Foehn Wind Analysis using Unsupervised Deep Anomaly Detection

Abstract

<p>Foehn winds are accelerated, warm and dry winds that can have significant environmental impacts as they descend into the lee of a mountain range. For example, in the McMurdo Dry Valleys in Antarctica, foehn events can cause ice and glacial melt and destabilise ice shelves, which if lost, resulting in a rise in sea level. Consequently, there is a strong interest in a deeper understanding of foehn winds and their meteorological signatures. Most current automatic detection methods rely on rule-based methodologies that require static thresholds of meteorological parameters. However, the patterns of foehn winds are hard to define and differ between alpine valleys around the world. Consequently, data-driven solutions might help create more accurate detection and prediction methodologies. </p> <p>State-of-the-art machine learning approaches to this problem have shown promising results but follow a supervised learning paradigm. As such, these approaches require accurate labels, which for the most part, are being created by imprecise static rule-based algorithms. Consequently, the resulting machine-learning models are trained to recognise the same static definitions of the foehn wind signatures. </p> <p>In this paper, we introduce and compare the first unsupervised machine-learning approaches for detecting foehn wind events. We focus on data from the Mc Murdo Dry Valleys as an example, however, due to the unsupervised nature of these approaches, our solutions can recognise a more dynamic definition of foehn wind events and are therefore, independent of the location. The first approach is based on multivariate time-series clustering, while the second utilises a deep autoencoder-based anomaly detection method to identify foehn wind events. Our best model achieves an f1-score of 88%, matching or surpassing previous machine-learning methods while providing a more flexible and inclusive definition of foehn events. </p>

Characteristics and Establishment of Objective Identification Criteria and Predictors for Foehn Winds in Urumqi, China

“…Therefore, only considering potential temperature when identifying foehn can result in apparent deviations. In addition, many scholars have trained machine learning models to predict and identify foehn, but it is difficult to select effective predictors [23,24].…”

Characteristics and Establishment of Objective Identification Standard and Perdictor of Foehn Winds in Urumqi on the North Slope of Middle Tianshan Mountains of China

“…Generally, when the temperature is high, which maybe misidentify the local winds caused by other weather phenomena as foehn [18]; Atkinson used two-element identification method to analyze the foehn occurred in the east of the Rocky Mountains in America: one is that the upper-level wind must have a component perpendicular to the mountains; the other is, the windward side of the mountain shows a high pressure ridge and the lee side has a low pressure trough in the surface weather map [17]; Schuetz et al only used potential temperature as the identification standard, ignoring the element of wind, and resulting in various WD [19]. Therefore, considering potential temperature alone has significant difference to identify the foehn; In addition, many scholars have trained machine learning models to predict and identify the foehn, but it is difficult to select effective predictors [23,24].…”