NETL uses machine learning to prevent and predict oil spills


ETL’s ORM suite comprises tools that can be used individually, or coupled together, to provide a full scope of data to evaluate, explore and ultimately prevent different spill scenarios.

Jared Ciferno, Kelly Rose, NETL

The National Energy Technology Laboratory (NETL) is the U.S. Department of Energy’s (DOE) national laboratory that, in partnership with DOE’s Office of Fossil Energy (DOE-FE), produces technological solutions for America’s fossil energy challenges. From developing creative innovations and efficient energy systems that make coal more competitive, to advancing technologies that enhance oil and natural gas extraction and transmission processes, NETL research is providing breakthroughs and discoveries. These efforts support domestic energy initiatives, stimulate a growing economy, and improve the health, safety, and security of all Americans.


At the center of much of the energy research carried out by NETL is DOE’s advanced computer system. According tothelatestrankingsbyTOP500fromJune2020,theJoule2.0supercomputerremainsamongthemostpowerful in the world, ranking 69thoverall.

Supercomputing is essential in achieving NETL’s support to DOE-FE’s program focus to discover, integrate and mature technology solutions that enhance the nation’s energy foundation and protect the environment for future generations. By expediting technology development through computational science and engineering, Joule 2.0 helps NETL cut costs, save time, and spur valuable economic investments with a global impact. A $16.5-million upgrade in 2019 boosted Joule’s computational power by nearly eight-fold, enabling researchers to tackle more challenging problems, as they work to make more efficient use of the nation’s vast fossil fuel resources.


Fig. 1. The Joule system’s computational power is 5.767 petaflops (PFLOPS), allowing it to perform more than 5 quadrillion calculations per second, equivalent to the efforts of 54,658 desktop computers, combined.

Fig. 1. The Joule system’s computational power is 5.767 petaflops (PFLOPS), allowing it to perform more than 5 quadrillion calculations per second, equivalent to the efforts of 54,658 desktop computers, combined.

Named for the familiar unit of energy, Joule allows researchers to model energy technologies, simulate challenging phenomena, and solve complex calculations, using computational tools that save time and money to ensure that technology development ultimately proves successful. The 2019 upgrade to Joule 2.0 boosted the system’s computational power to 5.767 petaflops (PFLOPS), meaning that it can perform more than 5 quadrillion calculations per second – equivalent to roughly 54,658 desktop computers combined, Fig. 1.


NETL also has created a Center for Data Analytics and Machine Learning (ML), which allows researchers to explore problems, using computer-based artificial intelligence (AI), ML, data mining, and advanced data-driven analytics. In addition to access to Joule, the center features a peta-scale machine (WATT), designed to house, transport and process up to 37 petabytes of data, using cutting-edge algorithms developed in-house and with external collaborators.

WATT links 104 graphics processing units (GPUs) with 16 petabytes of storage to provide unparalleled opportunities for the use of AI and ML, to enable scientific discovery and R&D acceleration. This facility enhances NETL’s ability to accumulate ML knowledge in analytical projects; enhance data handling functions, including curation, management and transformation of data; and combine physics-based modeling and AI/ML to address previously unanswerable problems or achieve fast, robust results.

DOE-fueled AI is already being applied broadly across the agency’s core missions. This research is strengthening national security and cybersecurity; improving grid resiliency and emergency response; and increasing energy efficiency and environmental sustainability. NETL’s recent AI- and ML-related achievements include rapidly screening advanced carbon capture materials to decrease the cost of the technology, and leading a multi-lab initiative called eXtremeMAT, which is using data analytics to develop and deploy new alloy materials that are affordable and perform reliably under the harsh environments encountered in high-temperature applications.


ML and AI are also used effectively in NETL’s execution of DOE-FE’s Offshore Research Program, which is pursuing research to enable the safe exploration and recovery of resources that are increasingly harder to locate and produce, especially in deeper water. Today, offshore facilities and infrastructure in federal waters are comprised of more than 5,300 mi of pipelines, more than 7,100 risers and platforms, and more than 120,000 subsea installations and wellbores. This infrastructure is exposed to some of the deepest, hottest, and most chemically challenging geologic conditions. In these complex environments, optimization of new infrastructure to leverage new materials while reducing cost, and addressing challenges associated with aging offshore infrastructure, requires technology innovations.

Current DOE offshore research has its roots in the Ultra-Deepwater and Unconventional Natural Gas and Other Petroleum Resources Research (UDW) Program launched in 2007 by EPAct 2005. The public/private partnership was designed to benefit consumers by developing environmentally friendly technologies, to increase America’s domestic oil and gas production while reducing the nation’s dependency on imports. Subsequent to the Macondo (Deepwater Horizon) incident, DOE-FE’s UDW Program became focused on safety and environmental sustainability. This partnership and synergistic research form the basis of the current Offshore Program.

