Synopsis

Storms and other severe weather events can cause both public health and economic problems for communities and municipalities. Many severe events can result in fatalities, injuries, and property damage, and preventing such outcomes to the extent possible is a key concern.

In this document the Storm Data is analysed: an official publication of the National Oceanic and Atmospheric Administration (NOAA). The main objective is to show those events that cause the major human and economical damage in the USA between the year 1950 and end in November 2011.

Data processing

The following code is for loading the data, available in the repdata_data_StormData.csv.bz2 file. The packages that are used will also be loaded with the library function.

## Warning: package 'plyr' was built under R version 3.5.3

1.-Across the United States, which types of events (as indicated in the EVTYPE variable) are most harmful with respect to population health?

First of all I will split the dataset regarding the harmful events. This includes the amount of people died (fatalities) and the amount of people injured (injuries).

The following are the most harmful events that caused mortalities:

head(fatalities[,c(1,2)])
##             EVTYPE fatalities
## 834        TORNADO       5633
## 130 EXCESSIVE HEAT       1903
## 153    FLASH FLOOD        978
## 275           HEAT        937
## 464      LIGHTNING        816
## 856      TSTM WIND        504

The following are the most harmful events that caused injuries:

head(injuries[,c(1,3)])
##             EVTYPE injuries
## 834        TORNADO    91346
## 856      TSTM WIND     6957
## 170          FLOOD     6789
## 130 EXCESSIVE HEAT     6525
## 464      LIGHTNING     5230
## 275           HEAT     2100

2.- Across the United States, which types of events have the greatest economic consequences?

I will focus in “PROPDMG”, “PROPDMGEXP”, “CROPDMG”, “CROPDMGEXP” variables, which are the variables of property and crop damage. “PROPDMGEXP” and “CROPDMGEXP” contain the exponential value for every one, being “H” for hundreds, “K” for thousands, “M” for millions, and “B” for billions (as it is documented).

The following code generates a dataframe that contains the total cost for every observation in a new variable.

Unifying it by event type:

The following are the events that caused more property economical costs:

head(propCost[c(1,2)])
##                EVTYPE     propCost
## 170             FLOOD 1.225222e+14
## 411 HURRICANE/TYPHOON 6.550381e+13
## 670       STORM SURGE 4.256076e+13
## 402         HURRICANE 5.706168e+12
## 834           TORNADO 5.351637e+12
## 848    TROPICAL STORM 5.152554e+12

The following are the events that caused more crop economical costs:

head(cropCost[c(1,3)])
##                EVTYPE     cropCost
## 590       RIVER FLOOD 5.000029e+12
## 427         ICE STORM 5.000022e+12
## 95            DROUGHT 1.512473e+12
## 411 HURRICANE/TYPHOON 1.511098e+12
## 275              HEAT 4.000015e+11
## 192            FREEZE 2.002462e+11

RESULTS

The following graph shows which are the most harmful events for people. In the left graph, those events that most fatalities have caused are represented. In the right graph, the events represent the injuries caused.

fat <- head(fatalities[,c(1,2)])
inj <- head(injuries[,c(1,3)])

par(mfrow=c(1,2))
barplot(fat$fatalities, title="Total fatalities by event type", ylab= "Total", beside=TRUE, col=rainbow(5))
## Warning in plot.window(xlim, ylim, log = log, ...): "title" is not a
## graphical parameter
## Warning in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...):
## "title" is not a graphical parameter
## Warning in axis(if (horiz) 1 else 2, cex.axis = cex.axis, ...): "title" is
## not a graphical parameter
legend("topright", as.character(fat$EVTYPE), cex=0.6, bty="n", fill=rainbow(5))

barplot(inj$injuries, main="Total injuries by event type", ylab= "Total", beside=TRUE, col=rainbow(5))
legend("topright", as.character(inj$EVTYPE), cex=0.6, bty="n", fill=rainbow(5))

The next graph shows the economical damages caused by the events. In the left side the events that caused the higher property damage cost are represented (in dollars), and in the right side those events that caused the higher crop damage cost.

prop <- head(propCost[c(1,2)])
crop <- head(cropCost[c(1,3)])


par(mfrow=c(1,2))
barplot(prop$propCost, title="Total property damage cost (in $) by event", ylab= "Total", beside=TRUE, col=rainbow(5))
## Warning in plot.window(xlim, ylim, log = log, ...): "title" is not a
## graphical parameter
## Warning in title(main = main, sub = sub, xlab = xlab, ylab = ylab, ...):
## "title" is not a graphical parameter
## Warning in axis(if (horiz) 1 else 2, cex.axis = cex.axis, ...): "title" is
## not a graphical parameter
legend("topright", as.character(fat$EVTYPE), cex=0.6, bty="n", fill=rainbow(5))

barplot(crop$cropCost, main="Total crop damage cost (in $) by event", ylab= "Total", beside=TRUE, col=rainbow(5))
legend("topright", as.character(inj$EVTYPE), cex=0.6, bty="n", fill=rainbow(5))