(A306) <<< Previous | Back to Abstracts Page | Next >>> (A308)
Ecological connectivity among coral reefs: Quantitative predictions of laval fish dispersal
Pierluigi Pantalone, Barry Ruddick, Bruce Hatcher, Jinyu Sheng
Dalhousie University
pierluigi_pantalone@hotmail.com(Abstract received 04/30/2005 for session A)
ABSTRACT
In this study we apply (1st order) Markov Chains theory to investigate the source-sink relationship between discrete areas of the MesoAmerican Barrier Reef System (MBRS; NW Caribbean Sea) in terms of bioparticles exchange. A 2-D numerical model with horizontal resolution of 1/18 degree and a time-step of 6 hours is used to perform Lagrangian tracking of an ensamble of passive (drifters) and active (larvae) particles over 60 days. The model is forced by monthly mean (near-surface) wind driven circulation with current velocities vertically integrated over the top 10 meters. Sub-grid scale dispersion effects are introduced by a random walk technique to simulate unresolved circulation features such as tidal currents or small scale eddies. Reflective boundary conditions are applied in the model to prevent beaching of the particles; the accuracy of the particle tracking is increased by bivariate linear interpolation (time-space) and 4th order Runge Kutta scheme.The biological parameters introduced in the model define the sensory and swimming abilities of an average (late-stage) coral reef fish larva with a planktonic larval duration (PLD) of 30 days. Ecological connectivity (units: # larvae/(unit area*unit time)) is estimated both for the passive and active case scenarios through the computation of time-dependent transition probability matrices (P_(k) and PB_(k)) derived from the particles trajectories in the MBRS. Time-independent transition probability matrices (P_(k)) are also computed for the passive runs using a constrained nonlinear optimization that provides the best fit to P_(k). Analysis of the results for specific reef areas in the MBRS are illustrated both for the passive and active case scenarios in order to highlight the importance of biological behaviors on larval fish dispersal patterns. The predictive capability of (P^(k)) in quantifying the above properties for each reef area is also shown through comparison with the estimates of retention and exchange based on the trajectories of the tracked particles each time-step (P_(k)).
(A306) <<< Previous | Back to Abstracts Page | Next >>> (A308)
2005 LAPCOD Meeting, Lerici, Italy, June 13-17, 2005