Regular issues
Special Issues

Southeastern Naturalist
    SENA Home
    Range and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Other EH Journals
    Northeastern Naturalist
    Caribbean Naturalist
    Neotropical Naturalist
    Urban Naturalist
    Eastern Paleontologist
    Journal of the North Atlantic
    Eastern Biologist

EH Natural History Home

Episodic Flooding of The Ouachita River: Levee-mediated Mortality of Trees and Saplings in a Bottomland Hardwood Restoration Area
Matthew L. Reid and Joydeep Bhattacharjee

Southeastern Naturalist, Volume 13, Issue 3 (2014): 493–505

Full-text pdf (Accessible only to subscribers.To subscribe click here.)


Site by Bennett Web & Design Co.
Southeastern Naturalist 493 M.L. Reid and J. Bhattacharjee 22001144 SOUTHEASTERN NATURALIST 1V3o(3l.) :1439,3 N–5o0. 53 Episodic Flooding of The Ouachita River: Levee-mediated Mortality of Trees and Saplings in a Bottomland Hardwood Restoration Area Matthew L. Reid1 and Joydeep Bhattacharjee2,* Abstract - The Mollicy Farms Unit of Upper Ouachita National Wildlife Refuge, LA, consists of former agricultural land replanted with traditional bottomland hardwood species. Much of it is surrounded by a containment levee built to hold back the annual floodwaters of the Ouachita River. In 2009, two extreme floods, with water levels over 4 m above the flood stage, breached the levee, leaving the area inside the levee inundated for an extended period of time. We investigated the mortality of trees and saplings following these floods. During the initial reforestation efforts, which began in 1998, trees were planted both inside and outside the levee, allowing us to compare tree and sapling mortality based on location, inside or outside the levee. The average mortality of all trees was 40.59%, and the average mortality of all saplings was 48.23%. Both tree and sapling mortality resulted from a significant interaction between elevation and location inside or outside the levee. Overall, results indicated increased mortality at lower elevations for the area inside the levee. Outside the levee, mortality was unaffected by elevation because floodwaters were able to recede naturally. Levee removal would restore a more traditional flooding regime, likely reducing tree and sapling mortality during future floods. Introduction Bottomland hardwood forests are deciduous forested wetlands found in broad floodplain areas that border river systems (Louisiana Natural Heritage Program 2009). These forested wetlands occur throughout the central and southeastern US (Hodges 1997). Riverine floodplains, which include bottomland hardwood forests, provide diverse ecosystem services, including disturbance regulation, waste filtration, and water supply and regulation (Brauman et al. 2007, Costanza et al. 1997, The Nature Conservancy 1992). They also provide productive habitat for a variety of wildlife species (The Nature Conservancy 1992, Taylor et al. 1990, Tiner 1984). Each of these ecological services is maintained by a natural flood regime and the floodplain hydrology of these forests. Bottomland hardwood forests are characterized by a hydrologic regime of alternating wet and dry cycles, which are driven by changes in water level of the associated river system, and from changes in groundwater levels (Wharton et al. 1982). During periods of heavy rainfall or spring snowmelt, rivers overtop their banks and floodwaters spill into floodplains. The flood pulse from river discharge is the major force controlling the biota in the riverine floodplains (Junk et al. 1989). 1Department of Biology, University of Louisville, KY 40292. 2Plant Ecology Laboratory, Department of Biology, University of Louisiana, Monroe, LA 71209. *Corresponding author - Manuscript Editor: Roger D. Applegate Southeastern Naturalist M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 494 Flood pulses deposit dissolved nutrients, organic matter, and fertile sediment into the floodplain. Further, hydrology influences most processes in bottomland hardwood forests, including seed dispersal (Nilsson et al. 1991; Schneider and Sharitz 1986, 1988), seed germination (Middleton 2000), growth (Harms et al. 1980, Reily and Johnson 1982, Wallace et al. 1996, Young et al. 1995), and survival of mature trees (Jones et al. 1994, Keeland et al. 1997), all of which affect the overall species composition in these plant communities (Hook 1984, Tanner 1986, Wharton et al. 1982). Bottomland hardwood forests are predominately flat, but because of the nature of