Impact of Weather Events on the US Population and Economy (1950-2011)

Executive Summary

This executive summary ranks weather-related fatalities, injuries and damages in billions of US dollars ($B) caused by severe weather during 1950-2011.

The top five fatality-related weather events were: 1) tornadoes, 2) heat waves, 3) floods, 4) lightning and 5) thunderstorm winds.

Weather Event Fatalities
Tornado 5,633
Heat 3,012
Flash Flood 1,488
Lightning 816
TSTM Wind 706

The top five injury-related weather events were: 1) tornadoes, 2) thunderstorm winds 3) floods, 4) heat waves and 5) lightning.

Weather Event Injuries
Tornado 91,346
TSTM Wind 6,957
Flood 6,789
Excessive Heat 6,525
Lightning 5,230

Floods accounted for $145B in property damages, hurricanes/typhoons $69B, tornadoes $43B and storm surges $43B.

Drought caused $14B in crop damages, floods and river flood $11B, ice storms $5B and hail storms $3B.

A. Storm Events Database

The Storm Events Database contains records on various types of severe weather, as collected by NOAA’s National Weather Service (NWS).

Bulk data download: repdata-data-StormData.csv (8/15/2014 10:52:50 AM)

Event Types Available:

  1. Tornado: From 1950 through 1954, only tornado events were recorded.

  2. Tornado, Thunderstorm Wind and Hail: From 1955 through 1992, only tornado, thunderstorm wind and hail events were keyed from the paper publications into digital data. From 1993 to 1995, only tornado, thunderstorm wind and hail events have been extracted from the Unformatted Text Files.

  3. All Event Types (48 from Directive 10-1605): From 1996 to present, 48 event types are recorded as defined in NWS Directive 10-1605.

Altogether there are 902,297 observations of 37 variables. Seven variables related to health outcomes and economic consequences by event type were selected from the database:

VARIABLE DESCRIPTION
EVTYPE Event type
FATALITIES Number of fatalities per event
INJURIES Number of injuries per event
PROPDMG Property damage per event
PROPDMGEXP (K/M/B) in thousands, millions or billions of dollars
CROPDMG Crop damage per event
CROPDMGEXP (K/M/B) in thousands, millions or billions of dollars 
setwd("C:/Users/d2i2k/RepData_PeerAssessment2")
StormData <- read.csv("repdata-data-StormData.csv",header=TRUE)
dim(StormData)
## [1] 902297     37
StormData2 <- StormData[,c(8,23:28)]
dim(StormData2)
## [1] 902297      7
summary(StormData2)
##                EVTYPE         FATALITIES          INJURIES        
##  HAIL             :288661   Min.   :  0.0000   Min.   :   0.0000  
##  TSTM WIND        :219940   1st Qu.:  0.0000   1st Qu.:   0.0000  
##  THUNDERSTORM WIND: 82563   Median :  0.0000   Median :   0.0000  
##  TORNADO          : 60652   Mean   :  0.0168   Mean   :   0.1557  
##  FLASH FLOOD      : 54277   3rd Qu.:  0.0000   3rd Qu.:   0.0000  
##  FLOOD            : 25326   Max.   :583.0000   Max.   :1700.0000  
##  (Other)          :170878                                         
##     PROPDMG          PROPDMGEXP        CROPDMG          CROPDMGEXP    
##  Min.   :   0.00          :465934   Min.   :  0.000          :618413  
##  1st Qu.:   0.00   K      :424665   1st Qu.:  0.000   K      :281832  
##  Median :   0.00   M      : 11330   Median :  0.000   M      :  1994  
##  Mean   :  12.06   0      :   216   Mean   :  1.527   k      :    21  
##  3rd Qu.:   0.50   B      :    40   3rd Qu.:  0.000   0      :    19  
##  Max.   :5000.00   5      :    28   Max.   :990.000   B      :     9  
##                    (Other):    84                     (Other):     9

B. Storm Data Dictionary

Storm data by event type (EVTYPE) were assigned to officially sanctioned weather event names in Table 1 of Section 2.1.1 of NWS Directive 10-1605 (http://www.nws.noaa.gov/directives/sym/pd01016005curr.pdf). For example, gsub(“LIGHTNING”, “Lightning”, StormData$EVTYPE“”) replaces two event types “LIGHTNING” and “LIGHTNING.” with a single weather event named “Lightning”.

