Distribution and population characteristics of African jewelfish and brown hoplo in modified wetlands in South Florida.
Invasive species represent one of the greatest threats to biodiversity worldwide. Nonnative animals can disrupt ecosystem processes and habitats, introduce diseases, and alter trophic interactions through predation and competitive exclusion (Cucherousset and Olden, 2011). These effects are particularly well documented for fish invaders. For example the introduction of the Nile perch (Lates niloticus) in Lake Victoria led to the extinction or extirpation of as many as 200 indigenous fish species (Kitchell et al., 1997). Likewise, the introduction of mosquitofish of the genus (Gambusia) has caused population declines and decreased growth and survival rates of amphibians in Europe, Australia, and North America (Gamradt and Kats, 1996; Morgan and Buttemer, 1996; Lawler et al., 2001; Denoel et al., 2005; Wilson, 2006).
Because of its warm subtropical climate and abundant interconnected wetlands, Florida has a particularly severe problem with establishment of nonnative freshwater fish. However, a lack of quantitative data on many aquatic communities in the region has hindered efforts to understand the effects of these invaders (Trexler et al., 2000). Of particular concern are members of the family Cichlidae which are native to Africa and the neotropics. They comprise one of the most speciose vertebrate families and many of the traits that have contributed to their evolutionary success predispose them to thrive in novel environments. Among these traits are phenotypic plasticity, physiological tolerance to a wide range of conditions, high reproductive potential, parental care, and aggressive behavior (Meyer, 1990; Balshine-Earn and Earn, 1998; Schofield et al., 2009; Langston et al., 2010). The Mayan cichlid (Cichlasoma urophthalmus) and the African jewelfish (Hemichromis letourneuxi) are now among the most abundant species in parts of the Everglades (Faunce and Lorenz, 2000; Loftus et al., 2006).
The range of the African jewelfish has recently expanded from a few populations in the Everglades to include most of the southern half of Florida, reaching as far north as Tampa (Langston et al., 2010). Although they are likely limited by cold weather (Schofield et al., 2009), their full colonization potential and ability to successfully invade different aquatic communities is still unknown. It is important to understand which communities are vulnerable so that management plans can be devised to prevent further spread (Vander Zanden and Olden, 2008).
Given that agricultural lands account for nearly 27% of the total land use in Florida (Clouser, 2005), agriculturally modified wetlands provide important habitat for many freshwater species. These wetlands are often highly interconnected via man made ditches, vary in size and length of hydroperiod, and provide suitable habitat for amphibians that breed in both seasonal ponds and permanent water bodies (Babbitt and Tanner, 2000; Knutson et al., 2004). Additionally, they can support species rich fish and macroinvertebrate communities and may be important seasonal habitat for various fish species (Leslie et al., 1997; Baber et al., 2002). Community assemblages in agricultural wetlands appear to be driven by large scale colonization processes, such as proximity to source populations and wetland connectivity, whereas extinction processes seem to be less important (Baber et al., 2002). The addition of exotic fish could alter these community dynamics and negatively affect the distribution and abundance of native organisms. Thus, it is important to understand the extent to which exotic fish species are able to invade modified wetlands and to determine the factors influencing their spread.
The Archbold Reserve, located in south-central peninsular Florida, is an example of a hydrologically altered, agricultural landscape undergoing invasion by exotic fish species. At least four exotic fish species occur on the Reserve, including the African jewelfish, which was first recorded in 2008. Another cichlid, the black acara (Cichlasoma bimaculatum), was also captured in 2008 but has not been detected since. The walking catfish (Clarias batrachus) has been established since 1980, and the brown hoplo(Hoplosternum littorale) was first reported in 2003. Our main objective was to determine the current distribution and relative abundance (measured by catch-per-unit-effort, CPUE) of exotic fish species on the Archbold Reserve and to compare the condition and diet of African jewelfish in these recently established populations with data published on other populations in their native and introduced ranges. We expected that exotic fish distribution would be limited to wetlands in close proximity to permanent streams. Further, we expected that wetland connectivity (i.e., distance from wetland edge to nearest major ditch) would influence exotic fish distribution and CPUE because ditches may facilitate colonization by acting as dispersal corridors. We also examined potential relationships between exotic fish occurrence and number of species and abundance (i.e., CPUE) of native fish, anurans, and macroinvertebrates. In particular we expected CPUE of native organisms and the total number of native species captured would be inversely correlated with CPUE of exotic fish because exotic species may adversely affect native species through predation and competition.
Our study was conducted on the 1476.9 ha Archbold Reserve, located adjacent to Archbold Biological Station at the southern end of the Lake Wales Ridge in Highlands County, Florida (Fig. 1). The major land use on the Reserve is cattle pasture (planted with bahiagrass Paspalum notatum). Approximately one-third of the Reserve is currently undergoing hydrological restoration. Historically, the Reserve was dominated by wet prairie communities on seepage slopes. Nearly all of the original wet prairie communities and depressional wetlands, including a large freshwater marsh at the head of Mary's Creek, were drained and converted to agricultural use by extensive ditching in 1969-1981 (Fig. 1). In addition to reducing hydroperiods (i.e., the number of days the wetlands are inundated during the year), ditching increased the overall connectivity among wetlands. The ditches also connect remaining seasonal wetlands to several permanent ponds created to provide water for cattle. We sampled cattle ponds in addition to modified wetlands because of their potential to serve as refuge for exotic species during the Dec.-May dry season. Wetlands on the Reserve are connected to the greater Fisheating Creek watershed via two headwater streams known as Mary's and Frances creeks.
