SEARCH · Search NASA
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.
This presentation is during a course on Machine Learning in Mechanics at the Society of Experimental Mechanics Annual Conference 2025.
ULYSSES is a python package that calculates the baryon asymmetry produced from leptogenesis in the context of a type-I seesaw mechanism. The code solves the semi-classical Boltzmann equations for points in the model parameter space as specified by the user. We provide a selection of predefined Boltzmann equations as well as a plugin mechanism for externally provided models of leptogenesis. Furthermore, the ULYSSES code provides tools for multi-dimensional parameter space exploration. The emphasis of the code is on user flexibility and rapid evaluation. It is publicly available at https://github.com/earlyuniverse/ulysses
Interactive machine learning (ML) systems are difficult to design because of the "Two Black Boxes" problem that exists at the interface between human and machine. Many algorithms that are used in interactive ML systems are black boxes that are presented to users, while the human cognition represents a second black box that can be difficult for the algorithm to interpret. These two black boxes create cognitive gaps between the user and the interactive ML model when a user interacts with the system. In this paper, we identify several cognitive gaps that exist in a previously-developed interactive visual analytics (VA) system, Andromeda. These cognitive gaps that we are addressing in Andromeda are representative of common problems in other VA systems. Our goal with this work is to open both black boxes and bridge these cognitive gaps by making improvements to the original Andromeda system, including designing new visual features to help people better understand how Andromeda processes and interacts with data and improving the underlying algorithm so that the Andromeda system can better understand the intent of the user during the data exploration process. We evaluate our designs through both qualitative and quantitative analysis (i.e., user study and simulation analysis), and the results confirm that the improved Andromeda system outperforms the original version significantly in a series of high-dimensional data understanding tasks.
In today’s virtual reality (VR), users have various ways to influence their VR experience, including alphanumeric input. While typing characters and numbers is straightforward on desktop computers, it presents challenges and opportunities in head-mounted display VR due to specific interaction methods and a lack of real-world visual stimuli. Addressing these open questions, our work implements and evaluates ten approaches to alphabetic and numeric inputs in VR. Here, we describe the design motivation behind these input methods and evaluate them in a user study with 40 participants divided into groups for alphabetic and numeric keyboards. This comparison investigates each method’s performance and user interactions. Our findings suggest that different input methods significantly impact words per minute and error rates, and that certain keyboard designs may receive better subjective evaluations despite poorer objective performance.
PVDeg version 0.7.0, released in December 2025, introduced major enhancements to improve usability and performance. This update reorganized tutorials and tool notebooks to create a more intuitive experience, enabling users to easily follow and adapt workflows for their specific analyses. In addition to structural improvements, both the notebooks and core logic underwent significant optimization for efficiency, robustness, and style. These refinements were supported by new testing frameworks built on nbval and pytest, adherence to PEP8 standards, and extensive code refactoring, which collectively simplify onboarding for new developers. Looking ahead, version 0.8.0 will deliver advanced AI-driven capabilities. The primary focus is to further develop and automate the degradation workflow, designed to analyze PV module degradation across diverse locations and system configurations. By integrating large language models (LLMs) to scan literature and compile a comprehensive database of materials and degradation rates, this feature will enable modeling of multiple materials and mechanisms within a single, streamlined workflow. Users will be able to evaluate degradation impacts on different system architectures under varying environmental conditions, facilitating informed decisions on bill-of-materials optimization for specific deployment scenarios. These advancements position PVDeg as a powerful, user-friendly tool for accelerating PV reliability research and system design.
