<div> <p><span data-contrast="auto">Forecasting winds at the local scale can be challenging due to the highly variable and complex nature of wind patterns, particularly in the case of complex terrain. In such cases, the accuracy of numerical weather prediction models (NWPs) is often limited by the quality of their initial conditions and their grid resolution. This is where the use of observational data through statistical postprocessing techniques can help to improve the quality of forecasts.</span><span data-ccp-props="{&quot;134233117&quot;:false,&quot;134233118&quot;:false,&quot;201341983&quot;:0,&quot;335551550&quot;:1,&quot;335551620&quot;:1,&quot;335559685&quot;:0,&quot;335559737&quot;:0,&quot;335559738&quot;:0,&quot;335559739&quot;:160,&quot;335559740&quot;:259}">&#160;</span></p> </div> <div> <p><span data-contrast="auto">Statistical postprocessing is nowadays an established component in operational weather forecasting that is used to improve the accuracy, resolution, and calibration of NWP ensemble forecasts with historical observations. In recent years, machine learning techniques have shown great potential in the field of postprocessing, thanks to their ability to deal with increasingly large volumes of data, and the capacity to capture complex relationships between forecasts and observations that are not explicitly represented in traditional postprocessing methods.</span><span data-ccp-props="{&quot;201341983&quot;:0,&quot;335559739&quot;:160,&quot;335559740&quot;:259}">&#160;</span></p> </div> <div> <p><span data-contrast="auto">To capitalize on machine learning for weather applications, and for it to gain acceptance and become a reliable technology for operational use, it is also crucial to consider the technical and engineering challenges that arise when implementing machine learning in a productive environment. MLOps, or Machine Learning Operations, is a set of practices that are used to manage and streamline the deployment, monitoring, and maintenance of machine learning models in production.&#160;</span><span data-ccp-props="{&quot;201341983&quot;:0,&quot;335559739&quot;:160,&quot;335559740&quot;:259}">&#160;</span></p> </div> <div> <p><span data-contrast="auto">We will present our recent experience with the development and operationalization of a statistical postprocessing system based on the use of neural networks to predict the probability distribution of forecasts of surface winds. Following MLOps best practices, our framework aims to improve the reproducibility and automation of most common tasks in a machine learning-based system, such as efficient data loading and manipulation, the monitoring and visualization of prediction quality, and the automation of model training and deployment pipelines.</span><span data-ccp-props="{&quot;201341983&quot;:0,&quot;335559739&quot;:160,&quot;335559740&quot;:259}">&#160;</span></p> </div>
<div> <p><span>Correctly representing surface wind is critical for applications such as renewable energy, snow modelling or warning systems. However, numerical weather prediction models with their limited resolution cannot fully represent the strong variability due to complex topography. Downscaling techniques &#8211; functionally equivalent to postprocessing when the ground truth is given by observational data - can achieve remarkable results in reducing systematic biases of raw models and can be calibrated to yield accurate probabilistic information at any point in space.</span><span>&#160;</span></p> </div><div> <p><span>These techniques can be further improved at analysis time by including real-time measurements, allowing to produce a probabilistic sub-grid resolution analysis of surface wind. Such a product would enable other interesting applications, such as detailed climatologies or nowcasting, and could serve as a ground truth for training deep learning-based postprocessing models with generative approaches, allowing to model spatially and temporally consistent ensembles.&#160;</span><span>&#160;</span></p> </div><div> <p><span>The first important challenge is to integrate measurements in a statistically optimized and efficient way. Here, we share our ongoing work and preliminary results in a comparative analysis of different approaches, from na&#239;ve interpolations to geostatistical techniques or novel approaches based on neural networks. The analysis is based on a multi-year archive of hourly wind observations and NWP analyses from the operational COSMO-1E model over Switzerland.</span><span>&#160;</span></p> </div>
<p>Traditional post-processing methods aim at minimizing forecast error. This often leads to predictions that violate physical principles and disregard dependencies between variables. However, for various impact-based applications such as hydrological forecasting or heat indices, it is important to provide forecasts that not only have high univariate accuracy, but also are physically consistent, in the sense of respecting physical principles and variable dependencies. Achieving physical consistency remains an open problem in the post-processing of weather forecasts, while this question has recently gained a lot of attention in the wider deep learning community and climate field. Recent contributions show that physical consistency may be pursued by applying different forms of constraints to deep learning models. The most widely used approaches are to incorporate physics via regularization, by defining physics-based losses in addition to common metrics such as mean absolute error, or to define custom-designed model architectures, such that the physical constraints are strictly enforced. Including constraints also has the potential to help the training procedure by restraining the hypothesis space of the model and improving generalization capabilities.</p><p>This work investigates the application of the aforementioned approaches for the postprocessing of a set of variables related to surface temperature and humidity, specifically temperature, dew point, surface pressure, relative humidity and water vapor mixing ratio. As baseline, we use an unconstrained fully connected neural network. We consider the simple case of postprocessing at a single location, and we show how it is possible to incorporate domain knowledge, specifically thermodynamic relationships, via analytic constraints, to obtain physically consistent postprocessed prediction. We compare different approaches and show that we can enforce physical consistency without degrading performance, or even improving it. Furthermore, we discuss additional advantages and disadvantages of these approaches in the context of post-processing, besides error reduction.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.