Printer Friendly

Composition and structure of a mixed-hardwood bottomland forest in the West Cross Timbers of north-central Texas.

Hardwood bottomland forests are some of the most widely distributed, biologically diverse, and productive of tree-dominated communities throughout southern regions of North America (Braun, 1964; Messina and Conner, 1998; Baker et al., 2004). Diamond et al. (1987) reported that bottomland forests dominated by sugarberry (Celtis laevigata variety laevigata), cedar elm (Ulmus crassifolia), and pecan (Carya illinoinensis) comprised the most widely distributed forests in Texas. These forests have been highly modified by humans, including flood-control projects. Increasing populations of invasive, exotic plants threaten structure and function of southern forests (Miller, 2003). Remaining parcels of floodplain forests that have less anthropogenic impacts typically are small and irregularly shaped. Currently, there is considerable interest in preserving remaining tracts of less-altered bottomland forests.

Description of the natural vegetation is an important phase of preservation and management of riparian and floodplain forests. This includes vegetational analysis to determine extents to which native forests have been invaded by nonnative plants. To date, there has been limited description and analysis of such communities. Eyre (1980) included limited qualitative coverage of these forests in descriptions of forest cover types, but this treatment largely ignored the understory. Although lower layers of bottomland-hardwood forests are valuable as natural pastures, descriptions of rangeland cover types (Shiflet, 1994) have excluded most types of forested range. This avoided redundancy but resulted in relatively scant treatment of grazable-browsable layers of these cover types, especially hardwood forests.

There are several cover types in bottomland forests in the southern region that have relatively open canopies and well-developed lower layers of vegetation including two recognized by Eyre (1980): sugarberry-American elm-green ash (C. laevigata-Ulmus americana-Fraxinus pennsylvanica) and sycamore-sweetgum-American elm (Platanus occidentalis-Liquidambar styraciflua-U. americana). Sycamore-sweetgum-American elm previously was designated as sycamore-pecan-American elm (P. occidentalis-C. illinoinensis-U. americana; Eyre, 1954). Descriptions of these cover types (Eyre, 1954, 1980) included little consideration of herbaceous components of the understory and presented no quantitative information.

Earlier descriptive studies and subsequent qualitative reports concerning southern-floodplain forests (Diamond et al., 1987; Meadows and Stanturf, 1997; Twedt and Best, 2004; Lockhart and Kellum, 2006; Twedt et al., 2010) showed the widespread sugarberry-elm-pecan type to be highly variable in composition, especially where it is ecotonal to adjacent cover types. None of these descriptions included quantitative measures of vegetation. Also, the published qualitative descriptions of vegetation for sugarberry-American elm-green ash and sycamore-sweetgum-American elm (Eyre, 1980) and natural alliances of forests (Hoagland, 2000) are limited. This general forested community in Texas was described variously as elm-hackberry parks-woods (McMahan and Frye, 1987; C. A. McMahan et al., in litt.), sugarberry-elm series (Diamond et al., 1987; D. D. Diamond, in litt.), sugarberry-cedar elm temporarily flooded forest, in part (A. S. Weakley et al., in litt.), sugarberry-elm floodplain forest (Bezanson, 2000), sugarberry-cedar elm, calcareous-floodplain forest (NatureServe, http://www., and central-Texas, floodplain, hardwood forest (L. Elliott, in litt.). The most applicable classification could be the C. illionensis-C. laevigata alliance given by Hoagland (2000) for Oklahoma. Braun (1964) and Shelford (1963) generally described this bottomland vegetation as associated with other floodplain-forest communities.

Descriptions of grazable aspects of woodlands in eastern and southern regions of the United States commonly have been in conjunction with soil surveys under leadership of the Natural Resources Conservation Service (Soil Conservation Service, 1967, 1976; Natural Resources Conservation Service, 2003). Prior to the current Natural Resources Conservation Service (2003) handbook, ecological units of forested range were treated as woodland-suitability groups in soil surveys with primary emphasis on trees, canopy, and production of wood. Woodland-suitability groups currently comprise most available descriptions of hardwood bottomland forests in private-lands states. Expanded descriptions of grazable woodlands are written as forest-land ecological sites (Natural Resources Conservation Service, 2003). Descriptions of forest-land ecological sites require greater detail regarding vegetation of forests, including that of the understory. Likewise, classification of natural communities such as forest alliances and series (Diamond et al., 1987; McMahan and Frye, 1987; Bezanson, 2000; Hoagland, 2000), primarily have been descriptive with limited quantitative information provided.

We conducted our study to provide quantitative knowledge for descriptions and analyses of sugarberry-elm-pecan bottomland forest, a variant of a forest cover type (or a composite of two forest cover types) and forest alliance widely distributed in south-central and southeastern USA. Our results were intended to be used for descriptions of forest cover types, ecological or range sites, alliances, natural-riparian-floodplain vegetation, and similar natural units. A current pressing need for quantitative data of this vegetation, which is lacking for much of Texas (Diamond et al., 1987), is in classification of natural communities of plants. Additionally, such data are essential for on-going ground-truthing efforts (L. Elliott, in litt.). Bezanson (2000) noted that sugarberry-elm-pecan bottomland forests were widespread in Texas. This forest cover type is valuable as grazable woodland (Bedell, 1998) that provides multiple uses (grazing by livestock, protection of watersheds, biodiversity, aesthetics and recreation, habitats for wildlife and fish). Our investigation provided the first quantitative data for sugarberry-elm-pecan bottomland forests, including features of the herbaceous layer, in the West Cross Timbers of Texas. It also quantified establishment of introduced invasive plants (Miller, 2003) into this forest cover type.

