Impact of Severe Weather Events in the United States
Synopsis
In this report we explore the U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database which tracks characteristics of major storms and weather events in the United States, including when and where they occur, as well as estimates of any fatalities, injuries, and property damage. Our analysis objective is to identify the types of weather events that are most harmful with respect to population health, and those with greatest economic consequences. To assess impact on population health, we looked at fatalities and injuries. To assess economic impact, we looked at property damage and crop damage. We used all the history on file (1950-2011) for the analysis. Aggregating this data across the US, we found that tornados caused the most fatalities and injuries, floods caused the highest property damage, and drought caused the most crop damage.
Data Processing
First we download the data from the URL, then we unzip and read the file into R.
## include the required libraries
library(R.utils)
library(dplyr)if (!file.exists("./data/storm.csv"))
{
fileurl <- "https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2"
if (!file.exists("./data")){dir.create("./data")}
download.file(fileurl, destfile = "./data/storm.bz2")
bunzip2("./data/storm.bz2", "./data/storm.csv", overwrite=TRUE, remove=FALSE)
}
storm <- read.csv("./data/storm.csv")
dim(storm)## [1] 902297 37head(storm)## STATE__ BGN_DATE BGN_TIME TIME_ZONE COUNTY COUNTYNAME STATE
## 1 1 4/18/1950 0:00:00 0130 CST 97 MOBILE AL
## 2 1 4/18/1950 0:00:00 0145 CST 3 BALDWIN AL
## 3 1 2/20/1951 0:00:00 1600 CST 57 FAYETTE AL
## 4 1 6/8/1951 0:00:00 0900 CST 89 MADISON AL
## 5 1 11/15/1951 0:00:00 1500 CST 43 CULLMAN AL
## 6 1 11/15/1951 0:00:00 2000 CST 77 LAUDERDALE AL
## EVTYPE BGN_RANGE BGN_AZI BGN_LOCATI END_DATE END_TIME COUNTY_END
## 1 TORNADO 0 0
## 2 TORNADO 0 0
## 3 TORNADO 0 0
## 4 TORNADO 0 0
## 5 TORNADO 0 0
## 6 TORNADO 0 0
## COUNTYENDN END_RANGE END_AZI END_LOCATI LENGTH WIDTH F MAG FATALITIES
## 1 NA 0 14.0 100 3 0 0
## 2 NA 0 2.0 150 2 0 0
## 3 NA 0 0.1 123 2 0 0
## 4 NA 0 0.0 100 2 0 0
## 5 NA 0 0.0 150 2 0 0
## 6 NA 0 1.5 177 2 0 0
## INJURIES PROPDMG PROPDMGEXP CROPDMG CROPDMGEXP WFO STATEOFFIC ZONENAMES
## 1 15 25.0 K 0
## 2 0 2.5 K 0
## 3 2 25.0 K 0
## 4 2 2.5 K 0
## 5 2 2.5 K 0
## 6 6 2.5 K 0
## LATITUDE LONGITUDE LATITUDE_E LONGITUDE_ REMARKS REFNUM
## 1 3040 8812 3051 8806 1
## 2 3042 8755 0 0 2
## 3 3340 8742 0 0 3
## 4 3458 8626 0 0 4
## 5 3412 8642 0 0 5
## 6 3450 8748 0 0 6Next we take a subset of the storm data, with just the variables relevant for our analysis. We need the columns: EVTYPE, FATALITIES, INJURIES, PROPDMG, PROPDMGEXP, CROPDMG, and CROPDMGEXP.
sub_storm <- select(storm, EVTYPE, FATALITIES:CROPDMGEXP)
str(sub_storm)## 'data.frame': 902297 obs. of 7 variables:
## $ EVTYPE : Factor w/ 985 levels " HIGH SURF ADVISORY",..: 834 834 834 834 834 834 834 834 834 834 ...
## $ FATALITIES: num 0 0 0 0 0 0 0 0 1 0 ...
## $ INJURIES : num 15 0 2 2 2 6 1 0 14 0 ...
## $ PROPDMG : num 25 2.5 25 2.5 2.5 2.5 2.5 2.5 25 25 ...
## $ PROPDMGEXP: Factor w/ 19 levels "","-","?","+",..: 17 17 17 17 17 17 17 17 17 17 ...
## $ CROPDMG : num 0 0 0 0 0 0 0 0 0 0 ...
## $ CROPDMGEXP: Factor w/ 9 levels "","?","0","2",..: 1 1 1 1 1 1 1 1 1 1 ...Assessing impact on population health
To assess impact on population health, we looked at fatalities and injuries.
