Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Causality”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,369 records · Page 76

Conceptual design of an integrated technology model for carbon policy assessment.

This report describes the conceptual design of a technology choice model for understanding strategies to reduce carbon intensity in the electricity sector. The report considers the major modeling issues affecting technology policy assessment and defines an implementable model construct. Further, the report delineates the basis causal structure of such a model and attempts to establish the technical/algorithmic viability of pursuing model development along with the associated analyses.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Impact of post-event avoidance behavior on commercial facilities sector venues-literature review.

The terrorist attacks of September 11, 2001 (9/11), focused a great deal of interest and concern on how individual and social perceptions of risk change behavior and subsequently affect commercial sector venues. Argonne conducted a review of the literature to identify studies that quantify the direct and indirect economic consequences of avoidance behaviors that result from terrorist attacks. Despite a growing amount of literature addressing terrorism impacts, relatively little is known about the causal relationships between risk perception, human avoidance behaviors, and the economic effects on commercial venues. Nevertheless, the technical and academic literature does provide some evidence, both directly and by inference, of the level and duration of post-event avoidance behaviors on commercial venues. Key findings are summarized in this Executive Summary. Also included as an appendix is a more detailed summary table of literature findings reproduced from the full report.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Performance Analysis: Work Control Events Identified January - August 2010

This performance analysis evaluated 24 events that occurred at LLNL from January through August 2010. The analysis identified areas of potential work control process and/or implementation weaknesses and several common underlying causes. Human performance improvement and safety culture factors were part of the causal analysis of each event and were analyzed. The collective significance of all events in 2010, as measured by the occurrence reporting significance category and by the proportion of events that have been reported to the DOE ORPS under the ''management concerns'' reporting criteria, does not appear to have increased in 2010. The frequency of reporting in each of the significance categories has not changed in 2010 compared to the previous four years. There is no change indicating a trend in the significance category and there has been no increase in the proportion of occurrences reported in the higher significance category. Also, the frequency of events, 42 events reported through August 2010, is not greater than in previous years and is below the average of 63 occurrences per year at LLNL since 2006. Over the previous four years, an average of 43% of the LLNL's reported occurrences have been reported as either ''management concerns'' or ''near misses.'' In 2010, 29% of the occurrences have been reported as ''management concerns'' or ''near misses.'' This rate indicates that LLNL is now reporting fewer ''management concern'' and ''near miss'' occurrences compared to the previous four years. From 2008 to the present, LLNL senior management has undertaken a series of initiatives to strengthen the work planning and control system with the primary objective to improve worker safety. In 2008, the LLNL Deputy Director established the Work Control Integrated Project Team to develop the core requirements and graded elements of an institutional work planning and control system. By the end of that year this system was documented and implementation had begun. In 2009, training of the workforce began and as of the time of this report more than 50% of authorized Integration Work Sheets (IWS) use the activity-based planning process. In 2010, LSO independently reviewed the work planning and control process and confirmed to the Laboratory that the Integrated Safety Management (ISM) System was implemented. LLNL conducted a cross-directorate management self-assessment of work planning and control and is developing actions to respond to the issues identified. Ongoing efforts to strengthen the work planning and control process and to improve the quality of LLNL work packages are in progress: completion of remaining actions in response to the 2009 DOE Office of Health, Safety, and Security (HSS) evaluation of LLNL's ISM System; scheduling more than 14 work planning and control self-assessments in FY11; continuing to align subcontractor work control with the Institutional work planning and control system; and continuing to maintain the electronic IWS application. The 24 events included in this analysis were caused by errors in the first four of the five ISMS functions. The most frequent cause was errors in analyzing the hazards (Function 2). The second most frequent cause was errors occurring when defining the work (Function 1), followed by errors during the performance of work (Function 4). Interestingly, very few errors in developing controls (Function 3) resulted in events. This leads one to conclude that if improvements are made to defining the scope of work and analyzing the potential hazards, LLNL may reduce the frequency or severity of events. Analysis of the 24 events resulted in the identification of ten common causes. Some events had multiple causes, resulting in the mention of 39 causes being identified for the 24 events. The most frequent cause was workers, supervisors, or experts believing they understood the work and the hazards but their understanding was incomplete. The second most frequent cause was unclear, incomplete or confusing documents directing the work. Together, these two causes were mentioned 17 times and contributed to 13 of the events. All of the events with the cause of ''workers, supervisors, or experts believing they understood the work and the hazards but their understanding was incomplete'' had this error in the first two ISMS functions: define the work and analyze the hazard. This means that these causes result in the scope of work being ill-defined or the hazard(s) improperly analyzed. Incomplete implementation of these functional steps leads to the hazards not being controlled. The causes are then manifested in events when the work is conducted. The process to operate safely relies on accurately defining the scope of work. This review has identified a number of examples of latent organizational weakness in the execution of work control processes.

