India's Rainfall analysis¶
Data visualisations made by Janhavi Pimplikar
A student at Pimpri Chinchwad College of Engineering
In [83]:
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
In [84]:
import seaborn as sns
import plotly as py
import cufflinks as cf
In [85]:
import plotly.express as px
import plotly.graph_objects as go
import plotly.figure_factory as ff
from plotly.subplots import make_subplots
In [86]:
py.offline.init_notebook_mode(connected=True)
In [87]:
cf.go_offline()
from plotly.offline import iplot
Accessing the number of districts in each state/union territories¶
In [88]:
rainfall=pd.read_csv('IndiaRainfall.csv')
In [89]:
rainfall
Out[89]:
In [90]:
states=rainfall[['STATE_UT_NAME','DISTRICT']]
In [91]:
statedata=rainfall['STATE_UT_NAME'].value_counts()
In [92]:
#above plot using plotly module
statedata.iplot(kind='bar',color='blue',legend=True,xTitle='Number of districts',yTitle='States and Union Territories',title='Number of districts in each state and union territory of India',orientation='v',bargap=0.7,theme='solar')
Analysis of rainfall received by each state/union territory (annual average and quarterly averages) as a whole¶
In [93]:
annual=rainfall[['STATE_UT_NAME','ANNUAL','Jan-Feb','Mar-May','Jun-Sep','Oct-Dec']]
In [94]:
annual
Out[94]:
In [95]:
annual=annual.groupby('STATE_UT_NAME')[['ANNUAL','Jan-Feb','Mar-May','Jun-Sep','Oct-Dec']].mean() #The .mean() function will give us the average rainfall received by the state
In [96]:
annual.reset_index(inplace=True)
In [97]:
annual.iplot(x='STATE_UT_NAME',y='ANNUAL',mode='markers',bargroupgap=0.2,orientation='v',legend=True,xTitle='State/Union territory',yTitle='Rainfall (in mm)',title='Average annual rainfall in each state/union territory',color='red',theme='solar')
In [98]:
annual.iplot(x='STATE_UT_NAME',y='Jan-Feb',kind='bar',bargroupgap=0.2,orientation='v',legend=True,xTitle='State/Union territory',yTitle='Rainfall (in mm)',title='Average rainfall in each state/union territory (Jan-Feb)',color='pink',theme='solar')
In [99]:
annual.iplot(x='STATE_UT_NAME',y='Mar-May',kind='bar',bargroupgap=0.2,orientation='v',legend=True,xTitle='State/Union territory',yTitle='Rainfall (in mm)',title='Average rainfall in each state/union territory (Mar-May)',color='lightgreen',theme='solar')
In [100]:
annual.iplot(x='STATE_UT_NAME',y='Jun-Sep',kind='bar',bargroupgap=0.2,orientation='v',legend=True,xTitle='State/Union territory',yTitle='Rainfall (in mm)',title='Average rainfall in each state/union territory (Jun-Sep)',color='orange',theme='solar')
In [101]:
annual.iplot(x='STATE_UT_NAME',y='Oct-Dec',kind='bar',bargroupgap=0.2,orientation='v',legend=True,xTitle='State/Union territory',yTitle='Rainfall (in mm)',title='Average rainfall in each state/union territory (Oct-Dec)',color='purple',theme='solar')
In [ ]:
Top 10 states/union territories contributing towards the heaviest rainfall (annually and quarterly)¶
In [102]:
top10=annual.sort_values('ANNUAL',ascending=False).head(10)
In [103]:
top10
Out[103]:
In [104]:
fig,ax=plt.subplots(figsize=(8,8))
lab=top10['STATE_UT_NAME']
top10.plot.pie(y='ANNUAL',legend=False,shadow=True,ax=ax,autopct="%1.2f%%",colors=sns.color_palette('Accent'),fontsize=15,labels=lab)
plt.title('Percentage share of top 10 states/union territories receiving the most rainfall',size=15)
