Revising Species Status Assessments for endangered species decision making (hosted by TWS)

Symposium
ROOM: RSCC, D7
SESSION NUMBER: 8214
 
Decisions under the U.S. Endangered Species Act (ESA) require input from a scientific risk assessment that analyzes a compilation of best available information to predict the species future condition.  A series of critiques on the role of science in ESA decisions have called for improved consistency and transparency in species risk assessments.  To address that need and to propose a path forward, the Service developed a new framework for a species status assessment (SSA) with has three successive stages: 1) documentation of the species life history and ecological relationships to provide the foundation for the assessment, 2) description and hypothesized causes of the species current condition, and 3) forecasts of the species future condition to predict how the species will respond to scenarios of possible change in those hypothesized causes and conservation efforts while incorporating relevant sources of uncertainty.  The framework applies the conservation biology principles of representation, resiliency, and redundancy to evaluate the current and future condition of the species and emphasizes modeling for prediction and scenario planning as a way to contrast species response to plausible futures.  We present several case studies based on assessments of varied taxa in varied geographies with diverse ecologies and data quality to illustrate the applicability and flexibility of the new framework. Beyond these example applications the framework leads to scientific assessments of species status that will be helpful for all of the endangered species management decisions.

1:10PM Revising the Species Status Assessment for Decision Support in Endangered Species Management
  Bjorn Erickson
The U.S. Endangered Species Act (ESA) and the decisions made under it are predicated on a scientific understanding of the status of a particular species and the risk to that species over time. Assessing these factors accurately for diverse taxa presents many challenges, including significant life history and ecological differences between species, high levels of uncertainty, and the myriad factors that may influence extinction risk. In response to internal and external feedback about how we had been conducting these analyses, the U.S. Fish and Wildlife Service has recently developed and begun implementing a new framework for Species Status Assessments (SSAs), designed to improve consistency and transparency in our process, among other objectives. The framework applies the conservation principles of resiliency, redundancy, and representation to characterize the viability of a given species. These principles are incorporated into each of three successive stages of the analysis: 1) documentation of the species life history and ecological relationships to provide the foundation for the assessment, 2) description and hypothesized causes of the species current condition, and 3) forecasts of the species future condition to predict how the species will respond to scenarios of possible change in those hypothesized causes and conservation efforts while incorporating relevant sources of uncertainty. Beyond increasing transparency in the assessment process and consistency in analysis across taxa, the SSA framework promotes the development of a single analysis that can then be used for multiple ESA decisions affecting a species, further increasing consistency while allowing for greater efficiency in carrying out the ESA. Here we present the framework and introduce this symposium, which will highlight the applicability and flexibility of SSAs in making endangered species management decisions.
1:30PM Incorporating Genetic Considerations into Species Status Assessments
  David J. Berg
Species status assessments consider the life history and ecology of target species, and the causes of current conditions to predict the probability of persistence into the future. The process utilizes the 3 Rs of conservation biology: resiliency, redundancy, and representation. We consider how population genetic/genomic theory and practice, based primarily on the Hardy-Weinberg Principle, can inform understanding of the 3 Rs in the SSA process. Resiliency measures population health, with highly resilient populations exhibiting positive demographic factors, high genetic variation, and large population size. Within populations, genetic factors contributing to resiliency include inbreeding and assortative mating, and genetic drift. Genetic metrics for evaluating resiliency include descriptive measures (allelic richness and heterozygosity), inbreeding coefficients (FIS), genetically effective population size (Ne), and detection of bottlenecks. Management activities increasing resiliency include actions that decrease mortality and increase recruitment, and population augmentation from captive or wild sources. Redundancy is primarily concerned with maintaining multiple populations. As such, the focus is on among-population genetic variation and gene flow connecting populations at various spatial scales. Redundancy can be evaluated using measures of population divergence (FST, Jost’s D), STRUCTURE diagrams quantifying admixture among populations, and phylogeographic methods. Management activities that can increase redundancy include population re-establishment and assisted migration. Representation focuses on the responses of a species to changing environments using diversity as a proxy for adaptive potential, which itself is determined by genetic variation in fitness-related traits. Identification of such traits and measurement of selection coefficients for genes coding for these traits are the primary metrics for evaluating representation and these typically rely on newly emerging genomic techniques. Preservation of variation in loci under selection is the primary management activity for maintaining adaptive potential. We illustrate the application of population genetic/genomic theory and practice within SSAs using case-studies derived primarily from work with freshwater mollusks.
1:50PM Incorporating Genetic Diversity Information into a Representation Analysis for the Eastern Hellbender Species Status Assessment
  Trisha Crabill
