Short-Time-Base Studies of Turnover in Breeding Bird Populations on the California Channel Islands

Abstract

Estimates of faunal turnover rates based on surveys many years apart are likely to be underestimates because of the in-and-out effect (a population immigrating and then going extinct or vice versa, between surveys). Surveys of breeding land and freshwater bird species on the 8 California Channel Islands [USA] were carried out annually since 1973, for comparison with earlier surveys. Sources of information on Channel Islands birds, survey strategies and criteria for breeding presence or absence are summarized. As depicted graphically by a turnover pyramid, populations of different species on the same island fall into categories that turn over at very different rates: zero-turnover populations, consisting of either regular breeders or absentees (members of the mainland species pool that never breed on the island for any of 5 reasons); terminal immigrants, that colonized once in the survey interval and persisted; terminal extinctions, species that disappeared in the survey interval and did not recolonize; in-and-out populations, that immigrated and became extinct repeatedly; and flukes, populations that once colonized unexpectedly and soon disappeared. Numerous examples of Channel Islands populations in each category are given. Man has acted to increase and decrease turnover in several direct and indirect ways. The in-and-out effect causes turnover values based on comparison of surveys several decades apart to underestimate the true values by about an order of magnitude. Conservative estimates of turnover rates for the Channel Islands, based on 1-yr survey intervals, range from 0.9%/yr for the larger islands to 5.6%/yr for the smaller islands. Relative and absolute turnover rates decrease with island area. Despite this turnover, species number on most islands has remained approximately constant because the immigration rate approximately equals the extinction rate. Extinction rates decrease steeply with population size. Relatively few immigrant populations succeed in achieving large numbers. On the basis of these observations, a model of the stages in colonization is proposed. Criticisms directed by Lynch and Johnson at previous turnover studies are considered in an appendix.