IMPLICATIONS OF ENVIRONMENTAL UNCERTAINTY FOR CALIFORNIA NATURAL RESOURCE MANAGEMENT

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1016692
• Project Start Date: Oct 1, 2018
• Project End Date: Sep 30, 2023
• Program Code: [(N/A)]- (N/A)

Recipient Organization
UNIVERSITY OF CALIFORNIA, DAVIS
410 MRAK HALL
DAVIS,CA 95616-8671

Performing Department
Environmental Science and Policy

Non Technical Summary
This project supports the mission of the Agricultural Experiment Station by addressing the Hatch Act areas of: plant and animal production, protection, and health. California has set a high bar for improvements in environmental quality. As vast as the state's agricultural and natural resources are, there is also a strong incentive to provide for mitigation of existing and anticipated threats to those resources from concerns like over-exploitation, invasive species and increased environmental variability. This project is concerned with developing insights and tools for addressing these needs against a backdrop of economic uncertainty and environmental change. Making systems and policies robust to uncertainty and change typically involves short term sacrifice. This research will examine these tradeoffs, seeking to find the least costly approach to ensuring that management strategies are intelligently constructed for economic and environmental shocks. How do we better take into account not just the biomass stock in a fishery but also stochastic ecosystem dynamics (e.g. evolving carrying capacity) and economic system dynamics (e.g. evolving demand for harvest) to ensure sustainable resource use? In the context of endangered species, how do we identify optimal management strategies to ensure viability of vulnerable populations over time in the face of threats, such as predation from invasive species? How do we understand the changing dynamics of import-driven invasive species introductions given expanded risk from wider and deeper trade connections but diminished risk from the advent of prevention measures? Economic and econometric theory will be extended to address these questions. Decision tools will be evaluated using probabilistic modeling and Monte Carlo simulation. This research is expected to contribute in general to environmental economics and decision making under uncertainty as well as generate applied tools for mitigating environmental risk.

Animal Health Component

50%

Research Effort Categories

Basic

50%

Applied

50%

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
136	0860	3100	50%
605	0899	3010	50%

Knowledge Area
136 - Conservation of Biological Diversity; 605 - Natural Resource and Environmental Economics;

Subject Of Investigation
0899 - Wildlife and natural fisheries, general/other; 0860 - Endangered species;

Field Of Science
3010 - Economics; 3100 - Management;

Keywords

bioeconomic

uncertainty

risk

population viability

dynamic optimization

non-native species

maximum likelihood estimation

Goals / Objectives
California faces enormous challenges to successfully mitigate key natural and agricultural resource risks under environmental and economic uncertainty. Of course this is not a new problem--essentially all decisions made with a long term planning horizon involve some degree of uncertainty. Contributing to the current sense of urgency, however, is the understanding that increased economic activity and connectivity are enhancing environmental stressors. Balancing this stark assessment are advancements in our ability to conceptualize, measure and respond to changes in economic-environmental systems. This project will entail two case studies to explore and advance methods for resource management that are robust to uncertainty:1. Multidimensional Natural Assets (MNA). The first case study will focus on managing MNA under uncertainty. Resource management models usually require the marriage of social-environmental modeling and optimization techniques. A standard analysis of efficient fishery management, for example, might involve solving an integrated model of biology and user behavior using dynamic programming tools like value function iteration (VFI). To date, these tools have generated useful intuition by providing reliable solutions to relatively simple problems. However such standard approaches are not well positioned to address evolving future needs. The frontier of natural resource management is moving towards an ever richer representation of social-environmental systems, involving an expanding appetite for incorporating more states, increasingly evoking the curse of dimensionality. Furthermore, capacity to address uncertainty is central. Two particular tracks will be pursued involving management of (A) harvestable stocks (e.g. fisheries) in pursuit of maximizing economic returns and (B) endangered species population in pursuit of maintaining a viable population. The objectives will be to improve decision support tools to better account for the risks and uncertainty involved in managing California's natural resources. Specifically, the research will:a. Identify the management implications of accounting for multiple interacting stocks of natural assets (or liabilities/"bad stocks"); andb. Identify ideal management strategies for viability problems involving threshold avoidance (e.g. extinction). 2. Invasives. The second case study problem will focus on assessing the socioeconomic drivers and dynamics of invasive non-native species (NNS) establishments. Specifically the research will examine the link between plant material imports to California (and the U.S.) and the establishment of hitchhiking NNS in order to understand how introduction risks are evolving (over time and by source of imports) and what species ecosystem managers should be prepared to face going forward.

