arXiv subjects

Gregor Skok

Publications and source records attributed to Gregor Skok.

6 recordsLinked to original sources

arXiv2026

WP-MIP: An Artificial Intelligence, Hybrid, and Physically Based Model Intercomparison Project for Weather Prediction

Rapid progress in the field of machine-learning for weather prediction has led to the emergence of algorithms whose forecasting skill can exceed that of traditional physically based models. This development represents an opportunity to improve the quality of forecasting services provided by operational centers, particularly given the speed at which machine-learning based models generate predictions. Despite the clear promise of these systems, questions remain about the ability of the current generation of machine-learning models to generate physically consistent predictions of the full suite of required forecast fields under all conditions. Answering these questions will require careful comparisons between the well-understood physically based models, current state-of-the-art machine-learning models, and the hybrid models that combine elements of these two archetypes. The Weather Prediction Model Intercomparison Project (WP-MIP) is a World Meteorological Organization-supported initiative whose initial goal is to create a centralized database of physically based, machine-learning, and hybrid model forecasts to enable a distributed assessment and evaluation effort. The first instance of WP-MIP focuses on global deterministic predictions using both center-specific and common initializations to facilitate sensitivity studies. Forecasts contributed by institutions across six continents will be used to develop AI-ready verification techniques that highlight the strengths and weaknesses of each class of prediction system, with the goal of establishing best-practice guidance to model developers and national weather centers. The broad engagement of the operational and forecast-evaluation communities in WP-MIP will ensure that the project results are highly relevant to the development and deployment of next-generation weather prediction systems.

physics.ao-ph↗

arXiv2025

A Neural-Network Model-Measurement-Based Observation Operator For Weather Radar Reflectivity Assimilation

In three-dimensional variational data assimilation (3DVar) for numerical weather prediction (NWP), the observation operator $\mathcal{H}$ plays a central role by mapping model state variables to an observation equivalent. For weather radar, however, specifying $\mathcal{H}$ is particularly challenging: reflectivity is a nonlinear, microphysics-dependent diagnostic quantity that only indirectly relates to the model's prognostic variables, making traditional parameterised radar operators complex, regime-dependent and difficult to tune. In this study, we propose a neural-network (NN)-based observation operator for radar reflectivity and apply it within a 3DVar framework. Using five years (2019-2023) of radar reflectivity data from the Lisca radar and 4.4 km-resolution short-range forecasts from ALADIN model over Slovenia, we train a convolutional encoder-decoder neural network to map model temperature, humidity, horizontal wind components and surface pressure fields to radar reflectivity. Across independent test cases spanning clear-sky, stratiform, and convective regimes, the NN-based operator accurately reproduces the spatial structure and intensity of observed reflectivity, relying primarily on the model state near the observation point. In the extreme precipitation case, which caused widespread floods in Slovenia on August 4, 2023, assimilating the full radar disc reduces the domain-averaged reflectivity root-mean-square error from 5.99 dBZ to 3.47 dBZ and improves the alignment between the analysed and observed convective bands. Embedded within 3DVar, the Jacobian of the NN observation operator allows radar reflectivity observations to inform model state variables, producing corresponding analysis increments. The proposed NN radar observation operator offers a flexible alternative to traditional parameterised radar operators for improving convective-storm forecasts.

physics.ao-ph↗

arXiv2025

Approximating the universal thermal climate index using sparse regression with orthogonal polynomials

The Universal Thermal Climate Index (UTCI) is a measure of thermal comfort that quantifies how humans experience environmental conditions. Due to its robustness and versatility as a bioclimatic indicator, it has been extensively employed across a wide range of studies in bioclimatology and is increasingly used as an operational measure of outdoor thermal comfort. Calculating the UTCI value from the relevant environmental parameters is nominally not straightforward, which is why using a 6th-degree polynomial approximation has become the standard way to calculate UTCI values. Although it is computationally efficient, the error of this polynomial approximation can be substantial. The goal of this study was to develop an improved version of the polynomial approximation - one that retains comparable computational efficiency but is more robust in terms of numerical stability and substantially more accurate, particularly in reducing the frequency of larger errors. This goal was achieved using sparse orthogonal regression, namely sparse regression with an orthogonal polynomial basis, which not only substantially reduces the average errors (i.e., the mean error, the mean absolute error, and the root mean square error) but also drastically reduces the frequency of large errors. By leveraging Legendre polynomial bases, approximation models could be constructed that efficiently populate a Pareto front of accuracy versus complexity and exhibit stable, hierarchical coefficient structures across varying model capacities. Training the new approximation models over only 20% of the data, with the testing performed over the remaining 80%, highlights successful generalization, with the results being robust under bootstrapping. The decomposition effectively approximates the UTCI as a Fourier-like expansion in an orthogonal basis, yielding results near the theoretical optimum in the L2 (least squares) sense.