flooding and sedimentation patterns, small changes in elevation result in considerable differences in soil drainage and vegetation distribution (Reid 2013, Tanner 1986). It is estimated that prior to European settlement, the extent of bottomland hardwood forests in the Lower Mississippi Alluvial Valley (LMAV) was 8–10 million ha. By 1979, approximately 2 million ha remained (MacDonald et al. 1979). Conversion to agriculture (primarily soybeans) was responsible for as much as 96% of the loss of bottomland hardwood forests in the LMAV (MacDonald et al. 1979, Newling 1990). Extensive conversion of bottomland hardwood forests to cultivation can be attributed to the rich agricultural potential of the forested wetlands in the LMAV (Tobin 1995). The LMAV has rich soils that are supplemented annually by alluvial deposits during flooding events (Hodges 1997); this enrichment leads to rapid growth rates and overall high productivity of the associated vegetation (Newling 1990). Annual flooding of the LMAV was a hindrance to farming operations, which prompted many farmers to build levees around their agricultural lands to protect them from flooding. Catastrophic flooding in the early 20th century resulted in the Flood Control Acts, which called for the construction of additional levees (Tobin 1995). These levees altered the hydrology of the bottomland hardwood forest systems and allowed for the conversion of more bottomland hardwood forest to agricultural land (Newling 1990). Louisiana has experienced a 50–75% decline of bottomland hardwood forest (Lester et al. 2005) since European settlement, which can mainly be attributed to conversion for agricultural production and alteration of traditional floodplain hydrology (US Fish and Wildlife Service 2008). In addition to the destruction of the forests, Louisiana floodplains have also been altered through the construction of impoundments, levees and canals, and the channelization and dredging of rivers (Gergel et al. 2002), leaving most floodplain forests hydrologically cut off from the river (Opperman et al. 2010). In recognition of the severe decline of these floodplain forests along the Mississippi River, Creasman et al. (1992) declared that the bottomland hardwood forest is an ecosystem in crisis. In the late 1970s, widespread abandonment of agricultural land provided opportunities for bottomland hardwood forest restoration. The US Fish and Wildlife Service (USFWS) and other federal agencies, various state agencies, The Nature Conservancy (TNC), and other organizations began getting involved in restoration efforts. From 1985 to 1995, approximately 75,000 ha of former agricultural land were reforested in the LMAV and another 108,000 ha were proposed for reforestation through 2005 (Allen 1997, Stanturf et al. 2000). Restoration efforts are Southeastern Naturalist 495 M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 currently underway throughout the southeastern US; Louisiana has undertaken one of the largest bottomland-hardwood reforestation efforts in the nation (Opperman et al. 2010, Weber et al. 2012). The inclusion of hydrological restoration is critical to riparian restoration projects because flooding regimes are essential to the proper functioning of these systems (Bhattacharjee et al. 2006, Seavy et al. 2009). However, severe flooding events can have a negative impact on the system. Holland and Burk (2000) reported tree mortality of over 30% in a floodplain forest in Massachusetts following catastrophic flooding of the Connecticut River. Extreme flooding can lead to elevated tree mortality in riparian forest systems (Acker et al. 2003, Demasceno et al. 2009). The primary objective of this study was to assess the mortality of trees and saplings in a bottomland hardwood restoration area in northeastern Louisiana following severe flooding of the Ouachita River in 2009. We conducted vegetation sampling simultaneously in a reforested area surrounded by a containment levee and a reforested area located in the natural floodplain of the river. A secondary objective was to compare mortality between the two sampling areas and among species. Materials and Methods Study area We conducted our research at the Mollicy Farms Unit portion of Upper Ouachita National Wildlife Refuge (Upper Ouachita NWR). The Mollicy Farms Unit is located in Morehouse Parish, approximately 48 km north of Monroe, LA, and includes all areas of Upper Ouachita NWR east of the Ouachita River (Fig. 1). The study site contains approximately 6500 ha of