Most Harmful and impacting events in severe weather conditions
Sergio Fernandez Bertolin
Tuesday, February 17, 2015
Synopsis
The following project is based on the NOAA public data. On it, we are going to analyze the most harmful events related with meteorological events. The data supplied is registered from 1950 until 2011. First, we made a pre-processing and loading of the data. Then, we group some results and present it in a suitable mode.
Basically, we present the most harmful events based on two different effects: the personal damage caused and the economic repercussion.
Data processing
Reading the data
First we extract the data from the NOAA
if(!file.exists("./storm_data.csv")){
download.file("http://d396qusza40orc.cloudfront.net/repdata%2Fdata%2FStormData.csv.bz2","./storm_data.csv",mode="wb")
}
data <- read.csv("./storm_data.csv",header=TRUE)Pre-processing the data
Initially, we are formatting the data, converting rows with dates to a proper format. Then, we extract and create a new column with the year of each register
Dates <- as.character(data$BGN_DATE)
Dates_2 <- as.Date(Dates,"%m/%d/%Y")
year <- format(Dates_2,"%Y")
data_bind <- cbind(data,year)
data_bind$year <- as.integer(as.character(data_bind$year))Simplifying the dataset
As we can see in the code below, the vast majority of the registers are only within the period between years 1996 and 2012, from first quartile to the end. Added to this, we have to consider that years 1994 and 1995 have a lot of bad or incongruent registers, so the idea of left this data out of the scope of the study is reinforced.
summary(data_bind$year)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1950 1995 2002 1999 2007 2011Next step is creating the new subset only from 1996.
data_post_1995 <- data_bind[data_bind[,38]>1995,] As we are only interested in personal damage, now we are groupping and adding the results by event type
fatalities_by_events <- aggregate(data_post_1995$FATALITIES,list(data_post_1995$EVTYPE),sum)
injuries_by_events <- aggregate(data_post_1995$INJURIES,list(data_post_1995$EVTYPE),sum)
injuries_fatalities <- cbind(fatalities_by_events,injuries_by_events$x)Selecting only the events with at least one fatality
injuries_fatalities_2 <- injuries_fatalities[(injuries_fatalities[,2])>0|(injuries_fatalities[,3])>0,]Examining only the Event types
print(injuries_fatalities_2$Group.1) ## [1] AVALANCHE BLACK ICE
## [3] BLIZZARD blowing snow
## [5] BRUSH FIRE COASTAL FLOOD
## [7] Coastal Flooding COASTAL FLOODING
## [9] COASTAL FLOODING/EROSION Coastal Storm
## [11] COASTAL STORM COASTALSTORM
## [13] Cold COLD
## [15] COLD AND SNOW Cold Temperature
## [17] COLD WEATHER COLD/WIND CHILL
## [19] DENSE FOG DROUGHT
## [21] DROWNING DRY MICROBURST
## [23] Dust Devil DUST DEVIL
## [25] DUST STORM EXCESSIVE HEAT
## [27] EXCESSIVE SNOW Extended Cold
## [29] Extreme Cold EXTREME COLD
## [31] EXTREME COLD/WIND CHILL EXTREME WINDCHILL
## [33] FALLING SNOW/ICE FLASH FLOOD
## [35] FLOOD FOG
## [37] FREEZING DRIZZLE FREEZING RAIN
## [39] Freezing Spray FROST
## [41] FUNNEL CLOUD GLAZE
## [43] GUSTY WIND Gusty winds
## [45] Gusty Winds GUSTY WINDS
## [47] HAIL HAZARDOUS SURF
## [49] HEAT Heat Wave
## [51] HEAVY RAIN HEAVY SEAS
## [53] HEAVY SNOW Heavy snow shower
## [55] Heavy Surf HEAVY SURF
## [57] Heavy surf and wind HEAVY SURF/HIGH SURF
## [59] HIGH SEAS High Surf
## [61] HIGH SURF HIGH SWELLS
## [63] HIGH WATER HIGH WIND
## [65] HURRICANE Hurricane Edouard
## [67] HURRICANE/TYPHOON HYPERTHERMIA/EXPOSURE
