你是不是做实验经常性的需要一些气象数据,例如PM2.5、相对湿度、月均温度等等……
但是当你开始寻找数据时就遇到困难了,由于权限、数据网站之类的麻烦你会花费大量无用时间,甚至有时候一无所获得不偿失,这就很头疼了!那么我们就去了解一个好的气象数据网免费获取站吧,保证你有所收获。
气象数据一直是一个价值较高的数据,它被广泛用于各个领域的研究当中。气象数据包括有气温、气压、相对湿度、降水、蒸发、风向风速、日照等多种指标,但是包含了这些全部指标的气象数据却较难获取,即使获取到了也不能随意分享。
1级目录 | 文件名 |
PRS | SURF_CLI_CHN_MUL_DAY-PRS-10004-YYYYMM.TXT(本站气压) |
TEM | SURF_CLI_CHN_MUL_DAY-TEM-12001-YYYYMM.TXT(气温) |
RHU | SURF_CLI_CHN_MUL_DAY-RHU-13003-YYYYMM.TXT(相对湿度) |
PRE | SURF_CLI_CHN_MUL_DAY-PRE-13011-YYYYMM.TXT(降水) |
EVP | SURF_CLI_CHN_MUL_DAY-EVP-13240-YYYYMM.TXT(蒸发) |
WIN | SURF_CLI_CHN_MUL_DAY-WIN-11002-YYYYMM.TXT(风向风速) |
SSD | SURF_CLI_CHN_MUL_DAY-SSD-14032-YYYYMM.TXT(日照) |
GST | SURF_CLI_CHN_MUL_DAY-GST-12030-0cm-YYYYMM.TXT(0cm地温) |
气象数据一直是一个价值较高的数据,它被广泛用于各个领域的研究当中。气象数据包括有气温、气压、相对湿度、降水、蒸发、风向风速、日照等多种指标,但是包含了这些全部指标的气象数据却较难获取,即使获取到了也不能随意分享。
想要大规模爬取的话,需要自己写爬虫,我之前写过一个爬取深圳市数据的爬虫。对深圳市的天气数据爬取基本没有问题。
import requests
import demjson
import re
import calendar
import csv
headers = {
'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_4) AppleWebKit/537.36\
(KHML, like Gecko) Chrome/52.0.2743.116 Safari/537.36',
}
def get_url(date):
url = 'https://www.timeanddate.com/scripts/cityajax.php?n=china/shenzhen&mode=historic'
url += '&hd=' + date
url += '&month=' + str(int(date[4:6]))
url += '&year=' + date[:4] + '&json=1'
return url
# input: type(str) eg:'20170601'
def crawl_single_day(date):
response = requests.get(get_url(date), headers=headers)
response_list = demjson.decode(response.text)
for weather in response_list:
w_time = re.compile(r'^\d+:\d+').search(weather['c'][0]['h']).group(0)
w_temperature = re.compile(
r'^-?\d+').search(weather['c'][2]['h']).group(0)
w_weather = re.compile(
r'^(.*?)\.').search(weather['c'][3]['h']).group(1)
if weather['c'][4]['h'] == 'No wind':
w_wind_speed = '0'
else:
w_wind_speed = re.compile(
r'^\d+').search(weather['c'][4]['h']).group(0)
w_wind_direction = re.compile(
r'title=\"(.*?)\"').search(weather['c'][5]['h']).group(1)
w_humidity = weather['c'][6]['h']
w_barometer = re.compile(r'^\d+').search(weather['c'][7]['h']).group(0)
w_visibility = weather['c'][8]['h']
if w_visibility != 'N/A':
w_visibility=re.compile(r'^\d+').search(w_visibility).group(0)
yield [date, w_time, w_temperature, w_weather, w_wind_speed, w_wind_direction,
w_humidity, w_barometer, w_visibility]
# input: type(int) eg: year=2017, month=6
def crawl_single_month(year, month):
_, num_day = calendar.monthrange(year, month)
month_str = str(year)
if month < 10:
month_str += '0' + str(month)
else:
month_str += str(month)
day_list = list(range(1, num_day + 1))
for day in day_list:
if day < 10:
for weather in crawl_single_day(month_str + '0' + str(day)):
yield weather
else:
for weather in crawl_single_day(month_str + str(day)):
yield weather
if __name__ == "__main__":
with open('weather0.csv', 'w', encoding='utf-8', newline='') as file:
writer = csv.writer(file)
writer.writerow('date time temperature weather wind_speed wind_direction humidity barometer visibility'.split())
for month in range(7, 13):
writer.writerows(crawl_single_month(2017, month))
with open('weather1.csv', 'w', encoding='utf-8', newline='') as file:
writer = csv.writer(file)
writer.writerow('date time temperature weather wind_speed wind_direction humidity barometer visibility'.split())
writer.writerows(crawl_single_day('20210401'))
对 20210401的深圳天气数据爬取获得的 csv 文件如下图所示:
当然啦,中国地面气象日值数据集发布的气象数据包括有气温、气压、相对湿度、降水、蒸发、风向风速、日照太阳辐射等等多种指标,全国共2400+站点。
原文链接:https://bbs.csdn.net/forums/gisrs?spm=1001.2014.3001.6682