99 GENERAL AND MISCELLANEOUS↗

Modeling Abrupt Change in Global Sea Level Arising from Ocean - Ice-Sheet Interaction

It is proposed to develop, validate, and apply a coupled ocean ice-sheet model to simulate possible, abrupt future change in global sea level. This research is to be carried out collaboratively between an academic institute and a Department of Energy Laboratory (DOE), namely, the PI and a graduate student at New York University (NYU) and climate model researchers at the Los Alamos National Laboratory (LANL). The NYU contribution is mainly in the area of incorporating new physical processes into the model, while the LANL efforts are focused on improved numerics and overall model development. NYU and LANL will work together on applying the model to a variety of modeling scenarios of recent past and possible near-future abrupt change to the configuration of the periphery of the major ice sheets. The project's ultimate goal is to provide a robust, accurate prediction of future global sea level change, a feat that no fully-coupled climate model is currently capable of producing. This proposal seeks to advance that ultimate goal by developing, validating, and applying a regional model that can simulate the detailed processes involved in sea-level change due to ocean ice-sheet interaction. Directly modeling ocean ice-sheet processes in a fully-coupled global climate model is not a feasible activity at present given the near-complete absence of development of any such causal mechanism in these models to date.

54 ENVIRONMENTAL SCIENCES↗

Background Report on Eastern Democratic Republic of the Congo

Each month, approximately 45,000 people die from violence, hunger, disease, and other effects of displacement as a result of war in the Democratic Republic of the Congo (DRC). The country is often said to be plagued by a 'resource curse.' During each period in history since its discovery by the West, the DRC has possessed the resources the world craves and the world has sought these without regard for the consequences to the Congolese people. The catastrophic consequences of Congo's history of natural resource exploitation are the direct and indirect death of millions of Congolese people. The current war in Congo is multi-causal in nature but explanations are often reduced to describing it as an ethic conflict based on objective grievance. Objective grievance such as inequality, ethnic tensions, land disputes, and lack of democracy do exist, but they are neither necessary nor sufficient to explain the cause of the violent conflict, and more importantly, they fall short in explaining why this conflict has continued for years. The reality is the conflict is an economic war in which the trade of conflict minerals, gold and the 3Ts (tin, tantalum, tungsten), is directly linked to the financial sustainability of the groups fighting each other in eastern DRC. Objective grievance is a by-product of the conflict, used to create a false but plausible moral justification to continue violence. This paper examines the history of conflict in the DRC and the socio-economic variables contributing to the current war fought over conflict minerals.

99 GENERAL AND MISCELLANEOUS↗

Computational Models of Intergroup Competition and Warfare

This document reports on the research of Kenneth Letendre, the recipient of a Sandia Graduate Research Fellowship at the University of New Mexico. Warfare is an extreme form of intergroup competition in which individuals make extreme sacrifices for the benefit of their nation or other group to which they belong. Among animals, limited, non-lethal competition is the norm. It is not fully understood what factors lead to warfare. We studied the global variation in the frequency of civil conflict among countries of the world, and its positive association with variation in the intensity of infectious disease. We demonstrated that the burden of human infectious disease importantly predicts the frequency of civil conflict and tested a causal model for this association based on the parasite-stress theory of sociality. We also investigated the organization of social foraging by colonies of harvester ants in the genus Pogonomyrmex, using both field studies and computer models.

99 GENERAL AND MISCELLANEOUS↗

Integrating large-scale functional genomics data to dissect metabolic networks for hydrogen production

The goal of this project is to identify gene networks that are critical for efficient biohydrogen production by leveraging variation in gene content and gene expression in independently isolated Rhodopseudomonas palustris strains. Coexpression methods were applied to large data sets that we have collected to define probabilistic causal gene networks. To our knowledge this a first systems level approach that takes advantage of strain-to strain variability to computationally define networks critical for a particular bacterial phenotypic trait.