Note: This is the last version (a2.6.1) of NRAP-Open-IAM released during NRAP Phase II in 2022. The latest version of NRAP-Open-IAM is available here: https://edx.netl.doe.gov/dataset/phase-iii-nrap-open-iam NRAP-Open-IAM is an open-source software product that enables quantification of containment effectiveness and leakage risk at storage sites in the context of system uncertainties and variability. NRAP-Open-IAM represents the next-generation in a line of systems-based computational models developed for quantitative geological carbon storage (GCS) risk assessment. The model comprises a set of reduced-order and analytical models of various components of the GCS system, potential leakage pathways, receptors of concern including impact to groundwater resources and the atmosphere, a framework to support stochastic simulation, time stepping, uncertainty quantification, other analytical functionality for scenario and risk-performance evaluation, and a basic graphical user interface to support scenario development, data input simulation definition, and basic post-processing and results display. As the NRAP Open-IAM functionality continues to evolve, we continue to add to its capability to develop quantitative, probabilistic, and time-dependent profiles of the evolution of risk at a GCS site and evaluate the influence of uncertain parameters on uncertainty in predicted risk. It can be used to quantify the dynamics of reservoir saturation plume and pressure-affected area, for evaluation of the area of potential groundwater impact (i.e., Area of Review) and monitoring requirements to support cost and regulatory analysis, and for consideration of different post-injection site care and closure scenarios. This submission contains the current version of NRAP-Open-IAM available for evaluation and testing. To use the NRAP-Open-IAM, download the source code (https://edx.netl.doe.gov/dataset/08f396c8-bc5f-44ad-a028-6e98a6ea6d70/resource/4c24a3da-3b40-4ffe-9892-c807ae9f8760) then open the NRAP-Open-IAM user's guide (https://edx.netl.doe.gov/dataset/08f396c8-bc5f-44ad-a028-6e98a6ea6d70/resource/8b27335a-343c-4836-b8a3-3ad0bdc9e669) to read more about the tool. Installation instructions for Windows, Mac, and Linux can be found in the "installers" folder of the extracted NRAP-Open-IAM folder and describe setup of environment (e.g., Python libraries) needed for proper work of the tool. Test of installation can be done by running "python openiam_setup_tests.py" in the "setup" folder. The installation test also runs a test suite to see if the NRAP-Open-IAM has been installed correctly. To run the test suite separately, run "python iam_test.py" in the "test" folder. User's guide: https://edx.netl.doe.gov/dataset/08f396c8-bc5f-44ad-a028-6e98a6ea6d70/resource/8b27335a-343c-4836-b8a3-3ad0bdc9e669 Developer's guide: https://edx.netl.doe.gov/dataset/08f396c8-bc5f-44ad-a028-6e98a6ea6d70/resource/3bc6ee7d-609d-4eb6-80ba-fa6130ee0313 Reservoir simulation data used in some examples distributed with NRAP-Open-IAM: - Kimberlina: https://edx.netl.doe.gov/dataset/08f396c8-bc5f-44ad-a028-6e98a6ea6d70/resource/eb62cece-61b2-4037-9b6d-32407dde2ab8 - Kimberlina (compartmentalized): https://edx.netl.doe.gov/dataset/08f396c8-bc5f-44ad-a028-6e98a6ea6d70/resource/366f9530-3b32-4b84-affe-ab2df1d9a8b5 - FutureGen 2.0: https://edx.netl.doe.gov/dataset/futuregen-2-0-1008-simulation-reservoir-lookup-table NRAP-Open-IAM GitLab repository: https://gitlab.com/NRAP/OpenIAM Related publications: - Bacon, D., Yonkofski, C., Brown, C., Demirkanli, D. and Whiting, J., 2019. Risk-based post injection site care and monitoring for commercial-scale carbon storage: Reevaluation of the FutureGen 2.0 site using NRAP-Open-IAM and DREAM. International Journal of Greenhouse Gas Control 90: 102784. - Bacon, D. Demirkanli, D., and White, S., 2020. Probabilistic risk-based Area of Review (AoR) determination for a deep-saline carbon storage site. International Journal of Greenhouse Gas Control 102: 103153. - Harp, D., Oldenburg, C., and Pawar, R., 2019. A metric for evaluating conformance robustness during geologic CO2 sequestration operations. International Journal of Greenhouse Gas Control 85: 100-108. - Lackey, G., Vasylkivska, V., Huerta, N., King, S., and Dilmore, R., 2019. Managing well leakage risks at a geologic carbon storage site with many wells, International Journal of Greenhouse Gas Control, 88 :182-194. - Vasylkivska, V., Dilmore, R., Lackey, G., Zhang, Y., King, S., Bacon, D., Chen, B., Mansoor, K., and