Shawn E. Liston¹, Kristine J. Dunker^1,2, Jennifer S. Rehage^1,2 and William F. Loftus^2
1Audubon of Florida, Tavernier Science Center, Tavernier, FL, USA
²U.S. Geological Survey, Florida Integrated Science Center, Everglades National Park Field Station, Homestead, FL, USA

The abundance of non-indigenous fishes in the Rocky Glades region of Everglades National Park (ENP) has increased markedly in recent years. The Rocky Glades is a short-hydroperiod karst wetland with numerous solution holes that vary from shallow, isolated depressions to deep, interconnected complexes. Solution holes provide refuges for fishes during seasonal drying events, and may serve as a source of marsh colonists upon re-flooding of the marsh surface. The non-indigenous African jewelfish (Hemichromis letourneuxi) was first collected in Everglades solution holes in 2000; subsequently, its relative abundance has increased rapidly. This increase was especially apparent in intermediate (41 to 80-cm max. depth) and deep (> 80-cm max. depth) solution holes. Data collected from fish-community monitoring efforts in 2003-04 indicated native fishes (e.g., Gambusia holbrooki, Fundulus confluentus, Jordanella floridae) were abundant in shallow solution holes (≤ 40-cm max. depth), but were uncommon and often absent in deep solution holes where H. letourneuxi was common (comprising ≈ 30% of total catch).

In this study, we conducted a predator inclusion/exclusion cage experiment in medium to deep solution holes in ENP to discern whether the absence of small native fishes in deep solution holes is a result of poor physicochemical conditions (e.g., low dissolved oxygen, high ammonia, etc.) or predator-induced mortality by H. letourneuxi. We installed two cages in each of 15 solution holes and conducted four predation trials through the course of the dry season as abiotic conditions deteriorated. For each trial, we placed 10 G. holbrooki (eastern mosquitofish) in one cage, and 10 G. holbrooki and 1 H. letourneuxi in the second cage; trials were run for 7 days. Physicochemical data were collected at the beginning and end of each trial. We found significantly higher mortality of G. holbrooki in the presence of H. letourneuxi than when the predator was absent. There was no consistent impact of deteriorating physicochemical conditions on either predator-induced mortality of G. holbrooki, or on H. letourneuxi survival during the trials.

We also performed a survival analysis of G. holbrooki in solution holes through the course of the dry season. Those data were collected from a third cage installed in each solution hole, stocked with 10 G. holbrooki (no predator present), and censused weekly until either the hole re-flooded to the marsh surface or there was 100% mortality. Our data suggest that predator-induced mortality by non-indigenous fishes may be deleterious to native fishes in refuges in southern Florida, and may be more limiting to native fish survival than concomitant abiotic stress.

Contact Information: Shawn E. Liston, Audubon of Florida, Tavernier Science Center, 115 Indian Mound Trail, Tavernier, FL 33070 USA; Phone: 305-242-7842; Fax: 305-242-7836; Email: shawn_liston@usgs.gov

(This abstract is from the 2006 Greater Everglades Ecosystem Restoration Conference.)

Back to Project Homepage