Southeastern Naturalist 561 C. Smith, E.H. Boughton, and S. Pierre 22001155 SOUTHEASTERN NATURALIST 1V4o(3l.) :1546,1 N–5o7. 63 Pomacea maculata (Island Apple Snail) Invasion in Seasonal Wetlands on Florida Ranchland: Association with Plant- Community Structure and Aquatic-Predator Abundance Colleen Smith1,3,*, Elizabeth H. Boughton1, and Steffan Pierre2 Abstract - The invasive Pomacea maculata (= Pomacea insularum) (Perry) (Island Apple Snail) is becoming increasingly abundant in the southeastern US with potentially detrimental effects on wetland vegetation and water quality. Here, we investigate the association of plant-community structure and aquatic-predator abundance with Island Apple Snail invasion in seasonal wetlands on a cattle ranch in south-central Florida. We found a negative correlation between Island Apple Snail abundance and abundance of its preferred forage species, raising concerns that the snail may have already altered plant communities in these seasonal wetlands. We also found a negative correlation between Island Apple Snail abundance in wetlands and distance to the nearest ditch. In a snail-growth experiment, we found a negative effect of species evenness and a positive effect of total vegetation cover on Island Apple Snail weight gain. To understand Island Apple Snail invasion success in seasonal wetlands, more research is needed on the relative importance of landscape- versus local-scale wetland characteristics and how resources such as preferred forage versus egglaying sites affect snail-population growth. Introduction Invasive species are a major driver of global environmental change. In addition to being economically harmful, species invasions reduce native biodiversity and change community composition, often interacting with other global-change drivers to do so (Vitousek et al. 1996). The introduction of invasive species into wetlands is an issue of particular concern to humans because wetlands provide many important ecosystem services including mitigating the impacts of storms and floods and supplying water (Costanza et al. 1997), and acting as sources of biodiversity (Zedler and Kercher 2005). The introduction of invasive species into wetlands has the potential to disrupt these ecosystem services by changing community composition and nutrient levels and decreasing wetland diversity (e.g, Fickbohm and Zhu 2006, Silliman and Bertness 2004). Invasive invertebrates in wetlands can impact multiple trophic levels in a community through competition, predation, and grazing, and can even have cascading effects on nutrient levels (e.g., Carlsson et al. 2004). Invasive apple snails (Pomacea spp., see Appendix 1 for common and scientific names and authority of all study species) are invertebrates known to change wetland structure and function, and they have been ranked first in pest potential out of 1MacArthur Agro-ecology Research Center, 300 Buck Island Ranch Road, Lake Placid, FL 33852. 2Department of Biology, University of Central Florida, Orlando FL 32816. 3Current address - Department of Ecology, Evolution, and Natural Resources, Rutgers University, New Brunswick, NJ 08901. *Corresponding author - colleen.m.smith@rutgers.edu. Manuscript Editor: Scott Markwith Southeastern Naturalist C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 562 non-native mollusks in the US (Cowie et al. 2009). Native to South America, Pomacea maculata (Island Apple Snail), is becoming a non-native of growing concern in Florida, Georgia, and Texas (Rawlings et al. 2007). The first genetically confirmed specimen of the snail was collected in Florida in 2002, but the snail’s eggs may have been observed in the state as early as 1991 (Rawlings et al. 2007). Island Apple Snail was first observed on Buck Island Ranch, a commercial cattle ranch in Lake Placid, FL, in 2010. The snail’s spread into agricultural areas of Florida may have detrimental economic implications for both private landowners and public agencies through vegetation changes, disease spread, and water-quality alterations. The closely related Pomacea canaliculata (Golden Apple Snail) is a significant pest of Oryzopsis sativa L. (Rice) crops in southeast Asia (Halwart 1994) and has also been shown to drastically alter ecosystems, changing a wetland in Thailand from macrophyte- to algae-dominated (Carlsson et al. 2004). Less is known about the ecology of the Island Apple Snail in its invaded range, although it is possible that studies of the Golden Apple Snail have confused the 2 morphologically similar species where their ranges overlap, for example in southeast Asia (Horgan et al. 2014). Laboratory studies conducted on the Island Apple Snail have demonstrated that the species is a generalist herbivore that eats several types of wetland plants and has high rates of consumption and fecundity (Baker et al. 2010, Barnes et al. 2008, Burlakova et al. 2008, Morrison and Hay 2011). In some cases, the consumption of vegetation is so high that it can be detrimental to water quality. In humanmade Typha sp. (cattail) marshes functioning as storm-water treatment areas (STA) in South Florida, the snail defoliated enough submerged aquatic vegetation