Establishing a Research andDemonstration Area Initiated byManagers: The SharkeyRestoration Research andDemonstration Site

Emile Gardiner, John Stanturf, Theodor Leininger, Paul Hamel,Lamar Dorris Jr., Jeffery Portwood, and James Shepard

As forest scientists increase their role in the process of science delivery, many research organizations aresearching for novel methods to effectively build collaboration with managers to produce valued results. Thisarticle documents our experience with establishment of a forest restoration research and demonstration areain the Lower Mississippi Alluvial Valley (LMAV), a region that has experienced extensive afforestation of formeragricultural land over the past 15 years. Although basic establishment techniques for production plantationshad been developed and applied on small areas, land managers lacked critical knowledge to implementoperational-scale afforestation practices that would accommodate multiple forest restoration objectives. In1993, managers with the US Fish and Wildlife Service made a 1,700-ac agricultural tract available to scientistsinterested in partnering research and demonstrating various aspects of bottomland hardwood ecosystemrestoration. Through collaborative efforts, resource managers and scientists have installed numerous exper-iments on the Sharkey Restoration Research and Demonstration Site to address relevant issues in afforestationand restoration of bottomland hardwood ecosystems. Development of this research and demonstration areahas provided a science-based resource for educating landowners, foresters, wildlife managers, and the generalpublic on afforestation techniques appropriate for restoration of bottomland hardwood forests; has served asa platform for scientists and land managers to cooperate on the development of innovative approaches toforest restoration; and has provided a venue for education and debate among policymakers active in theLMAV. Early results showed the viability of low-cost techniques such as direct seeding oaks, as well asintroducing the interplanting technique for rapid development of forest conditions. We recognize that the valueof the research and demonstration site is attributed in part to site characteristics and experimental design, andexpect continued work at the location to contribute to improved afforestation practices that will fosterestablishment of sustainable bottomland hardwood forests.

Keywords: science delivery, forest restoration, afforestation, LMAV, bottomland hardwood

T he need for effective ways to moveresearch results into practice havelong been discussed and the proper

roles of scientists and managers have beendebated. Increasingly, agencies that fund re-search are mandating that scientists engagein some form of knowledge transfer beyondpublishing journal articles. The value bothto the individual professionally and to theorganization corporately of acquiring newknowledge to stay competitive, increase pro-ductivity, or comply with regulations iswidely recognized by research and manage-ment communities. Structural and financialbarriers, however, preclude simple answersto the question of how to efficiently accom-plish this task. A well-developed theory ofinnovation diffusion (Rogers 1995) pro-vides a conceptual basis for this task, whichhas variously been called technology trans-fer, technology delivery, and, more recently,science delivery. The basic structure of the

Received April 11, 2008; accepted June 26, 2008.

Emile Gardiner (egardiner@fs.fed.us) is research forester, US Forest Service Southern Research Station, Center for Bottomland Hardwoods Research, P.O. Box 227,Stoneville, MS 38776. John Stanturf (jstanturf@fs.fed.us) is project leader, US Forest Service Southern Research Station, Center for Forest Disturbance Science, 320Green St., Athens, GA 30602. Theodor Leininger (tleininger@fs.fed.us) is project leader, and Paul Hamel (phamel@fs.fed.us) is research wildlife biologist, US ForestService Southern Research Station, Center for Bottomland Hardwoods Research, P.O. Box 227, Stoneville, MS 38776. Lamar Dorris Jr. (lamar_dorris@fws.gov)

Journal of Forestry • October/November 2008 363

AB

ST

RA

CT

education & communication

innovation process consists of three steps:research to produce an innovation, dissemi-nating knowledge of the innovation outsideof scientific circles, and implementation ofthe innovation within a user community.Presently, technology transfer (the dissemi-nation step) is a participatory model of sci-ence delivery with a push–pull dynamic: sci-entists push out generated knowledge andrespond to pulls from users and the market-place of ideas for researchable topics.

Rogers (1995) connected the dissemi-nation and implementation stages of the in-novation process through the decision toimplement a new technology. In our experi-ence, the closer (physically as well as psycho-logically) researchers and managers are whenat the crucial step of deciding to try an inno-vation the more likely that knowledge willbe effectively and efficiently transferred.Studies in forestry and other natural re-sources management professions supportthe idea that closely linked social networksfacilitate effective transmission of innova-tions (Baldwin and Haymond 1994, Hag-gith et al. 2003, Stanturf et al. 2003).

