James S. Clark - US grants

Biomass burning is a key component of past and future global change. Fire profoundly affects atmospheric chemistry, the carbon cycle, and climate and may have done so for millions of years. Literature estimates place carbon sequestration during biomass burning in excess of 20% of the total anthropogenic C releases to the atmosphere, which, failing oxidative losses, could tie up the globe's surface C in less than 106 yr. Past fire is suspected to have been one of the dominant forces shaping existing patterns of biodiversity, past changes in C and O2 cycles, air quality, and climate. Changes in the structure, distribution, and productivity of biomes forecast for the near future are contingent on poorly understood changes in burning practice. Analysis of burning transitions that have attended past environmental change has become a tool for assessing its potential importance with climate and cultural change. I propose to determine the role of fire and to quantify biomass burning emissions that have attended regional changes in climate and changing cultural practices of last 15,000 yr. This interval spans rapid changes in global climate at the end of Pleistocene, changes in seasonality that have forced the glacial/interglacial modes of atmosphere and ocean circulation and therefore fire seasons, subcontinental scale changes in fuels with shifting vegetation types, cultural burning patterns, including slash-and-burn (swidden), modern agriculture, and industrialization. The role of fire during past times of rapid climate change is especially relevant to challenges faced by modern resource managers in the face of environmental transformations of the next decade. Lake sediments contain a legacy of past biomass burning in the form of combustion products that can be extracted from sediment cores. By analyzing changes in the abundances of those constituents in successively older sediments we can determine how the importance of fire has changed together with other aspects of the environment. I plan to address several aspects of the paleofire record, including i) improvement of extraction methods for combustion products in sediments, ii) expansion of an incipient global data base on past emissions, and iii) extension of fire analyses from those currently underway in eastern North America and Siberia to tropical woodlands and grasslands in Brazil and east Africa, currently threatened by changing climate and intensive management, including widespread burning. Together with previous results, we expect to establish a global picture of biomass burning back through the past. Those estimates will be used with our knowledge of changing climate and cultural transition to determine whether fire was among the agents responsible for rapid environmental change. These results will aid our interpretation of its potential role in future global change.

Abstract ATM-9408864 Clark, James S. Duke University Title: Sediment Records of Biomass Burning Biomass burning profoundly affects atmospheric chemistry, the carbon cycle, and climate of biomass burning may have been greater in the past, before large increases in fossil fuel combustion. A principal obstacle to progress on understanding the past role of biomass burning rests with the interdisciplinary nature of the problem. This award supports a NATO Advanced Research Workshop to assemble the paleo records of past combustion and atmospheric and terrestrial expertise needed to interpret the effects of emissions on past global changes. Specific goals of the workshop are to: i) Better characterize sedimentary constituents and parameterize transport to sediments; ii) Calibrate lacustrine, marine, and ice-core records of biomass burning; iii) Identify the Data sets that could be used to quantify past biomass burning; iv) Estimate past mass fluxes of combustion-derived particulates and organic tracers; and v) Estimate past rates of biomass burning The program includes a balanced representation of fire ecologists, paleoecologists, and atmospheric chemists.

Antonovics 9520754 The majority of plant species have both male and female flowers. However, some species have male and female flowers on different plants. Plant ecologists and geneticists are keenly interested in the forces that determine the transitions between these two types of breeding systems. The proposed research will study how this transition is being manifested in an herbaceous plant, Astilbe biternata, which has an intermediate breeding system. Understanding this transition will increase knowledge of how genes can be transmitted in plants and will help to plan long-term conservation strategies for dioecious species.

Fire is an important component of the functioning of ecosystems. Fire also is currently being used extensively by human populations worldwide in the conversion of forest to pasture and agricultural land. The historic role of fire and the frequency of fire in different ecosystem types is important in assessing present-day management decisions. This project looks at 6,000 years of fire history in changing landscapes of the boreal forests and grasslands of North America. Pollen records and charcoal records will be combined to examine interplay between vegetation and fire. This will provide a comparison of how modern emissions from modern fires differ from those of the past. This research is important because it places the role of fire in a regional perspective. Methods are to be developed and tested which will allow fire history from individual lakes to be extended to a regional view of how fire structures and affects ecosystems.