08 HYDROGEN↗

Use of limited data to construct Bayesian networks for probabilistic risk assessment.

Probabilistic Risk Assessment (PRA) is a fundamental part of safety/quality assurance for nuclear power and nuclear weapons. Traditional PRA very effectively models complex hardware system risks using binary probabilistic models. However, traditional PRA models are not flexible enough to accommodate non-binary soft-causal factors, such as digital instrumentation&control, passive components, aging, common cause failure, and human errors. Bayesian Networks offer the opportunity to incorporate these risks into the PRA framework. This report describes the results of an early career LDRD project titled %E2%80%9CUse of Limited Data to Construct Bayesian Networks for Probabilistic Risk Assessment%E2%80%9D. The goal of the work was to establish the capability to develop Bayesian Networks from sparse data, and to demonstrate this capability by producing a data-informed Bayesian Network for use in Human Reliability Analysis (HRA) as part of nuclear power plant Probabilistic Risk Assessment (PRA). This report summarizes the research goal and major products of the research.

Groth, Katrina M.↗

Simulating Collisions for Hydrokinetic Turbines

Evaluations of blade-strike on an axial-flow Marine Hydrokinetic turbine were conducted using a conventional methodology as well as an alternative modeling approach proposed in the present document. The proposed methodology integrates the following components into a Computational Fluid Dynamics (CFD) model: (i) advanced eddy-resolving flow simulations, (ii) ambient turbulence based on field data, (iii) moving turbine blades in highly transient flows, and (iv) Lagrangian particles to mimic the potential fish pathways. The sensitivity of blade-strike prob- ability to the following conditions was also evaluated: (i) to the turbulent environment, (ii) to fish size and (iii) to mean stream flow velocity. The proposed methodology provided fraction of collisions and offered the capability of analyzing the causal relationships between the flow environment and resulting strikes on rotating blades. Overall, the conventional methodology largely overestimates the probability of strike, and lacks the ability to produce potential fish and aquatic biota trajectories as they interact with the rotating turbine. By using a set of experimental correlations of exposure-response of living fish colliding on moving blades, the occurrence, frequency and intensity of the particle collisions was next used to calculate the survival rate of fish crossing the MHK turbine. This step indicated survival rates always greater than 98%. Although the proposed CFD framework is computationally more expensive, it provides the advantage of evaluating multiple mechanisms of stress and injury of hydrokinetic turbine devices on fish.

CFD↗

An Evidence-Based Evaluation of the Cumulative Effects of Tidal Freshwater and Estuarine Ecosystem Restoration on Endangered Juvenile Salmon in the Columbia River: Final Report