We propose that one highly effectivetechnique for delivering needed results tousers is by establishing a research and dem-onstration site where managers and re-searchers can work collaboratively to answerpressing practical questions while develop-ing a test bed to explore new innovativetechniques or theoretical developments. Thebasic idea of a demonstration site is not nov-el; it is the primary method used in agricul-ture by extension specialists and agribusinessservice providers (Leeuwis and van den Ban2004). A well-known example in forestry isthe model forest network (Forestry Canada1992). In these examples, however, the pur-pose is usually to demonstrate or validate re-search conducted under experimentally con-trolled conditions. We suggest that an evenmore effective approach is to involve re-source managers in the design of the researchand installation of the research and demon-stration site. In the following discussion,we will illustrate this approach by presentingour experience with the Sharkey Restora-tion Research and Demonstration Site(SRRDS), where, at the urging of a regionalprogram manager (US Fish and Wildlife

Service [USFWS]), a research and demon-stration area was initiated in which foresters,refuge managers, and scientists are workingcollaboratively to address multiple questionsrelevant to restoration of bottomland hard-wood ecosystems.

ContextThe Lower Mississippi Alluvial Valley

(LMAV) in the southern United States isexperiencing significant afforestation (Stan-turf et al. 2000, Schoenholtz et al. 2001) formany of the same reasons that afforestationis escalating globally (Food and AgricultureOrganization 2001, Weber 2005). The driv-ing forces behind current forest restorationvary by region and ownership, but factorssuch as increased agricultural efficiency,biodiversity conservation, soil and waterprotection, carbon sequestration, demandfor timber resources, and reduction of land-scape fragmentation are among the mostprevalent contributors to this shift in landuse (Stanturf and Madsen 2005). Counter-acting the deforestation of more than 70%of the 24 million ac in the LMAV that oc-curred over the last two centuries has beenstimulated by Federal “Farm Bills” begin-ning in the mid-1980s and early 1990s(Kennedy 1990, Stanturf et al. 2000). Sincethe enactment of this legislation, bottom-land hardwood tree plantations have beenestablished on over 500,000 ac of former ag-ricultural land in the LMAV, and additionalarea is pending approval and funding(Gardiner and Oliver 2005).

The significant afforestation on privateland that began in earnest in the 1990s waspreceded on public holdings as early as the1960s. These earlier afforestation effortsconcentrated on establishing relatively fewspecies, particularly species that producehard mast such as the bottomland oaks(Quercus spp.), which would provide a com-ponent of wildlife habitat (Allen 1990, Stan-turf et al. 1998). Relying on an active re-search program on artificial regenerationtechniques for bottomland oaks, managerswere able to show successful establishmentof forest stands through a few, relativelysmall-scale plantings (Kennedy 1993, Stan-turf et al. 2001). Because of this success,these early afforestation techniques and

practices were adopted as the model ap-proach and were applied extensivelythroughout the region when current conser-vation programs were initiated in the late1980s (Stanturf et al. 1998, Gardiner et al.2002, Haynes 2004).

Despite the proven success of the ap-proach developed in the 1960s and 1970sprimarily for timber production, its defi-ciencies relative to changing priorities to-ward multiple resource management andbiodiversity objectives became apparent bythe mid-1990s. Allen (1997) questioned thesuitability of this basic approach to providefor biodiversity. The approach was modifiedunder the Wetlands Reserve Program(WRP) and the stocking density of theplanted species was much reduced; diversityand additional stocking was to come fromcolonization by light-seeded species fromnearby forests. Unfortunately, many of therestoration plantings were beyond the effec-tive dispersal range of light-seeded species.Stanturf et al. (1998) argued that the lowstocking densities common in WRP standswould produce inferior timber quality andsubsequently reduce future stand-manage-ment options. Still, others argued that im-proved afforestation systems could be devel-oped to catalyze various forest restorationprocesses and enhance certain ecosystemfunctions on bottomland sites such as habi-tat for neotropical migratory bird species,water quality protection, and carbonsequestration (Twedt and Portwood 1997,Stanturf et al. 2000, 2001). Thus, tremen-dous effort and resources were being allo-cated to establish forest stands that weremarginal or inadequate for producing de-sired outputs, and it was also clear that thisapproach would not lead to the establish-ment of a sustainable forest resource in theregion. Forest managers throughout theLMAV voiced a need to establish an infra-structure of afforestation and restorationresearch to expand existing knowledge anddevelop a forum for disseminating new af-forestation practices appropriate for restora-tion of bottomland hardwood ecosystems.

