In the complex world of forest ecology, one of the enduring scientific quests has been to unravel the mechanisms that sustain biodiversity within ecosystems and to predict their future trajectories. A groundbreaking study recently published in the journal Science propels this endeavor forward by introducing a sophisticated model that harnesses both census and genomic data to forecast species abundance fluctuations in forest communities. Spearheaded by James O’Dwyer, a plant biology professor at the University of Illinois Urbana-Champaign, alongside collaborators Andy Jones from Oregon State University and James Lutz from Utah State University, this work exemplifies an innovative fusion of ecological monitoring and population genomics.
Decades of painstaking fieldwork in forests have long contributed to our understanding of species diversity dynamics, but predicting how species populations rise and fall over time has remained fraught with complexity. Forests are inherently dynamic systems where myriad factors—from competition among neighboring trees for sunlight and nutrients to varying environmental conditions such as rainfall and soil quality—interplay to shape community composition. The team’s research harnesses the rare advantage of extensive longitudinal data from the Wind River Forest Dynamics plot in southern Washington, one of the Smithsonian Forest Global Earth Observatory’s many long-term ecological monitoring sites.
Previous efforts by O’Dwyer and colleagues laid the foundation for this current model. Their 2023 investigation, published in Nature, employed life history traits—species-specific timelines of growth, reproduction, and mortality—to develop matrices that estimate each species’ “effective population size.” This concept, rooted in evolutionary biology, encapsulates the number of individuals contributing genes to future generations, which is generally lower than the straightforward census count due to differential reproductive success. The 2023 study illuminated how combinations of life history parameters could determine whether multiple species coexist stably, thereby maintaining biodiversity.
Building on this insight, their follow-up work broadened the scope from pairs of species to entire multispecies communities in tropical forests. There, the effective population size became a predictive tool for short-term population fluctuations, highlighting its applicability across diverse forest types. However, the necessity of extensive life history data spanning decades posed practical challenges, impeding broader implementation in less well-studied ecosystems. This bottleneck catalyzed the development of a more streamlined approach leveraging genomic data.
Andy Jones led efforts to sequence partial genomes from approximately 100 individuals per species for eight dominant tree species within the Wind River plot. Unlike full-genome sequencing, this method targeted key gene regions hypothesized to retain imprints of species’ life history traits and evolutionary dynamics. Genetic variation patterns, particularly the balance between random and nonrandom gene associations—termed linkage disequilibrium—reflect the historical reproductive success and population structure of species. As such, genomic data become a powerful, integrative record reflecting the composite effects of life history and environmental interactions.
By integrating this genomic information with census data from the Wind River Forest census of 2011, the researchers constructed a predictive model capable of accurately forecasting species abundance changes in 2016 and 2021. The model outperformed others by capturing the complexities of interspecific interactions, demographic stochasticity, and environmental variability, illustrating the immense potential of genomic demography. According to O’Dwyer, the balance encoded in the genomes acts as a hidden archive of ecological history, from which predictions about community dynamics emerge with unprecedented clarity.
James Lutz, who has annually surveyed the Wind River plot since 2010, emphasized the ecological significance of preserving species diversity, especially in Western U.S. forests where diversity is relatively lower. Losing species in such settings can cascade through the ecosystem, reducing productivity and threatening the survival of understory plants and wildlife dependent on a diverse, healthy forest. The model’s capacity to identify species at risk offers a critical tool for conservationists and forest managers striving to anticipate and mitigate the impacts of environmental stressors and biological invasions.
In addition to ecological forecasting, this research represents a methodological leap by demonstrating that genomic data—a resource underutilized in ecological studies—can serve as a surrogate or complement for extensive demographic data. This approach dramatically reduces the time and resources traditionally necessary for generating life-history based models, making predictive ecology more accessible and scalable. The team envisions broadening this framework to incorporate other forest sites where comprehensive longitudinal datasets are not available, possibly revolutionizing how biodiversity dynamics are studied worldwide.
Behind these achievements lies the fundamental evolutionary biology concept of effective population size, initially conceptualized nearly a century ago. It acknowledges that not all individuals contribute equally to future generations—a principle that underpins the genetic diversity and adaptive capacity of populations. By quantifying variation in reproductive output and its genomic signatures, researchers can infer how populations might respond to ongoing environmental changes, such as climate shifts or pathogen pressures.
Ultimately, the integration of genomic demography with census data charts a path toward predictive models capable of informing forestry policy, conservation strategies, and ecosystem management on unprecedented scales. James O’Dwyer and his team continue to refine their models, aiming to capture more nuanced environmental interactions and validate predictions across diverse habitats. The implications extend beyond forests, potentially influencing biodiversity monitoring and ecological forecasting across a range of natural systems.
This pioneering research was supported by the National Science Foundation and the Simons Foundation, highlighting the critical role of interdisciplinary collaboration and funding in advancing frontier ecological science. It is an exemplary case of how modern tools from genomics and data analytics, combined with decades of ecological observations, can unlock insights into the complex web of life sustaining our planet’s vital ecosystems.
For researchers, conservationists, and policymakers alike, this study redefines what is possible in anticipating the future of forest biodiversity. It underscores the remarkable power of genetic data not just to reveal a species’ past but to illuminate its potential futures, offering a new lens through which to safeguard ecological resilience in a rapidly changing world.
Subject of Research: Not applicable
Article Title: Genomic demography predicts community dynamics in a temperate montane forest
News Publication Date: 18-Sep-2025
Web References:
DOI link
Smithsonian ForestGEO
Nature 2023 study
PNAS tropical forest study
References:
O’Dwyer et al., Science (2025) DOI: 10.1126/science.adu6396
Previous works referenced within article
Image Credits: Photo by James A. Lutz
Keywords: Forest ecology, genomic demography, effective population size, biodiversity prediction, species abundance fluctuations, ForestGEO, Wind River Forest, ecological modeling, population genomics, conservation biology, ecosystem resilience
Tags: biodiversity forecasting modelscompetition among tree speciesecological monitoring techniquesenvironmental factors in forestsforest community dynamicsforest ecology predictionsinnovative research in plant biologyJames O’Dwyer research initiativeslong-term ecological data analysispopulation genomics in forestsSmithsonian Forest Global Earth Observatoryspecies abundance fluctuations
