Geology topics

Michael C. Runge

Publications and source records attributed to Michael C. Runge.

At least 73 records · Page 4Linked to original sources

USGS2021

A framework for allocating conservation resources among multiple threats and actions

Land managers decide how to allocate resources among multiple threats that can be addressed through multiple possible actions. Additionally, these actions vary in feasibility, effectiveness, and cost. We sought to provide a way to optimize resource allocation to address multiple threats when multiple management options are available, including mutually exclusive options. Formulating the decision as a combinatorial optimization problem, our framework takes as inputs the expected impact and cost of each threat for each action (including do nothing) and for each overall budget identifies the optimal action to take for each threat. We compared the optimal solution to an easy to calculate greedy algorithm approximation and a variety of plausible ranking schemes. We applied the framework to management of multiple introduced plant species in Australian alpine areas. We developed a model of invasion to predict the expected impact in 50 years for each species‐action combination that accounted for each species’ current invasion state (absent, localized, widespread); arrival probability; spread rate; impact, if present, of each species; and management effectiveness of each species‐action combination. We found that the recommended action for a threat changed with budget; there was no single optimal management action for each species; and considering more than one candidate action can substantially increase the management plan's overall efficiency. The approximate solution (solution ranked by marginal cost‐effectiveness) performed well when the budget matched the cost of the prioritized actions, indicating that this approach would be effective if the budget was set as part of the prioritization process. The ranking schemes varied in performance, and achieving a close to optimal solution was not guaranteed. Global sensitivity analysis revealed a threat's expected impact and, to a lesser extent, management effectiveness were the most influential parameters, emphasizing the need to focus research and monitoring efforts on their quantification.

Conservation Biology↗

USGS2021

Risks posed by SARS‐CoV‐2 to North American bats during winter fieldwork

The virus that causes COVID‐19 likely evolved in a mammalian host, possibly Old‐World bats, before adapting to humans, raising the question of whether reverse zoonotic transmission to bats is possible. Wildlife management agencies in North America are concerned that the activities they authorize could lead to transmission of SARS‐CoV‐2 to bats from humans. A rapid risk assessment conducted in April 2020 suggested that there was a small but significant possibility that SARS‐CoV‐2 could be transmitted from humans to bats during summer fieldwork, absent precautions. Subsequent challenge studies in a laboratory setting have shed new information on these risks, as has more detailed information on human epidemiology and transmission. This inquiry focuses on the risk to bats from winter fieldwork, specifically surveys of winter roosts and handling of bats to test for white‐nose syndrome or other research needs. We use an aerosol transmission model, with parameter estimates both from the literature and from formal expert judgment, to estimate the risk to three species of North American bats, as a function of several factors. We find that risks of transmission are lower than in the previous assessment and are notably affected by chamber volume and local prevalence of COVID‐19. Use of facemasks with high filtration efficiency or a negative COVID‐19 test before field surveys can reduce zoonotic risk by 65 to 88%.

Conservation Science and Practice↗

USGS2021

Causes of delayed outbreak responses and their impacts on epidemic spread

Livestock diseases have devastating consequences economically, socially and politically across the globe. In certain systems, pathogens remain viable after host death, which enables residual transmissions from infected carcasses. Rapid culling and carcass disposal are well-established strategies for stamping out an outbreak and limiting its impact; however, wait-times for these procedures, i.e. response delays, are typically farm-specific and time-varying due to logistical constraints. Failing to incorporate variable response delays in epidemiological models may understate outbreak projections and mislead management decisions. We revisited the 2001 foot-and-mouth epidemic in the United Kingdom and sought to understand how misrepresented response delays can influence model predictions. Survival analysis identified farm size and control demand as key factors that impeded timely culling and disposal activities on individual farms. Using these factors in the context of an existing policy to predict local variation in response times significantly affected predictions at the national scale. Models that assumed fixed, timely responses grossly underestimated epidemic severity and its long-term consequences. As a result, this study demonstrates how general inclusion of response dynamics and recognition of partial controllability of interventions can help inform management priorities during epidemics of livestock diseases.

