Project Flowchart

The proposed stock assessment model will be developed to represent the dynamics of the whole spiny lobster stock in the southeast US. At every time step, a certain number of recruits will grow into the model (from the pre-molt size smaller than the minimum size to the post-molt size larger than the minimum size). This group of recruits is defined as model recruitment. During that time step, these recruits will experience one or some of the following processes: dying due to natural mortality, being caught in the fishery, spawning or molting. The surviving substock will repeat these processes again and again with new recruits in the subsequent time steps. Specifically, the stock assessment model will consist of five submodels:

1) A recruitment submodel that describes the recruitment dynamics.

Caribbean spiny lobster (Panulirus argus) has a long pelagic larval stage [1][2][3][4]. This character makes it extremely hard to model the recruitment of the spiny lobster. In this project, we try to use microsatellite as genetic markers to track the recruitment, therefore perform a better estimation of the recruitment.

2) A growth submodel that simulates the growth of the whole stock.

Caribbean spiny lobster does not grow continuously. They grow stepwise as a result of molting process. Therefore we need to consider the probability of molting in the growth model. In our project, we use the mark recapture data collected by FWC from South Florida during 1967 to 2003. We compare different growth methods that used in pervious crustacean growth research. We choose the model with the lowest uncertainty to use in the stock assessment model.

3) A catch-at-size submodel that predicts the size-based landings data based on the dynamics of stock structure.

In our project, the catch-at-size submodel can be derived from commonly used equation: the Baranov catch equation [5][6].

4) An observational submodel that describes the relationship between stock abundance and catch-per-unit-efforts estimated from fisheries.

Catch-per-unit-efforts (CPUE) are commonly used as the abundance indicator. In our project, a model will be developed to predict the abundance by CPUE.

5) An observational submodel that describes the relationship between predicted and observed landing size composition.

In the observational submodel, the predicted landings and the size compositions will be linked to the corresponding observed fisheries data via Bayesian likelihood functions. The maximum likelihood functions are used to optimize the parameters in the stock assessment model.

Those five submodels cooperate with each other. They compose the major skeleton of the size structure stock assessment model. A Bayesian approach will be used to fit the stock assessment model to data. Then we will use the stock assessment model to estimate fisheries status and predict the future population trends.

Reference:

[1] Lewis, J.B. (1951) The phyllosoma larvae of the spiny lobster, Panulirus argus. Bull. Mar. Sci. 1:89–103.

[2] Menzies, R.A. (1980) Biochemical population genetics and the spiny lobster larval recruitment problem: an update. Proc. Gulf Caribb. Fish. Res. Inst. 33:230–243.

[3] Yeung, C. (1996) Transport and Retention of Lobster Phyllosoma Larvae in the Florida Keys. PhD dissertation, Coral Gables, FL, USA: University of Miami, pp. 217.

[4] Matsuda, H., Goldstein, J.S., Takenouchi, T. and Butler, M.J. IV (2008) A description of the complete development of larval Caribbean spiny lobster Panulirus argus (LATREILLE, 1804) in culture. J. Crust. Biol. 28:306–327.

[5] Ricker, W. E. (1987). Computation and interpretation of biological statistics of fish populations.

[6] Hilborn, R., & Walters, C. J. (Eds.). (1992). Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty/Book and Disk. Springer.