physics.ao-ph↗

arXiv2025

Forecast error diagnostics in neural weather models

Deep-learning (DL) weather prediction models offer some notable advantages over traditional physics-based models, including auto-differentiability and low computational cost, enabling detailed diagnostics of forecast errors. Using our convolutional encoder-decoder model, ConvCastNet, we systematically relax selected subdomains of the forecast fields towards "true" weather states (ERA5 reanalyses) and monitor the forecast skill gain in other regions. Our results show that a medium-range mid-latitude forecast improves substantially when the stratosphere and boundary layer are relaxed, while relaxation of the tropical atmosphere has a negligible effect. This underscores the need for a more accurate representation of the stratosphere and the planetary boundary layer to improve medium-range weather predictability. Additionally, we investigate the relationship between the forecast error sensitivity to initial conditions and relaxation experiments. By utilising auto-differentiability, we identify overlapping regions of large error sensitivity and strong forecast skill improvement from relaxation. Average mid-latitude error sensitivity to initial conditions shows negligible influence from the tropics, corroborating the results of the tropical relaxation experiments. The error sensitivity shows a physically consistent influence of upstream weather dynamics and sea surface temperatures on forecast accuracy. The latter also highlights the importance of accurately representing the atmosphere--ocean coupling in numerical weather prediction models. This combined approach could provide valuable heuristics for diagnosing neural model errors and guiding targeted model improvements.

physics.ao-ph↗

arXiv2024

Smoothing and spatial verification of global fields

Forecast verification plays a crucial role in the development cycle of operational numerical weather prediction models. At the same time, verification remains a challenge as the traditionally used non-spatial forecast quality metrics exhibit certain drawbacks, with new spatial metrics being developed to address these problems. Some of these new metrics are based on smoothing, with one example being the widely used Fraction Skill Score (FSS) and its many derivatives. However, while the FSS has been used by many researchers in limited area domains, there are no examples of it being used in a global domain yet. The issue is due to the increased computational complexity of smoothing in a global domain, with its inherent spherical geometry and non-equidistant and/or irregular grids. At the same time, there clearly exists a need for spatial metrics that could be used in the global domain as the operational global models continue to be developed and improved, along with the new machine-learning-based models. Here, we present two new methodologies for smoothing in a global domain that are potentially fast enough to make the smoothing of high-resolution global fields feasible. Both approaches also consider the variability of grid point area sizes and can handle missing data appropriately. This, in turn, makes the calculation of smoothing-based metrics, such as FSS and its derivatives, in a global domain possible, which we demonstrate by evaluating the performance of operational high-resolution global precipitation forecasts provided by the European Centre for Medium-Range Weather Forecasts.

physics.ao-ph↗

arXiv2024

Spatial verification of global precipitation forecasts

Verification of global high-resolution precipitation forecasts is challenging. Spatial verification techniques address some shortcomings of traditional verification. However most existing methods do not account for the non-planar geometry of a global domain, or their computational complexity is too large for global assessments. We present an adaptation of the recently developed Precipitation Attribution Distance (PAD) metric, designed for verifying precipitation, enabling its use on the Earth's spherical geometry. PAD estimates the magnitude of location errors in the forecasts employing the mathematical theory of Optimal Transport, as it provides a close upper bound for the Wasserstein distance. The method is fast and flexible with time complexity $O(n \log(n))$. Its behavior is analyzed using idealized cases and 7 years of operational global deterministic 6-hourly precipitation forecasts from the Integrated Forecasting System (IFS) of the European Centre for Medium-Range Weather Forecasts. The summary results for the whole period show how location errors in the IFS model grow steadily with increasing lead time for all analyzed regions. Moreover, by examining the time-series of the results, we can determine the trends in the score's value and identify the regions where there is a statistically significant improvement (or worsening) of the forecast performance. The results can also be analyzed separately for different intensities of precipitation. Overall, the PAD provides meaningful results for estimating location errors in global high-resolution precipitation forecasts at an affordable computational cost.

physics.ao-ph↗