## [69] Hypothermia/Exposure HYPOTHERMIA/EXPOSURE
## [71] ICE ON ROAD ICE ROADS
## [73] ICE STORM ICY ROADS
## [75] LANDSLIDE LANDSLIDES
## [77] LIGHT SNOW LIGHTNING
## [79] Marine Accident MARINE HIGH WIND
## [81] MARINE STRONG WIND MARINE THUNDERSTORM WIND
## [83] MARINE TSTM WIND MIXED PRECIP
## [85] Mudslide Mudslides
## [87] NON-SEVERE WIND DAMAGE NON TSTM WIND
## [89] OTHER RAIN/SNOW
## [91] RECORD HEAT RIP CURRENT
## [93] RIP CURRENTS RIVER FLOOD
## [95] River Flooding RIVER FLOODING
## [97] ROGUE WAVE ROUGH SEAS
## [99] ROUGH SURF SMALL HAIL
## [101] Snow SNOW
## [103] SNOW AND ICE SNOW SQUALL
## [105] Snow Squalls STORM SURGE
## [107] STORM SURGE/TIDE STRONG WIND
## [109] Strong Winds STRONG WINDS
## [111] THUNDERSTORM THUNDERSTORM WIND
## [113] THUNDERSTORM WIND (G40) TIDAL FLOODING
## [115] TORNADO Torrential Rainfall
## [117] TROPICAL STORM TSTM WIND
## [119] TSTM WIND (G35) TSTM WIND (G40)
## [121] TSTM WIND (G45) TSTM WIND/HAIL
## [123] TSUNAMI TYPHOON
## [125] UNSEASONABLY WARM URBAN/SML STREAM FLD
## [127] WARM WEATHER WATERSPOUT
## [129] Whirlwind WILD/FOREST FIRE
## [131] WILDFIRE WIND
## [133] WINDS WINTER STORM
## [135] WINTER WEATHER WINTER WEATHER MIX
## [137] WINTER WEATHER/MIX WINTRY MIX
## 985 Levels: HIGH SURF ADVISORY COASTAL FLOOD ... WNDThe events shown have a lot of causes which can be considered as very similar. Next, we create new groups with closer events to reduce the number of inputs. Then, we obtain results across the new groups created.
injuries_fatalities_2$Group.1 <- gsub("(.*)COLD(.*)","COLD",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)Cold(.*)","COLD",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)SURF(.*)","SURF",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)Surf(.*)","SURF",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)Tornado(.*)","TORNADO",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)TORNADO(.*)","TORNADO",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)THUNDER(.*)TORM(.*)","THUNDERSTORM",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)Thunder(.*)","THUNDERSTORM",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)TSTM(.*)","THUNDERSTORM",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)FLOOD(.*)","FLOOD",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)Flood(.*)","FLOOD",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)HURRICANE(.*)","HURRICANE",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)Hurricane(.*)","HURRICANE",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)LIGHTNING(.*)","LIGHTNING",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)FIRE(.*)","FIRE",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)[Ww]ind(.*)","WIND",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)WIND(.*)","WIND",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)Heat(.*)","HEAT",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)HEAT(.*)","HEAT",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)Dust(.*)","DUST",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)DUST(.*)","DUST",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)MIX(.*)","WINTER WEATHER/MIX",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)RIP CURRENT(.*)","RIP CURRENT",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)ICE(.*)","ICE",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)ICY(.*)","ICE",injuries_fatalities_2$Group.1)
injuries_fatalities_2$Group.1 <- gsub("(.*)[Ss][Nn][Oo][Ww](.*)","SNOW",injuries_fatalities_2$Group.1)