Materials and Methods--The study area was within the West Cross Timbers ecoregion (29c on the ecoregion map of G. Griffith et al., in litt.) and Cross Timbers and Prairies vegetational area (Correll and Johnston, 1979) in Erath County, Texas. It was a tract of relict (Daubenmire, 1968; Zerbe, 1998), mixed-hardwood, bottomland forest with an herbaceous understory on a portion of primary floodplain of the Bosque River. Other than five species of invasive woody plants that comprised ca. 13.5% of relative importance, this climax forest was a natural area (Helms, 1998) evaluated as a reference area (Laycock, 1975) for potential natural vegetation (Zerbe, 1998). This tract was bisected by the Bosque River so that there was one portion north and another portion south of the river. The locality for the investigation was in the eastern city limits of Stephenville, Texas (32.22360[degrees]N, 98.19935[degrees]W). Based on the modified Koppen system, climate was mesothermal in all years, dry season in winter, with occasional desert years (Russell, 1945). Precipitation zone was subhumid, elevation was 402 m, and soil was the Bunyan series (fine-loamy, mixed, nonacid, thermic typic ustifluvents; Soil Survey Staff, classification/osd/index.html). Neither a woodland-suitability group nor a current ecological description was in the county soil survey (Soil Conservation Service, 1973).

This tract had not been subjected to grazing by livestock for [greater than or equal to] 50 years. A resident herd of white-tailed deer (Odocoileus virginianus) occurred in the study area. There was evidence of past browsing (felling and girdling) by American beavers (Castor canadensis), but the most recent feeding had occurred ca. 4years prior to sampling of vegetation. Eastern cottontails (Sylvilagus floridanus) and eastern fox squirrels (Sciurus niger) were other mammals that likely had some herbivorous influences (e.g., leaf-feeding, bud-feeding, and nut-planting). Wild turkeys (Meleagris gallopavo) also were present.

We used the step-point method (Costello and Schwan, 1946; Brown, 1954; Evans and Love, 1957; Cook, 1962; Cook and Stubbendieck, 1986; Bonham, 1989) to determine composition of species from foliar cover. Plants were sampled randomly with a sharp-pointed copper pin and total and relative numbers of hits were recorded. We sampled a total of 3,494 points. We estimated peak standing crop in November by clipping all herbaceous material (ground level) rooted in 0.09-[m.sup.2] plots followed by weighing oven-dried (100[degrees]C) herbage (Bonham, 1989). There was a total of 50 plots on the larger north portion and 30 plots on the smaller south portion. Peak standing crop consisted of broadleaf woodoats (Chasmanthium latifolium)at grain-ripe stage, dead herbage of Canada wildrye (Elymus canadensis) with intact spikes, autumn growth of Canada wildrye, dead herbage of annual grasses and forbs, and early autumn growth of cool-season annual grasses and forbs.

For woody vegetation, we used two rectangular quadrants, 25 by 100 m and 15 by 100 m, with the longest dimension parallel to the river bank as described by Ford and Van Auken (1982) and Wood and Wood (1988, 1989). The smaller quadrant was necessary for use at the outer limit where the forest adjoined an old field. The two areas sampled were <300 m apart, with one on the north side and the other on the south side of the Bosque River. Species of plants were identified and classified using Diggs et al. (1999), which also served as the reference for common and scientific names. We deposited voucher specimens in the herbarium at Tarleton State University in Stephenville, Texas.

We identified all woody plants in each quadrant and measured diameter at breast height (dbh) of all that were >1.0 cm. The dbh was used to calculate basal area. We calculated density (plants/ha), dominance (basal area/ha), and relative-importance values as described by Ford and Van Auken (1982) and Wood and Wood (1988, 1989).

RESULTS--The communities had three well-developed layers: mature tree with widely to moderately open canopy, sporadic small tree-shrub, and more or less continuous herbaceous understory. Woody vines frequently grew into crowns of trees, thereby extending through all layers of vegetation. The three-layer structure of this bottomland forest was consistent throughout, although species composition varied locally (Tables 1 and 2).

Of the number of herbaceous species, nine were native and five were introduced (Table 1). Native perennial grasses dominated herbaceous vegetation in the floodplain. The most abundant herbaceous species were Canada wildrye and broadleaf woodoats (Table 1). Two forbs, giant ragweed (Ambrosia trifida variety texana) and pigeonberry (Rivina humilis), also were relatively common and generally more abundant than other forbs such as frostweed (Verbesina virginica) and hedge-parsley (Torilis arvensis). Introduced grasses and forbs were only 4.0% of total hits (Table 1).

There was a total of eight species of shrubs of which six were woody vines. Three were nonnative and invasive. The most common species of shrub was Japanese honeysuckle (Lonicera japonica), a naturalized invasive exotic that was four times more plentiful than the native poison oak-ivy (Toxicodendron radicans) and saw greenbriar (Smilax bona-nox), the next most-abundant shrubs (Tables 1 and 2). Other native woody vines included Virginia creeper (Parthenocissus quinquefolia), trumpet-creeper (Campsis radicans), and mustang grape (Vitis mustangensis). The bottomland herbage (mean peak standing crop; oven-dry biomass) was 1,597 [+ or -] 1,769 kg/ ha. The two native perennial grasses comprised 58% of the herbaceous standing crop (926 kg/ha).

There were 13 species of trees in this forest of which two were nonnative and invasive. For all woody species >1.0 cm in diameter, sugarberry had the highest relative-importance value (56.0%) and greatest dominance (10,024.4 [m.sup.2]/ha). Sugarberry also had the greatest relative cover <1.0 cm (1.3%) of any species of tree. In all measured attributes, sugarberry was the most common tree. After sugarberry, the most common trees >1.0 cm overall were pecan, bois d'arc (Maclura pomifera), and cedar elm with relative-importance values of 9.0, 7.0, and 6.0%, dominance values of 1,577.5, 620.1, and 146.0 [m.sup.2]/ ha, and with percentage hits <1.0 cm of 0.1, 0, and 0.9, respectively. Tree-of-heaven (Ailanthus altissima), one of the invasive nonnative species, had the same relative-importance value (6.0%) as cedar elm, but lower dominance (46.9 [m.sup.2]/ha) than pecan, while bois d'arc had a higher relative-importance value (7.0%) than cedar elm. There was no hit for bois d'arcs <1.0 cm in diameter.

