Synopsis:

The objective of this study was analised a Storm data documentation in order to find out what type of weather is most dangerous for human health and has the greatest economic consequences.

The report will show how Tornados are the most dangerous type of event, both economic and health wise. Heat are the second most harmful and floods the ones that give the greatest economics damages.

Data Processing:

The data were collected from the course web site, and the documentation can be find at NATIONAL WEATHER SERVICE.

The code for loading the data is below:

setwd("C:/Users/Master/Desktop/PA_TESTE/")
data <- read.csv("repdata-data-StormData.csv")

Understanding the Data

Direct Fatalities/Injuries

A direct fatality or injury is defined as a fatality or injury directly attributable to the hydro-meteorological event itself, or impact by airborne/falling/moving debris, i.e.,missiles generated by wind, water, ice, lightning, tornado, etc. In these cases, the weather event was an “active” agent or generated debris which became an active agent. Generalized examples of direct fatalities/injuries would include:

-a. Thunderstorm wind gust causes a moving vehicle to roll over; -b. Blizzard winds topple a tree onto a person; and -c. Vehicle is parked on a road, adjacent to a dry arroyo. A flash flood comes down the arroyo and flips over the car. The driver drowns.

-Building the dataset that will be used:

Harmful<- data[,c(8,23,24,37)]
head(Harmful)
##    EVTYPE FATALITIES INJURIES REFNUM
## 1 TORNADO          0       15      1
## 2 TORNADO          0        0      2
## 3 TORNADO          0        2      3
## 4 TORNADO          0        2      4
## 5 TORNADO          0        2      5
## 6 TORNADO          0        6      6
T_Missing <- sum(is.na(Harmful))
T_Missing
## [1] 0

-Understanding the dataset

Total_Injuries<- tapply(Harmful$INJURIES,Harmful$EVTYPE, sum, na.rm=TRUE)
Total_Fatalities<- tapply(Harmful$FATALITIES,Harmful$EVTYPE, sum, na.rm=TRUE)

Harmful$HARMFUL<- Harmful$INJURIES + Harmful$FATALITIES
Total_Harmful<- tapply(Harmful$HARMFUL,Harmful$EVTYPE, sum, na.rm=TRUE)


Total_Injuries<-Total_Injuries[order(-Total_Injuries)][c(1:5)]
Total_Fatalities<-Total_Fatalities[order(-Total_Fatalities)][c(1:5)]
Total_Harmful<-Total_Harmful[order(-Total_Harmful)][c(1:5)]

Economical Damage:

Property damage estimates should be entered as actual dollar amounts, if a reasonably accurate estimate from an insurance company or other qualified individual is available.

-Building the dataset that will be used:

Damage<- data[,c(8,25,27)]
head(Damage)
##    EVTYPE PROPDMG CROPDMG
## 1 TORNADO    25.0       0
## 2 TORNADO     2.5       0
## 3 TORNADO    25.0       0
## 4 TORNADO     2.5       0
## 5 TORNADO     2.5       0
## 6 TORNADO     2.5       0
T_Missing <- sum(is.na(Damage))
T_Missing
## [1] 0

-Understanding the dataset

Total_Prop<- tapply(Damage$PROPDMG,Damage$EVTYPE, sum, na.rm=TRUE)
Total_Crop<- tapply(Damage$CROPDMG,Damage$EVTYPE, sum, na.rm=TRUE)

Damage$DAMAGE<- Damage$PROPDMG + Damage$CROPDMG
Total_Damage<- tapply(Damage$DAMAGE,Damage$EVTYPE, sum, na.rm=TRUE)


Total_Prop<-Total_Prop[order(-Total_Prop)][c(1:5)]
Total_Crop<-Total_Crop[order(-Total_Crop)][c(1:5)]
Total_Damage<-Total_Damage[order(-Total_Damage)][c(1:5)]

Results

Tornados are the most harmful event (both injuries and fatalities), followed by Excessive Heat. In general we can assume 3 types of events that are most harmful across USA: Tornados and Winds, Heat, Flood and lightning.

barplot(Total_Harmful,
        horiz=TRUE,
        axisname=TRUE,
        cex.names=0.5,
        col=c("plum4",
              "palevioletred4",
              "lightpink3",
              "mediumseagreen",
              "limegreen"),
        main="Total of Injuries + Fatalities by Event Type",
        xlab="Total of Injuries + Fatalities")

Total_Harmful
##        TORNADO EXCESSIVE HEAT      TSTM WIND          FLOOD      LIGHTNING 
##          96979           8428           7461           7259           6046
Total_Injuries
##        TORNADO      TSTM WIND          FLOOD EXCESSIVE HEAT      LIGHTNING 
##          91346           6957           6789           6525           5230
Total_Fatalities
##        TORNADO EXCESSIVE HEAT    FLASH FLOOD           HEAT      LIGHTNING 
##           5633           1903            978            937            816

Tornados are the event that has the greatest economic consequences (both properties and crop), followed by Flash Flood. In general we can assume 3 types of events that cause the greatest economic damages across USA: Tornados and Winds, Flood and Hail.

barplot(Total_Damage,
        horiz=TRUE,
        axisname=TRUE,
        cex.names=0.5,
        col=c("plum4",
              "palevioletred4",
              "lightpink3",
              "mediumseagreen",
              "limegreen"),
        main="Total Damages (Prop + Crop) by Event Type",
        xlab="Total Damages (Prop + Crop)")

Total_Prop
##           TORNADO       FLASH FLOOD         TSTM WIND             FLOOD 
##         3212258.2         1420124.6         1335965.6          899938.5 
## THUNDERSTORM WIND 
##          876844.2
Total_Crop
##        HAIL FLASH FLOOD       FLOOD   TSTM WIND     TORNADO 
##    579596.3    179200.5    168037.9    109202.6    100018.5
Total_Damage
##     TORNADO FLASH FLOOD   TSTM WIND        HAIL       FLOOD 
##     3312277     1599325     1445168     1268290     1067976