EXOTIC FISH SURVEY
We sampled 45 wetlands on the Reserve using minnow traps between Sep. 19, 2011 and Oct. 17, 2011. Traps were set during the afternoon and checked and emptied the next morning. On the first day of trapping in all wetlands, we deployed six Gee minnow traps (23 cm diameter x 44 cm length, 2.5 cm funnel opening, 6.4 mm mesh) in haphazardly chosen locations throughout the wetlands. If African jewelfish were detected following this initial trapping effort, then we removed the traps. If African jewelfish were not detected after the first day of trapping, then we continued to trap that wetland until it accrued 5 trap-nights/1000 [m.sup.2], at which point we considered the wetland free of African jewelfish. This target allowed us to sample all wetlands within a reasonable amount of time while providing a reliable determination of cichlid occupancy. During our study, nearly all traps in occupied wetlands detected African jewelfish even at a trapping density as low as one trap-night/7289 [m.sup.2]. Additional traps were set in some wetlands after the first day of trapping to decrease the number of days needed to reach the 5 trap-nights/ 1000 [m.sup.2] goal, and traps were moved between days to ensure that every trap-night represented an independent effort. Total sampling effort during the study was 581 trap-nights.
All exotic fish captured were euthanized in the field with a 250-mg/L solution of MS-222 (tricaine methanesulfonate) and then kept on ice while transported to the lab. We also collected all trapped invertebrates and preserved them in 70% ethanol for identification in the lab. Native vertebrates were identified and released.
To standardize relative abundance, we calculated CPUE by dividing the total captures for each species by the number of trap-nights for each wetland. Despite well documented biases (see He and Lodge, 1990; Jackson and Harvey, 1997), minnow traps were chosen for this study because they are effective at sampling communities of small bodied fish in small natural wetlands (Tonn, 1985; Jackson and Harvey, 1997) and they are effective at retaining deep bodied fish (Obaza et al., 2011). Further, the weight and length frequency distributions of African jewelfish captured during this study suggest that the minnow traps exerted little or no size bias. However, the minnow traps may have been less effective for sampling brown hoplos because larger individuals may have been excluded from the traps. Similarly, minnow traps were not effective for sampling walking catfish, which grow much larger than the opening of the traps used in this study.
Because the minnow traps may have been less effective at capturing a representative sample of brown hoplos, we restricted the detailed analysis of population characteristics to African jewelfish. We measured total length (to nearest mm) and weighed (to nearest 0.01 g) all African jewelfish. We also made a ventral incision in all African jewelfish to visually determine sex and to determine if fish were gravid (large, well-developed, unfertilized yellow eggs) or carrying undeveloped (small and white) eggs (Hyslop, 1987). Finally, to gain preliminary insights into which taxa may be affected by African jewelfish predation, we analyzed the gut contents of a subset of African jewelfish by removing the digestive tract and examining the contents under a dissecting microscope. Because of the small size of these fish, we used only individuals greater than 5 g for these analyses. This eased removal of the digestive tract and allowed us to identify a reasonable proportion of the stomach contents. We recorded all identifiable material to the lowest possible taxon.
We used ArcMap 10 software (Esri, Inc.) and existing spatial data layers for the Archbold Reserve (e.g., 2011 aerial imagery, shapefiles of wetland boundaries and ditches) to determine the shortest straight line distance from the edge of each wetland to a major ditch (ditches [greater than or equal to] 40 cm deep). We also calculated the total area of each wetland polygon. Maximum depth of each wetland was measured in the field.
Exotic fish distribution and abundance.--We used stepwise binary logistic regression (backward elimination method, SPSS v. 17) to determine if wetland area, wetland depth, or distance to nearest ditch were significant predictors of the presence of either African-jewelfish or brown hoplo. These three potential predictor variables were not significantly correlated (P > 0.10). For this analysis, we excluded the largest wetland which was an extreme outlier in terms of wetland area (174,936 [m.sup.2] versus a mean size of 4476 [m.sup.2] for all other wetlands). We also divided wetland area by 1000 to aid interpretation of the resulting odds ratios.
The CPUE for African jewelfish and brown hoplo were pooled to create the "exotic fish abundance" used in some analyses because the small number of wetlands occupied by each species precluded meaningful species-specific analyses. We used Spearman rank correlations to examine relationships between exotic fish CPUE and wetland area, distance to nearest ditch, and wetland depth. We also examined correlations between the pooled CPUE of native fishes and wetland characteristics.