Harp, D., 2021. NRAP-Open-IAM: A flexible open-source integrated assessment model for geologic carbon storage risk assessment and management, Environmental Modelling & Software, 143: 105114. Presentations: - Chen, B., Harp, D., and Pawar, R., A data assimilation approach (ES-MDA) coupling with NRAP-Open-IAM for quantifying uncertainty reduction in geological CO2 sequestration. AGUFM 2019: T44A-02. - Chen, B., and Harp, D., Improving risk analysis precision for geologic CO2 sequestration by quantifying the uncertainty reduction before and after acquiring monitoring data. 14th Greenhouse Gas Control Technologies Conference, Melbourne, Australia, 2018, pp. 21-26. - Harp, D., National Risk Assessment Partnership Task 2: Containment Assurance. No. LA-UR-19-28654, Los Alamos National Laboratory (LANL), Los Alamos, NM (United States), 2019. - Vasylkivska, V., King, S., Bacon, D., Harp, D., Chen, B., Mansoor, K., Onishi, T., Yang, Y., Zhang, Y., and Keating, E., NRAP-Open-IAM: An open-source integrated assessment model, poster, Mastering the Subsurface Through Technology Innovation, Partnerships and Collaboration: Carbon Storage and Oil and Natural Gas Technologies Review Meeting, Pittsburgh, PA, August 13-16, 2018. - Vasylkivska, V., Lackey, G., King, S., Wentworth, A., Huerta, N., Creason, C., DiGiulio, J., Yang, Y., and Dilmore, R., Long-term risk analysis of a geologic CO2 storage project during the post-injection period, SIAM Conference on Computational Science and Engineering, Spokane, WA, February 25-March 1, 2019. - Vasylkivska, V., Overview of the NRAP-Open-IAM tool for carbon storage (beta release), 2019 Annual NRAP Tool Users Meeting, Pittsburgh, PA, August 27, 2019. - Vasylkivska, V., Bacon, D., Chen, B., Dilmore, R., Harp, D., King, S., Lackey, G., Lindner, E., Liu, G., Mansoor, K. and Zhang, Y., NRAP-Open-IAM: A new, open-source code for integrated assessment of geologic carbon storage containment effectiveness and leakage risk, poster, American Geophysical Union Fall Meeting 2020 (virtual meeting), December 2020. - Vasylkivska, V., NRAP open-source integrated assessment model and relevant application, oral presentation, NRAP workshop "NRAP Tools for Geologic Carbon Storage Risk-Based Decision Making" held in conjunction with Groundwater Protection Council (GWPC) 2021 Annual Forum (virtual meeting), Salt Lake City, UT, September 2021. - Vasylkivska, V., NRAP-Open-IAM: open-source integrated assessment model, digital poster/demonstration, software demonstration session, 2022 Carbon Management Project Review Meeting, August 16, 2022
Ytopt is a Python machine-learning-based autotuning software package developed within the ECP PROTEAS-TUNE project. The ytopt software adopts an asynchronous search framework that consists of sampling a small number of input parameter configurations and progressively fitting a surrogate model over the input-output space until exhausting the user-defined maximum number of evaluations or the wall-clock time. libEnsemble is a Python toolkit for coordinating workflows of asynchronous and dynamic ensembles of calculations across massively parallel resources developed within the ECP PETSc/TAO project. libEnsemble helps users take advantage of massively parallel resources to solve design, decision, and inference problems and expands the class of problems that can benefit from increased parallelism. In this paper we present our methodology and framework to integrate ytopt and libEnsemble to take advantage of massively parallel resources to accelerate the autotuning process. Specifically, we focus on using the proposed framework to autotune the ECP ExaSMR application OpenMC, an open source Monte Carlo particle transport code. OpenMC has seven tunable parameters some of which have large ranges such as the number of particles in-flight, which is in the range of 100,000 to 8 million, with its default setting of 1 million. Setting the proper combination of these parameter values to achieve the best performance is extremely time-consuming. Therefore, we apply the proposed framework to autotune the MPI/OpenMP offload version of OpenMC based on a user-defined metric such as the figure of merit (FoM) (particles/s) or energy efficiency energy-delay product (EDP) on Crusher at Oak Ridge Leadership Computing Facility. In conclusion, the experimental results show that we achieve the improvement up to 29.49% in FoM and up to 30.44% in EDP.
