Tornado Time Research Articles

Dynamical analysis and visualization of tornadoes time series.

In this paper we analyze the behavior of tornado time-series in the U.S. from the perspective of dynamical systems. A tornado is a violently rotating column of air extending from a cumulonimbus cloud down to the ground. Such phenomena reveal features that are well described by power law functions and unveil characteristics found in systems with long range memory effects. Tornado time series are viewed as the output of a complex system and are interpreted as a manifestation of its dynamics. Tornadoes are modeled as sequences of Dirac impulses with amplitude proportional to the events size. First, a collection of time series involving 64 years is analyzed in the frequency domain by means of the Fourier transform. The amplitude spectra are approximated by power law functions and their parameters are read as an underlying signature of the system dynamics. Second, it is adopted the concept of circular time and the collective behavior of tornadoes analyzed. Clustering techniques are then adopted to identify and visualize the emerging patterns.

Family Emergency Preparedness Plans in Severe Tornadoes

Tornadoes, with warnings usually issued just minutes before their touchdowns, pose great threats to properties and people's physical and mental health. Few studies have empirically investigated the association of family emergency preparedness planning and observed protective behaviors in the context of tornadoes. The purpose of this study was to examine predictors for the action of taking shelter at the time of tornadoes. Specifically, this study investigated whether having a family emergency preparedness plan was associated with higher likelihood of taking shelter upon receiving tornado warnings. This study also examined the effects of socioeconomic status and functional limitations on taking such actions. A telephone survey based on random sampling was conducted in 2012 with residents in Tuscaloosa AL and Joplin MO. Each city experienced considerable damages, injuries, and casualties after severe tornadoes (EF-4 and EF-5) in 2011. The working sample included 892 respondents. Analysis was conducted in early 2013. Logistic regression identified emergency preparedness planning as the only shared factor that increased the likelihood of taking shelter in both cities and the only significant factor in Joplin. In Tuscaloosa, being female and white also increased the likelihood of taking shelter. Disability was not found to have an effect. This study provided empirical evidence on the importance of having a family emergency preparedness plan in mitigating the risk of tornadoes. The findings could be applied to other rapid-onset disasters.

Detection of Tornado Frequency Trend Over Ontario, Canada

An upward trend in Ontario tornado frequency (about 1.6 tornadoes/decade with the statistically significant level at least at 95%) is identified using three independent approaches. The first method is the conventional linear regression method that had no disturbance to the original tornado time series. The second approach is to employ the Mann-Kendall test with consideration of removing a lag one autoregressive process. The trend is further firmed up using the Monte Carlo simulation. The robustness of the detected tornado frequency trend presented in this work offers an example for tornado frequency trend analysis over other regions of the world. It is shown in this study that the Ontario tornado frequency is linked to ENSO (El Nino/Southern Oscillation) signals with a statistically significant level at 99%.

Support vector machines for spatiotemporal tornado prediction

The use of support vector machines (SVMs) for predicting the location and time of tornadoes is presented. In this paper, we extend the work by Lakshmanan et al. (Proceedings of 2005 IEEE international joint conference on neural networks (Montreal, Canada), 3, 2005a, 1642–1647) to use a set of 33 storm days and introduce some variations that improve the results. The goal is to estimate the probability of a tornado event at a particular spatial location within a given time window. We utilize a least-squares methodology to estimate shear, quality control of radar reflectivity, morphological image processing to estimate gradients, fuzzy logic to generate compact measures of tornado possibility and SVM classification to generate the final spatiotemporal probability field. On the independent test set, this method achieves a Heidke's skill score of 0.60 and a critical success index of 0.45.

The Teton-Yellowstone Tornado of 21 July 1987

The Teton-Yellowstone Tornado, rated F4, crossed the Continental Divide at 3070 m, leaving behind a damage swath 39.2-km long and 2.5-km wide. A detailed damage analysis by using stereo-pair and color photos revealed the existence of four spinup swirl marks and 72 microburst outflows inside the damage area. The tornado was spawned by a mesocyclone that formed at the intersection of a mesohigh boundary and a warm front. The parent cloud of the tornado, tracked on eight infrared-temperature maps from GOES East and West, moved at 25 m s-1 and the number of cold temperature pixels below -60 C reached a distinct peak during the tornado time. Identified and tracked also are two warm spots enclosed inside the cold anvil cloud. On the basis of their identity and movement, an attempt was made to explain the cause of these spots as being the stratospheric cirrus clouds.

Lightning Rates Relative to Tornadic Storm Evolution on 22 May 1981

Lightning and Doppler radar data for two tornadic storms in Oklahoma on May 22, 1981 are used to analyze ground flash rates relative to the time of tornadoes. It is found that the ground flash rates had no obvious relationship with the tornado times, although the stroke rate in both storms was greatest after the tornadic stage ended. The variations in the cyclone shear and the intracloud flash rates within 10 km of the mesocyclone region are examined. The results suggest that most tornadic storms have an increase in total flash rates near the time of the tornado and that this increase is often dominated by intracloud flashes.

Directional VLF sferics from the Union City, Oklahoma, tornadic storm

The Naval Electronics Laboratory Center (now Naval Ocean Systems Center) made directional measurements of very low frequency sferics in the Union City, Oklahoma, tornadic storm on May 24, 1973. These data verify the interesting findings of two earlier directional sferics studies. Whereas sferics activity in tornado‐producing storms is known to peak near tornado time, all these studies show conclusively that sferics do not originate in the tornado itself but rather throughout the parent thunderstorm.

Tornadic Storm Airflow and Morphology Derived from Single-Doppler Radar Measurements

Abstract Single-Doppler Velocity data reveal that a dominant feature in the Union City, Okla., tornadic thunderstorm is a core mesocyclonic circulation, 2–6 km in diameter, extending to at least 9 km above ground. There is an apparent flow through the precipitation echo at low levels and divergence at high levels. Considerable similarity appears between mid-level flow structure around the mesocyclone core and that observed around a solid rotating cyclinder embedded in classical potential flow. As tornado time approaches, core circulation tangential velocities increase while diameter decreases. Simultaneously, the collapse of storm top and extensive echo overhang suggest updraft weakening.

A Study of Upper Air Parameters at the Time of Tornadoes

Abstract An objective analysis routine was used to obtain the values of selected upper air parameters at the location and time of tornadoes. Parameters used were the temperature, dew point, wind direction and wind speed for the 850 mb, 700 mb, 500 mb levels, the 200 mb wind direction and wind speed, the mean mixing ratio in the lowest 100 mb layer, the tropopause temperature and pressure, and the 500–850 mb thickness. Some computed stability indices are also presented; these are the Showalter Index, Total-Totals Index, and the Air Force Sweat Index. Tables are presented giving the average value and the standard deviation for each of the parameters by states and by months. Similar information is presented for the longer track tornadoes and they are compared with very short track tornadoes.

Identification of Tornadoes by Observation of Waveform Atmospherics

Research on the characteristics of atmospherics peculiar to tornadoes has been in progress at the Oklahoma Agricultural and Mechanical College since 1947. It has been discovered that high-energy thunderstorms which develop into tornadoes generate discharges with a preponderance of frequencies in the 200- to- 400-kc band. The number of these discharges increases as tornado time is approached.