9701088 Clark Issues related to management and the potential change in forest structure due to climate change demand increased knowledge of the basic biology underlying forest dynamics. Theoretical studies have demonstrated the potential for variations in life history strategy to explain both the co-existence of species in forest communities and the patterns of forest stand development. Validation of such studies have been hampered by lack of species specific field data. The PIs seek to explore whether the level of variation actually seen in the field for two life history parameters, growth and mortality, can explain compositional differences observed between forest types. To these ends, the PIs are using tree ring data to derive species specific functions relating growth and size class and growth and mortality for tree species growing in five types of southern Appalachian forests. Simple models will be used to explore the implications of observed variations in growth and mortality for forest stand development.

9972595
Clark
Fires play an integral role in the vegetation and climate dynamics of the boreal forest and contribute significantly to the amount of aerosols and gases in the atmosphere both today and in the past. An understanding of the interaction between long-term climate changes and fire regime is critical to our understanding of how future fire dynamics will be effected by climate and vegetation. In order to understand how fire dynamics will be affected by future climate change and vegetation changes, it is imperative to determine how past climate and vegetation shifts have changed past fire regimes. This dissertation research focuses on the prevalence of fire in different climate regions and during past climate changes in Western Canada and Alaska, in order to determine sensitivity of fire regimes to climate changes. The research will reconstruct fossilized charcoal and pollen records across modern climate moisture gradients and past climate and vegetation changes. This will allow assessment of the importance of fire under these different climate and vegetation conditions.

0073171
Clark
The range expansion by forest trees following rapid climatic warming during the early Holocene provides a useful analog for predicting the ability of trees to respond to future climate change. Currently, however, early Holocene migration rates determined from fossil pollen records can not be easily reconciled with models of seed dispersal. This dissertation research is an attempt to reconcile the historical record and the dispersal biology of common eastern deciduous forest trees. First, migration routes and glacial refugia based on the distribution of molecular markers (cpDNA haplotypes) throughout the modern range of Fagus grandifolia, Acer rubrum, and Quercus rubra will be reconstructed. Maps of Holocene range expansion based on molecular and pollen data will provide improved estimates of climate driven migration rates. This study will then evaluate the ability of seed dispersal models to account for these migration rates using a dispersal model developed by this research group. The model will be parameterized using the distribution of seedlings established in old-fields and closed forests. By integrating fossil pollen and molecular data, this study will develop a more complete record of historical change. Explaining that change in terms of dispersal biology will allow a mechanistic basis for evaluating future change.

9981392
Clark
Understanding how forests respond to environmental change requires models that extrapolate from short-term observations to long-term dynamics. The current gap dynamic paradigm does this extrapolation under the assumption that recruitment is governed by conditions within canopy gaps 30 meters in diameter. Much empirical evidence and experimental tests of the paradigm suggest that diversity in many forests may depend on much larger canopy gaps than assumed in the models now used to forecast forest response to global change. If the underlying assumptions of the models are unrealistic, then predictions under novel environments of the future are likely misleading. In this project, a series of field experiments and model test will be conducted to test (1) gap sizes at which recruitment limitation is severe, and (2) whether maintenance of forest diversity depends on large gaps. Field studies will both test for specific life history stages at which recruitment limitation occurs and the factors responsible for limitation, while also providing parameter estimates for modeling studies. The extent to which fecundity, seed dispersal, predation in seed banks and availability of soil microbes affect recruitment in the closed forest understory and in gaps will be tested. Two modeling approaches will be used to determine the population and community consequences of experimental results. To determine the effects of demography observed in experimental and natural gaps on growth rates the PIs will use parameters in a stage structured population model. Analysis of the model will provide rate estimates implied by demography for each gap type and the contributions of each life history stage to population growth. An individual based model will be used to incorporate spatial elements, including seed dispersal and competition for light. Analysis of this model will provide estimates for the effects of gap size on forest diversity. Together, field experiments and modeling studies will allow the assessment of the contribution of disturbance to forest dynamics and thus, assessment of how forests may respond to novel environments in the future.

