Raw HURDAT

Tim Trice

2020-01-21

Introduction

HURDAT is a re-analysis project of all tropical cyclones by the National Hurricane Center (NHC) Hurricane Research Division(NRD). Known as HURDAT2, the text files are typically updated in late winter, early spring to include the previous season’s activity. These text files will not contain data on current tropical systems. For that I would recommend my Hurricanes package (not yet available in CRAN).

As of this vignette data up to and including 2016 has been added.

There are two core datasets available: one for the Atlantic (“AL”) basin (northern hemisphere) and Pacific basin (northwestern hemisphere). The Pacific basin is divided into two sections: east Pacific and central Pacific. However, both sections are included in the “EP” dataset.

str(EP)
## 'data.frame':    28537 obs. of  21 variables:
##  $ Key     : chr  "EP011949" "EP011949" "EP011949" "EP011949" ...
##  $ Name    : chr  "UNNAMED" "UNNAMED" "UNNAMED" "UNNAMED" ...
##  $ DateTime: POSIXct, format: "1949-06-11 00:00:00" "1949-06-11 06:00:00" ...
##  $ Record  : chr  NA NA NA NA ...
##  $ Status  : chr  "TS" "TS" "TS" "TS" ...
##  $ Lat     : num  20.2 20.2 20.2 20.3 20.4 20.5 20.6 14.1 14.6 15 ...
##  $ Lon     : num  -106 -106 -107 -108 -109 ...
##  $ Wind    : int  45 45 45 45 45 45 45 45 45 45 ...
##  $ Pressure: int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NE34    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SE34    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SW34    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NW34    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NE50    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SE50    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SW50    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NW50    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NE64    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SE64    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SW64    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NW64    : int  NA NA NA NA NA NA NA NA NA NA ...
str(AL)
## 'data.frame':    51346 obs. of  21 variables:
##  $ Key     : chr  "AL011851" "AL011851" "AL011851" "AL011851" ...
##  $ Name    : chr  "UNNAMED" "UNNAMED" "UNNAMED" "UNNAMED" ...
##  $ DateTime: POSIXct, format: "1851-06-25 00:00:00" "1851-06-25 06:00:00" ...
##  $ Record  : chr  NA NA NA NA ...
##  $ Status  : chr  "HU" "HU" "HU" "HU" ...
##  $ Lat     : num  28 28 28 28.1 28.2 28.2 28.3 28.4 28.6 29 ...
##  $ Lon     : num  -94.8 -95.4 -96 -96.5 -96.8 -97 -97.6 -98.3 -98.9 -99.4 ...
##  $ Wind    : int  80 80 80 80 80 70 60 60 50 50 ...
##  $ Pressure: int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NE34    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SE34    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SW34    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NW34    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NE50    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SE50    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SW50    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NW50    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NE64    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SE64    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ SW64    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ NW64    : int  NA NA NA NA NA NA NA NA NA NA ...

Usage

The datasets are extracted from online with the function get_hurdat. The basin parameter takes either “AL” for Atlantic or “EP” for northeastern and central Pacific.

AL <- get_hurdat(basin = "AL")

get_basin does not require basin be provided; by default it will access both datasets and return a dataframe approximately 80,000 x 21.

Most of the early storms will have missing data.

The Atlantic HURDAT codebook contains the definitions of the various fields. A codebook also exists for the northeast and central Pacific.

There are significantly more storms for the Atlantic than in the eastern Pacific. There are only a few storms for the central Pacific.

Warnings

Perform QA on the dataset; particularly latitude and longitude. I have found some errors in previous datasets that do not appear to be resolved. They have been reported. Additionally, there may be some other bad data I’m not aware of.

Regardless, if you find anything that is incorrect (either on my end or in the original dataset), please submit an issue and I will review it.