2007 SOUTHEASTERN NATURALIST 6(2):217–234 Hydrogeomorphology and Forest Composition of Sunrise Towhead Island in the Lower Mississippi River Sabine Greulich1, Scott Franklin2,4,*, Thad Wasklewicz3,4, and Jack Grubaugh2,4 Abstract - Forest vegetation was studied in relation to hydrogeomorphology on a large fluvial island in the meandering section of the Lower Mississippi River. The island has a relatively wide topographic gradient, including a former channel of the Mississippi River. Vegetation patterns were related to geomorphologic features, elevation, flood duration, and characteristics of surficial sediment. Overstory vegetation was species rich for the island as a whole and dominated by the pioneer taxa Populus deltoides (eastern cottonwood) and Salix nigra (black willow). Both pioneer species dominated the old channel. Vegetation at higher elevations was characterized by tree species from later successional stages such as Celtis laevigata (sugarberry), Acer saccharinum (silver maple), Liquidambar styraciflua (sweetgum) and Ulmus spp. (elm). Seedlings and saplings were dominated by sugarberry and other later successional species, whereas cottonwood was infrequent. Our results suggest that on the Lower Mississippi River, coexistence in the floodplain of pioneer stages and later successional stages will not be perpetuated under the prevailing hydrologic and geomorphologic regimes. Introduction Anthropogenic features, especially levees, have isolated many alluvial floodplains of the Lower Mississippi River (LMR) (Johnson 1992, Knutson and Klaas 1998, Nelson 1997, Peck and Smart 1986, Scott and Udouj 1999). Large islands along the LMR that are still influenced by the river’s hydrologic cycles harbour much of the remaining alluvial forests and may be important reservoirs of biotic diversity. We examined one such island, Sunrise Towhead Island, in an effort to reconstruct the natural history of the island and the spatial association of biotic and abiotic conditions. Many studies have documented modification or loss of floodplain forests during the 20th century (Johnson 1998, Johnson et al. 1976, Knutson and Klaas 1998, Nelson et al. 1998, Schnitzler 1996, Trémolières et al. 1998, Yin et al. 1997). Large rivers have undergone major human-induced changes, including river straightening, modification of cross-section morphology, and construction of dams and levees (Dynesius and Nilsson 1994, Müller 1995), which confine channel meandering and disconnect major portions of 1Centre de Recherche (Ville Société Territoire), UMR CNRS 6173 CITERES, Université François Rabelais, Parc de Grandmont, bât. B, F-37200 TOURS, France. 2Department of Biology, University of Memphis, 3700 Walker Avenue, Memphis, TN 38152. 3Department of Earth Sciences, University of Memphis, Memphis, TN 38152. 4Edward J. Meeman Biological Field Station, University of Memphis, Millington, TN 38058.* Corresponding author - sfrankli@memphis.edu. 218 Southeastern Naturalist Vol. 6, No. 2 floodplain forests (Delaney and Craig 1997, Schnitzler 1996, Trémolières et al. 1998). Floodplain forests remaining within the levees are affected by altered hydrology (Franklin et al. 2003, Grubaugh and Anderson 1989, Poff et al. 1997, Sparks 1995) and altered geo-fluvial dynamics (Galay 1983, Johnson 1992, Wasklewicz et al. 2004). In many engineered river systems, floodplain forests have become confined to islands (Dykaar and Wigington 2000, Nelson et al. 1998) that are generally too dynamic for anthropogenic use. Despite their potentially important role in floodplain forest conservation and landscape diversity, fluvial islands have been studied only recently as distinct floodplain units (Dykaar and Wigington 2000, Gurnell and Petts 2002, Merigliano 1998, Osterkamp 1998). Osterkamp (1998) has defined fluvial islands as land units surrounded by water, elevated higher than the mean water level, and persisting long enough to permit establishment of vegetative cover. The creation and evolution of many fluvial islands has been directly related to past river modifications (Lekarczyk 2003). While most studies of fluvial islands are limited to braided river sections of the Upper Mississippi River or smaller rivers (Bayard and Schweingruber 1991, Dykaar and Wigington 2000, Kollmann et al. 1999, Merigliano 1998, Merritt and Cooper 2000), detailed data on vegetation zonation and associated abiotic conditions of larger meandering river islands is lacking. This paper examines environmental gradients, forest composition, and structure of a large fluvial island (Sunrise Towhead Island) in the meandering section of the Lower Mississippi River. The island encompasses a substantial topographic gradient with frequently to rarely flooded sites and a previous river channel. Our goal was to characterize formation and structure of a large island, including vegetation gradients, in the LMR, and compare to current floodplain classification systems. Study Site Sunrise Towhead Island is situated at Mississippi River mile 775, approximately 40 km upstream from Memphis, TN, and just upstream from the mouth of the Hatchie River (Fig. 1). At low flows of the Mississippi River, the island measures approximately 6050 m long and 3000 m wide, with an elevation range of 􀂧 10 m (Fig. 1). Sunrise Towhead was initiated as a point bar that was progressively cut (we think naturally) from the shore during the 1930s and 1940s (Fig. 2a). Judging from several years of aerial photography, a large