to increase phosphorous outflow and cause 1 STA to be shut down until the snails could be controlled (Andreotta et al. 2015). Along with its potentially harmful impacts on wetland structure and function, the invasive snail has been found to serve as a food source for endangered bird species such as Rostrahamus sociabilis (Snail Kite) and Aramus guarauna (Limpkin) (Cattau et al. 2010, Tanaka et al. 2006). Compared to data from historical bird surveys, both of these birds have been more frequently observed at Buck Island Ranch since the arrival of the invasive snail (E.H. Boughton, pers. observ.). It is not clear if other predators of the Island Apple Snail are present in Florida, although some mammals (i.e., Procyon lotor [Raccoon] and Lontra canadensis [North American River Otter, hereafter, Otter]) appear to prey on the snail. Aquatic predators such as fish and Belostoma spp. (giant water bugs) could also prey on apple snail young. Little is known about Island Apple Snail ecology in the ditches and seasonal wetlands where it is introduced. Most research on the snail to date was conducted in the laboratory (e.g., Burlakova et al. 2008, Morrison and Hay 2011). Feeding trials show that the Island Apple Snail is an indiscriminate consumer of aquatic wetland plants that consumed 46 out of 52 macrophyte species tested (Horgan et al. 2014). Many of these plant species are present in the ditches and seasonal wetlands embedded in cattle ranches, an important land use in Florida that composes ~1/6th of the state’s land and ~404,686 ha in the Northern Everglades, where the Southeastern Naturalist 563 C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 hydrology of the watershed is modified by miles of ditches and canals (Swain et al. 2013). The successful invasion of apple snails into wetlands and ditches on cattle ranches is likely affected by connectivity of water sources and wetland abiotic and biotic attributes. Because the Island Apple Snail is an herbivore, its invasion success into some wetlands may be affected by attributes of the wetland-plant community such as richness. Plant communities that are species rich may have a higher probability of containing preferred food items. The Island Apple Snail may be a potential prey item for a multitude of species, so the wetland aquatic-animal community could also affect snail invasion success. Given that little is known about the factors that affect Island Apple Snail invasion success, we investigated the association of landscape characteristics, plant diversity, and potential aquatic-predator abundance with Island Apple Snail abundance and growth in seasonal wetlands on a cattle ranch in south-central Florida. We also surveyed 2 large ditches to examine the plant species utilized by the Island Apple Snail for egg-laying. Methods Site selection We conducted our study at MacArthur Agro-ecology Research Center (MAERC; a division of Archbold Biological Station) located on Buck Island Ranch, a commercial cattle ranch in Lake Placid, FL. The ranch contains 2 pasture types, improved pasture and semi-native pasture, which differ in species composition, grazing pressure, and history. Both contain seasonal wetlands that typically hold water for ~6 months of the year. Improved pastures are dominated by the introduced forage grass Paspalum notatum (Bahia Grass), are fertilized annually with nitrogen, and were fertilized with phosphorous until 1986. They are also grazed more heavily than semi-native pastures. Semi-native pastures were historically unfertilized, and their wetlands are more species rich than those in improved pastures (Boughton et al. 2010). Island Apple Snail egg masses were first observed on Buck Island Ranch in 2010 on culverts in Harney Pond Canal, which borders about half of the property (Fig. 1). They were subsequently observed on culverts and emergent vegetation in internal ditches and then in wetlands. The snails’ egg masses appeared to be especially abundant in the ranch’s East Marsh, so we chose wetlands in and around this area of the ranch for both parts of our study (Fig. 1). We chose a total of 12 wetlands, 6 in improved pasture and 6 in semi-native pasture (Table 1). In 2007, all wetlands involved in the study were fenced to exclude cattle in 2007 as part of a different study. We selected fenced wetlands to insure that cattle would not interfere with our experimental enclosures. To understand the snail’s distribution along the ditches in the East Marsh, we conducted preliminary ditch surveys by walking along ~3300 m of the 2 primary ditches in the East Marsh (Fig. 1) and recording the locations of each Island Apple Snail egg mass observed as well as the plant species or structure (i.e., culvert or fence post) on which the egg mass occurred. Island Apple Snails lay bright pink Southeastern Naturalist C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 564 eggs on emergent vegetation or structures, making the egg masses relatively easy to spot from ditch banks. Figure 1. Map of the wetlands surveyed in the East Marsh of Buck Island Ranch. Filled wetlands indicate that Island Apple Snail was encountered in traps and/or empty shells were