Search NASA⌕ Search

Engineering topics

Kashinath, Karthik

Publications and source records attributed to Kashinath, Karthik.

Huge ensembles – Part 1: Design of ensemble weather forecasts using spherical Fourier neural operators

Abstract. Simulating low-likelihood high-impact extreme weather events in a warming world is a significant and challenging task for current ensemble forecasting systems. While these systems presently use up to 100 members, larger ensembles could enrich the sampling of internal variability. They may capture the long tails associated with climate hazards better than traditional ensemble sizes. Due to computational constraints, it is infeasible to generate huge ensembles (comprised of 1000–10 000 members) with traditional, physics-based numerical models. In this two-part paper, we replace traditional numerical simulations with machine learning (ML) to generate hindcasts of huge ensembles. In Part 1, we construct an ensemble weather forecasting system based on spherical Fourier neural operators (SFNOs), and we discuss important design decisions for constructing such an ensemble. The ensemble represents model uncertainty through perturbed-parameter techniques, and it represents initial condition uncertainty through bred vectors, which sample the fastest-growing modes of the forecast. Using the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System (IFS) as a baseline, we develop an evaluation pipeline composed of mean, spectral, and extreme diagnostics. With large-scale, distributed SFNOs with 1.1 billion learned parameters, we achieve calibrated probabilistic forecasts. As the trajectories of the individual members diverge, the ML ensemble mean spectra degrade with lead time, consistent with physical expectations. However, the individual ensemble members' spectra stay constant with lead time. Therefore, these members simulate realistic weather states during the rollout, and the ML ensemble passes a crucial spectral test in the literature. The IFS and ML ensembles have similar extreme forecast indices, and we show that the ML extreme weather forecasts are reliable and discriminating. These diagnostics ensure that the ensemble can reliably simulate the time evolution of the atmosphere, including low-likelihood high-impact extremes. In Part 2, we generate a huge ensemble initialized each day in summer 2023, and we characterize the simulations of extremes.

Mahesh, Ankur↗

Huge ensembles – Part 2: Properties of a huge ensemble of hindcasts generated with spherical Fourier neural operators

Abstract. In Part 1, we created an ensemble based on spherical Fourier neural operators. As initial condition perturbations, we used bred vectors, and as model perturbations, we used multiple checkpoints trained independently from scratch. Based on diagnostics that assess the ensemble's physical fidelity, our ensemble has comparable performance to operational weather forecasting systems. However, it requires orders-of-magnitude fewer computational resources. Here in Part 2, we generate a huge ensemble (HENS), with 7424 members initialized each day of summer 2023. We enumerate the technical requirements for running huge ensembles at this scale. HENS precisely samples the tails of the forecast distribution and presents a detailed sampling of internal variability. HENS has two primary applications: (1) as a large dataset with which to study the statistics and drivers of extreme weather and (2) as a weather forecasting system. For extreme climate statistics, HENS samples events 4σ away from the ensemble mean. At each grid cell, HENS increases the skill of the most accurate ensemble member and enhances coverage of possible future trajectories. As a weather forecasting model, HENS issues extreme weather forecasts with better uncertainty quantification. It also reduces the probability of outlier events, in which the verification value lies outside the ensemble forecast distribution.

Mahesh, Ankur↗

Application of the AI2 Climate Emulator to E3SMv2's Global Atmosphere Model, With a Focus on Precipitation Fidelity

Abstract Can the current successes of global machine learning‐based weather simulators be generalized beyond 2‐week forecasts to stable and accurate multiyear runs? The recently developed AI2 Climate Emulator (ACE) suggests this is feasible, based upon 10‐year simulations with a network trained on output from a physics‐based global atmosphere model using a grid spacing of approximately 110 km and forced by a repeating annual cycle of sea‐surface temperature. Here we show that ACE, without modification, can be trained to emulate another major atmospheric model, EAMv2, run at a comparable grid spacing for at least 10 years with similarly small climate biases—a prerequisite to wider applicability. With an analysis that combines multiple temporal, spatial, and frequency domain perspectives, we show that ACE faithfully represents the spatiotemporal structure of EAMv2 precipitation and related variables. Finally, we show that a pretrained ACE network is able to adapt to a new global climate model simulation data set with 10 fewer training steps than when starting from random initialization, all while still maintaining low levels of climate bias. Further analysis of these fine‐tuning experiments reveal ACE's intriguing ability to interpolate between distinct global climate models.