Anticipated rapid climatic change over the next decades to centuries raises the concern that plant populations will be unable to migrate fast enough to track changing environmental conditions. Migration rates depend on dispersal of seed and on barriers to population spread, such as mountain ranges, large water bodies, and patterns of urban and agricultural land use. Range expansions at the end of the last ice age provide evidence that populations tracked global change in the past, but current evidence does not indicate the speed of these migrations. The proposed research will determine the pathways of past population spread and provide insights into rates of corresponding migrations. We will construct and analyze maps of chloroplast DNA variation across the ranges of common eastern North American tree species. Such maps reveal the "genetic fingerprint " of late-glacial refugia and post-glacial migration routes, and complement existing data for fossil pollen. The maps of post-glacial migration we will provide a framework for analysis of population expansion. Results will be used to test hypotheses concerning how growth of trees and dispersal of seeds affect the potential of plants to track rapid environmental change.

EIA-0131905 -James Clark-Duke University-BDEI: Computation and Uncertainty in Ecological Forecasting.

Planning for global change and decision making will be improved by access to reliable forecasts of ecosystem change. A recent initiative by the Ecological Society of America identifies three challenges that must be met for forecasts to be successful: 1) computational approaches (algorithm development and data structures) that would permit simulation of complex systems, 2) feasible methods to track statistical uncertainty, and 3) data inadequacy.

The project "BDEI: Computation and uncertainty in ecological forecasting" is an incubation that addresses these three challenges with an integrated approach. We propose to develop new computational and statistical techniques that will provide the capacity to forecast at broader spatial and temporal extents than possible with current approaches. The success of the proposed techniques will be evaluated using the data gathered from field experiments.

A team of three researchers-an ecologist (Clark), a computer scientist (Agarwal), and a statistician (Lavine)- propose a working group for Fall and Spring 2002 that will focus on forecasting forest compositional change. The working group will integrate new computational techniques into stand simulators and use models to estimate
uncertainty. The working group will develop an agenda for a broad initiative in ecological forecasting as basis for future proposals.

This study will utilize high-resolution analysis of cores from closed-basin lakes in the northern Great Plains to determine the frequency, intensity, duration, periodicity, and synchrony of droughts during the Holocene and to examine how vegetation and fire responded to drought. Since vegetation and lakes in the Northern Great Plains are highly sensitive to drought, this region offers an extraordinary opportunity to document decadal- to-century scale climate cycles in the mid-continental United States.

The working hypothesis is that the signal of drought from a single site is regional and that droughts were synchronous across the region. To test this hypothesis, the study will utilize lake-sediment mineralogy, fossil pollen, carbon isotopes of charcoal, and charcoal abundance to develop detailed reconstruction of drought cycles from the Northern Great Plains at a decadal scale. This will facilitate evaluation of the teleconnections between mid-continental climate and the North Atlantic region, where recent investigations have linked century to millennial scale climate oscillations to variations in solar irradiance.

This research has the potential for broad impact in range of physical and social sciences. Assessment of drought impacts forecast by atmospheric models for the Northern Great Plains requires understanding of natural drought variability. Evidence from paleoscience suggests that 20th century droughts (e.g., the Dust Bowl) do not provide perspective on the range of severe droughts that have occurred in even the recent past. Historical evidence is incomplete and paleoscience data have not yet been assembled at appropriate temporal and spatial scales to assess the intensity, periodicity, and impacts of past droughts.

A Summer School: Uncertainty and Variability in Ecological Inference, Forecasting, and Decision -- An Introduction to Modern Statistical Computation
Clark, James S.
Duke University

A two-week, graduate/post-graduate level summer school is proposed to introduce ecologists and earth scientists to modern statistical computation techniques that have emerged over the last decade. Ecological inference and forecasting are limited by large and diverse sources of variability that operate at a range of scales. Hierarchical Bayes and Markov chain Monte Carlo simulation provide powerful tools for analyzing processes characterized by multiple sources of uncertainty and variability. However, adoption of these techniques in ecology has been hindered due to limited training options. In the proposed summer school, leading statisticians and ecologists will provide day-long presentations and hands-on training with computation techniques. Students will include advanced graduate students and postdoctoral associates selected by an open application process. They will participate in small working groups that will each produce a chapter to be included in a published volume, together with lecture notes. By training a select group of young, quantitative ecologists and earth scientists, who will in turn train others as well as utilize this training in their research, this summer school will expedite the dissemination of modern statistical computation techniques. In addition, the volume that will be published as a result of this course will serve as a useful reference for a much broader audience. Adoption of these modern methods by ecologists and earth scientists will strengthen the field of ecological forecasting, enhance its reputation, and increase it usefulness to resource managers and policy makers.