current and former agricultural land; ~27 km of levee were built to protect ~5500 ha of land from Ouachita River flooding (Weber et al. 2012). The remaining 1000 ha of the Mollicy Farms Unit, including reforested areas, are outside the containment levee. The range of elevation of the site is approximately 17 m–23 m, excluding the levee, which averages about 9 m high (Weber et al. 2012) The USFWS acquired the Mollicy Farms Unit in 1997 and has since replanted more than 4400 ha with more than 3 million trees using species characteristic of bottomland hardwood forests, making this one of the largest bottomland-hardwood restoration efforts in the nation (Opperman et al. 2010, Weber et al. 2012). The tree species planted in the Mollicy Farms Unit include Fraxinus pennsylvanica Marshall (Green Ash), Taxodium distichum (L.) Rich (Bald Cypress), Liquidambar styraciflua L. (Sweetgum), Carya aquatica (Michx. F.) (Water Hickory), and various Quercus spp. (oaks), including Q. phellos L. (Willow Oak), Q. lyrata Walter (Overcup Oak), and Q. texana Buckley (Texas Red Oak). Additionally, Q. nigra L. (Water Oak) and Celtis laevigata Willd. (Sugarberry) were planted, but in very small numbers (Gypsy Hanks, USFWS, Monroe, LA, pers. comm.). Plant nomenclature follows USDA-NRCS (2010). There was some initial seedling mortality in the first few years after planting (1999–2000), with no substantial mortality occurring since approximately 2003 (Dan Weber, TNC, Monroe, LA, pers. comm.). Southeastern Naturalist M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 496 Extreme water volume and sustained flooding of the Ouachita River during 2009 caused the levee to breach naturally on 24 May. After the levee broke, water rushed into the floodplain, and several low-lying portions of the site were inundated to Figure 1. Map of Upper Ouachita National Wildlife Refuge, LA, showing sampling locations inside and outside the containment levee. Inset shows the location within the state. Southeastern Naturalist 497 M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 a depth of 18 m (Ferber 2010). Following this initial flood in spring 2009, a second round of intense flooding ensued, lasting from October 2009 to March 2010 (Fig. 2). During this period, the entire Mollicy Farms Unit was flooded and major parts of it remained inundated for about 4–5 months. Experimental design Because our experiment took advantage of natural conditions, we had no control of the time, duration or intensity of the flood of the Ouachita River. The experiment began following the recession of floodwaters. During the summer of 2010, we sampled the Mollicy Farms Unit, located in the floodplain of the Ouachita River. Specifically, we sampled 2 areas: one inside the breached containment levee and the other outside the levee. Both areas were replanted in 1999 and 2000, and each is ~1000 ha. We set up sampling plots in a grid, with plots stratified throughout the study areas, and uploaded coordinates of plot centers for all sampling plots to a GPS (Magellan®), which we used to help us find the plots in the field. We chose plot locations a priori to remove any bias in plot establishment. We established a total of fifty-three 10-m-radius circular sampling plots, covering an area of 16,642 m2 along the floodplain. This included an area of 8478 m2 (27 plots) sampled inside the levee, and an area of 8164 m 2 (26 plots) sampled outside the levee. Figure 2. Graph of the Ouachita River level during the flooding events of 2009 and 2010. The horizontal dotted line represents the flood stage of the river. The horizontal dashed line represents the record high flood stage. The vertical dashed line indicates the date the floodwaters breached the levee surrounding the Mollicy Farms Uni t. Southeastern Naturalist M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 498 To evaluate any differences in elevation, we obtained detailed data from digital elevation models (DEM) based on LIDAR images of the site (LSU 2009). DEMbased elevation models are often used in restoration assessments in a variety of wetland settings (i.e., Kupfer et al. 2010, Millard et al. 2013). Elevation of the sampling areas inside and outside the levee ranged from 18.5 m to 22.2 m and 18.9 m to 21.4 m, respectively. Sampling procedure We counted all stems within each plot and classified stems with a diameter at breast height (dbh; measured 1.37 m from the ground) ≥10 cm as trees, and stems with a dbh less than 10 cm as saplings. For each tree or sapling, we recorded the