Convection Event Type (EVTYPE)
Lightning LIGHTNING,LIGHTNING.
Tornado TORNADO,TORNADOES,TSTM WIND,HAIL
Thunderstorm THUNDERSTORM WIND,THUNDERSTORM WINDS,TSTM WIND
High Wind HIGH WIND,HIGH WINDS,HIGH WIND/SEAS
Strong Wind STRONG WIND,STRONG WINDS
Hail HAIL

Extreme Temperatures
Cold/Wind Chill:COLD/WINDCHILL; Cold:COLD,Cold; Heat:EXCESSIVE HEAT,HEAT,HEAT WAVE

Flood Flood:FLOOD; Flash Flood:FLASH FLOOD,FLASH FLOODING,FLASH FLOODING/FLOOD; Coastal Flood:COASTAL FLOOD,COASTAL FLOODING,Coastal Flooding

Marine
Rip current:RIP CURRENT,RIP CURRENTS,RIP CURRENT/HEAVY SURF; Marine Strong Wind:MARINE STRONG WIND; Marine Thunderstorm Wind:MARINE THUNDERSTORM WIND,MARINE TSTM WIND; Marine High Wind:MARINE HIGH WIND; Storm Surge/Tide:STORM SURGE,STORM SURGE/TIDE

Tropical Cyclones
Hurricane (Typhoon):HURRICANE,HURRICANE ERIN; Tropical Storm:TROPICAL STORM,TROPICAL STORM GORDON; Tsunami:TSUNAMI

Winter
Avalanche:AVALANCHE; Blizzard:BLIZZARD; Ice Storm:ICE STORM,ICE; Winter Storm:WINTER STORM,WINTER STORMS; Winter Weather:WINTER WEATHER,WINTER WEATHER/MIX

Other
Heavy Rain:HEAVY RAIN; Heavy Snow:HEAVY SNOW; High Surf:HIGH SURF

D. Health Outcomes by Weather Event

1) Deaths by Weather Event

Convection

Weather event-related fatalities were recorded in the storm database from 1950 to 2011. During these six decades, tornado-related deaths accounted for the most fatalities, totaling 5,633. Excessive heat, flash floods, heat and lightning were responsible for 1,903, 978, 937 and 816 weather-related deaths, respectively.

(http://en.wikipedia.org/wiki/April_25%E2%80%9328,_2011_tornado_outbreak)

“The April 25-28, 2011 tornado outbreak was the largest and one of the deadliest tornado outbreaks ever recorded, affecting the Southern, Midwestern, and Northeastern United States and leaving catastrophic destruction in its wake. While the states that were hardest hit were Alabama and Mississippi, the outbreak also produced destructive tornadoes in Arkansas, Georgia, Tennessee and Virginia, and affected many other areas throughout the Southern and Eastern United States. In total, 355 tornadoes were confirmed by the National Weather Service (NWS).”"

library(data.table)

Fatalities.dt <- data.table(Fatalities)
Fatalities.dt[,list(Deaths=sum(FATALITIES)),by="EVTYPE"]
##                         EVTYPE Deaths
##   1:                   TORNADO   5633
##   2:                 TSTM WIND    504
##   3:                      HAIL     15
##   4:              WINTER STORM    206
##   5: HURRICANE OPAL/HIGH WINDS      2
##  ---                                 
## 164:        MARINE STRONG WIND     14
## 165:             COASTAL FLOOD      3
## 166:          STORM SURGE/TIDE     11
## 167:          MARINE HIGH WIND      1
## 168:                   TSUNAMI     33

2) Injuries by Weather Event

Convection

Tornado, TSTM wind and hail-related injuries numbered 91,346. 6,957 and 1,361, respectively.

