About

Sarah J. Converse is an Associate Professor and Leader at the Washington Cooperative Fish and Wildlife Research Unit within the School of Environmental and Forest Sciences at the University of Washington. Her research areas include conservation science, statistics, spatial analysis, modeling, wildlife science, quantitative population ecology, demographic estimation and modeling, spatial demography, movement ecology, conservation biology, and decision analysis. She holds a B.S. in Fisheries and Wildlife from Michigan State University, an M.S. in Natural Resource Sciences from the University of Nebraska, and a Ph.D. in Wildlife Biology from Colorado State University. Her scholarly work involves applying quantitative methods to address ecological and conservation challenges, with a focus on population dynamics, species conservation strategies, and ecological decision-making.

Research topics

• Biology
• Computer Science
• Ecology
• Environmental science
• Geography
• Mathematics
• Statistics
• Environmental resource management

Selected publications

• Management strategy evaluation for salmon habitat restoration and hatchery supplementation

  Conservation Science and Practice · 2026-05-13

  articleOpen accessSenior author

  Abstract A primary challenge in conservation is the evaluation of management actions. For species with complex life histories, this evaluation is particularly challenging, and quantitative models are needed to understand how effects on particular life stages translate to objectives such as an acceptably low extinction risk or large abundance. We modeled the effects of increasing juvenile survival (as would result from habitat restoration) and hatchery supplementation on the achievement of recovery objectives for endangered Chinook salmon ( Oncorhynchus tshawytscha ) in the Wenatchee River Basin of Washington State. We evaluated these actions using simulations with a model that accounted for density‐dependence operating in juvenile rearing habitat. Simulations indicated that there were compounding benefits of increasing survival in both natal streams, where juveniles are born, and in downstream areas where a portion of juveniles rear. We also found that the marginal benefit of producing more hatchery fish declined as juvenile survival improved. Simulations suggested that achieving recovery targets for abundance of spring‐run Chinook salmon in the Wenatchee River would require more than a 150% increase in juvenile survival, barring increases in survival in other portions of the life cycle. Therefore, comprehensive management focusing on multiple factors affecting survival may be required to meet recovery criteria.

  PublisherDOI

• Columbia Basin Pygmy Rabbit Recovery Planning through Structured Decision Making

  bioRxiv (Cold Spring Harbor Laboratory) · 2026-04-14

  articleOpen accessSenior author

  EXECUTIVE SUMMARY The endangered Columbia Basin pygmy rabbit (CBPR) faces multiple threats, particularly increasing risk of larger and more intense wildfires due to climate change, emerging disease, and sagebrush habitat loss due to agriculture and development. Through the use of conservation breeding, the CBPR wild population grew from a low of 16 individuals captured in 2001 to over 100 individuals in two subpopulations in 2024. However, these subpopulations are geographically proximate, with potential risk that both subpopulations could be affected by a single wildfire or disease event. Additionally, a succession of setbacks in the breeding program has prompted a natural re-evaluation point for the CBPR conservation program. We undertook a structured decision-making (SDM) process with participants from both Washington Department of Fish and Wildlife (WDFW) and US Fish and Wildlife Service (USFWS) to develop a strategy that is sustainable and implementable for guiding management in the coming decades across the range of the CBPR, taking into account changing conditions and updated information. A population model that incorporated both demographic and high impact event uncertainties was developed to test how alternative strategies – defined by conservation breeding program, vaccination, and translocation components – affect CBPR population growth and cost objectives. Based on analysis of the model results, we identified the following actions that appear to have the greatest potential to allow WDFW and USFWS to meet their conservation objectives for CBPR: Continue conservation breeding program, and possibly expand to include an island subpopulation (an isolated, unfenced area that can serve as a source for rabbits while requiring fewer management inputs) Continue Rabbit Hemorrhagic Disease Virus (RHDV2) annual vaccinations in both breeding and wild populations When juveniles are available for translocation, prioritize recovery areas that are in the establishing phase In addition, while not analyzed explicitly in the model, discussions during the SDM process led to the identification of the following actions, which the group considered to have potential to benefit the CBPR either directly or indirectly: Increase the amount of suitable habitat available to pygmy rabbits Increase protections for existing and potential recovery areas Design future monitoring to better estimate survival and reproduction, with an emphasis on understanding how these vital rates vary between wild and semi-captive individuals, between vaccinated and unvaccinated individuals, and as a function of habitat factors