In the early 1990s the USDA Farm Ser-vice Agency (formerly Farmers Home Ad-ministration) acquired through foreclosureand transferred several holdings of agricul-

is administrative forester, US Fish and Wildlife Service, Theodore Roosevelt National Wildlife Refuge Complex, 728 Yazoo Refuge Rd., Hollandale, MS 38748.Jeffery Portwood (JeffPortwood@forestargroup.com) is manager, Forester Real Estate Group, Diboll, TX 75941. James Shepard (jshepard@cfr.msstate.edu) is chair,Department of Forestry, Mississippi State University, Mississippi State, MS 39762. The scientists, technicians, and managers who contributed to this work are toonumerous to mention. Two, however, were critical to establishing the site and developing the cooperation. The authors give special thanks to Dr. Ronnie Haynes andMr. Tim Wilkins of the US Fish and Wildlife Service; without their support, the Sharkey Restoration Research and Development Site would never have happened.

364 Journal of Forestry • October/November 2008

tural land in the LMAV to the US FWS,National Wildlife Refuge System. Thetransfer to the National Wildlife Refuge Sys-tem mandated that this agricultural land berestored to wildlife habitat. The Service wasinterested in improving the success of theirafforestation; in particular, they faced diffi-culty on wetter sites. Recognizing the poten-tial value of such properties to advancingforest restoration knowledge in the LMAV,personnel with the Southeast Region of theUS FWS proposed that a 1,700-ac tract lo-cated in Sharkey County, Mississippi, re-cently transferred to the Theodore RooseveltNational Wildlife Refuge Complex (for-merly Yazoo National Wildlife RefugeComplex) be reserved for use as a researchand demonstration area highlighting bot-tomland hardwood afforestation and forestrestoration practices (Table 1).

The invitation by the US FWS to workon a site reserved for research and demon-stration attracted parties from several orga-nizations and launched collaboration be-tween managers and researchers. The initialplanning sessions in 1993 and subsequentannual meetings provided a forum for ex-ploring research topics, sharing resources,and coordinating installation of experi-mental infrastructure that established theSharkey Restoration Research and Demon-stration Site (SRRDS). Designation of theresearch and demonstration site eventuallydrew together an extensive group of cooper-ators whose primary interest is “seeking toimprove our ability to restore bottomlandhardwood forests after agriculture abandon-ment” (Shepard 1996). Among those activeon the site were federal agencies, representedby the USFWS, US Forest Service, NaturalResources Conservation Service (NRCS),US Geological Survey, and the US ArmyCorps of Engineers (USCOE); State agen-

cies and universities including MississippiState University and Stephen F. Austin StateUniversity; nongovernmental organizationsincluding the National Council for Air andStream Improvement (NCASI); and privateindustry including Crown Vantage Corpo-ration and International Paper Company(Table 2). From the beginning, governancewas loosely structured; despite holding vetopower, the FWS only required that activitiesbe coordinated and within general refugeregulations.

Because the site was large enough to ac-commodate all needs and scientists in gen-eral used their own resources, there were fewconflicts. Scientists from the several agencieshad their own research agendas and devel-oped their questions and studies in collabo-ration with their own set of managers. TheNRCS, lacking their own research staff,funded universities to look at its specificquestions that were not being addressed byscientists from other agencies.

Beginning with initial field installationsin 1994, a broad range of research and dem-onstration experiments have been estab-lished on the site over the past 10 years (Fig-ure 1). A prominent installation on the site isa large-scale (240 ac) study of alternative af-forestation practices for forest restoration.This experiment contrasts four afforestationoptions that allowed us to compare passive(native recolonization) with active restora-tion (direct seeding and planting of Nuttalloak [Quercus nuttallii Palmer]), operationaltechniques on the same site (direct seedingand planting), and standard treatments withthe most intensive treatment of interplant-ing two species that differed in successionalstatus (early successional eastern cotton-wood [Populus deltoides Bartram ex Marsh.ssp. deltoides], with the midsuccessionalNuttall oak). A second installation on thesite is a 25-ac study of natural regenerationon abandoned agricultural land. This re-search focuses on the patterns of natural in-vasion by woody tree and shrub species onformer agricultural land relative to distanceand orientation from existing forest edge.The SRRDS also holds a 20-ac fenced area(for protection from animal damage) forstudying silvicultural techniques to establishnontraditional afforestation species. Thisarea is currently being used to investigateblack willow (Salix nigra L.) establishmenttechniques for rapid growth and carbon se-questration on frequently flooded wetlandsites. Another valuable component of thesite’s infrastructure is an impoundment sys-

tem designed with 12, 1-ac cells that can beindependently flooded and drained. ThisFlooding Research Facility (Lockhart et al.2006) was designed to enable researchers toinvestigate the effect of different levels andtiming of flooding on woody plants estab-lished on native soils. Initially, university re-searchers used the facility to look at severalquestions of immediate concern to theNRCS, including planting material, meth-ods, planting date, and flooding regimes.Currently, a team of US Forest Service re-searchers is investigating the influences ofsoil inundation and light availability onphysiology and growth of the endangeredbottomland shrub, pondberry (Lindera me-lissifolia [Walt.] Blume) at the request of theUSACE.

