Synopsis

In this report we present an analysis of the extend of natural disaster on the economy and public health. The analysis looks into the frequncy injuries and fitalities associated different weather events types. It turns out the Tornado events leads to hightest frequency of fitalities and enjouries making it the most harmful natural extreme event. The report also presents estimates of ecomonic damage that can be associated with various types of naturel desasters. The data leads to an insight that Tonados,Hail, flesh flood and fold claim a significant amount of ecomony with Tonados being the most distructive on both property and crops.

Processing

In order to carry out the extreme weather impacts analysis on both population and econmony, we select the relevant data columns. The Storm data is also available online from the NOAA’s National Weather Service. The data uploading also takes care that we are dealing with a large file.

Adata <- read.csv('file:///D:/Cousera/reproducible-research/repdata_data_StormData.csv')
Adata<-Adata[,c("EVTYPE","COUNTYNAME","STATE","INJURIES","PROPDMG","PROPDMGEXP","CROPDMG","CROPDMGEXP","FATALITIES")]

Data Transformation

The property damage estimates expoments are expressed through their respective abbreviations. In orger to aggregate the damage, we transform the property damage magnitute column (PROPDMG) and the (PROPDMGEXP) and introduce a new column on the data frame called PROPDMGEXPN and Finaly we multiply the (PROPDMG) and PROPDMGEXPN to obtain the property damage value that we can be aggregated.

#propterty damage data
AdataB<- Adata[1,c("EVTYPE","COUNTYNAME" ,"STATE","INJURIES","PROPDMG","PROPDMGEXP","FATALITIES")]
AdataB$PROPDMGN<-""
AdataB$CROPDMGN<-""
AdataB<- AdataB[-1,]
EXP_index<-c("M","m","K","0","","1","2","h","H","3","4","5","6","7","8")
EXP<-c(10^6,10^6,10^3,1,1,10^1,10^2,10^2,10^2,10^3,10^4,10^5,10^6,10^7,10^8)
for(i in 1:15){
  AdataTMP <-Adata[Adata$PROPDMGEXP==EXP_index[i],]
  AdataTMP$PROPDMGN<-EXP[i]*AdataTMP[,"PROPDMG"]
  AdataTMP$CROPDMGN<-EXp[i]*(AdataTMP[,"CROPDMG"])
  AdataB<- rbind(AdataB,AdataTMP)
  
}

Results

  1. Ranking the most harmful weather events

The frequency of reported publica health cases associated with extreme weather invenents is indicative of how harmful such cases can be the public. To the effect In this section we study at the reported injuries and fatalities per event type. The goal is gain a perspective on which of the extreme weather events lead to more harm on human public health.

For planning and development of resilience strategies, it is important to have reliable estimates of the damage associated with each of the extreme events. This is chieved by plotting the most costly events.

library("dplyr")
AdataB_new <- AdataB%>% select(STATE, EVTYPE, INJURIES, FATALITIES) %>%group_by(EVTYPE) %>%
summarise(Deaths = sum(FATALITIES), Injury = sum(INJURIES), Harmful = Deaths+Injury) %>%
    arrange(desc(Harmful));



head(AdataB_new) 

AdataB_new<-melt(AdataB_new,id.var="EVTYPE",measure.var=c("Deaths" ,"Injury","Harmful"))
AdataB_new<-AdataB_new %>%arrange(desc(value))
q <- ggplot(AdataB_new[AdataB_new$value>=537,], aes(EVTYPE, value)) + 
     geom_bar(stat = "identity")+
     facet_wrap(variable~.)+ theme_bw()+ coord_flip()+
     labs(title = " Types of events leading to injury or Mortality", x = "Event Type", 
         y = "Weather Events Number");
q

The events with frequencies higher than 537 are plotted in this graph. Tonados is the most dominating event.

  1. Estimates of loss due to the harmful events

In this section we potray events according to their associated ecomomic loss. Such an insight is mostly importat for development of strategy and resiliance measures for the population.


DMG_df <- AdataB %>% select(EVTYPE, PROPDMGN, CROPDMGN) %>%
  group_by(EVTYPE) %>% summarise(VPDMG = sum(PROPDMGN), VCDMG = sum(CROPDMGN), Total_Exp = VPDMG+VCDMG)

head(DMG_df)
  
DMG_df<-melt(DMG_df,id.var="EVTYPE",measure.var=c("VPDMG" ,"VCDMG","Total_Exp"))
DMG_df<-DMG_df %>%arrange(desc(value))

q <- ggplot(DMG_df[DMG_df$value>=1882229972,], aes(EVTYPE, value)) + 
     geom_bar(stat = "identity")+
     facet_wrap(variable~.)+ theme_bw()+ coord_flip()+
     labs(title = " Types of events leading to property or Crop Damage", 
         x = " Type of Event",y = "Estimated Event Damage" );
q

In this plot we view events that cost above 1.8 billon. It turns out Tonados lead to the most ecomonic loss associated with the crop and property damage. Hail, flod and Flash flood. Are also having a dominat contribution on the estimated loss.

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