Project Methods
1. Multidimensional Natural Assets (MNA). For the MNA case study, we will develop optimization tools with the capacity to handle multiple dimensions without falling victim to the curse of dimensionality. Specifically, we will explore the potential for using an emerging tool from operations research and macroeconomics called "forward dynamic programming". This simulation-based method has been shown to have strong advantages for problems that are concerned with multiple state variables (stocks of natural resource assets [e.g. fish] and/or liabilities [e.g. pollution or NNS]). In order to address species population viability problems we will develop a penalty-based approach for handling the difficult constraint of avoiding extinction over a long time horizon. In this approach, a penalty imposed on the undesirable outcome (e.g. hitting a population extinction threshold) substitutes for a thorny, multi-period probabilistic constraint on the population level. This, in turn, allows for the application of otherwise standard dynamic programing techniques for solving the management problem.2. Invasives. For the invasives case study, we will develop and estimate a structural, maximum likelihood model of the import-driven NNS establishment process that accounts for trade volume, the delay from NNS establishment to discovery, differences in import product type and source region. This project will focus on imports of plant material to tightly link a particular population of established species (Hemipter, or "true bugs") with the specific dominant vector of introduction. The model will account for imports from several world regions as well as temporally varying rates of post-introduction discovery of non-native insects. This approach combines (1) a model of import-driven non-native species introduction that differentiates between several source regions with (2) a probabilistic model of post-introduction discovery. Because we don't observe introductions of NNS but rather discoveries, this model will inform the delay-to-discovery process with a unique ecological dataset of species identification effort (versus assuming that discovery effort is homogenous over many decades). Second, the model will differentiate between product categories (versus aggregating all imports together). Multiple datasets are already in hand: (1) non-native species discoveries in the USA, consisting of "first records" of all introduced phytophagous Hemiptera species discovered between 1850 and 2012; (2) import values of plant material spanning 159 years (1854 - 2012) disaggregated by 7 global regions and two product categories (nursery stock and seeds); and annual native Hemiptera species discoveries in the U.S. (1749-2012). Current efforts are focused on assembling a team of interdisciplinary collaborators and developing a detailed outline of the modeling approach.

Progress 10/01/19 to 09/30/20

Outputs
Target Audience:· National- and state-level agencies (United States: USDA, USGS; California: CA Department of Food and Agriculture; United Kingdom: Department for Environment, Food & Rural Affairs) · Stakeholders (participants in NSF funded center working groups) · Academic economists, ecologists and natural scientists Changes/Problems:Covid-19 travel restrictions led to cancellation of in-person stakeholder meetings (e.g. with US Forest Service) and working group meetings. Where possible activities have transitioned to online/virtual settings. What opportunities for training and professional development has the project provided?Training and professional development activities for the MNA component have included ongoing conference calls with a Ph.D. student and scientists at USGS at the Grand Canyon Monitoring and Research Center in Flagstaff. The PI is disseminating economic expertise (applied decision methods and analysis methods) while USGS personnel share expertise in complex species interactions in the Colorado River. For the Invasives component, a post-doc has being trained in the use of panel data analysis and hazard modeling for the large (multi-country, multi-species, multi-century) data set described above. Also see the summary of the "Adaptive Management" tutorial (workshop) above under "Other Products". How have the results been disseminated to communities of interest?Outreach activities for the MNA component: * Research findings were presented and discussed with researchers working on related problems (1) in person at the USGS Southwest Biological Science Center, and (2) virtually at the summer conference of the Association of Environmental and Resource Economics. Outreach activities for the Invasives component: * Research findings were presented and discussed (online) with researchers working on related problems at (1) a National Socio-Environmental Synthesis Center (SESYNC) working group meeting; and (2) The Society for Benefit Cost Analysis Annual Conference. What do you plan to do during the next reporting period to accomplish the goals?Research activities in each component involving a post-doc and current/former graduate students is ongoing. In addition to communication and discussion through standard academic meetings and outlets, planned activities include (1) new invasive species research focused on identifying leading indicators of spikes in invasive species contamination of international trade developed with access to expanded data sets on trade flows; and (2) further meetings with representatives from USGS to discuss model development related to management of non-native and native species interactions in the Colorado River.

Impacts
What was accomplished under these goals? The major goals of this project were to develop policy insights and tools for decision-making which support managing and protecting California's natural resources in a setting of uncertainty and environmental change. 1. Multidimensional Natural Assets (MNA). For the MNA component, in the context of species population viability problems (i.e. managing to avoid extinction) we developed a penalty-based approach (in the previous reporting period) for handling the difficult constraint of avoiding extinction over a long time horizon (which allows for the application of otherwise standard dynamic programing techniques for solving the management problem). This model was extended to the marine fisheries setting to get traction on the problem of balancing commercial catch with the bycatch of endangered/threatened species. A manuscript detailing this work has now been completed and is under review. 2. Invasives. For the invasives component, a large new dataset was (constructed in the previous reporting period) was used to assess the role of trade in the global spread of invasive insects. The dataset merges approximately 200 years of records spanning 200 countries, with measures of trade, invasive species discoveries, and species traits. The model's unique contribution is to merge two historical approaches to estimating invasive species risk. The first is a species-specific model that focuses on species traits (e.g. reproduction rate) as predictors of invasion risk. The second focuses on the invasion propagule pressure created by international trade, while ignoring species-specific attributes. The dataset has now been used to leverage the power of both approaches in one unified model, offering a new, integrated lens on relative drivers of invasion risk from previously separate literatures. A full manuscript has been drafted and is expected for submission in February 2021.

