Safely Transporting Hazardous Liquids and Gases in a Changing U.S. Energy Landscape (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

This report reviews the response of the U.S. long-distance transportation sector to the sharp increase in the domestic production of crude oil, ethanol, and natural gas over the past decade. The demands faced by the nation’s inland pipeline, rail, and barge operators to move these new energy sup- plies economically and safely from new origins, across new transportation routes, and by new means of conveyance had been largely unexpected. To the credit of these operators, as well as the safety regulators who oversee them, the vast majority of these domestic supplies has been transported without incident, enabling the country to capitalize on its new energy re- sources and to reduce imported energy and the safety risks associated with its transportation. When this study commenced in late 2015, a national debate was under way about whether the domestic energy revolution was creating demands on the transportation system that would sacrifice safety. Railroad tank cars and waterborne tank barges were hauling oil and fuel ethanol in increasingly larger quantities and over longer distances, often on routes passing through communities that had little, if any, experience with regular and large quanti- ties of flammable liquids traffic. Some high-profile incidents had occurred, and there was skepticism about whether more pipelines would be built to provide a safer means of transportation, in part because of highly publicized opposition to new pipelines on grounds that their safety could not be assured. In retrospect, skepticism about the ability of the pipeline sector to respond to an increasing and changing geographic demand for the move- ment of oil and gas was unwarranted. The sector has attracted investment, extending the country’s pipeline capacity in a relatively short period for 5 Summary Observations and Recommendations 111

112 SAFELY TRANSPORTING HAZARDOUS LIQUIDS AND GASES such an infrastructure-intensive mode. Meanwhile, the domestic energy revolution has taken new turns in response to volatility in oil and natu- ral gas prices, which has led to marked reductions in the demand for the transportation of these energy liquids and gases by rail and by barge. Even the demand for transporting ethanol, which is met largely by railroads, has passed from a period of growth to one of stability as ethanol production has approached the levels targeted by federal policy. As pipeline capacity has grown and the fundamentals of the country’s energy markets have changed, trains and barges continue to transport crude oil and ethanol but in volumes that are down markedly since 2014. In this short time, the transportation sector has once again had to adapt to unforeseen conditions, demonstrating its robustness and flexibility in the face of the fast-changing domestic energy landscape. The transformation of the country’s energy markets is still playing out, and therefore future demands on the transportation system remain unclear. Trying to predict the safety implications of these demands is therefore tantamount to chasing a moving target, bound to result in a report whose content is more retrospective than prospective. Indeed, this report focuses largely on safety issues that were prominent during the period when the transportation sector was responding to the fast-increasing demand for the movement of energy liquids and gases and when government safety regulators were trying to catch up. Nevertheless, such a retrospective review can be important because of the insights it can provide for recognizing and responding in a faster and more effective manner to future changes in the demand for energy transportation and associated safety risks. In sponsoring this study, the Transportation Research Board (TRB) Execu tive Committee requested a review and comparison of how the country’s three major long-distance modes—pipelines, railroads, and barges—have fared in safely transporting the new traffic from the domestic energy revolution. Their respective safety performance over the past decade can be gauged to a limited degree by examining trends in unintended releases of oil, gas, and ethanol relative to the volume of these products transported. While data of this kind are provided in this report, they are poorly suited to making cross-modal comparisons of safety performance because of modal differences in incident-reporting criteria; the difficulty of representing rare, high-consequence events in statistical comparisons; and the potential for some of the new traffic to create second-order safety risks, such as the possibility of the new energy trains being involved in rail- highway grade-crossing accidents. Moreover, cross-modal safety compari- sons can have questionable purpose for decision making in cases when the modes are not viable alternatives. For instance, statistics suggesting that one mode is safer than another are immaterial if traffic volumes are too low to make investing in the seemingly safer mode economical.