ABSTRACT
Forests can potentially support many species, but not all do. Understanding why forests support different numbers of species and how forest diversity can be threatened by climate change, land-cover change, and rising atmospheric CO2 requires data and modeling approaches that account the for complex interactions among species and with their environment. We must assimilate information derived from many sources, including field experiments, monitoring, and remote sensing. Current methods for prediction cannot take advantage of the disparate types of information currently available, and they necessarily ignore many of the important variables. They cannot be expected to yield useful predictions.

This project proposes to exploit new computational methods to incorporate the many types of information available and to assess the factors that affect biodiversity. To accomplish these goals the project involves combining powerful new techniques in computational statistics, algorithms, and data structures with extensive data on ecological interactions. We integrate these new tools to test the assumptions that diversity is maintained specific competitive relationships among species, disturbance, or a combination thereof. We use the same models to predict potential outcomes of current global change.

The intellectual merits of the proposed activity include new insight into the factors that control diversity in nature (as opposed to in simple models), which is a basic requirement for wise stewardship of forest resources and biotic diversity. Moreover, the technical advances, particularly the integration of modern computation with ecological process, will be of immediate application for a range of environmental issues.

The broader impacts of the proposed research will include understanding of forest management and conservation, carbon sequestration, responses to climate change and rising CO2, and invading species. Our research will also establish stronger links between ecology and computer science.

Abstract
NSF-0430693


Consequences of global change for land cover, carbon cycles, and biodiversity loss involve complex interactions at fine scales, such as resource availability in forest understories, to regional land-cover, climate, and CO2. Global change research requires models developed through careful study of local phenomena that can be extended to landscape, regional, and global scales. Unfortunately, environmental scientists have been limited in their ability to determine how factors that operate at different scales impact landscapes.

The primary long-term goal of the research is to enhance the ability of biology and geoscience research programs to acquire, analyze, and distribute high-resolution GIS databases of important environmental attributes. In support of this goal the computer science team will develop new techniques to extract forest attributes in the form of GIS databases from remotely sensed data. The computer science team will build an aerial remote sensing platform and a suite of analysis tools for creating GIS databases of environmental attributes with sub-meter geo-registration and elevation accuracies.

The image acquisition, analysis and GIS tools developed by UMass and MHC
provide the critical broad-scale, yet high-resolution, data needed to parameterize and validate models used to study global change. The products of these analyses will be integrated within a modeling framework at Duke University that includes extensive field data, application of new statistical computation methods, and development of a stand simulator. The combined effort will be used to determine how diversity is maintained in forest stands based on a comprehensive accounting of environmental impacts and uncertainties.

A two-week Summer Institute for graduate and post-graduate students will be conducted to introduce environmental scientists to modern statistical techniques that have emerged over the last decade. Ecological inference and predictions are currently limited by the many sources of variability that operate at a range of scales. There are, however, powerful tools available for analyzing processes characterized by these sources of uncertainty and variation, including hierarchical Bayes and Markov chain Monte Carlo simulation. However, the widespread use of these techniques in ecology has been hindered due to limited training options. The 2nd Summer Institute will follow the format of the highly successful 1st Institute held in June 2004. Participants will include advanced graduate student and postdoctoral associates selected by an open application process. Leading statisticians and ecologists will provide day-long presentations and hands-on training with computational techniques. Students will participate in small working groups, with the aim of producing a published volume and lecture notes. The 2nd Summer Institute will take place in 2006, retaining aspects of the format that worked well in the 1st Institute, and incorporating modifications that will improve its effectiveness.

By training a select group of quantitative environmental scientists who will in turn train others and use this training in their research, this 'summer school' will expedite the spread and use of modern statistical computation techniques. Adoption of these modern methods by ecologists and earth scientists will strengthen the field of ecological forecasting, enhance its reputation, and increase its usefulness to resource managers and policy makers. Many of the techniques that will be covered should directly enhance capacity to use network information available in distributed databases.

