A model for amphipod (Talitrus saltator) population dynamics

Abstract

This study provides a population dynamics model for one of the most common species (Talitrus saltator) in sandy beaches. Sensitivity analysis showed that in the model the abundance of T. saltator is deeply affected by variations in the minimum recruitment day length, the minimum temperature at recruitment, and the period between recruitments. The mortality rate, as expected, also had a profound effect on model performance. The abundance (IST, ind m-1) of T. saltator was most sensitive to the initial number of individuals in age class 5 (ca. 4 months old) and least sensitive to the initial number in age class 9 (ca. 8 months old). Data used for the calibration of the model were obtained in the Lavos beach in central Portugal. The model is based on differential equations and it was constructed using Stella simulation software. Five different model versions were tested corresponding to different modes of recruitment that could fit the data. It was found that the best model was obtained when considering reproduction dependent on temperature and photoperiod and occurring on a semi-lunar basis. Two out of the five different model versions tested gave a good statistical performance. Only one of these two model versions makes an effective use of causality mechanisms. This was also the model with the best intercept and slope of the simulated vs. observed regression equation. The other statistically satisfactory model used continuous reproduction between two dates. These dates do not match a photoperiod threshold. The final model version (e) may serve as a useful tool, when used alone or when coupled with other models. One of its uses could be in assessing the effects of human actions upon a sandy beach ecosystem.