Journal of the Royal Society Interface↗

USGS2021

Reconstructing population dynamics of a threatened marine mammal using multiple data sets

Models of marine mammal population dynamics have been used extensively to predict abundance. A less common application of these models is to reconstruct historical population dynamics, filling in gaps in observation data by integrating information from multiple sources. We developed an integrated population model for the Florida manatee ( Trichechus manatus latirostris ) to reconstruct its population dynamics in the southwest region of the state over the past 20 years. Our model improved precision of key parameter estimates and permitted inference on poorly known parameters. Population growth was slow (averaging 1.02; 95% credible interval 1.01–1.03) but not steady, and an unusual mortality event in 2013 led to an estimated net loss of 332 (217–466) manatees. Our analyses showed that precise estimates of abundance could be derived from estimates of vital rates and a few input estimates of abundance, which may mean costly surveys to estimate abundance don’t need to be conducted as frequently. Our study also shows that retrospective analyses can be useful to: (1) model the transient dynamics of age distribution; (2) assess and communicate the conservation status of wild populations; and (3) improve our understanding of environmental effects on population dynamics and thus enhance our ability to forecast.

Florida↗

USGS2021

Incorporating climate change in a harvest risk assessment for polar bears Ursus maritimus in Southern Hudson Bay

Arctic marine mammals are harvested by Indigenous people for subsistence and are socially and culturally important. For ice-dependent species like the polar bear Ursus maritimus , management and conservation require understanding interactions between harvest and sea-ice loss due to climate change. We developed a demographic model to evaluate harvest risk for polar bears in Southern Hudson Bay, Canada, where the annual ice-free season has increased by approximately one month in recent decades. The model was based on the theta-logistic equation and allowed for density-dependent changes (through carrying capacity [ K ]) and density-independent changes (through population growth rate [ r ]). Model parameters were estimated using a Bayesian Monte Carlo method that included capture-recapture, aerial survey, and harvest data. Harvest management followed a state-dependent approach under which new estimates of abundance were used to update the harvest level every five years. Under a middle-of-the-road environmental scenario that assumed K and r would decline in proportion to projected sea-ice declines, annual removal of 0.02–0.03 of females resulted in a 0.8 probability of maintaining subpopulation abundance above maximum net productivity level for three polar bear generations (~34 years), our primary criterion for sustainability. Under more pessimistic and optimistic environmental scenarios, comparable female harvest rates were 0.01 and 0.055, respectively. Our coupled modeling-management framework can be used to inform tradeoffs between protection and sustainable use for wildlife populations experiencing habitat loss.

Southern Hudson Bay↗

USGS2021

Navigating the science-policy interface

As a wildlife population ecologist who wants to conduct useful science, I find the Endangered Species Act (ESA), like other federal wildlife statutes, an intriguing read. The topic is in my wheelhouse—fish, wildlife, and plants, with a focus at the population and species levels. There is an emphasis on science, in fact, the “best scientific and commercial data available.” And there are intriguing questions: what threats does a species face? What habitat would be critical for its survival? Could any federal actions put the species or critical habitat in greater peril? I am not alone in this attraction. Hundreds of scientists continue to consider the types of scientific analysis suggested by the ESA. The enthusiasm is palpable. If the first cursory read of the ESA is intriguing to me as a scientist, the second close read is tantalizing—I realize that something very attractive is just out of reach. I know how to estimate the probability of extinction, but I do not know what “in danger of extinction” means. I know how to evaluate the incremental change in status that might arise from some level of proposed take, but I do not know what “is not likely to jeopardize the continued existence” of a species means. The standards expressed in the statute are not stated in purely scientific terms. Thus, ESA decisions cannot be based solely on science, and require additional policy interpretation. Clarity about these policy interpretations—even awareness that they are needed—can be hard to find, leaving a gap between what I can provide as a scientist and what an ESA decision maker needs. This awareness of the interaction between science and policy is also occurring in the larger field of conservation science, where there has been an increasing recognition of a research-implementation gap, the need for actionable science, and the promise of translational ecology. All of these terms emphasize that science alone does not result in action; instead, action arises out of decisions that are informed both by science and by values. At the interface of science and policy, a scientist can deliver relevant knowledge, and a decision maker can explain the policy context in which that science is needed. As a scientist wanting to conduct useful science, I crave this two-way conversation. But how can this conversation be structured in a meaningful and appropriate way? In this chapter, I explore how decision analysis can be used to navigate the science-policy interface for ESA decisions. Decision analysis is a large, well-established field that studies how decisions are made and how they could be made, with explicit attention given to clarifying and separating the values-based and science-based elements of a decision; identifying the impediments that make a decision difficult; and providing tools to overcome those impediments. There have been concerted efforts to apply formal decision analysis to ESA decisions, but the practice is not yet widespread across both the U.S. Fish and Wildlife Service and the National Marine Fisheries Service (the Services). The chapter begins with an introduction to decision analysis and how it seeks to bridge the science-policy interface. In subsequent sections, I explore how a decision analyst might frame listing and reclassification decisions, recovery planning, section 7 consultation, budget allocations, and a few other ESA decisions, with an emphasis on two questions: for each type of decision, what policy clarifications does the decision maker need to make; and knowing the policy context, what type of scientific assessment is needed. In the final discussion, I identify common themes among the types of decisions, and offer thoughts on how decision analysis could be more widely used to integrate science into ESA decisions.

