NOAA Storm Database Analysis: Health and Economic Impact of Severe Weather Events
Synopsis
This analysis explores the U.S. National Oceanic and Atmospheric Administration (NOAA) storm database to identify which types of severe weather events are most harmful to population health and which have the greatest economic consequences across the United States. The data spans from 1950 to November 2011. Population health impact is measured by aggregating fatalities and injuries by event type. Economic impact is measured by aggregating property damage and crop damage values, accounting for the magnitude codes (K, M, B) attached to each damage estimate. The results show that tornadoes are by far the most harmful event type for population health, responsible for the greatest number of fatalities and injuries combined. For economic consequences, floods cause the highest total property and crop damage. These findings suggest that emergency preparedness resources should prioritize tornado response for human safety and flood mitigation for economic protection.
Data Processing
Loading the Data
The raw data file is loaded directly from the CSV file. No preprocessing is done outside this document.
data <- read.csv("repdata_data_StormData1.csv", stringsAsFactors = FALSE)
dim(data)## [1] 902297 37names(data)## [1] "STATE__" "BGN_DATE" "BGN_TIME" "TIME_ZONE" "COUNTY"
## [6] "COUNTYNAME" "STATE" "EVTYPE" "BGN_RANGE" "BGN_AZI"
## [11] "BGN_LOCATI" "END_DATE" "END_TIME" "COUNTY_END" "COUNTYENDN"
## [16] "END_RANGE" "END_AZI" "END_LOCATI" "LENGTH" "WIDTH"
## [21] "F" "MAG" "FATALITIES" "INJURIES" "PROPDMG"
## [26] "PROPDMGEXP" "CROPDMG" "CROPDMGEXP" "WFO" "STATEOFFIC"
## [31] "ZONENAMES" "LATITUDE" "LONGITUDE" "LATITUDE_E" "LONGITUDE_"
## [36] "REMARKS" "REFNUM"Processing Health Data
We aggregate total fatalities and injuries by event type.
fatalities_by_event <- tapply(data$FATALITIES, data$EVTYPE, sum, na.rm = TRUE)
injuries_by_event <- tapply(data$INJURIES, data$EVTYPE, sum, na.rm = TRUE)
health_df <- data.frame(
EVTYPE = names(fatalities_by_event),
FATALITIES = as.numeric(fatalities_by_event),
INJURIES = as.numeric(injuries_by_event)
)
health_df$TOTAL <- health_df$FATALITIES + health_df$INJURIES
health_df <- health_df[order(-health_df$TOTAL), ]
top10_health <- head(health_df, 10)
rownames(top10_health) <- NULL
print(top10_health)## EVTYPE FATALITIES INJURIES TOTAL
## 1 TORNADO 5633 91346 96979
## 2 EXCESSIVE HEAT 1903 6525 8428
## 3 TSTM WIND 504 6957 7461
## 4 FLOOD 470 6789 7259
## 5 LIGHTNING 816 5230 6046
## 6 HEAT 937 2100 3037
## 7 FLASH FLOOD 978 1777 2755
## 8 ICE STORM 89 1975 2064
## 9 THUNDERSTORM WIND 133 1488 1621
## 10 WINTER STORM 206 1321 1527Processing Economic Data
The damage columns (PROPDMG, CROPDMG) have exponent codes in PROPDMGEXP and CROPDMGEXP. We convert these using a vectorized lookup.