The greatest involvement of scientistsfrom multiple disciplines and diverse agen-cies was directed toward the comparison ofafforestation methods, in particular, the in-terplanting treatment. The public land man-agers and program managers (USFWS andNRCS) posed the core question of whetherit was better to plant oak seedlings or sowacorns; although both techniques had beensuccessful in research plots, there was mixedsuccess in operational plantings and, indeed,little research had evaluated the techniquesin side-by-side comparisons. The researcherswho took on this question (US Forest Ser-vice and NCASI) added a “do nothing”treatment to provide a baseline of old fieldsuccession, with the same starting point asactive restoration treatments. The research-ers also wanted to include a treatment forestablishing a mixture of species. A privateland manager (Crown Vantage), who wasnot involved in the initial discussions, sug-gested trying a technique he was using onfarmers’ land, interplanting oak with cot-tonwood. Once the study was installed,additional questions were asked and otherscientists became involved: what effect dosmall mammals have, especially on thedirect-seeding treatment? Does the rapidaccretion of vertical structure by the cotton-wood attract birds? Whether the cotton-wood benefits or harms the interplantedoak, what is the physiological response of theoak relative to seedlings planted in the open?The US Forest Service scientist leading theinstallation of the study recruited additionalscientists (US Forest Service) to addressthese questions.

The development of the SRRDS hasproceeded over 10 years. Because the entiresite was not used immediately, scientists

Table 1. Key characteristics of the SharkeyRestoration Research and DemonstrationSite.

Location: 32°58�N, 90°44�W; � 5 mi east ofAnguilla, Sharkey County, MS

Site area: 1,700 acPhysiographic region: Mississippi Alluvial PlainDominant soil: Sharkey Clay (very fine, smectitic,

thermic, chromic, Epiaquerts)Climate: Humid subtropical region of the Northern

Temperate ZoneMean annual precipitation: 52 in.Mean temperature—July: 82.0°FMean temperature—January: 45.5°FMean growing season length: 229 d

Journal of Forestry • October/November 2008 365

have been able to stage installation of studiesas resources became available and to fol-low-up on initial findings or new questionsthat arose. The design of the larger experi-mental areas has allowed for numeroussmaller-scaled studies to be comprehensivelyimposed over larger-scaled projects. TheFlooding Research Facility was designed toaddress questions that could be answeredwith short-term studies of relatively smallplots, such as early stand development. Thisdesign provided flexibility to liquidate a

study and install another. The large-scalestudy to examine alternative afforestationtechniques, on the other hand, was designedas a long-term study and the large plot sizewas chosen to allow us to study utilization bywildlife. The large treatment plots also pro-vided space to impose smaller studies of ef-fects of four cottonwood clones on inter-planted oak and disking 1 year versus 2 yearsto establish cottonwood. One particular ad-vantage of using the fast-growing cotton-wood is that we have completed a full rota-

tion (12 years) and imposed alternativeharvesting treatments (clearcut, thinning,and coppice). Because of the varied develop-ment of short-term and long-term researchat the SRRDS, even repeating visitors findsomething new to see.

OutcomesThe effort on the SRRDS has resulted

in many products; naturally, articles pub-lished in the scientific literature have been aprimary output. Beyond peer-reviewed pub-lications, at least three beneficial outcomeshave developed from establishment of thesite. First, the SRRDS has provided a plat-form for land managers and scientists to co-operate on the development of innovativeapproaches to forest restoration. Second, thesite has provided a science-based resource foreducating landowners, resource managers,and the general public on afforestation tech-niques and practices appropriate for restora-tion of bottomland hardwood forests. Fi-nally, the site has been a venue for educationand debate among landowners, nongovern-mental organizations, and policymakers ac-tive in the LMAV. This section will high-light these outcomes providing additionaldetail of how some of these benefits havebeen realized on the SRRDS.