Injuries.dt <- data.table(Injuries)
Injuries.dt[,list(Injuries=sum(INJURIES)),by="EVTYPE"]
##                        EVTYPE Injuries
##   1:                  TORNADO    91346
##   2:                TSTM WIND     6957
##   3:                     HAIL     1361
##   4:    ICE STORM/FLASH FLOOD        2
##   5:                DENSE FOG      342
##  ---                                  
## 154:          COLD/WIND CHILL       12
## 155: MARINE THUNDERSTORM WIND       26
## 156:       MARINE STRONG WIND       22
## 157:         MARINE HIGH WIND        1
## 158:                  TSUNAMI      129

E. Economic Consequences by Weather Event

Floods accounted for $145B in property damages, hurricanes/typhoons $69B, tornadoes $43B and storm surges $43B.

Drought caused $14B in crop damages, floods and river flood $11B, ice storms $5B and hail storms $3B.

1) Property Damage (in billions) by Weather Event

PropDamage1 <- subset(PropDamage,PROPDMGEXP=="K",c(EVTYPE,PROPDMG))
PropDamage1$PROPDMG <- PropDamage1$PROPDMG/1000000  # Property Damage (in billions)
PropDamage2 <- subset(PropDamage,PROPDMGEXP=="M",c(EVTYPE,PROPDMG))
PropDamage2$PROPDMG <- PropDamage2$PROPDMG/1000  # Property Damage (in billions)
PropDamage3 <- subset(PropDamage,PROPDMGEXP=="B",c(EVTYPE,PROPDMG))
Property_Damage <- rbind(PropDamage1,PropDamage2,PropDamage3)

2) Crop Damage (in billions) by Weather Event**

CropDamage1 <- subset(CropDamage,CROPDMGEXP=="K",c(EVTYPE,CROPDMG))
CropDamage1$CROPDMG <- CropDamage1$CROPDMG/1000000  # Crop Damage (in billions)
CropDamage2 <- subset(CropDamage,CROPDMGEXP=="M",c(EVTYPE,CROPDMG))
CropDamage2$CROPDMG <- CropDamage2$CROPDMG/1000  # Crop Damage (in billions)
CropDamage3 <- subset(CropDamage,CROPDMGEXP=="B",c(EVTYPE,CROPDMG))
Crop_Damage <- rbind(CropDamage1,CropDamage2,CropDamage3)

Figure 3. Histogram of Property and Crop Damage by Weather Event

The top five total property and crop damage-related weather events were: 1) drought, 2) floods and river floods, 4) hail and 5) ice storms.

x1 <- tapply(Property_Damage$PROPDMG,INDEX=Property_Damage$EVTYPE,FUN=sum,na.rm=TRUE)
y1 <- subset(x1,x1>30)

x2 <- tapply(Crop_Damage$CROPDMG,INDEX=Crop_Damage$EVTYPE,FUN=sum,na.rm=TRUE)
y2 <- subset(x2,x2>3)

df <- data.frame(rbind(y1,y2),rownames=c("Property Damage","Crop Damage"))
## Warning in rbind(y1, y2): number of columns of result is not a multiple of
## vector length (arg 1)
df
##      DROUGHT     FLOOD      HAIL ICE.STORM RIVER.FLOOD        rownames
## y1 144.65771 69.305840 43.323536 56.925660  144.657710 Property Damage
## y2  13.97257  5.661968  3.025537  5.022113    5.029459     Crop Damage
barplot(as.matrix(df),main="Histogram of Property and Crop Damage",ylab="Billions of Dollars",xlab="Weather Event",legend=df$rownames,cex.names=0.75)
## Warning in apply(height, 2L, cumsum): NAs introduced by coercion