# Vectorized multiplier function
get_multiplier <- function(exp_vec) {
exp_vec <- toupper(trimws(as.character(exp_vec)))
mult <- ifelse(exp_vec == "K", 1e3,
ifelse(exp_vec == "M", 1e6,
ifelse(exp_vec == "B", 1e9,
ifelse(exp_vec == "H", 1e2,
ifelse(exp_vec == "2", 1e2,
ifelse(exp_vec == "3", 1e3,
ifelse(exp_vec == "4", 1e4,
ifelse(exp_vec == "5", 1e5,
ifelse(exp_vec == "6", 1e6,
ifelse(exp_vec == "7", 1e7, 1))))))))))
return(mult)
}
data$PROP_DAMAGE <- data$PROPDMG * get_multiplier(data$PROPDMGEXP)
data$CROP_DAMAGE <- data$CROPDMG * get_multiplier(data$CROPDMGEXP)
data$TOTAL_DAMAGE <- data$PROP_DAMAGE + data$CROP_DAMAGE
prop_by_event <- tapply(data$PROP_DAMAGE, data$EVTYPE, sum, na.rm = TRUE)
crop_by_event <- tapply(data$CROP_DAMAGE, data$EVTYPE, sum, na.rm = TRUE)
total_by_event <- tapply(data$TOTAL_DAMAGE, data$EVTYPE, sum, na.rm = TRUE)
econ_df <- data.frame(
EVTYPE = names(total_by_event),
PROP_DAMAGE = as.numeric(prop_by_event),
CROP_DAMAGE = as.numeric(crop_by_event),
TOTAL_DAMAGE = as.numeric(total_by_event)
)
econ_df <- econ_df[order(-econ_df$TOTAL_DAMAGE), ]
top10_econ <- head(econ_df, 10)
rownames(top10_econ) <- NULL
top10_econ$PROP_B <- round(top10_econ$PROP_DAMAGE / 1e9, 2)
top10_econ$CROP_B <- round(top10_econ$CROP_DAMAGE / 1e9, 2)
top10_econ$TOTAL_B <- round(top10_econ$TOTAL_DAMAGE / 1e9, 2)
print(top10_econ[, c("EVTYPE", "PROP_B", "CROP_B", "TOTAL_B")])## EVTYPE PROP_B CROP_B TOTAL_B
## 1 FLOOD 144.66 5.66 150.32
## 2 HURRICANE/TYPHOON 69.31 2.61 71.91
## 3 TORNADO 56.95 0.41 57.36
## 4 STORM SURGE 43.32 0.00 43.32
## 5 HAIL 15.74 3.03 18.76
## 6 FLASH FLOOD 16.82 1.42 18.24
## 7 DROUGHT 1.05 13.97 15.02
## 8 HURRICANE 11.87 2.74 14.61
## 9 RIVER FLOOD 5.12 5.03 10.15
## 10 ICE STORM 3.94 5.02 8.97Results
Question 1: Which event types are most harmful to population health?
The chart below shows the top 10 weather event types by total casualties (fatalities + injuries) from 1950 to 2011.
par(mar = c(5, 11, 4, 2))
colors_health <- c("#D85A30", rep("#B4B2A9", 9))
bp <- barplot(
rev(top10_health$TOTAL),
names.arg = rev(top10_health$EVTYPE),
horiz = TRUE,
las = 1,
col = rev(colors_health),
main = "Top 10 Weather Events by Population Health Impact",
xlab = "Total Casualties (Fatalities + Injuries)",
cex.names = 0.8,
cex.axis = 0.85,
xlim = c(0, max(top10_health$TOTAL) * 1.15)
)
text(
x = rev(top10_health$TOTAL) + max(top10_health$TOTAL) * 0.01,
y = bp,
labels = format(rev(top10_health$TOTAL), big.mark = ","),
cex = 0.7,
adj = 0
)
Figure 1: Top 10 weather event types by total population health impact (fatalities + injuries), 1950-2011.
Tornadoes are the most harmful event for population health, with 96,979 total casualties — over 11.5x more than the second-ranked event (EXCESSIVE HEAT).
Question 2: Which event types have the greatest economic consequences?
The chart below shows the top 10 event types by total economic damage (property + crop) in billions of USD.
par(mar = c(5, 12, 4, 2))
damage_matrix <- rbind(
rev(top10_econ$PROP_B),
rev(top10_econ$CROP_B)
)
bp2 <- barplot(
damage_matrix,
names.arg = rev(top10_econ$EVTYPE),
horiz = TRUE,
las = 1,
col = c("#185FA5", "#3B6D11"),
main = "Top 10 Weather Events by Economic Damage",
xlab = "Total Damage (Billions USD)",
cex.names = 0.8,
cex.axis = 0.85,
legend.text = c("Property Damage", "Crop Damage"),
args.legend = list(x = "bottomright", cex = 0.8),
xlim = c(0, max(top10_econ$TOTAL_B) * 1.2)
)
text(
x = rev(top10_econ$TOTAL_B) + max(top10_econ$TOTAL_B) * 0.01,
y = bp2,
labels = paste0("$", rev(top10_econ$TOTAL_B), "B"),
cex = 0.7,
adj = 0
)
Figure 2: Top 10 weather event types by total economic damage in billions USD, 1950-2011.
Floods caused the most total economic damage at approximately $150.32 billion, followed by hurricanes/typhoons. Drought stands out as the leading cause of crop damage specifically.