Synopsis

This report analyzes the impact of severe weather events in the United States using NOAA storm data. The analysis identifies which event types are most harmful to population health and which have the greatest economic consequences. The results show that tornadoes cause the most harm to human life, while floods and hurricanes result in the highest economic damage.

Data Processing

The dataset was loaded from a compressed CSV file and processed in R. Relevant variables such as event type, fatalities, injuries, and property and crop damages were extracted. Damage values were converted into numeric form using appropriate multipliers (K, M, B) to calculate total economic impact.

storm_data <- read.csv("repdata_data_StormData.csv.bz2")
storm_clean <- storm_data[, c("EVTYPE", "FATALITIES", "INJURIES", 
                             "PROPDMG", "PROPDMGEXP", 
                             "CROPDMG", "CROPDMGEXP")]
storm_clean$PROPDMGEXP <- toupper(storm_clean$PROPDMGEXP)
storm_clean$PROPDMG_MULT <- ifelse(storm_clean$PROPDMGEXP == "K", 1e3,
                            ifelse(storm_clean$PROPDMGEXP == "M", 1e6,
                            ifelse(storm_clean$PROPDMGEXP == "B", 1e9, 1)))

storm_clean$CROPDMGEXP <- toupper(storm_clean$CROPDMGEXP)
storm_clean$CROPDMG_MULT <- ifelse(storm_clean$CROPDMGEXP == "K", 1e3,
                            ifelse(storm_clean$CROPDMGEXP == "M", 1e6,
                            ifelse(storm_clean$CROPDMGEXP == "B", 1e9, 1)))

storm_clean$TOTAL_PROPDMG <- storm_clean$PROPDMG * storm_clean$PROPDMG_MULT
storm_clean$TOTAL_CROPDMG <- storm_clean$CROPDMG * storm_clean$CROPDMG_MULT
storm_clean$TOTAL_DAMAGE <- storm_clean$TOTAL_PROPDMG + storm_clean$TOTAL_CROPDMG

Results

health_data <- aggregate(cbind(FATALITIES, INJURIES) ~ EVTYPE, 
                         data = storm_clean, sum)

health_data$TOTAL_HARM <- health_data$FATALITIES + health_data$INJURIES
health_data <- health_data[order(-health_data$TOTAL_HARM), ]
top10 <- head(health_data, 10)

barplot(top10$TOTAL_HARM,
        names.arg = top10$EVTYPE,
        las = 2,
        col = "steelblue",
        main = "Top 10 Harmful Weather Events",
        ylab = "Total Harm (Fatalities + Injuries)")

Tornadoes are the most harmful weather events in terms of population health, causing significantly higher fatalities and injuries compared to other event types.

economic_data <- aggregate(TOTAL_DAMAGE ~ EVTYPE, 
                           data = storm_clean, sum)

economic_data <- economic_data[order(-economic_data$TOTAL_DAMAGE), ]
top10_econ <- head(economic_data, 10)

barplot(top10_econ$TOTAL_DAMAGE,
        names.arg = top10_econ$EVTYPE,
        las = 2,
        col = "darkgreen",
        main = "Top 10 Events by Economic Damage",
        ylab = "Total Economic Damage")

Floods cause the greatest economic damage among all weather events, followed by hurricanes and tornadoes.