Community Microgrids are increasingly gaining popularity worldwide for their efficiency, cost-effectiveness, and local resilience improvement. Microgrid planning tools play a crucial role in their deployment. In the process, tentative designs of the microgrid are simulated, analyzed, and optimized. Due to the complexity of the problem, planning tools must carefully balance computational efficiency while seeking the most optimal solutions. This paper investigates the impact of algorithm selection on CPU and memory utilization of a Community Microgrid planning tool that is specifically designed for non-technical users. We compare two version of the code with two alternatives for Community Microgrid planning tools: a Greedy Algorithm, and a Linear Programming Approach. This comparison examines scenarios spanning from 2 to 50 buildings. Our findings revealed a significant difference in performance between the two algorithms, underscoring the critical role of algorithm selection in optimizing the efficiency of Community Microgrid Planning Tools.
HPC systems typically rely on the fixed-lifetime (FLT) data retention strategy, which only considers temporal locality of data accesses to parallel file systems. However, our extensive analysis based on the leadership-class HPC system traces suggests that the FLT approach often fails to capture the dynamics in users' behavior and leads to undesired data purge. In this study, we propose an activeness-based data retention (ActiveDR) solution, which advocates considering the data retention approach from a holistic activeness-based perspective. By evaluating the frequency and impact of users' activities, ActiveDR prioritizes the file purge process for inactive users and rewards active users with extended file lifetime on parallel storage. Our extensive evaluations based on the traces of the prior Titan supercomputer show that, when reaching the same purge target, ActiveDR achieves up to 37% file miss reduction as compared to the current FLT retention methodology.
INL has evaluated and refurbished General Purpose Heat Source (GPHS) Radioisotope Thermoelectric Generator (RTG) Flight Unit 5 (GPHS-RTG F5R or F5R) that was defueled in 2005. The generator has passed all the INL internal reviews, an end item data package has been delivered to NASA and is ready fueling for a mission. NASA requested a user’s guide for F5R with enough detail for a prospective mission proposal. This report documents the development of the power prediction model used in the user’s guide. The prediction evaluated the thermoelectric “burn-in” and the graceful degradation modes of the missions and developed a thermal inventory dependent equation that predicts the power that matches Galileo, Cassini, and PNH within a three-standard deviation of ±1%. This prediction is only valid for the 30 VDC load voltage of these missions.
INL has evaluated and refurbished General Purpose Heat Source (GPHS) Radioisotope Thermoelectric Generator (RTG) Flight Unit 5 (GPHS-RTG F5R or F5R) that was defueled in 2005. The generator has passed all the INL internal reviews, an end item data package has been delivered to NASA and is ready fueling for a mission. NASA requested a user’s guide for F5R with enough detail for a prospective mission proposal. This report documents the development of the power prediction model used in the user’s guide. The prediction evaluated the thermoelectric “burn-in” and the graceful degradation modes of the missions and developed a thermal inventory dependent equation that predicts the power that matches Galileo, Cassini, and PNH within a three-standard deviation of ±1%. This prediction is only valid for the 30 VDC load voltage of these missions.
US Ecology Idaho (USEI) operates a Resource Conservation and Recovery Act (RCRA) Subtitle C, Hazardous Waste Treatment, Storage, and Disposal (TSD) Facility on a 640-acre property in Owyhee County, Idaho. USEI accepts a wide variety of RCRA-exempt low activity radioactive wastes, including naturally occurring radioactive material (NORM) and technologically enhanced NORM (TENORM) (USEI 2009). The performance of the landfill with regard to radiological operating permit requirements was previously assessed using a model constructed with the residual radioactivity (RESRAD) computer program, developed by Argonne National Laboratory for the U.S. Department of Energy (DOE). RESRAD supports the evaluation of several environmental transport pathways related to a radionuclide-contaminated soil source term, but it was not conceived as a model for the evaluation of landfill radiological performance. A more comprehensive and realistic model is desired in order to better support submissions to both the State of Idaho's Department of Environmental Quality as well as the U.S. Nuclear Regulatory Commission. A Performance Assessment (PA) computer model was developed using GoldSim software for the USEI RCRA Subtitle C landfill. The RESRAD computer model was specifically developed for calculating soil cleanup criteria and radiological dose and cancer risk from residual radioactive material in soil. To evaluate site-specific conditions, users may select from among a number of available environmental transport and exposure pathways and modify 'default' parameter values. In GoldSim, a user must construct the model 'from scratch.' However, this modeling is facilitated by a number of specialized elements available in GoldSim to support a radiological mass transport model. These include elements for defining radiological decay and ingrowth, container failure and radiological release, advective and diffusive transport, and other processes. Advantages afforded by the use of RESRAD for modeling radiological dose for a RCRA Subtitle C