A Cooperative Platform for Innova-tive Bottomland Hardwood EcosystemRestoration. A key reason to develop theSRRDS was to provide a platform wheremanagers and researchers worked togetherto address important problems or questionson afforestation and restoration of bottom-land hardwood forests. A cooperative spirithas resulted in the development of innova-tive practices for restoration of bottomlandhardwood ecosystems. For example, forest-ers with Crown Vantage Corporation pro-vided ideas that led to testing the innovative

Figure 1. Aerial photo map of the SRRDS, including prominent experimental infrastructure.

Table 2. Partial list of collaborative organizations, their activities, and outcomes from their involvement on the Sharkey RestorationResearch and Demonstration Site.

Core group Activities Outcomes

US Forest Service Research, education, site development, management Publications, tours, new technology, policy changedevelopment, management

US Fish and Wildlife Service Site development, management, funding Site restoration, partnershipsUS Geological Survey Research PublicationsUS Natural Resources Conservation Service Site development, funding Publications, staff trainingUS Army Corps of Engineers Research PublicationsCrown Vantage Corporation Site development, management, research Proof of concept, landowner acceptanceInternational Paper Funding Proof of conceptNational Council for Air and Stream Improvement Site development, funding Proof of conceptMississippi State University Research Publications, student trainingStephen F. Austin State University Research Publications, student trainingMississippi Valley State University Research Publications, student training

366 Journal of Forestry • October/November 2008

interplanting system for restoring bottom-land hardwood ecosystems (Hamel 2003,Gardiner et al. 2004). This system estab-lishes a complex plantation that promotesrapid stand development (Fisher et al.2002), catalyzes development of ecosystemprocesses (Hamel 2003, Gardiner et al.2004), and accommodates multiple man-agement objectives (Gardiner et al. 2004).

A Science-based Demonstration Areafor Bottomland Hardwood EcosystemRestoration. A primary outcome of estab-lishing the site has been the provision of ascience-based demonstration area for bot-tomland hardwood ecosystem restoration.Annually, numerous field trips and work-shops are hosted by the researchers and landmanagers at the site where landowners,managers, students, scientists, and policy-makers observe firsthand operationally in-stalled large-scale afforestation (Table 3).Although these operationally installed large-scale treatments are invaluable for demon-stration purposes, supplementing these ob-servations with research findings on criticaltopics such as stand productivity, winteringbird use, small mammal use, and changes insoil quality are essential for linking specificafforestation practices to restoration of par-ticular ecosystem processes. Because theSRRDS was designed with both objectivesin mind, visitors to the site receive visualimages of various afforestation practices, aswell as comprehensive knowledge of impactson selected ecosystem processes. The effortsto promote the site as a demonstration areato educate interested parties on afforestationand restoration of bottomland hardwoodecosystems have been recognized jointly bythe US Forest Service and Ducks Unlimited,Incorporated, through their “Taking WingAward” for Public Awareness in 1997. Theeducational value of the site and accompa-nying research was realized in unexpectedways; a group of special needs children in astructured living center volunteered to assistin the wildlife fieldwork (Figure 2), earning

them an award for volunteerism from theDirector of the Southern Research Station.

A Venue for Policy Formation. Theeducation of policymakers is of paramountconcern to resource managers trying toadopt and implement innovative afforesta-tion and forest restoration practices. Inform-ing policymakers is particularly importantbecause the primary thrust behind forest res-toration in the LMAV is conservation ease-ments offered and administered by agenciesof the Federal Government. The SRRDShas provided policymakers with a conve-nient location to observe various afforesta-tion practices and forest restoration ap-proaches that are subject to scientificresearch (Figure 3). The value of the SRRDSfor affecting policy was shown recently by achange that should increase the economicviability and sustainability of afforestationpractices in the LMAV. In May 2005, theFarm Service Agency, which administers theConservation Reserve Program, amendedtheir guidelines for Conservation Practice31, Bottomland Timber Establishment onWetlands, to allow for establishment of cot-tonwood (Populus spp.) stands interplantedwith oak (USDA, Farm Service Agency No-tice CRP-496). Onsite visits to the SRRDS,extension of knowledge gained from re-search on the site, and debates between landmanagers, scientists, and Washington officestaff visiting the site were pivotal to imple-mentation of this recent policy change. Todate, this practice is still not allowed onWRP.

Planning for SuccessSeveral features of the SRRDS have

contributed to its value as a source of infor-mation and usefulness as a demonstrationarea. In this section we highlight three im-portant features associated with site selec-tion and three features related to experimen-tal design that have contributed to thesuccess of the research and demonstrationarea. Although some of these features wereof intentional design, the importance of oth-

ers was realized retrospectively. These note-worthy aspects of the SRRDS are discussedas recommendations for consideration bymanagers and researchers who may plansimilar sites in the future.