portion of the western section of the island was formed from deposition in the former main channel during cutting of the bar, and by the successive abandonment of the old channel (Fig. 2b). The old channel and other low areas of the island are still connected to the river hydrology due to porosity of island geology. Thus, during medium river stages (e.g., 8–10 m), while the island may not be overtopped by high water, low areas in the island are inundated. Erosion of the east shore by lateral migration of the Mississippi River main 2007 S. Greulich, S. Franklin, T. Wasklewicz, and J. Grubaugh 219 channel was abated by rip-rap and articulated concrete mattresses (i.e., revetments) that were laid out along the shore in the 1950s (Wasklewicz et al. 2004). Mean annual low stages occur in fall and mean annual high stages in late spring and early summer (Fig. 3), showing a flow pattern Figure 1. Topographic map of Sunrise Towhead Island (Lower Mississippi River, TN). 220 Southeastern Naturalist Vol. 6, No. 2 Figure 2. Present and former position of Sunrise Towhead Island (Lower Mississippi River, TN) with regard to the Mississippi River channels. The localization of sampling transects is indicated by dashed (1937) or solid (1983) lines. Figure 3. Hydrology (stage data standardized to meters NVGD 1929) for the Fulton, TN, gauge from 1981 to 1989. Crosses denote the mean (horizontal line) and standard deviation (vertical line) for the four major dominance types described on the island (see Table 3), as well as the total elevation range studied on Sunrise Towhead Island. 2007 S. Greulich, S. Franklin, T. Wasklewicz, and J. Grubaugh 221 typical of southeastern coastal plain rivers (Hupp 2000). The island is situated on a section of the Mississippi River where the channel bed has been degrading since the 1930s, following artificial meander cut-offs further downstream (Smith and Winkley 1996, Winkley 1994). The closest gauging station is Fulton, 500 m upstream of Sunrise Towhead Island. Methods General abiotic factors Two transects were established in 1999, crossing the island perpendicular to river flow, approximately at 1/3 (close to the largest island width) and 2/3 of the island length (Fig. 2a). Both transects intersect the 1930s channel and the higher elevations of the island core. Forest plots were established on each transect at 80-m intervals. A total of 61 plots were established. Elevations of the highest and lowest plots were 76.2 m and 67.1 m, respectively. A sediment core sample of the upper 5 cm was extracted at the center of each plot. Sediments were analyzed for organic matter content by loss through ignition at 430 °C for 24 h, after having been oven dried overnight at 105 °C. The sediment fraction < 2 mm was separated by sieving and its particle size distribution (fractions of sand, clay, and silt) was evaluated by the hydrometer method (Foth et al. 1980). Flooding frequency (number of times a particular site was inundated each year) and duration (number of days of inundation per year) at the different sampling plots were estimated by comparing topographic elevation of the plots according to a 1:24,500 map (Osceola quadrangle, US Army Engineer District Memphis, TN, 1994) and river-stage records from 1980 to 1989 (records after this point are not available) at the Fulton gauge. Elevation data were obtained after mapping topography at a 1.5-m scale in GIS software and overlaying plot points. Stage records were acquired from the US Army Corps of Engineers, Memphis District. Vegetation-environment relationships Vegetation data were collected from each plot along the two transects. Vegetation sampling was accomplished using the point-centered quarter method (Mueller-Dombois and Ellenberg 1974). Within each quarter and within 20 m, the nearest stem > 10 cm diameter at breast height (dbh) was measured both for distance from plot center and for diameter, from which basal area and density were calculated. At each of the four cardinal directions, a 4-m radius circular plot (0.005 ha; all four = 0.02 ha for each site) was established. Within these nested plots, all woody stems of trees and shrub species were identified to species and counted as either saplings (dbh 􀂕 2.54 and < 10 cm) or seedlings (dbh < 2.54 cm). Data analysis Relative frequency (number of plots per species/total number of plots), relative dominance (basal area of species/total basal area), relative density (individuals of species/total individuals), and importance value (I.V. = sum 222 Southeastern Naturalist Vol. 6, No. 2 of relative density and relative basal area) were calculated for each species for all sample points. Environmental structure of sample plots was analysed through a normalized principal component analysis (nPCA). Plots were characterized by environmental variables, elevation (elev), average event flooding duration between 1980 and 1989 (flood), percentage of clay (clay), silt (silt), and sand (sand), percentage of organic matter (OM), distance from channel in 1937 (dist_old), and closest distance to one of the two present channels, west or east shore (dist_new). The latter variable was included because island margins may be subjected to the most intense flood disturbance, and may trap the highest amount of water-dispersed seeds. Both factors could potentially influence vegetation patterns. Overstory vegetation composition was analyzed by correspondence analysis (CA). As a result of the point-centered method in