Pecan had the largest trees as shown by dominance (basal area of 1,577.5 [m.sup.2]/ha) and comparatively low density (42.5 plants/ha). In contrast, cedar elm had lower dominance (basal area of 146.0 [m.sup.2]/ha) yet higher density (82.5 plants/ha). Pecan had greatest proportion of mature-to-senescing individuals as evident by numerous dead and dying limbs, as well as lowest proportion of plants <1.0 cm in diameter among the common species of trees (Table 1). Tree-of-heaven had lower dominance than sugarberry, pecan, or cedar elm with higher density (95.0 plants/ha) than pecan and about the same density as cedar elm. Bois d'arc had third-highest dominance (less than sugarberry and pecan) and fourth-highest density (behind sugarberry, tree-of-heaven, and cedar elm), but the lowest composition (0%) of species measured as plants <1.0 cm in diameter (Table 1).

Discussion--Quantitative data for herbaceous and woody vegetation in mixed-hardwood, bottomland forest in the West Cross Timbers, including nonnative species, are provided for the first time. Comparison of herbaceous and woody vegetation to other studies in the region, possible successional status within the bottomland forest, invasion by introduced species, and the use and value of bottomland forests are discussed.

Herbaceous Vegetation--Previous investigations, as well as qualitative descriptions of bottomland forests (Bush and Van Auken, 1984; Diamond et al., 1987; McMahan and Frye, 1987; Bezanson, 2000; D. D. Diamond, in litt.; C. A. McMahan et al., in litt.; A. S. Weakley et al., in litt.), generally placed less emphasis on shrubs and provided little data and analysis of herbaceous layers of bottomland-hardwood forests. By contrast, our investigation of a sugarberry-cedar elm-pecan forest along the Bosque River emphasized the understory of the relict vegetation (Daubenmire, 1968) of this natural area (Helms, 1998) because these vegetational layers were of obvious value for grazing and browsing animals, constituting forest range or grazable woodland (Bedell, 1998).

Density and dispersion of trees combined with the lianas and relatively low number of seedlings and saplings of trees formed a canopy sparse enough for development of an herbaceous understory dominated by native perennial grasses and of relatively high biomass. Sugarberry-cedar elm-pecan forests are valuable for grazing because they represent the most widespread forest cover type over a large region although this vegetation typically is limited in expanse (Bezanson, 2000).

Herbaceous layers of the forest along the Bosque River were similar to those reported for bottomland forests of north-central Oklahoma (Rice, 1965) where wingstem (Verbesina alternifolia), instead of frostweed and Virginia wildrye (Elymus virginicus) instead of Canada wildrye, was reported. Rice (1965) listed more species of shrubs that included more forested communities, some of which were in areas of higher precipitation than the forested floodplain described herein. Nixon et al. (1991) reported about the same number of shrubs for a creek-bottom forest, but most of these shrubs, other than woody vines, were absent from the forest adjacent to the Bosque River.

Canada wildrye and broadleaf woodoats were herbaceous co-dominants in the forest we studied. These two species were listed as some of the most common grasses in the potential natural vegetation of bottomland-hardwood forests (J. Shedd et al., in litt.). They were viewed as dominants in late-seral to climax vegetation along streams and floodplains throughout much of Oklahoma (Tyrl et al., 2008) and Texas (Gould, 1975). Occurrence of charming caric sedge (Carex blanda) as the common grasslike plant (Table 1) suggested similarity of this bottomland forest to a pecan-cedar elm-sugarberry forest along a meander of the Brazos River (A. Lovelace et al., in litt.) in which Cherokee caric sedge (C. cherokeensis) was the common grasslike plant.

Dominant forbs, based on a combination of percentage of hits and biomass at peak standing crop, varied locally among giant ragweed, pigeonberry, and frostweed. Giant ragweed, the most common annual forb, was most abundant on local areas along the riverbank recently disturbed by flooding, where it grew in exclusive local colonies. This was the response ofa pioneer or colonizing species and a phenomenon widely documented for giant ragweed (Hartnett et al., 1987; Bazzaz, 1996) including formation of pure stands (Abul-Fatih and Bazzaz, 1979). Allelochemicals from giant ragweed were hypothesized (Stoller and Wax, 1973) and demonstrated to reduce growth of neighboring species of plants (Rasmussen and Einhellig, 1979). This may have accounted for local dominance by this annual forb along the Bosque River.

Perennial forbs (primarily pigeonberry and small stands offrostweed) were most common on environments that had not been disturbed recently and that had less cover of perennial grasses. Frostweed usually has been viewed as a species typical of middle stages of succession (Tyrl et al., 2008), but a study (Thompson and McKinney, 2006) of various successional stages in a river bottomland forest in the Nashville Basin of Tennessee with similar composition and structure to the one reported herein reported frostweed to be associated with higher seral states. On the bottomland forest described herein, frostweed often was associated with the cool-season perennial grasses, although in more open spots within the herbaceous layer.

Frostweed, pigeonberry, and broadleaf woodoats were important herbaceous species in a cedar elm-pecan-sugarberry forest that developed along the Brazos River (A. Lovelace et al., in litt.). Frostweed was listed as a common forb for pecan-sugarberry and cedar elm-sugarberry forests in central and southern Texas (NatureServe, A similar species, wingstem, was listed as one of several common forbs in the herbaceous layer in two similar bottomland forests (R. A. Chastain et al., in litt.). The importance of wingstem in these communities was consistent with occurrence of frostweed as a common forb on the sugarberry-cedar elm-pecan forest and related forests.

Woody Vegetation--Patterns of density and importance values for sugarberry, cedar elm, and pecan along the Bosque River corresponded closely to those reported for similar forested communities in this region. Composition and structure of the forest along the Bosque River was similar to various woody communities in the Trinity River basin of northern Texas as described by Nixon et al. (1990) and an old-growth bottomland forest along a creek in northeastern Texas (Nixon et al., 1991). In these forests, as in the one that developed on the floodplain of the Bosque River, sugarberry and cedar elm were consistently common or dominant trees. Nixon et al. (1991) reported that in a creek-floodplain forest, sugarberry and cedar elm had highest importance values and greatest densities of trees, whereas on slopes above the creek pecan had the greatest importance values. Bush and Van Auken (1984) also determined that sugarberry had greatest importance values and highest density of trees in a gallery forest on the floodplain of the San Antonio River.

