The global climate is undergoing changes in temperature and precipitation that are altering the abundance and distribution of species around the world. Species that are important from conservation, ecological, cultural, and economic perspectives may face decline and extinction in coming decades. Phenology, the timing of biological events, has emerged as a key indicator of species response to global warming and other aspects of climate change. Phenology includes the flowering and leafing out dates of plants, first arrival dates of birds, and other events throughout the year. The ability of scientists to predict how climate change will affect species and biological communities remains very limited, especially at the scale of countries and continents. Researchers in this project will study the year-round phenological trends since 1953 of numerous plant and animal species using data from 176 meteorological sites across Japan and South Korea. These records cover more species, a longer time period, and more sites than any comparable data set from the United States, and can be used to develop innovative methods of analysis that can be applied to species in the United States and elsewhere. At field sites in Japan, the researchers will examine ecological interactions, e.g., between plants and pollinators, to determine the potential for ecological mismatches among species with different responses to climate change. The U.S. researchers will train undergraduate and graduate students and will work with scientists from Japan, South Korea, China, and Europe, and the new U.S. National Phenological Network, thereby strengthening international connections among climate change researchers.
The overall objectives of this project were to understand and predict how a range of different plant and animal species respond phenologically to variation in climate and weather across geographical regions over time, and how they may continue to change in the future. Phenology is defined as periodic plant and animal life cycle events, including such things as: bud burst and flowering of plants in the spring, and leaf color, drop and plant dormancy in the fall; the arrival of animal migrants, or the onset of activity in hibernating animals in the spring, along with vocalization and breeding, or the cessation of animal activities in the fall; etc. Our research results demonstrated that we were able to accurately and consistently predict a range of phenological responses to variation in past climate and weather for a spectrum of plants and animals. We extended this study to include community and landscape-level phenological responses, focusing on woody plant species. To achieve these results we developed a suite of novel statistical and analytical tools. The results of our research have been presented in eight peer reviewed scientific publications to date, and at over fourteen different regional, national and international scientific meetings and colloquia. Key outcomes of our research include: 1. Our finding that for most species, spring phenology has been shifting earlier while autumn phenology has been shifting later over the past few decades, with the timing of events changing more quickly in autumn compared to the spring. 2. Cherry blossom timing in Japan has been shifting earlier in response to progressively earlier physiological chilling and heating requirements being met (breaking winter plant dormancy) with warmer winters and spring seasons. Moreover, earlier flowering cherry species and varieties were found to be more sensitive to heating than later flowering groups. Climate projections suggest that each cherry species or variety we examined will flower about 30 days earlier on average by 2100. Such dramatic shifts in the flowering times of cherry trees could have significant economic implications for important cultural festivals in Japan and East Asia, which are multi-billion dollar industries. 3. Warmer daily temperatures in the winter and early spring, observed over the past 15 years in New England, have result in earlier green-up of deciduous forest landscapes, compared to cooler winters and springs. This reflects an earlier meeting of physiological heating and chilling requirements and the breaking of winter dormancy for these woody species. However, chilling temperature requirements had a larger influence on green-up than heating temperatures. We also found species-specific differences in forest green-up patterns. Greater oak dominance in forests led to later green-up. Projecting into the future (i.e. 2046-2065) our model output suggested advanced green-up (8-48 days earlier) driven by changes of heating and chilling temperatures, but green-up in some south coastal areas may show delays in green-up. 4. Focusing on fall phenology in New England forests, timing of dormancy (i.e. leaf drop) responds not only to decreasing temperatures (chilling) in fall, but also to soil moisture condition, drought, frost, heat and other extreme weather events throughout the growing season. Moreover, interactions among these variables also contribute to fall phenological timing. Deciduous forests, dominated by maples and birches, showed higher sensitivity to weather condition changes affecting fall phenology, than forests dominated by oaks. These findings and projected changes in the future may have significant economic implications for the timing of the leaf-peeping season in New England and the maple sugaring season, which are multi-billion dollar industries. 5. Our study of selected woody, East Asian invasive species (oriental bittersweet and Japanese barberry) showed significantly longer spring through fall phenoperiods than ecologically comparable native species (spice bush and fox grape) – up to 4 weeks longer. This finding may have implications on the successful spread and persistence of these invasive species in the landscapes of northeastern North America. This research supported two Ph.D. graduate student dissertations and the training of four undergraduates in the science of phenology and climate change. This project involved collaboration with researchers at several institutions internationally in Japan, Korea and China, as well as nationally in the US, including Duke University, Boston University and the University of Michigan. The success of the project was also dependent in part on interdisciplinary collaborations with statisticians and climate scientists at the University of Connecticut and at Duke University.