Duncan, James P. C.↗

ClimSim: A large multi-scale dataset for hybrid physics-ML climate emulation

Modern climate projections lack adequate spatial and temporal resolution due to computational constraints. A consequence is inaccurate and imprecise predictions of critical processes such as storms. Hybrid methods that combine physics with machine learning (ML) have introduced a new generation of higher fidelity climate simulators that can sidestep Moore’s Law by outsourcing compute-hungry, short, high-resolution simulations to ML emulators. However, this hybrid ML-physics simulation approach requires domain-specific treatment and has been inaccessible to MLexperts because of lack of training data and relevant, easy-to-use workflows. Wepresent ClimSim, the largest-ever dataset designed for hybrid ML-physics research. It comprises multi-scale climate simulations, developed by a consortium of climate scientists and ML researchers. It consists of 5.7 billion pairs of multivariate input and output vectors that isolate the influence of locally-nested, high-resolution, high-fidelity physics on a host climate simulator’s macro-scale physical state. The dataset is global in coverage, spans multiple years at high sampling frequency, and is designed such that resulting emulators are compatible with downstream coupling into operational climate simulators. We implement a range of deterministic and stochastic regression baselines to highlight the ML challenges and their scoring. The data (https://huggingface.co/datasets/LEAP/ClimSim_high-res2) and code(https://leap-stc.github.io/ClimSim)arereleasedopenlytosupport the development of hybrid ML-physics and high-fidelity climate simulations for the benefit of science and society.

artificial intelligence, machine learning↗

Towards physics-inspired data-driven weather forecasting: integrating data assimilation with a deep spatial-transformer-based U-NET in a case study with ERA5

Abstract. There is growing interest in data-driven weather prediction (DDWP), e.g., using convolutional neural networks such as U-NET that are trained on data from models or reanalysis. Here, we propose three components, inspired by physics, to integrate with commonly used DDWP models in order to improve their forecast accuracy. These components are (1) a deep spatial transformer added to the latent space of U-NET to capture rotation and scaling transformation in the latent space for spatiotemporal data, (2) a data-assimilation (DA) algorithm to ingest noisy observations and improve the initial conditions for next forecasts, and (3) a multi-time-step algorithm, which combines forecasts from DDWP models with different time steps through DA, improving the accuracy of forecasts at short intervals. To show the benefit and feasibility of each component, we use geopotential height at 500 hPa (Z500) from ERA5 reanalysis and examine the short-term forecast accuracy of specific setups of the DDWP framework. Results show that the spatial-transformer-based U-NET (U-STN) clearly outperforms the U-NET, e.g., improving the forecast skill by 45 %. Using a sigma-point ensemble Kalman (SPEnKF) algorithm for DA and U-STN as the forward model, we show that stable, accurate DA cycles are achieved even with high observation noise. This DDWP+DA framework substantially benefits from large (O(1000)) ensembles that are inexpensively generated with the data-driven forward model in each DA cycle. The multi-time-step DDWP+DA framework also shows promise; for example, it reduces the average error by factors of 2–3. These results show the benefits and feasibility of these three components, which are flexible and can be used in a variety of DDWP setups. Furthermore, while here we focus on weather forecasting, the three components can be readily adopted for other parts of the Earth system, such as ocean and land, for which there is a rapid growth of data and need for forecast and assimilation.

54 ENVIRONMENTAL SCIENCES↗

Testing the reliability of interpretable neural networks in geoscience using the Madden–Julian oscillation

Abstract. We test the reliability of two neural network interpretation techniques, backward optimization and layerwise relevance propagation, within geoscientific applications by applying them to a commonly studied geophysical phenomenon, the Madden–Julian oscillation. The Madden–Julian oscillation is a multi-scale pattern within the tropical atmosphere that has been extensively studied over the past decades, which makes it an ideal test case to ensure the interpretability methods can recover the current state of knowledge regarding its spatial structure. The neural networks can, indeed, reproduce the current state of knowledge and can also provide new insights into the seasonality of the Madden–Julian oscillation and its relationships with atmospheric state variables. The neural network identifies the phase of the Madden–Julian oscillation twice as accurately as a linear regression approach, which means that nonlinearities used by the neural network are important to the structure of the Madden–Julian oscillation. Interpretations of the neural network show that it accurately captures the spatial structures of the Madden–Julian oscillation, suggest that the nonlinearities of the Madden–Julian oscillation are manifested through the uniqueness of each event, and offer physically meaningful insights into its relationship with atmospheric state variables. We also use the interpretations to identify the seasonality of the Madden–Julian oscillation and find that the conventionally defined extended seasons should be shifted later by 1 month. More generally, this study suggests that neural networks can be reliably interpreted for geoscientific applications and may thereby serve as a dependable method for testing geoscientific hypotheses.

