Spatio-temporal mismatches in species' responses to climate change in Japan and South Korea

Background/Question/Methods Ecological relationships may be disrupted when species respond in different ways to climate change. However, evidence for such mismatches in response is generally lacking because of the difficulty of obtaining informative data. Here we present an analysis of the phenological responses to climate variation of twelve species over a three month period from mid-winter to spring: included are first flowering or budburst of six plant species (Camellia, Ginkgo, Morus, Prunus, Taraxacum, Wisteria), and the first appearance or singing of six animal species, including two insects (Pieris, Polistes), a frog (Rana) and three birds (Alauda, Cettia, Hirundo). These species are not directly linked ecologically, but they are distributed among three different trophic levels. Phenology was monitored at 176 government meteorological stations varying in latitudes and elevations across Japan and South Korea from 1953 to 2005, and in some cases even longer. We developed a hierarchical Bayesian model to examine the complex interactions of temperature, site effects, and latitude on phenology. Results/Conclusions Winter temperatures (November through March) have risen by an average of 1.2 °C over the region from 1953 to 2005, with the most warming occurring in urban areas. As expected in response to warming, three plant species are flowering earlier and two others are leafing out earlier over time at most sites; however, the phenology of the six animal species and one plant species either is changing inconsistently, or contrary to expectations, is occurring later over time at most sites. Results show species-specific variation in the magnitude and the direction of their responses to increasing temperature, which also differ from site to site. This variation among species and sites reflects the complexity of species' responses to change. At five representative sites the differences in phenology among species are forecast to become greater with warmer temperatures; at a sixth site, differences between species will become less. As a consequence, the phenological relationships among the species are changing, and will continue to change in the future with unknown cascading effects for the communities in which they reside. More field studies are urgently needed to document and track these changing ecological relationships. Our results clearly demonstrate that it is difficult to generalize from one site to another because of differences that exist between the phenological responses of individual species and sites.