fatalities_new <- aggregate(injuries_fatalities_2$x,list(injuries_fatalities_2$Group.1),sum)
injuries_new <- aggregate(injuries_fatalities_2[,3],list(injuries_fatalities_2$Group.1),sum)
injuries_fatalities_def <- cbind(fatalities_new,injuries_new[,2])
colnames(injuries_fatalities_def) <- c("Event","fatalities","injuries")Here we can see a summary of the new results
print(injuries_fatalities_def)## Event fatalities injuries
## 1 AVALANCHE 223 156
## 2 BLIZZARD 70 385
## 3 Coastal Storm 0 1
## 4 COASTAL STORM 3 1
## 5 COASTALSTORM 1 0
## 6 COLD 372 127
## 7 DENSE FOG 9 143
## 8 DROUGHT 0 4
## 9 DROWNING 1 0
## 10 DRY MICROBURST 3 25
## 11 DUST 13 415
## 12 FIRE 87 1458
## 13 FLOOD 1309 8441
## 14 FOG 60 712
## 15 FREEZING DRIZZLE 2 13
## 16 FREEZING RAIN 2 0
## 17 Freezing Spray 1 0
## 18 FROST 1 3
## 19 FUNNEL CLOUD 0 1
## 20 GLAZE 1 212
## 21 HAIL 7 713
## 22 HEAT 2036 7683
## 23 HEAVY RAIN 94 230
## 24 HEAVY SEAS 1 0
## 25 HIGH SEAS 3 7
## 26 HIGH SWELLS 1 0
## 27 HIGH WATER 3 0
## 28 HURRICANE 125 1323
## 29 HYPERTHERMIA/EXPOSURE 1 0
## 30 Hypothermia/Exposure 4 0
## 31 HYPOTHERMIA/EXPOSURE 3 0
## 32 ICE 90 366
## 33 LANDSLIDE 37 52
## 34 LANDSLIDES 1 1
## 35 LIGHTNING 651 4141
## 36 Marine Accident 1 2
## 37 Mudslide 4 2
## 38 Mudslides 1 0
## 39 OTHER 0 4
## 40 RIP CURRENT 542 503
## 41 ROGUE WAVE 0 2
## 42 ROUGH SEAS 8 5
## 43 SMALL HAIL 0 10
## 44 SNOW 118 754
## 45 STORM SURGE 2 37
## 46 STORM SURGE/TIDE 11 5
## 47 SURF 142 240
## 48 THUNDERSTORM 398 5164
## 49 TORNADO 1511 20667
## 50 Torrential Rainfall 0 4
## 51 TROPICAL STORM 57 338
## 52 TSUNAMI 33 129
## 53 TYPHOON 0 5
## 54 UNSEASONABLY WARM 0 17
## 55 URBAN/SML STREAM FLD 28 79
## 56 WARM WEATHER 0 2
## 57 WATERSPOUT 2 2
## 58 WIND 404 1513
## 59 WINTER STORM 191 1292
## 60 WINTER WEATHER 33 343
## 61 WINTER WEATHER/MIX 31 243And the last pace is ordering the results in decreasing order of injuried people and dead people. We consider only the 15 most harmful events considered by our new groups
injuries_fatalities_def <- injuries_fatalities_def[order(injuries_fatalities_def$injuries,injuries_fatalities_def$fatalities,decreasing=TRUE),]
most_harmful_personal <- head(injuries_fatalities_def,15)
print(most_harmful_personal)## Event fatalities injuries
## 49 TORNADO 1511 20667
## 13 FLOOD 1309 8441
## 22 HEAT 2036 7683
## 48 THUNDERSTORM 398 5164
## 35 LIGHTNING 651 4141
## 58 WIND 404 1513
## 12 FIRE 87 1458
## 28 HURRICANE 125 1323
## 59 WINTER STORM 191 1292
## 44 SNOW 118 754
## 21 HAIL 7 713
## 14 FOG 60 712
## 40 RIP CURRENT 542 503
## 11 DUST 13 415
## 2 BLIZZARD 70 385Processing the data for calculating the economic impact
To make the required computations, we use the data from 1995, for the reasons explained previously. Using the dataset from 1995 as a source data, first we filter out only the events which cause important economic damage, corresponding to a PROPDMG and CROPDMG Columns with a value greater than zero
data_economic <- data_post_1995[(data_post_1995[,25]>0)|(data_post_1995[,27]>0),]We have to interpret the PROPDMGEXP and CROPDMGEXP as a correct multiplyer of PROPDMG and CRPDMG. To do this, now we change the levels of the factors corresponding to PROPDMGEXP and CROPDMGEXP
levels(data_economic$PROPDMGEXP)## [1] "" "-" "?" "+" "0" "1" "2" "3" "4" "5" "6" "7" "8" "B" "h" "H" "K"
## [18] "m" "M"levels(data_economic$CROPDMGEXP)## [1] "" "?" "0" "2" "B" "k" "K" "m" "M"levels(data_economic$PROPDMGEXP) <- c("0","1","1","1","1","10","100","1000","10000","100000","1000000","10000000","100000000","1000000000","100","100","1000","1000000","1000000")
levels(data_economic$CROPDMGEXP) <- c("0","1","1","100","1000000000","1000","1000","1000000","1000000")