landfill include ease of use and reasonable flexibility in specifying site-specific conditions. GoldSim allows for considerably more flexibility and site-specificity than RESRAD, including evaluation of potentially relevant environmental transport processes not supported in RESRAD. The probabilistic modeling capabilities of GoldSim also far exceed those of RESRAD. The pros and cons of RESRAD and GoldSim for this modeling problem are explored with a focus on identifying approaches and critical factors in identifying the appropriate platform. The capabilities of RESRAD and GoldSim for mathematically modeling the disposal system will be contrasted and compared. For example, the existing RESRAD radiological safety assessment evaluated potentially complete exposure pathways related to infiltration to groundwater and upwards diffusion of radon. Additional transport pathways identified in the USEI Idaho facility Conceptual Site Model include deposition of radon decay products in cover material, root uptake of radionuclides in disposed wastes by native plants, and mixing of cover material by burrowing animals. Results of the two models and how the different transport and dose pathways affect results will be discussed. (authors)
This tool, the Non-Powered Dam Hydropower Development and Ranking Opportunity Tool (NPD HYDRO), allows users to prioritize or rank NPD sites for future development based on user-defined priorities. Benefits can be evaluated in several categories: the grid, community, industry, and environment (represented by the GCIE impact scores in the tool. In addition, the tool provides the user with a qualitative measure for the feasibility of adding energy storage during NPD conversion, specifically battery, hydrogen electrolysis, and pumped-storage hydropower (PSH). By taking a holistic approach to the potential range of benefits provided by NPD conversions with results tailored to the user’s priorities, NPD HYDRO provides the opportunity for national-level screening and enables users to focus on the sites that are most closely aligned with their interests.
The EvaluateFlux program is a post-processing utility program for DIF3D, specifically DIF3D-VARIANT which handles Cartesian and hexagonal geometries. The EvaluateFlux program was developed to allow users to obtain flux and power traverses through the geometry domain, and its initial purpose was to facilitate foil analysis by evaluating the flux solution from DIF3D-VARIANT and combining it with foil cross section data. The EvaluateFlux program can calculate the neutron flux, as well as the reaction rates, at any user provided evaluation point. It does this by identifying the spatial mesh associated with the evaluation point and then evaluates the polynomial based neutron flux moments stored in the NHFLUX file at that point. The output of EvaluateFlux varies depending on the input setup. The maximum output includes the neutron flux and microscopic and macroscopic reaction rates at each evaluation point. The purpose of this work is to verify the outputs of EvaluateFlux. Simple models that have hand calculatable results are first defined and used to verify the EvaluateFlux outputs. More complex cases are then added where a duplicate program of EvaluateFlux that uses PrintTables outputs of the binary files is used to verify the EvaluateFlux outputs. In those complex cases, hand calculations of selected evaluation points were also displayed to confirm the software verification. For all the tests done, the hand calculations agreed well with those calculated by EvaluateFlux. For the larger complex problems, the duplicate program that can process hundreds of evaluation points was able to identify that zero points within some meshes have large errors. This aspect was attributed to the truncation error on the input provided to the duplicate program and is not a concern for the accuracy of the EvaluateFlux software.
The EvaluateFlux program is a post-processing utility program for DIF3D, specifically DIF3D-VARIANT which handles Cartesian and hexagonal geometries. The EvaluateFlux program was developed to allow users to obtain flux and power traverses through the geometry domain, and its initial purpose was to facilitate foil analysis by evaluating the flux solution from DIF3D-VARIANT and combining it with foil cross section data. The EvaluateFlux program can calculate the neutron flux, as well as the reaction rates, at any user provided evaluation point. It does this by identifying the spatial mesh associated with the evaluation point and then evaluates the polynomial based neutron flux moments stored in the NHFLUX file at that point. The output of EvaluateFlux varies depending on the input setup. The maximum output includes the neutron flux and microscopic and macroscopic reaction rates at each evaluation point. The purpose of this work is to verify the outputs of EvaluateFlux. Simple models that have hand calculatable results are first defined and used to verify the EvaluateFlux outputs. More complex cases are then added where a duplicate program of EvaluateFlux that uses PrintTables outputs of the binary files is used to verify the EvaluateFlux outputs. In those complex cases, hand calculations of selected evaluation points were also displayed to confirm the software verification. For all the tests done, the hand calculations agreed well with those calculated by EvaluateFlux. For the larger complex problems, the duplicate program that can process hundreds of evaluation points was able to identify that zero points within some meshes have large errors. This aspect was attributed to the truncation error on the input provided to the duplicate program and is not a concern for the accuracy of the EvaluateFlux software.
