SvrWeatherEventsAssess
Chilly Amador
June 20, 2015
This is an R Markdown document.
NOAA - SEVERE WEATHER EVENTS ASSESSMENT (1950-2011)
Synopsis
The purpose of this document is to identify the weather events that have the most impact across the USA, both in human and material terms, and answer two questions: 1. Which are the most Harmful events with respect to population health 2. Which events have the greatest economic consequences
This document is divided in two sections. The first, Data Processing to determine the most harmful events and the second Data Processing to determine the events with greatest economic consequences. In each section we explain the origin of the data that was used for the analysis, and any transformations and processes that were applied to it in order to obtain our results. Each part has its own Results section, and there is a Summary at the end of the document.
ENVIRONMENT SET UP
options(warn=-1)
library(knitr)
opts_chunk$set(echo = TRUE, results = 'hold')
library(ggplot2)
library(dplyr)##
## Attaching package: 'dplyr'
##
## The following object is masked from 'package:stats':
##
## filter
##
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionoptions(warn=0)
workingDirectory <- "~/Documents/COURSERAReproducible/RepData_PeerAssessment2"
setwd(workingDirectory)DATA LOADING
Download file from U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database.
WD <- read.csv("repdata-data-StormData.csv.bz2", header=TRUE, sep = ",")
# Original Dataset
head(WD, 10)## 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
## 7 1 11/16/1951 0:00:00 0100 CST 9 BLOUNT AL
## 8 1 1/22/1952 0:00:00 0900 CST 123 TALLAPOOSA AL
## 9 1 2/13/1952 0:00:00 2000 CST 125 TUSCALOOSA AL
## 10 1 2/13/1952 0:00:00 2000 CST 57 FAYETTE 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
## 7 TORNADO 0 0
## 8 TORNADO 0 0
## 9 TORNADO 0 0
## 10 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
## 7 NA 0 1.5 33 2 0 0
## 8 NA 0 0.0 33 1 0 0
## 9 NA 0 3.3 100 3 0 1
## 10 NA 0 2.3 100 3 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
## 7 1 2.5 K 0
## 8 0 2.5 K 0
## 9 14 25.0 K 0
## 10 0 25.0 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 6
## 7 3405 8631 0 0 7
## 8 3255 8558 0 0 8
## 9 3334 8740 3336 8738 9
## 10 3336 8738 3337 8737 10DATA PREPROCESSING TO DETERMINE MOST HARMFUL EVENTS
Create data subset with variables of interest and eliminate instances with value of zero.
A data subset was created with the relevant variables for the analysis. This also helped to reduce the size of the dataset and speed up processing. Instances with relevant variables that were 0, were eliminated because they didn’t contribute any information. A standarizing process was run on the data to convert EVTYPE attribute to uppercase and eliminate blanks. Columns PROPDMG and PROPDMGEXP were eliminated to create a WDinjury subset to calculate only events’ injuries and fatalities. They will be recover in the second part on the processing. Reducing the number of attributes helped to minimize processing time and made dataset easier to visualize.
WDset <- select(WD, EVTYPE, FATALITIES, INJURIES, PROPDMG, PROPDMGEXP)
WDset <- filter(WDset, FATALITIES!=0 | INJURIES!=0 | PROPDMG!=0)
WDTidy <- WDset %>% mutate(EVTYPE = sub("^ +", "", EVTYPE))
WDTidy <- group_by(WDTidy, EVTYPE)
WDTidy <- arrange(WDTidy, desc(EVTYPE))
WDinjury <- WDTidy
WDinjury$PROPDMG <- NULL
WDinjury$PROPDMGEXP <- NULL
# First cleaning process to reduce the dataset size by reducing typos, eliminating extra spaces
# and grouping eventData <- group_by(WDinjury, EVTYPE)
eventData$EVTYPE <- toupper(eventData$EVTYPE)
eventData$EVTYPE <- sub("^ +","", eventData$EVTYPE)
eventData <- group_by(eventData, EVTYPE)
eventData <- summarise(eventData,
SumFATALITIES = sum(FATALITIES),
SumINJURIES = sum(INJURIES))
eventData <- arrange(eventData, EVTYPE)Create Reclassification Event Table based on NOAA’s Event Table
In order to group all the events under a simplified event code, we created an Event Table based on NOAA’s. However, some events were not clearly classified in the official table and they required to be reclassified using common sense, especially when the event name was incorrectly spelled, and it was a common mistake. Reclassification criteria can be reviewed below.
