentation, light, nutrient, temperature andother physical and chemical regimes have beenshown to favor different taxa of primaryproducers and decomposers at the base of thefood web. These basal species will havedifferential vulnerabilities and nutritional valuesfor primary consumers (grazers and detritus-feedingorganisms), which in turn will supportdifferent levels of production of predators higherin the food web. By monitoring the succession oflower trophic levels on experimental substratesover periods of weeks and months we hope toprovide some indication of how well conditionsin different reaches of the river are supportingthe food base of desirable fishes and other predators.

As a group, bats may also be useful as ecologicalindicators for river-riparian systems becausetheir response to habitat alteration is integratedover a broader spatial scale than many otherspecies currently being studied. Bats account fora substantial fraction of the native mammaldiversity on the Sacramento Valley floor, and allare insectivorous with the foraging of manyspecies being concentrated near rivers, streamsand riparian vegetation. Foraging styles andmovements differ among species, but range froma fraction of a kilometer to several tens ofkilometers per night. On the alluvial SacramentoValley floor most natural cavity roosts are inlarge defective trees in mature riparian forestremnants, and thus the persistence (orrestoration) of local bat populations is intimatelytied to riparian forest dynamics. Existingresearch and historic site records indicate that thewestern red bat (Lasiurus blossevillii) reliesheavily upon cottonwood and sycamore riparianforests, and recent findings suggest thatnaturalistic flood regimes may benefit thisspecies (Pierson, Rainey, and Corben, unpublished data).

In recognition of this, the Western Bat WorkingGroup recently passed a resolution supportingfurther research, inventory, conservation, andmaintenance of existing stands, and restorationand reestablishment of historic cottonwood andsycamore ecosystems across western NorthAmerica (Pierson, pers. comm., 2002). Tofurther our knowledge of how bats interface withthe Sacramento River ecosystem we areconducting reconnaissance-level acousticsurveys during summer and fall. With the aid ofpassive acoustic detectors we are collectingbaseline information on bat diversity andforaging activity at a broad range of habitatsincluding: main channel, tributary channel,oxbow lake, mature cottonwood riparian forest,horticultural restoration sites, orchards, rowcrops and grasslands. We also intend to dofollow-up surveys at selected sites in winter andspring to test for seasonal differences in bat activity.

Over larger (watershed) scales, stable nitrogenand carbon isotope analyses of aquatic organismsand terrestrial insectivores, such as bats,swallows, and spiders show promise as potentialindicators of river-to-watershed fluxes ofnutrients and contaminants (e.g., nutrients andcontaminants exported from watersheds to riverscan be incorporated back into terrestrial foodwebs via aquatic insect emergence) (Cabana andRasmussen 1996, Cabana et al. 2000). Emerginginsects from contaminated aquatic areas candeliver contaminants to terrestrial insectivores,which may be concentrated in riparian habitats.Stable isotopic signatures of nutrients orcontaminants of aquatic versus terestrial origincan then be followed through the food webbecause the uptake of stable isotopes acrosstrophic levels occurs at a known rate offractionation (Vander Zanden and Rasmussen1999). We are surveying 15N:14N and 13C:12Cratios in widely distributed organisms at differenttrophic levels to construct a picture of isotopeenrichment at selected sites within the Projectarea that can be correlated with present andhistorical land use. These data will form aninvaluable baseline reference for future studiesof environmental change, and permitcomparisons of watershed impacts to be drawnbetween the Sacramento River and otherwatersheds where similar data are beingcollected (e.g., the South Fork Eel and CosumnesRivers). Initial results from the BiologicalIndicators Project are detailed in StillwaterSciences et al. (2003).

CONCLUSION

In this paper we have introduced an incompletesampling of the research and monitoring studiesthat are currently being conducted to evaluate

383


how the Sacramento River ecosystem isresponding to management actions (full reportsfrom many of these studies are postedelectronically at http://www.sacramentoriverportal.org).Initial results suggest that current horticultural approaches to restoration are advancing our conservation goals,but that much additional benefit may be gainedfrom implementing complementary naturalprocessrestoration strategies. Although there are many importantstudies underway, there is a clear need to increase scientific scrutiny upon this important riverine ecosystem.We are at a pivotal time in the Great Valley’s history as a great deal of resources are being directed at salvagingthe degraded Sacramento River ecosystem. It is our goal to build a science program that can effectively contribute to this revitalization.