Book chapter↗

USGS2020

Research priorities for migratory birds under climate change—A qualitative value of information assessment

The mission of the U.S. Geological Survey National Climate Adaptation Science Center is to provide actionable, management-relevant research on climate change effects on ecosystems and wildlife to U.S. Department of the Interior bureaus. Providing this kind of useful scientific information requires understanding how natural-resource managers make decisions and identifying research priorities that support those decision-making processes. Migratory bird management and conservation of migratory bird habitat are central components of the U.S. Department of the Interior’s mission. In particular, the U.S. Fish and Wildlife Service has an intensive, complex decision-making process for identifying high-priority parcels of land that will contribute to migratory bird conservation through permanent acquisition or easement. Climate change introduces several uncertainties into this decision-making process, and additional climate change research should help to support more informed decision making regarding habitat acquisition. Not all climate change related uncertainties, however, will have a meaningful effect on acquisition decisions; therefore, understanding which uncertainties have the most potential to alter decision making is crucial. This document summarizes a multiyear effort to clarify the major sources of climate change uncertainty that affect migratory bird management and to articulate related research priorities. We worked with U.S. Fish and Wildlife Service staff to assess the primary ways in which climate change is likely to affect migratory birds and their habitats; to clarify uncertainties surrounding these effects; and to assess how uncertainties may affect habitat acquisition decisions. Using a modified structured decision-making approach, we assessed a set of hypotheses about how climate change will affect migratory birds and their habitats. Then, we used a qualitative value of information assessment to rank the most important topics for future research. The ranking process was built on an assessment of three primary characteristics: the magnitude of uncertainty, the topic’s relevance to habitat acquisition decision making, and the feasibility of reducing the uncertainty. Based on the results of this process, high-priority topics for future research include the following: The effects of rising temperatures on spatial distributions of migratory birds during the breeding and nonbreeding seasons; Climate-driven changes to avian community composition through homogenization and loss of specialists; The effects of decreased precipitation on abundance in the breeding season; and The effects of rising temperatures on abundance in the nonbreeding season. In addition to describing high-priority research needs, this document provides a summary of the methodology used to identify, assess, and rank uncertainties. This method was developed for a climate change related topic where a full quantitative value of information approach may not be feasible. The results and methodology described here may be useful for U.S. Geological Survey and other science-funding agencies interested in improving the applicability of their research to natural-resource management decision making.

