Machine Learning Applications to Dust Storms: A Meta-Analysis

Abstract

Dust storms are natural hazards that affect both people and properties. Therefore, it is important to mitigate their risks by implementing an early notification system. Different methods are used to predict dust storms, such as observing satellite images, analyzing meteorological data, and using numerical weather prediction model forecasts. However, recent studies have shown that machine learning algorithms have higher capacities to predict dust storms in less time and with fewer processing operations compared to numerical weather models. This paper conducted a meta-analysis review to examine studies that addressed the areas associated with the application of machine learning to dust storm prediction. It aims to compare the applied models and the types of data used in the literature under study. Given that the location of a dust storm event is essential, the properties of dust storms are discussed in relation to the region. The output classes and the various performance metrics observed in each reviewed paper are also summarized. Subsequently, the present paper offers a detailed analysis highlighting the capabilities of machine learning models in predicting dust storms. The analysis shows two main categories: early detection and dust storm prediction. Most models used for dust storm early detection from satellite images are support vector machines (SVM). In contrast, the most used models for dust storm prediction are SVM and random forests that predict the occurrence of dust storms from meteorological data. Finally, the paper highlights the challenges and future trends in the field, illustrating the potential directions for applying deep learning algorithms and providing longrange predictions with assessments of dust storm duration and intensity.

Yes, dust storms can possibly be forecasted now!!

“…To address this challenging task, several studies have been tried in the past. These approaches can broadly be categorized as the use of artificial intelligence/machine learning (AI/ML) (40,41), physical factors and modelling techniques (42,43), remote sensing, statistical predictive models, and game theory (44), or the Air Quality Early Warning System (45). After thorough examination (41), it was concluded that the application of AI/ML approach has various inherent constraints, including quantity and quality of the training dataset along with the complexity of the processes.…”

“…Moreover, major benefit of the HDWI approach lies in its computational efficiency, which stands in stark contrast to the computationally demanding and resource-intensive AI/ML approach, remote sensing, statistical predictive models, game theory, or the Air Quality Early Warning System. The HDWI method exclusively utilizes physical factors, rather than depending on statistical approaches and/or pattern identification (41). Another advantage of HDWI method is its capacity to function independently, and requiring less dependency on external resources.…”