The listing of 13 salmon and steelhead stocks in the Columbia River basin (hereafter collectively referred to as “salmon”) under the Endangered Species Act of 1973, as amended, has stimulated tidal wetland restoration in the lower 235 kilometers of the Columbia River and estuary for juvenile salmon habitat functions. The purpose of the research reported herein was to evaluate the effect on listed salmon of the restoration effort currently being conducted under the auspices of the federal Columbia Estuary Ecosystem Restoration Program (CEERP). Linking changes in the quality and landscape pattern of tidal wetlands in the lower Columbia River and estuary (LCRE) to salmon recovery is a complex problem because of the characteristics of the ecosystem, the salmon, the restoration actions, and available sampling technologies. Therefore, we designed an evidence-based approach to develop, synthesize, and evaluate information to determine early-stage (~10 years) outcomes of the CEERP. We developed an ecosystem conceptual model and from that, a primary hypothesis that habitat restoration activities in the LCRE have a cumulative beneficial effect on juvenile salmon. There are two necessary conditions of the hypothesis: • habitat-based indicators of ecosystem controlling factors, processes, and structures show positive effects from restoration actions, and • fish-based indicators of ecosystem processes and functions show positive effects from restoration actions and habitats undergoing restoration. Our evidence-based approach to evaluate the primary hypothesis incorporated seven lines of evidence, most of which are drawn from the LCRE. The lines of evidence are spatial and temporal synergies, cumulative net ecosystem improvement, estuary-wide meta-analysis, offsite benefits to juvenile salmon, landscape condition evaluation, and evidence-based scoring of global literature. The general methods we used to develop information for the lines of evidence included field measurements, data analyses, modeling, meta-analysis, and reanalysis of previously collected data sets. We identified a set of 12 ancillary hypotheses regarding habitat and salmon response. Each ancillary hypothesis states that the response metric will trend toward conditions at relatively undisturbed reference sites. We synthesized the evidence for and against the two necessary conditions by using eleven causal criteria: strength, consistency, specificity, temporality, biological gradient, plausibility, coherence, experiment, analogy, complete exposure pathway, and predictive performance. Our final evaluation included cumulative effects assessment because restoration is occurring at multiple sites and the collective effect is important to salmon recovery. We concluded that all five lines of evidence from the LCRE indicated positive habitat-based and fish-based responses to the restoration performed under the CEERP, although tide gate replacements on small sloughs were an exception. Our analyses suggested that hydrologic reconnections restore access for fish to move into a site to find prey produced there. Reconnections also restore the potential for the flux of prey from the site to the main stem river, where our data show that they are consumed by salmon. We infer that LCRE ecosystem restoration supports increased juvenile salmon growth and enhanced fitness (condition), thereby potentially improving survival rates during the early ocean stage.

54 ENVIRONMENTAL SCIENCES↗

A Deeper Look at the Visualization of Scientific Discovery in the Federal Context (Final Report)

The visualization of scientific discovery has reached an intriguing point of development. Researchers in the field are producing fascinating representations that are catching the attention of program officers and policymakers. The accomplishments are being driven by the availability of both large scale data sets and the computing power and algorithms to analyze them. At the same time, skeptics are not sure what reality the visualizations represent and have concerns about the quality of the underlying data. Without an underlying statistical basis, it is hard to tell when a real change has occurred or when differences across teams, programs, or countries represent signal rather than noise. The almost exclusive reliance on publication, patent, and citation data is seen as too narrow and not addressing many of the crucial questions of causal relationships between federal policies and programs and outcomes for research, the economy, and society. Nonetheless, the participants in the Workshop on Visualization of Scientific Discovery agreed that this set of tools should be developed further, especially through interdisciplinary interaction among research users, experts in visual analytics, and science map producers. Interaction with users and development in relation to their needs are crucial in this process. The workshop discussion focused on the following important steps in turning the promise of today’s maps into the next generation of policy information.

99 GENERAL AND MISCELLANEOUS↗

Investigation of Microscopic Materials Limitations of Superconducting RF Cavities

The high-field performance of SRF cavities is often limited by breakdown events below the intrinsic limiting surface fields of Nb, and there is abundant evidence that these breakdown events are localized in space inside the cavity. Also, there is a lack of detailed understanding of the causal links between surface treatments and ultimate RF performance at low temperatures. An understanding of these links would provide a clear roadmap for improvement of SRF cavity performance, and establish a cause-and-effect ‘RF materials science’ of Nb. We propose two specific microscopic approaches to addressing these issues. First is a spatially-resolved local microwave-microscope probe that operates at SRF frequencies and temperatures to discover the microscopic origins of breakdown, and produce quantitative measurements of RF critical fields of coatings and films. Second, RF Laser Scanning Microscopy (LSM) has allowed visualization of RF current flow and sources of nonlinear RF response in superconducting devices with micro-meter spatial resolution. The LSM will be used in conjunction with surface preparation and characterization techniques to create definitive links between physical and chemical processing steps and ultimate cryogenic microwave performance. We propose to develop RF laser scanning microscopy of small-sample Nb pieces to establish surface-processing / RF performance relations through measurement of RF current distributions on micron-length scales and low temperatures.