African jewelfish population characteristics.--For African jewelfish, we calculated average length and weight, and determined the percent of individuals that were gravid. We used the allometric equation: W = [aL.sup.b], to determine the length-weight relationship, where W = weight, L = total length, b = a growth exponent, and a is the population condition factor (Ricker, 1975). To estimate a and b, we log-transformed this equation to make the relationship linear (Log W = b Log L + Log a) and performed regression to estimate the slope (b) of the line and the y-intercept (Log a). Because we calculated a in the same units (kg/cm) as Fulton's condition factor, K, the two estimates are comparable, facilitating comparison across studies.
If the weight and length of individual African jewelfish increase proportionally during growth, weight should theoretically increase as a cube of length (Froese, 2006); therefore, we used a Student's t-test to compare our estimated growth exponent (b) to three (Zar, 1999). Major deviations from three may indicate significant ontogenetic changes in body shape or well being (Froese, 2006). Because cichlids exhibit strong phenotypic plasticity in body size and shape (Trapani, 2003), it is reasonable to expect populations in different wetland communities to respond to local conditions. Differences in b may reflect changes in body proportion in response to local conditions, while differences in a may reflect differences in well being.
Community effects.--Because transformation of CPUE data did not improve normality, we used nonparametric Spearman rank correlations to examine associations between CPUE of exotic fish and CPUE of native fish, larval anurans, and macroinvertebrates. We limited these tests to species that were captured in at least 20 wetlands. We also tested the effects of African jewelfish presence, distance to the nearest ditch [greater than or equal to] 40 cm deep (re-classified as a categorical variable, i.e., <50 m or >50 m), and the interaction between African jewelfish presence and nearest ditch on total number of species using analysis of covariance (ANCOVA; Type III SS), with wetland area as the covariate. African jewelfish was excluded from the total number of species. Only species detected on the first day of trapping were used in the analyses to avoid bias associated with trapping wetlands for varying lengths of time. Wetland area was log-transformed and significance was assessed at [alpha] = 0.05. Unless otherwise noted, values shown are means [+ or -] 1 SE.
EXOTIC FISH DISTRIBUTION AND ABUNDANCE
Exotic fish were detected in 48% of wetlands. The 10 wetlands with African jewelfish were all within either the Mary's Creek or Frances Creek systems, whereas brown hoplos were captured in 15 wetlands distributed throughout the Reserve (Fig. 1). Wetland area was the only significant variable (likelihood ratio [chi square] = 16.061, df = 1, P < 0.0001) retained in the final logistic model for presence of African jewelfish (Hosmer-Lemeshow goodness-of-fit [chi square] = 11.968, P = 0.153). Wetland area had a positive effect on probability of occurrence (Fig. 2) with the odds of occurrence of African jewelfish increasing by approximately 1.6 times for every 1000-[m.sup.2] increase in wetland area (odds ratio = 1.575). The final model for brown hoplo did not include wetland area, wetland depth, or distance to nearest ditch (using [alpha] = 0.15 as the criterion for removal), indicating that none of these variables were significant predictors of occurrence for this species.
Capture rates of African jewelfish were second only to eastern mosquitofish (Gambusia holbrooki) (Fig. 3). In wetlands in which they were detected, African jewelfish were captured much more frequently than any other species (8.30 [+ or -] 1.73 individuals/trap-night; Table 1). Although brown hoplos were more widespread than African jewelfish, capture rates were very low (0.34 [+ or -] 0.12 individuals/trap-night). Neither exotic fish CPUE nor native fish CPUE was related to wetland depth, but both were positively correlated with wetland area and negatively correlated with distance to the nearest major ditch (Table 2).
AFRICAN JEWELFISH POPULATION CHARACTERISTICS
We obtained size measurements for 864 African jewelfish captured in nine wetlands (mean length 56.0 [+ or -] 11.2 SD mm; mean mass 3.14 [+ or -] 1.80 SD g; samples from one wetland were not measured because they were poorly preserved). The length-weight relationship was significant (8 = 3.013, [t.sub.862] = 166, P < 0.001) and length explained a significant proportion of the variation in weight ([R.sup.2] = 0.97, [F.sub.1,862] = 27,870, P < 0.001). The growth exponent (b) was not significantly different from 3 ([t.sub.862] = 0.750, P = 0.453), indicating an isometric relationship between weight and length. The average population condition factor (a) for all African jewelfish was 1.58 [+ or -] 0.01 and ranged from 1.44 to 1.82 among wetlands.
The length of the smallest reproductive individual was 47 mm, so we used this as the cutoff length to classify fish as juvenile or adult. Juveniles comprised 18.9% of the sample. The adult sex ratio was 51:49 (M:F). Mean total length of males (60.3 [+ or -] 0.5 mm) was slightly greater than that of females (59.5 [+ or -] 0.3 mm). The length-weight relationship was similar between sexes based on ANCOVA [Length (covariate) [F.sub.1,698] = 7394.503, P < 0.0001; Sex [F.sub.1,698] = 0.014, P = 0.906]. Over half (57%) of females were gravid and the rest were subgravid (carried unfertilized eggs that were less developed than in gravid females). All stomachs examined (N = 46) had at least some partially digested content. More than half (56.5%) contained evidence of fish predation (e.g., scales, fins, eggs). Arthropods (15.2%) and plant material (8.7%) comprised the rest of the diet; however, 34.7% of stomachs had unidentifiable contents.