Circular↗

USGS2020

Assessing the risks posed by SARS-CoV-2 in and via North American bats — Decision framing and rapid risk assessment

The novel β-coronavirus, SARS-CoV-2, may pose a threat to North American bat populations if bats are exposed to the virus through interaction with humans, if the virus can subsequently infect bats and be transmitted among them, and if the virus causes morbidity or mortality in bats. Further, if SARS-CoV-2 became established in bat populations, it could possibly serve as a source for new infection in humans, domesticated animals, or other wild animals. Wildlife management agencies in the United States are concerned about these potential risks and have begun to issue guidance regarding work that brings humans into contact with bats, but decision making is difficult because of the high degree of uncertainty about many of the relevant processes that could lead to virus transmission and establishment. The risk assessment described in this report was undertaken to provide management agencies with an understanding of the likelihood that the various steps in the causal pathways would lead to SARS-CoV-2 infection of North American bats from people. This assessment focused on the active season for bats in the temperate zone of North America (April 15 through November 15), and used Myotis lucifugus (little brown bats) as a surrogate species. At the time of this work (April 2020), no empirical data about the effects of SARS-CoV-2 on North American bats were available, so a formal process of expert judgment was used to elicit estimates of the underlying parameters. Twelve experts in bat ecology, epidemiology, virology, and wildlife disease from the United States, United Kingdom, and Australia participated in the elicitation. A Monte Carlo simulation model was used to integrate the parameter estimates elicited from the experts and to predict the likelihood of exposure and infection in bats through a series of transmission pathways, with particular attention to capturing uncertainty in the predictions. Given the current state of knowledge as expressed by the expert panel, the results of this assessment indicate that there is a non-negligible risk of transmission of SARS-CoV-2 from humans to bats. For example, if a research scientist were shedding SARS-CoV-2 virus while handling bats under the field protocols used in North America prior to the COVID-19 pandemic, the risk model indicates that 50 percent (uncertainty, 15–84 percent) of those bats could be exposed to virus, and 17 percent (uncertainty, 3–51 percent) could become infected. Use of personal protective equipment, especially a respirator, is expected to reduce the exposure risk. The expert panel estimated that exposure risk from research scientists could be reduced 94–96 percent (uncertainty, 86–99 percent) through proper use of appropriate N95 respirators (a type of mechanical filter worn over the nose and mouth), dedicated clothing (such as Tyvek coveralls), and gloves. Should any North American bats become infected with SARS-CoV-2, the expert panel estimated that there is an approximately 33-percent chance the virus could spread within a bat population. This study, conducted by the U.S. Geological Survey in cooperation with the U.S. Fish and Wildlife Service, identified several critical uncertainties that could affect the estimate of risks associated with SARS-CoV-2 entering bat populations—notably, the underlying probability that a human would be shedding virus while working with bats, the likelihood of the virus replicating in bat tissue, and the likelihood of transmission of the virus within bat populations. Ongoing empirical work during May–October 2020 may shed light on these issues. Follow-up work is needed to better understand the probability of transmission of SARS-CoV-2 to bats from the general public; the manner in which the probabilities of exposure, infection, and transmission would differ during hibernation compared to the breeding season; and the likelihood of important effects, like morbidity and mortality in bats, the possibility of zoonosis from a North American bat reservoir, and effects of and on other wildlife.

North America↗

USGS2020

Using value of information to prioritize research needs for migratory bird management under climate change: A case study using federal land acquisition in the United States