encountered along study transects. Outlined wetlands with no fill indicate that the Island Apple Snail was not encountered. Southeastern Naturalist 565 C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 Wetland surveys We measured relative Island Apple Snail abundance and the relative abundance of its potential predators using un-baited crayfish traps, which are effective tools for measuring apple snail abundance when mark–recapture methods are used (Darby et al. 1997). We were able to conduct trapping for apple snails in 6 wetlands (4 in improved pasture and 2 in semi-native pasture) given time constraints imposed by the drying down of wetlands at the end of Florida’s wet season in November. We placed 8 crayfish traps in each of the 6 wetlands for 6 consecutive days between 21 October 2013 and 26 November 2013 at water depths between 15 cm and 45 cm. We checked traps daily for the presence of the Island Apple Snail as well as any omnivorous or carnivorous organisms, which we considered the snail’s potential predators. We marked with identification numbers all Island Apple Snail individuals caught in traps so that we could calculate recapture probability. However, we did not recapture any individuals, and thus, we used the number of Island Apple Snails trapped as a measure of relative abundance. We released trapped individuals 10 steps away from the trap. We also recorded the identity and number of potential predators (snakes, sirens, turtles, fish, and crayfish) and identified fish and crayfish to species before we released them. To obtain additional data on the relative abundance of potential Island Apple Snail aquatic predators in each of the 12 wetlands, we dip-netted for macroinvertebrates, crayfish, and fish at 10 random points, evenly distributed among the center, north, south, east, and west quadrants in each wetland. We dip-netted twice at each Table 1. Size, water-quality, and dominant vegetation of wetlands used in the surveys and enclosure experiment. E = enclosure experiment, S = shell survey, T = trapping survey, SN = semi-native pasture, I = improved pasture, TP = total phosphorous, TN = total nitrogen, ALTPHI = Alternanthera philoxeroides, AMPMUH = Amphicarpum muehlenbergianum, CENASI = Centella asiatica, CLAJAM = Cladium jamaicense, DIOVIR = Diodia virginiana, HYMAMP = Hymenachne amplexicaulis, LEMMIN = Lemna minor, PANHEM = Panicum hemitomon, PASACU = Paspallum acuminatum, POLPUN = Polygonum punctatum, RHYINU = Rhynchospora inundata, and SALMIN = Salvinia minima. Wetland Area Depth TP TN Snails ID Used for Pasture (m2) (cm) pH (ppm) (ppm) (Y/N) Dominant vegetation 49 S, E SN 2.02 21 4.53 0.06 2.41 N AMPMUH, RHYINU 84 S SN 0.78 25 4.84 0.08 1.89 Y CLAJAM, PANHEM 91 T, S SN 1.27 47 5.17 0.10 2.83 Y PANHEM, PASACU 148 T I 2.50 60 5.22 0.29 4.22 N ALTPHI, SALMIN 151 T I 3.08 53 4.52 0.55 1.02 N PANHEM, ALTPHI 196 S I 1.67 37 6.03 0.23 3.33 N ALTPHI, HYMAMP 210 S, E SN 1.06 45 4.32 0.16 3.08 N CENASI, RHYINU 225 S SN 2.20 62 4.6 0.09 2.84 N PANHEM, DIOVIR, CENASI 258 T, S, E SN 2.02 48 5.04 0.13 1.76 Y ALTPHI, HYMAMP 271 T, E I 1.85 38 5.38 0.17 3.48 Y ALTPHI, LEMMIN 295 T, S, E I 1.10 46 6.15 0.42 4.33 Y ALTPHI, PANHEM 317 T I 2.31 50 5.98 0.71 3.72 Y LEMMIN, ALTPHI, POLPUN Southeastern Naturalist C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 566 point and recorded the number and identity of all fish, crayfish, and water beetles. We considered the total number of potential aquatic predators caught by trapping and by dip-netting in a wetland to be the wetland’s relative abundance of potential aquatic predators. Because wetlands were fenced, many terrestrial predators were likely excluded, including Otters and Raccoons, and our sampling technique did not permit us to measure the abundance of these and potential avian predators such as Limpkins and Snail Kites. To measure plant-species richness and community composition in the 12 wetlands, we recorded the identity of all plant species in 1-m2 quadrats at 15 random stratified points (stratified by center, north, south, east, and west) in each wetland between 25 September and 31 October 2013. To obtain additional data on Island Apple Snail population sizes and predation in the wetlands, we counted the number of empty shells observed along random transects that crossed 9 of the wetlands that were sufficiently dry to conduct sampling by 9 December 2013. For each wetland, we recorded the number of empty Island Apple Snail shells encountered within 1 m on either side of the transect. Given that the empty shells we encountered were generally large in size, and that shells were often grouped around bird nests, these shells were probably the remains of bird predation. Limpkins are sighted frequently in wetlands at Buck Island Ranch. We used the number of Island Apple Snail shells encountered in each wetland as an indirect measure of the snail’s abundance, assuming that birds predate upon Island Apple Snails primarily in the wetland where their nest is sited and most heavily in wetlands where the snail is common. This measure of snail abundance is not ideal because snail shells can persist for a long time, and predatory activity many not