Although abundances of native organisms did not vary consistently according to the presence or absence of African jewelfish, capture rates for 16 of 25 native taxa, including all anurans, were higher in wetlands lacking African jewelfish (Table 1). The CPUE of larval southern leopard frogs (Lithobates sphenocephalus), Everglades crayfish (Procambarus alleni) and giant water bugs (Lethocerus spp). were not significantly correlated with exotic fish CPUE; however, the CPUE of eastern mosquitofish, native topminnows (Fundulus spp.) and the giant diving beetle (Cybisterfimbriolatus) were positively correlated with exotic fish CPUE (Table 3). Neither African jewelfish presence nor distance to nearest ditch affected the total number of species (Table 4). The covariate, wetland area, had a significant positive effect (Table 4), although it explained only a small proportion of the variation in number of species ([R.sup.2] = 0.216).
EXOTIC FISH DISTRIBUTION AND ABUNDANCE
Exotic fish occupied nearly half of the wetlands sampled, but we captured African jewelfish in only 22% of wetlands on the Archbold Reserve. Despite their limited distribution, African jewelfish were the second-most abundant fish species detected, and in wetlands in which they occurred they were by far the most abundant vertebrate captured. The trapping methods used in this study may have influenced this result, because minnow traps can be species-selective (Layman and Smith, 2001). Capture rates may also be biased towards smaller, active species because only organisms that encounter the traps and are small enough to enter may be captured (Obaza et al., 2010). Therefore, comparisons between the abundance of African jewelfish and larger species, such as bluegill (Lepomis macrochirus), or more sedentary species may be unreliable. However, similarly high relative abundances have been reported in other cichlid populations in the Everglades (Faunce and Lorenz, 2000; Loftus et al., 2006).
African jewelfish were restricted to wetlands within the Mary's Creek and Frances Creek systems (Fig. 1). In each case, their ability to spread south appears limited by a raised service road, however, the much wider distribution of the brown hoplo suggests that future spread of African jewelfish is possible. The strong association between the distribution of African jewelfish and the two creeks was expected because the creeks are presumed to be the initial avenue of colonization by this species. Small creeks may aid exotic fish dispersal by connecting smaller isolated wetlands, such as those on the Archbold Reserve, to larger regional watersheds. We suspect colonization of the headwaters of Fisheating Creek was facilitated by extensive flooding in the aftermath of multiple hurricanes (Charley, Frances, Ivan and Jeanne) in 2004.
As we expected, wetland area was a useful predictor of the occurrence of African jewelfish and positively correlated with the combined abundance of African jewelfish and brown hoplo. Larger wetlands tend to support more diverse plant communities and microhabitats (Moller and Rordam, 1985), which increase the probability that a species will encounter a suitable niche (Tilman, 2004). Larger wetlands may also support greater numbers and abundances of invertebrate prey than smaller wetlands (Semlitsch and Bodie, 1998) which would benefit both native and exotic fish predators. The increase in fish prey would confer an additional benefit to the piscivorous African jewelfish. Together, these effects reduce the probability of population extinction in larger wetlands.
Wetland depth was not an important predictor of the distribution of African jewelfish or brown hoplo and was not correlated with exotic fish abundance. Deep water often serves as a refuge for fish in south Florida and aids the persistence of tropical species during bouts of cold weather (Schofield et aI., 2009). However, the average depth of wetlands on the Reserve during our study was 0.75 m (only one was >2 m deep), which is probably too shallow to provide adequate refuge during dry downs and cold weather. Thus, even the larger wetlands on the Reserve likely experience periodic extinction.
In our study area, the probability of recolonization is likely dependent on wetland connectivity, with dispersal mediated by the seasonal inundation of ditches. Although distance to nearest ditch was not a significant predictor of African jewelfish or brown hoplo distribution, we found a negative correlation between fish abundance (exotic and native) and distance to the nearest major ditch. This result is consistent with other studies showing that ditches are important for fish dispersal to seasonal wetlands (Hohausava et al., 2010) and that wetland connectivity enhances the persistence and abundance of fish (Bouvier et al., 2009). In the subtropical climate of south Florida, however, periodic extreme flooding events may eventually expose most wetlands to the exotic species that occur in the region, diluting the effect of ditch-mediated connectivity. As may be the case with brown hoplo on the Reserve, species-specific tolerance to environmental conditions and interspecific interactions within wetlands may become more important predictors of long-term distribution. Longer term studies and more refined measures of hydrological connectivity are needed to understand colonization and extinction dynamics and factors governing exotic fish distribution in these types of agricultural landscapes.
AFRICAN JEWELFISH POPULATION CHARACTERISTICS
The mean condition factor for African jewelfish on the Archbold Reserve was similar to those reported for (Hemichromis) in native populations (Laleye, 2006; Ayoade and Ikulala, 2007). African jewelfish density, may continue to increase until the population reaches a saturation point, at which point condition and reproduction will decline and stabilize, consistent with the pattern exhibited by African jewelfish in long established populations in the Everglades (Lopez et al., 2012).
