UC Davis

In the Northern Hemisphere, springtime events are frequently reported as advancing more rapidly at higher latitudes, presumably due to an acceleration of warming with latitude. However, this assumption has not been investigated in an analytical framework that simultaneously examines acceleration of warming with latitude while accounting for variation in phenological time series characteristics that might also co-vary with latitude. We analyzed 743 phenological trend estimates spanning 86 years and 42.6 degrees of latitude in the Northern Hemisphere, as well as rates of Northern Hemisphere warming over the same period and latitudinal range. We detected significant patterns of co-variation in phenological time series characteristics that may confound estimates of the magnitude of variation in trends with latitude. Notably, shorter and more recent time series tended to produce the strongest phenological trends, and these also tended to be from higher latitude studies. However, accounting for such variation only slightly modified the relationship between rates of phenological advance and latitude, which was highly significant. Furthermore, warming has increased non-linearly with latitude over the past several decades, most strongly since 1998 and northward of 59°N latitude. The acceleration of warming with latitude has likely contributed to an acceleration of phenological advance along the same gradient.