Estimating the Per Capita Rate of Population Change: How Well Do Life-History Surrogates Perform?

Abstract

Abstract A method of estimating per capita rate of change for experimental populations in the absence of adult survival and fecundity schedules (r′) was proposed previously (J. Anim. Ecol. 53: 573–580; 1984). The method has been used repeatedly, although there has been no attempt to verify the relationship between r′ and the per capita growth rate in populations. This method was tested with laboratory populations of the mosquito Aedes albopictus (Skuse) (Diptera: Culicidae) growing at three larval densities. Survival and fecundity schedules enabled calculation of per capita growth rate, r, for each population, whereas measuring emergent females enabled the calculation r′. The capacity of r′ to predict r was significant, although a substantial amount of variation in r remained unaccounted for, and r′ overestimated r. The degree of overestimation of r by r′ was consistent across the three densities, and the regressions off and r′ versus density were similar. Overestimation of per capita rate of change by r′ is attributed largely to extensive adult longevity in laboratory conditions for this species, coupled with a lengthy reproductive period, in combination with prereproductive adult mortality, both of which are assumed to be negligible with the r′ method. Separate indicators of larval success were also tested for their predictions of r; only first day emergence and average development time were significant predictors although neither performed as well as r′. The r′ index accounted for the largest fraction of variation in r. Implications of overestimating r by r′ to estimating other population parameters are discussed.