years and regions, so age ratios could still provide a valuable index of productivity.
However, if bias varied with time or place it could invalidate the evaluation of
trends or regional differences. A study of ratios of young (juvenile and first-
winter) to older Blue Tits Parus caeruleuswhen first captured in mist nets by
volunteer bird ringers in two winters, 1970–71 and 1971–72, showed that
young to old ratios in July–September were implausibly high, presumably
because of some combination of the effects listed above. However, from October
to the following spring there was a consistent difference between the two winters
such that sampling at any time within them would have identified that the
young:old ratio was higher in 1970–71 than 1971–72 (Krebs and Perrins 1978).
Similarly, the relative number of juvenile Bullfinches Pyrrhula pyrrhulaper adult
in mist net catches in October appeared to be a reliable index of productivity
(Newton 1999a). The ratio of juvenile to adult Kirtland’s Warblers Dendroica
kirtlandiicaptured in mist nets in late summer was found to correlate well with
an independent measure of annual productivity (Bart et al. 1999). However,
juveniles were found to be about 1.7 times more susceptible to capture than
adults. Age ratios were found to vary considerably among study areas, so a large,
representative sample of areas is needed.
An analysis of the proportion of juveniles relative to adults in a 9-year series of
season-long catches from a standardized mist-netting program carried out by
volunteer ringers showed considerable homogeneity across widely distributed
sites and habitats in Britain in the magnitude of year to year changes (Peach,
Buckland, and Baillie 1996). This indicated that it may be feasible to use age ratios
in samples of birds trapped by ringers as an annual, regional index of productivity.
3.6.4Intensive studies of breeding
Detailed studies of individually marked birds can produce reliable estimates
of annual productivity because successive breeding attempts of the same female
can be identified. Such studies are most practicable where potential nest sites
are restricted and can all be checked. There are several examples of this from
nest box studies of hole-nesting species in which it is known that there is a negli-
gible chance of breeding attempts occurring undetected in natural sites and where
ringed individuals are identified during every breeding attempt by capturing
them in the nest box. The same applies for many raptors and seabirds with
restricted or highly traditional nesting sites. For species that make nests concealed
in vegetation, only the most intensive studies can produce full details of every
breeding attempt. Because of the early stage at which they found most nests, Roth
and co-workers considered that they had located almost all breeding attempts
by Wood Thrushes Hylocichla mustelinain their study area in a 22-year period
Measuring annual productivity| 75