The Ecology of Death
Death is a hallmark of life. It is a fundamental, inexorable phenomenon that drives ecological and evolutionary processes on every branch of the tree of life; it causes populations to rise and fall, alters ecosystem functions and nutrient cycling, and is the engine of natural selection. Yet, we know remarkably little about the causes and consequences of death because deaths are rarely directly observed in the wild. Current ecological approaches measure mortality only at population scales, missing the heterogeneity in mortality risk across individuals. However, this invisible variation in mortality risk is precisely what generates the population dynamics and evolutionary patterns that ecological and evolutionary theory seeks to explain. Consequently, we lack a unifying theory about the mechanisms leading to death and how variation in those deaths shapes ecological and evolutionary processes on the grandest scales. However, recent technological advances in the fields of animal tracking, remote sensing, and artificial intelligence have created a transformational opportunity to directly observe mortality events in the wild with remarkable detail. The Ecology of Death program leverages this revolutionary capability to study the causes and consequences of mortality across the tree of life to create and test novel ecological theory and reveal the drivers of population dynamics to inform conservation efforts.
Global Animal Mortality Project
A cross-taxa, international synthesis effort building a global dataset of animal tracking records paired with field-verified mortality events. Led in partnership with Ruth Oliver (UC Santa Barbara), the project investigates how human activities shape mortality. Data are contributed by collaborators across institutions and taxa - contact us if you are interested in participating!
Mortality-based biodiversity indicators
In collaboration with the Animalk Movement Biodiversity Observation Network (MoveBON), we are developing global indicators of wildlife mortality for use in conservation monitoring and policy. The goal is create standardized metrics that extend the utility of tracking data beyond movement science into biodiversity policy.
Drivers of Decline in North American Birds
North American bird populations have lost more than three billion individuals over the past half-century, but the demographic mechanisms driving those losses remain a mystery for most species. Population trajectories reflect the combined outcome of survival and reproduction — and identifying which vital rates are changing, where, and why is essential for targeted and effective conservation Action. Several projects in the lab address this directly, using longe-term datasets as well as continental-scale monitoring programs like the Monitoring Avian Productivity and Survivorship (MAPS) program.
Macro-demography: spatial patterns in vital rate change
Using long-term capture-mark-recapture data from MAPS stations distributed across North America, we are modeling geographic variation in apparent survival and reproductive productivity across landbirds. Preliminary results reveal substantial spatial heterogeneity in the demographic drivers of decline: the dominant mechanism varies across the range for many species, and single-mechanism explanations are insufficient. The goal is to produce spatially resolved demographic maps that can directly inform geographically targeted conservation planning.
Morphological change and survival
Birds across North America and the world have been getting smaller and changing shape over recent decades. In collaboration with the Weeks Lab at the University of Michigan, we use multiple approaches to ask what's causing these changes and what the consequence could be for populations and ecosystems.
Life history of migration
The factors that shape when, where, and how far animals migrate remain poorly understood. Our research asks how migratory behaviors evolve to track spatiotemporal variation in resources and what that means for life history tradeoffs across the annual cycle. out work has shown that seasonal resource dynamics constrain viable life histories across terrestrial taxa, and that migratory behavior reflects an evolved solution to navigating those constraints. A current focus is linking migratory behavior to fecundity and survival in North American landbirds using MAPS data, with the goal of identifying whether migratory strategy itself is a demographic mechanism underlying population declines in long-distance migrants.
Long-term field studies
We collaborate with the Linkhart Lab at Colorado College in a long-term demographic (nearly 5 decades) study of Flammulated Owls in central Colorado's ponderosa pine and mixed-conifer forests. These fire-dependent ecosystems are home to this small, insectivorous owl and partially migratory owl . The study has generated one of the longest continuous mark-recapture datasets for any North American owl, enabling analyses of survival, productivity, habitat association, and how this species' migratory behavior evolved to track seasonal resource dynamics in montane systems. This work also provides a direct empirical thread connecting early career field experience to current questions about mortality mechanisms, life history tradeoffs, and the demographic consequences of disturbance-maintained environments.
Quantitative approaches
Much of the lab's work depends on methods that don't yet exist off the shelf. We develop and apply Bayesian hierarchical models (primarily in NIMBLE) for survival and movement analysis, including novel approaches to estimating seasonal survival from automated telemetry data — a problem that has constrained the utility of the Motus Wildlife Telemetry System for demographic inference. We use simulation modeling to evaluate study designs prior to data collection and to test theoretical predictions under controlled conditions. More recently, we have been exploring the potential of emerging AI tools — particularly computer vision and acoustic detection pipelines — to scale mortality detection and attribution across large tracking datasets. Across all of this, the goal is rigorous inference: connecting individual-level observations to population-level conclusions in ways that are honest about uncertainty.
