Reproducible Research Peer Assessment 2

Lyn Yang

May 21, 2023

Title:Analysis on weather’s effects on health and economic consequences

Synopsis: This data analysis synthesizes weather’s effects on health and economic consequences and uses bar charts to display the effects of weather. Based on the current analysis, we can tell that tornado is the most harmful weather to population health. Tornado and hail are the weather that leads to the most severe economic consequences.

Data Processing:

Loading the data:

        url <- "https://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2"
        download.file(url, destfile = "Data.csv.bz2", method = "curl")
        NOAA <- read.csv(bzfile("Data.csv.bz2"), header = TRUE, 
                         stringsAsFactors = FALSE)

Loading some R packages

        library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

Results

Across the United States, which types of events are most harmful with respect to population health?

Based on the dataset description, we can tell that there are two variables relate to being harmful to the population health. They are the fatalities number and the injuries number. So we’ll take a look at them.

Total Fatalities

        Total_Fatalities <- aggregate(NOAA$FATALITIES, by = list(NOAA$EVTYPE),
                                      "sum", na.rm = TRUE)
        names(Total_Fatalities) <- c("Event", "Total Fatalities")
        Total_Fatalities_sorted <- Total_Fatalities[order(-Total_Fatalities$`Total Fatalities`), ][1:10, ]

Total Injuries

        Total_Injuries <- aggregate(NOAA$INJURIES, by = list(NOAA$EVTYPE),
                                      "sum", na.rm = TRUE)
        names(Total_Injuries) <- c("Event", "Total Injuries")
        Total_Injuries_sorted <- Total_Injuries[order(-Total_Injuries$`Total Injuries`), ][1:10, ]

Visualization

        par(mfrow = c(2, 1), mar = c(10, 4, 2, 2), las = 3, cex = 0.7, 
            cex.main = 1.4, cex.lab = 1.2)
        barplot(Total_Fatalities_sorted$`Total Fatalities`, 
                names.arg = Total_Fatalities_sorted$Event, col = 'black', 
                main = 'Top 10 Weather Events for Fatalities', ylab = 'Fatalities Number')
        barplot(Total_Injuries_sorted$`Total Injuries`, 
                names.arg = Total_Injuries_sorted$Event, col = "grey", 
                main = 'Top 10 Weather Events for Injuries', ylab = 'Injuries Number')

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

We can tell from the description that there’re two varaibles show the economic consequences, which are the property damage and the crop damage. So we’ll take a look at them.

Total Property Damage

        Total_PDMG <- aggregate(NOAA$PROPDMG, by = list(NOAA$EVTYPE),
                                      "sum", na.rm = TRUE)
        names(Total_PDMG) <- c("Event", "Total Property Damage")
        Total_PDMG_sorted <- Total_PDMG[order(-Total_PDMG$`Total Property Damage`), ][1:10, ]

Total Crop Damage

        Total_CDMG <- aggregate(NOAA$CROPDMG, by = list(NOAA$EVTYPE),
                                      "sum", na.rm = TRUE)
        names(Total_CDMG) <- c("Event", "Total Crop Damage")
        Total_CDMG_sorted <- Total_CDMG[order(-Total_CDMG$`Total Crop Damage`), ][1:10, ]

Visualization

        par(mfrow = c(2, 1), mar = c(10, 4, 2, 2), las = 3, cex = 0.7, 
            cex.main = 1.4, cex.lab = 1.2)
        barplot(Total_PDMG_sorted$`Total Property Damage`, 
                names.arg = Total_PDMG_sorted$Event, col = 'blue', 
                main = 'Top 10 Weather Events for Property Damge', 
                ylab = 'Property Damage')
        barplot(Total_CDMG_sorted$`Total Crop Damage`, 
                names.arg = Total_CDMG_sorted$Event, col = "yellow", 
                main = 'Top 10 Weather Events for Crop Damage', 
                ylab = 'Crop Damage')