Effects of Severe Weather Events over Public Health and Economy across United States from 1950 to 2011
Among the many factors involved in the cause of diseases or morbidities are those related to the environment in which they live susceptible subject. Storm and other severe weather events are among them.
This design checks across the United States, which types of these events are more harmful to public health and which bring greater economic losses.
The results show that among these events, tornadoes appear to be associated with the largest measurement of fatalities, flooding with high measures of injury, hurricanes / typhoons with the greatest damages to agriculture and storm with higher damage to property.
Project Objectives
Check weather events which are related to the main events of fatalities, injuries and property damage, in order to plan public prevention and risk reduction.
Data Processing
The data analyzed is from storm database of the U.S. National Oceanic and Atmospheric Administration’s (NOAA), which can be downloaded via the link https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2 (but the code can do it automatically). For more informations, see National Weather Service Storm Data (Documentation) and National Climatic Data Center Storm Events (FAQ).
Once downloaded and uncompressed data file, an initial exploration showed 902,297 observations on 37 variables, from which were selected 8 of interest:
- BGN_DATE (factor) as date of event.
- EVTYPE (factor) as event type.
- FATALITIES (numeric) as measure of harm to human health.
- INJURIES (numeric )as measure of harm to human health.
- PROPDMG (numeric) as measure of property damage and hence economic damage.
- PROPDMGEXP (factor) as magnitude of property damage in USD.
- CROPDMG (numeric) as measure of crop damage and hence economic damage in USD.
- CROPDMGEXP (factor) as magnitude of crop damage.
filename = "repdata-data-StormData.csv.bz2"
if (!file.exists(filename)) {
fileURL = "https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2"
download.file(fileURL, filename)
}
if (!file.exists("repdata-data-StormData.csv")) {
library(R.utils)
bunzip2(filename)
}
storm = read.csv("repdata-data-StormData.csv")
dim(storm)## [1] 902297 37str(storm)## 'data.frame': 902297 obs. of 37 variables:
## $ STATE__ : num 1 1 1 1 1 1 1 1 1 1 ...
## $ BGN_DATE : Factor w/ 16335 levels "1/1/1966 0:00:00",..: 6523 6523 4242 11116 2224 2224 2260 383 3980 3980 ...
## $ BGN_TIME : Factor w/ 3608 levels "00:00:00 AM",..: 272 287 2705 1683 2584 3186 242 1683 3186 3186 ...
## $ TIME_ZONE : Factor w/ 22 levels "ADT","AKS","AST",..: 7 7 7 7 7 7 7 7 7 7 ...
## $ COUNTY : num 97 3 57 89 43 77 9 123 125 57 ...
## $ COUNTYNAME: Factor w/ 29601 levels "","5NM E OF MACKINAC BRIDGE TO PRESQUE ISLE LT MI",..: 13513 1873 4598 10592 4372 10094 1973 23873 24418 4598 ...
## $ STATE : Factor w/ 72 levels "AK","AL","AM",..: 2 2 2 2 2 2 2 2 2 2 ...
## $ EVTYPE : Factor w/ 985 levels " HIGH SURF ADVISORY",..: 834 834 834 834 834 834 834 834 834 834 ...
## $ BGN_RANGE : num 0 0 0 0 0 0 0 0 0 0 ...
## $ BGN_AZI : Factor w/ 35 levels ""," N"," NW",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ BGN_LOCATI: Factor w/ 54429 levels "","- 1 N Albion",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ END_DATE : Factor w/ 6663 levels "","1/1/1993 0:00:00",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ END_TIME : Factor w/ 3647 levels ""," 0900CST",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ COUNTY_END: num 0 0 0 0 0 0 0 0 0 0 ...
## $ COUNTYENDN: logi NA NA NA NA NA NA ...
## $ END_RANGE : num 0 0 0 0 0 0 0 0 0 0 ...
## $ END_AZI : Factor w/ 24 levels "","E","ENE","ESE",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ END_LOCATI: Factor w/ 34506 levels "","- .5 NNW",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ LENGTH : num 14 2 0.1 0 0 1.5 1.5 0 3.3 2.3 ...