Representation (or adaptive capacity) is the ability of a species to adapt to changing environmental conditions over time, and it can be characterized by the breadth of adaptive diversity within and among populations. Because adaptive diversity can be challenging to measure, the U.S. Fish and Wildlife Service often uses genetic diversity and ecological diversity as proxies for adaptive diversity when assessing a species’ adaptive capacity. In conducting the Species Status Assessment (SSA) for the Eastern hellbender (Cryptobranchus alleganiensis alleganiensis), we utilized results from recent genetic analyses which indicated that the subspecies consists of four deeply divergent lineages that largely align with major watersheds. The analyses also indicated that there is high genetic variation among lineages, low variation within lineages, and that restricted gene flow within these lineages is likely due to limited migration of hellbenders and other life history factors. The major sources of genetic diversity in this subspecies, therefore, are likely due to natural selection and genetic drift, with gene flow playing a minor role in driving evolutionary processes. Given this information, we characterized the Eastern hellbender’s representation, both currently and in the future, by assessing the distribution and health of populations within the four lineages.
2:10PM Candy Darter Status Assessment: A Case Study in Scientific Collaboration
  Keith Hastie
Abstract: Species status assessments (SSAs) inform decisions made under the US Endangered Species Act and therefore must be based on the “best available science.” In developing our SSA for the candy darter, a small freshwater fish endemic to the Upper Kanawha basin of Virginia and West Virginia, we worked collaboratively with species experts, aquatic ecologists, geneticists, and land managers from the state wildlife agencies, other federal agencies, and academia to help: 1) identify and collect the most current information on the species and its habitat; 2) identify and evaluate the stressors known or suspected to be affecting the species; and 3) ensure that our interpretation and application of the information was scientifically reasonable. In this presentation we discuss our process for engaging with these outside experts, from our initial in-person “kick-off” meeting to the ongoing communications and exchange of information as the SSA progressed.
2:30PM Monarch Butterfly Population Viability Modeling to Predict Future Viability in the Eastern and Western Populations
  Jennifer Szymanski
We describe the analytical approach to analyzing available data to assess current and future condition across the range of the Monarch butterfly using the U.S. Fish and Wildlife Service Species Status Assessment framework. North American populations of the monarch butterfly (Danaus plexippus) have shown declines over the past two decades and are currently being evaluated for listing under the Endangered Species. While extensive data and research exist on past and present trends of monarch abundance and population growth throughout North America, there is limited information on how such trends will continue into the future. In particular, we lack specific information on how populations respond to threats or conservation, the magnitude of these responses, and how these responses aggregate over multiple possible influences. We built stochastic geometric growth models to address these information gaps for the eastern and western North American monarch populations. Each model estimated the probability of quasi-extinction in 50 years using population-specific previously estimated trend data (growth rates and environmental variability) and expert elicited estimates of population responses to multiple influences (threats and conservation efforts). Both models incorporate population responses to changes in the amount of available breeding and overwintering habitat, insecticide use, and climate change. We explored how changes in the magnitude of different threats and conservation actions influenced population-level quasi-extinction risk using multiple, plausible future scenarios.
2:50PM Refreshment Break
3:20PM Using Expert Input to Inform Species Status Assessments Under the ESA
  Jennifer Szymanski
Often, we are data limited in assessing the status of species, yet the decision must be made within a statutory time-frame thereby excluding an option collecting additional data. Furthermore, some uncertainties are irreducible such as future habitat conditions. In these situations, expert knowledge is indispensable for making robust, science-based decisions. We frequently use expert input in conducting species status assessments to, among other reasons, better understand the ecology of species, identify and circumscribe areas of unique adaptive variation, and predict population response to changes in state conditions. The value of expert elicited information, however, depends upon the rigor of the process used to elicit the information. To illustrate the applicability of expert knowledge in species status assessments, we describe the expert elicitation process used for the monarch butterfly (Danaus plexippus plexippus) species status assessment (SSA). Specifically, we highlight how expert knowledge was obtained and used in the monarch SSA to identify and circumscribe areas of unique adaptive variation and to predict the species’ response to a range of changes in state conditions.
3:40PM Eastern Black Rail Dynamic Occupancy and Patch Persistence Modeling to Predict Future Redundancy and Resiliency
  Conor McGowan