The next generation of wireless sensor networks will be dynamic systems with the potential to
revolutionize understanding of environmental change, provided they can assimilate large amounts of heterogeneous data in real time, rapidly assess (optimize) the relative value and costs of new data collection, and schedule subsequent measurements accordingly. Thus, they are Dynamic Data Driven Application Systems that integrate sensing with modeling in an adaptive framework. Keen interest in broad application of wireless sensing of the environment, as in NEON and CLEANER, awaits DDDAS technology that can estimate the value of future data in terms of its contribution to understanding against the costs of deployment, acquisition, transmission, and storage. This balance is especially important for environmental data, because networks will typically be deployed in remote locations without access to infrastructure (e.g., power), and sampling intervals will range from meters and seconds to landscapes and years, depending on the process, the current state of the system, the uncertainty about that state, and the perceived potential for rapid change. Network control must be dynamic and driven by models capable of
learning about both the environment and the network. The focus of this project is the dynamic sensor network application involving understanding how biodiversity and carbon storage are influenced by global change. Specifically, this project is designed to learn how the growth, survival, and reproduction of forest trees are influenced by changes in climate, CO2 and disturbance, in the context of these and other variables that can fluctuate rapidly. This goal involves models of how tree growth and resource allocation are influenced by variables that can be understood through adaptive sampling across diverse scales in both time and space. The project will enable a general framework for dynamic data-driven wireless network control that combines environmental modeling and sensor network modeling both in and out of the network. Out of the network, environmental modeling entails full assimilation of all information, with exploitation of computing resources available there. Environmental modeling in the network is based on simplified representations that provide real-time, approximate answers. The in-network control model provides rapid scheduling for new measurements, and it communicates network information to the server, for diagnostics, supervisory control, and data assimilation. Periodically, the in-network model is updated
based on this most complete understanding of the environmental variables, parameters, and battery life. Specific goals are (i) to construct a wireless sensing and networking infrastructure that supports a new paradigm of joint in-network and supervisory measurement, modeling, and prediction, (ii) to develop the modeling strategy needed to combine system understanding with costs for efficient wireless sensing of the environment, (iii) to make significant progress in understanding the maintenance of biodiversity and in measuring ecosystem properties, and (iv) to improve collaboration between computer sciences, engineering, statisticians and environmental scientists.

Plant pathogens have long been suspected to play a role in maintaining tree diversity by controlling seedling growth and survival, yet relatively little is known about pathogens in unmanaged temperate forests. To better understand how forest pathogens influence coexistence of tree species, mixed-species plots of tree seedlings were planted in two North Carolina forests. Seedling mortality is monitored frequently, and pathogens are identified from dead seedlings using cultural and DNA-based methods. Other relevant data about tree demography (seedling mortality, adult density) and environmental conditions (light, soil moisture) are collected simultaneously. These data are combined to identify the major fungal and oomycete pathogens in these forests, the environmental factors underlying their distributions, and the role they play in shaping forest communities. This project provides insight into understanding the spatial and temporal dynamics of plant pathogens, which may help to reduce their negative impact in natural and managed forest ecosystems. This information is of economic significance and can be used to improve decisions about the density, location, and timing of planting and harvesting in forests or restoration areas such that new disease management strategies can be developed that promote increased plant productivity and sustainability. This study has the potential to enhance current management practices and to inform models that predict future tree distributions associated with global change. This project also promotes outreach, teaching and training by involving undergraduate and high school students as work-study assistants and summer research technicians.

Predicting biodiversity, i.e., abundance of species, in response to climate change is a goal of environmental change research. Despite recent valuable advances in understanding biodiversity and climate, the current grasp is limited. There are two widely recognized obstacles: first, because of the complexity of the underlying processes, the existing models intended for understanding and prediction are not (computationally) scalable. Second, the coarse-scale environment models fail to capture interactions among species, which control biodiversity, and the models based on fine-scale, short-term observations are unable to make long-term predictions. This project aims to develop a prediction framework that coherently combines broad-scale pattern data with fine-scale data on species interactions and that is computationally scalable. It focuses on prediction at the geographic scale and in using geographic-scale data to better understanding at the scales where species interactions occur.

