arXiv subjects

Leif Denby

Publications and source records attributed to Leif Denby.

6 recordsLinked to original sources

arXiv2026

IT-DPC-SRI: A Cloud-Optimized Archive of Italian Radar Precipitation (2010-2025)

We present IT-DPC-SRI, the first publicly available long-term archive of Italian weather radar precipitation estimates, spanning 16 years (2010--2025). The dataset contains Surface Rainfall Intensity (SRI) observations from the Italian Civil Protection Department's national radar mosaic, harmonized into a coherent Analysis-Ready Cloud-Optimized (ARCO) Zarr datacube. The archive comprises over one million timesteps at temporal resolutions from 15 to 5 minutes, covering a $1200\times1400$ kilometer domain at 1 kilometer spatial resolution, compressed from 7TB to 51GB on disk. We address the historical fragmentation of Italian radar data - previously scattered across heterogeneous formats (OPERA BUFR, HDF5, GeoTIFF) with varying spatial domains and projections - by reprocessing the entire record into a unified store. The dataset is accessible as a static versioned snapshot on Zenodo, via cloud-native access on the ECMWF European Weather Cloud, and as a continuously updated live version on the ArcoDataHub platform. This release fills a significant gap in European radar data availability, as Italy does not participate in the EUMETNET OPERA pan-European radar composite. The dataset is released under a CC BY-SA 4.0 license.

physics.ao-ph↗

arXiv2025

DANRA: The Kilometer-Scale Danish Regional Atmospheric Reanalysis

The DANish regional atmospheric ReAnalysis (DANRA) is a novel high-resolution (2.5 km) reanalysis dataset covering Denmark and its surrounding regions over a 34-year period (1990-2023). Denmark's complex coastline, with over 400 islands and an extensive 7,400 km coastline, means that most municipalities experience mixed land-sea variability. This complexity requires a regional climate reanalysis that can resolve fine-scale coastal and inland features, as well as their impact on climate variability. DANRA is based on the HARMONIE-AROME Numerical Weather Prediction (NWP) model and assimilates a comprehensive set of observations, with a particular focus on Denmark. Compared to global reanalyses such as the ECMWF Reanalysis v5 (ERA5), DANRA demonstrates superior performance in representing essential climate variables, including near-surface weather parameters during both extreme and ordinary conditions. We illustrate these improvements in the representation of several extreme weather cases over Denmark, such as the December 1999 hurricane-force storm, the July 2022 national temperature record, and the August 2007 cloudburst in South Jutland. DANRA is made to support climate adaptation, impact modelling, and the training of next-generation data-driven atmospheric forecasting models. DANRA is distributed as Zarr dataset freely accessible from an object store, maximizing its usability for climate adaptation, impact modelling, and data-driven research.

physics.ao-ph↗

arXiv2025

Building Machine Learning Limited Area Models: Kilometer-Scale Weather Forecasting in Realistic Settings

Machine learning is revolutionizing global weather forecasting, with models that efficiently produce highly accurate forecasts. Apart from global forecasting there is also a large value in high-resolution regional weather forecasts, focusing on accurate simulations of the atmosphere for a limited area. Initial attempts have been made to use machine learning for such limited area scenarios, but these experiments do not consider realistic forecasting settings and do not investigate the many design choices involved. We present a framework for building kilometer-scale machine learning limited area models with boundary conditions imposed through a flexible boundary forcing method. This enables boundary conditions defined either from reanalysis or operational forecast data. Our approach employs specialized graph constructions with rectangular and triangular meshes, along with multi-step rollout training strategies to improve temporal consistency. We perform systematic evaluation of different design choices, including the boundary width, graph construction and boundary forcing integration. Models are evaluated across both a Danish and a Swiss domain, two regions that exhibit different orographical characteristics. Verification is performed against both gridded analysis data and in-situ observations, including a case study for the storm Ciara in February 2020. Both models achieve skillful predictions across a wide range of variables, with our Swiss model outperforming the numerical weather prediction baseline for key surface variables. With their substantially lower computational cost, our findings demonstrate great potential for machine learning limited area models in the future of regional weather forecasting.

physics.ao-ph↗

arXiv2025

Multi-field Visualization: Trait design and trait-induced merge trees

Feature level sets (FLS) have shown significant potential in the analysis of multi-field data by using traits defined in attribute space to specify features in the domain. In this work, we address key challenges in the practical use of FLS: trait design and feature selection for rendering. To simplify trait design, we propose a Cartesian decomposition of traits into simpler components, making the process more intuitive and computationally efficient. Additionally, we utilize dictionary learning results to automatically suggest point traits. To enhance feature selection, we introduce trait-induced merge trees (TIMTs), a generalization of merge trees for feature level sets, aimed at topologically analyzing tensor fields or general multi-variate data. The leaves in the TIMT represent areas in the input data that are closest to the defined trait, thereby most closely resembling the defined feature. This merge tree provides a hierarchy of features, enabling the querying of the most relevant and persistent features. Our method includes various query techniques for the tree, allowing the highlighting of different aspects. We demonstrate the cross-application capabilities of this approach through five case studies from different domains.

cs.LG↗

arXiv2023

Charting the Realms of Mesoscale Cloud Organisation using Unsupervised Learning

Quantifying the driving mechanisms and effect on Earth's energy budget, of mesoscale shallow cloud organisation, remains difficult. Partly because quantifying the atmosphere's organisational state through objective means remains challenging. We present the first map of the full continuum of convective organisation states by extracting the manifold within an unsupervised neural networks's internal representation. On the manifold distinct organisational regimes, defined in prior work, sit as waymarkers in this continuum. Composition of reanalysis and observations onto the manifold, shows wind-speed and water vapour concentration as key environmental characteristics varying with organisation. We show, for the first time, that mesoscale shallow cloud organisation produces $\pm 1.4\%$ variations in albedo in addition to variations from cloud-fraction changes alone. We further demonstrate how the manifold's continuum representation captures the temporal evolution of organisation. By enabling study of states and transitions in organisation (in simulations and observations) the presented technique paves the way for better representation of shallow clouds in simulations of Earth's future climate.

physics.ao-ph↗

arXiv2021

Characterising the shape, size and orientation of cloud-feeding coherent boundary layer structures

This paper presents two techniques for characterisation of cloud-feeding coherent boundary layer structures through analysis of large-eddy simulations of shallow cumulus clouds, contrasting conditions with and without ambient shear. The first technique is a generalisation of the two-point correlation function where the correlation length-scale as well as orientation can be extracted. The second technique decomposes the vertical transport by coherent structures by the shape, size and orientation of these structures. It is found that the structures dominating the vertical flux are plume-like in character (extending from the surface into cloud), show small width/thickness asymmetry and rise near-vertically in the absence of ambient wind. The planar stretching and tilting of boundary layer structures caused by the introduction of ambient shear is also quantified, demonstrating the general applicability of the techniques for future study of other boundary layer patterns.

physics.ao-ph↗