|projects > effects of hydrological restoration on manatees: integrating data and models for the ten thousand islands and everglades > abstract
Linking a Manatee Individual-Based Model with the TIME Hydrology Model to Assess Restoration Effects in the Everglades and Ten Thousand Islands
Brad Stith1, Eric D. Swain2, Jim Reid1 and Catherine Langtimm1
A significant population of Floridas manatees (Trichechus manatus latirostris) occurs in southwest Florida, including extensive protected areas within the Ten Thousand Islands (TTI) and Everglades National Park (ENP). Restoration projects are planned within ENP and a smaller restoration (Picayune Strand) is an Accerler-8 project that is being fast-tracked. These restoration efforts are expected to significantly change freshwater flow throughout the region, and may have effects on manatees using freshwater, estuarine, and near-offshore areas downstream from these projects. To help evaluate these potential impacts we are integrating an individual-based manatee model with the TIME hydrology model (see Swain et al.). The TIME model will generate key hydrologic parameters that manatees respond to, including salinity, water level, and water temperature. To improve the value of the TIME model for evaluating manatees in this region, the hydrology model will be applied in the TTI region south of US41 to include the Acceler-8 Picayune Strand restoration area and the Big Cypress region, both important to manatees.
The individual-based model is parameterized with telemetry data from manatees tagged in TTI between June 2000 and June 2005, miscellaneous tagged manatees that used the area (e.g. rescued animals), as well as manatees tagged in ENP starting in March 2005. A network data structure is used to model manatee movement between nodes representing destination sites for feeding, drinking, and thermal sheltering, all connected by arcs representing travel corridors. The movement of manatees between different zones is simulated using a Markov Chain approach to transition manatees between different landscape zones (offshore, bays, river systems). These transition probabilities are generated using multi-state modeling, with the raw telemetry data formatted to fit into a mark-recapture framework. Virtual manatees are allocated home ranges derived from the telemetry data, comprising different portions of the total network that includes freshwater sites, thermal refugia, and feeding areas. Salinities, water temperature, and water depth along the landscape network will be derived from the TIME model to reflect changes due to restoration. Simulated manatees can shift their home range to different parts of the network if freshwater, thermal refugia, or seagrass become unavailable within their home range. These shifts are modeled using a reinforcement model which controls how manatees respond to changes in the availability of critical resources. Initially, pre-restoration hydrologic conditions will be simulated to develop manatee distribution estimates that can be validated with independent aerial survey data. The final set of simulations will run different hydrologic restoration scenarios and will compare the resulting manatee distributions for the dry, wet, and cold seasons.
(This abstract is from the 2006 Greater Everglades Ecosystem Restoration Conference.)
U.S. Department of the Interior, U.S. Geological Survey
This page is: http://sofia.usgs.gov/projects/integrating_manatee/manlnkabgeer06.html
Comments and suggestions? Contact: Heather Henkel - Webmaster
Last updated: 15 January, 2013 @ 12:43 PM(TJE)