be uniform across wetlands, therefore results based on empty shells must be interpreted with caution. In September and October 2006 and again in October 2008, we took pH, phosphorous, and nitrogen measurements at 3 stratified points. We collected a grab sample at each of the 3 points using a 1-m-long sampling probe. The grab samples were collected at the central staff gauge, 10 m from the staff gauge and 2–3 m from the edge. At each sampling point, acid-washed 100-ml bottles were filled with unfiltered water and preserved with H2SO4. We measured pH with a protable YSI 556 multi-probe meter. Data analysis: wetland surveys We used Spearman’s rank correlation to examine the relationship of habitat factors (plant richness, relative abundance of fish and insects, relative abundance of structural plants and preferred plants [defined below], and distance to nearest ditch) with abundance of trapped Island Apple Snails in each wetland (n = 6). We used linear regression with a model-selection approach to determine which factors were associated with the number of empty Island Apple Snail shells encountered along transects in each wetland (n = 9). We selected the model with the lowest corrected Akaike information criterion (AICc) value and the highest weight (Burnham and Anderson 2002). In total, we assessed 8 models including a null Southeastern Naturalist 567 C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 model. All analyses were conducted in the R statistical program (R Development Core Team 2008). The independent variables were plant species richness, distance to the nearest ditch, wetland size, fish and insect relative abundance, relative frequency of preferred forage species (Sagittaria lancifolia [Bulltongue Arrowhead], Eicchornia crassipes [Water Hyacinth], Limnobium spongia [American Spongeplant], Panicum repens L. [Torpedo Grass], Utricularia spp. [bladderworts], and Bacopa caroliniana (Walter) B.L. Rob. [Blue Waterhyssop]; Baker et al. 2009, Burlakova et al. 2008, Morrison and Hay 2011), and relative frequency of structural plants (Hymenachne amplexicaulis [West Indian Marsh Grass], Water Hyacinth, Myrica cerifera [Wax Myrtle], Pontederia cordata [Pickerelweed], Bulltongue Arrowhead, Thalia geniculata [Bent Alligator-flag], American Spongeplant, Cephalanthus occidentalis [Buttonbush], Eupatorium capillifolium [Dog-fennel], Sabal palmetto [Cabbage Palm], Sesbania herbacea [Bigpod Sesbania], and Juncus effusus [Common Rush]). These structural plant species have suitably rigid stems for Island Apple Snail to use for egg laying and include all plant species on which we encountered 5 or more egg masses during the initial ditch survey, in addition to species with rigid stems and/or leaves that are common in wetlands but not ditches, and thus would not have been observed in the ditch survey. We also examined the effect of pasture type on snail abundance measures using Mann-Whitney U tests. Enclosure experiment We conducted enclosure experiments to directly test the role of plant-species richness on the growth rate of the Island Apple Snail. We placed three 1.6-m tall x 0.5 m2 circular wire-mesh enclosures in each of 6 wetlands—3 in improved pasture and 3 in semi-native pasture. We used garden staples to keep enclosures in the ground and covered the tops of the enclosures with mesh cloth to keep snails from escaping and to keep birds from consuming snails. For each pasture type, the number of plant species in enclosures ranged from 1 to 9. Each wetland contained 3 enclosures, 1 each of low, medium, and high species richness (1–3, 4–6, and 7–9 plant species, respectively). In addition, we established three 0.5- m2 control plots in each wetland with equivalent plant species richness. We chose enclosure sites and control plots by surveying wetlands with 0.5-m2 hoops to find the appropriate richness levels at water depths that ranged from 20 to 40 cm. To account for differences between enclosures, we measured total plant percent cover in each and included this as a covariate. We collected Island Apple Snails for the enclosure experiment from wetlands and ditches on Buck Island Ranch that were not part of the study. Before adding snails to enclosures, we weighed individuals and measured shell length and height. We grouped snails into a large-size class and a small-size class. Snails in the smallsize class ranged in weight from 2.0 g to 9.1 g, and snails in the large-size class ranged in weight from 35.8 g to 80.0 g. In October of 2013, we marked 1 large and 1 small snail with nail polish for each enclosure, placed them in the enclosure; left them for 4–6 weeks, and then removed and weighed them. Southeastern Naturalist C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 568 Data analysis: enclosure experiment We used Spearman’s rank correlation to determine which factors were associated with Island Apple Snail percent weight change in enclosures. The covariates were plant-species richness, total plant percent cover, evenness, pH, total P, and total N. We used Mann-Whitney U tests to analyze the effect of pasture type (improved pasture or semi-native pasture). In this analysis, we averaged snail response