The goal is to develop a multiscale modeling framework and to design algorithms that make environmental models computationally scalable. The approach hinges upon strong interplay of algorithmic and statistical techniques. Statistical inference brings stochastic modeling sophistication in space and time, yielding improved characterization of the process and the possibility of full inference. Sophisticated algorithms make models and processes scalable and provide trade-offs between accuracy and efficiency.

The project draws on a wide range of topics in computer science and statistics, including geometric algorithms, approximation algorithms, hierarchical specifications within a Bayesian framework, and space-time process modeling. The problem areas address in the proposed prototypical example indicate more broadly applicable consequential challenges for both computer science and statistics. These include maintaining/updating distributions and summaries, dynamic algorithms, data driven algorithms, stochastic optimization, and assessing uncertainty and multi-scale nonlinear interactions in inference. Techniques for obtaining trade-offs between conflicting goals are needed in order to optimize the overall performance of the model.

Fungal pathogens are thought to control biodiversity of forest trees through selective mortality of competitively dominant species that would otherwise threaten less competitive species. Climate warming may thus increase the extent and severity of disease. As pathogens increasingly survive mild winters, reproductive rates increase and plant defenses suffer from drought and temperature stress. This research will assess the extent to which pathogens regulate tree seedling health, identify the fungi involved and their effects on tree growth and survival, and determine how those interactions are affected by the temperature changes predicted for mid-century. The study will determine how temperatures affect their hosts when temperatures increase, within the context of simultaneous competition from other tree species, using a warming experiment where tree seedlings are exposed to soil and air temperature increases of 3°C to 5°C in North Carolina and Massachusetts. Once seedlings emerge, germination, survival, disease symptoms, and growth will be monitored. Pathogenic fungi and oomycetes residing on both seeds and seedlings will be identified using a combination of culture and DNA sequence-based methods. The effects of warming on pathogen incidence, infection, and seedling demography will be assessed using models that account for the combination of temperature, pathogens, and competition with other trees. By incorporating different sources of data into the analyses, this research will provide insight into the basic biology of forest risk as the natural capacity of trees to defend themselves changes under a warmer climate. The proposed research will provide insight into spatial and temporal dynamics of plant pathogens in natural systems, and how these dynamics might change under a warmer climate. An improved understanding of the diversity and composition of populations of plant pathogens and environmental factors can both enhance current management practices and inform models of future tree distributions. Teaching, training, and learning will be an integral part of this study, as it will require the assistance of undergraduate and high school students as work-study assistants and summer research technicians. In addition, a survey-based data collection component will be initiated to engage K-12 and the interested public, which combines this diversity research with outreach activities supported by Duke Forest. The results of this study will be disseminated to high school students during the annual Duke Forest Research Days, and to the general public in a series of forest pathology hikes in the Duke Forest.

Climate change is rapidly transforming forests over much of the globe in ways that are not anticipated by current science. Large scale forest diebacks, apparently linked to interactions involving drought, warm winters, and other species, are becoming alarmingly frequent. Models of biodiversity and climate have not provided guidance on if, where or when such responses will occur. Instead models tend to provide potential numbers of extinctions, but such forecasts are not linked in any mechanistic way to the processes that could cause them. Both modeling and field studies rely on aggregate measures of species presence or absence, or their relative abundance at regional scales. However, climate acts on individuals. Aggregating data on individual trees to the level of a whole species hides or may even change predictions of climate effects. This study aims to link individual scale tree processes to regional species level responses by sampling and analyzing data about individuals across their entire range and corresponding range in climate conditions. It will use data from existing research sites, plus the platform of sites that form the core of the new National Ecological Observatory Network. These data will be collectively synthesized and used to develop computer models that can help determine when and where predicting climate impacts on biodiversity is a plausible goal. The models will also reveal where surprises are likely to occur and can provide feedback to expectations of individual tree health and vulnerability to environmental changes.

