Search

Statistical post-processing of global ensemble weather forecasts is revisited by leveraging recent developments in machine learning. Verification of past forecasts is exploited to learn systematic deficiencies of numerical weather predictions in order to boost post-processed forecast performance. Here, we introduce PoET, a post-processing approach based on hierarchical transformers. PoET has 2 major characteristics: 1) the post-processing is applied directly to the ensemble members rather than to a predictive distribution or a functional of it, and 2) the method is ensemble-size agnostic in the sense that the number of ensemble members in training and inference mode can differ. The PoET output is a set of calibrated members that has the same size as the original ensemble but with improved reliability. Performance assessments show that PoET can bring up to 20% improvement in skill globally for 2m temperature and 2% for precipitation forecasts and outperforms the simpler statistical member-by-member method, used here as a competitive benchmark. PoET is also applied to the ENS10 benchmark dataset for ensemble post-processing and provides better results when compared to other deep learning solutions that are evaluated for most parameters. Furthermore, because each ensemble member is calibrated separately, downstream applications should directly benefit from the improvement made on the ensemble forecast with post-processing.

Wind farm modelling has been an area of rapidly increasing interest with numerous analytical as well as computational-based approaches developed to extend the margins of wind farm efficiency and maximise power production. In this work, we present the novel ML framework WakeNet, which can reproduce generalised 2D turbine wake velocity fields at hub-height over a wide range of yaw angles, wind speeds and turbulence intensities (TIs), with a mean accuracy of 99.8% compared to the solution calculated using the state-of-the-art wind farm modelling software FLORIS. As the generation of sufficient high-fidelity data for network training purposes can be cost-prohibitive, the utility of multi-fidelity transfer learning has also been investigated. Specifically, a network pre-trained on the low-fidelity Gaussian wake model is fine-tuned in order to obtain accurate wake results for the mid-fidelity Curl wake model. The robustness and overall performance of WakeNet on various wake steering control and layout optimisation scenarios has been validated through power-gain heatmaps, obtaining at least 90% of the power gained through optimisation performed with FLORIS directly. We also demonstrate that when utilising the Curl model, WakeNet is able to provide similar power gains to FLORIS, two orders of magnitude faster (e.g. 10 minutes vs 36 hours per optimisation case). The wake evaluation time of wakeNet when trained on a high-fidelity CFD dataset is expected to be similar, thus further increasing computational time gains. These promising results show that generalised wake modelling with ML tools can be accurate enough to contribute towards active yaw and layout optimisation, while producing realistic optimised configurations at a fraction of the computational cost, hence making it feasible to perform real-time active yaw control as well as robust optimisation under uncertainty., Comment: 16 Pages, 18 Figures, 3 Tables

Throughout history, coastal zones have been vulnerable to the dual risks of erosion and flooding. With climate change likely to exacerbate these risks in the coming decades, coasts are becoming an ever more critical location on which to apply hydro-morphodynamic models blended with advanced numerical and statistical techniques, to assess risk. We implement a novel depth-averaged hydro-morphodynamic model using a discontinuous Galerkin based finite element discretisation within the coastal ocean model {\em Thetis}. Our model is the first with this discretisation to simulate both bedload and suspended sediment transport, and is validated for test cases in fully wet and wet-dry domains. These test cases show our model is more accurate, efficient and robust than industry-standard models. Additionally, we use our model to implement the first fully flexible and freely available adjoint hydro-morphodynamic model framework which we then successfully use for sensitivity analysis, inversion and calibration of uncertain parameters. Furthermore, we implement the first moving mesh framework with a depth-averaged hydro-morphodynamic model, and show that mesh movement can help resolve the multi-scale issues often present in hydro-morphodynamic problems, improving their accuracy and efficiency. We present the first application of the multilevel Monte Carlo method (MLMC) and multilevel multifidelity Monte Carlo method (MLMF) to industry-standard hydro-morphodynamic models as a tool to quantify uncertainty in erosion and flood risk. We use these methods to estimate expected values and cumulative distributions of variables which are of interest to decision makers. MLMC, and more notably MLMF, significantly reduce computational cost compared to the standard Monte Carlo method whilst retaining the same level of accuracy, enabling in-depth statistical analysis of complex test cases that was previously unfeasible. The comprehensive toolkit of techniques we develop provides a crucial foundation for researchers and stakeholders seeking to assess and mitigate coastal risks in an accurate and efficient manner. Open Access