ACKNOWLEDGEMENTS

Rick Soehren and an anonymous reviewerprovided valuable criticisms on an earlier versionof this manuscript. We thank CALFED and theUSFWS for supporting many of the researchprojects described herein. The PRBO studieswere funded by The Nature Conservancy, theU.S. Fish and Wildlife Service, the David andLucille Packard Foundation, the William andFlora Hewlett Foundation, the National Fish andWildlife Foundation, CALFED, the Bureau ofReclamation, and the Natural ResourceConservation Service. S. Greco’s work wasfunded by a grant from the CA Department ofWater Resources (DWR). We thank the NationalScience Foundation Biocomplexity program andthe University of California MulticampusResearch Initiative program for funding theBiocomplexity Project. Aaron Holmes, andGrant Ballard provided assistance in studydesign and analysis of the PRBO studies. Wethank Jim DeStaebler, Glen Lubcke, JackCampbell, and Shelly Stephens and many othersfor dedicated field assistance.

LITERATURE CITED

Alpert, P., F.T. Griggs, and D.R. Peterson. 1999.Riparian forest restoration along large rivers:initial results from the Sacramento River Project.Restoration Ecology 7:360-368.

Ballard, G. and G.R. Geupel. 1999. Songbirdmonitoring on the San Luis National WildlifeRefuge, 1995-1997. Report to the U.S. Fish andWildlife Service. Point Reyes Bird Observatory,Stinson Beach, CA.

Bornette, G., C.A. Amoros, H. Piegay, J, Tachet, andT. Hein. 1998. Ecological complexity of wetlandswithin a river landscape. Biological Conservation 85:35-45.

Braatne, J.H., S.B. Rood, and P. Heilman. 1996. Lifehistory, ecology, and conservation of ripariancottonwoods. Pages 57-80 in: R.F. Stettler, H.D.Bradshaw, P.E. Heilman, and T.M. Hinckley

(Editors). Biology of populus and its implicationsfor management and conservation. NRCResearch Press, Ottawa, Ontario, Canada.

Brown, D.L. and Wood, D.M. 2002. Measure keyconnections between the river and floodplain.Report to The Nature Conservancy. Available athttp://www.sacramentoriverportal.org.

Buer, K., D. Forwalter, M. Kissel, and B. Stohler.1989. The middle Sacramento River: humanimpacts on physical and ecological processesalong a meandering river. Pages 22-32 in: D.L.Abell (Editor). Proceedings of the Californiariparian systems conference: protectionmanagement and restoration for the 1990s.General Technical Report PSW-110. PacificSouthwest Forest and Range Experiment Station,Forest Service, U. S. Department of Agriculture.Berkeley, CA.

Bunn, S.E., P.M. Davies, and T.D. Mosisch. 1999.Ecosystem measures of river health and theirresponse to riparian and catchment degradation.Freshwater Biology 41:333-345.

Cabana, G. and J. Rasmussen. 1996. Comparison ofaquatic food chains using nitrogen isotopes.Proceedings of the National Academy ofSciences of the United States of America 93:10844-10847.

Cabana, G., W. Giroux, M.E. Power, W.E. Rainey,J.C. Finlay, and C. Kendall. 2000. Tracingtrophic exchanges and contaminant flow betweenaquatic and terrestrial food webs using stableisotopes. 2000 INTECOL Congress, Quebec City, Canada.

CA Department of Water Resources. 1998. WoodsonBridge State Recreation Area long-term solutionsstudy working draft. State of California, theResources Agency, Department of WaterResources. Northern District Office, Sacramento, CA.

CA Resources Agency. 2000. Sacramento Riverconservation area handbook. Sacramento, CA.

Crone, E.E. and K.D. Holl. 2002. Biocomplexityincubation activities: linking hydrological andbiological dynamics in restored riparian forests.Report to the Nature Conservancy. Available athttp://www.sacramentoriverportal.org.