We found that the weight-length ratio remains the same during growth of an individual African jewelfish in this population. In contrast, the weight-length ratio in (Hemichromis) populations in Africa has been shown to decrease during growth (Laleye, 2006; Ayoade and Ikulala, 2007). This growth difference may be the result of differences in habitat. (Hemichromis) populations in Africa were sampled in streams, where thinner individuals may experience less drag in the stream environment and thereby conserve energy. In the absence of this selection pressure, individuals in more stagnant habitats, such as the Reserve wetlands, may attain relatively greater mass. This difference could also be explained by the 'naive prey' hypothesis, which suggests that a nonnative predator may be able to exploit an abundance of prey that lack the ability to detect or respond appropriately to the unfamiliar predator (Sih et al., 2010). If invasive African jewelfish populations encounter abundant, easily exploitable food sources, then they may be able to attain greater weight per unit length than (Hemichromis) in native habitats. However, the results of predator-prey interaction studies of African jewelfish in south Florida have not supported this hypothesis (Rehage et al., 2009; Porter-Whitaker et al., 2012). More research is needed to address whether prey naivete influences the success of invasive African jewelfish.
The results of the gut content analysis are consistent with the results from other studies of introduced populations in Florida (Loftus et al., 2006; Lopez et al., 2012). Levels of fish predation in our study were likely overestimated. Although fish likely make up a considerable proportion of the African jewelfish diet in Florida, in our study, intra-trap predation probably occurred during the time fish were confined in the minnow traps. Nonetheless, this result suggests that African jewelfish could have a strong predatory effect on small native fishes and macroinvertebrates. We also suspect that African jewelfish have a negative predatory effect on larval anurans. African jewelfish can attack tadpoles larger than their gape size (unlike native centrarchids) and are able to remove portions of soft tissue from the body and tail of tadpoles, sometimes causing death (J. O'Connor, unpublished data). These predation events likely go undetected in gut content analyses because soft tissue is nondescript and digested rapidly. Additional studies are needed to determine if African jewelfish are important predators of larval anurans in natural settings.
Contrary to the effects of predatory fish introductions in other freshwater communities (e.g., Hecnar and M'Closkey, 1997; Walser et al., 2000), we found little evidence for a negative effect of exotic fish on native organisms using a correlational approach. Abundances of eastern mosquitofish, topminnows, Everglades crayfish, and giant diving beetles were positively correlated with exotic fish abundance (Table 3). Many native species appear to be responding similarly to the factors that positively influenced exotic fish abundance, namely wetland size and extent of connectivity. It is possible that the detritivorous brown hoplo may not have a negative predatory effect on native species (Hardin, 2007). Furthermore, the limited number of wetlands with African jewelfish and the lack of preinvasion data on community structure limit our conclusions. Although capture rates for anurans and several other native taxa tended to be higher in wetlands lacking African jewelfish (Table 1), presence of African jewelfish did not have a significant effect on total number of species (Table 4). Mesocosm studies or other controlled experiments would help to clarify the potential predatory and competitive relationships between African jewelfish and native organisms. Assuming exotic fish continue to spread and colonize new wetlands in our study area, the results of this initial survey could be useful for comparing community structure before and after invasion.
The results of our study provide evidence that African jewelfish are well adapted to invade modified agricultural wetlands in south Florida and their presence warrants concern because of their predatory nature and their ability to reach high abundances. From a management perspective, future plans to modify hydrology in areas known to harbor African jewelfish, especially for restoration purposes, should take into account their current distribution so that steps can be taken to minimize further spread. Future research into how African jewelfish affect aquatic communities in Florida may allow for the development of effective management strategies.
Acknowledgments.--We appreciate Roberta Pickert's assistance compiling data in ArcGIS. We are also grateful for the field and laboratory assistance provided by Joshua Daskin, Justin Dee, Zachery Forsburg, Heidi Henrichs and David Moldoff. Carl Ruetz III and two anonymous reviewers provided helpful suggestions for improving the manuscript. This research was conducted in accordance with FFWCC permit LSSC-10-00042A and accepted guidelines for use of fishes (American Fisheries Society 2004) and amphibians and reptiles (American Society of Ichthyologists and Herpetologists 2004) in field research.
SUBMITTED 30 MARCH 2012
ACCEPTED 21 JANUARY 2013
AYOADE, A. A. AND A. A. O. IKULALA. 2007. Length weight relationship, condition factor, and stomach contents of Hemichromis bimaculatus, Sarotherodon melanotheron and Chromidotilapia guentheri (Perciformes: Cichlidae) in Eleiyele Lake, Southwestern Nigeria. Int. J. Trop. Biol. Conserv., 55(4):969-977.
BABBITT, K.J. AND G. W. TANNER. 2000. Use of temporary wetlands by anurans in a hydrologically modified landscape. Wetlands, 20(2):313-322.
BABER, M.J., D. L. CHILDERS, K.J. BABBITT, AND D. H. ANDERSON. 2002. Controls on fish distribution and abundance in temporary wetlands. Can. J. Fish. Aquat. Sci., 59:1441-1450.
BALSHINE-EARN, S. AND D.J. EARN. 1998. On the evolutionary pathway of parental care in mouth-brooding cichlid fish. Proc. R. Soc. Lond. B., 265:2217-2222.