Sites Should Be Located on a Neutraland Secure Area. The SRRDS was estab-lished on an area within the National Wild-life Refuge System. No hurdles due toagency policies were encountered, possiblybecause the site was being restored from ac-tive farming for research purposes. Cer-tainly, the enthusiastic support of the refugemanager and his staff also was a factor. Twoprimary advantages associated with publicownership of the site are neutrality and se-curity. The neutrality of our site has at-tracted interest from and spawned collabo-ration between multiple organizationsincluding federal and state agencies, univer-sities, private industry and nongovernmen-tal organizations. Furthermore, because theproperty is under federal management aspart of the refuge system, researchers havepractical assurance that the site will be securefor establishment of long-term experiments.

Selected Sites Should Be Representa-tive of the Problem. To ensure the value offield research and extension of findings and

Figure 2. High school children assisting re-search scientists with wintering bird usesampling in habitat created by alternativeafforestation practices.

Figure 3. Field discussion on conservationprogram policy between forest scientists,landowner advocate groups, and policy-making officials.

Table 3. Categories of visitors to the Sharkey Restoration Research and DemonstrationSite and outcomes relevant to the knowledge and awareness gained through site visits.

Visitor category Outcomes

Educational groups Student awareness of practical and technical issuesPolicymakers Awareness of potential user needs and changes in federal

policiesPotential users (consultants, landowners) Increased acceptance of innovation and alternative practicesScientists Experimental approaches, knowledge advancement

Journal of Forestry • October/November 2008 367

recommendations to the intended audience,the selected demonstration site must be rep-resentative of the targeted problem area. Thesite chosen for installation of our researchand demonstration area can be consideredtypical of afforestation sites throughout theLMAV for several reasons. First, because ofthe soil and hydrologic regime on the site, itcan be classified as a farmed wetland and itsland-use history is representative of mostlow-lying areas cleared in the 1960s and1970s for soybeans. Second, the predomi-nant soils on the site, Sharkey Clay, is also adominant soil series in the LMAV (Pettryand Switzer 1996) and probably the mostprevalent soil enrolled in conservation ease-ment programs in this region. Additionally,the site is characteristic of other afforestationtracts in the LMAV because land adjacent toour site has remained in agricultural produc-tion, largely isolating the site from other for-ested tracts. These site characteristics in-crease the value of knowledge gained fromexperimentation established on the site bymaking it directly applicable to the majorityof land enrolled in conservation easementprograms in the LMAV.

The Research and DemonstrationArea Should Be Easily Accessible. A dem-onstration area inaccessible to the intendedaudience is of little value. Our experienceindicates that many visitors to our site areeither unprepared or would be unwilling totrek long distances across inhospitable ter-rain to make observations on experimentalplantings. This is particularly true of visitors,such as policymakers, who routinely do notwork in the field. Thus, access should be ac-counted for during site selection as well asduring the layout and configuration of ex-perimental plantings. Because some of theexperimentation on our site is adjacent to apublic road, people traveling through thearea can readily observe forest developmentin the agricultural landscape (Figure 1). Ad-ditionally, interior portions of the site can beeasily accessed on foot through an estab-lished road and trail system. A disadvantageof such ready access is potential vandalism.For this reason, we did not install signage onthe site because we considered it too tempt-ing a target for local sharpshooters.

Sites Should House Large-Scale Re-search and Demonstrations. Establish-ment of large-scale research and demonstra-tion has greatly benefited the educationalvalue of our site. Although establishment oflarge-scale research will generally increasethe installation and measurement costs over

smaller-scale projects, we have observed sev-eral advantages of the large-scale installa-tions on our site. First, the visual impact of atreatment effect is enhanced when experi-mental units encompass a relatively largearea. Next, these large-scale experimentalunits enable visitors to visualize the treat-ment on other landscapes. Additionally, es-tablishment of large experimental units al-lows for building comprehensiveness bylayering of additional studies within theframework of the primary study. This couldinclude examination of response variablesthat are influenced by area (e.g., certainwildlife species), temporal variation (e.g.,forest stand development patterns), or vari-ous nested treatments applied in the future(e.g., intermediate stand practices or har-vests).