event = list()
event["astronomical low tide"] = list(c("astronomical low tide","ASTRONOMICAL LOW"))
event["avalanche"] = list(c("avalanche","AVALANCE", "AVALANCHE"))
event["blizzard"] = list(c("blizzard", "BLIZZARD", "BLIZZARD/WINTER STORM", "GROUND BLIZZARD", "BLIZZ"))
event["coastal flood"] = list(c("coastal flood", "COASTAL FLOODING/EROSION", "COASTAL FLOOD", "COASTAL EROSION","EROSION/CSTL FLOOD", "COASTAL SURGE"))
event["coastal storm"] = list(c("coastal storm", "COASTAL STORM", "COASTALSTORM"))
event["cold/wind chill"] = list(c("cold/wind chill","COLD","LOW TEMPERATURE"))
event["debris flow"] = list(c("debris flow","LANDSLIDES", "LANDSLIDE", "LANDSLUMP", "LANDSPOUT", "MUD SLIDES", "MUDSLIDE", "MUDSLIDES"))
event["dense fog"] = list(c("dense fog","FOG"))
event["dense smoke"] = list(c("dense smoke","DENSE SMOKE"))
event["drought"] = list(c("drought","DROUGHT"))
event["dust devil"] = list(c("dust devil","DUST DEVIL"))
event["dust storm"] = list(c("dust storm","DUST STORM"))
event["excessive heat"] = list(c("excessive heat","EXTREME HEAT", "HYPERTHERMIA"))
event["excessive cold/wind chill"] = list(c("excessive cold/wind chill","EXCESSIVE COLD", "EXTREME COLD", "EXTREME WINDCHILL"))
event["flood"] = list(c("flood","MAJOR FLOOD", "FLOOD", "MINOR FLOOD", "RIVER FLOOD", "RIVER FLOODING", "RURAL FLOOD", "SEVERE RAISING WATER",
"URBAN", "DAM BREAK"))
event["flash flood"] = list(c("flash flood","FLASH FLOOD", "FLASH FLOODING", "FLASH FLOODS", "URBAN FLOOD", "URBAN FLOODS", "URBAN FLOODING"))
event["frost/freeze"] = list(c("frost/freeze","FREEZING", "FROST", "FREEZE"))
event["funnel cloud"]= list(c("funnel cloud","FUNNEL CLOUD"))
event["freezing fog"]= list(c("freezing fog", "FREEZING FOG"))
event["hail"] = list(c("hail","HAIL", "HAILSTORM"))
event["heat"] = list(c("heat","HEAT"))
event["heavy rain"] = list(c("heavy rain","HEAVY RAIN", "HEAVY RAINS", "PRECIPITATION", "PRECIP", "HEAVY SHOWER", "RAIN",
"RAINSTORM", "RAINFALL"))
event["heavy snow"] = list(c("heavy snow","HEAVY SNOW","HEAVY SNOWPACK", "Y SNOW", "SIVE SNOW", "ING SNOW",
"RECORD SNOW", "SNOW$", "SNOW SQUALL"))
event["heavy surf"] = list(c("heavy surf","HEAVY SURF", "SWELLS"))
event["high surf"] = list(c("high surf","HIGH SURF", "HEAVY SEAS"))
event["high wind"] = list(c("high wind","HIGH WIND", "HIGH WINDS"))
event["hurricane/typhoon"] = list(c("hurricane/typhoon","HURRICANE", "TYPHOON"))
event["extreme cold/wind chill"] = list(c("extreme cold/wind chill","HYPOTHERMIA"))