NETL’s research is intended to investigate these aspects of offshore oil and gas resources:

  • Geologic Uncertainty—Prediction and early detection of subsurface conditions andgeohazards
  • Drilling & Completion Systems—Reducing risk and supporting informeddecision-making
  • Surface Facilities and Umbilicals—Improving systems and performance within extremeenvironments
  • Subsea Systems Reliability—Improving systems reliability through automation and advancedtechnology.

NETL maintains research partnerships with industrial, academic and research entities designed to catalyze development and demonstration of innovative technologies and methodologies for offshore oil and natural gas development.

As NETL strives to develop technological solutions to the nation’s energy challenges, the agency is exercising its expertise to lead the development of advanced data computing solutions for offshore oil and gas applications.

To address the unique challenges associated with offshore hydrocarbon exploration, researchers from NETL’s Geo- Analysis & Monitoring Team created the award-winning Offshore Risk Modeling (ORM) suite to evaluate and reduce the risk of oil spill events. ORM is a 2019 R&D World Magazine R&D100 award winner. Consisting of digital modeling and visualization tools using ML and AI, the ORM suite represents over six years of development, innovation and validation, resulting in a robust suite of advanced tools that are easily accessible for use by researchers and operators.

This suite of ML, big data, and advanced visualization tools is flexible, and can be paired with commercial and other open-source capabilities, offering stakeholders efficient, flexible, data-driven methods to assess and reduce risks associated with energy exploration and production offshore. The suite provides a comprehensive framework for future predictions, analyses and visualizations surrounding oil spill scenarios, to better inform offshore drilling efforts.

The ORM suite incorporates terabytes of data from Energy Data eXchange™ (EDX)—the DOE-FE’s virtual data library and laboratory—which allows for rapid large-scale predictions. EDX is an online collection of capabilities and resources that advance research and customize energy-related needs. It was created to support collaborative research efforts and technology transfer of research products. All the tools available for licensing can be accessed via EDX (

The data used by the ORM suite account for several factors that could impact the fate and transport of an oil spill, from information surrounding the water column and ocean currents to emergency response availability, oil particulate behavior, and more. The tools can simulate 4-D oil spill and blowout scenarios, identify critical subsurface characteristics, such as pressure and porosity during drilling activities, evaluate emergency response preparedness and assess the integrity of offshore infrastructure.

The ORM suite is unique, in that the constituent tools can be used either individually or coupled together to provide a full scope of data to evaluate, explore and ultimately prevent different spill scenarios. The suite’s analysis tools are listed in the following paragraphs.

Fig. 2. A hypothetical marine cross-section displaying BLOSOM’s capabilities, including the simulation of uncontrolled hydrocarbon release events, such as surface spills and subsurface blowouts, throughout the water column.

Fig. 2. A hypothetical marine cross-section displaying BLOSOM’s capabilities, including the simulation of uncontrolled hydrocarbon release events, such as surface spills and subsurface blowouts, throughout the water column.

BLOSOM™ (Blowout Spill Occurrence Model) is an open-source, comprehensive model that predicts how, and where, oil will travel following offshore blowout and spill events. BLOSOM offers a flexible suite of modeling modules designed to work together as a single system, to assess the multiple environmental uncertainties associated with deepwater and ultra-deepwater environments, blowouts and spills, Fig. 2. All components are designed to be explicitly three-dimensional and use equations best suited for high-pressure environments while maintaining the flexibility to operate with limited or highly uncertain data. BLOSOM incorporates the capabilities of severalmodules into one tool:

  • Jet/Plume Module simulates the initial oil and gas jet rising from the wellhead during an underwater blowout discharge. This module tracks the blowout’s physical properties,including the crude oil, gases, and water within a conceptual control volume, until it reaches a terminal level, at which point the plume is converted into individual oilparticles.
  • Conversion Module transfers elements from the Jet/Plume and Transport modules and amalgamates the two contrasting approaches in each, while converting control volumes ofmixed fluids into particles. This model also appropriately distributes oil droplet sizes to best capture subsurface plume formation, optionally simulating the direct application of dispersants at the source of the blowout (a practiced impact mitigationtechnique).
  • TransportModule simulatesthelong-termfateandtransportofthespill.Beachingandsinking events are also monitored. Surface spills can be simulated, using the TransportModule.
  • DispersantModuleoptionallypartitionstheactiveoilintotwoproportionsofoil—thatwhichhas been exposed to dispersants, and that which has yet to be affected. The methodologies used  to determine when, and where, dispersant is applied in the simulation will become more refined as development of BLOSOM progresses.
  • Weathering Module simulates oil weathering and degradation processes, including spreading, evaporation, emulsification, dissolution and dispersion. Other processes, such as biodegradation,photolysis,sedimentation,andsurfaceapplicationofdispersants,areplannedfor futureincorporation.
  • Crude Oil Module simulates changes to the oil’s physical and chemical properties in high-pressureenvironmentslikethedeepocean. Thismodulealsosimulateschangestotheoil,dueto degradation, using a pseudo-components approach. The components may be built with detailed crude information or interpolated from more readily obtainable crude assaydata.
  • Gas/Hydrates Module simulates gas properties, dissolution, and the formation and decomposition of hydrates for methane and ethane gases that may be present in an oil or gas wellblowout.
  • Hydrodynamic Handler handles ocean data for use in the other modules and can provide its own correlations and interpolations from the available data. It is designed to be flexible with multiple file formats and outputtypes.

