Synopsis

This report analyzes the NOAA Storm Database to identify the most impactful weather events. We found that Tornadoes are the leading cause of fatalities and injuries. In contrast, Floods and Hurricanes cause the most significant economic damage.

Data Processing

library(dplyr)
## Warning: package 'dplyr' was built under R version 4.5.3
library(ggplot2)
## Warning: package 'ggplot2' was built under R version 4.5.3
# Load data directly from the bz2 file
data <- read.csv("C:/Users/saram/Downloads/repdata_data_StormData.csv.bz2")

# Function to handle damage multipliers
calc_dmg <- function(dmg, exp) {
  if (exp %in% c('k', 'K')) return(dmg * 1000)
  if (exp %in% c('m', 'M')) return(dmg * 1e6)
  if (exp %in% c('b', 'B')) return(dmg * 1e9)
  return(dmg)
}

# Apply the function to get real dollar values
data$prop_total <- mapply(calc_dmg, data$PROPDMG, data$PROPDMGEXP)
data$crop_total <- mapply(calc_dmg, data$CROPDMG, data$CROPDMGEXP)

# Create summary totals
data$health_total <- data$FATALITIES + data$INJURIES
data$econ_total <- data$prop_total + data$crop_total

The analysis begins by loading the raw NOAA Storm Database. To ensure the analysis is reproducible, the data is read directly from the compressed .csv.bz2 format. Because the database uses character identifiers (K, M, B) to represent the magnitude of economic damage, a custom function was applied to multiply the base damage estimates by their respective exponential values (\(1,000\), \(1,000,000\), or \(1,000,000,000\)). Finally, we created two aggregate variables: health_total (sum of fatalities and injuries) and econ_total (sum of property and crop damage) to simplify the comparison across event types.

Results

1. Impact on Population Health

health_results <- data %>%
  group_by(EVTYPE) %>%
  summarise(Total_Health = sum(health_total, na.rm = TRUE)) %>%
  arrange(desc(Total_Health)) %>%
  slice(1:10)

ggplot(health_results, aes(x = reorder(EVTYPE, Total_Health), y = Total_Health)) +
  geom_bar(stat = "identity", fill = "red") +
  coord_flip() +
  labs(title = "Top 10 Weather Events: Population Health", x = "Event Type", y = "Total Injuries/Fatalities")

Based on the analysis of the NOAA database (Figure 1), Tornadoes are the clear outlier and the most significant threat to population health in the United States. With over 90,000 combined injuries and fatalities, the impact of tornadoes is nearly ten times greater than the next leading cause. Excessive Heat ranks second, followed closely by Thunderstorm Winds (TSTM WIND) and Flooding. The data suggests that while high-frequency events like wind and rain cause steady harm, the extreme intensity of tornadoes accounts for the vast majority of weather-related casualties.

2. Economic Consequences

econ_results <- data %>%
  group_by(EVTYPE) %>%
  summarise(Total_Econ = sum(econ_total, na.rm = TRUE)) %>%
  arrange(desc(Total_Econ)) %>%
  slice(1:10)

ggplot(econ_results, aes(x = reorder(EVTYPE, Total_Econ), y = Total_Econ)) +
  geom_bar(stat = "identity", fill = "blue") +
  coord_flip() +
  labs(title = "Top 10 Weather Events: Economic Damage", x = "Event Type", y = "Total Damage (USD)")

When evaluating financial impact (Figure 2), the data reveals that Flooding is the primary driver of economic loss, accounting for approximately $150 billion in combined property and crop damage. This is followed by Hurricanes/Typhoons, which caused over $70 billion in damage, and Tornadoes at approximately $57 billion. It is notable that while Tornadoes are the deadliest events, Floods are the most ‘expensive.’ This indicates that water-related disasters, including Storm Surges and River Flooding, represent the greatest long-term financial risk to US infrastructure and agriculture.

Interpretations

Based on the data, there is a clear distinction between ‘High-Health Risk’ and ‘High-Economic Risk’ events. Tornadoes are overwhelmingly the most dangerous to human life, likely due to their sudden onset and high intensity. However, when looking at the checkbook, Floods and Hurricanes are the primary drivers of economic loss. This suggests that while tornado warnings save lives, better flood-plain management and coastal reinforcement are what save the economy.