event["ice storm"] = list(c("ice storm","ICE STORM", "ICE"))
event["lake-effect snow"] = list(c("lake-effect snow","LAKE EFFECT", "LAKE-EFFECT"))
event["lakeshore flood"] = list(c("lakeshore flood","LAKE FLOOD", "LAKESHORE FLOOD"))
event["lightning"] = list(c("lightning","LIGHTINING"))
event["marine hail"] = list(c("marine hail","MARINE HAIL"))
event["marine high wind"] = list(c("marine high wind","MARINE HIGH WIND"))
event["marine strong wind"] = list(c("marine strong wind","MARINE STRONG WIND"))
event["marine thunderstorm wind"] = list(c("marine thunderstorm wind","MARINE THUNDERSTORM WIND"))
event["microburst winds"] = list(c("high wind","MICROBURST"))
event["rip current"] = list(c("rip current","RIP CURRENTS", "RIP CURRENT"))
event["seiche"]= list(c("seiche","SEICHE"))
event["sleet"] = list(c("sleet","SLEET", "GLAZE"))
event["storm surge/tide"] = list(c("storm surge/tide","STORM SURGE"))
event["thunderstorm wind"] = list(c("thunderstorm wind","THUNDERSTORM", "THUNDERSTORMS", "TSTM", "TSTMW", "THUDERSTORM WINDS", "THUNDEERSTORM WINDS",
"THUNDERESTORM WINDS","THUNDERSNOW", "THUNDERSTORMW", "THUNDERSTORMWINDS", "THUNDERSTROM WIND",
"THUNDERTORM WINDS", "THUNERSTORM WINDS", "THUNDERSTORM", "THUNDERSTORMs", "TUNDERSTORM", "WIND STORM"))
event["tornado"] = list(c("tornado","TORNADO", "TORNDAO", "TORNADOES"))
event["tropical depression"] = list(c("tropical depression","TROPICAL DEPRESSION"))
event["tropical storm"] = list(c("tropical storm","TROPICAL STORM"))
event["tsunami"] = list(c("tsunami","TSUNAMI"))
event["volcanic ash"] = list(c("volcanic ash","VOLCANIC ASH"))
event["excessive heat"] = list(c("excessive heat","WARM"))
event["waterspout"] = list(c("waterspout","WATERSPOUT"))
event["wind"] = list(c("high wind","STRONG WIND", "STRONG WINDS", "WIND", "WIND DAMAGE", "WINDS", "WIND AND WAVE",
"EXTREME WIND", "GUSTY", "GUSTY WIND", "GUSTY WINDS", "DRY MIRCOBURST WINDS", "DRY MICROBURST", "GUSTNADO"))
event["wildfire"] = list(c("wildfire","WILDFIRE", "FOREST FIRE", "FOREST", "FIRES", "WILDFIRES"))
event["winter storm"] = list(c("winter storm","WINTER STORM"))
event["winter weather"] = list(c("winter weather","WINTER WEATHER", "WINTRY MIX"))Events Reclassification for Injuries/Fatalities Dataset
Events’ names were standarized using the following process and the event table to replace the event’s name (uppercase) with the official name (lowercase).
# Create reclassification flag-column
eventData = mutate(eventData, NEVTYPE = "unclassified")