Science to support wildlife management decision making often requires developing useful data analyses and predictive models to fill information gaps and provide useful results to decision makers. The US Fish and Wildlife Service has initiated a re-envisioned approach for providing decision makers with the best available science and synthesis of that information called the Species Status Assessment (SSA) for endangered species decision making. The SSA is a descriptive document that provides decision makers with an assessment of current status and predicts future status of the species, to support all manner of decisions under the US Endangered Species Act, such as listing, reclassification, recovery planning, etc. We describe the analytical approach to analyzing available data to assess current condition across the range in a dynamic occupancy analysis for the eastern black rail, a secretive marshbird. We used the results of the analysis to develop a site occupancy projection model where the model parameters (initial occupancy, site persistence, colonization) were derived from the data analysis and were linked to environmental covariates, such as land management, and land cover change (sea-level rise, development, etc.). We used the projection model to predict future conditions under multiple scenarios. Occupancy probability and site colonization were low in all regions and site persistence was also low, suggesting low redundancy and resiliency currently. Extinction probability was high for all regions in all simulated scenarios except one with significant effort to preserve existing habitat, suggesting low future resiliency and redundancy . With results of these data analyses and predictive modeling, the US Fish and Wildlife Service concluded that protections of the Endangered Species Act were warranted for this species.
4:00PM Incorporating Climate Science into Species Status Assessments
  Ryan Boyles
Species Status Assessments (SSAs) require an assessment of current and future needs or requirements of a target species. Climate can directly and indirectly impact these needs. Unfortunately, developing robust climate change analyses for use in SSAs can be daunting. The complexity of the topic, the myriad data sources, the large uncertainties (both physical and biological), and the fact that climate science is a diverse discipline unto itself to which most wildlife biologists may have little or no prior exposure, all speak to the magnitude of the challenge. For example, the output of most climate models used to simulate conditions decades into the future (i.e. climate model projections) are easily misunderstood as climate model forecasts (similar to numerical weather model forecasts), rather than representing computer simulations that depict possible climatic responses to different greenhouse gas emission scenarios. Moreover, while there are currently more than 30 state-of-the-art earth system models developed by the climate modeling community, traditional model evaluation and selection methods used by biologists to increase predictive power are not appropriate for use with climate model projections. Here we provide a general introduction to climate models, discuss their use in the context of situations of varying level of biological and climatological complexity, and offer some suggestions to help improve the efficiency and rigor with which biologists incorporate historical and projected climate changes into SSAs.
4:20PM Applying Global Climate Model Projections of Sea Ice to Assess Future Habitat Conditions for Pacific Walrus
  William Beatty
In 2017, the U.S. Fish and Wildlife Service completed the Pacific Walrus Species Status Assessment (PWSSA) to inform the Endangered Species Act listing determination for this species. The Pacific walrus is a migratory pinniped that feeds on benthic macroinvertebrates in shallow sub-Arctic and Arctic continental shelf waters. Pacific walruses winter in the Bering Sea, but adult females and juveniles migrate with retreating ice in the spring to the Chukchi Sea while adult males remain in the Bering Sea. In the summer, sea ice provides resting platforms near productive foraging areas for adult females and juveniles in the Chukchi Sea. In the late summer and early autumn when sea ice is no longer present over the continental shelf in the Chukchi Sea, walruses aggregate in large groups on coastal haulouts. Consequently, recent and projected reductions in summer Arctic sea ice represent loss of potential resting habitat and a possible reduction in available foraging habitat to adult female and juvenile Pacific walruses. As part of the PWSSA, a team of experts convened to estimate the future cumulative impacts of various stressors on the population with a Bayesian belief network (BBN). One BBN input variable quantified the number of observed and model-projected ice-free months over shelf waters to capture broad trends in future sea ice loss. To generate data for this input variable, we used monthly sea ice projections from several global atmosphere-ocean general circulation models (AOGCMs). We also developed spatial analyses to identify potential foraging areas on the continental shelf that may be accessible to Pacific walruses from sea ice or coastal haulouts in the future. Here, we will discuss best practices to consider when developing scientific products for SSAs that use AOGCM projections or other model outputs to estimate wildlife habitat conditions in the future.
4:40PM Multispecies Approaches to Status Assessments to Inform Endangered Species Determinations
  Daniel Fitzgerald
The detailed description of species’ ecology, distribution, and population structure required to accurately estimate current and future conditions for species status assessments are time consuming and often data limited. One option for improving efficiency and accuracy may be multispecies assessments, in which several species with similar ecology, distribution, and threats are assessed simultaneously. While multispecies status assessments in support of endangered species determinations are already occurring in the US, no formal guidance or best practices currently exist. We reviewed the U.S. Federal Register for previous multispecies assessments conducted between 1994-2018 to identify current trends and key research areas needed to support the endangered species listing process. Most status assessments grouped species based on overlapping distributions and habitat requirements, although few assessments explicitly state the reason for assessing species together. Interestingly, taxonomic relatedness was not a major driver of multispecies approaches, with numerous assessments grouping species at the levels of Domain, Kingdom, or Phylum. The most common approach encountered was a combined threats analysis, with few assessments presenting joint population or distribution models. We discuss the implications of these patterns, and suggest key ecological and policy factors that should be taken into account when developing multispecies status assessments.

 
Organizers: Conor McGowan, David Smith, Nathan Allan, Peter Erickson
 

Symposium
Location: Reno-Sparks CC Date: September 30, 2019 Time: 1:10 pm - 5:00 pm