Abstract There is growing evidence that morphological change is a widespread response to the warming climate. The empirical basis for understanding this process has, to date, largely been correlative analyses of morphological time series. However, it is not clear what constitutes sufficient temporal sampling for assessing long-term morphological change. We analyzed two long-term high-quality morphological datasets, one including 33 species of birds collected over 37 years and one including 11 species of mammals collected over 40 years. We find that the number of years of data needed to accurately characterize trends in functional traits varies by trait and taxonomic group. For birds, accurately estimating the trend of morphological changes 95% of the time requires data from a minimum of 14 years for bill length, 19 years for tarsus length, and 20 years for wing length. For mammals, 31 years of hind-foot length data and 36 years of weight data are needed to achieve the same level of accuracy in trend estimates. Using these minimum sampling thresholds, we identify where in the world there are sufficient museum holdings to reconstruct long-term trends in morphology. Museum specimens have a critical benefit that is not available from other long-term data sources: collected, vouchered, specimens can be analyzed in new and standardized ways designed to understand morphological responses to climate change. We find that there are many opportunities to reconstruct morphological change in birds and, to a smaller degree, in mammals using museum specimens though sufficient sampling is not available for the vast majority of the globe. Most of the sites at which there is sufficient sampling are in the Northern Hemisphere and are concentrated in the United States and Europe. Expanding long-term animal capture efforts will be critical to enabling a more holistic understanding of biotic responses to global change in the future.
Weeks et al. (Mon,) studied this question.