Impact of weather on health and economic damage in US
Zoey Le
February 11, 2018
Synopsis
The purpose of this assignment is to look at the impacts on people and economic of some severe weather events by exploring the NOAA Storm Database. In particularly, the analysis will focus on the number of damage at the amount of injuries, fatalities, property damage and crop damage. Hence, we can answer some basic question, which is:
- Across the United States, which types of events are most harmful with respect to population health?
- Across the United States, which types of events have the greatest economic consequences?
Data Processing
First step is to read the data with only variables we considered including event type (EVTYPE), number of fatalities (FATALITIES), number of injuries (INJURIES), property damage (PROPDMG), exponent of property damage (PROPDMGEXP), crop damage (CROPDMG), exponent of crop damage (CROPDMGEXP).
data=read.csv("repdata_data_StormData.csv")
var=c("EVTYPE", "FATALITIES", "INJURIES", "PROPDMG", "PROPDMGEXP", "CROPDMG", "CROPDMGEXP")
data=data[var]
head(data,5)## EVTYPE FATALITIES INJURIES PROPDMG PROPDMGEXP CROPDMG CROPDMGEXP
## 1 TORNADO 0 15 25.0 K 0
## 2 TORNADO 0 0 2.5 K 0
## 3 TORNADO 0 2 25.0 K 0
## 4 TORNADO 0 2 2.5 K 0
## 5 TORNADO 0 2 2.5 K 0Results
1. Health Effects
1.1 Calculating the total number of injuries and fatalities accross the US. Then the top 5 worst impacted weather events will be extracted
#total value for fatality and injury
fatal <- aggregate(FATALITIES ~ EVTYPE, data, FUN = sum)
injury <- aggregate(INJURIES ~ EVTYPE, data, FUN = sum)
#top 5 largest values
fataltop <- fatal[order(-fatal$FATALITIES),][1:5, ]
injurytop <- injury[order(-injury$INJURIES),][1:5, ]
fataltop## EVTYPE FATALITIES
## 834 TORNADO 5633
## 130 EXCESSIVE HEAT 1903
## 153 FLASH FLOOD 978
## 275 HEAT 937
## 464 LIGHTNING 816injurytop## EVTYPE INJURIES
## 834 TORNADO 91346
## 856 TSTM WIND 6957
## 170 FLOOD 6789
## 130 EXCESSIVE HEAT 6525
## 464 LIGHTNING 52302.3 Picture the results by plotting
library(ggplot2)
library(gridExtra)
plot1=ggplot(injurytop, aes(y=INJURIES,x=EVTYPE,label=INJURIES)) +
geom_bar(fill="steelblue",stat="identity")+ coord_flip()+
ylab("Total number of injuries") + xlab("Event type") +
ggtitle("Injuries by weather events in the US - Top 5")
plot2=ggplot(fataltop, aes(x=EVTYPE,y=FATALITIES,label=FATALITIES)) +
geom_bar(fill="firebrick",stat="identity") +coord_flip()+
ylab("Total number of fatalities") + xlab("Event type") +
ggtitle("Fatalities by weather events in the US - Top 5")
grid.arrange(plot1, plot2, nrow =2)
2. Economic Effects
2.1 Proper value of damage
Since the value of damage is recorded as 2 variables: the number and the exponent, it is neccessary to produce another proper variable for the estimate property damage and crop damage. This value will be generated by multiplying the number of damage and its exponent. Moreover, the exponent variable, which is recorded as letter or unable to be recorded, need to convert to usable number. At the end, 2 more variables will be added in the data which are property exponent (PROPEXP) and crop exponent (CROPEXP). Now, value of property damage and crop damage can be calculated to property damage value(PROPDMGVAL) and crop damage value(CROPDMGVAL)
#Convert exponent of property damage to usable number
data$PROPDMGEXP=as.character(data$PROPDMGEXP)
exp=as.character(unique(data$PROPDMGEXP))
value=c(1000,1e+06,1,1e+09,1e+06,0,1,1e+05,1e+06,0,10000,100,1000,100,1e+07,100,0,10,1e+08)
for (i in 1:length(exp)){
data$PROPEXP[data$PROPDMGEXP==exp[i]]=value[i]
}
# Calculating the property damage value
data$PROPDMGVAL <- data$PROPDMG * as.numeric(data$PROPEXP)
# Convert exponent of crop damage to usable number
cropexp=as.character(unique(data$CROPDMGEXP))
cropvalue=c(1,1e+06,1000,1e+06,1e+09,0,1,1000,100)
for(i in 1:length(cropexp)){
data$CROPEXP[data$CROPDMGEXP == cropexp[i]] = cropvalue[i]
}
# calculating the crop damage value
data$CROPDMGVAL <- data$CROPDMG * data$CROPEXP
head(data,5)## EVTYPE FATALITIES INJURIES PROPDMG PROPDMGEXP CROPDMG CROPDMGEXP
## 1 TORNADO 0 15 25.0 K 0
## 2 TORNADO 0 0 2.5 K 0
## 3 TORNADO 0 2 25.0 K 0
## 4 TORNADO 0 2 2.5 K 0
## 5 TORNADO 0 2 2.5 K 0
## PROPEXP PROPDMGVAL CROPEXP CROPDMGVAL
## 1 1000 25000 1 0
## 2 1000 2500 1 0
## 3 1000 25000 1 0
## 4 1000 2500 1 0
## 5 1000 2500 1 02.2 Calculating the total damage value for each events across the US. Select top 5 worst damaged weather events on property and crop
#total value of damage on property and crop
propdmg <- aggregate(PROPDMGVAL ~ EVTYPE, data, FUN = sum)
cropdmg <- aggregate(CROPDMGVAL ~ EVTYPE, data, FUN = sum)
#top 5 worst damage events
propdmgtop <- propdmg[order(-propdmg$PROPDMGVAL),][1:5, ]
cropdmgtop <- cropdmg[order(-cropdmg$CROPDMGVAL), ][1:5, ]
propdmgtop## EVTYPE PROPDMGVAL
## 170 FLOOD 144657709807
## 411 HURRICANE/TYPHOON 69305840000
## 834 TORNADO 56947380617
## 670 STORM SURGE 43323536000
## 153 FLASH FLOOD 16822673979cropdmgtop## EVTYPE CROPDMGVAL
## 95 DROUGHT 13972566000
## 170 FLOOD 5661968450
## 590 RIVER FLOOD 5029459000
## 427 ICE STORM 5022113500
## 244 HAIL 30259544732.3 Picture the results by plotting
plot1=ggplot(data=propdmgtop, aes(y=PROPDMGVAL,x=EVTYPE,label=PROPDMGVAL)) +
geom_bar(fill="steelblue",stat="identity")+ coord_flip()+
ylab("Total value of propperty damage") + xlab("Event type") +
ggtitle("Property damaged by weather events in the US - Top 5")
plot2=ggplot(cropdmgtop, aes(x=EVTYPE,y=CROPDMGVAL,label=CROPDMGVAL)) +
geom_bar(fill="firebrick",stat="identity") +coord_flip()+
ylab("Total value of crop damage") + xlab("Event type") +
ggtitle("Crop damaged by weather events in the US - Top 5")
grid.arrange(plot1, plot2, nrow =2)
Conclusion
In conclusion, Tornado has the worst impact on population health regarding both number of injuries and number of fatalities. On the aspect of economic, Flood has the worst impact on property while Draught play the major factor on damage of crop.