DeSante, D.F. and T.L. George. 1994. Populationtrends in the landbirds of western North America.Pages 173-190 in: J.R. Jehl, Jr. and N.K. Johnson(Editors). A century of avifaunal change inwestern North America. Studies in AvianBiology, no. 15. Cooper Ornithological Society,Lawerence, KS.

Ernst, C.H., J.E. Lovich, and R.W. Barbour. 1994.Turtles of the United States and Canada.Smithsonian Institution Press, Washington.

Fineberg, S.E. 1970. The analysis of multidimensionalcontingency tables. Ecology 51:419-433.

Gaines, D.F. 1977. The valley riparian forests ofCalifornia: their importance to bird populations.Pages 57-85 in: A. Sands (Editor). Riparianforests in California: their ecology andconservation. Institute of Ecology publication 15.University of California, Davis, CA.

384

G.H. GOLET, ET AL.

Gibbons, J.W., J.L. Greene, and J.D. Congdon. 1990.Temporal and spatial movement patterns ofsliders and other turtles. Pages 201-215 in: J.W.Gibbons (Editor). Life history and ecology of theslider turtle. Smithsonian Institute Press Washington, DC.

Goodman, R.J., Jr. and G.R. Stewart. 2000. Aquatichome ranges of female western pond turtles,Clemmys marmorata, at two sites in southernCalifornia. Chelonian Conservation and Biology 3:743-745.

Gore, J.A. and F.D. Shields. 1995. Can large rivers berestored? Bioscience 45:142-152.

Greco, S.E. 1999. Monitoring riparian landscapechange and modeling habitat dynamics of theYellow-billed Cuckoo on the Sacramento River,California. Ph.D. dissertation. University ofCalifornia, Davis, CA.

Greco, S.E., R.E. Plant, and R.H. Barrett. 2002.Geographic modeling of temporal variability inhabitat quality of the Yellow-billed Cuckoo onthe Sacramento River, miles 196-219, California.Pages 183-195 in: J.M. Scott and P. Heglund(Editors). Predicting Species Occurrences: Issuesof Scale and Accuracy. Island Press, Covelo, CA.

Griggs, F.T. and G.H. Golet. 2002. Riparian valleyoak (Quercus lobata) forest restoration on themiddle Sacramento River. Pages 543-550 in:R.B. Standiford, D. McCreary, and K.L. Purcell(Technical Coordinators). Proceedings from thefifth symposium on oak woodlands: Oaks inCalifornia’s Changing Landscapes, October 22-25, 2001, San Diego, CA. USDA Forest ServiceGeneral Technical Report. PSW-GTR-184 Albany, CA.

Holl, K.D. and E.C. Crone. Testing islandbiogeography and landscape theory in restoredriparian forests. Ecological Applications. in review.

Holmes, L.C., J.W. Nelson, and Party. 1913.Reconnaissance soil survey of the SacramentoValley, California. Field operations of the Bureauof Soils, U.S. Department of Agriculture,Government Printing Office, Washington, DC.

Johannesson, H. and G. Parker. 1989. Linear theory ofriver meanders. Pages 181-213 in: S. Ikeda andG. Parker (Editors). River Meandering. WaterResources Monograph 12. American GeophysicalUnion, Washington, DC.

Johnson, D.H. 1979. Estimating nest success: theMayfield method and an alternative. Auk 96:651-661.

Jones and Stokes. 2000. Functional relationships forthe ecosystem functions model Sacramento-SanJoaquin Rivers Basin Comprehensive Study. U.S.Army Corps of Engineers, Sacramento, CA.

Knopf, F.L., R.R. Johnson, T. Rich, F.B. Samson, andR.C. Szaro. 1988. Conservation of riparianecosystems in the United States. Wilson Bulletin100:272-284.

Kondolf, G.M., T. Griggs, E. Larsen, S. McBain, M.Tompkins, J. Williams, and J. Vick. 2000. Flowregime requirements for habitat restoration alongthe Sacramento River between Colusa and RedBluff. CALFED Bay Delta Program IntegratedStorage Investigation. Sacramento, CA.

