Competitive interactions in chenier communities of Southwest Louisiana.
Key words: cheniers, Chinese Tallow tree, competition, introduced species, Sapium sebiferum.
The Chenier Plain in southwestern Louisiana and southeastern Texas was formed by Mississippi River sediments during the past 5,000 years (Coleman 1966). In places, the geological substrate of the plain is overlain by relatively recent inland "stranded" beaches called cheniers that run parallel to the coast (Gosselink et al. 1979). These cheniers mark periods of coastal retreat between intervals of coastal advance, resulting from shifts in position of the Mississippi River delta (Price 1955, Van Lopijk and McIntire 1957, Gould and McFarlan 1959, Coleman 1966, Orvos and Price 1979), and range in age from 2,800 years to less than 300. These ridges (relict "islands" in brackish and fresh marshes) range from 1 to 3 m in height, 30 to 450 m in width, and up to 50 km in length. Since Native Americans arrived in Louisiana at least 12,000 years ago (Kniffen 1968), cheniers have experienced anthropogenic effects throughout their history. Today, almost all of the chenier land is privately owned. Chenier acreage is used for pasturage of cattle, habitations, and roads. Only fragments remain of the woody vegetation that once covered the cheniers.
The introduced tree species Sapium sebiferum (L.) Roxb. is now common in the woodland remnants of the cheniers. Commonly called Chinese Tallow, it is a native of subtropical China. It has been planted in several areas of the world as an ornamental and for the production of vegetable tallow and stillingia oil from its fruits (Kahn et al. 1973). It was introduced to the southeastern U.S.A. by the U.S.D.A. Bureau of Plant Industry in the early 1900's to establish local soap industries (Jamieson and McKinney 1938). Although there are no records indicating precisely how long S. sebiferum has been present on the cheniers, local residents do not recall a time when it was absent (pers. obs.). Sapium sebiferum has some of the life-history characteristics expected of invasive tree species (Rejmanek and Richardson 1996). It grows rapidly and regenerates from stumps (Scheld and Cowles 1981), and its seeds are widely dispersed by birds (Jones and McLeod 1989). The overwhelming majority of successful plant invasions occur in disturbed habitats (Rejmanek 1989, Hobbs 1991), and it is in such environments that S. sebiferum is most abundant. Not surprisingly, the species has become naturalized and widespread in the Gulf Coastal Plain of Texas and Louisiana, in some cases replacing coastal prairies and pasturage with monospecific stands (Bruce et al. 1995). It often outgrows native bottomland hardwoods in sunlight, grows well in saline soils (Cameron and Spencer 1989), and thrives in poorly drained and intermittently flooded soils.
Neyland and Meyer (1997) characterized the woody vegetation of six remnant woodlands on Little Chenier and Grand Chenier in Cameron Parish in southwestern Louisiana. The tree species with the highest densities, coverages and importance values (sensu Cox 1996) were Quercus virginiana Miller, Sapium sebiferum, Celtis laevigata Willd., Ulmus americana L., and Diospyros virginiana L. The extent of S. sebiferum infiltration varied among the sites, from 85% of all trees on one site on Little Chenier, to complete absence on one site on Grand Chenier.
It was once thought that successful invaders appropriated "empty niches," and that such opportunism accounted for their success. Recent theory argues instead that exotics outcompete native species, especially in anthropogenically disturbed habitats (Herbold and Moyle 1986). The successful, and often most devastating, introductions are thus organisms that have strong competitive abilities. Rather than entering empty niches, they displace native species. The high densities reached in some chenier communities by Sapium sebiferum suggest that it might well have such an effect. The purpose of this study is to evaluate intraspecific and interspecific competitive interactions involving S. sebiferum in these woodlands. Definitive proof of competition in the field requires manipulative experiments, as has been done with some communities (e.g. Petren and Case 1996). However, since such an approach was not feasible for the adult tree communities studied here, a correlative observational technique developed by Yeaton and Cody (1976) and Yeaton et al. (1977) was used to evaluate competitive interactions. Given that competition often leads to a positive correlation between plant size and interplant distance (Pielou 1962), a correlation between the combined sizes of a pair of individuals and the distance between them is an indication of competition. This applies equally to intraspecific and interspecific competition, although in the latter case the methodology must account for differences in plant size.