## $ WIDTH : num 100 150 123 100 150 177 33 33 100 100 ...
## $ F : int 3 2 2 2 2 2 2 1 3 3 ...
## $ MAG : num 0 0 0 0 0 0 0 0 0 0 ...
## $ 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 ...
## $ WFO : Factor w/ 542 levels ""," CI","$AC",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ STATEOFFIC: Factor w/ 250 levels "","ALABAMA, Central",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ ZONENAMES : Factor w/ 25112 levels ""," "| __truncated__,..: 1 1 1 1 1 1 1 1 1 1 ...
## $ LATITUDE : num 3040 3042 3340 3458 3412 ...
## $ LONGITUDE : num 8812 8755 8742 8626 8642 ...
## $ LATITUDE_E: num 3051 0 0 0 0 ...
## $ LONGITUDE_: num 8806 0 0 0 0 ...
## $ REMARKS : Factor w/ 436774 levels "","-2 at Deer Park\n",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ REFNUM : num 1 2 3 4 5 6 7 8 9 10 ...library(dplyr)
s = select(storm,
BGN_DATE,
EVTYPE,
FATALITIES,
INJURIES,
PROPDMG,
CROPDMG,
PROPDMGEXP,
CROPDMGEXP)Especially, it was necessary to adjust the values of losses measures.
table(s$PROPDMGEXP, s$CROPDMGEXP)##
## ? 0 2 B k K m M
## 461616 2 3 1 4 0 3865 0 443
## - 1 0 0 0 0 0 0 0 0
## ? 8 0 0 0 0 0 0 0 0
## + 5 0 0 0 0 0 0 0 0
## 0 211 0 0 0 0 0 4 0 1
## 1 25 0 0 0 0 0 0 0 0
## 2 13 0 0 0 0 0 0 0 0
## 3 3 0 0 0 0 0 1 0 0
## 4 4 0 0 0 0 0 0 0 0
## 5 26 0 0 0 0 0 1 0 1
## 6 4 0 0 0 0 0 0 0 0
## 7 5 0 0 0 0 0 0 0 0
## 8 1 0 0 0 0 0 0 0 0
## B 16 0 0 0 2 0 11 0 11
## h 1 0 0 0 0 0 0 0 0
## H 6 0 0 0 0 0 0 0 0
## K 149067 4 16 0 3 21 274690 0 864
## m 6 0 0 0 0 0 0 1 0
## M 7395 1 0 0 0 0 3260 0 674s$PROPDMGEXP = as.character(s$PROPDMGEXP)
s$PROPDMGEXP = gsub("\\-|\\+|\\?","0",s$PROPDMGEXP)
s$PROPDMGEXP = gsub("B|b", "9", s$PROPDMGEXP)
s$PROPDMGEXP = gsub("M|m", "6", s$PROPDMGEXP)
s$PROPDMGEXP = gsub("K|k", "3", s$PROPDMGEXP)
s$PROPDMGEXP = gsub("H|h", "2", s$PROPDMGEXP)
s$PROPDMGEXP = as.numeric(s$PROPDMGEXP)
s$PROPDMGEXP[is.na(s$PROPDMGEXP)] = 0
s$PROPDAM = s$PROPDMG * 10 ^ s$PROPDMGEXP
s$CROPDMGEXP = as.character(s$CROPDMGEXP)
s$CROPDMGEXP = gsub("\\-|\\+|\\?","0",s$CROPDMGEXP)
s$CROPDMGEXP = gsub("B|b", "9", s$CROPDMGEXP)
s$CROPDMGEXP = gsub("M|m", "6", s$CROPDMGEXP)
s$CROPDMGEXP = gsub("K|k", "3", s$CROPDMGEXP)
s$CROPDMGEXP = gsub("H|h", "2", s$CROPDMGEXP)
s$CROPDMGEXP = as.numeric(s$CROPDMGEXP)
s$CROPDMGEXP[is.na(s$CROPDMGEXP)] = 0
s$CROPDAM = s$CROPDMG * 10 ^ s$CROPDMGEXPThe variables were transformed (log scale, per billion) and the data were arranged to enable basic data exploring (including a better graph visualization).