from each enclosure and considered snail response an independent replicate, resulting in 12 snail responses collected from 5 wetlands (Table 2). Note that we retrieved no snails in 1 wetland (wetland 49). Table 2. Plant-species composition and number of snails retrieved for each enclosure. An “e” in the “Snails retrieved” column denotes that the snail was found estivating. ID = enclosure ID, n = plant richness, ALTPHI = Alternanthera philoxeroides, AZOLLA = Azolla sp., CARABL = Carex albolutescens, COMDIF = Commelina diffusa, DIOVIR = Diodia virginiana, EUPCAP = Eupatorium capilllifolium, HYMAMP = Hymenachne amplexicaulis, JUNEFF = Juncus effusus, JUSANG = Justicia angusta, LEEHEX = Leersia hexandra, LEMMIN = Lemna minor, LIMSPO = Limnobium spongia, LUDREP = Ludwigia repens, PANHEM = Panicum hemitomon, PASACU = Paspalum acuminatum, PHYNOD = Phyla nodiflora, PONCOR = Pontedaria cordata, POLPUN = Polygonum punctatum, PROPEC = Proserpinaca pectinata, SACSTR = Sacciolepis striata, SAGLAN = Sagittaria lancifolia, THAGEN = Thalia geniculata, and UTRFOL = Utricularia foliosa. Wetland Total % Snails % weight ID ID n Plant species cover retrieved change 1 49 2 PANHEM, SAGLAN 60 1-e - 2 5 LEEHEX, PANHEM, PONCOR, SACSTR, SAGLAN 67 0 - 3 8 HYMAMP, LUDREP, PANHEM, POLPUN, PROPAL, 89 1-e - SACSTR, SAGLAN, UTRFOL 4 151 2 ALTPHI, PANHEM 85 1 2.68 5 5 ALTPHI, AZOLLA, LIMSPO, PANHEM, POLPUN 88 1 -0.02 6 8 ALTPHI, AZOLLA, JUNEFF, LEEHEX, LEMMIN, 137 2 0.42 PANHEM, POLPUN, SACSTR 7 210 3 HYMAMP, PASACU, POLPUN 32 1 -0.09 8 6 LEEHEX, DIOVIR, JUSANG, PANHEM, PASACU, 39 1 -0.17 SAGLAN 9 9 HYMAMP, DIOVIR, JUSANG, PASACU, PANHEM, 61 1 -0.06 PONCOR, PROPEC, SAGLAN, UTRFOL 1 0 258 1 LEMMIN 42 2 0.08 1 1 4 ALTPHI, HYMAMP, JUNEFF, POLPUN 77 1 0.08 12 7 ALTPHI, AZOLLA, HYMAMP, JUNEFF, LEMMIN, 82 1 0.05 POLPUN, UTRFOL 1 3 271 1 PANHEM 45 1 0.06 1 4 5 ALTPHI, HYMAMP, LEMMIN, PANHEM, POLPUN 34 1 -0.08 15 7 ALTPHI, CARABL, JUNEFF, LEMMIN, PHYNOD, 83 0 - POLPUN, THAGEN 1 6 295 3 PANHEM, POLPUN, LEMMIN 70 1 0.03 17 6 ALTPHI, COMDIF, EUPCAP, LEMMIN, PANHEM, 65 2-e - POLPUN 18 9 SAGLAN, EUPCAP, POLPUN, LEMMIN, PANHEM, 82 1-e - COMDIF, ALTPHI, CARABL, LEEHEX Southeastern Naturalist 569 C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 Results Ditch and wetland surveys We observed 509 egg masses on 17 plant species over 4 d of sampling between 12 August and 20 August 2013. The plants on which we observed egg masses tended to be either woody shrubs or herbaceous plants with sturdy stems (Table 3). In total, Island Apple Snail occurred in 6 out of the 12 wetlands studied (Table 1, Fig. 1). We encountered the snail in 4 out of the 6 wetlands in which we conducted trapping and in 5 out of the 9 wetlands in which we conducted transect surveys (Table 1). Wetland survey: trapping Island Apple Snail abundance obtained by trapping was negatively correlated with the frequency of preferred plant species (ρ = -0.91, P = 0.01; Fig. 2). There was no relationship between Island Apple Snail abundance and any other variable— plant richness (ρ = -0.09, P = 0.87), fish abundance (ρ = 0.46, P = 0.35), insect abundance (ρ = -0.44, P = 0.38, frequency of structural plants (ρ = 0.43, P = 0.39), or distance to the nearest ditch (ρ = 0.66, P = 0.15). There was no difference in snail abundance between wetlands embedded in improved or semi-native pastures (P = 0.24). Wetland survey: transects The model that best explained the abundance of empty Island Apple Snail shells per wetland contained the main effect of distance to the nearest ditch (Table 4). The Table 3. The number of egg masses encountered on plant species and manmade structures. Plant species or structure Common name Number of egg masses Hymenachne amplexicaulis West Indian Marsh Grass 223 Cephalanthus occidentalis Buttonbush 54 Myrica cerifera Wax Myrtle 54 Juncus effusus Common Rush 40 Cladium jamaicense Saw-grass 20 Culvert 18 Dead woody material 18 Eichhornia crassipes Water Hyacinth 18 Pontederia cordata Pickerelweed 16 Sesbania herbacea Bigpod Sesbania 14 Sabal palmetto Cabbage Palm 11 Sagittaria lancifolia Bulltongue Arrowhead 7 Eupatorium capillifolium Dogfennel 6 Fence 2 Quercus virginiana Live Oak 2 Alternanthera philoxeroides Alligator Weed 1 Andropogon virginicus Broomsedge Bluestem 1 Paspalum notatum Brook Crowngrass 1 Polygonum punctatum Dotted Smartweed 1 Spartina bakeri Sand Cordgrass 1 Southeastern Naturalist C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 570 coefficient of the main effect of distance from ditch was significant (coefficient = -0.4952, P < 0.001) and was negatively related to shell abundance. There was no difference in snail abundance between wetlands embedded in improved or seminative pastures (P = 0.59). Enclosure experiment Snails were kept in enclosures with different plant composition and richness in 6 different wetlands for 4–6 weeks (Table 2). Some wetlands had partially dried down by the end of this time period, causing snails in 5 enclosures to begin hibernating belowground. When possible, we obtained these snails by digging, but did not use these estivating snails’ weight changes as part of the data set because weight loss occurs during estivation (Table 2, Little 1968). We calculated a total recovery rate of 38.8%. Table 4. The top 4 models that explained abundance of Island Apple Snail shells per wetland. The best model was the direct effect of distance to the nearest ditch. K = number of parameters, Log(£) = maximum log-likelihood, AICc = corrected Akaike information criterion value, ΔI = difference between the lowest AIC value and AICi, wI = model weight, D = distance to the nearest ditch, M = macroinvertebrate abundance, and S = relative frequency of structural plants. Model df K Log(£) AICc ΔI wI D 7 2 -49.29 104.57 0 1.00 M 7 2 -58.59 123.18 18.61 9.09 x 10-5 S 7 2 -70.29 146.58 42.01 7.53 x 10-10 Null 8 1 -74.22 151.02 46.45 8.20 x 10-11 Figure 2. Snail abundance obtained from trapping has a negative relationship with the frequency of its preferred forage species (ρ = -0.91, P = 0.01). Southeastern Naturalist 571 C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 Of the snails that we collected, percent change in weight was positively related to the total amount of plant cover and negatively related to evenness (Table 5, Fig. 3). There was a marginal negative relationship between change in weight and total N (Table 5). There was no difference in weight change between snails from the two pasture types (P = 0.39). Discussion There is a growing concern about the invasive Island Apple Snail in Florida because of its potentially detrimental effects on wetland vegetation, water quality, and disease spread. In one of the only field studies conducted t o date on the Island Apple Snail in seasonal wetlands, we showed a negative association between snail abundance and the frequency of its preferred forage species, raising concerns about the potential for the snail to alter plant communities in seasonal wetlands in the southeastern US. We also found associations between the snail’s abundance and the wetland’s distance to the nearest ditch, and between its growth and plant-species evenness and percent cover. Below we discuss the implications of our findings and include some caveats. The results of the ditch survey suggest that the Island Apple Snail oviposits on thick-stemmed and woody vegetation. We found almost half of all egg-masses on the invasive West Indian Marsh Grass, which forms dense monocultures in wetlands and ditches (Diaz et al. 2013). Although lack of data on ditch plant-community composition did not allow us to assess the snail’s oviposition preferences, the results of our survey suggest that West Indian Marsh Grass management may also reduce the abundance of egg-laying substrate for Island Apple Snail. Interestingly, it is probable that much of the vegetation the snail uses for oviposition are not palatable to it, perhaps due to the thick cell walls and high lignin content of the emergent vegetation necessary to support egg masses above water. Burlakova et al. (2008) found that Island Apple Snails preferred to eat plant species with low dry-matter content, an indication of low cellulose and lignin content, and they recommended that plants with high dry-matter content be used for restoration. However, because it is likely that many of the plants on which Island Apple Snail oviposits have high dry-matter content, these restoration efforts may not actually aid in the snail’s suppression, depending on whether food or laying substrate is the limiting factor in a given wetland community. Future studies should examine what life-history stages Table 5. Spearman’s rank correlations between the independent variables and Island Apple Snail percent weight change in the enclosure experiment. Variable ρ P Plant richness -0.29 0.35 Total percent cover 0.64 0.03 Evenness -0.61 0.04 pH 0.30 0.34 Total P 0.18 0.59 Total N -0.55 0.06 Southeastern Naturalist C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 572 are most important to snail invasion and how abundance of emergent structures affects restoration efforts. In the wetland survey, we found a negative relationship between Island Apple Snail abundance and the frequency of its preferred forage species. This result Figure 3. Percent weight change of snails in enclosures (A) increases with percent cover (with outlier [removed from graph for clarity]) ρ = 0.64, P = 0.03 (without outlier ρ = 0.61 and P = 0.051) and (B) decreases with evenness (with outlier [removed from graph for clarity] ρ = -0.61, P = 0.04 and without outlier ρ = -0.545 and P = 0.087). Percent cover of vegetation can be greater than 100 because percent cover of submerged and emergent vegetation in enclosures was measured separately. Southeastern Naturalist 573 C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 suggests that the Island Apple Snail may impact vegetation in seasonal wetlands by consuming and thus driving down the abundance of certain wetland plants, especially submerged aquatic vegetation. This result is concerning because many of these preferred species are native, and because such reductions in vegetation could increase wetland nutrient levels (Andreotta et al. 2015). However, our sample size was small (6 wetlands), and trapping is generally influenced both by time-of-year and site-to-site variation (Elzinga et al. 2001). Importantly, this relationship is correlative, and we did not measure the direct cause of the negative relationship. More research on the snail’s effect on plant-community composition in seasonal wetlands is greatly needed. We conducted additional surveys of snail-shell abundance in 9 dried-down wetlands. Our analysis showed that distance to the nearest ditch was negatively related to the abundance of snail shells. More shells may be found in wetlands closer to ditches because wetlands closer to ditches experience higher propagule pressure; snails experience more predation in wetlands near ditches, leaving behind more shells; or wetlands close to ditches were invaded earlier and more shells accumulated prior to sampling. It is important to note that snail-shell abundance is an indirect measure of actual snail abundance and may be affected by wetland vegetation characteristics that affect apple snail burrowing. More research is needed to understand how ditch connectivity and other landscape-scale characteristics affect snail invasion. Contrary to expectations, plant-species richness in the enclosure experiment, was not related to Island Apple Snail growth. There was a significant positive relationship between total plant percent cover and snail growth, and an unexpected negative effect of plant-species evenness on Island Apple Snail weight change. One explanation for the negative trend is that, although this snail is a generalist, it exhibits preferences and performs optimally when consuming certain plant species over others (Morrison and Hay 2011). Using our experimental design, we were not able to determine which plant species the Island Apple Snail consumes, making it difficult to assess why there was a negative trend between evenness and weight gain. Implications of Island Apple Snail invasion in ranchland wetlands Our results suggest that managers aiming to prevent introductions of the Island Apple Snail should focus on wetlands close to propagule sources such as ditches. Additionally, to mitigate the snail’s impact, special concern should be paid to conserving the snail’s preferred forage species. The results of this study are preliminary and more work is needed to understand Island Apple Snail invasion in seasonal wetlands. Future studies should examine the relative importance of food versus egg-laying resources for snail-population growth, the effect of the snail on wetland nutrient levels, and the effect of landscape characteristics and propagule pressure on the snail’s invasion success. Acknowledgments Thank you to Julia Maki, Jessica Franks, and Vi Bui for help with field work. We would also like to thank 2 anonymous reviewers for thoughtful comments that improved Southeastern Naturalist C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 574 the manuscript. C. Smith was funded as a research intern by Archbold Biological Station. USDA funding (USDA CSREES FLAR-2006-01378) was used to fence the study wetlands in 2007. This is contribution #162 from the MacArthur Agro-ecology Research Center. Literature Cited Andreotta, H., M. Chimney, T. DeBusk, B. Garrett, L. Gerry, J. Henry, D. Ivanoff, M. Jerauld, M. Kharbanda, M. Kirkland, N. Larson, S. Miao, T. Piccone, K. Pietro, L. Schwartz, D. Sierer-Finn, L. Toth, S.K. Xue, Y. Yan, M. Zamorano, and H. Zhao. 2015. Performance of the Everglades stormwater treatment areas Pp. 5B–11, In M. Chimney (Ed.). 2015 South Florida Environmental Report. 83 pp. Baker, P., F. Zimmanck, and S.M. Baker. 2010. Feeding rates of an introduced freshwater gastropod Pomacea insularum on native and nonindigenous aquatic plants in Florida. Journal of Molluscan Studies 76:138–143. Barnes, M.A., R.K. Fordham, and R.L. Burks. 2008. Fecundity of the exotic apple snail Pomacea insularum. Journal of the North American Benthological Society, 27(3):738–745. Boughton, E.H., P.F. Quintana-Ascencio, P.J. Bohlen, D.G. Jenkins, and R. Pickert. 2010. Land-use and isolation interact to affect wetland-plant assemblages. Ecography 22:461–470. Burlakova, L.E., A.Y. Karatayev, D.K. Padilla, L.D. Cartwright, and D.N. Hollas. 2008. Wetland restoration and invasive species: Apple snail (Pomacea insularum) feeding on native and invasive aquatic plants. Restoration Ecology 17:433–440. Burnham, K.P., and D.R. Anderson. 2002. Model Selection and Multimodel Inference: Apractical Information-theoretic Approach. Springer Publishing, New York, NY. 488 pp. Carlsson, N.O.L., C. Bronmark, and L.A. Hansson. 2004. Invading herbivory: The Golden Apple Snail alters ecosystem functioning in Asian wetlands. Ecology 85:1575–1580. Cattau, C.E., J. Martin, and W.M. Kitchens. 2010. Effects of a non-native prey species on a native specialist: Example of the Snail Kite. Biological Conservation 143:513–520. Costanza, R., R. d’Arge, R. de Groots, S. Farber, M. Grasso, B. Hannon, K. Limburg, S. Naeem, R.V. O’Neill, 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. Cowie, R.H., R.T. Dillon Jr., D.G. Robinson, and J.W. Smith. 2009. Alien non-marine snails and slugs of priority quarantine importance in the United States: A preliminary risk assessment. American Malacological Bulletin 27:113–132. Darby, P.C., P.L. Valentine-Darby, R.E. Bennetts, J.D. Croop, H.F. Percival, and W.M. Kitchens. 