species and whether the tree or sapling was dead or alive. Because of the inherent difficulty of accurately identifying dead oaks to the species level, we grouped all oaks as Quercus for analyses. We excluded newly recruited tree seedlings (less than 30 cm tall) from the sample because these seedlings likely established after the flooding events, thereby avoiding inaccuracy in estimating flood-caused mortality. Data regarding pre-flood tree numbers were not available because no prior efforts had been made to assess the reforestation efforts at the Mollicy Farms Unit by any of the agencies involved. Although there was initial mortality of some of the newly planted tree seedlings, refuge managers have informed us that no substantial mortality had been documented since approximately 2003 (Dan Weber, TNC, pers. comm.). Statistical analyses We analyzed these data using a one-factor logistic model with a covariate. For each sampling plot, we coded the number of living and dead trees. Location (inside or outside the levee) was the factor tested, with plot elevation included as the covariate. We specified use of quasi-binomial distribution (logit link) to correct for significant over-dispersion in the data. We analyzed significant interactions using linear contrasts and repeated the analysis for sapling data. Differences in mortality among species and the diversity of naturally regenerating species between sites were evaluated using descriptive statistics. Analyses were carried out using R v. 3.0.2 (R Core Team 2013). Results Tree mortality The average mortality of all trees was 40.59%. The logistic model of tree mortality indicated a significant interaction between location and elevation (P = 0.011; Table 1). We examined this interaction by using contrasts. The slope of the logit line for the sampling locations inside the levee was 2.22, indicating that tree mortality was significantly higher at lower elevations (P = 0.004). In contrast, the slope of the logit line for the sampling locations outside the levee was not significantly different from zero (P = 0.932), indicating no effect of elevation on tree mortality outside the levee (Table 1). Southeastern Naturalist 499 M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 Sapling mortality The average mortality of all saplings was 48.23%. The logistic model of sapling mortality indicated a significant interaction between location and elevation (P = 0.021; Table 2). We again used contrasts to understand the interaction. The slope of the logit line for the sampling locations inside the levee was 0.65, indicating that sapling mortality was marginally higher at lower elevations (P = 0.067). The slope of the logit line for the sampling locations outside the levee was not significantly different from zero (P = 0.466), indicating no effect of elevation on sapling mortality outside the levee (Table 2). Mortality by species To analyze differences in mortality among species, we combined trees and saplings for analyses (Table 3). The 3 most commonly sampled species found in both sampling locations were oaks, Green Ash, and Bald Cypress. Of the species that were planted, oaks had the highest overall mortality inside and outside the levee Table 2. Results of logistic regression model of sapling mortality as a function of location and elevation, with results of the contrasts on the significant interacti on term. Parameter Scaled deviance P Slope P Location 5.7281 0.017 - - Elevation 4.8405 0.029 - - Location:Elevation 5.2930 0.021 - - Elevation | Inside levee - - 0.6519 0.067 Elevation | Outside levee - - -0.3680 0.466 Table 1. Results of logistic regression model of tree mortality as a function of location and elevation, with results of the contrasts on the significant interaction ter m. Parameter Scaled deviance P Slope P Location 6.8705 0.009 - - Elevation 18.0128 less than 0.001 - - Location:Elevation 6.3616 0.012 - - Elevation | Inside levee - - 2.2214 0.004 Elevation | Outside levee - - -0.2277 0.933 Table 3. Mortality among select species based on location inside or outside the levee. For status, PL = planted and NR = naturally regenerating. The number of trees and saplings sampled (n) is included in parentheses. Species Status Combined Inside levee Outside levee Diospyros virginiana NR 16.0% (81) 0.0% (7) 17.6% (74) Fraxinus pennsylvanica PL 40.0% (98) 28.3% (53) 53.3% (45) Ilex decidua NR 21.2% (47) - 21.2% (47) Nyssa sylvatica NR 33.3% (33) - 33.3% (33) Pinus taeda NR 71.1% (38) 68.6% (35) 100.0% (3) Quercus spp. PL 