head(data_economic)## STATE__ BGN_DATE BGN_TIME TIME_ZONE COUNTY
## 248768 1 1/6/1996 0:00:00 08:00:00 PM CST 1
## 248769 1 1/11/1996 0:00:00 06:35:00 PM CST 31
## 248770 1 1/11/1996 0:00:00 06:45:00 PM CST 31
## 248771 1 1/11/1996 0:00:00 07:05:00 PM CST 45
## 248772 1 1/11/1996 0:00:00 07:38:00 PM CST 67
## 248774 1 1/18/1996 0:00:00 06:00:00 PM CST 1
## COUNTYNAME STATE EVTYPE BGN_RANGE BGN_AZI
## 248768 ALZ001>038 AL WINTER STORM 0
## 248769 COFFEE AL TORNADO 5 N
## 248770 COFFEE AL TSTM WIND 0
## 248771 DALE AL TSTM WIND 0
## 248772 HENRY AL TSTM WIND 0
## 248774 ALZ001>018 - 020 - 022 AL HIGH WIND 0
## BGN_LOCATI END_DATE END_TIME COUNTY_END COUNTYENDN
## 248768 1/7/1996 0:00:00 03:00:00 PM 0 NA
## 248769 KINSTON 1/11/1996 0:00:00 06:36:00 PM 0 NA
## 248770 ENTERPRISE 1/11/1996 0:00:00 06:45:00 PM 0 NA
## 248771 PINCKARD 1/11/1996 0:00:00 07:05:00 PM 0 NA
## 248772 HEADLAND 1/11/1996 0:00:00 07:38:00 PM 0 NA
## 248774 1/19/1996 0:00:00 05:00:00 AM 0 NA
## END_RANGE END_AZI END_LOCATI LENGTH WIDTH F MAG FATALITIES
## 248768 0 0 0 NA 0 0
## 248769 5 N KINSTON 1 75 1 0 0
## 248770 0 ENTERPRISE 0 0 NA 0 0
## 248771 0 PINCKARD 0 0 NA 0 0
## 248772 0 HEADLAND 0 0 NA 0 0
## 248774 0 0 0 NA 40 0
## INJURIES PROPDMG PROPDMGEXP CROPDMG CROPDMGEXP WFO
## 248768 0 380 1000 38 1000 BMX
## 248769 0 100 1000 0 0 TAE
## 248770 0 3 1000 0 0 TAE
## 248771 0 5 1000 0 0 TAE
## 248772 0 2 1000 0 0 TAE
## 248774 0 400 1000 0 0 BMX
## STATEOFFIC
## 248768 ALABAMA, Central
## 248769 ALABAMA, Southeast
## 248770 ALABAMA, Southeast
## 248771 ALABAMA, Southeast
## 248772 ALABAMA, Southeast
## 248774 ALABAMA, Central
## ZONENAMES
## 248768 LAUDERDALE - LAUDERDALE - COLBERT - FRANKLIN - LAWRENCE - LIMESTONE - MADISON - MORGAN - MARSHALL - JACKSON - DEKALB - MARION - LAMAR - FAYETTE - WINSTON - WALKER - CULLMAN - BLOUNT - ETOWAH - CALHOUN - CHEROKEE - CLEBURNE - PICKENS - TUSCALOOSA - JEFFERSON - SHELBY - ST. CLAIR - TALLADEGA - CLAY - RANDOLPH - SUMTER - GREENE - HALE - PERRY - BIBB - CHILTON - COOSA - TALLAPOOSA - CHAMBERS
## 248769
## 248770
## 248771
## 248772
## 248774 LAUDERDALE - LAUDERDALE - COLBERT - FRANKLIN - LAWRENCE - LIMESTONE - MADISON - MORGAN - MARSHALL - JACKSON - DEKALB - MARION - LAMAR - FAYETTE - WINSTON - WALKER - CULLMAN - BLOUNT - ETOWAH - CHEROKEE - PICKENS
## LATITUDE LONGITUDE LATITUDE_E LONGITUDE_
## 248768 0 0 0 0
## 248769 3116 8608 3116 8608
## 248770 3119 8551 3119 8551
## 248771 3119 8533 3119 8533
## 248772 3121 8521 3121 8521
## 248774 0 0 0 0
## REMARKS
## 248768 A winter storm brought a mixture of freezing rain, sleet, and snow to the northern two-thirds of Alabama. Precipitation began as freezing rain and sleet but quickly changed to snow. The precipitation coated roads and caused serious travel problems across the northern sections of thestate that lasted into Monday morning (the 8th). Some higher elevations of the northeast corner of Alabama had travel problems into Tuesday. Amounts were generally light with the highest snowfall reported at Huntsville International Airport with 2 inches. Most other locations across North Alabama reported one-quarter of an inch to an inch and a half. On Sunday the 7th, one fatality occurred in an automobile/train collision in Calhoun County that was attributed to icy roads. The teenage driver of the car was not wearing a seat belt and was thrown from the vehicle.