BOUVIER, L. D., K. COTTENIE, AND S. E. DOKA. 2009. Aquatic connectivity and fish metacommunities in wetlands of the lower Great Lakes. Can. J. Fish. Aqua. Sci., 66:933-948.
CLOUSER, R. L. 2005. Issues at the Rural-Urban Fringe: Land Use Conflicts. EDIS Document No. FE549, Department of Food and Resource Economics, Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida, Gainesville, Florida.
CUCHEROUSSET, J. AND J. D. OLDEN. 2011. Ecological impacts of nonnative freshwater fishes. Fisheries, 36(5):215-230.
DENOEL, M., G. DZUKIC, AND M. L. KALEZIC. 2005. Effects of widespread fish introductions on paedomorphic newts in Europe. Conserv. Biol., 19(1):162-170.
FAUNCE, C. H. AND J. J. LORENZ. 2000. Reproductive biology of the introduced mayan cichlid, Cichlasoma urophthalmus, within an estuarine mangrove habitat of Southern Florida. Enviro. Biol. Fishes, 58(2):215-225.
FROESE, R. 2006. Cube law, condition factor, and weight-length relationships: history, meta-analysis, and recommendations. J. Appl. Ichthyol., 22:241-253.
GAMRADT, S. C. AND L. B. KATS. 1996. Effect of introduced crayfish and mosquitofish on California newts. Conserv. Biol., 19:1155-1162.
HARDIN, S. 2007. Managing non-native wildlife in Florida: State perspective, policy and practice, p. 43-52. In: G. Witmer, W. Pitt and K. Fagerstone (eds.). Managing Vertebrate Invasive Species: Proceedings of an International Symposium, USDA/APHIS/WS, National Wildlife Research Center, Fort Collins, Colorado.
HE, X. AND D. M. LODGE. 1990. Using minnow traps to estimate fish population size: the importance of spatial distribution and relative species abundance. Hydrobiologia, 190:9-14.
HECNAR, S.J. AND R. T. M'CLOKEY. 1997. The effects of predatory fish on amphibian species richness and distribution. Biol. Conserv., 79:123-131.
HOHAUSOVA, E., R. J. LAVOV, AND M. S. ALLEN. 2010. Fish dispersal in a seasonal wetland: influence of anthropogenic structures. Mar. Freshwater Res., 61:682-694.
HYSLOP, E.J. 1987. Aspects of the biology of Hemichromis bimaculatus. J. Fish Biol., 31 (6):745-751.
JACKSON, D. A. AND H. H. HARVEY. 1997. Qualitative and quantitative sampling of lake fish communities. Can. J. Fish. Aquat. Sci., 54:2807-2813.
KITCHELL, J. F., D. E. SCHINDLER, R. OGUTU-OHWAYO, AND P. N. REINTHAL. 1997. The Nile Perch in Lake Victoria: Interactions between predation and fisheries. Ecol. App., 7:653-664.
KNUTSON, M. G., W. B. RICHARDSON, D. M. REINEKE, B. R. GRAY, J. R. PARMELEE, AND S. E. WEICK. 2004. Agricultural ponds support amphibian populations. Ecol. App., 14(3):669-684.
LALEYE, P. A. 2006. Length-weight and length-length relationships of fishes from the Oueme River in Benin (West Africa). J. App. Ichthyology, 22(4):330-333.
LANGSTON, J. N., P.J. SCHOFIELD, J. E. HILL, AND W. F. LOFTUS. 2010. Salinity tolerance of the African jewelfish Hemichromis letourneuxi, a non-native cichlid in south Florida, USA. Copeia, 2010:475-480.
LAWLER, S. P., O. DRITZ, T. STRANGE, AND M. HOLYOAK. 2001. Effects of introduced mosquitofish and bullfrogs on the threatened California red-legged frog. Conserv. Biol., 13(3):613-622.
LAYMAN, C. A. AND D. E. SMITH. 2001. Sampling bias of minnow traps in shallow aquatic habitats on the eastern shore of Virginia. Wetlands, 21(1):145-154.
LESLIE, A.J., T. L. CRISMAN, J. P. PRENGER, AND K. C. EWEL. 1997. Benthic macroinvertebrates of small Florida pondcypress swamps and the influence of dry periods. Wetlands, 17(4):447-455.
LOFTUS, W. F., J. C. TREXLER, K. DUNKER, S. E. LISTON, AND J. S. REHAGE. 2006. Introduced fishes in short-hydroperiod wetlands: Evaluation of sampling, status, and potential effects. Final Report from USGS to Everglades National Park for agreement #CESI IA F5284-04-0039.
LOPEZ, D. P., A. A. JUNGMAN, AND J. S. REHAGE. 2012. Normative African jewelfish are more fit but not bolder at the invasion front: a trait comparison across an Everglades range expansion. Biol. Invasions, 14:2159-2174.
MEYER, A. 1990. Ecological and evolutionary consequences of the trophic polymorphism in Cichlasoma citrinellum (Pisces: Cichlidae). Biol. J. Linnean Soc., 39:279-299.