vegetation sampling, the score of each tree species at a sampling plot was assigned values between 0 (i.e., none of the nearest trees belonged to the considered species) and 4 (i.e., each of the four sampled trees on the plots belonged to the considered species). Groups delineated through ordination analysis were used as community types to document vegetation-environment relationships. Ordination by environmental factors and ordination of overstory vegetation were compared by co-inertia analysis (Dolédec and Chessel 1994). Statistical significance of the co-structure was tested with a random permutation test (Monte Carlo test). All multivariate analyses were undertaken using the ADE-4 software package (Thioulouse et al. 1997). Results General abiotic factors Mean flooding frequency ranged from 0 at the highest elevation (76 m) to 50 times per year on the lowest elevation (67 m). Mean number of flood days ranged from 0 to 181 days per year. Surface sediment texture was mostly silty clay or clay. Sand dominated in the most marginal plots on both transects and in the western third of the north transect. Texture ranged as follows: sand constituted 0% to 100%, silt 0% to 63%, and clay 0% to 68% of surface sediment. Organic matter content ranged from 0.15% to 3.9%. Average annual flooding was positively correlated with percent clay and organic matter. Plots close to the former channel were situated at low elevations and appeared to have low sand, but high clay and organic matter content, and a relatively high basal area. Distance from the present channel showed few correlations with other abiotic factors. Vegetation and environment relations A total of 23 tree species were encountered. Populus deltoides Bartram ex Marsh (eastern cottonwood) was the dominant species, representing half of the total basal area (Table 1), followed by Salix nigra Marsh (black willow) and Celtis laevigata Willd (sugarberry), which were similar in density and frequency. Relative frequencies of the most dominant species were similar on both transects. 2007 S. Greulich, S. Franklin, T. Wasklewicz, and J. Grubaugh 223 Ordination of sampling plots by correspondence analysis of overstory vegetation revealed a direct gradient, with three main vegetation communities coinciding with topography (Figs. 4 and 5). Co-inertia between overstory vegetation structure and environmental factors was highly significant (p < 0.0001; Fig. 4). Group A, the Ulmus-Celtis laevigata community based on dominance (Table 2), comprised species centered far from the old channel, on unflooded or only briefly flooded plots with sandy sediment (Table 3). The canopy was generally sparse, with other dominant species including Juniperus virginiana L. (eastern red cedar) and Diospyros virginiana L. (common persimmon). The understory was dominated by sugarberry and was more diverse (seedling richness = 5) than other communities (seedling richness 􀂔 3; Table 4). Group B, the Celtis laevigata-Ulmus-Acer saccharinum community, was composed of species that were generally located outside the old channel at sites with intermediate annual flooding and intermediate to fine-grained sediment, relatively far from the present river channels (Table 3, Fig. 5). This group had fairly high richness for all strata, especially the canopy (richness = 11), compared to other groups (canopy richness 􀂔 8; Tables 2 and 4). Table 1. Overstory (trees > 10 cm dbh) composition on the Mississippi Island Sunrise Towhead. N = total number of individuals on island; Rel. dom. = relative dominance; Rel. den. = relative density; I.V. = importance value (= relative dominance + relative density); Plot Frequency = number of plots in which a species occurs with regard to total plot number. Species are listed in order by I.V. Abbrev. Species Rel. Rel. % plot name common name (scientific name) N dom. den. I.V. frequency Popdel Eastern cottonwood (Populus deltoides) 43 50.0 21.4 71.4 37.7 Salnig Black willow (Salix nigra) 42 19.2 20.9 40.1 32.8 Cellae Southern hackberry (Celtis laevigata) 38 9.6 18.9 28.5 36.1 Ulmus1 Elm (Ulmus spp.) 23 7.7 11.5 19.1 26.2 Acesaci Silver maple (Acer saccharinum) 21 6.7 10.4 17.2 21.3 Aceneg Boxelder (Acer negundo) 7 1.8 3.5 5.3 11.5 Carya2 Hickory (Carya spp.) 6 1.2 3.0 4.2 9.8 Gletri Honey locust (Gleditsia triacanthos L.) 2 1.4 1.0 2.4 3.3 Junvir Eastern red cedar (Juniperus virginiana) 3 0.8 1.5 2.3 4.9 Diovir Common persimmon (Diospyros virginiana) 4 0.3 2.0 2.3 4.9 Liqsty Sweetgum (Liquidambar styraciflua) 3 0.2 1.5 1.7 1.6 Quenig Water Oak (Quercus nigra L.) 2 0.2 1.0 1.2 1.6 Iledec Possumhaw holly (Ilex decidua)Walter 2 0.1 1.0 1.1 3.3 Frapen Ash (Fraxinus pennsylvanica Marsh.) 1 0.6 0.5 1.1 1.6 Betnig River birch (Betula nigra L.) 1 0.1 0.5 0.6 1.6 Nyssyl Sourgum (Nyssa sylvatica L.) 1 0.1 0.5 0.6 1.6 Morrub Red mulberry (Morus rubra L.) 1 0.1 0.5 0.6 1.6 Malus Crabapple (Malus sp.) 1 0.0 0.5 0.5 1.6 Totals 201 100.0 100.0 200.0 100.0 1Ulmus species included Ulmus rubra Muhl (slippery elm), Ulmus americana L.(American elm), and Ulmus alata Michx (winged elm). 2Carya species included Carya illinoensis (Wangenh) K. Koch (pecan) (CARILL), Carya laciniosa (Michx.) Loudon (shellbark hickory) (CARLAC), and Carya tomentosa (Poir.) Nutt (mockernut hickory) (CARTOM). 