Note that, there are other tools to test for significant difference such as simple regression or multiple regression<em> but when comparing the results of two groups in terms of which is higher or lower</em>, then a two sided t-test would be best appropriate.
Since we want to know if there is a significant difference between the number of tornadoes in August and the number of tornadoes in October on average we use the two sided t-test because a Two sided t-test can show the possibility of positive or negative differences.
