We investigate how to make statistical predictions of extreme weather such that events predicted to occur with a probability of 1 % will occur 1 % of the time. We apply the methods we describe to a standard extreme weather attribution example from the recent climate literature. We find that the methods we describe imply that extremes are roughly twice as likely as when estimated using maximum likelihood. We have developed a software package to make it easy to apply these methods.

Correctly forecasting weather is crucial for decision-making in various fields. Standard multivariate verification tools have limitations, and a single tool cannot fully characterize predictive performance. We formalize a framework based on aggregation and transformation to build interpretable verification tools. These tools target specific features of forecasts, improving predictive performance characterization and bridging the gap between theoretical and physics-based tools.

This paper assesses the skill of probabilistic forecasts of solar irradiance in the northern regions of Chile. Raw ensemble forecast are calibrated using a parametric and a novel non-parametric machine-learning-based method. As the reference approach, the ensemble model output statistics are considered. We verify the superiority of the proposed non-parametric neural-network-based ensemble correction, resulting in more than 50 % improvement in prediction performance compared to the raw forecasts.

We employ hierarchical statistical models to investigate spatiotemporal differences between sea surface temperature, marine air temperature, and their anomalies in the tropical Pacific.

This paper presents a novel study on boxplot-valued data in climatological applications. Our methodologies are applied to the Berkeley Earth Surface Temperature Study. We validate our approaches through comprehensive simulations, comparing Bayesian and frequentist estimators for efficiency and accuracy. The results provide robust insights into climatic trends, particularly summer average temperatures across European countries. 

Southeasterm Australia, including the Murray–Darling Basin, is a highly productive agricultural region largely dependent on adequate rainfall, providing irrigation water needed for crops.

The Australian Bureau of Meteorology uses linear methods and provides seasonal forecasts expressed as the probability of exceeding median rainfall. I use expanded methods, including more ocean data and deep learning neural networks that provide nonlinear estimates of the rainfall as measured by rain gauges.

Drought frequently affects southern Ethiopia, and inadequate rainfall data complicate monitoring. Satellite data from three sources were evaluated for accuracy, showing the Climate Hazards Group InfraRed Precipitation with Stations (CHIRPS) to be the most reliable. We assessed drought severity using the standardized precipitation index, aligning results with historical drought patterns. The study emphasizes that the CHIRPS enhances drought monitoring and early-warning systems in the region.

Meteorological data recorded at high frequency by automatic sensors are often marred by multiple forms of error. Existing validation techniques, in isolation, are sub-optimal for such error-prone data. We propose a new data-standardization procedure for the validation of strongly correlated series which commonly arise in meteorology. We show the procedure to be more comprehensive in the simultaneous detection of solitary spikes, shifts in series means, and irregular diurnal patterns.

Weather generators efficiently create realistic weather data based on historical records. This study introduces a daily temperature generator for large regions, separating deterministic factors (trends, seasonality) from random variations modeled using space-time interactions. Validated on French weather station data, it replicates observed patterns, including heatwaves. It offers a practical solution for generating realistic weather data, for applications such as climate impact assessments.

This study investigates soil moisture–temperature coupling during the extreme warm conditions in May–August 2018 in southern and central Sweden using the merged GLEAM-E-OBS dataset and four simulations from the Weather Research and Forecasting model coupled with the Community Terrestrial Systems Model (WRF-CTSM). Based on changes in surface soil moisture, evaporative fraction, and daily maximum 2 m temperature, on average across the region and five datasets, the coupling lasted for 22 d.

We develop a spatial statistical calibration of wind gust observations for the region of Germany with an interpolation to unobserved locations. Furthermore, the model is spatially adaptive and includes the station altitude both as explanatory variable and as offset to increase the distance between stations. This offset allows us to include mountain stations into the training data. Compared to a spatially constant model, the adaptive model improves the representation of extreme wind gusts.