70.3% (573) 56.5% (246) 80.7% (327) Taxodium distichum PL 17.7% (62) 28.2% (39) 0.0% (23) Southeastern Naturalist M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 500 combined, at 70.3%. Both oaks and Green Ash had higher mortality outside the levee than inside the levee. Bald Cypress had 28.2% mortality inside the levee and 0% mortality outside (Table 3). We observed Pinus taeda L. (Loblolly Pine), a species not typically associated with floodplain forests, to be regenerating in both sampling locations, though most of the regeneration was inside the levee. Mortality of Loblolly Pine trees and saplings inside the levee was 68.6%. We sampled only 3 Loblolly Pines outside the levee, all of which were dead. Most of the naturally regenerating species had lower overall mortality than the planted species. Only Nyssa sylvatica Marshall (Blackgum), Ilex decidua Walter (Possumhaw), and Diospyros virginiana L. (Common Persimmon) experienced high mortality outside the levee at 33.3%, 21.2%, and 17.6% respectively (Table 3). Mortality was less than 10% for all other species regenerating outside the levee. Natural regeneration We calculated richness (S) of naturally regenerating species (species not planted during reforestation efforts) for both sampling areas. The plots outside the levee (S =11) had greater richness of naturally regenerating species than the plots inside the levee (S = 4). Total abundance of naturally regenerating species was also lower inside the levee than outside of it: 16 and 31trees; 28 and 490 saplings, repectively. Discussion Severe flooding for a prolonged period during the growing season resulted in heavy mortality of well-established trees in the floodplain of the Ouachita River. Other studies that have evaluated mortality following severe flooding report similar findings. However, tree and sapling mortality values from this study are higher than those reported elsewhere (Acker et al. 2003, Damasceno et al. 2009, Holland and Burk 2000), and we attribute this to the magnitude and the intensity of the flooding event of the Ouachita River. Overall, trees and saplings had significantly lower mortality in the sampling location outside the levee. The levee that surrounds the Mollicy Farms Unit has been known to trap water inside during periods of heavy rain or river flooding (Ferber 2010). For this reason, farming was often unsuccessful in the area. During the floods of 2009–2010, the levee was breached and floodwater rushed into the floodplain. As the water level of the Ouachita River dropped, floodwaters were able to recede in the areas not surrounded by the containment levee. However, the levee that surrounds much of the Mollicy Farms Unit prevented the recession of the floodwaters in those areas, leaving the area inside the levee inundated for a longer period of time, as is often the case when levee breaches occur (Tobin 1995). The sampling area outside the levee was part of the original floodplain of the Ouachita River and experienced the river’s more traditional hydrologic regime. The area inside the levee was subjected to an altered hydrologic regime because the levee imposed barriers to the recession of floodwaters. While controlled flooding has been shown to be beneficial to natural seedling recruitment, especially in the riparian systems of large rivers (Bhattacharjee et al. 2006), the Southeastern Naturalist 501 M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 increased period of inundation of the site during the growing season likely resulted in greater physiological stress due to waterlogging and anaerobic soil conditions (Kozlowski 2002), which would explain the higher mortality of trees and saplings inside the levee. Our analyses indicated a significant interaction between location and elevation on the mortality of both trees and saplings. Trees and saplings at lower elevations had higher mortality than those at higher elevations inside the levee, but there was no relationship between mortality and elevation for either trees or saplings outside the levee. Inside the levee, areas at lower elevations were exposed to a longer duration of flooding than those at higher elevations. This situation would have resulted in greater stress for trees and saplings at lower elevations, contributing to their higher mortality. Mortality analyses by species indicated that oaks and Green Ash had high mortality both inside and outside the levee. Bald Cypress, which has a naturally higher flood tolerance than any other species planted at the Mollicy