224 Southeastern Naturalist Vol. 6, No. 2 Group C, dominated by eastern cottonwood and black willow, was located on parts of the island with the greatest annual flooding and was the dominant community (31 of 61 plots). This group was common in the old channel, where soils were fine-textured and soil organic matter content fairly high. A fourth group, lacking an overstory, was dominated by Prunus angustifolia Marsh (chickasaw plum) shrubs at higher elevations on sandy, droughty soils with very low organic matter (Tables 3 and 4, Fig. 5). Eastern red cedar and Ulmus spp. were the only taxa found taller than 1.5 m, and individuals were scarce. Cornus drummondii C.A. Mey (roughleaf dogwood) was also common in these sites. Flooding was infrequent and of short duration. Figure 4. Factorial maps of overstory vegetation (a) and abiotic factors (b) on Sunrise Towhead Island, Lower Mississippi River, TN. Maps were created by co-inertia analysis that linked ordination of sampling plots according to vegetation to their ordination according to abiotic factors. Species are abbreviated by the first three letters of their generic name and specific epithet (see Table 1). See text for explanation of abiotic factors. 2007 S. Greulich, S. Franklin, T. Wasklewicz, and J. Grubaugh 225 Dominant tree species showed different patterns throughout the island with regard to size-class distribution (Fig. 6). In all species but eastern cottonwood, tree abundance was greatest in the smallest size classes, Figure 5. Vegetation communities (based on ordination analysis) along sampling transects on Sunrise Towhead Island, Lower Mississippi River, TN. The channel prior to island formations (i.e., old channel) is shown for both transects. Elevation is meters above mean sea level. S = surface sediments with > 50% sand. 226 Southeastern Naturalist Vol. 6, No. 2 reflecting recruitment subsequent to overstory formation. The great majority of eastern cottonwood trunks were larger than 40 cm dbh, while the smallest size class was absent. The largest trees on the island were eastern cottonwood and black willow. The major portion of small understory vegetation (stems < 2.54 cm dbh) consisted of the shrub or vine Toxicodendron radicans L. Kuntze (poison ivy; Table 4). Sugarberry was the most frequent and most abundant species of tree seedlings and saplings, followed by Acer negundo L. (boxelder) and silver maple in the seedling size class, and by black willow in the sapling size class. Eastern cottonwood, the dominant overstory species, was infrequent in the understory. Table 3. Mean (± standard deviation) values of various environmental variables of the four major dominance types on Sunrise Towhead Island, Lower Mississippi River. Flood frequency is the average number of times a particular site was inundated. Flood duration is the number of days a particular site remained inundated. Soil texture and organic matter are from the top 10 cm of soil profile. Celtis laevigata- Prunus Ulmus-Celtis Ulmus-Acer Populus deltoides- Variable shrubland laevigata saccharinum Salix nigra Elevation (m) 73.9 (0.8) 73.5 (1.3) 71.9 (2.3) 68.9 (1.7) Flood duration (days) 12 (7.1) 17 (14) 45 (45.5) 112 (44.9) Flood frequency < 1 5 (3.9) 12 (12.5) 31 (12.3) Percent organic matter 0.2 (0.1) 1.4 (1.1) 2.3 (1.1) 2.5 (0.7) Percent clay 1 (2.5) 21 (19.2) 38 (21.9) 45 (16.8) Percent sand 96 (5.7) 57 (38.5) 28 (38.5) 11 (25.7) Distance to old channel (m) 1467 (141) 1020 (484) 286 (353) 111 (216) Distance to new channel (m) 360 (150) 816 (491) 678 (421) 590 (330) Table 2. Relative basal area for select species of the four major dominance types on Sunrise Towhead Island, Lower Mississippi River. All Prunus stems were < 10 cm diameter at breast height, and thus had negligible basal area. Celtis laevigata- Prunus Ulmus- Ulmus-Acer Populus deltoides- Species shrubland Celtis laevigata saccharinum Salix nigra Juniperus virginiana 60.0 8.4 Diospyros virginiana 3.5 0.1 Celtis laevigata 19.2 32.4 3.1 Other hardwoods 9.3 6.0 0.8 Liquidambar styraciflua 2.3 Ulmus spp. 40.0 46.4 26.1 Acer saccharinum 19.1 4.4 Acer negundo 9.0 0.2 Populus deltoides 5.0 66.1 Salix nigra 2.4 25.4 Sample size (number of sites) 6 5 19 31 Total richness (all sites) 2 8 11 7 Average basal area (4 stems) 0.03 0.35 0.23 0.58 2007 S. Greulich, S. Franklin, T. Wasklewicz, and J. Grubaugh 227 Discussion We found evidence for one major environmental gradient controlling vegetation patterns on Sunrise Towhead Island. As with most floodplain systems, vegetation was strongly related to inundation and associated soil characteristics, e.g., soil texture (Johnson et al. 1995, Mitsch and Gosselink 1993). This study shows that pioneer tree species eastern cottonwood and black willow presently dominate the forest canopy on Sunrise Towhead Island, but succession is proceeding toward more shade tolerant hardwoods, e.g., sugarberry. Sunrise Towhead Island, situated in the Lower Mississippi Valley (LMV), belongs to the southern floodplain forest region (Küchler 1964). This region covers the Mississippi alluvial plain northward from the Gulf Coast to the southern tip of Illinois, and along the lower reaches of several Mississippi tributaries (Braun 1950, Putnam et al. 1960). We found four major vegetation-environment zones on the island. The lowest, most frequently flooded parts of the island were characterized by fine-grained sediments (silty clay, clay). Their very low organic matter content (< 3.5%) suggests that those sediments had been deposited by recent low-energy flows such as waning floods and had not yet been amended in situ by organic material (Hupp 2000). The soil and flood characteristics seem to match floodplain zone III as described by Clark and Benforado (1981) and Wharton et al. (1982). Canopy dominants were eastern cottonwood and black willow, matching most closely to the P. deltoides Temporarily Flooded Forest Alliance (I.B.2.N.d.160) of The Nature Conservancy’s (TNC) National Vegetation Classification System (NVCS, Table 4. Total sapling (2.54 < dbh < 10 cm) density, average sapling density in parentheses, and diversity for the four major dominance types on Sunrise Towhead Island, Lower Mississippi River. Celtis laevigata- Prunus Ulmus- Ulmus-Acer Populus deltoides- Species shrubland Celtis laevigata saccharinum Salix nigra Cornus drummundii 23 (4) 18 (4) 32 (2) 1 (< 1) Prunus angustifolia Marsh. 