We analyze the applicability of convolutional neural network (CNN) architectures for downscaling of short‐range forecasts of near‐surface winds on extended spatial domains. Short‐range wind forecasts (at the 100 m level) from European Centre for Medium Range Weather Forecasts ERA5 reanalysis initial conditions at 31 km horizontal resolution are downscaled to mimic high resolution (HRES) (deterministic) short‐range forecasts at 9 km resolution. We evaluate the downscaling quality of four exemplary CNN architectures and compare these against a multilinear regression model. We conduct a qualitative and quantitative comparison of model predictions and examine whether the predictive skill of CNNs can be enhanced by incorporating additional atmospheric variables, such as geopotential height and forecast surface roughness, or static high‐resolution fields, like land–sea mask and topography. We further propose DeepRU, a novel U‐Net‐based CNN architecture, which is able to infer situation‐dependent wind structures that cannot be reconstructed by other models. Inferring a target 9 km resolution wind field from the low‐resolution input fields over the Alpine area takes less than 10 ms on our graphics processing unit target architecture, which compares favorably to an overhead in simulation time of minutes or hours between low‐ and high‐resolution forecast simulations.
4 December 2017 in IEEE Transactions on Visualization and Computer Graphics
This article surveys the history and current state of the art of visualization in meteorology, focusing on visualization techniques and tools used for meteorological data analysis. We examine characteristics of meteorological data and analysis tasks, describe the development of computer graphics methods for visualization in meteorology from the 1960s to today, and visit the state of the art of visualization techniques and tools in operational weather forecasting and atmospheric research. We approach the topic from both the visualization and the meteorological side, showing visualization techniques commonly used in meteorological practice, and surveying recent studies in visualization research aimed at meteorological applications. Our overview covers visualization techniques from the fields of display design, 3D visualization, flow dynamics, feature-based visualization, comparative visualization and data fusion, uncertainty and ensemble visualization, interactive visual analysis, efficient rendering, and scalability and reproducibility. We discuss demands and challenges for visualization research targeting meteorological data analysis, highlighting aspects in demonstration of benefit, interactive visual analysis, seamless visualization, ensemble visualization, 3D visualization, and technical issues.
Despite the potential of neural scene representations to effectively compress 3D scalar fields at high reconstruction quality, the computational complexity of the training and data reconstruction step using scene representation networks limits their use in practical applications. In this paper, we analyse whether scene representation networks can be modified to reduce these limitations and whether such architectures can also be used for temporal reconstruction tasks. We propose a novel design of scene representation networks using GPU tensor cores to integrate the reconstruction seamlessly into on‐chip raytracing kernels, and compare the quality and performance of this network to alternative network‐ and non‐network‐based compression schemes. The results indicate competitive quality of our design at high compression rates, and significantly faster decoding times and lower memory consumption during data reconstruction. We investigate how density gradients can be computed using the network and show an extension where density, gradient and curvature are predicted jointly. As an alternative to spatial super‐resolution approaches for time‐varying fields, we propose a solution that builds upon latent‐space interpolation to enable random access reconstruction at arbitrary granularity. We summarize our findings in the form of an assessment of the strengths and limitations of scene representation networks for compression domain volume rendering, and outline future research directions. Source code: https://github.com/shamanDevel/fV‐SRN
Radio interferometers do not measure the sky brightness distribution directly, but measure a modified Fourier transform of it. Imaging algorithms therefore need a computational representation of the linear measurement operator and its adjoint, regardless of the specific chosen imaging algorithm. In this paper, we present a C++ implementation of the radio interferometric measurement operator for wide-field measurements that is based on so-called improved w-stacking. It can provide high accuracy (down to ≈10−12), is based on a new gridding kernel that allows smaller kernel support for given accuracy, dynamically chooses kernel, kernel support, and oversampling factor for maximum performance, uses piece-wise polynomial approximation for cheap evaluations of the gridding kernel, treats the visibilities in cache-friendly order, uses explicit vectorisation if available, and comes with a parallelisation scheme that scales well also in the adjoint direction (which is a problem for many previous implementations). The implementation has a small memory footprint in the sense that temporary internal data structures are much smaller than the respective input and output data, allowing in-memory processing of data sets that needed to be read from disk or distributed across several compute nodes before.
CLEAN, the commonly employed imaging algorithm in radio interferometry, suffers from a number of shortcomings: In its basic version, it does not have the concept of diffuse flux, and the common practice of convolving the CLEAN components with the CLEAN beam erases the potential for super-resolution; it does not output uncertainty information; it produces images with unphysical negative flux regions; and its results are highly dependent on the so-called weighting scheme as well as on any human choice of CLEAN masks for guiding the imaging. Here, we present the Bayesian imaging algorithm resolve , which solves the above problems and naturally leads to super-resolution. We take a VLA observation of Cygnus A at four different frequencies and image it with single-scale CLEAN, multi-scale CLEAN, and resolve. Alongside the sky brightness distribution, resolve estimates a baseline-dependent correction function for the noise budget, the Bayesian equivalent of a weighting scheme. We report noise correction factors between 0.4 and 429. The enhancements achieved by resolve come at the cost of higher computational effort.