Model projections of changes in precipitation express large uncertainties, partly owing to climate noise from internal variability dominating the greenhouse gas forced signal. In a high-resolution initial-condition regional climate model ensemble, we investigate this signal-to-noise problem for precipitation changes in major river basins in Western Europe, focusing on how it depends on different methods of spatial aggregation. This is motivated by the common need for spatially aggregated climate change information. We find that the forced climate signal is only moderately dependent on spatial aggregation. However, uncertainties (and likewise the signal-to-noise ratio) strongly depend on the aggregation method. Computing uncertainties at the grid scale – as typically provided in maps in climate change studies – and averaging those to larger areas, leads to large uncertainty estimates and corresponding low signal-to-noise ratios. This method does not account for spatiotemporal dependencies in the underlying data, and is representative for the uncertainties at a specific location. Computing changes at the grid scale, spatially averaging those per ensemble member, and computing the uncertainty in these spatial averages results in much smaller uncertainty estimates, up to a factor 2 for 1-day precipitation maxima in summer. This information is useful for stakeholders with an aggregated uncertainty perspective, e.g. insurance. Spatial aggregation of precipitation itself and computing uncertainties in basin-scale precipitation changes – relevant for e.g. river flooding – also yields smaller uncertainties, up to a factor 1.5 for 1-day summer maxima. This paper is intended to give guidance on processing uncertainty information depending on the application-specific needs.

To simulate the elastic wave propagations in 3-dimensional (3D) anisotropic media, a 3D anisotropic lattice model is proposed. The derivation of the general relationship between the anisotropic lattice model and the anisotropic continuum model is presented. With this lattice model, the dynamic lattice method (DLM) is able to simulate elastic wave propagation in arbitrary anisotropic media. Numerical tests on different anisotropic models show that the elastic waves simulated by the DLM coincide with those simulated by the finite difference method, indicating that the dynamic lattice method represents a lattice-based choice for the elastic wave simulation in complex anisotropic media. Our research demonstrates the robustness of the anisotropic lattice model in describing anisotropic media, and lays the foundation for lattice-based methods in terms of elastic wave modeling in 3D anisotropic media.

Ribbon collision is a process that can rapidly disturb the symmetry of subduction zones. Previous studies have demonstrated how ribbon collision causes rotation at the surface and contortion in the slab, but have focused on the surface kinematics. We use three-dimensional mechanical models to investigate how the dynamic evolution of ribbon collision perturbs the strain and stress field at the surface, the deformation style in the slab and the force balance in intra-plate regions. In our numerical simulations, we vary the angle between the trench and the ribbon to explore the slab response to ribbons colliding at different orientations. Our numerical simulations show that ribbon collision causes significant heterogeneity of stress, strain rate and vorticity in both the overriding and subducting plates surface and the slab. Slab deformation shows compartmentalization into low and high strain-rate regions around a high vorticity zone, with strain-rate variations of up to an order of magnitude occurring in the along-strike and down-dip directions. In the context of our idealised oceanic-continental subduction system, the simulations show that intra-plate stresses are affected to a similar degree by buoyancy contrasts (i.e. gravitational potential energy variations), slab-pull and ribbon collision. This partitioning allows for significant heterogeneity in the intra-plate stress regime. This work highlights how the rapid changes in strain-rate within the slab, caused by ribbon collision, can explain the seismicity gaps observed in collisional margins, which are often interpreted as slab-tears.

The application of a new approach to bias-correct Earth System Model (ESM) driving data for regional Climate Model (RCM) downscaling is presented. This approach employs a novel Empirical Runtime Bias Correction (ERBC) of the ESM, designed to self-consistently reduce climatological biases in the driving data. The impact of such ESM bias reduction on RCM downscaling is evaluated through an experimental protocol where a single ESM and its ERBC counterpart drive two different RCMs. A continental-scale analysis of these results in a North American regional domain over the historical period indicates that the impact of global model biases on RCM downscaling products can be mitigated significantly by employing ERBC driving data. A similar series of ESM downscaling simulations is conducted for future projections of climate change following the Coupled Model Intercomparison Project phase 6 SSP3-7.0 scenario of anthropogenic forcings. Unlike diagnostic bias corrections applied to model output, ERBCs have the potential to improve climate-change circulation responses, thereby reducing uncertainty. This reduction would be reflected in a narrower spread of responses within a multi-model ensemble of ESMs and RCMs. Since this study involves only one ESM, we investigate the necessary but not sufficient condition for uncertainty reduction: that ERBCs can alter climate change responses compared to their uncorrected counterparts. The results show that ERBCs induce statistically significant changes in the climate-change circulation responses in both the global and regional models.

Accurate forecasts on subseasonal (S2S) timescales are essential for the preparation and mitigation of the impacts of high-impact events, such as flash droughts. To improve the accuracy of soil moisture forecasts —a critical factor in identifying flash droughts— we present a hybrid modeling framework that combines dynamical forecasts from the European Centre for Medium-Range Weather Forecasts (ECMWF) with deep learning (DL) models. This approach not only corrects biases in numerical weather prediction models but also improves spatial resolution, increasing the accuracy of S2S forecasts. By using deterministic inputs, such as the ensemble mean and spread, we further assess the uncertainty of forecasts through dropout neural networks via Monte Carlo sampling. Our results demonstrate that the DL models outperform baseline methods, offering skillful S2S forecasts of soil moisture. This advanced hybrid framework provides more accurate soil moisture predictions, ultimately supporting improved strategies for managing and mitigating the impacts of flash droughts.

Heterogeneity in geometry, stress, and material properties is widely invoked to explain the observed spectrum of slow earthquake phenomena. However, the effects of length scale of heterogeneity on macroscopic fault sliding behavior remain underexplored. We investigate this question for subduction megathrusts, via linear stability analysis and quasi-dynamic simulations of slip on a dipping fault characterized by rate-and-state friction (RSF). Frictional heterogeneity is imposed through alternating velocity-strengthening (VS) and velocity-weakening (VW) patches, over length scales spanning from those representative of basement relief (several km) to the entrainment of contrasting lithologies (100s of m). The resulting fault behavior is controlled by: (1) the average frictional properties of the fault, and (2) the size of VW blocks relative to a critical length scale. Reasonable ranges of these properties yield sliding behaviors spanning from stable sliding, to slow and seismic slip events that are confined within VW blocks or propagate along the entire fault.