Resource availability and goshawk offspring sex ratio variation:: a large-scale ecological phenomenon

1. Local population studies have shown that sex allocation among many birds and mammals seems to be partly non-random and in connection to surrounding factors, such as environmental or parental quality. In this scenario, if environmental quality varies in space and time, it is feasible that environmental quality also comes to influence offspring sex ratio on larger geographical scales. 2. Investigating this idea - using nation-wide data sets on size-dimorphic Finnish northern goshawks Accipiter gentilis from 1989 to 1998 - we found that offspring sex ratio is related to spatial and temporal variation in availability of their main prey, woodland grouse species. 3. In a majority of locations (n = 73), broods are large and male-biased when grouse density is high, while small and female-biased broods are found in areas and times with low grouse density. When adjusted for grouse density, the correlation between brood size and sex ratio disappears, but the correlation between grouse density and sex ratio remains. 4. We also found evidence indicating a mechanism balancing local offspring sex ratios towards the previously rarer sex with a time lag of 1 year. That is, local sex ratios in subsequent years are negatively correlated. 5. Our study highlights the importance of inclusion of both time and space in sex ratio studies. Ignorance of these dimensions could explain a great deal of the inconsistency in many earlier local sex ratio studies. 6. Irrespective of whether the observed pattern is the result of facultative manipulation, sex biased mortality, or both, it is suggested that the male-biased sex ratio during periods of high grouse availability maximizes goshawk reproductive success.