The SCALE Leap-In for Cores at Equilibrium (SLICE) method leverages capabilities available in the SCALE modeling and simulation suite to facilitate rapid estimation of equilibrium core inventories for flowing-pebble reactor systems in support of a variety of fuel cycle applications. New capabilities that compliment the SLICE method have been developed for the ORIGAMI interface to ORIGEN in SCALE to facilitate rapid depletion calculations for flowing-pebble systems. New features include a more generalized and flexible means of specifying interpolation dimensions, the ability to evaluate pebbles moving through user-defined “axial zones” in each of the pebble’s multiple passes through the core, and the treatment of differential velocities across radial channels (representing both pebble-to-pebble and wall-to-pebble friction effects). ORIGAMI thus provides an efficient user interface to define a pebble’s path and irradiation histories of its multiple passes through the core to calculate the pebble’s time-dependent inventories, which can be useful in various applications domains such as safeguards, criticality safety, and disposal analyses. The consistency of new ORIGAMI approach was verified against the solution obtained using the SLICE method with ORIGEN-ARP, showing excellent agreement.
Principal Component Analysis can be used to reduce the cost of Computational Fluid Dynamics simulations of turbulent reacting flows by reducing the dimensionality of the transported variables through projection of the thermochemical state onto a lower-dimensional manifold. However, because of the nonlinearity of the principal component source terms, nonlinear regression techniques must be utilized for the source terms in terms of the principal components. Unfortunately, widely available and utilized nonlinear regression techniques can have prohibitive computational requirements and/or accuracy that is highly dependent on user experience in ad hoc tuning of model architecture and hyperparameters. Here, in this work, a new nonlinear regression approach is proposed that is both computationally efficient and automated so does not require any user input. The approach is evaluated through a priori prediction of principal component source terms using data from a Direct Numerical Simulation of a turbulent nonpremixed n-heptane/air jet flame. In particular, the proposed framework consists of local regressions whose complexity is adapted according to the local nonlinearity of the data: local linear regression when accurate enough and local Artificial Neural Networks when nonlinear regression is required. The number of local clusters for local regression is determined automatically using the Davies-Bouldin index. In addition, Bayesian optimization is utilized for model training (i.e., to select the best architectures and hyperparameters of the nonlinear regressions in an unsupervised fashion), eliminating ad hoc hand-tuning and/or expensive grid searches. Overall, compared to a single, global neural network, the new local adaptive regression approach is shown to have comparable accuracy but 69% less training time due to the utilization of local linear regression and faster training of local neural networks.
Abstract Over the last decades, climate science has evolved rapidly across multiple expert domains. Our best tools to capture state‐of‐the‐art knowledge in an internally self‐consistent modeling framework are the increasingly complex fully coupled Earth System Models (ESMs). However, computational limitations and the structural rigidity of ESMs mean that the full range of uncertainties across multiple domains are difficult to capture with ESMs alone. The tools of choice are instead more computationally efficient reduced complexity models (RCMs), which are structurally flexible and can span the response dynamics across a range of domain‐specific models and ESM experiments. Here we present Phase 2 of the Reduced Complexity Model Intercomparison Project (RCMIP Phase 2), the first comprehensive intercomparison of RCMs that are probabilistically calibrated with key benchmark ranges from specialized research communities. Unsurprisingly, but crucially, we find that models which have been constrained to reflect the key benchmarks better reflect the key benchmarks. Under the low‐emissions SSP1‐1.9 scenario, across the RCMs, median peak warming projections range from 1.3 to 1.7°C (relative to 1850–1900, using an observationally based historical warming estimate of 0.8°C between 1850–1900 and 1995–2014). Further developing methodologies to constrain these projection uncertainties seems paramount given the international community's goal to contain warming to below 1.5°C above preindustrial in the long‐term. Our findings suggest that users of RCMs should carefully evaluate their RCM, specifically its skill against key benchmarks and consider the need to include projections benchmarks either from ESM results or other assessments to reduce divergence in future projections.