SUMMARY OBSERVATIONS AND RECOMMENDATIONS 113 Although there is limited value in judging the relative safety perfor- mance of pipelines, railroads, and barges using available incident data, a number of insights can be gleaned from comparing the safety-assurance systems of the three modes. Pipelines and barges have accommodated major portions of the growth in domestic energy liquids and gases, and they have done so without creating major new safety concerns and within the basic framework of their longstanding regulatory and safety-assurance systems. Special attention is given in this report to the new regulations and other policies put in place to assure the safety of rail shipments of crude oil and ethanol, mainly because this traffic did not exist before the domestic energy revolution and has created several new safety concerns. The study committee is charged with making recommendations, as needed, on policies that can help reduce the likelihood of future incidents involving the transportation of these domestic energy supplies and to ensure an effective emergency response when incidents do occur. In responding to the domestic energy revolution, the nation’s rail, pipeline, and water transportation modes have demonstrated a high degree of innovation and adaptability, but with safety-assurance systems that have been challenged at times, particularly in the case of rail. The experience of the waterways, however, shows how safety can be assured when new traffic demands arise unexpectedly; in this case by a concerted strategy, put in place over the last three decades, to create a more robust and anticipatory safety assurance system. Based on information presented in this report, the committee offers its overarching observations on how the three modes, their safety assur- ance systems, and the emergency preparedness community have fared in responding to the challenges and demands of the domestic energy revolu- tion. In light of these observations, several recommendations are made to strengthen aspects of this response. OBSERVATIONS Expansion of Oil and Gas Pipelines Will Require Safety Vigilance Pipelines have always transported a large majority of the country’s domestic and imported crude oil and natural gas, as this mode, more than any other, is dedicated almost exclusively to the transport of hazardous liquids and gases. Although pipelines are not used to any significant degree to transport ethanol, pipeline operators have a high degree of familiarity with trans- porting oil and gas. Large investments have been made in new oil and gas pipeline capacity over the past decade. Oil transmission pipeline mileage grew by more than 40 percent between 2010 and 2016 despite volatility in oil prices and some high-visibility pipeline routing and construction permit- ting controversies.

114 SAFELY TRANSPORTING HAZARDOUS LIQUIDS AND GASES The committee did not find evidence that pipelines have encountered special safety challenges in accommodating the new demands of the grow- ing domestic energy supply. The pipeline incident rate has been generally stable for hazardous liquids and gas shipments, with year-to-year fluctua- tions in total release volumes affected mainly by the periodic occurrence of high-consequence incidents that are sufficiently rare to limit judgments about changes in their underlying risk. Nevertheless, substantially more pipeline mileage and higher traffic volumes can be expected to result in more pipeline releases over time, simply because of the increase in expo- sure. The safety impact, however, is likely to depend on the extent to which new pipeline technologies, leak-monitoring systems, and more vigilant and capable integrity management programs are effective in protecting both the newer pipelines and the older ones that connect to them. While no new safety problems have emerged as a result of the increased use of pipelines in transporting the larger volumes of domestic oil and gas, the potential for adverse safety effects from changes to the pipeline network will require careful monitoring. This need for monitoring was evident from 2008 through 2010 when Pipeline and Hazardous Materials Safety Administration (PHMSA) inspectors identified urgent pipeline construc- tion issues, which over time would have likely affected pipeline integrity. Time-dependent failure mechanisms such as corrosion and cracking, as well as outside-force damage, will undoubtedly lead to releases in some of the newer pipelines as time passes, requiring vigilance in leak monitoring, maintenance, and integrity management, as well as an effective emergency response capacity. Where the increase in domestic supplies of crude oil and natural gas are being transported in pipelines that have undergone flow reversals or been repurposed from carrying other commodities, an inven- tory of these lines may be warranted to monitor for potential problems associated with changes in their stress and operating profiles. Marine Transportation System Offers a Model for Robust Safety Assurance Tank barges have a long history of transporting crude oil and comparatively small volumes of ethanol and natural gas liquids (NGLs). Nevertheless, crude oil traffic moved by tank barge grew at a rapid pace after 2010 as hydraulic fracturing activity increased to boost demand for crude oil trans- portation on the inland and intracoastal waterways. Although the increase in barge movements of the domestic oil supply has not attracted much pub- lic or policy maker attention, the traffic volumes exceed those of rail, par- ticularly because of the large quantities transported along the Gulf Coast. The safety record of energy liquids movements by barge has been ex- emplary. There are no reports of ethanol or NGL releases from tank barges