MATERIALS AND METHODS
Three sites were selected from the six surveyed by Neyland and Meyer (1997): one on Little Chenier (LC1) and two on Grand Chenier (GC1 and GC2). The densities, relative densities, and coverage of the three most common species (Neyland and Meyer 1997) at each site are shown in Table 1. These sites represented high (LCI), moderate (GC2), and no (GC1) S. sebiferum colonization. For the three most common species at each site, data were obtained on the sizes and interplant distances of individuals of all possible intraspecific and interspecific pair combinations. For each pair combination, 10 breast height diameters and distances to the nearest neighbor (measured from center of trunk to center of trunk) were determined. The nearest neighbor was only accepted if no individual of an undesired species (including overhead crown foliage) intercepted the line between it and the first member. All interplant distances were less than 16 m. These data were collected during April 1998. For interspecific combinations, the values for the species with the smaller variance in size were rescaled by multiplying them by [s.sub.1]/[s.sub.2], where [s.sub.1] is the standard deviation for the species with the higher variance. This created sets of sample values with equal variances. For all intraspecific and interspecific combinations, correlations were calculated between the sum of the size values of the two pair members and the interplant distance for these individuals. Competitive interaction is indicated by a significant (P<0.05) positive correlation (Yeaton and Cody 1976, Yeaton et al. 1977, Cox 1996).
From 108 Sapium sebiferum individuals (obtained at LC2, GC2 and one other Little Chenier site), a regression of age (tree rings) with breast height diameter was calculated. This was used to determine the age structure of the 307 S. sebiferum individuals censused by Neyland and Meyer (1997).
In the fall of 1997, three 25 [m.sup.2] plots were cleared of all tree seedlings and juveniles at each of the three sites. On 7 March 1999 the plots were censused for Sapium sebiferum seedlings and juveniles. Other seedling species were enumerated but not identified.
The woodland remnant at LC1 measured 250 m X 50 m and was bounded by intermediate marsh (salinity approximately 2.2 ppt). It is located at 92 [degrees] 56'W, 29 [degrees] 49' N. The center of the site and its north-facing boundary were in a natural state. The southern boundary was defined by paved road. Mesic-type herbaceous ground cover was present (e.g. Heliotropium indicum L., Myosotis macrosperma Engelm., Sisyrinchium langloisii E. Greene). Low areas of the interior were characterized by several north-south sloughs that supported herbaceous swamp plants (e.g. Iris virginica L., Saururus cernuus L.).
Site GC1 measured 300 m X 150 m and was bounded by intermediate to brackish marsh (salinity approximately 2.2 to 5.1 ppt). It was located at 92 [degrees] 49' W, 29 [degrees] 44' N. The center of the site and its north-facing edge were in a natural state. The south-facing edge was defined by a paved road, the east edge by a private shell road, and the west boundary by a commercial shell road. Mesic-type herbaceous ground cover was present. Sabal minor (Jacquin) was dense. A single clump of Cornus drommondii C. Meyer was present.
Site GC2 measured 300m X 100 m and was bounded by intermediate marsh (salinity approximately 2.2 ppt). It was located at 92 [degrees] 56' W, 29 [degrees] 44' N. The wooded interior and all edges were in a natural state. There was evidence of frequent use by cattle. Mesic-type herbaceous ground cover was present.
At all three sites, the correlation coefficient, r, indicated a significant intraspecific interaction for Quercus virginiana and Celtis laevigata (Table 2). Sapium sebiferum and Diospyros virginiana did not show any evidence of intraspecific interaction (Table 2). Evidence of significant interspecific competition was found in only two of the nine species combinations, both at GC2 (Table 3). Sapium sebiferum competed interspecifically only with Celtis laevigata at GC2.
Age and breast height diameter of Sapium sebiferum were found to be significantly correlated (r=0.7032, p=0.0143). The age structure of S. sebiferum at the chenier sites is shown in Table 4. A large majority of the individuals found were under 6 years old. Only three were over 15 years old.
Sapium sebiferum seedlings colonized the cleared plats at LC1 and GC2 (Table 5). They were not found at GC1. where the species is virtually absent as an adult.
Sapium sebiferum was the most abundant tree species in chenier woodland at all sites where it was found (Neyland and Meyer 1997). However, unlike two other dominant chenier species, Quercus virginiana and Celtis laevigata, S. sebiferum showed no evidence of intraspecific competition. Evidence for interspecific interactions between Sapium sebiferum and other dominant chenier trees was found only with Celtis laevigata at GC2. The relative density of S. sebiferum was substantially lower at GC2 than at LC1 (Table 1). Furthermore, at GC2, S. sebiferum was less obviously clumped in dispersion and more intermixed with Q. laevigata than at LC1, where it was concentrated in several sloughs (Neyland and Meyer 1997). Sapium sebiferum in monospecific clumps at GC2 would be less likely to compete interspecifically with C. laevigata. These differences between LC1 and GC2 may account for the differences in interspecific interactions.