# Reclassifing events in accordance to NOAA's table.
for (i in 1:nrow(eventData)) {
for (ii in 1:length(event)) {
listLength = length(event[[ii]])
for (iii in 1:listLength) {
if ( grepl(event[[ii]][iii], eventData$EVTYPE[i], ignore.case=TRUE) & eventData$NEVTYPE[i] == "unclassified") {
eventData$NEVTYPE[i] <- event[[ii]][1]
}
}
}
}
# Dataset reclassified
head(eventData, 10, addrownums = FALSE)
# Check number of unclassified events
unclassifiedEvents <- filter(eventData, NEVTYPE =="unclassified")
# Number of unclassified events
nrow(unclassifiedEvents)
# Unclassified Events Dataset
head(unclassifiedEvents)## Source: local data frame [10 x 4]
##
## EVTYPE SumFATALITIES SumINJURIES NEVTYPE
## 1 ? 0 0 unclassified
## 2 APACHE COUNTY 0 0 unclassified
## 3 ASTRONOMICAL HIGH TIDE 0 0 unclassified
## 4 ASTRONOMICAL LOW TIDE 0 0 astronomical low tide
## 5 AVALANCE 1 0 avalanche
## 6 AVALANCHE 224 170 avalanche
## 7 BEACH EROSION 0 0 unclassified
## 8 BLACK ICE 1 24 ice storm
## 9 BLIZZARD 101 805 blizzard
## 10 BLIZZARD/WINTER STORM 0 0 blizzard
## [1] 31
## Source: local data frame [6 x 4]
##
## EVTYPE SumFATALITIES SumINJURIES NEVTYPE
## 1 ? 0 0 unclassified
## 2 APACHE COUNTY 0 0 unclassified
## 3 ASTRONOMICAL HIGH TIDE 0 0 unclassified
## 4 BEACH EROSION 0 0 unclassified
## 5 BLOWING DUST 0 0 unclassified
## 6 BRUSH FIRE 0 2 unclassifiedCalculate Most Harmful Events Across the US. Calculated total injuries and fatalities.
# Preprocessing Injuries/Fatalities Subset to reduce redundancy and unidentified events.
injuryData <- eventData
injuryData$EVTYPE <- NULL
head(injuryData, 10)
injuryData <- group_by(injuryData, NEVTYPE)
injuryData <- summarise(injuryData,
FATALITIES = sum(SumFATALITIES),
INJURIES = sum(SumINJURIES))
injuryData <- arrange(injuryData, NEVTYPE)
injuryData <- mutate(injuryData, TOTAL = 0)
injuryData <- mutate(injuryData, TOTAL = FATALITIES + INJURIES)
injuryData <- arrange(injuryData, desc(TOTAL))## Source: local data frame [10 x 3]
##
## SumFATALITIES SumINJURIES NEVTYPE
## 1 0 0 unclassified
## 2 0 0 unclassified
## 3 0 0 unclassified
## 4 0 0 astronomical low tide
## 5 1 0 avalanche
## 6 224 170 avalanche
## 7 0 0 unclassified
## 8 1 24 ice storm
## 9 101 805 blizzard
## 10 0 0 blizzardRESULTS: MOST HARMFUL EVENTS ACROSS THE US
The analysis found that the ten most dangerous events across the US are: tornado, flood, thunderstorm wind, excessive heat, lightning, heat, high wind, ice storm, wildfire and winter storm. The 10 Most Harmful Events account for 92% (143,742) of the total injuries and fatalities (155,673).
# Total number of injuries and fatalities in all the events: 155,673
TotalInjury <- sum(injuryData$TOTAL)
# 10 Most harmful events across the US
injuryPlot <- injuryData[1:10,]
# Total number of injuries and fatalities in the 10 Most Harmful Events: 143,742
TotalInjury_10 <- sum(injuryPlot$TOTAL)
