MORTALITY ESTIMATES FROM REPEATED VISUAL CENSUSES OF A PARROTFISH (SPARISOMA-VIRIDE) POPULATION - DEMOGRAPHIC-IMPLICATIONS

Repeated visual censuses of different categories (juveniles, females, territorial and group males) of the stoplight parrotfish (Sparisoma vi ride, a protogynous hermaphrodite) over a 3-year period indicated a re latively stable size and structure of the adult population. This allow ed estimates of size-specific mortality, sex change, and territory acq uisition probabilities from previously reported growth rates. Comparis on of the predicted number of survivors, sex changers, and territory t ake overs with field observations indicates that our estimates are qui te reliable. However, rather large differences in mortality are obtain ed for the largest three size classes (>25 cm), which may be due to re duced accuracy of length estimates of large fish. A pooled mortality f or these classes is therefore suggested as a more realistic estimate. The life-history implications of our findings are investigated by comp aring the predicted survival and future reproductive success of fish t hat change sex at different sizes. Ten percent of the adults reach an age of ca. 17 years, once a length of 20 cm is attained, as predicted from the pooled estimates. Calculations based on these estimates indic ate that the predicted reproductive output of a 15-year-old fish (2500 matings) is the same for early (i.e. at 20 cm) and late (>30 cm) sex changers and for an average fish subject to the estimated daily sex ch ange and territory acquisition probabilities. These findings suggest t hat S. viride individuals are able to flexibly adjust the timing of se x change in an adaptive way to unpredictable local conditions. However , independent mortality estimates are needed to corroborate our presen t findings.