Michael B. Richman

To assist water managers in south‐central Oklahoma prepare for future drought, reliable place‐based drought forecasts are produced. Past‐, present‐, and future‐forecasted climate indices (Multivariate ENSO Index, Pacific Decadal Oscillation index, and Atlantic Multidecadal Oscillation index) and past and present Palmer Drought Severity Index (PDSI) are employed as predictor variables to forecast PDSI using a multivariate regression technique. PDSI is forecasted 18 months in advance with sufficient skill to provide water managers early warning of drought. Using a training data set obtained from the period January 1901 to November 2021, a second‐order model equation that contains, without any restriction, all the predictors and their interaction terms is built to predict drought intensity. Significant predictors are selected through stepwise regression, with cross‐validation producing the simplest restricted model that describes the data well. PDSI values are predicted using 1000 fitted restricted models produced from bootstrapping, then averaged monthly. The technique found the best‐fitting model and estimated the model coefficients that minimized the sum of squared deviations between the fitted model and the predictor variables. The adjusted R‐squared value of the restricted model is large enough to explain an adequately accurate model, and relatively low values of error measures point to good predictive ability of the model. Although the model slightly overestimates the PDSI forecast maxima and minima, it necessarily captures the timing of the periods of severe to exceptional drought.

This study supports efforts directed toward research on large‐scale atmospheric patterns and on the variability of tornado outbreaks. Specifically, we applied rotated principal component analysis to identify synoptic‐scale patterns of 500‐hPa geopotential height associated with tornado outbreaks in the United States. We created a database of historic tornado outbreaks using kernel density estimation on events composed of at least seven tornadoes of magnitude (E)F2 or higher (major tornado outbreaks) that occurred in May from 1950 to 2019 (91 events). Results of the analysis show that the first three principal components explained the majority, that is, 74% of the total variation. Based on the analysis of congruence coefficients, the Promax oblique transformation was chosen as the most representative rotation in portraying physically meaningful modes and resulted in three main atmospheric patterns. Two of these, to the best of our knowledge, have not been yet identified as the most representative of tornado outbreaks in any previous studies. Additionally, results suggest that although synoptic patterns associated with major May tornado outbreaks remain the same over time, partial variability in their geographic location exhibits some cyclical behaviour, especially on decadal and multidecadal scales. Identifying these patterns can serve as a first step in determining how they may change under anthropogenic climate change in the future.