Commentary - Improving and Advancing Predictability

Improving and Advancing Predictability

Much of the research conducted at Lawrence Livermore National Laboratory involves efforts to better predict outcomes to some of the most complex scientific challenges. The articles in this issue of Science & Technology Review describe widely disparate projects and programs that are improving our predictive abilities through testing, experimentation, and updating techniques and facilities.

The feature article highlights Livermore’s research in the “Microbes Persist” Soil Microbiome Scientific Focus Area. Soils are a critical natural resource and a global reservoir of organic carbon. The characteristics of soil persistence (residence time) and cycling (turnover) impact soil’s fertility, structure, and water-holding capacity and are largely shaped by information encoded in the genomes of soil microbes. To dig deeper into this topic, researchers at Livermore are using a variety of innovative measurement techniques and prediction tools.

Livermore’s microbial genomics and the terrestrial carbon cycle research includes ‘omics data (DNA, RNA, protein, and metabolite characterization), isotopic labeling, and submicrometer-scale isotopic imaging and chemical analyses. With these techniques, Livermore researchers can quantitatively determine in situ rates of microbial growth, death, and consumption for individual microbial populations under different conditions. This work has resulted in unprecedented measurements of the microbial traits that help to retain soil organic carbon, one that would not have been possible without the Laboratory’s state-of-the-art isotope measurement and micro-analytical facilities. The team has also used carbon dating of specific molecules to show that amino acids, lipids, and acid-insoluble organic matter persist for different lengths of time and that these pools change with soil depth. This information is critical to determining the durability of carbon in soils and improving biogeochemical models that can be integrated into Earth system models.

The issue’s first highlight features Livermore’s research to discover next-generation alloys and to advance processes to find future materials suitable for national security needs. Livermore researchers developed the Materials Acceleration Program, a design framework that identifies optimal alloy compositions and the related parameters to additively manufacture the materials. This work, which recently was awarded the Journal of Alloys and Compounds’ 2024 best paper award, is enabling Livermore’s materials development researchers to better identify and predict material compositions that can endure extreme thermal stress before deforming, a critical insight for the Laboratory’s hypersonic, nuclear stockpile, and fusion energy applications.

The second highlight details the use of AI, machine learning, and other tools to ensure the Scorpius accelerator’s predictability and reliability. Intended to improve understanding of nuclear weapons’ functions using plutonium, Scorpius consists of thousands of components and represents a collaborative effort between Lawrence Livermore, Los Alamos, and Sandia national laboratories, as well as the Nevada National Security Site. Livermore’s Scorpius team is responsible for developing and producing line replaceable units (LRUs), components that generate optimal pulses for high-resolution radiography. Livermore has developed a set of functional tests using AI and machine learning to evaluate the LRUs, LRU subsystems, and the equipment that controls each cluster to ensure that the equipment operates as expected and under conditions similar to those in the accelerator.

The final highlight introduces planned actions to maintain and prolong the life of the National Ignition Facility (NIF), ensuring better predictability for how the facility will operate over the course of many years. After continued regular activity, even with scheduled maintenance and refurbishment, NIF needs updates and upgrades to enable future operations. Livermore is deploying a NIF Sustainment Plan that includes many different projects, from upgrading equipment and maintenance processes to cleaning debris that can damage optics systems to addressing challenges in finding spare parts.

As Livermore looks to the future, our research across every part of the Laboratory enables us to more effectively understand how complex systems and feedback loops interact with different variables. These examples of Livermore’s work exemplify the many ways the Laboratory is ensuring we can predict outcomes and provide the best resources to meet our mission.