Title: “NOAA Storm Data Assignment”

output: html_document: default

SYNOPSIS

This analysis examine the U.S National Oceanic and Atmospheric Administartion storm database to determine which severe waether events have the high effect on public health and the economy. the impact of public health is measure using injuries and fatalities.property and crop damage is used to measure economic impact.The analyis begins with the raw compressed storm data file and process the data indside R for reproucible result.total public health, economic impact are calculate d for each event type and this event types are group using EVTYPE variables. The results are presented using the summary tables and plot. This result is inteneded to identify the must harmful waether event accross the united states based on the avialable NOAA storm data.

Data Processing

The data for this analysis were obtained from the NOAA Storm Database. The original data file is a compressed comma-separated file with a .csv.bz2 extension. The analysis starts from the raw compressed file, loads the data into R, and selects the variables needed to answer the public health and economic impact questions.

options(scipen = 999)

file_url <- "https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2"
file_name <- "StormData.csv.bz2"

if (!file.exists(file_name)) {
        download.file(file_url, file_name, method = "auto")
}

storm_data <- read.csv(bzfile(file_name), stringsAsFactors = FALSE)

dim(storm_data)

## [1] 902297     37

names(storm_data)[1:20]

##  [1] "STATE__"    "BGN_DATE"   "BGN_TIME"   "TIME_ZONE"  "COUNTY"    
##  [6] "COUNTYNAME" "STATE"      "EVTYPE"     "BGN_RANGE"  "BGN_AZI"   
## [11] "BGN_LOCATI" "END_DATE"   "END_TIME"   "COUNTY_END" "COUNTYENDN"
## [16] "END_RANGE"  "END_AZI"    "END_LOCATI" "LENGTH"     "WIDTH"

The variables needed for the analysis include event type, fatalities, injuries, property damage, crop damage, and the damage exponent variables. For the public health question, fatalities and injuries were combined into one total health-impact measure.

storm_selected <- storm_data[, c("EVTYPE", 
                                 "FATALITIES", 
                                 "INJURIES", 
                                 "PROPDMG", 
                                 "PROPDMGEXP", 
                                 "CROPDMG", 
                                 "CROPDMGEXP")]

storm_selected$EVTYPE <- toupper(storm_selected$EVTYPE)

storm_selected$health_damage <- storm_selected$FATALITIES + storm_selected$INJURIES

health_summary <- aggregate(health_damage ~ EVTYPE, 
                            data = storm_selected, 
                            sum)

health_summary <- health_summary[order(-health_summary$health_damage), ]

top_health <- head(health_summary, 10)

top_health

##                EVTYPE health_damage
## 758           TORNADO         96979
## 116    EXCESSIVE HEAT          8428
## 779         TSTM WIND          7461
## 154             FLOOD          7259
## 418         LIGHTNING          6046
## 243              HEAT          3037
## 138       FLASH FLOOD          2755
## 387         ICE STORM          2064
## 685 THUNDERSTORM WIND          1621
## 888      WINTER STORM          1527

For the economic impact question, property damage and crop damage were converted into dollar values using their corresponding exponent variables. The exponent variables indicate whether the reported damage should be multiplied by hundreds, thousands, millions, or billions.

convert_exponent <- function(exp) {
        exp <- toupper(as.character(exp))
        
        multiplier <- rep(0, length(exp))
        
        multiplier[exp == ""] <- 1
        multiplier[exp == "0"] <- 1
        multiplier[exp == "1"] <- 10
        multiplier[exp == "2"] <- 100
        multiplier[exp == "3"] <- 1000
        multiplier[exp == "4"] <- 10000
        multiplier[exp == "5"] <- 100000
        multiplier[exp == "6"] <- 1000000
        multiplier[exp == "7"] <- 10000000
        multiplier[exp == "8"] <- 100000000
        multiplier[exp == "9"] <- 1000000000
        multiplier[exp == "H"] <- 100
        multiplier[exp == "K"] <- 1000
        multiplier[exp == "M"] <- 1000000
        multiplier[exp == "B"] <- 1000000000
        
        multiplier
}

storm_selected$property_damage_value <- storm_selected$PROPDMG * 
        convert_exponent(storm_selected$PROPDMGEXP)

storm_selected$crop_damage_value <- storm_selected$CROPDMG * 
        convert_exponent(storm_selected$CROPDMGEXP)

storm_selected$economic_damage <- storm_selected$property_damage_value + 
        storm_selected$crop_damage_value

economic_summary <- aggregate(economic_damage ~ EVTYPE, 
                              data = storm_selected, 
                              sum)

economic_summary <- economic_summary[order(-economic_summary$economic_damage), ]

top_economic <- head(economic_summary, 10)

top_economic

##                EVTYPE economic_damage
## 154             FLOOD    150319678257
## 372 HURRICANE/TYPHOON     71913712800
## 758           TORNADO     57362333886
## 599       STORM SURGE     43323541000
## 212              HAIL     18761221986
## 138       FLASH FLOOD     18243991078
## 84            DROUGHT     15018672000
## 363         HURRICANE     14610229010
## 529       RIVER FLOOD     10148404500
## 387         ICE STORM      8967041360

Results

Events Most Harmful to Population Health

The first research question asks which types of weather events are most harmful to population health. Population health impact was measured by combining total fatalities and injuries for each event type. The event types with the highest combined fatalities and injuries are shown below.

barplot(top_health$health_damage,
        names.arg = top_health$EVTYPE,
        las = 2,
        main = "Top 10 Weather Events by Population Health Impact",
        ylab = "Total Fatalities and Injuries",
        cex.names = 0.7)

Based on the results, tornadoes have the greatest overall impact on population health. Tornadoes produced the highest combined number of fatalities and injuries in the NOAA storm database.

Events With the Greatest Economic Consequences

The second research question asks which types of weather events have the greatest economic consequences. Economic impact was measured by combining property damage and crop damage after converting the damage exponent variables into dollar-value multipliers.

barplot(top_economic$economic_damage / 1000000000,
        names.arg = top_economic$EVTYPE,
        las = 2,
        main = "Top 10 Weather Events by Economic Damage",
        ylab = "Total Economic Damage in Billions of Dollars",
        cex.names = 0.7)

Based on the results, flood-related events have some of the greatest economic consequences. Floods caused very large property and crop damage across the United States.

Conclusion

This analysis shows that the most harmful weather events depend on whether the concern is population health or economic loss. Tornadoes are the most harmful event type for population health when fatalities and injuries are combined. Flood-related events are among the most economically damaging event types when property and crop damage are combined. These findings suggest that emergency preparedness planning should consider both human health effects and financial losses when prioritizing severe weather risks.