ECOLOGICAL SUCCESSION OF BREEDING BIRD POPULATIONS …

[Pages:6]ECOLOGICAL

SUCCESSION OF BREEDING

POPULATIONS

IN NORTHWESTERN

ARKANSAS

BIRD

HERMANHENRYSHUGARTJ,R.,ANDDOUGLAJSAMES

T?s study was designedto determinethe spedes and densitiesof breeding birds in seral stages of upland subsere successionin the Ozark Highlandsat Pea Ridge National Military Park in Benton County, Arkansas.Ecologicalsuccessioonf uplandplants in the Ozark Highlands (Figure 1) varies regionally with the nature of the bedrock (Cozzens, 1940). Dale and Fullerton (1964) describedthe early stagesof the uplandsubserenear the presentstudy area.

The first stageof the secondarysuccessioins highly variabledepending on prior land use. Fields burned and abandonedoften have horseweed (Erigeron canadensis); land heavily grazed before abandonmenthas much bitterweed (Helenium tenuifolium); land that was plowed and abandonedhas hayfever weed (Ambrosia artemisiifolia) as a dominant plant species(scientificnamesof plantswere taken from Fernaid, 1950). Within 1-3 yearsthis initial variable stageis followedpredominantlyby white heath aster (Aster ericoides). The aster stageis replacedafter a year or two by broom sedge (Andropogonvirginicus) grassland. Five or moreyearslater the broomsedgedominancegraduallybeginsto yield to an early tree stagecharacterizedby persimmon(Diospyrosvirginiana), sassafras(Sassafrasalbidum), or wingedelm ( Ulmusalata) (Hite, 1960; Bullington, 1961). This stage can persist over 40 years (Hite, 1960). On favorablesites, the early tree stage developsinto forest dominated by white oak (Quercusalba) and black oak (Q. velutina), the climax community in the region (Cozzens, 1940). On xeric sites the early tree stagebecomespost oak (Q. stellata) forest,which ultimately transforms to climax forest as moisture conditions improve with soil buildup (Turner, 1935).

A glade with red cedar (Juniperusvirginiana) developson limestone (Steyermark, 1940; Quarterman, 1950), which is the pioneer woody stage on outcroppings(Bellman and Brenner, 1951). Cedar glades slowly transform to post oak subclimax (Kucera and Martin, 1957).

We thank the administrationand staff of the Pea Ridge National Military Park for their aid and for providing protected study plots. Travel to and from the study plots was financed in part by a grant to the seniorauthor from the JosselynVan Tyne Memorial fund of the American Ornithologists'Union.

62

The Auk 90: 62-77. January 1973

January1973]

BreedingBird Populations

63

CLIMAX FOREST STAGE

Figure 1. Ecologicalsuccessionof upland plant communitiesin northwestern

Arkansas.

METHODS

Ten plotsrangingfrom 4 to 24 acresand averaging16 acreswerechosenfor uniformityand for representativenoefsusplandsubserseuccessioTnr.ee-ringdata were obtainedwith incrementborersto determinestudy plot agessinceagricultural abandonment.Plots also were dated by Ivan Tolley and Alvin Seamster,who provided information about past cultivation and land abandonment.

As the plotsvariedin physiognomiyt, wasnecessartyo useseveravl egetationsamplingtechniquedsependinogn the life formsof the plants.A sightingtube (Winkworth and Goodall,1962) sampledgroundcover in the field plots. In foreststhe first 20 groundplantslessthan 4.5 feet tall were identifiedalonga

64

SHUGART^NV JAMES

[Auk, Vol. 90

line having a random starting point and a random direction (Baur, 1943). Understorystems (4 3 inchesDBH and ?> 4.5 feet tall) were countedand identified in 0.01-acre,paced,arm-lengthrectangles(Rice and Penfound,1955). Trees (?> 3 inchesDBH) were countedand identifiedin 0.1-acrerectangularsamplesand were grouped into 3-inch size classes(Diameter Breast High). Canopy density in the forest plots was determinedby noting if canopy cover was present or absent at 20 randompoints. In fact all samplesitesin eachhabitat were locatedrandomly.

