From Small to Large: Modeling at the Scale of Ecological Processes to Understand Temperate Forest Range Limits, Biomass, and Traits
Over the next century, global change is expected to fundamentally alter ecological communities. Forecasting the consequences of this change requires an understanding of current drivers of ecosystem form and function, yet the scale of available data is often mismatched to the scale of the ecological processes of interest. In this dissertation, I employ novel statistical modeling on continent-spanning observational datasets, national forest inventories from temperate regions, with the aim to understand how forest communities may respond to future climate. In each chapter, the methodological focus on scale is conceptually linked with the ecological questions posed.
In the first two research chapters, we investigated the relationship between forest biomass and environment across stages of stand succession and levels of species richness. These complementary analyses both estimate that climate exerts strong controls on productivity, and that these controls are dependent on complex interactions. When these interactions are taken into account, we find that differences in plot-level productivity across levels of species richness are small in comparison to the effects of climate, disturbance, and forest management (Ch. 2). Chapter 3 takes a different approach to estimating biomass and biomass productivity through a forest structure framework. The forest structure approach estimates changes in water availability could be a dominant control on future forest biomass, but that the impacts of climate change may be spatially variable across the eastern U.S. because they depend on stand age.
Finally, we synthesized eight national forest inventory datasets from North America and Western Europe to determine whether community-weighted mean (CWM) traits have congruent and predictable relationships with environment. Combining species-level mean traits extracted from the literature with a multivariate species distribution modeling approach, we show that variation in traits can be better explained by environmental covariates in North America than Western Europe. As a result, predictions of CWM traits on one continent based on the model fit from the other tend to have high predictive error and are weakly correlated with observed values. We conclude that differences in evolutionary history and human management may weaken the generality of CWM trait responses to environment in temperate forests.
Overall, this dissertation offers insights into how temperate forests may respond to climate change. Bayesian hierarchical modeling is demonstrated to be an effective way to align the disparate scales between data and ecological process. From an ecological perspective, these results suggest that changes in climatic water availability and drought-related disturbances may be the dominant force driving the function of future temperate forests.
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.
Rights for Collection: Duke Dissertations
Works are deposited here by their authors, and represent their research and opinions, not that of Duke University. Some materials and descriptions may include offensive content. More info