Global Earth Surface Temperature Data Visualization Using Python
Data Source : Berkeley Earth
Python Script :
import csv
import numpy as np
import datetime as dt
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
def read_dates(file_name):
dates = np.genfromtxt(file_name,delimiter = ",",skip_header=0,dtype=str,filling_values='null',usecols=(0,1) )
dates = dates[1:,0]
dates_list = [dt.datetime.strptime(date, '%Y-%m-%d').date() for date in dates]
return dates_list
def read_LandAvgTemp(file_name):
lat = np.genfromtxt(file_name,delimiter = ",",skip_header=0,dtype=str,filling_values='null',usecols=(0,1) )
lat = lat[1:,1]
return lat
def clean_data(dates_list,lat):
dump = np.nonzero(lat=='')[0]
dates_list = np.delete(dates_list,dump,axis=0)
lat = np.delete(lat,dump,axis=0)
return (dates_list,lat)
def sort_by_month(dates_list,lat,month):
month_index=[]
for i in range(0,len(dates_list)-1):
if(dates_list[i].month==month):
month_index.append(i)
return month_index
def plot_by_month(X,Y,month):
month_index=sort_by_month(dates_list,lat,month)
temp_month=[]
temp_lat=[]
for index in month_index:
temp_month.append(dates_list[index])
temp_lat.append(lat[index])
X = temp_month
Y = temp_lat
plt.plot(X,Y)
plt.xlabel(set_month(month),fontsize=20)
plt.ylabel('Avg Temperature')
plt.savefig('test.jpg')
plt.show()
def set_month(month):
year = ["January","February","March","April","May","June","July","August","September","October","November","December"]
set = year[month-1]
return set
dates_list=read_dates('GlobalTemperatures.csv')
lat=read_LandAvgTemp('GlobalTemperatures.csv')
dates_list,lat = clean_data(dates_list,lat)
lat=lat.astype(float)
X=dates_list
Y=lat
for month in range(1,13):
plot_by_month(X,Y,month)
General Intuition : There has been a fairly great reduction in the difference of maximum and minimum temperatures , Moreover rise of Earth's global surface temperature is clearly depicted in the figures .
Coming up next : Time-Series Analysis
