ORIGINAL PAPER Marsh Terraces in Coastal Louisiana Increase Marsh Edge and Densities of Waterbirds Jessica L. O���Connell & John A. Nyman Received: 4 September 2008 /Accepted: 4 September 2009 /Published online: 9 December 2009 # Society of Wetland Scientists 2009 Abstract We evaluated the influence of marsh terracing on waterbirds in Louisiana���s Chenier Plain. Terracing is a novel technique used to slow coastal marsh loss. Terracing increases marsh edge and is assumed to slow erosion, decrease pond depth, and encourage vegetation production. From April to September 2005, we monitored waterbirds in paired terraced and unterraced ponds in three sites dominated by Spartina patens. We additionally sampled submerged aquatic vegetation (SAV) biomass, nekton density, and water quality. Waterbird density and species richness were 3.8 and 1.4 times greater, respectively, in terraced ponds. By foraging guild, probers, aerial foragers, and dabbling foragers were more abundant in terraced ponds. Waders were frequently more abundant in terraced ponds. Diver density did not differ significantly between pond types. Terracing increased marsh edge in ponds 3.5 times. Nekton and SAV were more abundant in edge habitat than in open water, but water quality, water depth, SAV, and nekton did not differ significantly between pond types and did not influence bird density. Bird densities were higher in ponds with greater proportions of marsh edge, possibly because they are morphologically constrained to forage in shallow water or because of abundant food near edges. Keywords Habitat interspersion . Nekton . Submerged aquatic vegetation . Waterbird communities . Wetland restoration Introduction Conservation of coastal marshes is important because they provide many unique ecosystem services and constitute valuable wildlife habitat. Louisiana contains 40% of the remaining coastal marsh in the continental United States (Field et al. 1991), and provides critical habitat for water- birds (Michot 1996 Esslinger and Wilson 2001). However, many of these waterbirds are declining regionally from habitat loss. For example, 27 species classified as species of high or moderate concern by the Waterbird Conservation Council (Kushlan et al. 2002) are regularly seen in brackish marshes along the Gulf Coast. In 1984, coastal Louisiana contained more nesting colonies of seabirds and wading birds than any other state in the southeast US (Keller et al. 1984). However more recently, Louisiana colonies for most species have declined (Michot et al. 2003 Green et al. 2006). Wetlands in coastal Louisiana have been in rapid decline, accounting for 80% of US coastal wetland loss from 1950 to 1994 (Boesch et al. 1994), potentially imperiling dependent migratory birds. Coastal marsh loss in Louisiana is considerable and results from conversion of marsh to shallow open water. These marshes can be divided into two major sub-regions, the Deltaic Plain of the Mississippi River, and the Chenier Plain, directly to the west. The two regions differ in process of formation, geomorphology, hydrology, and marsh loss. Marshes in the Chenier Plain, where this study was conducted, are formed over riverine sediments that have been reworked and deposited through marine action, J. L. O���Connell (*) Department of Zoology, Oklahoma State University, Stillwater, OK 74078, USA e-mail: jessica.oconnell@okstate.edu J. A. Nyman School of Renewable Natural Resources, Louisiana State University Agricultural Center, Baton Rouge, LA 70803, USA e-mail: jnyman@lsu.edu Wetlands (2010) 30:125���135 DOI 10.1007/s13157-009-0009-y

creating new wetlands (Penland and Suter 1989). Unlike deltaic marshes, direct sediment deposition from riverine flow is unimportant in wetland creation in the Chenier Plain. Once created, emergent wetlands in the Chenier Plain depend upon vertical accumulation to offset sea-level rise and subsidence (Foret 2001). Vertical accretion proceeds via peat accumulation, which relies on nutrient supplements from sporadic over-wash events during large storms and autochthonous production (Foret 2001 Turner et al. 2006). Prior to 1956, these processes were sufficient to sustain uninterrupted expanses of emergent marsh throughout much of the Chenier Plain (Barras et al. 1994). Since then, dredging for navigation and petroleum exploration has directly or indirectly caused extensive interior marsh loss (Byrnes et al. 1995). Marsh loss rates in the Chenier Plain were 16.3 km2/year from 1978���2000 (Barras et al. 2003). Interior loss is often attributed to salt-water intrusion or prolonged flooding that causes vegetation die-off in isolated hot spots. This creates open water areas approximately 1-m deep that rarely drain because of lengthy, convoluted hydraulic connections to