Farms Unit (Hook 1984), had relatively low mortality inside the levee and no observed mortality outside the levee. The high mortality of most of the planted species in both locations could also be attributed to a lack of proper species–site matching when planting trees (Stanturf et al. 2001) because hydrology limits species to certain sites in bottomland floodplains (Hook 1984, Tanner 1986). In bottomland hardwood forests, it is not unusual to see marked vegetation changes in response to elevation differences of less than 1 m within a 1-ha area (Jones et al. 1994, Wharton et al. 1982). The high mortality of Green Ash and oaks in both sampling locations seems to indicate that these species may not have been matched to the proper sites for planting. For example, some species of oak such as Overcup Oak are moderately tolerant of flooding, but other oaks such as Willow Oak are less tolerant of flooding (Burns and Honkala 1991, Hook 1984). Thus, it is possible that many of the planted oaks included species not sufficiently tolerant of flooding to warrant planting in a flood-prone area. However, it is also possible that the flood intensity and duration would have caused high mortality of the planted species, regardless of the effects of site–species matching. Most of the naturally regenerating species had reduced mortality, with regeneration occurring mainly outside the levee. The lack of natural regeneration inside the levee could also be attributed to the levee because it prevented annual floodwaters access to the floodplain, effectively eliminating hydrochory, which is an important mechanism for bottomland hardwood forest regeneration (Schneider and Sharitz 1986, 1988). We did not test this explicitly, but our data clearly indicate lower richness of naturally regenerating species inside the levee, and richness is often limited by seed dispersal (Myers and Harms 2009). Based on the results of this study and supporting evidence, we can speculate that if the levee were removed, the portion of the Mollicy Farms Unit currently inside the levee would experience a flooding regime similar to the area outside the levee. This likely would result in reduced tree and sapling mortality during subsequent flooding events because, at our study site, the area exposed to the more traditional hydrologic regime had lower average mortality. Southeastern Naturalist M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 502 It should be noted that we sampled within two large reforested stands (approximately 1000 ha each) separated by a levee. Thus, we are not able to draw inferences regarding other bottomland restoration areas. However, given the similar nature of the reforested areas in this study, it is likely that differences between stands are greatly influenced by the levee. Additionally, given the relative paucity of treemortality data in response to severe flooding, it is important to utilize available data to identify trends in tree mortality and suggest management practices to maintain these riparian systems. Overall, our results indicate significantly higher mortality of trees and saplings inside the levee than outside following the extreme floods that occurred during 2009 and 2010. By comparing locations that were similar in most regards, other than the presence of the levee, we found that the differential mortality between the two areas is likely attributable to the presence of the levee. Although flooding events can seldom be predicted, more studies should be conducted to assess the ecological impacts of containment levees, whenever any such opportunity arises. With many thousands of kilometers of levees in the Mississippi River basin (Tobin 1995), there are ample opportunities to further address the effects of levees on reforested bottomland systems throughout the region. The increasing reforestation efforts in the LMAV and the potential for increasing frequency of severe floods (Milly et al. 2002, Palmer et al. 2008) necessitate a better understanding of the patterns of tree mortality following severe floods in bottomlands. Acknowledgements Funding for this project was provided by the United States Fish and Wildlife Service. Gypsy Hanks of USFWS provided information about the tree species planted in the study area. Dan Weber of TNC provided information on seedling mortality following initial planting. Sean Chenoweth helped obtain elevation data from LIDAR images of the site. Charles Battaglia helped create the map. We thank Alex Fotis and Sarah Rhoads for assistance collecting data in the field. Literature Cited Acker, S.A., S.V. Gregory, G. Lienkaemper, W.A. McKee, F.J. Swanson, and S.D. Miller. 