Exploratory Data Analysis
In regards to our initial exploration of the data, we were planning on investigating the pattern of severe storm based on time of the year/location. Following are some of our plots we have made:
We were surprised to see that there wasn’t a clear pattern in terms of severe storms and time of the year, but we do see that later in the year, from October to December,the occurrence of storms do slow down. On the other hand, southern and midwestern states like Oklahoma, Texas, Nebraska, Kansas, Illinois, and Arkansas are heavily affected by severe storms. The states are all part of an area known as “Tornado Alley” which spans from Texas to South Dakota. States in this area are more prone to Tornadoes.
Data Equity
Among the principles, beneficence is the most relevant to our data. We are able to maximize the benefits of the data by identifying patterns of severe storm events in specific areas. By having more information on this, we will be able to help people by giving them more time to evacuate areas faster. We have seen the lives of millions in New Orleans, Houston, and Oklahoma City be affected with very severe storms this decade. Our data will help these people by predicting severe storms more effectively. This is why we will be splitting our analysis into smaller subsets based on location. Our data exudes respect for all people which is another data equity principle. We offer people our storm data, but it is their decision to do something with it. For the people who do not have the power to make their own decision upon looking at the data, we will protect them by setting the data in a more friendly way and writing out reports for each general location. The data definitely can’t be used to take advantage of those who need protection. A limitation we will face would be the general unpredictable nature of weather. Knowing even a flap of a butterfly can snowball into radically different weather patterns, it is hard to say we will have super precise predictions for upcoming patterns of storm.