## 248769 A tornado destroyed 4 house trailers that were unoccupied. Debris was scattered for about 1 mile, according to county emergency management.
## 248770 Several trees were blown down and two backyard sheds were destroyed according to newspaper reports and county emergency management.
## 248771 County emergency management confirmed that three sheds were destroyed, and several houses received superficial damage.
## 248772 County emergency management reported that a porch roof was lifted off a home and a shed was destroyed.
## 248774 High wind over the northern third of Alabama generally along a cold front caused widespread scattered damage. Damage was primarily to downed trees, limbs, and power lines but some roof damage occurred in several locations.
## REFNUM year
## 248768 248768 1996
## 248769 248769 1996
## 248770 248770 1996
## 248771 248771 1996
## 248772 248772 1996
## 248774 248774 1996Now we are creating a new variable which summarizes the total damage
total_economic_damage <- data_economic$PROPDMG * (as.numeric(as.character(data_economic$PROPDMGEXP))) + data_economic$CROPDMG*(as.numeric(as.character(data_economic$CROPDMGEXP)))We are only interested in the new total damage variable and the type of event. For this reason, we create a new subset with only the two columns mentioned
economic_damage_event <- cbind(as.character(data_economic$EVTYPE),total_economic_damage)
colnames(economic_damage_event) <- c("Event","Damage")Next step is group the results by Event Types
economic_damage_by_event <- aggregate(as.numeric(economic_damage_event[,2]),list(economic_damage_event[,1]),sum)And then, as groups are very similar to the used on the personal damage case, we use ALMOST the same simplifying of groups. Here we add a processing for the name groups similar to “Storm Surge” to have only one group
economic_damage_by_event$Group.1 <- gsub("(.*)COLD(.*)","COLD",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)Cold(.*)","COLD",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)SURF(.*)","SURF",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)Surf(.*)","SURF",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)Tornado(.*)","TORNADO",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)TORNADO(.*)","TORNADO",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)THUNDER(.*)TORM(.*)","THUNDERSTORM",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)Thunder(.*)","THUNDERSTORM",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)TSTM(.*)","THUNDERSTORM",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)FLOOD(.*)","FLOOD",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)Flood(.*)","FLOOD",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)HURRICANE(.*)","HURRICANE",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)Hurricane(.*)","HURRICANE",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)LIGHTNING(.*)","LIGHTNING",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)FIRE(.*)","FIRE",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)[Ww]ind(.*)","WIND",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)WIND(.*)","WIND",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)Heat(.*)","HEAT",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)HEAT(.*)","HEAT",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)Dust(.*)","DUST",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)DUST(.*)","DUST",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)MIX(.*)","WINTER WEATHER/MIX",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)RIP CURRENT(.*)","RIP CURRENT",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)ICE(.*)","ICE",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)ICY(.*)","ICE",economic_damage_by_event$Group.1)
economic_damage_by_event$Group.1 <- gsub("(.*)[Ss][Nn][Oo][Ww](.*)","SNOW",economic_damage_by_event$Group.1)
#Extra modification
economic_damage_by_event$Group.1 <- gsub("(.*)STORM SURGE(.*)","STORM SURGE",economic_damage_by_event$Group.1)
damage_new <- aggregate(economic_damage_by_event$x,list(economic_damage_by_event$Group.1),sum)