Despite being highly abundant on most chenier sites, the relative coverage of Sapium sebiferum was always substantially less than Celtis laevigata and Quercus virginiana. The age structure of S. sebiferum shows that the overwhelming majority of individuals are young (Table 4). Although this is the first study to analyze chenier woodlands quantitatively, published surveys and anecdotal evidence from local inhabitants (pers. obs.) indicate that S. sebiferum has been abundant on the cheniers for at least half a century. This suggests that survivorship of this species at these sites is low. Most Q. virginiana and many C. laevigata at these sites are undoubtedly much older than most S. sebiferum. It is therefore perhaps not surprising that S. sebiferum shows only slight evidence of interspecific competition; indeed it seems unlikely that it would be able to supplant older, established trees of these other species. Unfortunately, we have no way to substantiate what tree species, if any, previously occupied the space currently held by S. sebiferum.
The absence of Sapium sebiferum from GC1 site is puzzling. The site does not differ in any obvious manner from other sites in the area and is only approximately 2.5 km from the nearest potential source. Neyland and Meyer (1997) speculated that the species might only now be invading the community. However, the failure of S. sebiferum to colonize cleared plots at GC1 does not support this speculation. Alternatively, as suggested in Neyland and Meyer (1997), given that S. sebiferum is bird dispersed (Jones and McLeod 1989) and currently established at other sites in the proximity of GC1, some unknown mechanism may be preventing the species from becoming established or even dispersing to the site.
This study investigated competitive interactions only among adult and juvenile trees with diameters at breast height of at least 1 cm. Therefore, we do not discount the possibility that Sapium sebiferum is a superior interspecific competitor as a seedling. Nevertheless, current evidence does not support the hypothesis that S. sebiferum in the cheniers of southwestern Louisiana outcompetes and displaces native tree species. Our results are more consistent with the hypothesis that S. sebiferum is an opportunistic species whose superior dispersal capability and affinity for disturbed and fragmented habitats allows it to colonize unoccupied terrain successfully. Given that much of our modern terrestrial landscape is disturbed and will become increasingly more so (Hannah et al. 1994), we can expect that the infiltration of this highly successful invader into these habitats will increase.
TABLE 1. Densities (number of individuals per 25 [m.sup.2] plot), relative densities (percent), and relative coverages (percent) of the most common tree species at three sites on the cheniers of southwest Louisiana (Neyland and Meyer 1997). Relative Relative density coverage Site Species Density (%) (%) LC1 Sapium sebiferum 3.40 61 2 Celtis laevigata 1.04 19 7 Quercus virginiana 0.52 9 87 GC1 Celtis laevigata 1.72 27 36 Diospyros virginiana 1.52 24 1 Quercus virginiana 1.04 16 34 GC2 Sapium sebiferum 1.28 39 11 Celtis laevigata 0.72 22 21 Quercus virginiana 0.48 15 50 TABLE 2. Intraspecific interactions among the three most common tree species at three chenier sites in southwest Louisiana. An asterisk indicates a significant interaction (P<0.05). Site Species r, correlation coefficient P GC1 Celtis laevigata 0.6608 0.0375 * Quercus virginiana 0.6726 0.0331 * Diospyros virginiana 0.2846 0.4255 GC2 Sapium sebiferum -0.1602 0.6584 Quercus virginiana 0.7492 0.0126 * Celtis laevigata 0.6689 0.0344 * LC1 Sapium sebiferum -0.1712 0.6363 Quercus virginiana 0.7686 0.0094 * Celtis laevigata 0.7408 0.0229 * TABLE 3. Interspecific interactions among the three most common tree species at three chenier sites in southwest Louisiana. An asterisk indicates a significant interaction (P<0.05). Site Species pair r, correlation coefficient P GC1 Celtis-Quercus 0.1429 0.6939 Celtis-Diospyros -0.1202 0.7408 Quercus-Diospyros 0.1364 0.7231 GC2 Sapium-Quercus -0.1705 0.6376 Sapium-Celtis 0.6638 0.0364 * Quercus-Celtis 0.6358 0.0482 * LC1 Sapium-Quercus -0.4861 0.1543 Sapium-Celtis -0.3555 0.3134 Quercus-Celtis 0.1429 0.6938 TABLE 4. Age structure of Sapium sebiferum on Little Chenier and Grand Chenier in southwest Louisiana. Age class (y) Number of individuals 0-5 254 6-10 33 11-15 17 >15 3 TABLE 5. Colonization by seedlings of Sapium sebiferum and other tree species of three cleared 25 [m.sup.2] plots at three chenier sites in southwest Louisiana. Mean number of Site Species seedlings per plot Standard deviation LC1 Sapium sebiferum 11.7 15.0 other 1.0 1.0 GC1 Sapium sebiferum 0 0 other 8.3 6.5 GC2 Sapium sebiferum 3.7 2.5 other 4.0 2.6
Financial assistance for this project was provided by McNeese State University through a 1997-1998 McNeese State University Shearman Academic Endowed Professorship. We thank Dean Robert for assistance in the field. We also thank Billy Doland for allowing us access to his property.