Large-Scale Experiments Should BeInstalled Operationally. When possible,large-scale experiments on the research anddemonstration site were installed operation-ally, through which we have realized at leastthree pragmatic advantages (Figure 4). Aprimary advantage is the necessary involve-ment by managers on the design and instal-lation of experiments. Second, it providesvisitors the assurance that particular treat-ments can be installed on their sites withexisting equipment and materials. Last, op-erational installation of experiments gener-ates realistic inputs for cost accounting mod-els of applied practices. Such information iscritically important to landowners and man-agers looking to invest in new afforestationtechnology.

Experimentation Should IncludeComparisons with Conventional Prac-tices. Including an operational control of aconventional practice in the experimentaldesign is very useful when conducting re-search or establishing a demonstration area

on alternative afforestation practices. Doingso provides a familiar baseline from which tomake direct comparisons of treatment in-puts and results on a common site. Addi-tionally, such experimental designs often al-low for striking visual contrast of knownpractices to innovative alternative treat-ments.

SummaryEstablishment of the SRRDS arose

through the vision of managers and the co-operative spirit of researchers who wereactively practicing and studying forest resto-ration in the LMAV. What began as an ini-tiative to improve bottomland hardwoodrestoration has grown to a collaborative ef-fort by land managers and scientists promot-ing the advancement of afforestation prac-tices and forest restoration in bottomlandhardwood ecosystems through comprehen-sive research and demonstration. Work ac-complished on the site has had a positiveimpact on afforestation in the LMAV astechniques and practices evolve toward es-tablishment of more complex plantationsthat are better suited to accomplish multipleobjectives. Active dialogue between resourcemanagers, scientists, landowners, and poli-cymakers has been pivotal in transferringknowledge and implementing innovativepractices developed on the site. The value ofthe site as a research and demonstration areais attributed to characteristics of the locationand features of experimental design. Contin-ued collaboration on the site should fosterdevelopment of economically viable prac-tices to establish ecologically sustainable for-ests for the advancement of restoration ofthis vital bottomland hardwood resource.

Literature CitedALLEN, J.A. 1990. Establishment of bottomland

oak plantations on the Yazoo National Wild-life Refuge Complex. South. J. Appl. For. 14:206–210.

ALLEN, J.A. 1997. Reforestation of bottomlandhardwoods and the issue of woody species di-versity. Restor. Ecol. 5:125–134.

BALDWIN, S.B., AND J.L. HAYMOND. 1994. A sys-tems approach to communication behavioramong scientists, foresters, and NIPF land-owners. South. J. Appl. For. 4:175–180.

FOOD AND AGRICULTURE ORGANIZATION (FAO).2001. Global forest resources assessment 2000—Main report. United Nations FAO ForestryPap. 140, Rome, Italy. 479 p.

FORESTRY CANADA. 1992. Canada’s model forestprogram: An initiative for sustainable develop-ment. Forestry Canada, Chalk River, Ontario,Canada. Available online at www.modelforest.net/cmfn/en/publications/publications/publica

Figure 4. Large-scale experimental plots,such as this 20-ac planting of cottonwoodand oak, allow for operational-scale treat-ments including selectively harvesting thecottonwood and releasing the oak.

368 Journal of Forestry • October/November 2008

tions_record.aspx?title_id�261; last accessedMar. 26, 2008.

FISHER, R.K., E.S. GARDINER, J.A. STANTURF,AND C.J. PORTWOOD. 2002. Disking effects onfifth-year volume production of four easterncottonwood clones established on an afforesta-tion site, Sharkey County, Mississippi. P. 515–519 in Proc. of the 11th biennial southern silvi-cultural research conf., Outcalt, K.W. (ed.). USFor. Serv. Gen.Tech. Rep. SRS-GTR-48,Asheville, NC.

GARDINER, E.S., D.R. RUSSELL, M. OLIVER, AND

L.C. DORRIS, JR. 2002. Bottomland hardwoodafforestation: state of the art. P. 75–86 in Proc.of a conf. on sustainability of wetlands and waterresources: How well can riverine wetlands con-tinue to support society into the 21st century?Holland, M.M., M.L. Warren, and J.A. Stan-turf (eds.). US For. Serv. Gen. Tech. Rep.SRS-GTR-50, Asheville, NC.

GARDINER, E.S., J.A. STANTURF, AND C.J.SCHWEITZER. 2004. An afforestation systemfor restoring bottomland hardwood forests:Biomass accumulation of Nuttall oak seedlingsinterplanted beneath eastern cottonwood. Re-stor. Ecol. 12:525–532.