Presence of such woody species as red mulberry (Morus rubra), American elm (U. americana), mustang grape, and chittamwood (Sideroxylon languinosum) showed similarity between floodplain forests of the Bosque and San Antonio rivers (Bush and Van Auken, 1984; Van Auken and Bush, 1985). Presence of native woody vines such as poison ivy, saw green-briar, Virginia creeper, and trumpet-creeper was similar to that on forests along the Brazos River (A. Lovelace et al., in litt.). These species also were characteristic shrubs on bottomland forests described for the Tamaulipan region (NatureServe, http://www. and the Ozark Plateau (R. A. Chastain et al., in litt.). Nixon et al. (1991) recorded poison ivy and Virginia creeper in an old-growth creek-bottom forest in the Texas Blackland Prairie, but they did not report green-briar, trumpet creeper, or the invasive Japanese honeysuckle.

Successional Status of the Bottomland Forest--We concluded that the sugarberry-cedar elm-pecan community was representative of the potential natural vegetation (Zerbe, 1998) or climax floodplain forest (Braun, 1964) for this ecological site. Composition and structure was consistent with traditional descriptions ranging from that of forest cover types (Eyre, 1980) to classification for conservation of plant communities (Diamond et al., 1987; Hoagland, 2000; D. D. Diamond, in litt.).

Quantitative studies of bottomland hardwood forests have been less common than qualitative analyses. Furthermore, quantitative investigations generally were limited to woody species as, for example, those of similar bottomland forests along the San Antonio River (Bush and Van Auken, 1984; Van Auken and Bush, 1985), Trinity River (Nixon et al., 1990), and Spring Creek in the Blackland Prairie (Nixon etal., 1991). Quantitative results herein closely resembled those in the few quantitative reports of similar bottomland forests. Barry and Kroll (1999) concluded that a sugarberry-cedar elm-green ash (Fraxinus pennsylvanica) bottomland forest was the climax community for a similar bottomland habitat. Nixon et al. (1991) also reported that green ash was a dominant tree along with species of elm, pecan, sugarberry, and oak. Green ash was a minor species in the relict forest of our study (Table 2). Instead, pecan was the third-most common species. Pecan was not reported as being important by Barry and Kroll (1999). Holcomb (2001) conducted studies following those of Barry and Kroll (1999) on the same tract and concluded that pecan was a dominant of the seral stage that preceded the sugarberry-elm-green ash community. This was consistent with the sugarberry-cedar-elm-pecan community in which pecan was, perhaps, a subclimax species that persisted into the climax forest.

Holcomb (2001) used as the model for his successional scheme the qualitative and generic pattern that Hodges (1997) proposed for bottomland forests of the Atlantic and Gulf coastal plains. It was not clear how relevant the general coastal model of Hodges (1997) would be for quantitative analysis of forests in the inland Cross Timbers and Prairies physiographic province. The forest inventoried by Barry and Kroll (1999) and Holcomb (2001) was along Elm Fork of the Trinity River in the more moist East Cross Timbers, whereas the relict forest along the Bosque River was in the West Cross Timbers (Diggs et al., 1999; G. Griffith et al., in litt.). Some differences in composition between adjoining ecoregions would be expected. Studies of forests along the Trinity River (Barry and Kroll, 1999; Holcomb, 2001) did not include the herbaceous layer. A pecan-cedar elm-sugarberry forest (plant association) was described (A. Lovelace et al., in litt.) for the lower Brazos River in the Southern Post Oak Savanna of the East Central Texas Plains ecoregion (G. Griffith et al., in litt.). That appraisal was for wildlife habitats and did not include quantitative data for vegetation.

Van Auken and Bush (1985) studied secondary succession of a climax forest on terraces of the San Antonio River and explained that sugarberry and cedar elm became dominant in advanced stages of development of forests, whereas pecan established earlier in the successional sequence and persisted into the climax forest. These results were consistent with the relict forest on the floodplain of the Bosque River wherein these three species were the apparent climax dominants. Most revealing, perhaps, in all of these studies of alluvial forests that developed along streams in Texas was the consistent trend for pecan to establish earlier in succession, yet persist into advanced successional stages with lower rates of recruitment, whereas sugarberry appeared and established much later in succession.

This suggested that sugarberry was more tolerant with adaptation to lower light conditions. Bush and Van Auken (1986) concluded that sugarberry is a sciophyte. They reported that seedlings of sugarberry achieved maximum growth under conditions of low light. Bush and Van Auken (1995) described seedlings of sugarberry as being tolerant of shade. Fertilization trials by Van Auken et al. (1985) indicated that biomass of sugarberry increased dramatically with addition of nitrogen (although not other nutrients). A later experiment indicated that nitrogen in soil was more important than intensity of light on growth of sugarberry (Bush and Van Auken, 1995). In another trial, Van Auken and Lohstroh (1990) discovered that competition of roots between adults and seedlings of sugarberry was not important. These numerous experiments strongly suggested that establishment of sugarberry in late-seral and climax floodplain forests, such as that along the Bosque River, was the result of increasing nitrogen content of soil and, secondarily, to progressively greater shade, while competition was of minor importance.

We interpreted sugarberry as the primary dominant species with pecan and cedar elm the secondary dominants. General dominance by sugarberry and cedar elm was based on relative importance values of larger trees (Table 2), as well as general recruitment as observed and demonstrated by species composition of plants <1.0 cm in diameter (Table 1). Sugarberry had the greatest number of trees with higher proportions in smaller (and apparently) younger age classes (Table 1). This was followed by cedar elm. Pecan had the highest number of mature trees of large size (e.g., basal area). Bush and Van Auken (1984) reported similar relationships for a gallery forest on terraces of the San Antonio River, although they encountered some species of trees that were not present in the forest along the Bosque River. We concluded that sugarberry, cedar elm, and pecan were dominants based on density, basal area, and apparent crown cover of individuals with diameter >1.0 cm. Dominance of species varied locally. We determined that recruitment of species (seedlings and individuals <1.0 cm) was greater for sugarberry and cedar elm than for pecan (Table 1).