It was adopted as those higher values greater than the 99th percentile for data no zero.
s = s[,-(5:8)]
s$BGN_DATE = as.Date(s$BGN_DATE, "%m/%d/%Y %H:%M:%S")
class(s$BGN_DATE)## [1] "Date"library(lubridate)
s$BGN_DATE = year(s$BGN_DATE)
library(reshape2)
dates = melt(s[,-2], id = "BGN_DATE",
measure.vars = c("FATALITIES", "INJURIES", "PROPDAM" ,"CROPDAM"))
dates = aggregate(value ~ BGN_DATE + variable, dates, sum)
summary(dates)## BGN_DATE variable value
## Min. :1950 FATALITIES:62 Min. :0.000e+00
## 1st Qu.:1965 INJURIES :62 1st Qu.:6.600e+01
## Median :1980 PROPDAM :62 Median :1.236e+03
## Mean :1980 CROPDAM :62 Mean :1.925e+09
## 3rd Qu.:1996 3rd Qu.:3.169e+08
## Max. :2011 Max. :1.219e+11dates$value = dates$value/1000000000
health = melt(s[,-1], id = "EVTYPE", measure.vars = c("FATALITIES", "INJURIES"))
summary(health)## EVTYPE variable value
## HAIL :577322 FATALITIES:902297 Min. :0.00e+00
## TSTM WIND :439880 INJURIES :902297 1st Qu.:0.00e+00
## THUNDERSTORM WIND:165126 Median :0.00e+00
## TORNADO :121304 Mean :8.63e-02
## FLASH FLOOD :108554 3rd Qu.:0.00e+00
## FLOOD : 50652 Max. :1.70e+03
## (Other) :341756health = health[health$value > 0, ]
quantile(health$value, c(.75, .8, .9, .95, .99, 1))## 75% 80% 90% 95% 99% 100%
## 3 4 10 20 90 1700health = health[(health$value > quantile(health$value, .99)), ]
health$value = log(health$value)
econ = melt(s[,-1], id = "EVTYPE", measure.vars = c("PROPDAM", "CROPDAM"))
summary(econ)## EVTYPE variable value
## HAIL :577322 PROPDAM:902297 Min. :0.000e+00
## TSTM WIND :439880 CROPDAM:902297 1st Qu.:0.000e+00
## THUNDERSTORM WIND:165126 Median :0.000e+00
## TORNADO :121304 Mean :2.645e+05
## FLASH FLOOD :108554 3rd Qu.:0.000e+00
## FLOOD : 50652 Max. :1.150e+11
## (Other) :341756econ = econ[econ$value > 0, ]
econ$value = log(econ$value)
quantile(econ$value, c(.75, .8, .9, .95, .99, 1))## 75% 80% 90% 95% 99% 100%
## 10.81978 11.00210 12.42922 13.65299 16.11810 25.46820econ = econ[econ$value > 18, ]
summary(econ$value)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 18.01 18.42 18.83 19.15 19.47 25.47We can see that last decade has more complete data.
library(ggplot2)
ggplot(dates,
aes(BGN_DATE, value)) +
geom_bar(stat = "identity") +
facet_grid(. ~ variable) +
xlab("Year") +
ylab("Value in USD (per billion)") +
ggtitle("Total of Damage Per Consequence Type Caused Per Severe Weather Events
Across United States from 1950 to 2011.") +
theme(axis.text.x = element_text(angle = 90))
Results
Below we can see that tornadoes were the major causes of fatalities and also had a strong impact on the number of injuries. However, when it comes to injuries, tornadoes are medially overcome by floods, hurricanes / typhoons and blizzard. Floods, high winds and hurricanes were the main cause of damage to crop and severe thunderstorms and storms surge were the most losses caused to property.
ggplot(health,
aes(EVTYPE, value, fill = variable)) +
geom_boxplot() +
xlab("Events Type") +
ylab("Value in USD (log scale)") +
ggtitle("Association Between Higher Values per Type Damage For Health and Type of Weather Events
Across United States from 1950 to 2011.") +
theme( axis.text.x = element_text(angle = 90)) +
guides(fill = guide_legend(title = "Type Damage", title.position = "top"))
ggplot(econ,
aes(EVTYPE, value, fill = variable)) +
geom_boxplot() +
xlab("Events Type") +
ylab("Value in USD (log scale)") +
ggtitle("Association Between Higher Values per Type Damage For Economy and Type of Weather Events
Across United States from 1950 to 2011.") +
theme( axis.text.x = element_text(angle = 90)) +
guides(fill = guide_legend(title = "Type Damage", title.position = "top"))