GARDINER, E.S., AND J.M. OLIVER. 2005. Resto-ration of bottomland hardwood forests in theLower Mississippi Alluvial Valley, U.S.A. P.235–251 in Restoration of boreal and temperateforests, Stanturf, J.A., and P. Madsen (eds.).CRC Press, Boca Raton, FL.

HAGGITH, M., R. PRABHU, C.J.P. COLFER, B.RITCHIE, A. THOMSON, AND H. MUDAVANHU.2003. Infectious ideas: Modeling the diffusionof ideas across social networks. Small-scale For.Econ. Manag. Policy 2:225–239.

HAMEL, P.B. 2003. Winter bird community dif-ferences among methods of bottomland hard-wood forest restoration: Results after sevengrowing seasons. Forestry 76:189–197.

HAYNES, R.J. 2004. The development of bottom-land forest restoration in the Lower MississippiRiver Alluvial Valley. Ecol. Rest. 22:170–182.

KENNEDY, H.E., JR. 1990. Hardwood reforestationin the South: Landowners can benefit from Con-servation Reserve Program Incentives. US For.Serv. Res. Note SO-RN-364, New Orleans,LA.

KENNEDY, H.E., JR. 1993. Artificial regenerationof bottomland hardwoods. P. 241–249 in Oakregeneration: Serious problems, practical recom-mendations, Loftis, D.L., and C.E. McGee(eds.). US For. Serv. Gen. Tech. Rep. SE-GTR-84, Asheville, NC. 6 p.

LEEUWIS, C., AND A. VAN DEN BAN. 2004. Com-munication for rural innovation: rethinking ag-ricultural extension, 3rd Ed. Blackwell Science,Ltd., Oxford. 424 p.

LOCKHART, B.R., E.S. GARDINER, T.D. LEIN-INGER, K.F. CONNOR, P.B. HAMEL, N.M.SCHIFF, A.D. WILSON, AND M.S. DEVALL.2006. Flooding facility helps scientists exam-ine the ecophysiology of floodplain speciesused in bottomland hardwood restorations.Ecol. Restor. 24:151–157.

PETTRY, D.E., AND R.E. SWITZER. 1996. Sharkeysoils in Mississippi. Mississippi Agric. and For.Exp. Stn., Bull. 1057. 37 p.

ROGERS, E.M. 1995. Diffusion of innovations.The Free Press, New York. 518 p.

SCHOENHOLTZ, S.H., J.P. JAMES., R.M. KAMIN-SKI, B.D. LEOPOLD, AND A.W. EZELL. 2001.Afforestation of bottomland hardwoods in the

Lower Mississippi Alluvial Valley: Status andtrends. Wetlands 21:602–613.

SHEPARD, J. 1996. Cooperative bottomland hard-wood restoration research project, SharkeyCounty, MS. Progress Rep., May 29, 1996,NCASI, Southern Regional Center, Gaines-ville, FL. 41 p.

STANTURF, J.A., C.J. SCHWEITZER, AND E.S.GARDINER. 1998. Afforestation of marginal ag-ricultural land in the Lower Mississippi RiverAlluvial Valley, U.S.A. Silva Fenneca 32:281–297.

STANTURF, J.A., E.S. GARDINER, P.B. HAMEL,M.S. DEVALL, T.D. LEININGER, AND M.W.WARREN JR. 2000. Restoring bottomlandhardwood ecosystems in the Lower MississippiAlluvial Valley. J. For. 98:10–16.

STANTURF, J.A., S.H. SCHOENHOLTZ, C.J.SCHWEITZER, AND J.P. SHEPARD. 2001. Achiev-ing restoration success: myths in bottomlandhardwood forests. Restor. Ecol. 9:189–200.

STANTURF, J.A., R. KELLISON, F.S. BROERMAN,AND S.B. JONES. 2003. Innovation and forestindustry: Domesticating the pine forests of thesouthern United States, 1920–1999. For. Pol-icy Econ. 5:407–419.

STANTURF, J.A., AND P. MADSEN. 2005. Restora-tion of boreal and temperate forests. CRC Press,Boca Raton, FL. 569 p.

TWEDT, D.J., AND J. PORTWOOD. 1997. Bottom-land hardwood reforestation for neotropicalmigratory birds: Are we missing the forest forthe trees? Wildl. Soc. Bull. 25:647–652.

WEBER, N. 2005. Afforestation in Europe: Les-sons learned, challenges remaining. P.121–135 in Restoration of boreal and temperateforests, Stanturf J.A., and P. Madsen (eds.).CRC Press, Boca Raton, FL.

Journal of Forestry • October/November 2008 369