  PublisherDOI

• Cook Inlet beluga whale calling varies by group characteristics, behavior, and tidal state

  Behavioral Ecology and Sociobiology · 2026-05-01

  articleOpen access

  Abstract Communication allows social species to exchange information among group members. In aquatic environments, acoustic signals are among the most effective forms of communication and are important for many species, including cetaceans. Beluga whales ( Delphinapterus leucas ) are highly social and vocal, yet little is known about the functionality of their social calls. To examine context-dependent vocal behavior in belugas, we collected passive acoustic data and fine-scale behavioral observations for the endangered Cook Inlet beluga population. The resulting dataset includes 1,720 annotated vocalizations collected over 21 behaviorally encoded encounters. We fit generalized linear mixed models to these data to investigate the effect of behavioral state, group size, calf presence, and tidal state on (1) calling rate (number of calls/minute) and (2) call category (whistles, pulsed calls, combined calls). Belugas were more likely to call when traveling and had higher calling rates during flood tides. Group-level calling rate increased sublinearly with group size, suggesting that individuals called less in larger groups, possibly reflecting increased listening, vocal coordination, or a strategy to avoid acoustic masking. Group calling rate increased before transitions between traveling and milling, suggesting a possible link between communication and behavioral transitions. Whistles were more prevalent when traveling, while pulsed calls were more prevalent when milling. Combined calls occurred only when calves were present, indicating the importance of these calls in communication with calves. Identifying these communication patterns and the contexts in which they occur can enhance our understanding of beluga whale ecology and aid in conservation efforts via passive acoustic monitoring.

  PublisherOA PDFDOI

• Forecasting Dynamics of a Recolonizing Wolf Population Under Different Management Strategies

  Animal Conservation · 2025-05-26

  article

  ABSTRACT Species recovery can be influenced by a wide variety of factors, such that predicting the spatiotemporal dynamics of recovering species can be exceedingly difficult. These predictions, however, are valuable for decision makers tasked with managing species and determining their legal status. We applied a spatially explicit projection model to estimate population growth and viability of gray wolves ( Canis lupus ) from 2021 to 2070 in the state of Washington, USA, where wolves have been naturally recolonizing since the establishment of the first resident pack in 2008. Using this model, we predicted the effects of 12 scenarios relating to management actions (e.g., lethal removals by the state agency, translocation, recreational harvest) and system uncertainties (e.g., immigration from out of state, disease) on the probability of meeting Washington's wolf recovery goals, the probability of extinction, and other metrics related to population status. Population recovery was defined under Washington's Wolf Conservation and Management Plan as four breeding pairs in each of three recovery regions and three additional breeding pairs anywhere in the state. The baseline and two translocation scenarios indicated a high (> 90%) probability of wolf recovery in Washington by 2070, but scenarios related to harvest mortality (removal of 5% of the population every 6 months), increased lethal removals (removal of 8.53% of the population across the state each year), and cessation of immigration from out of state resulted in probabilities of < 0.20 (0.01, 0.04, and 0.17, respectively) of meeting recovery goals by 2070. Only two scenarios of 12 (increased harvest and lethal removals scenarios) resulted in a geometric mean of population growth ≤ 1, indicating long‐term population stability or growth for most scenarios. Our results suggest that wolves will continue to recolonize Washington and that recovery goals are likely to be met so long as harvest and lethal removals are not at unsustainable levels and adjacent populations support immigration into Washington.

  PublisherDOI
• RETRACTED

  RETRACTED: Group dynamics, behavior, and tidal state drive calling in Cook Inlet beluga whales

  Research Square · 2025-10-23

  preprint
  PublisherOA PDFDOI

• Development and evaluation of the remote passive integrated transponder tag reader for customizable monitoring of wildlife