The bird populationws erecertsuseudsingthe territorymappingmethodof Williams (1936, 1947). At leastsix early morningcensusewsere taken on eachplot during April and May 1967. Care was taken to enter each plot from different sides eachtime and to visit eachplot at differenttimesin the morning. Two night

counts of nocturnal birds were made on all plots.

One-acregrids were usedfor territory mapping. Bird densitieswere converted to number of territorial males per 100 acres. Increasedaccuracyin determining populationdensitieson the smallerplots was obtainedby analyzingfractionsof

bird territories.

THE STm)? P?oTs

The Ozark Highlandscontainno large expansesof uniform habitat representingall stagesof secondarysuccessionT. his is due in part to the fact that land abandonmenptracticesare not randomthroughtime, but are influencedby climaticeventssuchas droughtsg, overnmental policiessuchas the soil bank program,and, in the caseof the present study,historicael ventssuchasthe Civil War, the GreatDepressiona,nd the comingof the railroads.Alsothe variabilityof successioanccording to the bedrockin the Ozarks (Cozzens,1940) makesit impossibleto searchlargeregionsfor suitablestudyplots. Theseconstraintlsimit any successionsatul dyin the OzarkHighlandsto smallerstudyplots than onewouldusein other areas. We recognizethe shortcomingosf using smallstudyplots,but feel that not conductingstudiesin areasthat are difficult to sampleis the lessViable alternative.

Bird populationosn land prior to abandonmenwtere studiedon a 16~acremowedhayfieldplot with no woodyvegetation.This plot's dominantplant was blue grass(Poa sp.).

A 16-acreburnedfield plot representetdhe initial variablestageof successionT.his formerfield of broomsedgeand sassafrasaplingshad burnedthe winter beforethe study. Most broomsedgeand sassafras wasdestroyebdutsomedeadsassafrasstemstillstood4-5 feettall and

were used as song perches. A 16-acreplot in a broom sedgefield was establishedfor the next

stageof successionB.lackberry(Rubussp.) and persimmonwereimportant woodyplants on this grasslandplot.

The earlytreestagewasrepresentebdy two studyplots. The 16-acre clonalpersimmopnlot wasan earlyphaseof the early treestage.Shade

January 1973]

Breeding Bird Populations

65

intolerant speciesformedgroupsof small trees separatedby expanses of broom sedgeplus other herbaceousspecies. Tree density was 122 trees per acre, stem density 322 stems per acre. Within the clonal clumpsof treesthegroundcoverwasrelativelysparse.Overallthe canopy was 83 percent open.

A woody field plot representeda later phasein the early tree stage. This was a 4-acre plot in which the separatedclonal clumpsof shade intoleranttrees,primarily sassafrash,ad expandedto the point of barely joining and thus were totally excludingthe grasses.A well-developed understorywith 4,550 stemsper acre was composedprimarily of winged sumac (Rhus copallina). The canopy was 75 percent open and there were 210 trees per acre.

The cedargladestageof successiownas studiedon a 4-acreplot. The soil was shallowand on limestone. Red cedar predominatedin both the tree and shrublayers,but the presenceof post oak in both layers demonstratedthe connectionbetweenthis stageand the post oak forest. Broom sedgewas the principal ground cover, but was accompaniedby several other species.With 126 treesper acre the canopywas 53 percent open; the understoryhad 270 stemsper acre.

A 7-acre forest edgeplot was includedin the study. This plot was bisectedby the junction of a post oak forest and a broom sedgefield. The junction included a denseunderstoryof winged sumac and blackberry.