In response to global habitat loss, many governmental and non‐governmental organizations have implemented land acquisition programs to protect critical habitats permanently for priority species. The ability of these protected areas to meet future management objectives may be compromised if the effects of climate change are not considered in acquisition decisions. Unfortunately, the effects of climate change on ecological systems are complex and plagued by uncertainty, making it difficult for organizations to prioritize research needs to improve decision‐making. Herein, we demonstrate the use of qualitative value of information analysis to identify and prioritize which sources of uncertainty should be reduced to improve land acquisition decisions to protect migratory birds in the face of climate change. The qualitative value of information analysis process involves four steps: ( i ) articulating alternative hypotheses; ( ii ) determining the magnitude of uncertainty regarding each hypothesis; ( iii ) evaluating the relevance of each hypothesis to acquisition decision‐making; and ( iv ) assessing the feasibility of reducing the uncertainty surrounding each hypothesis through research and monitoring. We demonstrate this approach using the objectives of 3 U.S. federal land acquisition programs that focus on migratory bird management. We used a comprehensive literature review, expert elicitation, and professional judgement to evaluate 11 hypotheses about the effect of climate change on migratory birds. Based on our results, we provide a list of priorities for future research and monitoring to reduce uncertainty and improve land acquisition decisions for the programs considered in our case study. Reducing uncertainty about how climate change will influence the spatial distribution of priority species and biotic homogenization were identified as the highest priorities for future research due to both the value of this information for improving land acquisition decisions and the feasibility of reducing uncertainty through research and monitoring. Research on how changes in precipitation patterns and winter severity will influence migratory bird abundance is also expected to benefit land acquisition decisions. By contrast, hypotheses about phenology and migration distance were identified as low priorities for research. By providing a rigorous and transparent approach to prioritizing research, we demonstrate that qualitative value of information is a valuable tool for prioritizing research and improving management decisions in other complex, high‐uncertainty cases where traditional quantitative value of information analysis is not possible. Given the inherent complexity of ecological systems under climate change, and the difficulty of identifying management‐relevant research priorities, we expect this approach to have wide applications within the field of natural resource management.

Biological Reviews↗

USGS2020

Anticipating future learning affects current control decisions: A comparison between passive and active adaptive management in an epidemiological setting

Infectious disease epidemics present a difficult task for policymakers, requiring the implementation of control strategies under significant time constraints and uncertainty. Mathematical models can be used to predict the outcome of control interventions, providing useful information to policymakers in the event of such an epidemic. However, these models suffer in the early stages of an outbreak from a lack of accurate, relevant information regarding the dynamics and spread of the disease and the efficacy of control. As such, recommendations provided by these models are often incorporated in an ad hoc fashion, as and when more reliable information becomes available. In this work, we show that such trial-and-error-type approaches to management, which do not formally take into account the resolution of uncertainty and how control actions affect this, can lead to sub-optimal management outcomes. We compare three approaches to managing a theoretical epidemic: a non-adaptive management (AM) approach that does not use real-time outbreak information to adapt control, a passive AM approach that incorporates real-time information if and when it becomes available, and an active AM approach that explicitly incorporates the future resolution of uncertainty through gathering real-time information into its initial recommendations. The structured framework of active AM encourages the specification of quantifiable objectives, models of system behaviour and possible control and monitoring actions, followed by an iterative learning and control phase that is able to employ complex control optimisations and resolve system uncertainty. The result is a management framework that is able to provide dynamic, long-term projections to help policymakers meet the objectives of management. We investigate in detail the effect of different methods of incorporating up-to-date outbreak information. We find that, even in a highly simplified system, the method of incorporating new data can lead to different results that may influence initial policy decisions, with an active AM approach to management providing better information that can lead to more desirable outcomes from an epidemic.

Journal of Theoretical Biology↗

USGS2020

Harnessing multiple models for outbreak management

The coronavirus disease 2019 (COVID-19) pandemic has triggered efforts by multiple modeling groups to forecast disease trajectory, assess interventions, and improve understanding of the pathogen. Such models can often differ substantially in their projections and recommendations, reflecting different policy assumptions and objectives, as well as scientific, logistical, and other uncertainty about biological and management processes (1). Disparate predictions during any outbreak can hinder intervention planning and response by policy-makers (2, 3), who may instead choose to rely on single trusted sources of advice, or on consensus where it appears. Thus, valuable insights and information from other models may be overlooked, limiting the opportunity for decision-makers to account for risk and uncertainty and resulting in more lives lost or resources used than necessary. We advocate a more systematic approach, by merging two well-established research fields. The first element involves formal expert elicitation methods applied to multiple models to deliberately generate, retain, and synthesize valuable individual model ideas and share important insights during group discussions, while minimizing various cognitive biases. The second element uses a decision-theoretic framework to capture and account for within- and between-model uncertainty as we evaluate actions in a timely manner to achieve management objectives.