Pecan persisted into the climax vegetation as mature and senescing individuals (old trees with dying branches), while recruitment of pecan was lower than that of sugarberry and cedar elm. Limited recruitment of pecan as one of the dominant species was consistent with Rice (1965). It appeared that sugarberry and cedar elm were the climax dominants, while pecan was a member of the climax as a persistent subclimax species. Rate of recruitment of pecan was not determined in our static study, but reproduction appeared to be adequate to maintain pecan at existing proportions. Burns and Honkala (1990) categorized pecan as subclimax. On the tolerance rating of the Society of American Foresters (Eyre, 1954), pecan was classified as intolerant (Wenger, 1984; Burns and Honkala, 1990). Loucks and Keen (1973) reported that pecan was intermediate in submersion tolerance. Age-class distribution of pecan in the forest we studied was consistent with these interpretations.

We observed that in the West Cross Timbers ecoregion, size (basal area) of mature pecan typically exceeded that of cedar elm and, even more so, sugarberry (Table 2). Sugarberry and, to a lesser extent, cedar elm frequently are parasitized heavily by Christmas mistletoe (Phoradendron tomentosum). This could be a factor that limits size of sugarberry or, alternatively, sugarberry and cedar elm may have been harvested in the past for firewood. The most likely explanation for differences in size was that throughout this area sugarberry trees die before they attain the size of pecan. This was obvious from snags of sugarberry that invariably were smaller than live trees and snags of pecan. Whatever reason for this phenomenon, sugarberry had the greatest number of individuals of various sizes on the forest we studied followed by cedar elm.

Braun (1964) concluded that sugarberry was nearly ubiquitous in bottomland forests throughout the southcentral region, whereas cedar elm was more restricted, being confined primarily to western parts of this region. Kellison et al. (1998) listed sugarberry as present throughout southern bottomland forests, whereas pecan and cedar elm had more restricted distributions on these floodplains. Pecan was listed previously in the title of this forest cover type by the Society of American Foresters (Eyre, 1954). Pecan was not included in the title of the revised forest cover type (Eyre, 1980). Burns and Honkala (1990) recognized the sycamore-pecan-American elm association (the previous forest cover type designated by the Society of American Foresters; Eyre, 1954) as a variant of the sycamore-sweetgum-American elm forest cover type (Eyre, 1980). Persistence of large mature pecan together with limited, although adequate, reproduction appeared to justify inclusion of this species in a variant of either current forest cover types classified as 93 or 94 by the Society of American Foresters (Eyre, 1980), both of which have considerable variation in composition of species. This also was consistent with the C. illionensis-C. laevigata alliance (Hoagland, 2000).

In some restricted parts of the tract of forest that we studied, other species of trees had densities similar to those of overall dominants. This included the introduced tree-of-heaven, which was sometimes a local dominant. Invasion by this exotic weedy species appeared to pose some threat to native vegetation.

Invasion by Nonnative Woody Plants--Five species of nonnative woody plants had invaded the bottomland forest. Japanese honeysuckle was the most common exotic species in this forest. It frequently covered

areas of this bottomland community to the exclusion of all other vascular plants. This aggressive invasive species threatened all native plants in the understory of this forest. Japanese honeysuckle was recognized as a significant threat to forests of eastern Texas by the Texas Invasive Plant and Pest Council ( invasives), Texas Forestry Association (http://www., and United States Forest Service (Miller, 2003). Burns and Honkala (1990) described Japanese honeysuckle as often present in forests where pecan was a common species. Kellison et al. (1998) concluded that Japanese honeysuckle was widespread and almost always present on terraces and levees of rivers in southern-floodplain forests. These workers reported that Japanese honeysuckle was associated with native lianas including trumpet creeper, greenbriar, Virginia creeper, and poison ivy, all common species on the floodplain of the Bosque River. Flory and Clay (2009) noted that Japanese honeysuckle was one of two nonnative invasive species on weed-control plots in a deciduous forest where dominant understory grasses included wildrye and broadleaf woodoats.

Chinese privet (Lignustrum sinense) was the other exotic shrub in this relict tract of forest. Chinese privet was much less abundant than Japanese honeysuckle, but relative importance of privet was greater than that of the native mustang grape. Chinese privet has become a troublesome invader on southeastern forests (Miller, 2003). It was ranked by the Texas Forestry Association ( as a greater threat to forests in Texas than Japanese honeysuckle.

Tree-of-heaven was less common than Japanese honeysuckle and did not form exclusive ground cover. Nonetheless, this exotic tree had obvious influences on the forest as it grew to sizes larger than many sugarberry and cedar elm. The most obvious effects from tree-of-heaven were on structure and composition of stands. Changes in the native-forest community resulted primarily from proliferation of fast-growing shoots off of rhizomes of established trees. Rapid development of offshoots and alleochemicals from roots were reported (Kaufman and Kaufman, 2007) as important impacts of this exotic invasive on forested ecosystems. Tree-of-heaven apparently does not pose as serious a threat to forests in Texas as some other introduced woody species such as Japanese honeysuckle and Chinese privet, but this tree was described by Moser et al. (2009) as a nonnative invasive plant with adverse impacts on forests.

China-berry (Melia azedarach) had a lower relative-importance value than that of tree-of-heaven. Bush and Van Auken (1984) referred to China-berry as a rapid-growing species along the San Antonio River and a species associated with sugarberry and cedar elm (Van Auken and Bush, 1985). The invasive China-berry appeared to be another species that could threaten vegetation of the natural forest.

Cause of invasion by these alien woody species along the Bosque River was not determined. Richardson et al. (2007) explained that rivers were susceptible to invasion by alien plants because hydrologic dynamics and frequent disturbances of streams make them especially effective for dispersal of plant propagules. This phenomenon was likely a partial explanation for the relatively high cover of alien woody species in this bottomland forest. Bush and Van Auken (1984) commented that China-berry along with sugarberry and native colonizing species of trees likely became established following flooding.

Another plausible factor leading to invasion by these exotic plants was location of the floodplain vegetation adjacent to a town where all exotic species except bitter orange (Poncirus trifoliata), a minor woody species (Table 2), had been used as introduced landscape plants. None of these nonnative species was recognized as being desirable for landscape or gardens by the National Arboretum (Heriteau and Cathey, 1990), but all are domestic species (Bailey, 1949) that at one time were used locally in landscaping or, in case of bitter orange, as root stock for the citrus industry.