CIAM (Climatological Isolation and Attraction Model) applies mathematical theories to  determine where oil and other particles in the ocean (e.g. debris, hazardous waste, plankton, etc.) are likely to be attracted orrepulsed.

CSIL™(CumulativeSpatialImpactLayers)isatool that integratesdatasetsrelatedtovarioussocial,economic and environmental information for a region, to rapidly assess potential impacts and inform environmental risk reductionefforts.

SWIM (Spatially Weighted Impact Model) is a decision support tool that evaluates relationships for various decision strategies and potential outcomes, and allows users to incorporate  weights as it ranks and compares scenarios, based on interactions among human and natural systems and othervariables.

STA™ (Subsurface Trend Analysis) which combines traditional petroleum geology methods with data science to improve prediction of subsurface properties that are critical to calculatinghazards.

VGM© (Variable Grid Method) communicates the uncertainty in data and modeled results. VGM is a novel approach to data visualization that employs geographic information system capabilities to simultaneously quantify and visualize spatial data trends and underlying data uncertainty in both two and three dimensions. The method provides a user-friendly, flexible and reliable tool toeffectively communicate spatial data, as well as the data’s inherent uncertainties, in a single, unified product. Applications that use big data, data analytics, and advanced computing run an inherent risk that results could be misleading or contain unseen error and uncertainty. The VGM is a simple, widely accepted tool for communicating that uncertainty or error with spatial data-driven products. VGM helps communicate the relationship between uncertainty and spatial data to effectively guide research, support advanced computation analyses, and inform management and policy decisions. The method is applicable to a wide range of end-users for reducing risks, improving decision-making, and reducing costs.


“The ORM suite introduces a new way of evaluating the offshore environment,” said Kelly Rose, Ph.D., with NETL’s Geologic & Environmental Systems team. “In the past, smaller-scale datasets were the primary focus in informing decisions. Using large-scale spatial and temporal data to inform local-scale needs is a groundbreaking approach that has the potential to greatly increase the safety of hydrocarbon exploration and ensure responsible stewardship of the environment.”

To date, ORM components have been used by numerous domestic and international stakeholders, including government agencies, academia and industry. NETL researchers use the ORM suite to supplement ongoing research by analyzing and predicting geologic properties, such as pressure, temperature, porosity and permeability in offshore regions, where little data is available. Other tools and models currently in use have yet to offer such a comprehensive and holistic view of offshore data and would require the use of several technologies to complete what the ORM suite can do at once.

NETL’s geo-data science experts developed novel data computing solutions available for licensing that address comprehensive earth, energy, environmental and engineering needs. These innovative tools offer accurate science-based data, analysis, and prediction for Earth-energy applications—including AI and data analytics capabilities—to industry, governments, researchers and other stakeholders requiring research-driven data for energy operations.

These tools, along with NETL’s efforts to study and characterize subsurface resources like the Gulf of Mexico, demonstrate NETL’s abilities to monitor, analyze and predict the physical, chemical and biological structures and functions of complex environments, from field-scale down to the molecular level. This meets theDOE’s Office of Fossil Energy’s mandate to produce technological solutions for America’s fossil energy challenges, including offshore risk mitigation.

Kelly Rose KELLY ROSE is a geology, geo-data science researcher with NETL’s Research Innovation Center. Her research focuses on using geologic and geospatial science to reduce uncertainty about, characterize and understand spatial relationships between energy and engineered-natural systems at a range of scales. Her work involves development of new data-driven methods and tools for analysis of offshore energy, oil & gas, rare earth elements, groundwater, carbon storage, and geothermal systems. She has served on various governmental and academic advisory committees and is associate editor for the Journal of Sustainable Energy Engineering, Rose has co-authored one patented, eight trademarked, and two copyrighted custom tools, and more than 100 published technologies and studies. She holds geology degrees from Denison University (BS), Virginia Tech (MS), and Oregon State University (Ph.D.).


Leave a Reply

This website uses cookies. By continuing to use this site, you accept our use of cookies.