A5026632303- Meteorological Phenomena and Simulations
- Soil Moisture and Remote Sensing
- Precipitation Measurement and Analysis
- Scientific Computing and Data Management
- Distributed and Parallel Computing Systems
- Climate variability and models
- Atmospheric aerosols and clouds
- Atmospheric and Environmental Gas Dynamics
- Geophysics and Gravity Measurements
- Plant Water Relations and Carbon Dynamics
- Remote Sensing in Agriculture
- Climate change and permafrost
- Atmospheric chemistry and aerosols
- Atmospheric Ozone and Climate
- Air Quality Monitoring and Forecasting
- Big Data and Business Intelligence
- Remote Sensing and Land Use
- Flood Risk Assessment and Management
- Advanced Computational Techniques and Applications
- Advanced Data Storage Technologies
- Research Data Management Practices
- Remote-Sensing Image Classification
- Tropical and Extratropical Cyclones Research
- Advanced Image and Video Retrieval Techniques
- Oceanographic and Atmospheric Processes
Oak Ridge National Laboratory
2012-2025
Office of Scientific and Technical Information
2022-2024
National Technical Information Service
2022
Oak Ridge Leadership Computing Facility
2021
Oak Ridge Associated Universities
2021
New York University
2019
Mississippi State University
2004-2013
This work documents the first version of U.S. Department Energy (DOE) new Exascale Earth System Model (E3SMv1). We focus on standard resolution fully coupled physical model designed to address DOE mission-relevant water cycle questions. Its components include atmosphere and land (110-km grid spacing), ocean sea ice (60 km in midlatitudes 30 at equator poles), river transport (55 km) models. base configuration will also serve as a foundation for additional configurations exploring higher...
Several recommendations have been proposed for detecting land use and cover change (LULCC) on the environment from, observed climatic records to modeling improve its understanding impacts climate. Researchers need detect LULCCs accurately at appropriate scales within a specified time period better understand their climate provide improved estimates of future The US Climate Reference Network (USCRN) can be helpful in monitoring LULCC near-surface atmospheric conditions, including temperature....
Abstract In an attempt to advance the understanding of Earth's weather and climate by representing deep convection explicitly, we present a global, four‐month simulation (November 2018 February 2019) with ECMWF's hydrostatic Integrated Forecasting System (IFS) at average grid spacing 1.4 km. The impact explicitly simulating on atmospheric circulation its variability is assessed comparing km equivalent well‐tested calibrated global simulations 9 without parametrized convection. explicit...
Triggered by the realization that AI emulators can rival performance of traditional numerical weather prediction models running on HPC systems, there is now an increasing number large address use cases such as forecasting, downscaling, or nowcasting. While parallel developments in literature focus foundation -- be effectively tuned to multiple, different and climate side largely single-use with particular emphasis mid-range forecasting. We close this gap introducing Prithvi WxC, a 2.3...
Gravity waves (GWs) make crucial contributions to the middle atmospheric circulation. Yet, their climate model representation remains inaccurate, leading key circulation biases. This study introduces a set of three neural networks (NNs) that learn predict GW fluxes (GWFs) from multiple years high-resolution ERA5 reanalysis. The NNs: 1x1 ANN, 3x3 ANN-CNN, and an Attention UNet embed different levels horizontal nonlocality in architecture are capable representing nonlocal effects missing...
Few hundred million to billion parameters of autoregressive transformer-based weather foundation models (FMs) have demonstrated generalizabilities for various downstream applications such as regional forecasting downscaling. They occasionally outperform traditional physics-based medium-range skills well enable significantly faster execution speeds. These ML approaches are designed timeframes ranging from hourly up 10 days, and sub-seasonal forecasting, defined a range spanning two weeks...
AI foundation models have already demonstrated their usefulness in harnessing potential a wide range of science application domains. They derive power from the large volumes data, along with computational methods used to exploit them using unprecedented amounts compute power. We are inundated data but managed only small fraction available data.  The Earth System Grid Federation (ESGF) is hosting nearly 16 PB collection Coupled Model Intercomparison Project (CMIP6), expected grow 5 -...
Abstract: Cloud cover poses a significant obstacle in harnessing multi-spectral satellite imagery for various earth observation applications including disaster response, land use and mapping. To address this issue, study investigates the potential of Prithvi WxC foundation model (Johannes Schmude et al., 2024), deep learning architecture designed weather climate applications, to perform cloud gap imputation. By leveraging its ability capture atmospheric dynamics predict missing data, offers...
The growing interest in deep learning and large language models (LLMs) recent years highlights their remarkable adaptability ability to generalize, drawing researchers from a wide array of disciplines. Despite promise, many instances, these advancements have exposed lack transparency rigor during development processes. Although this rapid pace research undoubtedly offers numerous benefits, it has also led an increasing prevalence works conducted without superficial way. Code that is not...
AI-based weather emulators have begun to rival the accuracy of traditional numerical solvers, for a fraction computational cost. The question whether they can be reliably deployed in all use cases (e.g., forecast extreme scenarios), however, is still open. We outline an ensembling strategy based on architectural variations Prithvi WxC foundation model (FM), highlighting impact each these physical and ability capture distributional extremes. A simple ensemble 100 models sufficient observe...
AI foundation models hold considerable promise for leveraging the vast and diverse datasets available in atmospheric geoscientific research. These have potential to advance scientific discovery by capturing complex spatial temporal relationships inherent earth system processes. However, development deployment of such is often hindered limited computational resources.Accurate reconstruction fine-scale features from coarse-resolution data a critical challenge modeling, as well benchmark...
Abstract A coupled atmosphere-ocean model is necessary for tropical cyclone (TC) prediction to accurately characterize ocean feedback on atmospheric processes within the TC environment. Here, ECMWF global run at horizontal resolutions from 9 km 1.4 in atmosphere, as well 25 and 8 ocean, identify how resolution impacts forecast accuracy of four observed major TCs Atlantic: Irma, Florence, Teddy Ida. Most used here are unprecedented models. GOES-16 SAR satellite images best track data...
Abstract Probable maximum precipitation (PMP), defined as the largest rainfall depth that could physically occur under a series of adverse atmospheric conditions, has been an important design criterion for critical infrastructures such dams and nuclear power plants. To understand how PMP may respond to projected future climate forcings, we used physics‐based numerical weather simulation model estimate across various durations areas over Alabama‐Coosa‐Tallapoosa (ACT) River Basin in...
Abstract This study integrates machine learning and particle‐resolved aerosol simulations to develop emulators that predict submicron mixing state indices from the Earth system model (ESM) simulations. The using only quantities are predicted by ESM, including bulk species concentrations, which do not themselves carry information. We used PartMC‐MOSAIC as NCAR's CESM ESM. trained for three different in terms of chemical abundance ( χ a ), optically absorbing nonabsorbing o hygroscopic...
Precipitation extremes have tangible societal impacts. Here, we assess if current state of the art global climate model simulations at high spatial resolutions (0.35° × 0.35°) capture observed behavior precipitation in past few decades over continental US. We design a correlation-based regionalization framework to quantify extremes, where samples extreme events for grid box may also be drawn from neighboring boxes with statistically equal means and significant temporal correlations....
In this paper, the link-based cluster ensemble (LCE) method is utilized to improve cloud classification and satellite precipitation estimation. High resolution Satellite Precipitation Estimation (SPE) based on from Remotely Sensed Imagery using an Artificial Neural Network Cloud Classification (PERSIANN-CCS) algorithm. This modified SPE with incorporation of LCE involves following four steps: 1) segmentation infrared images into patches; 2) patch feature extraction; 3) clustering patches...
Abstract By employing wavelet and selected features (WSF), median merging (MM), curve-fitting (SCF) techniques, the Precipitation Estimation from Remotely Sensed Imagery using an Artificial Neural Networks Cloud Classification System (PERSIANN-CCS) has been improved. The PERSIANN-CCS methodology includes following four main steps: 1) segmentation of satellite cloud images into patches, 2) feature extraction, 3) classification 4) derivation temperature–rain-rate (T–R) relationship for every...
A relationship between the likelihood of wildfires and various drought metrics (soil moisture-based fire potential indices) were examined over southern part Mississippi. The following three indices tested used to simulate spatial temporal wildfire probability changes: (1) accumulated difference daily precipitation evapotranspiration (P - E); (2) simulated moisture content top 10 cm soil; (3) Keetch-Byram Drought Index (KBDI). These estimated from gridded meterological data Mosaic-simulated...
Abstract Accurately simulating a geostationary hyperspectral infrared sounder is critical for quantitative applications. Traditional radiation simulations of such instruments often overlook the influence slant observation geometry by using vertical profile assumption, leading to inadequate simulation accuracy. By global atmospheric profiles with 1 km spatial resolution, slant‐path effects on brightness temperature are quantified. Experiments indicate that has less impact longwave and...
Abstract To better understand error and spatial variability sources of soil moisture simulated with land surface models, observed values (using offline simulations the Noah model four layers approximately 1-km horizontal grid spacing) were compared. This comparison between modeled daily was performed over Lower Mississippi Delta region during summer–fall months 2004–06. The covered 2.5° × latitude–longitude domain forced by North American Land Data Assimilation System (NLDAS) atmospheric...
<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> We examine the Naval Research Laboratory (NRL) blended satellite (NRL-Blend) High-Resolution Precipitation Product (HRPP) as a proxy for Global Mission (GPM)-era HRPP by using NRL-Blend precipitation forcing in land surface models (LSM). use existing (late 2008) constellation of low Earth orbiting (LEO) microwave-based platforms baseline to impact omitting several and sensor types from future...
Abstract. Using an observing system simulation experiment (OSSE), we investigate the potential soil moisture retrieval capability of National Aeronautics and Space Administration (NASA) Aquarius radiometer (L-band 1.413 GHz) scatterometer (L-band, 1.260 GHz). We estimate errors in retrievals identify sources that could cause those errors. The OSSE includes (i) a land surface model NASA Land Information System, (ii) radiative transfer backscatter model, (iii) realistic orbital sampling (iv)...