Rob Bennetts
The specific objectives for the project are as follows: 1) estimate apple snail abundance in wet prairie habitats (relatively short hydroperiod) relative to slough habitats (relatively longer hydroperiod) with potential linkages to snail kite foraging habitat selection; 2) estimate snail abundance in sites with foraging snail kites, and incorporate data from this study in to a larger coordinated effort to link snail kite habitat use and snail availability; and 3) document baseline seasonal apple snail egg production (not yet available in wetlands south of Lake Okeechobee) and assess the potential impacts of drying events on annual egg production.
U.S. Department of Agriculture - Natural Resources Conservation Service (NRCS) Department of the Interior - U.S. Geological Survey Department of Commerce - National Oceanic and Atmospheric Administration (NOAA) Environmental Protection Agency (EPA) Smithsonian Institution - National Museum of Natural History (NMNH)
11000 University Parkway
Kitchens, W. M., Dreitz, V. J.
Anderson, D. R.
Croop, J. D., Bennetts, R. E., Valentine-Darby, P.L., Kitchens, W. M.
Percival, H. F.
Bennetts, R. E.
We will characterize plant community composition (per cent species coverage for the area), substrate composition, plant stem density (1-m sq. quadrat samples, n=7) and periphyton abundance in all sampling sites. We will attempt to control for variation in all aforementioned habitat characteristics during sample site selection.
Three wet prairie/slough habitats known to support foraging snail kites (locations provided from our own surveys or by the snail kite monitoring field crew; P.I., Wiley Kitchens, USGS-BRD) will also be sampled for snail abundance using throw traps. Comparisons to the randomly selected sites within a wetland unit will be made to see if kites forage in sites with relatively higher snail abundance. However, understanding how snail kites respond to apple snail abundance requires comparisons to foraging habitats spread throughout the snail kite’s range. Snail kites are highly nomadic, even in generally high-water years (Bennetts and Kitchens 2000), but no snail abundance data has been linked to kite movements. Data from this 3-year study will be examined as part of a coordinated effort via concurrent studies in the Loxahatchee N.W.R., Kissimmee Chain-of-Lakes, and potentially other areas in order to correlate snail abundance with snail kite habitat use. Such coordination enables us to increase our scope of inference across areas with a range of hydrologic regimes and varying substrate types (e.g., sand versus peat) and over a temporal and spatial scale relevant to the nomadic snail kites’ foraging range.
Apple snail egg clusters provide critical information about the timing of the annual post-reproductive adult die off (Darby and Percival 2000). A 1m x 2.5m PVC quadrat will be flipped end over end 20 times in order to count egg clusters in each transect. Two egg cluster transects will be established in all of the aforementioned throw trap study sites (thus incorporating a hydrology effect) and sampled monthly from February - September each year (n=26 transects each month). Finally, 2 egg cluster transects will be established in each of 3 sites subject to frequent drying events and known to contain apple snails (e.g., portions of NESRS and northwestern WCA3A) in order to quantify recruitment losses related to drying events and/or potential resumption of oviposition after water levels rise following a dry down.
Apple snail density data from an approximately 50 m x 50 m area tends to be distributed as a negative binomial (Darby et al. 1999). A likelihood ratio testing approach (per White and Bennetts 1996, see snail data in Darby et al. 1999) will be used to model apple snail density as a function of community type, substrate type, snail kite presence or absence (or possibly some index of foraging success or extent of kite use, if data were available) and time (+ interactions). Final model selection will be based on comparing Akaike's information criteria (AIC) values from a suite of candidate models (Burnham and Anderson 1998). We will evaluate a suite of candidate models that include each main effect separately as well as all possible combinations of interactions. Egg cluster counts will be analyzed using a generalized linear models approach as described in Darby and Percival (2000). The model for egg cluster production will include temperature, water level, community type, substrate type, and time (+ interactions) as independent variables. Final model selection will be based on an AIC approach as described above.
11000 University Parkway
11000 University Parkway
U.S. Department of the Interior, U.S. Geological Survey, Center for
Coastal Geology
Comments and suggestions? Contact: Heather
Henkel - Webmaster
Generated by mp version 2.8.18 on Tue Jan 23 03:54:31 2007