Mamba Wind Time Series Representation: PCA and Mode Analysis Across Multiple Airports

Read full text Summarise with VIRAT
Authors: K.D. Pom
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This dataset and accompanying analysis documents the Mamba model representation of wind time series at three airports: Queen Beatrix Intl (TNCA, Aruba), Incheon Intl (RKSI, South Korea), and Duluth Intl (US, WBAN 14913). Mamba readouts (y_t) were reduced via PCA, clustered with KMeans, and analyzed through: PCA scatter plots Windrose visualizations Mode statistics (speed and direction) State transition matrices The dataset includes raw wind measurements (u, v, wspd, wdir), Mamba representations (Y), and metadata. This work demonstrates interpretable low-dimensional representations of wind dynamics, highlighting physical modes and transition dynamics. Data were obtained from the NOAA Local Climatological Data (LCD) archive (2025). All analyses, figures, and derived datasets are included for reproducibility. Keywords Mamba Wind Time Series PCA KMeans Windrose Mode Analysis Low-Dimensional Representation NOAA LCD

Fields of Study & Subjects

Meteorological Phenomena and Simulations Aerospace and Aviation Technology Time Series Analysis and Forecasting

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
20 March 2026
Language
EN
Document type
Dataset