14 (2) 7 (1) 0 1 (< 1) Juniperus virginiana 7 (1) 0 1 (1) 1 (< 1) Diospyros virginiana 0 7 (1) 2 (less than 1) 7 (< 1) Celtis laevigata 1 79 (16) 104 (6) 14 (1) Ulmus spp. 0 5 (< 1) 14 (1) 1 (< 1) Other hardwoods 0 31 (6) 45 (3) 20 (1) Toxicodendron radicansA 3 474 (95) 5233 (275) 302 (10) Acer negundo 0 1 (< 1) 3 (< 1) 26 (1) Populus deltoides 0 0 6 (< 1) 2 (< 1) Salix nigra 0 0 27 (2) 378 (12) Mean seedling richness 2 5 3 2 Mean seedling evenness 0.22 0.68 0.35 0.33 Mean sapling richness 2 3 3 1 Mean sapling evenness 0.42 0.76 0.53 0.18 ASeedlings, shown due to overwhelming dominance. 228 Southeastern Naturalist Vol. 6, No. 2 Figure 6. Size distribution of the five most dominant tree species on S u n r i s e Towhead I s l a n d , L o w e r M i s s i s - s i p p i R i v e r , TN. 2007 S. Greulich, S. Franklin, T. Wasklewicz, and J. Grubaugh 229 sensu Weakley et al. 1998). Lack of eastern cottonwood regeneration and predominance of smaller black willow suggests more frequent inundation at present than during the initial formation of the forest (Fralish and Franklin 2002). Compared to eastern cottonwood seedlings, seedlings of black willow are more tolerant to saturated soil conditions and long inundation (Hosner and Minckler 1963, Shelford 1954). Intermediate elevations and flooding regimes of the island matched with floodplain zone IV (Clark and Benforado 1981, Wharton et al. 1982), and were dominated by sugarberry and Ulmus species. Ulmus dominated on higher elevations with lower organic matter and higher percentages of sand, while sugarberry and Ulmus spp. shared dominance with silver maple and several other hardwood species at elevations between Ulmusdominated communities and Populus-dominated communities. These communities are most similar to the Sugarberry-Elm-Ash cover type developed by the Society of American Foresters (Eyre 1980), and the Acer saccharinum Temporarily Flooded Forest Alliance from TNC’s NVCS (Weakley et al. 1998). These Celtis-dominated forests are currently the most common community of the Mississippi Alluvial Valley (McWilliams and Rosson 1990), almost doubling in size since the 1930s (Rudis 2001). They now exhibit the strongest successional trend on the island, with large eastern cottonwood and black willow stems in the overstory and few to none in the understory. This successional sequence is common of southern floodplain forests, where seral communities following cottonwood-willow stands are typically dominated by Liquidambar styraciflua L. (sweetgum), sugar berry, and/or silver maple (Barnes 1985, Hosner and Minckler 1963, Putnam et al. 1960, Shelford 1954). Available aerial photographs suggest that cottonwood was established in the early 1950s, i.e., soon after the old channel was abandoned by the Mississippi River in the 1940s. Estimated age and location of the present cottonwood-willow stands suggest that they comprise the first colonizers of the old Mississippi channel (mainly willow) and adjacent sand bars (mainly cottonwood), ideal conditions for cottonwood-willow colonization (Fralish and Franklin 2002, Karrenberg et al. 2003, Kellison et al. 1998, Shull 1944). According to Shelford’s (1954) schedule of community development in the Lower Mississippi floodplain, the dominance of the Celtis-Liquidambar stage is achieved about 80 years after establishment of the first cottonwoodwillow seedlings. Our findings suggest that vegetation succession on Sunrise Towhead will support this timeframe. These seral communities (sugarberry is short-lived, < 150 years) may eventually be succeeded by an oak-hickory forest type, which would be determined by a combination of site conditions and disturbance regimes (Sharitz and Mitsch 1993, Shelford 1954). Later successional oak (mainly Quercus nigra L. [WATER OAK]) and hickory species (mainly pecan), while currently present, make up a small percentage of the overstory vegetation (a combined importance value of only 5.4). 230 Southeastern Naturalist Vol. 6, No. 2 The above-described vegetation and succession gradient explains why species richness on Sunrise Towhead is relatively high compared to findings from other studies on floodplain forests. We identified 23 overstory tree taxa (17 genera), and some additional species were encountered in the seedling/ sapling layer. The surveyed surface, however, was small (7.7 ha: 61 plots of 0.126 ha). Since the point-quarter centered method takes only four trees at each sampling plot into account, additional species may occur on the island. As a comparison, Johnson et al. (1976) studied 34 floodplain forest stands along a 130-km stretch of the Missouri River and found nine overstory species. Crites and Binger (1969) encountered 12 tree species in six floodplain forests in east-central Illinois. Knutson