2003. Composition, complexity, and tree mortality in riparian forests in the central Western Cascades of Oregon. Forest Ecology and Management 173:293–3 08. Allen, J.A. 1997. Reforestation of bottomland hardwoods and the issue of woody species diversity. Restoration Ecology 5:125–134. Bhattacharjee, J., J.P. Taylor, and L.M. Smith. 2006. Controlled flooding and staged drawdown for restoration of native cottonwoods in the Middle Rio Grande Valley, New Mexico, USA. Wetlands 26:691–702. Brauman, K.A., G.C. Daily, T.K. Duarte, and H.A. Mooney. 2007. The nature and value of ecosystem services: An overview highlighting hydrologic services. Annual Review of Environmental Resources 32:67–98. Burns, R.M., and B.H. Honkala. 1991. Silvics of North America, Vol. 2. Hardwoods. USDA Forest Service Agricultural Handbook No. 654. US Government Printing Office, Washington, DC. 877 pp. Southeastern Naturalist 503 M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 Costanza, R., R. d’Arge, R. de Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, S. Naeem, R.V. O-Niell, J. Paruelo, R.G. Raskin, P. Sutton, and M. van den Belt. 1997. The value of the world’s ecosystem services and natural capital. Nature 387:253–260. Creasman, L., N.J. Craig, and M. Swan. 1992. The forested wetlands of the Mississippi River: An ecosystem in crisis. The Nature Conservancy of Louisiana, Baton Rouge, LA. Damasceno, G.A., Jr., J. Semir, F.A.M. Santos, and H. Leitào-Filho. 2004. Tree mortality in a riparian forest at Rio Paraguai, Pantanal, Brazil, after an extreme flooding. Acta Botanica Brasilica 18:839–846. Ferber, D. 2010. Into the breach. The Nature Conservancy Magazine 60:30–39. Gergel, S.E., M.D. Dixon, and M.G. Turner. 2002. Consequences of human-altered floods: Levees, floods, and floodplain forests along the Wisconsin River. Ecological Applications 12:1755–1770. Harms, W.R., H.T. Schreuder, D.D. Hook, C.L. Brown, F.W. Shropshire. 1980. The effects of flooding on the swamp forest in Lake Ocklawaha, Florida. Ecol ogy 61:1412–1421. Hodges, J.D. 1997. Development and ecology of bottomland hardwood sites. Forest Ecology and Management 90:117–125. Holland, M.M., and C.J. Burk. 2000. Effects of catastrophic flooding on floodplain-forest succession. Verhandlungen der Internationalen Vereinigung fur Theoretische und Angewandte Limnologie 27:1435–1439. Hook, D.D. 1984. Waterlogging tolerance of lowland tree species of the South. Southern Journal of Applied Forestry 8:136–149. Jones, R.H., R.R. Sharitz, P.M. Dixon, D.S. Segal, and R.L. Schneider. 1994. Woody plant regeneration in four floodplain forests. Ecological Monographs 6 4:345–367. Junk, W.J., P.B. Bayley, and R.E. Sparks. 1989. The flood-pulse concept in river-floodplain systems. Canadian Special Publication of Fisheries and Aquatic Sciences 106:110–127. Keeland, B.D., W.H. Conner, and R.R. Sharitz. 1997. A comparison of wetland tree-growth response to hydrologic regime in Louisiana and South Carolina. Forest Ecology and Management 90:237–250. Kozlowski, T.T. 2002. Physiological–ecological impacts of flooding on riparian forest ecosystems. Wetlands 22:550–561. Kupfer, J.A., K.M. Meitzen, and A.R. Pipkin. 2010. Hydrogeomorphic controls of early post-logging successional pathways in a southern floodplain forest. Forest Ecology and Management 259:1880–1889. Lester, G.D., S.G. Sorenson, P.L. Faulkner, C.S. Reid, and I.E. Maxit. 2005. Louisiana Comprehensive Wildlife Conservation Strategy. Louisiana Department of Wildlife and Fisheries, Baton Rouge, LA. Louisiana Natural Heritage Program. 2009. The Natural Communities of Louisiana. Louisiana Department of Wildlife and Fisheries, Baton Rouge, LA. Lousiana State University (LSU). 2009. Atlas: The Louisiana Statewide GIS. LSU CADGIS Research Laboratory, Baton Rouge, LA, 114. Available online at http://atlas.lsu. edu. Accessed 9 September 2010. MacDonald, P.O., W.E. Frayer, and J.K. Clauser. 1979. Documentation, chronology, and future projections of bottomland hardwood habitat losses in the Lower Mississippi Alluvial Plain, Volumes 1 and 2. US Fish and Wildlife Service, Washington, DC. Middleton, B. 2000. Hydrochory, seedbanks, and regeneration dynamics along the landscape boundaries of a forested wetland. Plant Ecology 146:169–1 84. Millard, K., A.M. Redden, T. Webster, and H. Stewart. 