the Gulf of Mexico. Once formed, these ponds increase in size even after vegetation die-off ceases, presumably because of soil erosion around pond margins. This phenomenon may be exacerbated by a variety of factors, including sea level rise and sediment starvation from channelization of the Mississippi and other rivers (Boesch et al. 1994 Turner 1997). Pond terracing is a novel marsh restoration technique developed in response to open water conversion of interior marshes in the Chenier Plain (Underwood et al. 1991 Steyer 1993 Rozas and Minello 2001). Since 1990, 2,200 ha of Chenier Plain marsh have been restored using terraces (Stead and Hill 2004), and an evaluation of terraced marsh as habitat for dependent waterbirds is overdue. Terraces are discontinuous, narrow strips of created marsh. They are formed of dredge material stabilized by emergent vegetation such as Spartina alterniflora (Underwood et al. 1991 Steyer 1993 Rozas and Minello 2001). Sediment for terrace building usually is taken from pond bottoms and is piled using a backhoe, creating borrow pits within ponds. Terraces are thought to reduce wave energy and dampen the erosive force of water, potentially slowing lateral erosion of surrounding marsh edges (Underwood et al. 1991 Boesch et al. 1994). Additionally, reduction in wave energy may encourage sediment settling and increase water clarity, resulting in increased production of submerged aquatic vegetation (SAV). Increased sediment settling additionally may decrease pond depths, increase soil fertility, and provide a more hospitable environment for the expansion of emergent vegetation. Terraces do not return an area to a close approximation of its condition prior to marsh loss and thus are classified as management by some (NRC 1992) but as restoration by others (Boesch et al. 1994 Stead and Hill 2004 Feagin and Wu 2006). Terracing is thought to improve wildlife habitat by increasing the amount of edge (the boundary between emergent vegetation and open water) within a pond (Rozas and Minello 2001). Shallow marsh edge frequently has been noted as a highly productive zone for waterbird forage such as plants, nekton, and invertebrates (Gosselink 1979 Peterson and Turner 1994 Chesney et al. 2000 Minello and Rozas 2002). Adding terraces to open water ponds can also increase habitat interspersion (mixing of open water and vegetated marsh habitats) by increasing the amount of emergent vegetated edge in open water ponds. A relationship between interspersion and wetland use by waterbirds has been noted for marshes elsewhere (Weller and Spatcher 1965 Mack and Flake 1980 Kaminski and Prince 1981 Murkin et al. 1982 Fairbairn and Dinsmore 2001). Thus, increasing the propor- tion of marsh edge around the perimeter and interior portions of marsh ponds could improve habitat quality for waterbirds. However, experimental tests of this prediction are few, and we are unaware of any in coastal marshes. Previous studies have evaluated some effects of terrac- ing. Steyer (1993) showed that terracing at Sabine National Wildlife Refuge increased primary productivity through the creation and expansion of emergent marsh. Terracing also increased the amount of emergent marsh to water edge, although terrace fields do not mimic natural marsh in shape or habitat complexity (Feagin and Wu 2006). Four studies suggested that terraced edge had more nekton biomass than open water controls and changed nekton community composition (Rozas and Minello 2001 Thom et al. 2004 Rozas et al. 2005 La Peyre et al. 2007). Cannaday (2006) concluded that terracing increased SAV abundance at both marsh edge and whole-pond scales. The efficacy of terraces at improving habitat quality for waterbirds, which depend heavily on coastal marshes, has not been evaluated. Efficacy of terraces at improving marsh functions can be measured at two scales. First, effects can be compared between areas directly adjacent to terraces edges (restora- tion condition) and open water habitat far from any edge (unrestored condition). Additionally, effects could be evaluated at the whole-pond scale. Thus far, only Cannaday (2006) has conducted a whole-pond analysis of terrace effects (on SAV communities) using multiple sites. We evaluated the effect of pond terracing on waterbird habitat by comparing waterbird density and species richness at microhabitat and whole-pond scales between ponds restored with terraces and unrestored ponds. We also evaluated whether bird density varied by foraging guild between habitats. Finally, we compared SAV, nekton, and water quality at the microhabitat and whole-pond scales in restored and unrestored ponds, and evaluated whether terracing effects on nekton and SAV were influencing waterbird densities. 126 Wetlands (2010) 30:125���135