Synopsis

In this report, our goal is to observe the negative effects of weather phenomena in the United States. To do this we will observe the impact on health and the economy. We will analyze each phenomenon and in each area we will choose the 10 that have the greatest individual impact when it occurs. We will build on he U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database. This database tracks characteristics of major storms and weather events in the United States, including when and where they occur, as well as estimates of any fatalities, injuries, and property damage.

Data Processing

From the National Weather Service we obtained the U.S. National Oceanic and Atmospheric Administration’s (NOAA) storm database. The events in the database start in the year 1950 and end in November 2011.

We obtained the file Storm Data.

Packages

The packages to be used for the project are:

library(dplyr)
library(ggplot2)

Read Data

SData = read.csv('repdata_data_StormData.csv', header = T)

Processing the data for the section “The most damaging meteorological event with respect to the health of the population”

The data is filtered according to the catastrophic events column (EVTYPE) and the two columns related to population health (INJURIES and FATALITIES) are added together. The mean per event is obtained to find which event is more catastrophic each time it occurs.

PSData1 <- SData[,c(8,23,24)]
PSData1$harm <- PSData1$FATALITIES+PSData1$INJURIES
PSData1F <- aggregate(harm~EVTYPE, data = PSData1, FUN=mean, na.rm=TRUE)
PSData1F <- head(PSData1F[order(PSData1F$harm,decreasing = TRUE),], n = 10)
head(PSData1F)
##                         EVTYPE  harm
## 277                  Heat Wave 70.00
## 851      TROPICAL STORM GORDON 51.00
## 954                 WILD FIRES 38.25
## 821              THUNDERSTORMW 27.00
## 842 TORNADOES, TSTM WIND, HAIL 25.00
## 366         HIGH WIND AND SEAS 23.00

Processing the data for the section “The most damaging meteorological event with respect to the economic”

The data is filtered according to the catastrophic events column (EVTYPE) and the two columns related to the economy (PROPDMG and CROPDMG).

  • PROPDMG: Property Damage
  • CROPDMG: Crop damage

The two columns referring to the economy are separated, to obtain separate results. Although in parallel the mean per event is obtained to find which event is more catastrophic each time it occurs.

  • Property Damage
PDSData1 <- SData[,c(8,25)]
PDSData1 <- aggregate(PROPDMG~EVTYPE, data = PDSData1, FUN=mean, na.rm=TRUE)
PDSData1 <- head(PDSData1[order(PDSData1$PROPDMG,decreasing = TRUE),], n = 10)
head(PDSData1)
##                     EVTYPE PROPDMG
## 52         COASTAL EROSION     766
## 291   HEAVY RAIN AND FLOOD     600
## 589 RIVER AND STREAM FLOOD     600
## 445              Landslump     570
## 38   BLIZZARD/WINTER STORM     500
## 158           FLASH FLOOD/     500
  • Crop damage
CDSData1 <- SData[,c(8,27)]
CDSData1 <- aggregate(CROPDMG~EVTYPE, data = CDSData1, FUN=mean, na.rm=TRUE)
CDSData1 <- head(CDSData1[order(CDSData1$CROPDMG,decreasing = TRUE),], n = 10)
head(CDSData1)
##                    EVTYPE CROPDMG
## 118 DUST STORM/HIGH WINDS 500.000
## 190          FOREST FIRES 500.000
## 851 TROPICAL STORM GORDON 500.000
## 392       HIGH WINDS/COLD 401.000
## 407       HURRICANE FELIX 250.000
## 591        River Flooding 241.368

Results

The most damaging meteorological event with respect to the health of the population

Analyzing the damage generated by the occurrence of each meteorological event. From the data processing:

  • The event that has the greatest impact each time it occurs is the heat wave.

  • Second tropical storm gordon and third wild fires.

library(ggplot2)
ggplot(PSData1F,aes(reorder(EVTYPE,harm), harm)) +
      geom_bar(fill="#9F7EC4", color="#FAA3F4", stat = "identity") +
      scale_fill_brewer(palette = "Set1") +
      theme(axis.text.x = element_text(angle = 50, hjust = 1)) +
      xlab("Meteorological event") +
      ggtitle("Most catastrophic weather event")

The most damaging meteorological event with respect to the economic

Results are divided by Property Damage and Crop damage.

Property Damage

Analyzing the damage generated by the occurrence of each meteorological event. From the data processing:

  • The event that has the greatest impact each time it occurs is the coastal erosion.

  • Second heavy rain and flood and also river and stream flood.

ggplot(PDSData1,aes(reorder(EVTYPE,PROPDMG), PROPDMG)) +
      geom_bar(fill="#9F7EC4", color="#FAA3F4", stat = "identity") +
      scale_fill_brewer(palette = "Set1") +
      theme(axis.text.x = element_text(angle = 50, hjust = 1)) +
      xlab("Meteorological event") +
      ylab("Property Damage")+
      ggtitle("Most catastrophic weather event for the economy - property damage")

Crop damage

Analyzing the damage generated by the occurrence of each meteorological event. From the data processing:

  • The events that have the greatest impact each time they occur are: dust storm/high winds, forest fires and tropical storm Gordon.
ggplot(CDSData1,aes(reorder(EVTYPE,CROPDMG), CROPDMG)) +
      geom_bar(fill="#9F7EC4", color="#FAA3F4", stat = "identity") +
      scale_fill_brewer(palette = "Set1") +
      theme(axis.text.x = element_text(angle = 50, hjust = 1)) +
      xlab("Meteorological event") +
      ylab("Crop Damage")+
      ggtitle("Most catastrophic weather event for the economy - crop damage")