Publications

• Type: Journal Articles Status: Published Year Published: 2020 Citation: Rebecca Epanchin-Niell, Andrew Liebhold, Carol McAusland, Paul Mwebaze, Michael R. Springborn. Trade and Invasive Species: Tackling a Moving Target. Review of Economics and Environmental Policy, in press.

Progress 10/01/18 to 09/30/19

Outputs
Target Audience:· National- and state-level agencies (United States: USDA, USGS; California: CA Department of Food and Agriculture; United Kingdom: Department for Environment, Food & Rural Affairs) · Stakeholders (participants in NSF funded center working groups) · Academic economists, ecologists and natural scientists Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Training and professional development activities for the MNA component have included ongoing conference calls with a Ph.D. student and scientists at USGS at the Grand Canyon Monitoring and Research Center in Flagstaff, Arizona in July. The PI is disseminating economic expertise (applied decision methods and analysis methods) while USGS personnel share expertise in complex species interactions in the Colorado River. For the Invasives component, a post-doc has been hired and is being trained in the use of panel data analysis for the large (multi-country, multi-species, multi-century) data set described above. How have the results been disseminated to communities of interest?Outreach activities for the MNA component: * Research findings were presented and discussed with researchers working on related problems at the 2019 Association of Environmental and Resource Economics Summer Conference in Tahoe, CA. Outreach activities for the Invasives component: * Insights from completed research and new research needs were discussed with participants in working group meeting at National Socio-Environmental Synthesis Center (SESYNC) in Annapolis, MD. * Research findings were presented and discussed with researchers working on related problems at the 2019 Association of Environmental and Resource Economics Summer Conference in Tahoe, CA. What do you plan to do during the next reporting period to accomplish the goals?Research activities in each component involving a post-doc and current/former graduate students is ongoing. In addition to communication and discussion through standard academic meetings and outlets, planned activities include (1) meetings with the USDA in Maryland and USFS in Washington D.C. to present and garner feedback on research findings related to invasive species and (2) meetings with representatives from USGS to discuss model development related to management of non-native and native species interactions in the Colorado River.

Impacts
What was accomplished under these goals? The major goals of this project were to develop policy insights and tools for decision-making which support managing and protecting California's natural resources in a setting of uncertainty and environmental change. 1. Multidimensional Natural Assets (MNA). For the MNA component, an an emerging tool from operations research and macroeconomics called "approximate dynamic programming" was adapted for use in natural resource management (see Springborn and Faig (2019)). This simulation-based method has been shown to have strong advantages for problems that are concerned with multiple state variables (stocks of natural resource assets [e.g. fish] and/or liabilities [e.g. pollution or NNS]). The method was extended for use in a Pacific salmon management problem to capture management impacts on genetic diversity in decision making (mansucript under review). In the context of species population viability problems (i.e. managing to avoid extinction) we developed a penalty-based approach for handling the difficult constraint of avoiding extinction over a long time horizon (which allows for the application of otherwise standard dynamic programing techniques for solving the management problem)--see Donovan et al (2019) above. This model was extended to the marine fisheries setting to get traction on the problem of balancing commercial catch with the bycatch of endangered/threatened species. This model has been solved and a manuscript drafted. 2. Invasives. For the Invasives component, a large new dataset was constructed to assess the role of trade in the global spread of invasive insects. The dataset merges approximately 200 years of records spanning 200 countries, with measures of trade, invasive species discoveries, and species traits. The model's unique contribution is to merge two historical approaches to estimating invasive species risk. The first is a species-specific model that focuses on species traits (e.g. reproduction rate) as predictors of invasion risk. The second focuses on the invasion propagule pressure created by international trade, while ignoring species-specific attributes. The dataset will allow for leveraging the power of both approaches in one unified model, offering a new, integrated lens on relative drivers of invasion risk from previously separate literatures.

Publications

• Type: Journal Articles Status: Published Year Published: 2019 Citation: Michael R. Springborn and Amanda Faig. Moving forward: a simulation-based approach for solving dynamic resource management problems. Marine Resource Economics, 34(3), 199-224, 2019.
• Type: Journal Articles Status: Published Year Published: 2019 Citation: Julie L. Lockwood, Dustin J. Welbourne, Christina Romagosa, Phillip Cassey, Nicholas E. Mandrak, Angela Strecker, Brian Leung, Oliver C. Stringham, Bradley Udell, Diane J. Episcopio Sturgeon, Michael F. Tlusty, James Sinclair, Michael R. Springborn, Elizabeth F. Pienaar, Andrew Rhyne and Rueben Keller. When Pets Become Pests: The Role of the Exotic Pet Trade in Producing Invasive Vertebrate Animals. Frontiers in Ecology and the Environment, 2019.
• Type: Other Status: Published Year Published: 2019 Citation: Donovan, Pierce, and Michael R. Springborn. Maintaining the Long-Term Viability of the Humpback Chub in the Grand Canyon."�ARE Update Vol. 22, No. 5, May/June, 2019. URL: https://s.giannini.ucop.edu/uploads/giannini_public/a7/cc/a7cc1329-f61a-44e2-ac1c-751a6bbf921a/v22n5.pdf.