SUMMARY OBSERVATIONS AND RECOMMENDATIONS 115 over the past 10 years and rare reports of crude oil releases. The highly sen- sitive nature of the maritime environment has long demanded vigilance in preventing, containing, and mitigating oil releases from tank barges. Some 30 years ago, however, a series of catastrophic, high-profile oil spills from tanker vessels and tank barges caused the federal government to revamp the safety-assurance system in ways that have fundamentally altered the industry’s safety profile. In addition to requiring the use of double-hulled tankers and tank barges for shipping oil and refined products, several new statutory and regulatory requirements created a safety culture that has proven to be robust, enabling the marine transport industry to safely accommodate unanticipated fluctuations in the demand for oil and other energy liquids transportation. These reforms included the clear designation of the vessel operator as the responsible party for spills, requirements for operators to make preparations for emergency response over the length of the transportation route, and requirements for operators to immediately notify the U.S. Coast Guard of any oil discharged into water regardless of volume and including any sheens observed that may have been caused by other parties. These reforms have been credited with not only fundamentally chang- ing the set of safety incentives that maritime carriers and shippers face but also with improving the industry’s safety culture. An aim was to create a safety-assurance system that would be anticipatory so that safety would be reasonably assured in the face of unforeseen changes in demand, technol- ogy, and operating practices. As discussed next, the railroads were in a much different position when the domestic energy revolution commenced. Unlike the waterways, the railroads had virtually no experience trans- porting large volumes of crude oil and ethanol. The industry and regula- tors were compelled to react to incidents and the new safety issues they presented. As these safety issues have become better understood, and the demand for rail transportation of crude oil has slowed, the challenge for the rail industry and regulators is to develop a safety assurance system that has a high degree of robustness like that of the maritime sector. Railroads Have an Opportunity to Create a More Robust Safety Assurance System Railroads had little experience carrying ethanol and crude oil in large quantities until after 2005, when this traffic increased sharply in response to public policies to promote fuel ethanol and to new supplies of domestic crude oil produced from hydraulic fracturing in areas lacking sufficient pipeline takeaway capacity. Railroads responded by transporting these flammable liquids in tank car unit trains, despite limited experience trans- porting hazardous materials in such trainload volumes, an absence of

116 SAFELY TRANSPORTING HAZARDOUS LIQUIDS AND GASES tank cars designed specifically to carry flammable liquids, and many oil and ethanol shippers who lacked experience preparing flammable liquids shipments for long-haul movement by rail. It is notable that unlike in the pipeline and marine transportation sectors, in which the carriers have pri- mary responsibility for all major factors affecting safety, railroads share this responsibility with shippers who own or lease, load, and secure the tank cars used to transport flammable liquids by rail. In response to derailments of trains carrying ethanol and crude oil, the initial focus of the industry and regulators was on reducing the severity of incidents by making the tank cars that carry these flammable liquids more crashworthy and resistant to thermal failure. The new tank car design specifications developed to improve crashworthiness and thermal resistance were informed largely by the historical performance of tank cars that de- railed in mixed-car trains, as opposed to unit trains. Until recently, there has simply not been enough experience with tank car unit trains to know how the dynamics of their derailments, which are more likely to involve multiple tank cars, compare with the dynamics of tank cars that have derailed in mixed-car trains. The limited (less than a decade) experience with flammable liquids unit trains suggests the high kinetic energy of the derailments has been a factor in the ignition of released product, more so than the specific volatility characteristics of the product being transported, such as its vapor pressure. However, there has been limited modeling of derailment kinetic energy that results in the conversion of thermal energy associated with multitank car derailments. As the tank cars compliant with the new design specifications are being phased in, tank cars built to the older specifications that are less crashworthy and less resistant to thermal failures may continue to be used for flammable liquids traffic for a few more years. Preventing the derail- ment of these cars is therefore an imperative. Post-incident investigations of severe flammable liquids train derailments indicate track wear and de- fects are common causal factors. New regulations that establish maximum failure rates per track mile are intended to be performance based so as to encourage innovation in the inspection and repair of defects. Questions remain, however, about the technical basis for the allowable failure rate, the prioritization that should be given to repairing specific types of defects, and whether these allowable rates and repair priorities should be adjusted for routes with significant crude oil and ethanol traffic. Likewise, questions remain about the technical basis for regulatory requirements that establish maximum speed limits for this traffic, which were established based on speed limits imposed on other types of hazardous materials that are seldom transported in unit trains.