Science↗

USGS2020

Decision analysis for managing public natural resources

Examples like the Grand Canyon story are becoming more common—agencies and stakeholders are increasingly using decision tools to help navigate complex decisions regarding natural resources. Several books and many papers have made the case for the use of decision analysis in natural resource management settings and have described the tools available (Conroy and Peterson 2013; Gregory et al. 2012; Williams et al. 2002), but only a few applications have been described. Thus, many agencies understand why decision analysis is valuable, but wonder how to implement it. The primary purpose of this book is to show what decision analysis looks like in practice for natural resource management, as a means to guide decision makers in adopting these practices to improve the achievement of their objectives. To this end, the collected case studies in this book, all real natural resource management decisions, were chosen with three objectives in mind: (1) to demonstrate how structuring decisions can render them more tractable; (2) to illustrate the diversity of decision analysis tools useful at various stages of the structuring process; and (3) to provide guidance on how decision analysis processes can be carried out in public environmental institutions.

Book chapter↗

USGS2020

Introduction to risk analysis

Many decisions are made in the face of uncertainty that either cannot or will not be reduced, and the challenge to the decision maker is how to manage the risk imposed by that uncertainty. This chapter will introduce the field of risk analysis, focusing on both the scientific tasks (estimating the probabilities and magnitudes of possible outcomes) and the policy-relevant value judgments needed (understanding the risk tolerances of the decision makers and stakeholders). The three case studies that follow demonstrate a range of approaches to risk management in a natural resource setting.

Book chapter↗

USGS2020

Allocating funds under the National Fish Habitat Action Plan

Each year, the Director of the U.S. Fish and Wildlife Service (Service), with advice from a Fisheries Management Team, allocates funding to support the National Fish Habitat Action Plan. The Service distributes the funds to Fish Habitat Partnerships (FHPs), who, in turn, undertake projects that “protect, restore, or enhance fish and aquatic habitats or otherwise directly support habitat-related priorities of Fish Habitat Partnerships.” Initially, this allocation was made based on a simple formula: larger FHPs received twice the allocation of smaller FHPs. But as the number of partnerships grew, and as funding grew at a slower rate, inequities developed among the FHPs. In 2012, the Service convened a structured decision making process to develop a more equitable, transparent, and strategic formula for annual funding allocation. The initial decision analysis, which focused on strategic aspects of the allocation, is described in this chapter. Deliberate consideration of decision analysis concepts brought about two advances: a focus on the fundamental long-term objective of maximizing the sustainability of aquatic species populations; and recognition that the benefits of the relatively small investment by the Service occur through leveraging contributions from management partners and increasing the efficiency of on-the-ground projects. Four allocation strategies were evaluated, using formal expert judgment methods, against an array of ecological and administrative objectives. The resulting consequence table was presented to Service managers to illustrate the considerations that underlie an allocation strategy. The insights of this initial decision analysis led to further internal discussions within the Service, and development of a fully articulated allocation method. In December 2013, the Director of the Service approved this new, competitive, performance-based method for allocating funds to FHPs, and it has been used since then to guide decision making. This case study illustrates the power of problem framing, the importance of articulating fundamental objectives, and the value of making transparent the hidden predictions at the heart of any decision.

Book chapter↗

USGS2020

Introduction to linked and dynamic decisions

Often, a decision maker is faced with a series of linked decisions, rather than an isolated one-off decision. In natural resource management, it is common to make a similar type of decision on a regular basis (e.g., annually). Such linked decisions have two important properties: they are dynamic, that is, the actions taken early on affect immediate outcomes as well as the effects of actions taken later; and they may be adaptive, that is, early actions might generate learning that can be applied to later actions. This situation has given rise to the development of methods for adaptive management. This chapter provides an overview of the decision analytical methods available to support framing and solving dynamic decisions, and briefly reviews the case studies that follow.