2013. Use of GIS and high-resolution LiDAR in salt marsh restoration site-suitability assessments in the upper Bay of Fundy, Canada. Wetlands Ecology and Management 21:243–262. Southeastern Naturalist M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 504 Milly, P.C.D., R.T. Wetherald, K.A. Dunne, and T.L. Delworth. 2002. Increasing risk of great floods in a changing climate. Nature 415:514–517. Myers, J.A., and K.E. Harms. 2009. Seed arrival, ecological filters, and plant species richness: A meta-analysis. Ecology Letters 12:1250–1260. Newling, C.J. 1990. Restoration of bottomland hardwood forests in the Lower Mississippi Valley. Restoration Manager Notes 8:23–28. Nilsson, C., M. Gardfjell, and G. Grelsson. 1991. Importance of hydrochory in structuring plant communities along rivers. Canadian Journal of Botany 69:2 631–2633. Opperman, J.J., B.A. McKenney, M. Roberts, and A.W. Meadows. 2010. Ecologically functional floodplains: Connectivity, flow regime, and scale. Journal of the American Water Resources Association 46:211–226. Palmer, M.A., C.A.R. Lierman, C. Nilsson, M. Flörke, J. Alcamo, P.S. Lake, and N. Bond. 2008. Climate change and the world’s river basins: Anticipating management options. Frontiers in Ecology and the Environment 6:81–89. R Core Team. 2013. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Reid, M.L. 2013. A quarter century of plant succession in a bottomland hardwood forest in northeastern Louisiana. M.Sc. Thesis. University of Louisiana at Monroe, Monroe, LA. Reily, P.W., and W.C. Johnson. 1982. The effects of altered hydrological regime on tree growth along the Missouri River in North Dakota. Canadian Journal of Botany 60:2410–2423. Schneider, R.L., and R.R. Sharitz. 1986. Seedbank dynamics in a southeastern riverine swamp. American Journal of Botany 73:1022–1030. Schneider, R.L., and R.R. Sharitz. 1988. Hydrochory and regeneration in a Bald Cypress– Water Tupelo swamp forest. Ecology 69:1055–1063. Seavy, N.E., T. Gardali, G.H. Golet, F.T. Griggs, C.A. Howell, R. Kelsey, S.L. Small, J.H. Viers, and J.F. Weigand. 2009. Why climate change makes riparian restoration more important than ever: Recommendations for practice and research. Ecological Restoration 27:330–338. Stanturf, J.A., E.S. Gardiner, P.B. Hamel, M.S. Devall, T. Leininger, and M.E. Warren. 2000. Restoring bottomland hardwood ecosystems in the Lower Mississippi Alluvial Valley. Journal of Forestry 98:10–16. Stanturf, J.A., S.H. Schoenholtz, C.J. Schweitzer, and J.P. Shepard. 2001. Achieving restoration success: Myths in bottomland hardwood forests. Restoration Ecology 9:189–200. Tanner, J.T.1986. Distribution of tree species in Louisiana bottomland forests. Castanea 51:168–174. Taylor, J.R., M.A. Cardamone, and W.J. Mitsch. 1990. Bottomland hardwood forests: Their functions and values. Pp. 13–86, In J.G. Gosselink, L.C. Lee, and T.A. Muir (Eds.). Ecological Processes and Cumulative Impacts: Illustrated by Bottomland Hardwood Wetland Ecosystems. Lewis Publishers, Chelsea, MI. 708 pp. The Nature Conservancy. 1992. Restoration of the Mississippi River Alluvial Plain as a Functional Ecosystem. The Nature Conservancy, Baton Rouge, LA. Tiner, R.W. 1984. Wetlands of the United States: Current status and recent trends. US Fish and Wildlife Service, Washington, DC. 59 pp. Tobin, G.A. 1995. The levee love affair: A stormy relationship? Water Resource Bulletin 31:359–367. US Department of Agriculture–Natural Resources Conservation Service (USDA-NRCS). 2010. The PLANTS database. Available online at Accessed 30 June 2012. Southeastern Naturalist 505 M.L. Reid and J. Bhattacharjee 2014 Vol. 13, No. 3 US Fish and Wildlife Service (USFWS). 2008. Upper Ouachita and Handy Brake National Wildlife Refuges and the Louisiana Wetlands Management District, Louisiana: Comprehensive conservation plan. US Department of the Interior, Southeast Region, Atlanta, GA. 238 pp. Wallace, P.M., D.M. Kent, and D.R. Rich. 1996. Responses of wetland tree species to hydrology and soils. Restoration Ecology 4:33–41. Weber, D., J. McGowan, S. Haase, C. Rice, T. Kennedy, R. Martin, and K. Ouchley. 2012. Bringing floodplain restoration to scale: The Mollicy Farms Project on the Ouachita River, Upper Ouachita National Wildlife Refuge, Louisiana, USA. National Wetlands Newsletter 34:16–19. 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. USFWS, Biological Service Program, FWS/OBS-81/37, Washington, DC. 133 pp. Young, P.J., B.D. Keeland, and R.R. Sharitz. 1995. Growth response of Baldcypress (Taxodium distichum (L.) Rich.) to an altered hydrologic regime. The American Midland Naturalist 133:206–212.