1997. Ecological studies of apple snails. Final report submitted to the South Florida and St. Johns River Water Management Districts. Special Publication SJ98-SP6, Palatka, FL. Diaz, R., W.A. Overholt, B. Sellers, and J.P Cuda. Wetland weeds: West Indian Marsh Grass (Hymenachne amplexicaulis). EDIS #ENY693. Institute of Food and Agricultural Sciences, University of Florida. Available online at https://edis.ifas.ufl.edu/in491. Accessed 20 July 2015. Elzinga,C.L., D.W. Salzer, J.W. Willoughby, and J/P. Gibbs. 2001. Monitoring Plant and Animal Populations. Blackwell Science, Inc., Malden, MA. 360 pp. Fickbohm, S.S., and W.X. Zhu. 2006. Exotic Purple Loosestrife invasion of native cattail freshwater wetlands: Effects on organic matter distribution and soil nitrogen cycling. Applied Soil Ecology 32:123–131. Southeastern Naturalist 575 C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 Halwart, M. 1994. The Golden Apple Snail, Pomacea canaliculata, in Asian rice farming systems: Present impact and future threat. International Journal of Pest Management 40(2):199–206. Horgan, F.G., A.M. Stuart, and E.P. Kudavidanage. 2014. Impact of invasive apple snails on the functioning and services of natural and managed wetlands. Acta Oecologica 54:90–100. Little, C. 1968. Aestivation and ionic regulation in two species of Pomacea. Journal of Experimantal Biology 48:569–585. Morrison, W.E., and M.E. Hay. 2011. Feeding and growth of native, invasive, and noninvasive alien apple snails (Ampullariidae) in the United States: Invasives eat more and grow more. Biological Invasions 13:945–955. R Development Core Team. 2008. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available online at http:// www.R-project.org. Accessed 20 July 2015. Rawlings, T.A., K.A. Hayes, R.H. Cowie, and T.M. Collins. 2007. The identity, distribution, and impacts of non-native apple snails in the continental United States. BMC Evolutionary Biology 7:97. Silliman, B.R., and M.D. Bertness. 2004. Shoreline development drives invasion of Phragmites australis and the loss of plant diversity on New England salt marshes. Conservation Biology 18(5): 1424–1434. Swain, H.M., E.H. Boughton, P.J. Bohlen, and L.O. Lollis. 2013. Trade-offs among ecosystem services and disservices on a Florida ranch. Rangelands 5:75–87. Tanaka, M.O., A.L.T. Souza, and E.S. Módena. 2006. Habitat-structure effects on size selection of Snail Kites (Rostrhamus sociabilis) and Limpkins (Aramus guarana) when feeding on apple snails (Pomacea spp.) Acta Oecologica 30(1):88–96. Vitousek, P.M., C.M. D’Antonio, M. Carla, L.L. Loope, and R. Westbrooks. 1996. Biological invasions as global environmental change. American Scientist 84(5):468–478. Zedler, J.B., and S. Kercher. 2005. Wetland resources: Status, trends, ecosystem services, and restorability. Annual Review of Environment and Resources 30:39–74. Southeastern Naturalist C. Smith, E.H. Boughton, and S. Pierre 2015 Vol. 14, No. 3 576 Appendix 1. The scientific name, authority, and common name of all species discussed. Scientific name with authority Common name Alternanthera philoxeroides Mart. (Standl.) Alligator Weed Amphicarpum muehlenbergianum Chapm. Muhlenberg Maidencane Andropogon virginicus L. Broomsedge Bluestem Aramus guarauna L. Limpkin Azolla sp. Mosquito Fern Bacopa caroliniana (Walter) B.L. Rob. Blue Waterhyssop Carex albolutescens Schwein. Greenwhite Sedge Centella asiatica (L.) Urb. Spadeleaf Cephalanthus occidentalis L. Buttonbush Cladium jamaicense Crantz (Kük.) Jamaica Swamp Sawgrass Commelina diffusa Burm.f. Climbing Dayflower Diodia virginiana L. Virginia Buttonweed Eichhornia crassipes (Mart.) Solms Water Hyacinth Eupatorium capillifolium (Lam.) Small Dogfennel Hymenachne amplexicaulis (Rudge) Nees West Indian Marsh Grass Juncus effusus L. Common Rush Justicia angusta (Chapm.) Small Pineland Water-willow Leersia hexandra Sw. Southern Cutgrass Lemna minor L. Common Duckweed Limnobium spongia (Bosc) Rich. ex Steud. American Spongeplant Lontra canadensis Schreber North American River Otter Ludwigia repens J.R. Frost Creeping Primrose-willow Myrica cerifera (L.) Small Wax Myrtle Panicum hemitomon Schult Maidencane Panicum repens L. Torpedo Grass Paspalum acuminatum Raddi Brook Crowngrass Paspalum notatum Flüggé Bahia Grass Phyla nodiflora (L.) Greene Turkey Tangle Fogfruit Polygonum punctatum Elliott Dotted Smartweed Pomacea canaliculata Lamarck Golden Apple Snail Pomacea maculata Perry Island Apple Snail Pontederia cordata L. Pickerelweed Proserpinaca pectinata Lam. Combleaf Mermaidweed Pryocon lotor L. Racoon Quercus virginiana Mill. Live Oak Rhynchospora inundata (Oakes) Fernald Narrowfruit Horned Beaksedge Rostrhamus sociabilis Vieillot Snail Kite Sabal palmetto (Walter) Lodd. ex.Schult. & Schult. f. Cabbage Palm Sacciolepis striata Nash American Cupscale Grass Sagittaria lancifolia L. Bulltongue Arrowhead Salvinia minima Baker Water Spangles Sesbania herbacea (Mill.) McVaugh Bigpod Sesbania Spartina bakeri Merr. Sand Cordgrass Thalia geniculata L. Bent Alligator-flag Utricularia foliosa L. Leafy Bladderwort