colnames(damage_new) <- c("Event","Damage")
print(damage_new)## Event Damage
## 1 AGRICULTURAL FREEZE 28820000
## 2 ASTRONOMICAL HIGH TIDE 9425000
## 3 ASTRONOMICAL LOW TIDE 320000
## 4 AVALANCHE 3711800
## 5 Beach Erosion 100000
## 6 BLIZZARD 532718950
## 7 COASTAL EROSION 766000
## 8 Coastal Storm 50000
## 9 COLD 1370817900
## 10 DAM BREAK 1002000
## 11 Damaging Freeze 34130000
## 12 DAMAGING FREEZE 8000000
## 13 DENSE FOG 7319000
## 14 DENSE SMOKE 100000
## 15 DOWNBURST 2000
## 16 DROUGHT 14413667000
## 17 DRY MICROBURST 1747600
## 18 DUST 9257630
## 19 Early Frost 42000000
## 20 FIRE 8162704630
## 21 FLOOD 166047107120
## 22 FOG 13145500
## 23 Freeze 10500000
## 24 FREEZE 146425000
## 25 Freezing drizzle 15000
## 26 Freezing Drizzle 55000
## 27 FREEZING DRIZZLE 35000
## 28 FREEZING FOG 2182000
## 29 Freezing Rain 35000
## 30 FREEZING RAIN 591000
## 31 FROST 15000
## 32 Frost/Freeze 1100000
## 33 FROST/FREEZE 1103566000
## 34 FUNNEL CLOUD 134100
## 35 Glaze 90000
## 36 GLAZE 60000
## 37 HAIL 17071172870
## 38 HARD FREEZE 12900000
## 39 HEAT 501822200
## 40 HEAVY RAIN 1313034240
## 41 HIGH SEAS 15000
## 42 HIGH SWELLS 5000
## 43 HURRICANE 86467941810
## 44 ICE 3658252010
## 45 Ice jam flood (minor 1000
## 46 LANDSLIDE 344595000
## 47 LANDSLIDES 5000
## 48 Landslump 570000
## 49 LANDSPOUT 7000
## 50 LIGHT FREEZING RAIN 451000
## 51 LIGHTNING 749975520
## 52 Marine Accident 50000
## 53 MARINE HAIL 4000
## 54 Microburst 20000
## 55 Mixed Precipitation 235000
## 56 MUD SLIDE 100100
## 57 MUDSLIDE 1225000
## 58 Other 50000
## 59 OTHER 1039900
## 60 RAIN 550000
## 61 RIP CURRENT 163000
## 62 ROCK SLIDE 150000
## 63 SEICHE 980000
## 64 SMALL HAIL 20863000
## 65 SNOW 753113640
## 66 STORM SURGE 47835579000
## 67 SURF 110374500
## 68 THUNDERSTORM 8936415880
## 69 TORNADO 24900370720
## 70 TROPICAL DEPRESSION 1737000
## 71 TROPICAL STORM 8320186550
## 72 TSUNAMI 144082000
## 73 TYPHOON 601055000
## 74 UNSEASONABLY WARM 10000
## 75 UNSEASONAL RAIN 10000000
## 76 URBAN/SML STREAM FLD 66797750
## 77 VOLCANIC ASH 500000
## 78 WATERSPOUT 5730200
## 79 WET MICROBURST 35000
## 80 WIND 6148911240
## 81 WINTER STORM 1544687250
## 82 WINTER WEATHER 35866000
## 83 WINTER WEATHER/MIX 6997000
## 84 Wintry Mix 2500And the last pace is ordering the results in decreasing order of economic damaged produced by the events. We consider only the 15 most harmful events considered by our new groups
damage_new <- damage_new[order(damage_new$Damage,decreasing=TRUE),]
most_harmful_economic <- head(damage_new,15)
print(most_harmful_economic)## Event Damage
## 21 FLOOD 166047107120
## 43 HURRICANE 86467941810
## 66 STORM SURGE 47835579000
## 69 TORNADO 24900370720
## 37 HAIL 17071172870
## 16 DROUGHT 14413667000
## 68 THUNDERSTORM 8936415880
## 71 TROPICAL STORM 8320186550
## 20 FIRE 8162704630
## 80 WIND 6148911240
## 44 ICE 3658252010
## 81 WINTER STORM 1544687250
## 9 COLD 1370817900
## 40 HEAVY RAIN 1313034240
## 33 FROST/FREEZE 1103566000Results
Dead people by events
First of all we are going to show the 15 most harmful events condidering the amount of deaths involved in (number of people)
library("lattice")
barchart(most_harmful_personal$Event~most_harmful_personal$fatalities,xlab="Dead people by event",ylab="Events")
The top3 harmful events, considering the amount of people directly dead is:
- HEAT
- TORNADO
- FLOOD
Injuried people by events
Our second plot shows the 15 most harmful events condidering the amount of injuried people as a consequence of the event
barchart(most_harmful_personal$Event~most_harmful_personal$injuries,xlab="Injuried people by event",ylab="Events")
The top3 harmful events, considering the amount of people injuried is:
- TORNADO
- HEAT
- FLOOD
Economic impact caused by events
Now the 15 most harmful events condidering the total economic impact (in dollars) adding corporal and property damages
barchart(most_harmful_economic$Event~most_harmful_economic$Damage,xlab="Economic damage caused by event",ylab="Events")
The top3 harmful events, considering the economic impact caused is:
- FLOOD
- HURRICANE
- STORM SURGE