36 MATERIALS SCIENCE↗

Advanced Small Modular Reactor (SMR) Probabilistic Risk Assessment (PRA) Demonstration

A key area of the Advanced Small Modular Reactor (SMR) Probabilistic Risk Assessment (PRA) strategy is the development of methodologies and tools that will be used to predict the safety, security, safeguards, performance, and deployment viability of SMRs. The goal of the SMR PRA activity will be to develop quantitative methods and tools and the associated analysis framework for assessing a variety of risks. Development and implementation of SMR-focused safety assessment methods may require new analytic methods or adaptation of traditional methods to the advanced design and operational features of SMRs. We will need to move beyond the current limitations such as static, logic-based models in order to provide more integrated, scenario-based models based upon predictive modeling which are tied to causal factors. The development of SMR-specific safety models for margin determination will provide a safety case that describes potential accidents, design options (including postulated controls), and supports licensing activities by providing a technical basis for the safety envelope. This report documents the progress that was made to implement the PRA framework, specifically by way of demonstration of an advanced 3D approach to representing, quantifying and understanding flooding risks to a nuclear power plant.

97 MATHEMATICS AND COMPUTING↗

Towards the Understanding of Induced Seismicity in Enhanced Geothermal Systems

This DOE funded project was a collaborative effort between Array Information Technology (AIT), the University of California at Berkeley (UCB), the Helmholtz Centre Potsdam - German Research Center for Geosciences (GFZ) and the Lawrence Berkeley National Laboratory (LBNL). It was also part of the European research project “GEISER”, an international collaboration with 11 European partners from six countries including universities, research centers and industry, with the goal to address and mitigate the problems associated with induced seismicity in Enhanced Geothermal Systems (EGS). The goal of the current project was to develop a combination of techniques, which evaluate the relationship between enhanced geothermal operations and the induced stress changes and associated earthquakes throughout the reservoir and the surrounding country rock. The project addressed the following questions: how enhanced geothermal activity changes the local and regional stress field; whether these activities can induce medium sized seismicity M > 3; (if so) how these events are correlated to geothermal activity in space and time; what is the largest possible event and strongest ground motion, and hence the potential hazard associated with these activities. The development of appropriate technology to thoroughly investigate and address these questions required a number of datasets to provide the different physical measurements distributed in space and time. Because such a dataset did not yet exist for an EGS system in the United State, we used current and past data from The Geysers geothermal field in northern California, which has been in operation since the 1960s. The research addressed the need to understand the causal mechanisms of induced seismicity, and demonstrated the advantage of imaging the physical properties and temporal changes of the reservoir. The work helped to model the relationship between injection and production and medium sized magnitude events that have jeopardized, and in some cases suspended, the generation of energy from EGS systems worldwide.

15 GEOTHERMAL ENERGY↗

Assessing the Macroeconomic Importance of Gasoline and Vehicle Spending

Vector error correction (VEC) was used to test the importance of a theoretical causal chain from transportation fuel cost to vehicle sales to macroeconomic activity. Real transportation fuel cost was broken into two cost components: real gasoline price (rpgas) and real personal consumption of gasoline and other goods (gas). Real personal consumption expenditure on vehicles (RMVE) represented vehicle sales. Real gross domestic product (rGDP) was used as the measure of macroeconomic activity. The VEC estimates used quarterly data from the third quarter of 1952 to the first quarter of 2014. Controlling for the financial causes of the recent Great Recession, real homeowners’ equity (equity) and real credit market instruments liability (real consumer debt, rcmdebt) were included. Results supported the primary hypothesis of the research, but also introduced evidence that another financial path through equity is important, and that use of the existing fleet of vehicles (not just sales of vehicles) is an important transport-related contributor to macroeconomic activity. Consumer debt reduction is estimated to be a powerful short-run force reducing vehicle sales. Findings are interpreted in the context of the recent Greene, Lee, and Hopson (2012) (hereafter GLH) estimation of the magnitude of three distinct macroeconomic damage effects that result from dependence on imported oil, the price of which is manipulated by the Organization of Petroleum Exporting Countries (OPEC). The three negative macroeconomic impacts are due to (1) dislocation (positive oil price shock), (2) high oil price levels, and (3) a high value of the quantity of oil imports times an oil price delta (cartel price less competitive price). The third of these is the wealth effect. The VEC model addresses the first two, but the software output from the model (impulse response plots) does not isolate them. Nearly all prior statistical tests in the literature have used vector autoregression (VAR) and autoregressive distributed lag models that considered effects of oil price changes, but did not account for effects of oil price levels. Gasoline prices were rarely examined. The tests conducted in this report evaluate gasoline instead of oil.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

