Environmental ‘events' such as catastrophic winter mortality and reproduction failure influence age structure, sex ratios and survival rates of Sand martin populations in very specific ways. In order to recognize these influences we need to know normal or average values for these quantities, as well as normal disturbances likely to create sampling errors. Variance normal to Sand martins includes (1) that there is pronounced age segregation within colonies, (2) that survival rates of juveniles decrease with progressing fledging date, and (3) that adult survival rates vary by at least a factor of three in the course of 10–15 years.
In addition to heavy overall mortality, catastrophic winter conditions create biased sex ratios among older birds: females of age ≥ 3y gradually disappear. Stress on older birds due to their higher breeding contributions (double broods) is the probable cause. This differential mortality—a characteristic of the population as a whole, not of one colony-is compensated for by distinct, precisely timed (counter-)biased sex ratios among first-time breeders. So far there has been no reason to assume that these ratios are created by exclusion of members of either sex from breeding.
The age structure of a population becomes compressed during periods (prolonged over some years) of adverse weather, and expands during favourable periods. This brings a shift from general vulnerability to general stamina in its train. All changes are remarkably continuous, in a sense never really ‘catastrophic’, because of the moderating influences of age structure and sex ratios.
In south Sweden Sand martins are heavily preyed upon by badgers. Most of the predation occurs during periods of dry weather, when earthworms are difficult to procure. However, the observed oscillations (Persson, 1987a) cannot be the outcome of a predator-prey relation: several prerequisites are not fulfilled. It is assumed, instead, that the population involved is self- regulating, although this self-regulation may very well be accentuated by predation.