and Klaas (1998) conducted a floristic inventory of floodplain forests along the Upper Mississippi River (pools 6 to 10, a river stretch of 177 km) and listed 18 overstory species, a number similar to that found on Sunrise Towhead. Studies that were able to sample southern bottomland forest on wider topographic gradients and/or to include later successional stages have found a species richness exceeding that of Sunrise Towhead. Shelford (1954), in the Lower Mississippi River, and Robertson et al. (1978), in the Middle Mississippi River, for example, list some forty tree species each. Surface sediment of many of the highest, least flooded sites were mainly sand (88% to 100%) and had very low organic matter content (“no overstory” zone). This zone was occupied mainly by Prunus angustifolia (chickasan plum) shrubs, a drought-tolerant species that commonly grows in thickets on sandy soils (Gilman and Watson 1994). This differs from what one would expect underlaying undisturbed, later successional communities, and has no clear counterpart in typical floodplains. Scattered stems of eastern red cedar and common persimmon occurred in these xeric conditions. According to the Mississippi River hydrograph (Fulton gauge) of the recent past, the highest zone of the island is rarely flooded, and the vegetation exists mainly because of rainwater. There are two possible explanations for the origin of the sandy soil. First, these high elevation sands may have been deposited during major events (1927, 1937, and 1980 according to Fulton gauge records) and thus are above high-water levels of normal years. Secondly, sands may have been dumped on the island as dredge spoil, although dredge spoil along this section of the Mississippi River is almost always dumped back into the river to be carried downstream (Micheal Thron, US Army Corps of Engineers, Memphis District, Memphis, TN, pers. comm.). In addition, historical maps show this portion of the island intact back to the 1800s. This study shows that the hydrologic and geomorphologic context of Sunrise Towhead Island excludes development of new habitats for pioneer vegetation on the island. It is unlikely that this situation will change in the near future, since meandering of the river and erosion of the island shore will not continue to generate moist, open sandy surfaces ideal for pioneer seed germination due to rip-rap and articulated concrete mattresses. These 2007 S. Greulich, S. Franklin, T. Wasklewicz, and J. Grubaugh 231 engineered features are part of the regulation of the Mississippi River that started systematically on a large scale in the 1930s (Stevens et al. 1975, Theiling 1999, Wasklewicz et. al. 2001, Winkley 1994). Since Sunrise Towhead has assembled on its small surface ecological gradients typical of floodplain environments, the island might be representative of the whole floodplain sector. Our results suggest that on the Lower Mississippi River, the coexistence in the floodplain of pioneer stages and later successional stages that is valuable from a landscape diversity point of view (Amoros and Bornette 2002, Naiman et al. 1993, Osterkamp et al. 2001, Schnitzler 1996, Ward 1998, Ward et al. 1999) will not be perpetuated under the prevailing hydrologic and geomorphologic regimes. With the area of floodplain forests in the Mississippi Alluvial Plain steadily declining over the past 70 years (McWilliams and Rosson 1990), islands may be important diversity refuges, and should be incorporated into restoration proposals for the Lower Mississippi River (Llewellyn et al. 1996, Stanturf et al. 2000). Acknowledgments The authors are grateful to Darren Mitchell, Dennis Staley, Marcus Pearson, and Timothy Bills. Thanks to A. Schnitzler and K. Maier for sharing knowledge on southern floodplain forests, and K. Maier for field help. A special thanks to D. Marsh and T. Welch for allowing us to overnight on the island. This research was made possible by a post-doctoral grant of the University of Memphis’ Biology Department to S. Greulich, and by an internal University of Memphis Faculty Research Grant. The research would not have been possible without the support of the Meeman Biological Field Station (MBFS). MBFS supplied use of field equipment, materials, and boat travel to and from the island. Literature Cited Amoros, C., and G. Bornette. 2002. Connectivity and biocomplexity in waterbodies of riverine floodplains. Freshwater Biology 47:761–776. Barnes, W.J. 1985. Population dynamics of woody plants on a river island. Canadian Journal of Botany 63:647–655. Bayard, M., and F.H. Schweingruber. 1991. Ein baumgrenzstandort: Das Wildwasserbett der Maggia im Tessin, Schweiz. Eine dendroökologische studie. Botanica Helvetica 101:9–28. Braun, E.L. 1950. Deciduous Forests of Eastern North America. The Blakiston Company, Philadelphia, PA. Clark, J.R., and J. Benforado (Eds.) 1981. Wetlands of Bottomland Hardwood Forests. Elsevier, Amsterdam, The Netherlands. Crites, R.W., and J.E. Binger. 1969. Vegetation survey of floodplain forests in East Central Illinois. Transactions Illinois Academy Science 44:316–330. Delaney, R., and M.R. Craig. 1997. Longitudinal changes in Mississippi River floodplain structure. Upper Mississippi River Long Term Resource monitoring Program, US Geological Survey, Onalaska, WI. Project Status Report 97-02. Dolédec, S., and D. Chessel. 1994. Co-inertia analysis: An alternative method for studying species-environment relationships. Freshwater Biology 31:277–294. 