Abstract
The use of analog-similar weather patterns for weather forecasting and analysis is an established method in meteorology. The most challenging aspect of using this approach in the context of operational radar applications is to be able to perform a fast and accurate search for similar spatiotemporal precipitation patterns in a large archive of historical records. In this context, sequential pairwise search is too slow and computationally expensive. Here, we propose an architecture to significantly speed up spatiotemporal analog retrieval by combining nonlinear geometric dimensionality reduction (UMAP) with the fastest known Euclidean search algorithm for time series (MASS) to find radar analogs in constant time, independently of the desired temporal length to match and the number of extracted analogs. We show that UMAP, combined with a grid search protocol over relevant hyperparameters, can find analog sequences with lower mean square error (MSE) than principal component analysis (PCA). Moreover, we show that MASS is 20 times faster than brute force search on the UMAP embedding space. We test the architecture on real dataset and show that it enables precise and fast operational analog ensemble search through more than 2 years of radar archive in less than 3 seconds on a single workstation.
Highlights
The observation of repeating weather patterns has a long history [1], and the use of analogs has found its way in almost all aspect of meteorology, for the most diverse purposes
We evaluated the solution by comparing the straight cumulative mean square error (MSE) error between the sequences found by Mueen’s Algorithm for Similarity Search (MASS)-Uniform Manifold Approximation and Projection [17] (UMAP) and MASS-principal component analysis (PCA) and the sequences retrieved by MSE
We investigated some of the manifolds generated by UMAP projections and plotted the resulting embeddings for the search and the verification sets (Figure 4)
Summary
The observation of repeating weather patterns has a long history [1], and the use of analogs has found its way in almost all aspect of meteorology, for the most diverse purposes. Approaches based on analogs have been proposed for the postprocessing of numerical weather predictions [2], as a statistical downscaling technique [3] and for data assimilation in numerical models [4,5]. The most prolific use of analogs is by far forecasting: either as a proxy for predictability [6], or as prediction technique itself [7]. In this regard, one of the most used operational methods for analog-based forecasting is Analog Ensemble (AnEn) [8], which involves searching and using an ensemble of past analogs to generate new deterministic or probabilistic [8] predictions. In this paper we present a novel search method for spatio-temporal sequences and show how it meets many of these desirable qualities
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.