Date of Completion


Embargo Period


Major Advisor

John A. Silander, Jr.

Associate Advisor

Robin L. Chazdon

Associate Advisor

Eldridge S. Adams

Associate Advisor

Daniel L. Civco

Associate Advisor

Richard B. Primack

Field of Study

Ecology and Evolutionary Biology


Doctor of Philosophy

Open Access

Open Access


Shifts in plant phenology can have direct impacts on community and ecosystem-level processes as well as substantial economic impacts including declines in maple syrup production and changes in the timing and vividness in fall foliage as this affects ecotourism. Understanding how plant phenology mechanistically responds to environmental variation is vital to assessing the effects of climate change on ecological processes and to making predictions about the future. This dissertation focuses on understanding the mechanisms of phenological responses of Northeastern North American deciduous forests to climate change from ground-based individual level to the community at landscape scales. Mechanism-based plant phenological models were developed that incorporate spatio-temporal responses from local to regional scales. These models incorporated ground-based, visual phenological observations on individual trees, time-lapse digital photography, and remotely-sensed land surface phenology. A novel, mechanistic modeling framework that utilizes Bayesian survival analysis is developed in Chapter one. This model using remotely-sensed satellite data identified significant effects from chill and heat units on deciduous forest green-up and predicted the future change across the landscapes of the Northeast. In the second chapter, significant environmental factors affecting the timing of fall dormancy of deciduous forests in New England, based on remotely sensed data, were identified and quantified using multiple statistical variable selection methods. Future predictions of fall dormancy timing suggested complex effects from temperature, precipitation, drought-, heat-stress and floods on forests autumn phenology across the landscape. Chapter three focuses on the phenological responses of individual trees to climate/weather variables based on ground-based, visual observations. Linear mixed effects models revealed species-specific phenological responses of leaf coloration and leaf drop to precipitation, drought and floods. In fourth chapter, color indices derived from time-lapse camera imagery of tree canopies were analyzed to determine how fall foliage coloration responds to variation in climate/weather variables. The red color index matched better with visual defined autumn phenology across the dominant species than did the green color index. Linear mixed effects model results suggested that chill in autumn, drought stress in summer and autumn, and heat-stress in summer are all important factors of the timing of peak color in fall foliage.