No description available.

p. 2 (Abstract) missing from manuscript.

No description available.

No description available.

A radar study of two Alberta hailstorms has revealed the existence of small cellular radar echoes with durations of 10-30 minutes. These cells were found to group together to form storm families which existed for several hours. New cells invariably formed on the southern end of the storm family and migrated northward through the storm to dissipate on the northern boundary. [...]

Hail damage in canola has been characterized with symptoms such as randomly broken stems and loss of plants. The effect on stand reduction and stem cut-off in canola has yet to be addressed. Two separate experiments, stand reduction and stem cut-off, were conducted at six environments using a factorial arrangement, with four growth stages and five levels of intensity as the two factors, and four replications for each treatment. Data on seed yield, seed oil, and seed protein from each plot of five environments, and yield contributing traits from twenty plants per treatment from four environments were taken. Seed yield losses increased as both growth stage and level of stem cut-off/stand reduction increased. A regression equation was developed to estimate the yield reduction resulting from different levels of stand reduction/stem cut-off at four growth stages. Highest seed yield reductions were 82% and 43% at 90% stand reduction/stem cut-off, respectively. / Northern Canola Growers Association

Ten years of lightning data was used to examine the lightning climatology in the Mid-South and to create a model capable of predicting severe hail storms using CG lightning. Cloud to ground lightning peaked reached a maximum in July and a minimum in January. Positive CG accounted for 5.3% of all strikes. The percentage of positive strikes reached a maximum in December and a minimum in August. Artificial intelligence along with logistic regression models were used for hail prediction. The 95% confidence intervals of the contingency statistics were used to determine the performance of the models. The linear cost 100 model and logistic regression had the highest performance and were tested with an independent data set. The logistic regression model outperformed the linear cost 100 model. The performance by both models was under the median statistics but within the 95% confidence interval.

No description available.

Call number: LD2668 .T4 1950 S37 / Master of Science

Vertical wind shear is computed between the 28 possible pairs of the first 8 mandatory radiosonde levels. A hail severity index is defined. Statistically significant correlations between strong shear and hail-free days are found in 11 shear layers. Strong shears are also found to occur more frequently with low and high severity indices than with intermediate severity indices. [...]