A xeric foreststudyplot covered24 acresof post oak subclimaxforest. Post oak predominated,but the presenceof large and small white oaks and black oaks linked this stage to later successionasl tages. Ground coverwassparseand the understorylayer waswell-developed(730 stems per acre). The canopy was 36 percent open with 367 trees per acre.

A mesic forest plot of 7 acres representedthe climax forest. There were no dominantsin any of the three vegetationlayers, but white oak and black oak wereimportant trees. The plot had beenin standingtimber sincethe time of the Civil War. The understorylayer was sparse(223 stemsper acre). The canopywas 26 percent open with 193 trees per

acre.

RESULTS AND DISCUSSION

UPLAND AvIAN SUBSERE AND SUCCESSION

Upland successionin the region of the study involved three general stages:fields dominatedby grassesand forbs, fields dominatedby shrubs and shadeintolerant trees,and forestsdominatedby trees. These general stageseach have distinctivebreedingbird species(Table 1).

66

Suvc^s:r^?) J^?rs

[Auk, Vol. 90

i i

January 1973]

Breeding Bird Populations

67

68

SI-IUOARAT i?D JAMES

[Auk, Vol. 90

I I

January 1973]

BreedingBird Populations

69

GrasshoppeSr parrowsand Eastern Meadowlarks,characteristicof grasslandsw, ere found in both the mowed field and the burned field plots (Table 1). As the fieldswereinvadedby shrubsand smallertrees in the broomsedgefield plot and to a greaterextentin the two plots representativeof the early tree stage, the Field Sparrowbecamethe typicalbird (Table 1). Althoughnot foundat suchhigh densitiesas the Field Sparrow,the Bell's Vireo also seemedto be typical of this later field habitat. The two plots in the early tree stageof successiohnad a large complementof specieswith the Field Sparrow,Yellow-breasted Chat, and American Goldfinchbeing the most abundant. The BluewingedWarbler,Blue GrosbeakP, aintedBunting,and Mockingbirdwere lessnumerousbut typical. The woody field plot that was late in the early tree stagealsohad the Rufous-sidedTowbeeand Brown Thrasher associatedwith the densesitnvadingvegetation.Cardinalswere common in this latter habitat too, but also occurredin large numbersin the forest plots and thus could not be consideredexclusive to the field-toforest transition (Table 1). The cedar glade had several of the bird speciesassociatewd ith the early tree stageof successiobnu, t alsohad a goodcomplementof forest birds (Table 1).

The forestedplotssupportedtwo generalgroupsof birds: speciesthat

occurred in all forested habitats and those that were confined to the

mesic forest. The former were the Tufted Titmouse, Blue-gray Gnatcatcher, Carolina Chickadee, Eastern Wood Pewee, Red-bellied Woodpecker; the latter includedAcadian Flycatcher, Wood Thrush, WormeatingWarbler, Parula Warbler, CeruleanWarbler, Ovenbird,Kentucky Warbler,andAmericanRedstart(Table 1). The restrictedspeciegsenerally werelessplentifulin the mesicforestthan the wide-rangingspecies also presentthere.

'Certain speciesi,ncludingthe Cardinal,Bobwhite,MourningDove, and Blue Jay, were not limited to any singlemajor successionsatlage.

SUCCESSION AND BIRD POPULATION DENSITY

A generalincreasein avian speciesand density throughprogressive successioncaol mmunitiestowardclimaxvegetationwas foundby Saunders (1936) in New York, Kendeigh (1948) in Michigan, Karr (1968) in Illinois,Odum (1950) in North Carolina,Johnstonand Odum (1956) in Georgia,and Haapanen (1965) in Finland. This representssimilar findingsin diverseregionsincludingboth deciduousforestand coniferous forestclimax. Evenso,Kendeigh(1946) foundin the deciduous-coniferous forest ecotonethat the highestbird densitiesoccurredin the shrubby seralstagerather than in climax forestand Karr (1968) noted a decline in bird speciesand density in the last forest stage.

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download