Although there was invasion of native bottomland forest by introduced woody species, presence of herbaceous species (including native forbs combined with production of mast and cover of trees) provided valuable habitat for wildlife. Invasive species had not achieved sufficient cover to reduce herbaceous biomass of the native perennial grasses that dominated the understory and provided a valuable forage crop for livestock.

Use and Value of Forest Vegetation--Relatively high foliar cover of herbaceous species along with well-developed canopy and diverse species of shrubs resulted in a forested community capable of providing numerous goods and services to the ecosystem. These included general reduction of damage due to flash flooding (including accelerated erosion of the stream channel), provision of high-quality forage for grazing by livestock, refugia for propagules of numerous native species of plants, and diverse habitats for wildlife. For example, presence of dense populations of native grasses and native forbs combined with production of mast and cover of trees afforded desirable environments for native vertebrates (e.g., white-tailed deer and wild turkeys).

We thank R. E. Sosebee, J. R. McBride, and R. H.Johnson for numerous suggestions and insights that contributed to this work. An anonymous reviewer provided critical references and numerous comments that added immensely to the paper.

Submitted 4 February 2011. Accepted 27 October 2012. Associate Editor was Florence M. Oxley.


ABUL-FATIH, H. A., AND F. A. BAZZAZ. 1979. The biology of Ambrosia trifida L. 1. Influence of species removal on the organization of the plant community. New Phytologist 83:813-816.

BAILEY, L. H. 1949. Manual of cultivated plants. Macmillan Publishing Company, Inc., New York.

BAKER, M. B., JR., P. F. FIOLLIOT, L. F. DEBANO, AND D. G. NEARY, EDITORS. 2004. Riparian areas of the southwestern United States: hydrology, ecology, and management. CRC Press, Boca Raton, Florida.

BARRY, D., AND A. J. KROLL. 1999. A phytosociological description of a remnant bottomland hardwood forest in Denton County, Texas. Texas Journal of Science 51:309-316.

BAZZAZ, F. A. 1996. Plants in changing environments--linking physiological, population, and community ecology. Cambridge University Press, Cambridge, United Kingdom.

BEDELL, T. E., CHAIRMAN. 1998. Glossary of terms used in range management. Society for Range Management, Denver, Colorado.

BEZANSON, D. 2000. Natural vegetation types of Texas and their representation in conservation areas. M.S. thesis, University of Texas, Austin.

BONHAM, C. D. 1989. Measurements for terrestrial vegetation. John Wiley and Sons, New York.

BRAUN, E. L. 1964. Deciduous forests of eastern North America (facsimile of 1950 edition). Hafner Publishing Company, New York.

BROWN, D. 1954. Methods of surveying and measuring vegetation. Commonwealth Bureau of Pastures and Field Crops Bulletin 42:1-223.

BURNS, R. M., AND B. H. HONKALA. 1990. Silvics of North America, volume 2, hardwoods. United States Department of Agriculture Forest Service, Agriculture Handbook 654:1-711.

BUSH, J. K., AND O. W. VAN AUKEN. 1984. Woody-species composition of the upper San Antonio River gallery forest. Texas Journal of Science 36:139-148.

BUSH, J. K., AND O. W. VAN AUKEN. 1986. Light requirements of Acacia smallii and Celtis laevigata in relation to secondary succession on floodplains of South Texas. American Midland Naturalist 115:118-122.

BUSH, J. K., AND O. W. VAN AUKEN. 1995. Interaction between seedlings of an early and a late successional woody species. Southwestern Naturalist 40:379-387.

COOK, C. W. 1962. Basic problems and techniques in range research: a report of a joint committee of the American Society of Range Management and the Agricultural Board. National Academy of Sciences-National Research Council, Washington, D.C., Publication 890:1-341.

COOK, C. W., AND J. STUBBENDIECK. 1986. Range research: basic problems and techniques. Society for Range Management, Denver, Colorado.

CORRELL, D. S., AND M. C. JOHNSON. 1979. Manual of the vascular plants of Texas. University of Texas, Richardson.

COSTELLO, D. F., AND H. E. SCHWAN. 1946. Conditions and trends on ponderosa pine range in Colorado. United States Department of Agriculture Forest Service, Rocky Mountain Forest and Range Experiment Station, Washington, D.C.

DAUBENMIRE, R. 1968. Plant communities: a textbook of plant synecology. Harper and Row Publishers, New York.

DIAMOND, D. D., D. K. RISKIND, AND S. L. ORZELL. 1987. A framework for plant community classification and conservation in Texas. Texas Journal of Science 39:203-221.

DIGGS, G. M., JR., B. L. LIPSCOMB, AND R. J. O'KENNON. 1999. Shinners' and Mahler's flora of north central Texas. Botanical Research Institute of Texas Press, Fort Worth.

EVANS, R. A., AND R. M. LOVE. 1957. The step-point method of sampling: a practical tool in range research. Journal of Range Management 10:208-212.

EYRE, F. H. 1954. Forest cover types of North American (exclusive of Mexico). Society of American Foresters, Washington, D.C.

EYRE, F. H., EDITOR. 1980. Forest cover types of the United States and Canada. Society of American Foresters, Washington, D.C.

FLORY, S. L., AND K. CLAY. 2009. Invasive plant removal method determines native plant community responses. Journal of Applied Ecology 46:434-442.

FORD, A. L., AND O. W. VAN AUKEN. 1982. The distribution of woody species in the Guadalupe River floodplain forest in the Edwards Plateau of Texas. Southwestern Naturalist 27:388-392

GOULD, F. W. 1975. Grasses of Texas. Texas A&M University Press, College Station.

HARTNETT, D. C., B. B. HARTNETT, AND F. A. BAZZAZ. 1987. Persistence of Ambrosia trifida populations in old fields and responses to successional changes. American Journal of Botany 74:1239-1248.

HELMS, J. A. 1998. The dictionary of forestry. Society of American Foresters, Bethesda, Maryland.