MOLLER, T. R. AND C. P. RORDAM. 1985. Species numbers of vascular plants in relation to area, isolation, and age of ponds in Denmark. Oikos, 45(1):8-16.
MORGAN, L. A. AND W. A. BUTTEMER. 1996. Predation by the nonnative fish Gambusia holbrooki on small Litoria aurea and L. dentate tadpoles. Aust. Zool., 30(2):143-149.
OBAZA, A., D. L. DEANGELIS, ANDJ. C. TREXLER. 2011. Using data from an encounter sampler to model fish dispersal. J. Fish Biol., 78:495-513.
PORTER-WHITAKER, A., J. S. REHAGE, S. E. LISTON, AND W. F. LOFTUS. 2012. Multiple predator effects and native prey responses to two non-native everglades cichlids. Ecol. Freshwat. Fish, 21 (3):375-385.
REHAGE, J. S., K. L. DUNLOP, AND W. F. LOFTUS. 2009. Antipredator responses by native mosquitofish to non-native cichlids: An examination of the role of prey naivete. Ethol., 115(11):1046-1056.
RICKER, W. E. 1975. Computation and interpretation of biological statistics offish populations. Bull. Fish. Res. Board Can. No. 191. 382 p.
SCHOFIELD, P. J., W. F. LOFTUS, R. M. KOBZA, M. I. COOK, AND D. H. SLONE. 2009. Tolerance of nonindigenous cichlid fishes (Cichlasoma urophthalmus, Hemichromis letourneuxi) to low temperature: laboratory and field experiments in south Florida. Biol. Invasions, 12:2441-2457.
SEMLITSCH, R. D. AND J. R. BODIE. 1998. Are small, isolated wetlands expendable? Conserv. Biol., 12(5):1129-1133.
SIH, A., D. I. BOLNICK, B. LUTTBEG, J. L. ORROCK, S. D. PEACOR, L. M. PINTOR, E. PREISSER, J. S. REHAGE, AND J. R. VONESH. 2010. Predator-prey naivete, antipredator behavior, and the ecology of predator invasions. Oikos, 119(4):610-621.
TILMAN, D. 2004. Niche tradeoffs, neutrality, and community structure: A stochastic theory of resource competition, invasion, and community assembly. Proc. Natl. Acad. Sci. USA., 101:10854-10861.
TONN, W. M. 1985. Density compensation in Umbra-Perca fish assemblages of northern Wisconsin Lakes. Ecol., 66(2):415-429.
TRAPANI, J. 2003. Morphological variablility in the Cuatro Cienegas cichlid, Cichlasoma minckleyi. J. Fish Biol., 62:276-298.
TREXLER, J. C., W. F. LOFTUS, F. JORDAN, J. J. LORENZ, J. H. CHICK, AND R. M. KOBZA. 2000. Empirical assessment of fish introductions in a subtropical wetland: an evaluation of contrasting views. Biol Invasions, 2:265-277.
VANDER ZANDEN, M.J. AND J. D. OLDEN. 2008. A management framework for preventing the secondary spread of aquatic invasive species. Can. J. Fish. Aqua. Sci., 65:1512-1522.
WALSER, C. A., B. FALTERMAN, AND H. L. BART JR. 2000. Impact of introduced rough shiner (Notropis Bailey,) on the native fish community in the Chattahoochee river system. Am. Midl. Nat., 144(2):393-405.
WILSON, G. G. 2006. Impact of invasive exotic fishes on wetland ecosystems in the Murray- Darling Basin. p. 45-60. In: B. Phillips (ed.). (2006) Native fish and wetlands of the Murray-Darling Basin: action plan, knowledge gaps and supporting papers. Proceedings of a workshop held in Canberra ACT. 7-8 Jun. 2005. Murray-Darling Basin Commission, Canberra.
ZAR, J. H. 1999. Biostatistical analysis. 4th ed. Prentice Hall, Inc, Upper Saddle River, New Jersey. 929 p.