SUMMARY OBSERVATIONS AND RECOMMENDATIONS 117 Emergency Response Preparedness Has Improved But with Geographic Variability The emergency response community has had longstanding familiarity with energy liquids being transported by pipeline and waterways. Because unit trains of crude oil and ethanol are a relatively new phenomenon, many of the communities traversed by this traffic had no familiarity with responding to incidents involving trainload shipments of these flammable liquids. Pre- paredness for incidents involving this traffic has improved as knowledge of crude oil and ethanol behavior in derailments has grown and as emergency response procedures, resources, and information have improved, aided by grants, research, training, and educational efforts of industry and govern- ment safety agencies. Transportation operators have strengthened their connections with state and local officials, developed new communications tools to aid emergency response, and increased their offerings of emergency response training. Despite these important developments of emergency response proce- dures and training programs, opportunities for improvement remain. In- dustry and government authorities face a continuing challenge in ensuring that these procedures are widely known and that training opportunities are exploited, especially among rural communities that are served by vol- unteer fire departments. Not only are clear guidelines lacking on the kinds of traffic data that carriers should be providing state and local agencies to prepare for energy liquids and gas transportation emergencies, but the recipient agencies differ so much from state to state that it unclear if this information is being transmitted to those who need it. RECOMMENDATIONS A decade after the start of the domestic energy revolution, it is not possible to say definitively that the impacts on the transportation sector have stabi- lized or temporarily reached a steady state. However, as the pace of change has slowed, now is an opportune time to take a more strategic look at the functioning and performance of each mode’s safety assurance systems with an eye to their future ability to respond effectively to unanticipated safety developments. The committee recommends that the Pipeline and Hazardous Materials and Safety Administration (PHMSA) undertake a comprehensive review of the successes and failures during the past decade in responding promptly and effectively to the transportation safety challenges presented by the domestic energy revolution for the purpose of informing the development of more anticipatory and robust safety assurance systems, including regulatory approaches. It is the committee’s view that such a retrospective assessment

118 SAFELY TRANSPORTING HAZARDOUS LIQUIDS AND GASES will provide insight into requirements and practices that have worked well in ensuring the safety of some modes and that may be candidates for ap- plication in others. By way of example, PHMSA could examine the oil spill dispersion models that currently inform the emergency response plans of tank barge operators to determine whether similar modeling may be appli- cable in cases where crude oil trains and pipelines traverse highly sensitive environmental areas. The general goal should be to develop a more forward-looking and anticipatory approach to safety assurance. To this end, the committee recommends that PHMSA periodically consult with industry on develop- ments impacting energy liquids and gas transportation and report annually on steps that are being taken to monitor and assess the risk implications of such developments. In 2005, for instance, it should have been evident that ethanol movements by train were about to grow substantially in response to the federal Renewable Fuel Standard. In keeping abreast of such develop- ments, PHMSA will be in a better position to collect the information and undertake the analyses that are needed to assess potential risks and to begin managing them in ways that will lessen the potential for harm to the public, environment, and emergency response community. PHMSA’s approach to risk monitoring and control should have a strong data and analytical basis. To this end, the committee recommends that PHMSA evaluate the utility of existing incident- and traffic-reporting data for the purpose of identifying and assessing public safety and environ- mental risks associated with transporting energy liquids and gases, deter- mine whether new and improved incident- and traffic-reporting systems are needed, and ensure that these data and risk metrics are being shared with state emergency preparedness agencies and used by industry for safety assur- ance purposes. A relevant example of where such metrics can be used is in monitoring and assessing the risks associated with pipelines that have been converted for different commodities or had their flow directions reversed. To develop such metrics, regulators would benefit from more uniform sets of data on incidents and traffic across modes. Not only are such data nec- essary to enable effective monitoring of the safety performance of the differ- ent modes when transporting energy liquids and gases, but also for allo cating public funds for emergency response preparation. Accordingly, the committee recommends that PHMSA consult shippers and carriers on the kinds of data that are available and needed to improve incident- and traffic-reporting sys- tems for the purpose of developing risk metrics—in that, indicators to assist in setting safety policies—and consult with state emergency preparedness agencies on opportunities for presenting and sharing these data and metrics with local communities and their emergency responders. An analytical approach to safety assurance will require the use of more quantitative tools for risk analysis by regulators and industry. The commit-