Book chapter↗

USGS2020

Framework for a long-term strategic plan for the Capital Area Groundwater Conservation Commission

The Capital Area Groundwater Conservation Commission oversees the use of groundwater in six parishes in Louisiana. In carrying out its statutory responsibilities and authorities, the Commission recognizes the complexity of its decisions: the long-term objectives it is seeking are multifaceted; the actions it can choose from are numerous and interdependent; and the understanding of the hydrogeological, economic, and social systems affected by its actions is limited. To navigate this complexity, the Commission is developing a long-term strategic plan to guide its activities and to serve as a primary mode of communication to stakeholders and the public. The long-term strategic plan is intended to consider actions and outcomes over at least the next 50 years within the 6 parishes in the Commission’s jurisdiction and related to all the confined aquifers in the 3000 feet below the district. The primary purposes of the plan are to promote long-term sustainability of groundwater extraction, continuity of operations of the Commission, long-term planning by water users, and clear communication with the public. The plan will describe specific management actions to be taken over time by the Commission, and the conditions under which those actions are to be taken. It will include intermediate milestones the Commission intends to achieve on the way toward achieving its long-term objectives. The actions under consideration include regulation and monitoring of groundwater withdrawal, mitigation of the environmental effects of withdrawal, support of relevant scientific studies, as well as work with partner agencies to implement measures to conserve, develop, and supplement groundwater resources. The plan will have greater detail about short-term actions than mid-term and long-term actions, and the Commission anticipates updating the plan periodically to adapt to changing circumstances and knowledge. To develop its long-term strategic plan, the Commission is working with The Water Institute of the Gulf and the U.S. Geological Survey using a facilitated process based on the principles of structured decision making (Gregory et al., 2012). This document outlines the framework for the strategic plan by describing the legal, economic, and scientific context for the plan, the fundamental objectives the Commission seeks to achieve in the long term, and the strategic alternatives it is considering.

Report↗

USGS2019

Concurrent assessment of epidemiological and operational uncertainties for optimal outbreak control: Ebola as a case study

Determining how to best manage an epidemiological outbreak may be hindered by both epidemiological uncertainty (i.e. about epidemiological processes) and operational uncertainty (i.e. about the effectiveness of candidate interventions). These two uncertainties are rarely addressed concurrently in epidemic studies, impeding decision-making. We present an approach to simultaneously address both sources of uncertainty. Epidemiological uncertainty is represented by a large ensemble of models of the 2014 West African Ebola outbreak. Operational uncertainty about the effectiveness of three classes of intervention is assessed for a wide range of potential effectiveness for each intervention. We ranked each intervention in terms of caseload reduction in each model, initially assuming an unlimited budget. To explore the role of budget limitation, we assessed the influence of three candidate cost functions relating intervention effectiveness and cost for different budget levels. The Value of Information (VoI) to resolve uncertainty is generally high in this study; appropriate information gain could reduce expected caseload by more than 50%. The ranking of interventions is jointly determined by the underlying epidemiological process, the effectiveness of the interventions and the size of the budget available for the program. An epidemiologically effective intervention might not be optimal if its costs outweigh its epidemiological benefit. Under higher budget conditions, resolution of epidemiological uncertainty is most valuable. When budgets are tight, however, operational and epidemiological uncertainty are equally important. Overall, our study demonstrates that significant public health improvements could result from a careful examination of both epidemiological and operational uncertainties within the same modelling structure. This approach can be applied to decision-making for management of other diseases for which multiple models and multiple interventions are available.

Proceedings of the Royal Society B↗

USGS2019

Context matters: Using reinforcement learning to develop human-readable, state-dependent outbreak response policies

The number of all possible epidemics of a given infectious disease that could occur on a given landscape is large for systems of real-world complexity. Furthermore, there is no guarantee that the control actions that are optimal, on average, over all possible epidemics are also best for each possible epidemic. Reinforcement learning (RL) has been used to develop machine-readable context-dependent solutions for complex problems with many possible realisations ranging from video-games to the game of Go. RL could be a valuable tool to generate context-dependent policies for outbreak response, though translating the resulting policies into simple rules that can be read and interpreted by human decision-makers remains a challenge. Here we illustrate the application of RL to the development of context-dependent outbreak response policies to minimise outbreaks of foot-and-mouth disease. We show that control based on the resulting context-dependent policies, which adapt interventions to the specific outbreak, result in smaller outbreaks than static policies. We further illustrate two approaches for translating the complex machine-readable policies into simple heuristics that can be evaluated by human decision-makers.

Philosophical Transactions of the Royal Society B:↗