  Wildlife Society Bulletin · 2025-02-06

  articleOpen accessSenior author

  Abstract Increasing the quantity and quality of mark‐reencounter data can be important when individual identification data are needed to address research or management questions. Physical recapture methods tend to be labor‐intensive and therefore expensive, while remote detection methods are not easily applied to all taxa or are proprietary and thus difficult to customize. However, partnerships between wildlife scientists and engineers have the potential to provide innovative solutions that improve data collection while reducing costs. We describe a collaborative effort to improve the collection of mark‐reencounter data on an invasive reptile, the brown treesnake ( Boiga irregularis ). In laboratory trials conducted on Guåhan (CHamoru; Guam in English) in 2021, we assessed the feasibility of detecting snakes fitted with passive integrated transponder (PIT) tags using customized remote PIT‐tag readers (RePTaR) paired with baited snake traps. We evaluated differences in scanning success as a function of snake traits, location of PIT tags within snakes, and the distance of the PIT tag from the reader. We successfully scanned all individuals, on average 529 times but ranging from 3 to 6,436 times, during 12‐hour trials using a designed customizable RePTaR unit. Snake characteristics and marking location explained little of the variability in scanning success, with distance from the reader as the best‐supported covariate. By building cross‐disciplinary partnerships, we can develop customizable technological solutions and provide additional tools with which to address challenges in wildlife research and management.

  PublisherOA PDFDOI

• Self‐sustaining populations are a conservation vision, not an operational objective

  Conservation Science and Practice · 2025-03-27 · 4 citations

  articleOpen accessSenior author

  Abstract It is common for species conservation plans to identify the establishment or maintenance of a “self‐sustaining population” as an objective. However, this statement vaguely conflates different formulations and interpretations of population viability, management costs, and cultural preferences for non‐invasive population management. Hidden value judgments and assumptions about these components can create disagreement and conflict among partners. Thus, although a simple statement about “achieving self‐sustaining populations” can be a powerful strategic vision, evoking important shared values, it will not be effective as an operational objective for conservation decision making. Best practices in decision making emphasize the importance of fundamental objectives that are clear, unambiguous, and operational. Conservation planners may be better served by replacing the self‐sustaining concept with better‐defined fundamental objectives using quantitative statements about viability and clearly laying out ecological, economic, and cultural values.

  PublisherOA PDFDOI

• Author response for "Longer rorqual whale mothers produce more female offspring"

  2025-06-02

  peer-reviewSenior author
  PublisherDOI

• Author response for "Longer rorqual whale mothers produce more female offspring"

  2025-07-24

  peer-reviewSenior author
  PublisherDOI

• Measuring polar bear health using allostatic load

  Conservation Physiology · 2025-01-01 · 3 citations

  articleOpen access

  ) has been adversely affected by climate change and loss of sea ice habitat. Even though the sub-population is likely decreasing, it remains difficult to link individual polar bear health and physiological change to sub-population effects. We developed an index of allostatic load, which represents potential physiological dysregulation. The allostatic load index included blood- and hair-based analytes measured in physically captured southern Beaufort bears in spring. We examined allostatic load in relation to bear body condition, age, terrestrial habitat use and, over time, for bear demographic groups. Overall, allostatic load had no relationship with body condition. However, allostatic load was higher in adult females without cubs that used terrestrial habitats the prior year, indicating potential physiological dysregulation with land use. Allostatic load declined with age in adult females without cubs. Sub-adult males demonstrated decreased allostatic load over time. Our study is one of the first attempts to develop a health scoring system for free-ranging polar bears, and our findings highlight the complexity of using allostatic load as an index of health in a wild species. Establishing links between individual bear health and population dynamics is important for advancing conservation efforts.

  PublisherOA PDFDOI

Frequent coauthors

• Jane E. Austin

  William Paterson University

  37 shared

• Nathan J. Hostetter

  North Carolina State University

  37 shared

• Thomas Mueller

  29 shared

• Michael C. Runge

  Eastern Ecological Science Center

  28 shared

• Claire S. Teitelbaum

  United States Geological Survey

  23 shared

• John B. French

  21 shared

• Mark H. Sorel

  Washington Department of Fish and Wildlife

  21 shared

• Amy A. Yackel Adams

  Fort Collins Science Center

  20 shared

Labs

• Washington Cooperative Fish and Wildlife Research UnitPI

Education

• PhD, Fishery and Wildlife Biology

  Colorado State University

  2005

• MS, School of Natural Resource Sciences

  University of Nebraska-Lincoln

  1999

• BS, Fisheries and Wildlife

  Michigan State University

  1996