# The 10 Most Harmful Events account for 92% of the total injuries and fatalities.
injuryPlot## Source: local data frame [10 x 4]
##
## NEVTYPE FATALITIES INJURIES TOTAL
## 1 tornado 5636 91407 97043
## 2 flood 1547 8676 10223
## 3 thunderstorm wind 716 9421 10137
## 4 excessive heat 1960 6544 8504
## 5 lightning 817 5232 6049
## 6 heat 1212 2684 3896
## 7 high wind 438 1948 2386
## 8 ice storm 101 2153 2254
## 9 wildfire 90 1606 1696
## 10 winter storm 216 1338 1554Most harmful events results plot
# Creating labels and adjusting parameters
injuryTots <- injuryPlot$TOTAL
injuryNames <- injuryPlot$NEVTYPE
colors <-rainbow(length(injuryTots))
injuryLabs <- as.character(injuryTots)
injuryLabs <- paste(injuryLabs, "victims")
# Pie plotting
pie(injuryTots, main="Most Harmful Events across the US", col=colors,
labels=injuryNames, cex=0.7)
legend(-1.7, 1, injuryLabs, cex=0.7,
fill=colors)
CALCULATE THE EVENTS THAT HAVE THE GREATEST ECONOMIC CONSEQUENCES
DATA PROCESSING TO DETERMINE THE EVENTS THAT HAVE THE GREATEST ECONOMIC CONSEQUENCE ACROSS DE US
A data subset was created with the relevant variables to determine the events that have the greatest economic consequences. This also helped to reduce the size of the dataset and speed up processing. Instances with relevant variables that were 0, were eliminated because they didn’t contribute any information. To calculate the economic consequences we only considered Property Damage and used the exponetial provided to calculate the amount in US Dollars as instructed in the National Weather Service Instruction (https://d396qusza40orc.cloudfront.net/repdata%2Fpeer2_doc%2Fpd01016005curr.pdf). A standarizing process was run on the data to convert EVTYPE attribute to uppercase, eliminate blanks and eliminated unvalid values from exponent attribute PROPDMGEXP. INJURIES and FATALITIES columns were eliminated to create a WDinjury subset to calculate only events’ injuries and fatalities. They will be recover in the second part on the processing. Reducing the number of attributes helped to minimize processing time and made dataset easier to visualize.
# Eliminate unnecessary columns to expedite calculations
WDdamage <- WDTidy
WDdamage$INJURIES<- NULL
WDdamage$FATALITIES <- NULL
head(WDdamage, 10)
# Event reclassification processing. Use the event table generated in the first part.
WDdamage$EVTYPE <- toupper(WDdamage$EVTYPE)
WDdamage$EVTYPE <- sub("^ +","", WDdamage$EVTYPE)
WDdamage <- filter(WDdamage, PROPDMG !=0)
attach(WDdamage)
aggDMG <-aggregate(PROPDMG, by =list(EVTYPE,PROPDMGEXP), FUN=sum)
detach(WDdamage)
colnames(aggDMG) <- c("EVTYPE", "EXP", "PROPDMG")
aggDMG <- arrange(aggDMG, EVTYPE)## Source: local data frame [10 x 3]
## Groups: EVTYPE
##
## EVTYPE PROPDMG PROPDMGEXP
## 1 ? 5.0 K
## 2 APACHE COUNTY 5.0 K
## 3 ASTRONOMICAL HIGH TIDE 1.0 M
## 4 ASTRONOMICAL HIGH TIDE 5.0 M
## 5 ASTRONOMICAL HIGH TIDE 90.0 K
## 6 ASTRONOMICAL HIGH TIDE 2.5 M
## 7 ASTRONOMICAL HIGH TIDE 20.0 K
## 8 ASTRONOMICAL HIGH TIDE 800.0 K
## 9 ASTRONOMICAL HIGH TIDE 5.0 K
## 10 ASTRONOMICAL HIGH TIDE 10.0 KStandarize EXPTYPE (exponential values to K(103),M(106) and B(10^9))
# Create reclassification flag-column
aggDMG = mutate(aggDMG, EXPTYPE = "undefined")
# Eliminate undefined exp values that will allow to calculate the values in US dollars.
for (y in 1:nrow(aggDMG)) {
if (as.character(aggDMG[y,2]) == "K") {
aggDMG[y,4] <- "K"}
if (as.character(aggDMG[y,2]) == "M") {
aggDMG[y,4] <- "M"}
if (as.character(aggDMG[y,2]) == "B") {
aggDMG[y,4] <- "B"}
}Events Reclassification for Economic Consequences Dataset
Events’ names were standarized using the following process and the event table created before to replace the event’s name (uppercase) with the official name (lowercase).