BRUCE, K. A., G. N. CAMERON, AND P. A. HARCOMBE. 1995. Initiation of a new woodland type on the Texas Coastal Prairie by the Chinese Tallow tree (Sapium sebiferum (L.) Roxb.). Bull. Torrey Bot. Soc. 122:215-225.
CAMERON, G. N. AND R. SPENCER. 1989. Rapid leaf decay and nutrient release in a Chinese tallow forest. Oecologia 80:222-228.
COLEMAN, J. M. 1966. Recent coastal sedimentation: Central Louisiana Coast. Coastal studies Series No. 17. Louisiana University Press, Baton Rouge.
COX, G. W. 1996. Laboratory Manual for General Ecology. Seventh Edition. Wm. C. Brown Publishers.
GOSSELINK J. G., C. L. CORDES, AND J. W. PARSONS. 1979. An ecological characterization study of the Chenier Plain coastal ecosystem of Louisiana and Texas. U.S. Fish and Wildlife Service, Office of Biological Services.
GOULD, H. R. AND E. MCFARLAN. 1959. Geological history of the Chenier Plain, southwest Louisiana. Trans. Gulf Coast Assoc. Geol. Soc. 9:261-270.
HANNAH, L., D. LOHSE, C. HUTCHINSON, J. L. CARR, AND A. LAKERANI. 1994. A preliminary inventory of human disturbance of world ecosystems. Ambio 23:246-250.
HERBOLD, B. AND P. B. MOYLE. 1986. Introduced species and vacant niches. American Naturalist 128:751-760.
HOBBS, R. J. 1991. Disturbance as a precursor to weed invasion in native vegetation. Plant Protection Quarterly 6:99-104.
JAMIESON, G. S. AND R. S. MCKINNEY. 1938. Stillingia oil. Oil and Soap 15:295-296.
JONES, R. H. AND K. W. MCLEOD. 1989. Shade tolerance in seedlings of Chinese tallow tree, American sycamore, and cherrybark oak. Bull. Torrey Bot. Club 116:371-377.
KAHN, F. W., K. KHAN, AND M. N. MALIK. 1973. Vegetable tallow and stillingia oil from the fruits of Sapium sebiferum, Roxb. Pakistan J. Forestry 23:257-266.
KNIFFEN, F. B. 1968. Louisiana, its land and people. Louisiana University Press, Baton Rouge.
NEYLAND, R. AND H. A MEYER. 1997. Species diversity of the Louisiana chenier woody vegetation remnants. Bull. Torrey Bot. Soc. 124:254-261.
OTVOS, E. G. AND W. A. PRICE. 1979. Problems of chenier genesis and terminology--An overview. Marine Geol. 31:251-263.
PETREN, K. AND T. J. CASE. 1996. An experimental demonstration of exploitation competition in an ongoing invasion. Ecology 77:118-132.
PIELOU, E. C. 1962. The use of plant-to-neighbor distances for the detection of competition. J. Ecology 50:357-367.
PRICE, W. A. 1955. Environment and formation of the chenier plain. Quaternaria 2:75-86.
REJMANEK, M. 1989. Invasibility of plant communities. Pp. 369-388. In J. A. Drake, H. A. Mooney, F. di Castri, R. H. Groves, F. J. Kruger, M. Rejmanek, and M. Williamson (Eds.), Biography of Mediterranean invasions.
REJMANEK, M. AND D. M. RICHARDSON. 1996. What attributes make some plant species more invasive? Ecology 77:1655-1661.
SCHELD, H. W. AND J. R. COWLES. 1981. Woody biomass potential of the Chinese Tallow tree. Econ. Bot. 35: 391-397.
VAN LOPIJK, J. R. AND W. G. MCINTIRE. 1957. Cheniers of Vermillion Parish, La: Their relation to Mississippi River delta chronology. Trans. Gulf Coast Assoc. Geol. Soc. 7: 302.
YEATON, R. I. AND M. L. CODY. 1976. Competition and spacing in plant communities: The northern Mojave Desert. J. Ecology 64:689-686.
YEATON, R. I., J. TRAVIS, AND E. GILINSKY. 1977. Competition and spacing in plant communities: The Arizona upland association. J. Ecology 65:587-595.
Harry A. Meyer, Ray Neyland, and Eric Smith Department of Biological and Environmental Sciences Box 92000 McNeese State University Lake Charles, LA 70609
|Printer friendly Cite/link Email Feedback|
|Author:||Meyer, Harry A.; Neyland, Ray; Smith, Eric|
|Publication:||The Proceedings of the Louisiana Academy of Sciences|
|Article Type:||Statistical Data Included|
|Date:||Jan 1, 1999|
|Previous Article:||Initial investigation into the reproductive biology of the antelope-horn milkweed, asclepias viridis walter (asclepiadaceae).|
|Next Article:||Abstracts of papers presented at the 1999 annual meeting of the Louisiana Academy of Sciences.|
|National charity mailings work around suppressed ZIPS.|