The Price-Concentration Relationship in Early Residential Solar Third-Party Markets

The market for residential solar photovoltaic (PV) systems in the United States has experienced tremendous growth over the past decade, with installed capacity more than doubling between 2014 and 2016 alone (SEIA, 2016). As the residential market continues to grow, it prompts new questions about the nature of competition between solar installers and how this competition, or lack thereof, affects the prices consumers are paying. It is often assumed that more competition leads to lower prices, but this is not universally true. For example, some studies have shown that factors such as brand loyalty could lead to a negative relationship between concentration and price in imperfectly competitive markets (Borenstein, 1985; Holmes, 1989). As such, the relationship between prices and market concentration is an open empirical question since theory could predict either a positive or negative relationship. Determining a relationship between prices and market concentration is challenging for several reasons. Most significantly, prices and market structure are simultaneously determined by each other -- the amount of competition a seller faces influences the price they can command, and prices determine a seller's market share. Previous studies have examined recent PV pricing trends over time and between markets (Davidson et al., 2015a; Davidson and Margolis 2015b; Nemet et al., 2016; Gillingham et al., 2014; Barbose and Darghouth 2015). While these studies of solar PV pricing are able to determine correlations between prices and market factors, they have not satisfactorily proven causation. Thus, to the best of our knowledge, there is little work to date that focuses on identifying the causal relationship between market structure and the prices paid by consumers. We use a unique dataset on third-party owned contract terms for the residential solar PV market in the San Diego Gas and Electricity service territory to better understand this relationship. Surprisingly, we find that firms charged higher prices in more competitive markets in our sample. The finding is robust across multiple definitions of firm concentration. There are at least two potential explanations for our findings. First, firms could be conducting entry deterrence strategies. It is possible that firms are acting in a non-competitive way and setting prices lower than they would be otherwise. Setting low prices that are below potential competitors' marginal costs could deter entrants and ensure a larger market share. Second, there could be a group of dominant firms (with a competitive fringe), and the dominant firms may occasionally engage in price wars. If this is true, prices should be lower in more concentrated markets during the price wars (Salinger, 1990). As the rooftop PV market continues to grow, market structure will remain a relevant policy issue in consideration of the potential for rooftop solar to contribute to de-carbonization efforts or other policy objectives. This paper adds to a growing emphasis on understanding supply-side factors in scaling up solar markets in the residential sector. Generally, solar markets have become more competitive over the same time period that solar technology costs decreased. While solar system hard costs have come down, our research suggests that total costs are more nuanced in early solar system TPO markets. Policymakers should consider these findings when designing markets, and have the data needed to make informed decisions.

14 SOLAR ENERGY↗

Preliminary Assessment of Spatial Competition in the Market for E85: Presentation Supplement

Anecdotal evidence suggests retail E85 prices may track retail gasoline prices rather than wholesale costs. This indicates E85 prices may be higher than they would be if priced on a cost basis hence limiting adoption by some price-sensitive consumers. Using publicly available and proprietary E85 and regular gasoline price data, we examine pricing behavior in the market for E85. Specifically, we assess the extent to which local retail competition in E85 markets decreases E85 retail prices. Results of econometric analysis suggest that higher levels of retail competition (measured in terms of station density) are associated with lower E85 prices at the pump. While more precise causal estimates may be produced from more comprehensive data, this study is the first to our knowledge that estimates the spatial competition dimension of E85 pricing behavior by firms. This technical report elaborates on a related presentation.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Preliminary Assessment of Spatial Competition in the Market for E85

Anecdotal evidence suggests retail E85 prices may track retail gasoline prices rather than wholesale costs. This indicates E85 prices may be higher than they would be if priced on a cost basis hence limiting adoption by some price-sensitive consumers. Using publicly available and proprietary E83 and regular gasoline price data, we examine pricing behavior in the market for E85. Specifically, we assess the extent to which local retail competition in E85 markets decreases E85 retail prices. Results of econometric analysis suggest that higher levels of retail competition (measured in terms of station density) are associated with lower E85 prices at the pump. While more precise causal estimates may be produced from more comprehensive data, this study is the first to our knowledge that estimates the spatial competition dimension of E85 pricing behavior by firms. This is an initial presentation; a related technical report is also available.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