This study will provide the first forecasts of the vulnerability of forest biodiversity to changes in climate that are directly linked to the biological processes that are most sensitive. The goal is to provide forecasts of the distribution, growth, reproduction and risks of mortality for tree species making up the nation's forests. These predictions will help scientists, forest managers and policy makers anticipate the combined risks of increasing drought and longer growing seasons. Methods and results developed during this project will be disseminated through workshops for training resource managers, as well as graduate students and postdoctoral associates at a number of universities.

Global warming is expected to have strongly negative effects on many species. One way these effects might be reduced is if species can change their geographical ranges as climate changes; if species can migrate to cooler places such as higher latitudes or elevations as temperature rises, they may be able to stay in the same climate by changing place. However, the preliminary research for this project shows that species of trees in the eastern U.S. have not moved northward. The project will test two alternative hypotheses to explain this, both based on whether young and adult trees differ in their ability to tolerate a wide range of environmental conditions. Researchers will expand a current species distribution model to include more explicit effects of juvenile and adult responses to temperature, and use data from a national forest inventory and climate measurements to compare the abundances of juveniles and adults.

Anticipating the impacts of climate change on U.S. forests is an important issue for forest managers and for the nation as a whole. Results from this project will help plan strategies for maintaining forest productivity and for substitutions of alternative land uses. The project will also strengthen collaboration between Duke University and the USDA Forest Service, and train a Ph.D. student and an undergraduate student.

Drought is a nationwide threat to biodiversity, one of the challenges that motivated development of the National Ecological Observatory Network (NEON). Droughts are already affecting organisms from microbes to vertebrates and higher plants. The long-term consequences are unpredictable because each species responds to other species, as they too respond to drought. For example, a species that is insensitive to drought will still respond if the plants on which it feeds or the predators that consume it respond to drought. Ecologists could better anticipate drought effects by developing tools to integrate the information from many species as they respond both to climate and each other. With data from key taxonomic groups monitored by NEON the researchers will use a new approach, joint modeling of species and drought, to develop a predictive framework for biodiversity analysis.

Using relevant NEON, predictive models, which combine multiple species, functional types, and functional traits, will be developed to identify commonalities in species responses to changing drought, reducing the dimensionality of thousands of species to groups that can be predicted. Indirect effects of species interactions will be emphasized in planned modeling activities. For example, a tree species may increase in crowded stands only when others decrease. The effects of temperature usually depend on precipitation. The results of this collaboration involving ecology and statistics will be of interest not only ecologists, but for all global change scientists and policy makers.

With support from the Directorate for Geosciences and the NSF 2026 Fund Program in the Office of Integrated Activities, Professors Dana Hunt, Mark Borsuk, and James Clark at Duke University conduct research that provides new insights into the factors that shape microbial productivity and function in the oceans as well as how this change during extreme events such as hurricanes. The driver of this research comes from the fact that marine microbes provide essential ecosystem services, including primary production (photosynthesis) and organic matter turnover, that sustains all marine organisms. That said, it still remains unclear as to what extent microbiomes are shaped by environmental factors, such as temperature and primary productivity, that can be altered by season, disturbances, global change, and other factors. This research combines long-term observations at a coastal site at Beaufort Island, North Carolina and uses these data to capture annual changes in microbiomes and their environments using high frequency measurements that were taken before and after hurricanes Florence (2018) and Dorian (2019). Examining the impact of hurricanes on marine biomes is important because hurricanes are multi-factor disturbances that introduce both foreign freshwater and terrestrial microbes into a stable system while altering salinity, nutrients, and organic matter in the coastal ocean. This work combines information from field observations and modeling to develop new approaches that will allow the differentiation of factors that often co-occur in field samples, such as warmer temperatures and higher primary production that occur during the summer months in the coastal Atlantic Ocean. By integrating multiple aspects of microbiome research, this work deepens current understanding of the coastal ocean microbiome system and its functionality. It also develops new testable hypothesis to guide future research. Broader impacts of the work include advanced training for undergraduate, graduate, and postdoctoral students, as well as translating research results into products for K-12 students and the public. Additional impacts include the production of detailed user manuals and training materials for software developed in the course of the project to facilitate the use of research results for future microbiome research and undergraduate education.