JASON H. O'CONNOR (1) AND BETSIE B. ROTHERMEL (2)
Archbold Biological Station, 123 Main Drive, Venus, Florida 33960
(1) Present address: Department of Natural Resources and the Environment, University of Connecticut, Storrs, CT 06269
(2) Corresponding author: Telephone: (863) 465-2571; Fax: (863) 699-1927; Email: brothermel@ archbold-station.org
TABLE 1.--Mean ([+ or -] SE) number of captures per trap-night (across all sampling days) of fish, amphibians, reptiles, and macroinvertebrates in wetlands with and without African jewelfish. Names of exotic species are in bold. N = number of wetlands. For anurans, A = adult (postmetamorphic) and L = larval life stage Jewelfish absent Family Species Mean SE N Callichthyidae Hoplosternum littorale 0.34 0.12 11 Clariidae Clarias batrachus 0.17 -- 1 Fundulidae Fundulus spp. 0.58 0.19 13 Cyprinodontidae Jordanella floridae 1.29 0.88 6 Poeciliidae Gambusia holbrooki 1.79 0.34 25 Centrarchidae Lepomis macrochirus -- -- 0 Lepomis gulosus 0.10 0.07 2 Ranidae Lithobates sphenocephalus (A) 0.11 0.05 6 Lithobates sphenocephalus (L) 0.86 0.42 20 Lithobates grylio (A) 0.19 0.06 7 Lithobates grylio (L) 0.38 0.06 5 Lithobates catesbeianus (A) 0.25 0.12 5 Lithobates catesbeianus (L) 2.59 1.83 3 Hylidae Hyla femoralis (L) 0.13 0.04 3 Hyla cinerea (L) 0.04 -- 1 Hyla gratiosa (L) 0.04 -- 1 Sirenidae Siren intermedia 0.02 -- 1 Amphiumidae Amphiuma means 0.05 0.03 3 Colubridae Seminatrix pygaea 0.24 -- 1 Nerodia fasciata 0.03 -- 1 Thamnophis sauritus 0.10 0.07 2 Cambaridae Procambarus alleni 2.25 0.39 27 Dytiscidae Cybister fimbriolatus 0.82 0.15 33 Hydrophilidae Hydrophilus triangularis 0.09 0.03 2 Belostomatidae Lethocerus uhleri 0.21 0.03 18 Lethocerus griseus 0.24 0.05 11 Belastoma luterium 0.18 0.11 4 Nepidae Ranatra australis 0.37 0.20 5 Aeshnidae Gomphaeschna spp. 0.11 0.03 8 Coryphaeschna spp. 0.07 -- 1 Jewelfish present Family Species Mean SE N Callichthyidae Hoplosternum littorale 0.15 0.02 4 Clariidae Clarias batrachus -- -- 0 Fundulidae Fundulus spp. 1.08 0.54 7 Cyprinodontidae Jordanella floridae 0.15 0.03 3 Poeciliidae Gambusia holbrooki 3.06 1.08 9 Centrarchidae Lepomis macrochirus 0.83 0.50 2 Lepomis gulosus 0.25 0.05 4 Ranidae Lithobates sphenocephalus (A) 0.08 -- 1 Lithobates sphenocephalus (L) 0.50 0.33 2 Lithobates grylio (A) -- -- 0 Lithobates grylio (L) 0.26 0.09 5 Lithobates catesbeianus (A) -- -- 0 Lithobates catesbeianus (L) -- -- 0 Hylidae Hyla femoralis (L) -- -- 0 Hyla cinerea (L) -- -- 0 Hyla gratiosa (L) -- -- 0 Sirenidae Siren intermedia 0.08 -- 1 Amphiumidae Amphiuma means -- -- 0 Colubridae Seminatrix pygaea 0.25 0.08 2 Nerodia fasciata 0.06 0.02 2 Thamnophis sauritus -- 0 Cambaridae Procambarus alleni 1.78 0.59 10 Dytiscidae Cybister fimbriolatus 1.70 0.49 10 Hydrophilidae Hydrophilus triangularis 0.17 -- 1 Belostomatidae Lethocerus uhleri 0.19 0.07 3 Lethocerus griseus 0.17 0.00 3 Belastoma luterium -- -- 0 Nepidae Ranatra australis 0.10 0.06 2 Aeshnidae Gomphaeschna spp. -- -- 0 Coryphaeschna spp. -- -- 0 TABLE 2.--Spearman rank coefficients (one-tailed P-values) for correlations between CPUE of exotic and native fish and characteristics of wetlands on the Archbold Reserve. The largest wetland was excluded from correlations involving wetland area because it was an extreme outlier N Exotic fish CPUE Native fish CPUE Wetland area 44 0.484 (<0.001) 0.339 (0.012) Wetland depth 45 0.057 (0.356) 0.202 (0.092) Distance to nearest ditch 45 -0.263 (0.040) -0.235 (0.060) TABLE 3.--Spearman rank coefficients (one-tailed P-values) for correlations between CPUE of exotic fish and CPUE of native vertebrates and macroinvertebrates Species Spearman rank coefficient (P) Eastern mosquitofish (Gambusia holbrooki) 0.315 (0.018) Topminnows (Fundulus spp.) 0.380 (0.005) Southern leopard frog larvae (Lithobates sphenocephalus) -0.116 (0.224) Everglades crayfish (Procambarus alleni) 0.224 (0.070) Giant diving beetle (Cybister fimbriolatus) 0.298 (0.023) Giant waterbugs (Lethocerus spp.) -0.029 (0.425) TABLE 4.--Results of ANCOVA testing the effects of African jewelfish presence and distance to nearest ditch on the total number of species. All wetlands were included in the analysis, although results were qualitatively similar if the largest (outlier) wetland was omitted Source of variation df MS F P Log(wetland area) 1 72.032 12.500 0.001 Jewelfish presence 1 11.442 1.986 0.167 Nearest ditch 1 0.628 0.109 0.743 Jewelfish x nearest ditch 1 0.735 0.128 0.723 Error 40 5.762
|Printer friendly Cite/link Email Feedback|
|Author:||O'Connor, Jason H.; Rothermel, Betsie B.|
|Publication:||The American Midland Naturalist|
|Date:||Jul 1, 2013|
|Previous Article:||Habitat associations of stream fishes in protected tallgrass prairie streams.|
|Next Article:||Avian predation and the evolution of life histories in the garter snake Thamnophis elegans.|