Here we aggregate the fatalities by weather-event type and pick the top 10 weather events causing fatalities.
fe <- aggregate(FATALITIES ~ EVTYPE, sub_storm, sum)
fe_srt <- arrange(fe, desc(FATALITIES))
top10fe <- fe_srt[1:10,] Next we aggregate the Injuries by weather-event type and pick the top 10 weather events causing injuries.
ie <- aggregate(INJURIES ~ EVTYPE, sub_storm, sum)
ie_srt <- arrange(ie, desc(INJURIES))
top10ie <- ie_srt[1:10,]Assessing economic impact
To assess economic impact, we looked at property damage and crop damage.
In order to calculate property damage and crop damage totals, we first process the exponent columns: PROPDMGEXP and CROPDMGEXP by converting the factor values (k, m, b …) to their numeric equivalents (k = 1000, …). Then we multiply the damage amounts by the corresponding exponent to get the final value.
looking at the available exponents:
table(sub_storm$PROPDMGEXP)##
## - ? + 0 1 2 3 4 5
## 465934 1 8 5 216 25 13 4 4 28
## 6 7 8 B h H K m M
## 4 5 1 40 1 6 424665 7 11330table(sub_storm$CROPDMGEXP)##
## ? 0 2 B k K m M
## 618413 7 19 1 9 21 281832 1 1994Data has invalid exponents (“-”, “?”, and “+”). We will exclude those rows since they make up a very small proportion (21 rows out of the total 902,297 observations.)
Below is a function to map the exponents to their numeric equivalents.
fexp <- function(x) {
if(x %in% c("0","")) {x<- 1}
else if(x == "1") {x<- 10}
else if(x %in% c("2","h","H")) {x<-10^2}
else if(x %in% c("3","k","K")) {x<-10^3}
else if(x == "4") {x<- 10^4}
else if(x == "5") {x<- 10^5}
else if(x %in% c("6","m","M")) {x<-10^6}
else if(x == "7") {x<- 10^7}
else if(x == "8") {x<- 10^8}
else if(x %in% c("9","b","B")) {x<-10^9}
else x<-0 ## to zero out the values with invalid exponents
}Here we compute and aggregate property damage by weather-event type and pick the top 10 weather events causing property damage.
## Aapply the fexp function created abpve to compute the property damage amount (add new field PROPDMGAMT).
sub_storm <- mutate(sub_storm, PROPDMGAMT = PROPDMG * sapply(PROPDMGEXP,fexp ))
pe <- aggregate(PROPDMGAMT ~ EVTYPE, sub_storm, sum)
pe_srt <- arrange(pe, desc(PROPDMGAMT))
top10pe <- pe_srt[1:10,]Next we compute and aggregate crop damage by weather-event type and pick the top 10 weather events causing crop damage.
## Aapply the fexp function created above to compute the crop damage amount (add new field CROPDMGAMT).
sub_storm <- mutate(sub_storm, CROPDMGAMT = CROPDMG * sapply(CROPDMGEXP,fexp ))
ce <- aggregate(CROPDMGAMT ~ EVTYPE, sub_storm, sum)
ce_srt <- arrange(ce, desc(CROPDMGAMT))
top10ce <- ce_srt[1:10,] Results
- Population Health Impact
The plots below show the 10 most harmful weather events across the United States with respect to population health. The highest is tornado, which has caused the most fatalities and injuries.
par(mfrow = c(1,2), mar=c(8,5,4,2),cex = 0.7)
barplot(top10fe$FATALITIES,
main = "Top 10 harmful weather events
causing Fatalities",
names.arg = top10fe$EVTYPE, las = 2,
ylab = "Fatalities", col = "blue")
barplot(top10ie$INJURIES,
main = "Top 10 harmful weather events
causing Injuries",
names.arg = top10ie$EVTYPE, las = 2,
ylab = "Injuries", col = "blue")
- Economic Impact
The plots below show the 10 most harmful weather events across the United States with respect to economic impact. Floods have caused the highest property damage. and Drought had the worst impact on crop damage.
par(mfrow = c(1,2), mar=c(11,5,4,2),cex = 0.7)
barplot(top10pe$PROPDMGAMT / 10^9,
main = "Top 10 harmful weather events
causing Property Damage",
names.arg = top10pe$EVTYPE, las = 2,
ylab = "Property Damage ($billions)", col = "blue")
barplot(top10ce$CROPDMGAMT / 10^9,
main = "Top 10 harmful weather events
causing Crop Damage",
names.arg = top10ce$EVTYPE, las = 2,
ylab = "Crop Damage ($billions)", col = "blue")