58 GEOSCIENCES↗

ClimateNet: an expert-labeled open dataset and deep learning architecture for enabling high-precision analyses of extreme weather

Abstract. Identifying, detecting, and localizing extreme weather events is a crucial first step in understanding how they may vary under different climate change scenarios. Pattern recognition tasks such as classification, object detection, and segmentation (i.e., pixel-level classification) have remained challenging problems in the weather and climate sciences. While there exist many empirical heuristics for detecting extreme events, the disparities between the output of these different methods even for a single event are large and often difficult to reconcile. Given the success of deep learning (DL) in tackling similar problems in computer vision, we advocate a DL-based approach. DL, however, works best in the context of supervised learning – when labeled datasets are readily available. Reliable labeled training data for extreme weather and climate events is scarce. We create “ClimateNet” – an open, community-sourced human-expert-labeled curated dataset that captures tropical cyclones (TCs) and atmospheric rivers (ARs) in high-resolution climate model output from a simulation of a recent historical period. We use the curated ClimateNet dataset to train a state-of-the-art DL model for pixel-level identification – i.e., segmentation – of TCs and ARs. We then apply the trained DL model to historical and climate change scenarios simulated by the Community Atmospheric Model (CAM5.1) and show that the DL model accurately segments the data into TCs, ARs, or “the background” at a pixel level. Further, we show how the segmentation results can be used to conduct spatially and temporally precise analytics by quantifying distributions of extreme precipitation conditioned on event types (TC or AR) at regional scales. The key contribution of this work is that it paves the way for DL-based automated, high-fidelity, and highly precise analytics of climate data using a curated expert-labeled dataset – ClimateNet. ClimateNet and the DL-based segmentation method provide several unique capabilities: (i) they can be used to calculate a variety of TC and AR statistics at a fine-grained level; (ii) they can be applied to different climate scenarios and different datasets without tuning as they do not rely on threshold conditions; and (iii) the proposed DL method is suitable for rapidly analyzing large amounts of climate model output. While our study has been conducted for two important extreme weather patterns (TCs and ARs) in simulation datasets, we believe that this methodology can be applied to a much broader class of patterns and applied to observational and reanalysis data products via transfer learning.

54 ENVIRONMENTAL SCIENCES↗

Detection Uncertainty Matters for Understanding Atmospheric Rivers

The 3rd ARTMIP Workshop What: Over 30 participants from multiple universities and research insititutions met to discuss new results from the Atmospheric River Tracking Method Intercomparison Project. Where: Lawrence Berkeley National Lab, Berkeley, CA, USA When: 16-18 October 2019

O'Brien, Travis↗

The Atmospheric River Tracking Method Intercomparison Project (ARTMIP): Quantifying Uncertainties in Atmospheric River Climatology

Atmospheric rivers (ARs) are now widely known for their association with high-impact weather events and long-term water supply in many regions. Researchers within the scientific community have developed numerous methods to identify and track of ARs—a necessary step for analyses on gridded data sets, and objective attribution of impacts to ARs. These different methods have been developed to answer specific research questions and hence use different criteria (e.g., geometry, threshold values of key variables, and time dependence). Furthermore, these methods are often employed using different reanalysis data sets, time periods, and regions of interest. The goal of the Atmospheric River Tracking Method Intercomparison Project (ARTMIP) is to understand and quantify uncertainties in AR science that arise due to differences in these methods. This paper presents results for key AR-related metrics based on 20+ different AR identification and tracking methods applied to Modern-Era Retrospective Analysis for Research and Applications Version 2 reanalysis data from January 1980 through June 2017. We show that AR frequency, duration, and seasonality exhibit a wide range of results, while the meridional distribution of these metrics along selected coastal (but not interior) transects are quite similar across methods. Furthermore, methods are grouped into criteria-based clusters, within which the range of results is reduced. AR case studies and an evaluation of individual method deviation from an all-method mean highlight advantages/disadvantages of certain approaches. For example, methods with less (more) restrictive criteria identify more (less) ARs and AR-related impacts. Finally, this paper concludes with a discussion and recommendations for those conducting AR-related research to consider.

Rutz, Jonathan J.↗

Topological data analysis and machine learning for recognizing atmospheric river patterns in large climate datasets

Identifying weather patterns that frequently lead to extreme weather events is a crucial first step in understanding how they may vary under different climate change scenarios. Here, we propose an automated method for recognizing atmospheric rivers (ARs) in climate data using topological data analysis and machine learning. The method provides useful information about topological features (shape characteristics) and statistics of ARs. We illustrate this method by applying it to outputs of version 5.1 of the Community Atmosphere Model version 5.1 (CAM5.1) and the reanalysis product of the second Modern-Era Retrospective Analysis for Research and Applications (MERRA-2). An advantage of the proposed method is that it is threshold-free – there is no need to determine any threshold criteria for the detection method – when the spatial resolution of the climate model changes. Hence, this method may be useful in evaluating model biases in calculating AR statistics. Further, the method can be applied to different climate scenarios without tuning since it does not rely on threshold conditions. We show that the method is suitable for rapidly analyzing large amounts of climate model and reanalysis output data.

54 ENVIRONMENTAL SCIENCES↗

Atmospheric River Tracking Method Intercomparison Project (ARTMIP): project goals and experimental design

The Atmospheric River Tracking Method Intercomparison Project (ARTMIP) is an international collaborative effort to understand and quantify the uncertainties in atmospheric river (AR) science based on detection algorithm alone. Currently, there are many AR identification and tracking algorithms in the literature with a wide range of techniques and conclusions. ARTMIP strives to provide the community with information on different methodologies and provide guidance on the most appropriate algorithm for a given science question or region of interest. All ARTMIP participants will implement their detection algorithms on a specified common dataset for a defined period of time. The project is divided into two phases: Tier 1 will utilize the Modern-Era Retrospective analysis for Research and Applications, version 2 (MERRA-2) reanalysis from January 1980 to June 2017 and will be used as a baseline for all subsequent comparisons. Participation in Tier 1 is required. Tier 2 will be optional and include sensitivity studies designed around specific science questions, such as reanalysis uncertainty and climate change. High-resolution reanalysis and/or model output will be used wherever possible. Proposed metrics include AR frequency, duration, intensity, and precipitation attributable to ARs. Here, we present the ARTMIP experimental design, timeline, project requirements, and a brief description of the variety of methodologies in the current literature. Furthermore, we also present results from our 1-month proof-of-concept trial run designed to illustrate the utility and feasibility of the ARTMIP project.

58 GEOSCIENCES↗

Resolution dependence of precipitation statistical fidelity in hindcast simulations

This article is a U.S. Government work and is in the public domain in the USA. Numerous studies have shown that atmospheric models with high horizontal resolution better represent the physics and statistics of precipitation in climate models. While it is abundantly clear from these studies that high-resolution increases the rate of extreme precipitation, it is not clear whether these added extreme events are “realistic”; whether they occur in simulations in response to the same forcings that drive similar events in reality. In order to understand whether increasing horizontal resolution results in improved model fidelity, a hindcast-based, multiresolution experimental design has been conceived and implemented: the InitiaLIzed-ensemble, Analyze, and Develop (ILIAD) framework. The ILIAD framework allows direct comparison between observed and simulated weather events across multiple resolutions and assessment of the degree to which increased resolution improves the fidelity of extremes. Analysis of 5 years of daily 5 day hindcasts with the Community Earth System Model at horizontal resolutions of 220, 110, and 28 km shows that: (1) these hindcasts reproduce the resolution-dependent increase of extreme precipitation that has been identified in longer-duration simulations, (2) the correspondence between simulated and observed extreme precipitation improves as resolution increases; and (3) this increase in extremes and precipitation fidelity comes entirely from resolved-scale precipitation. Evidence is presented that this resolution-dependent increase in precipitation intensity can be explained by the theory of Rauscher et al. (), which states that precipitation intensifies at high resolution due to an interaction between the emergent scaling (spectral) properties of the wind field and the constraint of fluid continuity.

54 ENVIRONMENTAL SCIENCES↗

fastKDE

This software implements the fast, self-consistent probability density estimation described by O'Brien et al. (2014, doi: ). It uses a non-uniform fast Fourier transform technique to reduce the computational cost of an objective and self-consistent kernel density estimation method.

O'Brien, Travis A.↗