# Create flag-column for exponential type undefined that need to be eliminated
aggDMG <- filter(aggDMG, EXPTYPE !="undefined")
aggDMG$EXP <- NULL
# Reclassifiying events in accordance to the NOAA's table
aggDMG = mutate(aggDMG, NEVTYPE = "unclassified")
for (i in 1:nrow(aggDMG)) {
for (ii in 1:length(event)) {
listLength = length(event[[ii]])
for (iii in 1:listLength) {
if ( grepl(event[[ii]][iii], aggDMG$EVTYPE[i], ignore.case=TRUE) & aggDMG$NEVTYPE[i] == "unclassified") {
aggDMG$NEVTYPE[i] <- event[[ii]][1]
}
}
}
}
head(aggDMG, 10, addrownums = FALSE)
# Check unclassified events
unclassifiedDamageEvents <- filter(aggDMG, NEVTYPE =="unclassified")
# Number of unclassified events in damage calculations
nrow(unclassifiedDamageEvents)
# Unclassified events dataset
head(unclassifiedDamageEvents,10)## EVTYPE PROPDMG EXPTYPE NEVTYPE
## 1 ? 5.00 K unclassified
## 2 APACHE COUNTY 5.00 K unclassified
## 3 ASTRONOMICAL HIGH TIDE 925.00 K unclassified
## 4 ASTRONOMICAL HIGH TIDE 8.50 M unclassified
## 5 ASTRONOMICAL LOW TIDE 320.00 K astronomical low tide
## 6 AVALANCHE 1621.80 K avalanche
## 7 AVALANCHE 2.10 M avalanche
## 8 BEACH EROSION 100.00 K unclassified
## 9 BLIZZARD 24683.95 K blizzard
## 10 BLIZZARD 634.53 M blizzard
## [1] 23
## EVTYPE PROPDMG EXPTYPE NEVTYPE
## 1 ? 5.0 K unclassified
## 2 APACHE COUNTY 5.0 K unclassified
## 3 ASTRONOMICAL HIGH TIDE 925.0 K unclassified
## 4 ASTRONOMICAL HIGH TIDE 8.5 M unclassified
## 5 BEACH EROSION 100.0 K unclassified
## 6 BLOWING DUST 20.0 K unclassified
## 7 BRUSH FIRE 55.0 K unclassified
## 8 DOWNBURST 2.0 K unclassified
## 9 HEAVY MIX 705.0 K unclassified
## 10 HEAVY MIX 0.6 M unclassifiedCalculate total damage in dollars using the exponential code (K,M,B)
aggDMG$EVTYPE <- NULL
attach(aggDMG)
aggDMGtotal <-aggregate(PROPDMG, by =list(NEVTYPE,EXPTYPE), FUN=sum)
detach(aggDMG)
colnames(aggDMGtotal) <- c("EVTYPE", "EXPTYPE", "PROPDMG")
aggDMGtotal <- mutate(aggDMGtotal, PROPDMGDLS = 0.0)
aggDMGtotal <- arrange(aggDMGtotal, EVTYPE)
for (z in 1:nrow(aggDMGtotal)) {
if (as.character(aggDMG[z,2]) == "K") {
factorExp <- 10^3 }
if (as.character(aggDMG[z,2]) == "M") {
factorExp <- 10^6 }
if (as.character(aggDMG[z,2]) == "B") {
factorExp <- 10^9 }
aggDMGtotal[z,4] <- aggDMGtotal[z,3] * factorExp
}
damageData <- aggDMGtotal
damageData$EXPTYPE <- NULL
damageData$PROPDMG <- NULL
attach(damageData)
damageDatatotal <- aggregate(PROPDMGDLS, by=list(EVTYPE), FUN=sum)
detach(damageData)
colnames(damageDatatotal) <- c("EVTYPE","DAMAGEDLS")
damageDatatotal <- arrange(damageDatatotal, desc(DAMAGEDLS))RESULTS: EVENTS WITH GREATEST ECONOMIC CONSEQUENCES IN THE US
The analysis found that Total Economic Consequences Across the US are $8,578,573,119,270 USDS and 99% of this amount ($8,563,126,977,920 USDS) can be accounted by the 10 Event with the Greatest Economic Consequences: high win, thunderstorm wind, hail, wildfire, ice storms, heavy rain, tornados, flood, debris flows and coastal flood.