232 Southeastern Naturalist Vol. 6, No. 2 Dykaar, B.B., and P.J. Wigington. 2000. Floodplain formation and cottonwood colonization patterns on the Willamette River, Oregon, USA. Environmental Management 25:87–104. Dynesius, M., and C. Nilsson. 1994. Fragmentation and flow regulation of river systems in the northern third of the world. Science 266:753–762. Eyre, F.H. 1980. Forest-cover Types of the United States and Canada. Society of American Foresters, Washington, DC. Foth, H.D., L.V. Withee, H.S. Jacobs, and S.J. Thien. 1980. Laboratory Manual for Introductory Soil Science, 5th Edition. WCB Publishing, Dubuque, IA. Fralish, J.S., and S.B. Franklin. 2002. Taxonomy and Ecology of Woody Plants in North American Forests. John Wiley and Sons, Inc., New York, NY. Franklin, S.B., T. Wasklewicz, J.W. Grubaugh, and S. Gruelich. 2003. Temporal periodicity of the Mississippi River before and after systematic channel modifications. Journal of the American Water Resources Association 39:637–648. Galay, V.J. 1983. Causes of river bed degradation. Water Resources Research 19:1057–1090. Gilman, E.F., and D.G. Watson. 1994. Prunus angustifolia, Chickasaw Plum. US Forest Service Fact Sheet ST-504, Florida Cooperative Extension Service, University of Florida, Gainesville, FL. Grubaugh, J.W., and R.V. Anderson. 1989. Long-term effects of navigation dams on a segment of the upper Mississippi River. Regulated Rivers: Research and Management 4:97–104. Gurnell A.M., and G.E. Petts. 2002. Island-dominated landscapes of large floodplain rivers, a European perspective. Freshwater Biology 47:581–600. Hosner, J.F., and L.S. Minckler. 1963. Bottomland hardwood forests of Southern Illinois: Regeneration and succession. Ecology 44:29–40. Hupp, C.P. 2000. Hydrology, geomorphology, and vegetation of coastal plain rivers in the southeastern United States. Hydrologic Processes 14:2991–3010. Johnson, W.C. 1992. Dams and riparian forest: Case study from the Upper Missouri River. Rivers 3:229–242. Johnson, W.C. 1998. Adjustment of riparian vegetation to river regulation in the Great Plains, USA. Wetlands 18 (4): 608–618. Johnson, W.C., R.L. Burgess, and W.R. Kaemmerer. 1976. Forest overstory vegetation and environment on the Missouri River Floodplain in North Dakota. Ecological Monographs 46:59–84. Johnson, W.C., M. Dixon, R. Simons, S. Jenson, and K. Larson. 1995. Mapping the response of riparian vegetation to possible flow reductions in the Snake River, Idaho. Geomorphology 13:159–173. Karrenberg, S., P.J. Edwards, and J. Kollmann. 2003. The life history of Salicaceae living in the active zone of floodplains. Freshwater Biology 47:733–748. Kellison, R.C., M.J. Young, R.R. Braham, and E.J. Jones. 1998. Major alluvial floodplains. Pp. 219–323, In M.G. Messina and W.H. Conner (Eds.). Southern Forested Wetlands Ecology and Management, Lewis Publishers, Boca Raton, FL. Knutson, M.G., and E.E. Klaas.1998. Floodplain forest loss and changes in forest community structure in the Upper Mississippi River: A wildlife habitat at risk. Natural Areas Journal 18:138–150. Kollmann, J., M. Vieli, P.J. Edwards., K. Trockner, and J.V. Ward. 1999. Interactions between vegetation development and island formation in the Alpine river Tagliamento. Applied Vegetation Science 2:25–36. Küchler, A.W. 1964. Manual to Accompany the Map of Potential Vegetation of the Conterminous United States. Special Publication No. 36. American Geographical Society, New York, NY. 2007 S. Greulich, S. Franklin, T. Wasklewicz, and J. Grubaugh 233 Lekarczyk, M. 2003. Island morphometry: 55 years of change on the northern lower Mississippi. M.Sc. Thesis. University of Memphis, Memphis, TN. Llewellyn, D.W., G.P. Shaffer, N.J. Craig, L. Creasman, D. Pashley, M. Swan, and C. Brown. 1996. A decision-support system for prioritizing restoration sites on the Mississippi River. Conservation Biology 10:1446–1455. McWilliams, W.H., and J.F. Rosson, Jr. 1990. Composition and vulnerability of bottomland hardwood forests of the Coastal Plain Province in the south central United States. Forest Ecology and Management 33/34:485–501. Merigliano, M.F. 1998. Cottonwood and willow demography on a young island, Salmon River, Idaho. Wetlands 18:571–576. Merritt, D.M., and D.J. Cooper. 2000. Riparian vegetation and channel change in response to river regulation: A comparative study of regulated and unregulated streams in the Green River Basin, USA. Regulated Rivers: Research and Management 16:543–564. Mitsch, W.J., and J.G. Gosselink. 1993. Wetlands, 2nd Edition. John Wiley and Sons, Inc., New York, NY. Mueller-Dombois, D., and H. Ellenberg. 1974. Aims and Methods of Vegetation Analysis. John Wiley and Sons, New York, NY. Müller, N. 1995. River dynamics and floodplain vegetation and their alterations due to human impact. Archiv für Hydrobiologie Supplement 101:43–61. Naiman, R.J., H. Décamps, and M. Pollock. 1993. The role of riparian corridors in maintaining regional biodiversity. Ecological Applications 3:209–212. Nelson, J.C. 1997. Presettlement vegetation patterns along the 5th principal meridian, Missouri Territory, 1815. American Midland Naturalist 137:79–94. Nelson, J.C., L. DeHaan, R.E. Sparks, and L. Robinson. 