A method to estimate the probability of strong winds occurrence using weather radar data

Abstract

Weather radars are capable of detecting and displaying storm-related turbulence as well as precipitation rate with fine spatial resolution and reliable quality. Especially, weather radars can enable the detection of atmospheric conditions in the vicinity of cities and thereby help to notify the strong winds. This article investigates the application of weather radar measurements for predicting strong wind and presents a new method for very short-term storm prediction. In the proposed method, hidden Markov model (HMM) is used to classify atmospheric conditions to "potentially stormy" and "non-stormy" states using the available radar data, and semi-Markov theory is used to estimate the probability of storm occurrence with time. In fact, the probability of transition between "potentially stormy," "non-stormy," and "stormy" states is modeled by a semi-Markov model, to find the unconditional probability of storm occurrence with time. The model is implemented with the use of the data of Tehran C-band weather radar and anemometer of Tehran international airport.Verification results show that the precision (forecast accuracy) is around 0.19 and the recall is around 0.67 in the presented classification method.

A Review of Laboratory and Numerical Techniques to Simulate Turbulent Flows

“…Consequently, this lack of understanding of turbulent flows is possibly the most serious obstacle to reach a systematic description of many important engineering and environmental phenomena. Only to quote some, the reduction in energy losses in the movement of cars (Ferraris et al [9], Palanivendhan et al [10], Igali et al [11], Kurec and Piechna [12]) trains (Wang et al [13] and [14], Zhang et al [15]) and aircrafts (Bollt and Bewley [16], Varshney et al [17], Miller et al [18]), the improvement in the efficiency of engines (Kazmouz et al [19], Balmelli et al [20], Krastev and Di Ilio [21]) and the reduction in their noise (Zhong et al [22], Polacsek et al [23], Jun et al [24]), the dispersion of pollutants released in the atmosphere (Bellini et al [25], Huertas et al [26], Defforge et al [27], Di Bernardino et al [28]) or in the ocean (Liang et al [29], Chen et al [30], Lee et al [31]), the weather forecasts (Chiniforoush et al [32], Tsiringakis et al [33], Verbitsky and Khodzinskaya [34]), the atmospheric boundary layers (Di Bernardino et al [35], Lu and Li [36], Tkachenko et al [37]), Barbarosa et al [38]),the mixing and combustion in chemical reactors (Soto et al [39], Janoszek and Masny [40], Mularski and Modli ński [41]), the optimization of turbines (Pacciani et al [42], Duthè et al [43], Yang et al [44]), the study of the air flowing through the lungs (Wu et al [45]…”