# Calculate Total Economic Consequences Across the US
damageTotal <- sum(damageDatatotal$DAMAGEDLS)
# Total Economic Consequences Across the US: $8,578,573,119,270 USDS
damageTotal
# Dataset: Ten events with the greatest economic consequences
head(damageDatatotal, 10)
# Prepare data for plotting
dmgPlot <- damageDatatotal[1:10,]
# Total of Ten events with the greatest economic consequences
damageTotal_10 <- sum(dmgPlot$DAMAGEDLS)
# Total of Ten events with the greatest economic consequences: $8,563,126,977,920
damageTotal_10
# The ten first events with the greatest economic consequences account for 99% of the total amount
# of economic consequences## [1] 3.976507e+12
## EVTYPE DAMAGEDLS
## 1 tropical storm 2.564391e+12
## 2 hail 6.854443e+11
## 3 high wind 4.483987e+11
## 4 rip current 1.630000e+11
## 5 tornado 5.168160e+10
## 6 blizzard 2.538458e+10
## 7 hurricane/typhoon 1.093641e+10
## 8 thunderstorm wind 9.727984e+09
## 9 ice storm 3.962639e+09
## 10 unclassified 3.694309e+09
## [1] 3.966621e+12Plot events with the greatest economic consequences across the US
# Adjust number of digits to fit the plot size (showing only 4 significant digits)
dmgPlot$DAMAGEDLS <- dmgPlot$DAMAGEDLS/10^8
dmgPlot$DAMAGEDLS <- signif(dmgPlot$DAMAGEDLS,4)
dmgPlot
# Prepare plot labels
dmgTots <- dmgPlot$DAMAGEDLS
dmgNames <- dmgPlot$EVTYPE
colors <-rainbow(length(dmgTots))
dmgLabs <- as.character(dmgTots)
dmgLabs <- paste(dmgLabs, " *10^8 USDS")
# Plot histogram of Total Economic Consequences
barplot(dmgTots, main="Events w/Greatest Economic Consequences", xlab= "EVENT",
ylab="Economic Consequences (10^8 USDLS)", names.arg=dmgNames, col=colors,
space=0.1, cex.axis=0.6, cex.names=0.6, las=1)
legend("topright", dmgLabs, cex=0.6,
bty="n", fill=colors)
## EVTYPE DAMAGEDLS
## 1 tropical storm 25640.00
## 2 hail 6854.00
## 3 high wind 4484.00
## 4 rip current 1630.00
## 5 tornado 516.80
## 6 blizzard 253.80
## 7 hurricane/typhoon 109.40
## 8 thunderstorm wind 97.28
## 9 ice storm 39.63
## 10 unclassified 36.94SUMMARY
The ten most dangerous events across the US are: tornado, flood, thunderstorm wind, excessive heat, lightning, heat, high wind, ice storm, wildfire and winter storm. The 10 Most Harmful Events account for 92% (143,742) of the total injuries and fatalities (155,673). The Total Economic Consequences Across the US are $8,578,573,119,270 USDS and 99% of this amount ($8,563,126,977,920 USDS) can be accounted by the 10 Event with the Greatest Economic Consequences: high win, thunderstorm wind, hail, wildfire, ice storms, heavy rain, tornados, flood, debris flows and coastal flood.