Krebs, C.J. 1994. Ecology: the experimental analysisof distribution and abundance (4th edition).HarperCollins Collage, New York, NY.

Larsen, E.W. 1995. The mechanics and modeling ofriver meander migration. Ph.D. dissertation.University of California, Berkeley, CA.

Larsen, E.W. and S.E. Greco. 2002. Modeling channelmanagement impacts on river migration: A casestudy of Woodson Bridge State Recreation Area,Sacramento River, California, USA. EnvironmentalManagement 30:209-224.

Limm, M.P. and M.P. Marchetti. 2003. Contrastingpatterns of juvenile chinook salmon(Oncorhynchus tshawytschaw) growth, diet, andprey densities in off-channel and main stemhabitats on the Sacramento River. Report to TheNature Conservancy. Available at http://www.sacramentoriverportal.org.Lowrance, R.R., R.Leonard, and J. Sheridan. 1985. Managingriparian ecosystems to control nonpoint pollution.Journal of Soil and Water Conservation 40:87-91.

Mahoney, J.M. and S.B. Rood. 1998. Streamflowrequirements for cottonwood seedlingrecruitment—an integrative model. Wetlands 18:634-645.

Manley, P. and C. Davidson. 1993. A risk analysis ofneotropical migrant birds in California, U.S.Forest Service report, Region 5. San Francisco, CA.

Mayer, K.E. and W.F. Laudenslayer, Jr. 1988. A guideto the wildlife habitats of California. CADepartment of Forestry and Fire Protection, Sacramento, CA.

Mayfield, H.F. 1975. Suggestions for calculating nestsuccess. Wilson Bulletin 87:456-466.

Morken, I. and G.M. Kondolf. 2003. EvolutionAssessment and Conservation Strategies forSacramento River Oxbow Habitats. Report toThe Nature Conservancy. Available athttp://www.sacramentoriverportal.org.

Naiman, R.J., H. Décamps, and M. Pollock. 1993. Therole of riparian corridors in maintaining regionalbiodiversity. Ecological Applications 3:209-212.

Nur, N., T. Gardali, and G.R. Geupel. 1997a.Modeling the impact of cowbirds on LazuliBuntings: the source/sink equation in the CentralValley of California. Abstract from Conferenceon Management and Research on Impacts ofCowbirds in Western Landscapes, Sacramento,CA, October 1997.

Nur, N., S. Laymon, G. Geupel, and D. Evans. 1997b.Save our songbirds: songbird conservation inCalifornia’s riparian habitats; populationassessments and management recommendations.Final report, project 96-040. National Fish andWildlife Foundation, Washington, DC.

Patten, D.T. 1998. Riparian ecosystems of semi-aridNorth America: diversity and human impacts.Wetlands 18:498-512.

Plant, R.E., M.P. Vaysierres, S.E. Greco, M.R.George, and T.E. Adams. 1999. A qualitativespatial model of hardwood rangeland state-andtransitiondynamics. Journal of Range Management 52:51-59.

385


Ralph, C.J., J.R. Sauer, and S. Droege. 1995.Monitoring bird populations by point counts.USDA Forest Service Publication, PSW-GTR149, Albany, CA.

Richter, B.B., J.V. Baumgartner, J. Powell, and D.P.Braun. 1996. A method for assessing hydrologicalteration within ecosystems. ConservationBiology 10:1163-1174.

Roberts, M.D., D.R. Peterson, and D.E. Jukkola. 2002.A pilot investigation of cottonwood recruitmenton the Sacramento River. CALFED Final Report.

Roberts M.D., D.R. Peterson, D.E. Jukkola, V.L.Snowden. 2002. Pilot Study of FactorsInfluencing Riparian Forest Regeneration. Reportto CALFED. Available at http://www.sacramentoriverportal.org.

Rood, S.B. and J. Mahoney. 1990. Collapse of riparianpoplar forests downstream from dams in westernprairies: probable causes and prospects formitigation. Environmental Management 14:451- 464.