HERITEAU, J., AND H. M. CATHEY. 1990. The National Arboretum book of outstanding garden plants: the authoritative guide to selecting and growing the most beautiful, durable, and carefree garden plants in North America. Simon and Schuster, New York.

HOAGLAND, B. 2000. The vegetation of Oklahoma: a classification for landscape mapping and conservation planning. Southwestern Naturalist 45:385-420.

HODGES, J. D. 1997. Development and ecology of bottomland hardwood sites. Forest Ecology and Management 90:117-125.

HOLCOMB, S. S. 2001. An examination of the riparian bottomland forest in north central Texas through ecology, history, field study, and computer simulation. M.S. thesis, University of North Texas, Denton.

KAUFMAN, S. R., AND W. KAUFMAN. 2007. Invasive plants: guide to identification and the impacts and control of common North American species. Stackpole Books, Mechanicsburg, Pennsylvania.

KELLISON, R. C., J. J. YOUNG, R. R. BRAHAM, AND E. J. JONES. 1998. Major alluvial floodplains. Pages 291-323 in Southern forested wetlands: ecology and management (M. G. Messina and W. H. Conner, editors). CRC Press, Boca Raton, Florida.

LAYCOCK, W. A., CHAIRMAN. 1975. Rangeland reference areas. Society for Range Management, Denver, Colorado.

LOCKHART, B. R., AND J. E. KELLUM. 2006. A complex stand on the White River National Wildlife Refuge: implications for bottomland hardwood old growth. Journal of the Arkansas Academy of Science 60:81-184.

LOUCKS, W. L., AND R. A. KEEN. 1973. Submersion tolerance of selected seedling trees. Journal of Forestry 71:496-497.

MCMAHAN, C. A., AND R. G. FRYE. 1987. Bottomland hardwoods in Texas. Proceedings of an interagency workshop on status and ecology, May 6-7, 1986. Texas Parks and Wildlife Division PWD-RP-7100-133:1-170.

MEADOWS, J. S., AND J. A. STANTURF. 1997. Silvicultural systems for southern bottomland hardwood forests. Forest Ecological Management 90:127-140.

MESSINA, M. G., AND W. H. CONNER. 1998. Southern forested wetlands. CRC Press, Boca Raton, Florida.

MILLER, J. H. 2003. Nonnative invasive plants of southern forests: a field guide for identification and control. United States Department of Agriculture Forest Service, Southern Research Station, Asheville, North Carolina GTR-SRS-62:1-93.

MOSER, W. K., E. L. BARNARD, R. F. BILLINGS, S. J. CROCKER, M. E. DIX, A.N.GRAY, G.G. ICE, M. S. KIM, R. REID, S. U. RODMAN, AND W. H. MCWILLIAMS. 2009. Impacts of nonnative invasive species on US forests and recommendations for policy and management. Journal of Forestry 107:320-327. NATURAL RESOURCES CONSERVATION SERVICE. 2003. National range and pasture handbook. United States Department of Agriculture Natural Resources Conservation Service, Grazing Lands Technology Institute, Washington, D.C.

NIXON, E. S., J. R. WARD, E. A. FOUNTAIN, AND J. S. NECK. 1991. Woody vegetation of an old-growth creekbottom forest in north-central Texas. Texas Journal of Science 43:157-164.

NIXON, E. S., G. A. SULLIVAN, S. D. JONES, G. D. JONES, AND J. K. SULLIVAN. 1990. Species diversity of woody vegetation in the Trinity River basin, Texas. Castanea 55:97-106.

RASMUSSEN, J. A., AND F. A. EINHELLIG. 1979. Allelochemic effects of leaf extracts of Ambrosia trifida (Compositae). Southwestern Naturalist 24:637-644. RICE, E. L. 1965. Bottomland forests of north-central Oklahoma. Ecology 46:708-714.

RICHARDSON, D. P., M. HOLMES, K. J. ESLER, S. M. GALATOWITSCH, J. C. STROMBERG, S. P. KIRKMAN, P. PYSEK, AND R. J. HOBBS. 2007. Riparian vegetation: degradation, alien plant invasions, and restoration prospects. Diversity and Distributions 13:126-139.

RUSSELL, R. J. 1945. Climates of Texas. Annals of the Association of American Geographers 35:37-52. SHELFORD, V. E. 1963. The ecology of North America. University of Illinois Press, Urbana.

SHIFLET, T. N. 1994. Rangeland cover types of the United States. Society for Range Management, Denver, Colorado.

SOIL CONSERVATION SERVICE. 1967. National handbook for range and related grazing lands. United States Department of Agriculture Soil Conservation Service, Washington, D.C.

SOIL CONSERVATION SERVICE. 1973. Soil survey of Erath County, Texas. United States Department of Agriculture Soil Conservation Service, Washington, D.C.

SOIL CONSERVATION SERVICE. 1976. National range handbook--rangeland, grazable woodland, native pasture. United States Department of Agriculture Soil Conservation Service, Washington, D.C.

STOLLER, E. W., AND L. M. WAX. 1973. Periodicity of germination and emergence of some annual weeds. Weed Science 21:574-580

THOMPSON, R. L., AND L. E. MCKINNEY. 2006. Vascular flora and plant habitats of an abandoned limestone quarry at Center Hill Dam, DeKalb County, Tennessee. Castanea 71:54-64.

TWEDT, D. J., AND C. BEST. 2004. Restoration of floodplain forest for the conservation of migratory landbirds. Ecological Restoration 22:194-203.

TWEDT, D. J., S. G. SOMERSHOE, K. R. HAZLER, AND R. J. COOPER. 2010. Landscape and vegetation effects on avian reproduction on bottomland forest restorations. Journal of Wildlife Management 74:423-436.

TYRL, R. J., T. G. BIDWELL, R. E. MASTERS, AND R. D. ELMORE. 2008. Field guide to Oklahoma plants: commonly encountered prairie, shrubland, and forest species. Oklahoma State University, Stillwater.

VAN AUKEN, O. W., AND J. K. BUSH. 1985. Secondary succession on terraces of the San Antonio River. Bulletin of the Torrey Botanical Club 112:158-166.