This research leverages an established decade-long microbial time-series, the Piver?s Island Coastal Observatory (PICO, Beaufort Inlet, NC USA) to improve the modeling of microbial populations and their relationship to changing environments. With 10 years of weekly (or more frequent) microbial community SSU rRNA gene sequence datasets, coupled with the suite of sample, in-situ, and environmental parameters, the PICO dataset is one of the most complete, long-term datasets for coastal ocean microbiomes. The work carried out uses the application of Bayesian modeling to the PICO time series to improve understanding and predictions of microbiome responses to ocean conditions. Bayesian models are well suited to microbial systems because they have the ability to handle sparse datasets, capture non-linear responses to environmental changes, and include impacts of disturbances. This research integrates microbiome applications and the Bayesian model gjamTime. This combination has the potential to transform microbial ecology by leveraging advances in multivariate time-series methods that accommodate the dependence among individual taxa and their environment over time. One goal of the project is to test model predictions using time-series data from natural disturbances (i.e., hurricanes) at the Beaufort Inlet site and explore various key environmental parameters such as temperature (+3 °C) and primary production as key environmental parameters. Similar work will be done more broadly for the ocean. Impacts of the research extend beyond the targeted coastal dataset as, if successful, the approach can be applied to other diverse study systems such as soil and human microbiomes. It can also be used to address questions about environmental filtering, disturbance and stochasticity, each of which is critical to understanding the factors and processes that govern microbial responses to environmental change.

This project responds to the NSF2026 Idea Machine winning entries of "Global Microbiome in a Changing Planet" and "Imagine a Life with Clean Oceans"

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

The sustainability of North American forests depends on seed production by trees as well as seedlings that must establish for the next generation of trees. For most of North America, neither the amounts of seed that are produced, nor how much of that seed survives to become adult trees is known. Population spread beyond current frontiers will be governed by seed production of trees (fecundity), germination, and seedling survival—the capacity of trees to produce seed and disperse it to the habitats where populations can survive in the future. Planning for environmental change impacts requires this knowledge to anticipate tree species migrations and its impacts on the birds and mammals that depend on forests for habitat and food. Understanding these forest recruitment responses requires a methodological shift from the current method of monitoring of seeds, seedlings, and consumers on small plots to extensive sampling methods that can be implemented at biogeographic scales. This study combines continental scale tree fecundity estimates with a new generation of monitoring and synthesis methods for integrating tree fecundity, seedling success, and its impacts on animal consumers. This research will quantify current trends across the continent, the changes in forests that are happening now, and the habitat changes that are causing them. Development of a biogeographic network of tree fecundity and recruitment will provide the monitoring platform needed for science and management of future forests. Broader impacts will focus on stakeholder integration, including conservation and management planning, information transfer to stakeholders in federal and state agencies, and citizen science outreach. Products of the study will have immediate application to forest regeneration practices in the coming decades. Agency and NGO stakeholders will advise and disseminate products of the study. <br/><br/>New analytical tools will identify where tree recruitment is limited in North America, its rate of change, and what’s causing it. The project focuses on three recruitment stages, seed supply (seed mass per tree abundance), seedling establishment (seedlings per seed mass), and recruitment rate (advanced regeneration per seedling). Each recruitment stage will be linked to climate and habitat variables and to the vertebrate consumers of seeds, fruits, and nuts. Extensive gradient sampling (EGS) is a new approach to estimate the key demographic rates that are relevant at the scale of habitats or plant communities, while combining it with traditional data already available from the meter-scale intensive monitoring sampling (IMS). The project will include data collection based on this new approach, (EGS) of fecundity, tree recruitment, and vertebrates distributed across climate and habitats. Predictive vertebrate modeling (PVM) of activity based on camera traps (snapshot USA, NEON, and this study), live trapping (NEON sites) and bird point counts (BBS, NEON, and eBird) across North America will be conducted by the research team. By understanding tree recruitment and the vertebrates that depend on them, this study will i) identify the species that are limited by recruitment, including the habitats and stages where limitation occurs, ii) quantify the relationship with vertebrate activity, and iii) evaluate predictive distributions of change that account for climate-vertebrate interactions fitted to data. Quantifying tree fecundity and animal-consumer relationships at biogeographic scales will provide a foundation for the next generation of efforts to understand food web implications of environmental change.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.