Schlorff, R. W. 1997. Monitoring bank swallowpopulations on the Sacramento River: a decade ofdecline. Transactions of the Western Section ofthe Wildlife Society 33:40-48.

Schultz, C.B. and E.E. Crone. 1998. Burning prairie torestore butterfly habitat: a modeling approach tomanagement tradeoffs for the Fender's blue.Restoration Ecology 6:244-252.

Scott, L.B. and S.K. Marquiss. 1984. An historicaloverview of the Sacramento River. Pages 51-57in: R. Warner and K. Hendrix (Editors).California Riparian Systems: EcologyConservation and Productive Management.University of California Press, Berkeley, CA.

Scott, M.L., G.T. Auble, and J.M. Friedman. 1997.Flood dependency of cottonwood establishmentalong the Missouri River, Montana, USA.Ecological Applications 7:677-690.

Shafroth, P.B., G.T. Auble, J.C. Stromberg, and D.T.Patten. 1998. Establishment of woody riparianvegetation in relation to annual patterns ofstreamflow, Bill Williams River, Arizona.Wetlands 18:577-590.

Small, S.L., T. Gardali, and N. Nur (in review).Source/sink status of the Black-headed Grosbeakin three regions of California. The Condor.

Small, S.L., N. Nur, A. Black, G.R. Geupel, D.Humple, and G. Ballard. 2000. Riparian birdpopulations of the Sacramento River system:results from the 1993-1999 field seasons. PointReyes Bird Observatory. Report to The NatureConservancy and the U.S. Fish and WildlifeService. Stinson Beach, CA.

Sommer, T.R., M.L. Nobriga, W.C. Harrell, W.Batham, and W.J. Kimmerer. 2001. Floodplainrearing of juvenile chinook salmon: evidence ofenhanced growth and survival. Canadian Journalof Fisheries and Aquatic Sciences 58:325-333.

Stillwater Sciences, W. Rainey, E. Pierson, and C.Corben. 2003. Sacramento River ecologicalindicators pilot study. Report to The Nature

Conservancy. Available at http://www.sacramentoriverportal.org.

TNC. 2003. Methods for Evaluating EcosystemIntegrity and Monitoring Ecosystem Response.Report to CALFED. Available at http://www.sacramentoriverportal.org.

Van Horne, B. 1983. Density as a misleading indicatorof habitat quality. Journal of Wildlife Management47:893-901.

Vander Zanden, M.J. and J.B. Rasmussen. 1999.Primary consumer δ15N and δ13C and the trophicposition of aquatic consumers. Ecology 80:1395-1404.

Wang, J.C.S. 1986. Fishes of the Sacramento-SanJoaquin Estuary and adjacent waters, California:a guide to the early life histories. TechnicalReport 9 (FS/B10-4ATR 86-9). Berkeley DigitalLibrary Project, Berkeley, CA.

Ward, J.V. and J.A. Stanford. 1995. Ecologicalconnectivity in alluvial river ecosystems and itsdisruption by flow regulation. Regulated Rivers11:105-120.

Wilson, D.S. 1998. Nest-site selection: microhabitatvariation and its effects on the survival of turtleembryos. Ecology 79:1884-1892.

Wolman, M.G. and J.P. Miller. 1960. Magnitude andfrequency of forces in geomorphic processes.Journal of Geology 68:54-74.

Wood, D.M. 2003a. Pattern of woody speciesestablishment on point bars of the middleSacramento River, California. Report to TheNature Conservancy. Available at http://www.sacramentoriverportal.org.

Wood, D.M. 2003b. The distribution and compositionof woody species in riparian forests along themiddle Sacramento River, California. Report toThe Nature Conservancy. Available athttp://www. sacramentoriverportal.org.

Yoshiyama, R.M., F.W. Fisher, and P.B. Moyle. 1998.Historical abundance and decline of chinooksalmon in the Central Valley region of California.North American Journal of FisheriesManagement 18:487-521.

Documents

Using Science to Evaluate Restoration Efforts and Ecosystem …ibdev.mcb.berkeley.edu/labs/power/publications/Golet... · 2008. 9. 15. · 368 INTRODUCTION In 1988 The Nature Conservancy