VAN AUKEN, O.W., AND R. J. LOHSTROH. 1990. Influence of canopy position for growth of Celtis laevigata seedlings. Texas Journal of Science 42:83-89.

VAN AUKEN, O. W., E. M. GESE, AND K. CONNORS. 1985. Fertilization response of early and late successional species: Acacia smallii and Celtis laevigata. Botanical Gazette 146:564-569.

WENGER, K. F., EDITOR. 1984. Forestry handbook. John Wiley and Sons, Inc., New York.

WOOD, C. E., AND J. K. WOOD. 1988. Woody vegetation of the Frio River riparian forest, Texas. Texas Journal of Science 40:309-321

WOOD, C. E., AND J. K. WOOD. 1989. Riparian forests of the Leona and Sabinal rivers. Texas Journal of Science 41:395-412.

ZERBE, S. 1998. Potential natural vegetation: validity and applicability in landscape planning and nature. Applied Vegetation Science 1:165-172.


Department of Agribusiness, Agronomy, Horticulture, and Range Management, Tartelon State University, Stephenville, TX 76402 (RER)

Department of Biological Sciences, Tarleton State University, Stephenville, TX 76402 (AND, LPC) * Correspondent:
Table 1-Species composition of herbaceous and woody plants
([less than or equal to]1 cm diameter) of mixed-hardwood
bottomland forest along the Bosque River, Erath County,
Texas, as determined by the step-point method.
Asterisks indicate introduced species.

Category                                                of hits (%)


Broadleaf woodoats (Chasmanthium latifolium)            754 (21.6)
Canada wildrye (Elymus canadensis)                      1,273 (36.4)
Cheatgrass brome (Bromus tectorum variety tectorum) *   1 (<0.1)
Italian ryegrass (Lolium perenne subspecies             8 (0.2)
  multiflorum) *
Nimblewill (Muhlenbergia schreberi)                     32 (0.9)
Rescuegrass (Bromus catharticus) *                      45 (1.3)
Total grasses                                           2,113 (60.5)


Charming caric sedge (Carex blanda)                     96 (2.8)
Total grasslike                                         96 (2.8)


Catchweed bedstraw (Galium aparine)                     6 (0.2)
Four-O'Clock (Mirabilis jalapa)*                        11 (0.3)
Frostweed (Verbesina virginica)                         67 (1.9)
Giant ragweed (Ambrosia trifida variety texana)         215 (6.2)
Hedge-parsley (Torilis arvensis)*                       89 (2.6)
Pigeonberry (Rivina humilis)                            164 (4.7)
Pokeweed (Phytolacca americana)                         1(<0.1)
Total forbs                                             553 (15.8)


Chinese privet (Lignustrum sinense)*                    17 (0.5)
Common trumpet-creeper (Campsis radicans)               4 (0.1)
Japanese honeysuckle (Lonicera japonica)*               263 (7.5)
Mustang grape (Vitis mustangensis)                      4 (0.1)
Poison-ivy (Toxicodendron radicans)                     70 (2.0)
Saw greenbriar (Smilax bona-nox)                        62 (1.8)
Virginia creeper (Parthenocissus quinquefolia)          37 (1.1)
Total shrubs                                            457 (13.1)


American elm (Ulmus americana)                          2 (0.1)
Bur oak (Quercus macrocarpa)                            4 (0.1)
Cedar elm (Ulmus crassifolia)                           33 (0.9)
Chittamwood (Sideroxylon languinosum subspecies         22 (0.6)
Honey locust (Gleditsia tricanthos)                     3 (0.1)
Pecan (Carya illinoinensis)                             4 (0.1)
Red mulberry (Morus rubra)                              5 (0.1)
Sugarberry (Celtis laevigata variety laevigata)         44 (1.3)
Tree-of-heaven (Ailanthus altissima)*                   20 (0.6)
Total trees                                             137 (3.9)
Total bare ground (including wood)                      138 (4.0)
Total hits                                              3,494 (100.0)

Table 2-Density, dominance, and relative-importance values for woody
vegetation >1 cm diameter at breast height in mixed-hardwood
bottomland forest along the Bosque River, Erath County, Texas.
Asterisks indicate introduced species.

                                               Dominance   importance
                                   Density      ([m.sup      value
Taxon                            (plants/ha)    .2]/ha)       (%)

American elm (Ulmus americana)      10.0         12.2         0.5

Bitter orange                       57.5          1.9         4.0
(Poncirus trifoliata) *

Black walnut (Juglans nigra)        12.5         26.4         0.5

Bois d arc (Maclura pomifera)       65.0         620.1        7.0

Cedar elm (Ulmus crassifolia)       82.5         146.0        6.0

China-berry                         27.5         78.2         2.0
(Melia azedarach) *

Chinese privet                      27.5          2.3         1.5
(Lignustrum sinense) *

Chittamwood (Sideroxylon            22.5          5.9         1.5
languinosum subspecies

Common honey-locust                  5.0          2.0         0.5
(Gleditsia tricanthos)

Cottonwood (Populus deltoides)       2.5         20.1         0.3

Green ash                            7.5          0.1         0.5
(Fraxinus pennsylvanica)

Mustang grape                       20.0          4.4         0.5
(Vitis mustangensis)

Pecan (Carya illinoinensis)         42.5        1,577.5       9.0

Red mulberry (Morus rubra)          40.0         205.8        4.0

Sugarberry (Celtis laevigata        307.5      10,024.4       56.0
variety laevigata)

Tree-of-heaven                      95.0         46.9         6.0
(Ailanthus altissima) *

Virginia-creeper                    12.5          0.4         0.5
(Parthenocissus quinquefolia)

Total                               837.5      12,774.6      100.3
COPYRIGHT 2013 Southwestern Association of Naturalists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rosiere, R.E.; Nelson, Allan D.; Cowley, L. Paige
Publication:Southwestern Naturalist
Article Type:Report
Geographic Code:1USA
Date:Mar 1, 2013
Previous Article:Demographic features and habitat preferences of Osgoodomys banderanus (Osgood's deermouse) in Colima, Mexico.
Next Article:Associations between size and fitness of adult females in the model odonate: Enallagma civile (odonata: Coenagrionidae).

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters