ERA5
PulseAugur coverage of ERA5 — every cluster mentioning ERA5 across labs, papers, and developer communities, ranked by signal.
7 day(s) with sentiment data
-
New AI Model AUWave Reconstructs Ocean Wave Heights from Sparse Data
Researchers have developed AUWave, a novel deep learning model designed to reconstruct high-resolution significant wave height (SWH) fields from sparse buoy observations. This hybrid framework combines a station-wise en…
-
New AI critic learns to score realism in weather forecasts
Researchers have developed a novel method for evaluating the realism of machine learning weather forecasts by training a discriminator model. This "learned atmospheric critic" identifies and scores physical artifacts in…
-
New AsyncCouple-Flow method improves spatio-temporal forecasting with missing data
Researchers have developed a new method called AsyncCouple-Flow to improve multi-modal spatio-temporal forecasting. This approach addresses challenges such as different data sampling rates, missing modalities, and error…
-
New ML framework improves sea-ice type prediction using multi-label learning
Researchers have developed a novel framework for predicting sea-ice types by reframing the task as a weakly supervised multi-label proportion learning problem. This approach directly utilizes polygon-level ice chart lab…
-
New Earth Foundation Models Explore Geometry-Aware Representations
Researchers have proposed a new approach for Earth Foundation Models that explicitly incorporates physical typing and geometry-awareness. This method aims to improve prediction accuracy by distinguishing between differe…
-
InCommodities unveils Aries weather model, outperforming ECMWF on wind speed
Researchers at InCommodities have developed Aries, a new medium-range weather prediction model utilizing a SwinTransformer architecture. Trained on ERA5 reanalysis data, Aries predicts numerous atmospheric variables and…
-
Research paper analyzes ensemble complexity in photovoltaic forecasting
A new research paper explores the complexity of ensemble models used in photovoltaic forecasting. The study analyzes how adding components to an ensemble can improve predictions while also potentially increasing computa…
-
Deep learning improves 3D wind field retrieval from satellite data
Researchers have developed a novel method to improve the accuracy and efficiency of retrieving three-dimensional wind fields from satellite imagery. This new approach utilizes deep optical flow to replace traditional wi…
-
LLM Agent Reveals Gaps in Documenting Compound Climate Events
A new LLM-agent framework called CDEP Agent has been developed to bridge the gap between meteorological definitions of compound drought-to-extreme-precipitation (CDEP) events and their real-world documentation. In a cas…
-
New ClimPhyDM framework enhances weather forecasting with physics-informed AI
Researchers have introduced Climate Physics Dynamic Matching (ClimPhyDM), a novel framework designed to improve weather forecasting by integrating physics-based models with data-driven components. This variational, simu…
-
Dandelion: New Neural Network for Planetary Dynamics Simulation
Researchers have introduced Dandelion, a novel neural network architecture designed for simulating planetary dynamics on spherical surfaces. Unlike traditional models that struggle with spherical geometry, Dandelion uti…
-
AI weather model Aardvark goes probabilistic, attributing forecast uncertainty
Researchers have developed a probabilistic version of the Aardvark Weather model, an end-to-end AI system for weather forecasting. By incorporating learned noise in the observation encoder and Monte Carlo Dropout in the…
-
New neural field offers flexible regional temperature forecasting
Researchers have developed a novel neural field called the Continuous Spatiotemporal Temperature Forecaster (CSTF) to improve regional near-surface temperature forecasting. Unlike traditional models that produce fixed o…
-
AI models achieve high-resolution weather forecasting using data scaling and stretched grids
Two new research papers explore advanced machine learning techniques for high-resolution weather forecasting. The first paper introduces BaguanHR, a framework that uses super-resolution to synthesize high-resolution dat…
-
Earth observation embeddings enhance weather downscaling accuracy
Researchers have demonstrated that Earth observation embeddings can serve as effective descriptors for probabilistic weather downscaling. By integrating these embeddings, derived from TESSERA data at 10m resolution, int…
-
New framework links climate memory to European summer warming risk
A new research paper proposes an empirical framework to better understand and predict European summer warming. The framework decomposes regional warming indicators into inherited memory from slow-state climate patterns …
-
AI weather models show competitive skill in extreme conditions, study finds
A new study published on arXiv investigates the performance of AI weather models, specifically examining whether they underperform in extreme weather conditions. The research evaluated eleven different physical and AI f…
-
Frozen neural weather models adapted for climate simulation with Rescene
Researchers have developed Rescene, a novel method to adapt frozen neural weather models for climate simulation. By adding a deterministic wrapper and a spectrally shaped stochastic perturbation, Rescene enables a previ…
-
New generative model forecasts tropical cyclones with enhanced speed and accuracy
Researchers have developed a novel deep generative model for tropical cyclone forecasting that can jointly predict satellite imagery and atmospheric fields. This single-pass model, called Latent Rectified Flow, is signi…
-
VeinCast framework enhances weather forecasting with physics-guided graphs
Researchers have developed VeinCast, a novel framework for global medium-range weather forecasting that integrates physics-guided dynamic field graphs with graph-conditioned fusion. This approach combines predefined atm…