1998. Presettlement and contemporary vegetation patterns along two navigation reaches of the Upper Mississippi River. Pp. 51–60, In T.D. Sisks (Ed.). Perspectives on the Land Use of North America: A Context for Understanding Our Changing Environment. US Geological Survey, Biological Resources Division, Springfield, VA. Biological Science Report USGS/BRD/BSR 1998-003 (Revised September 1999). Osterkamp W.R. 1998. Processes of fluvial island formation, with examples from Plum Creek, Colorado and Snake River, Idaho. Wetlands 18:530–545. Osterkamp, W.R., W.C. Johnson, and M.D. Dixon. 2001. Biophysical gradients related to channel islands, Middle Snake River, Idaho. Pp. 73–83, In J.M Dorava, D.R. Montgomery, B.B. Palcsak, and F.A. Fitzpatrick (Eds.). Geomorphic Processes and Riverine Habitats, Water Science and Application Volume 4. American Geophysical Union, Washington, DC. Peck, J.H., and M.M. Smart. 1986. An assessment of the aquatic and wetland vegetation of the Upper Mississippi River. Hydrobiologia 136:57–76. Poff, N.L., J.D. Allan, M.B. Bain, J.R. Karr, K.L. Prestegaard, B.D. Richter, R.E. Sparks, and J.C. Stromberg. 1997. The natural flow regime. BioScience 47:764–784. Putnam, J.A., G.M. Furnival, and J.S. McKnight. 1960. Management and inventory of southern hardwoods. Agriculture Handbook 181. US Department of Agriculture Forest Service, Washington, DC. Robertson, P.A., G.T. Weaver, and J.A. Cavanaugh. 1978. Vegetation and tree species patterns near the northern terminus of the southern floodplain forest. Ecological Monographs 48:249–267. Rudis, V.A. 2001. Composition, potential old growth, fragmentation, and ownership of Mississippi Alluvial Valley bottomland hardwoods: A regional assessment of historic change. Pp. 28–48, In P.B. Hamel and T.L. Foti (Eds.). Bottomland Hardwoods of the Mississippi Alluvial Valley: Characteristics and Management of Natural Function, Structure, and Composition. General Technical Report SRS-42. USDA Forest Service, Asheville, NC. 234 Southeastern Naturalist Vol. 6, No. 2 Schnitzler, A. 1996. Comparison of landscape diversity in forests of the upper Rhine and middle Loire floodplains (France). Biodiversity and Conservation. 5:743–758. Scott, H.D., and T.H. Udouj. 1999. Spatial and temporal characterization of land use in the Buffalo National River Watershed. Environmental Conservation 26:94–101. Sharitz, R.R., and W.J. Mitsch. 1993. Southern floodplain forests. Pp. 311–372, In W.H. Martin, S.G. Boyce, and A.C. Echternacht (Eds.). Biodiversity of the Southeastern United States: Lowland Terrestrial Communities. John Wiley and Sons, New York, NY. Shelford, V.E. 1954 Lower Mississippi valley flood plain biotic communities: Their age and elevation. Ecology 35:126–142. Shull, C.A. 1944. Observations of general vegetation changes on a river island in the Mississippi River. American Midland Naturalist 32:771–776. Smith, L.M., and Winkley, B.R. 1996. The response of the lower Mississippi River to river engineering. Engineering Geology 45:433–455. Sparks, R.E. 1995. Need for ecosystem management of large rivers and their floodplains. BioScience 45:168–182. Stanturf, J.A., E.S. Gardiner, P.B. Hamel, M.S. Devall, T.D. Leininger, and M.E. Warren, Jr. 2000. Restorating bottomland hardwood ecosystems in the lower Mississippi Alluvial Valley. Journal of Forestry 98:10–16. Stevens, M.A., Simons, D.B., and Schumm, S.A. 1975. Man-induced changes of Middle Mississippi River. Journal of the Waterways Harbors and Coastal Engineering Division 101:119–133. Theiling, C. 1999. Important milestones in the human and ecological history of the Upper Mississippi River system. Pp. 3–1 to 3–12, In J.D. Cox and M.V. Winger (Eds.). Ecological Status and Trends of the Upper Mississippi River System 1998. LTRMP Report 99-T001. US Geological Survey, La Crosse, WI. Thioulouse, J., D. Chessel, S. Dolédec, and J.M. Olivier. 1997. ADE-4: A multivariate analysis and graphical display software. Statistics and Computing 7:75–83. Trémolières, M., J.M. Sanchez-Pérez, A. Schnitzler, and D. Schmitt. 1998. Impact of river management history on the community structure, species composition, and nutrient state in the Rhine alluvial forest. Plant Ecology 135:59–78. Ward, J.V. 1998. Riverine landscapes: Biodiversity patterns, disturbance regimes, and aquatic conservation. Biological Conservation 83:269–278. Ward, J.V., K. Tockner, and F. Schiemer. 1999. Biodiversity of floodplain river ecosystems: Ecotones and connectivity. Regulated Rivers: Research and Management 15:125–139. Wasklewicz, T.A., S.B. Franklin, S. Greulich, and J.W. Grubaugh. 2001. Changes in the hydrologic periodicity of Mississippi River, USA. Pp. 53–61, In R.A. Falconer and W.R. Blain (Eds.). River Basin Management, WIT Press, Southampton, UK. Wasklewicz, T.A., J.W. Grubaugh, S.B. Franklin, and S. Gruelich. 2004. 20th-century trends along the Mississippi River. Physical Geography 25:208–224. Weakley, A.S., K.D. Patterson, S. Landaal, M. Pyne, and others. 1998. Terrestrial Vegetation of the Southeastern United States. March 1998 working draft. The Nature Conservancy, Chapel Hill, NC. Wharton, C.H., W.M. Kitchens, E.C. Pendleton, and T.W. Sipe. 1982. The ecology of bottomland hardwood swamps of the southeast: A community Profile. FWS/OBS- 81/37. US Fish and Wildlife Service, Biological Service Program, Washington, DC. Winkley, B.R. 1994. Response of the lower Mississippi River to flood control and navigation improvements. Pp. 45–74, In S.A. Schumm and B.R. Winkley (Eds.). The Variability of Large Alluvial Rivers. American Society of Civil Engineers, New York, NY. Yin, Y., J.C. Nelson, and K.S. Lubinski. 1997. Bottomland hardwood forests along the Upper Mississippi River. Natural Areas Journal 17:164–173.