Quantifying the robustness of first arrival dates as a measure of avian migratory phenology

As the climate changes, many long-term studies have shown that the timing of bird migration is shifting, increasing the need for reliable measures of migratory phenology. Ideally, daily counts of birds at a site are used to calculate the mean arrival date (MAD) but, as this approach is not always possible and is very labour-intensive, simpler metrics such as first arrival date (FAD) have commonly been used. Here, we examine the relationship between FAD and MAD in 28 summer migrant bird species over a 42-year period (1970–2011) at Portland Bird Observatory, UK. Although significant correlations between FAD and MAD were detected, relationships were weak, particularly in long-distance migrants. We suggest that FAD, although a simple and straightforward measure, is not particularly robust as a proxy for overall migratory phenology at a population level.