python對NOAA天氣數據格式轉換

原文連接：http://tecdat.cn/?p=7679

轉換NOAA天氣數據文件「 .fly」爲Pandas DataFrame

 獲取數據 ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/daily

In [1]:

import matplotlib.pyplot as pltimport pandas as pdimport numpy as npimport reimport ftplib% matplotlib notebook

In [2]:

# download data from FTPdef download_file_from_ftp ( FTP_SERVER , FTP_PATH , FILENAME ):with ftplib . FTP ( FTP_SERVER ) as ftp :ftp . login ()ftp . cwd ( FTP_PATH )with open ( FILENAME , 'wb' ) as f :ftp . retrbinary ( 'RETR ' + FILENAME , f . write )

查詢站ID

In [3]:

def get_station_ID ( station_to_find , filename ):for line in open ( filename ):if station_to_find in line :line_with_station = linestation_ID = re . split ( " " , line_with_station )[ 0 ]return station_IDreturn None# warning, it is slow, download it only oncedownload_file_from_ftp ( "ftp.ncdc.noaa.gov" , "/pub/data/ghcn/daily" , "ghcnd-stations.txt" )station_to_find = "GUANGZHOU" # USE CAPSstation_ID = get_station_ID ( station_to_find , "ghcnd-stations.txt" )

下載天氣數據

In [4]:

weather_data_filename = station_ID + '.dly'# warning, it is slow, download it only oncedownload_file_from_ftp ( "ftp.ncdc.noaa.gov" , "/pub/data/ghcn/daily/all" , weather_data_filename )

將.fly轉換爲pandas Dataframe

In [7]:

df = convert_dly_to_dataframe ( weather_data_filename )df . head ()

Out[7]:

YEAR

MONTH

ELEMENT

VALUE1

VALUE2

VALUE3

VALUE4

VALUE5

VALUE6

VALUE7

...

VALUE22

VALUE23

VALUE24

VALUE25

VALUE26

VALUE27

VALUE28

VALUE29

VALUE30

VALUE31

0

1945

11

TAVG

NaN

NaN

NaN

NaN

NaN

NaN

NaN

...

NaN

NaN

NaN

NaN

NaN

NaN

NaN

NaN

107.0

NaN

1

1945

12

TAVG

123.0

136.0

152.0

144.0

146.0

189.0

219.0

...

179.0

146.0

128.0

107.0

104.0

112.0

122.0

127.0

129.0

156.0

2

1946

1

TAVG

150.0

150.0

123.0

117.0

112.0

121.0

125.0

...

146.0

153.0

173.0

196.0

211.0

212.0

218.0

201.0

156.0

131.0

3

1946

2

TAVG

114.0

112.0

147.0

181.0

195.0

192.0

149.0

...

201.0

196.0

231.0

226.0

221.0

229.0

240.0

NaN

NaN

NaN

4

1946

3

TAVG

237.0

162.0

142.0

133.0

183.0

187.0

160.0

...

183.0

192.0

205.0

216.0

223.0

238.0

207.0

195.0

233.0

228.0