SUMMARY OBSERVATIONS AND RECOMMENDATIONS 119 tee recommends that PHMSA and the modal safety regulators encourage pipeline, barge, and rail carriers to make greater use of quantitative risk analysis tools to inform decisions about the routing of energy liquids and gases and about priorities for maintenance and integrity management of the equipment and infrastructure used. A candidate application for such tools is for assessing the risks associated with standards for maintenance of track traversed by unit trains of energy liquids and pipeline flow reversals and conversions. Another early candidate for the use of these quantitative tools is to assess and seek to strengthen the analytic basis for the multifactor train- routing criteria and train speed restrictions in the high-hazard flammable trains (HHFT) rule. Indeed, the committee recommends that PHMSA and the Federal Railroad Administration (FRA) regularly and systematically assess the risk-reducing effects of the HHFT rule, perhaps starting with a review of the crash and thermal performance of the new DOT-117 tank car designs. As these new tank cars enter the fleet in larger numbers, their safety record will become observable, allowing for assessments of how de- sign features that were informed by past tank car crashes are faring in the newer environment of flammable liquids unit train service. PHMSA and FRA upgraded the crashworthiness and thermal resistance of tank cars used in flammable liquids service out of a recognition that derailments will occur. Of course, the upgrades will not prevent all flam- mable liquids releases from tank car derailments, and the risk of release remains higher for crashes involving the older designs being phased out for flammable liquids service. Under these circumstances, the prevention of derailments remains a safety imperative. To obtain a more comprehensive understanding of the full array of factors that can give rise to and affect the severity of flammable liquids train crashes, the committee recommends that FRA and PHMSA seek to model these factors systematically, giving attention, for instance, to the propagation of internal rail defects and the kinetics that arise from multicar derailments. This deeper understanding of crash-causation factors will, among other things, inform railroad track inspection programs. Ensuring that these programs spot track defects that can lead to failures is essential to ensuring the safe operation of flammable liquids unit trains. To strengthen these pro- grams, the committee recommends that FRA enable and incentivize more frequent and comprehensive inspections of rail routes with regular energy liquids traffic, particularly by enabling railroads to exploit new inspection capabilities made possible by advances in sensor, high-resolution imaging, and autonomous systems technologies. Finally, it is necessary to presume that failures of energy liquids and gas pipelines, trains, and barges will occur, and that emergency responders will be called to the scene to contain and mitigate their consequences.

120 SAFELY TRANSPORTING HAZARDOUS LIQUIDS AND GASES Because the domestic energy revolution has led to the opening of many new energy transportation routes, some of these incidents will occur in communities with limited relevant emergency response experience. The committee recommends that PHMSA make a concerted effort to ensure that federal emergency preparedness grants are being used to meet the plan- ning, training, and resource needs of communities that are facing new and unfamiliar risks as a result of the changes that have occurred in the rout- ing and volume of energy liquids and gas shipments. As a starting point, PHMSA should review the extent to which emergency responders in these communities, especially in rural areas, are taking advantage of relevant government and industry response training opportunities, and then use this information to tailor programs that will enable and incentivize higher levels of participation. CONCLUDING COMMENTS The committee believes, as reflected in the actions recommended above, that industry and regulators should strive to make the safety assurance system for energy liquids and natural gas transportation more anticipatory, responsive, and risk informed. Based on its review of responses to the safety challenges arising from the domestic energy revolution, the committee is optimistic about the ability of industry and government to achieve such an outcome, especially through more collaboration. Working together, indus- try, regulators, and the emergency response community will be in a better position to reduce the occurrence and the severity of incidents involving transportation of energy liquids and natural gas. To do so, however, they will need to share information and develop more robust risk analytics, cre- ate and apply incentives to further the use of automation and other tech- nological innovations for monitoring the safe operation and the condition of equipment and infrastructure, and regularly review the effectiveness of safety regulations. The actions recommended in this report represent first steps in meeting these needs.