Death is a hallmark of life. It is a fundamental, inexorable phenomenon that drives ecological and evolutionary processes on every branch of the tree of life; it causes populations to rise and fall, alters ecosystem functions and nutrient cycling, and is the engine of natural selection. Yet, we know remarkably little about the causes and consequences of death because deaths are rarely directly observed in the wild. Current ecological approaches measure mortality only at population scales, missing the heterogeneity in mortality risk across individuals. However, this invisible variation in mortality risk is precisely what generates the population dynamics and evolutionary patterns that ecological and evolutionary theory seeks to explain. Consequently, we lack a unifying theory about the mechanisms leading to death and how variation in those deaths shapes ecological and evolutionary processes on the grandest scales. However, recent technological advances in the fields of animal tracking, remote sensing, and artificial intelligence have created a transformational opportunity to directly observe mortality events in the wild with remarkable detail. The Ecology of Death program leverages this revolutionary capability to study the causes and consequences of mortality across the tree of life to create and test novel ecological theory and reveal the drivers of population dynamics to inform conservation efforts.
Global Animal Mortality Project
A cross-taxa, international synthesis effort building a global dataset of animal tracking records paired with field-verified mortality events. Led in partnership with Ruth Oliver (UC Santa Barbara), the project investigates how human activities shape mortality. Data are contributed by collaborators across institutions and taxa - contact us if you are interested in participating!
Mortality-based biodiversity indicators
In collaboration with the Animalk Movement Biodiversity Observation Network (MoveBON), we are developing global indicators of wildlife mortality for use in conservation monitoring and policy. The goal is create standardized metrics that extend the utility of tracking data beyond movement science into biodiversity policy.
Drivers of Decline in North American Birds
North American bird populations have lost more than three billion individuals over the past half-century, but the demographic mechanisms driving those losses remain a mystery for most species. Population trajectories reflect the combined outcome of survival and reproduction — and identifying which vital rates are changing, where, and why is essential for targeted and effective conservation Action. Several projects in the lab address this directly, using longe-term datasets as well as continental-scale monitoring programs like the Monitoring Avian Productivity and Survivorship (MAPS) program.
Macro-demography: spatial patterns in vital rate change
Using long-term capture-mark-recapture data from MAPS stations distributed across North America, we are modeling geographic variation in apparent survival and reproductive productivity across landbirds. Preliminary results reveal substantial spatial heterogeneity in the demographic drivers of decline: the dominant mechanism varies across the range for many species, and single-mechanism explanations are insufficient. The goal is to produce spatially resolved demographic maps that can directly inform geographically targeted conservation planning.
Morphological change and survival
Birds across North America and the world have been getting smaller and changing shape over recent decades. In collaboration with the Weeks Lab at the University of Michigan, we use multiple approaches to ask what's causing these changes and what the consequence could be for populations and ecosystems.
Life history of migration
The factors that shape when, where, and how far animals migrate remain poorly understood. Our research asks how migratory behaviors evolve to track spatiotemporal variation in resources and what that means for life history tradeoffs across the annual cycle. out work has shown that seasonal resource dynamics constrain viable life histories across terrestrial taxa, and that migratory behavior reflects an evolved solution to navigating those constraints. A current focus is linking migratory behavior to fecundity and survival in North American landbirds using MAPS data, with the goal of identifying whether migratory strategy itself is a demographic mechanism underlying population declines in long-distance migrants.
Long-term field studies
We collaborate with the Linkhart Lab at Colorado College in a long-term demographic (nearly 5 decades) study of Flammulated Owls in central Colorado's ponderosa pine and mixed-conifer forests. These fire-dependent ecosystems are home to this small, insectivorous owl and partially migratory owl . The study has generated one of the longest continuous mark-recapture datasets for any North American owl, enabling analyses of survival, productivity, habitat association, and how this species' migratory behavior evolved to track seasonal resource dynamics in montane systems. This work also provides a direct empirical thread connecting early career field experience to current questions about mortality mechanisms, life history tradeoffs, and the demographic consequences of disturbance-maintained environments.
Quantitative approaches
Much of the lab's work depends on methods that don't yet exist off the shelf. We develop and apply Bayesian hierarchical models (primarily in NIMBLE) for survival and movement analysis, including novel approaches to estimating seasonal survival from automated telemetry data — a problem that has constrained the utility of the Motus Wildlife Telemetry System for demographic inference. We use simulation modeling to evaluate study designs prior to data collection and to test theoretical predictions under controlled conditions. More recently, we have been exploring the potential of emerging AI tools — particularly computer vision and acoustic detection pipelines — to scale mortality detection and attribution across large tracking datasets. Across all of this, the goal is rigorous inference: connecting individual-level observations to population-level conclusions in ways that are honest about uncertainty.