fig.show()
In [105]:
top10_first=annual.sort_values('Jan-Feb',ascending=False).head(10)
top10_second=annual.sort_values('Mar-May',ascending=False).head(10)
In [106]:
top10_first
Out[106]:
In [107]:
top10_second
Out[107]:
In [108]:
fig,(ax1,ax2)=plt.subplots(nrows=1,ncols=2,figsize=(16,16))
fig.subplots_adjust(hspace=0.9)
labels1=top10_first['STATE_UT_NAME']
labels2=top10_second['STATE_UT_NAME']
patches,texts,autotexts=ax1.pie(top10_first['Jan-Feb'],shadow=True,labels=labels1,autopct='%1.2f%%')
plt.setp(autotexts,size=12)
ax1.set_title('% share (Jan-Feb)',size=15)
##############################################
patches,texts,autotexts=ax2.pie(top10_second['Mar-May'],shadow=True,labels=labels2,autopct='%1.2f%%')
plt.setp(autotexts,size=12)
ax2.set_title('% share (Mar-May)',size=15)
fig.show()
In [109]:
top10_third=annual.sort_values('Jun-Sep',ascending=False).head(10)
top10_fourth=annual.sort_values('Oct-Dec',ascending=False).head(10)
In [110]:
top10_third
Out[110]:
In [111]:
top10_fourth
Out[111]:
In [112]:
fig,(ax1,ax2)=plt.subplots(nrows=1,ncols=2,figsize=(16,16))
fig.subplots_adjust(hspace=0.9)
labels1=top10_third['STATE_UT_NAME']
labels2=top10_fourth['STATE_UT_NAME']
patches,texts,autotexts=ax1.pie(top10_third['Jun-Sep'],shadow=True,labels=labels1,autopct='%1.2f%%')
plt.setp(autotexts,size=12)
ax1.set_title('% share (Jun-Sep)',size=15)
##############################################
patches,texts,autotexts=ax2.pie(top10_fourth['Oct-Dec'],shadow=True,labels=labels2,autopct='%1.2f%%')
plt.setp(autotexts,size=12)
ax2.set_title('% share (Oct-Dec)',size=15)
fig.show()
In [ ]:
Analysing monthly rainfall for 6 states/union territories as a whole¶
In [113]:
monthly_rain=rainfall.groupby('STATE_UT_NAME')[['JAN','FEB','MAR','APR','MAY','JUN','JUL','AUG','SEP','OCT','NOV','DEC']].mean()
In [114]:
monthly_rain
Out[114]:
In [ ]:
In [115]:
mh=monthly_rain.loc['MAHARASHTRA']
In [116]:
mh.iplot(kind='line',mode='markers',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend in Maharashtra',color='blue',fill=True,legend=True,theme='solar')
In [117]:
ors=monthly_rain.loc['ORISSA']
In [118]:
ors.iplot(kind='line',mode='markers',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend in Orissa',color='pink',fill=True,legend=True,theme='solar')
In [119]:
pun=monthly_rain.loc['PUNJAB']
In [120]:
pun.iplot(kind='line',mode='markers',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend in Punjab',color='orange',fill=True,legend=True,theme='solar')
In [121]:
tam=monthly_rain.loc['TAMIL NADU']
In [122]:
tam.iplot(kind='line',mode='markers',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend in Tamil Nadu',color='green',fill=True,legend=True,theme='solar')
In [123]:
meg=monthly_rain.loc['MEGHALAYA']
In [124]:
meg.iplot(kind='line',mode='markers',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend in Meghalaya',color='red',fill=True,legend=True,theme='solar')
In [125]:
pon=monthly_rain.loc['PONDICHERRY']
In [126]:
pon.iplot(kind='line',mode='markers',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend in Pondicherry',color='purple',fill=True,legend=True,theme='solar')
Variation in rainfall for each district of selected states/union territories¶
1. Maharashtra¶
In [127]:
mh_district=rainfall.loc[rainfall['STATE_UT_NAME']=="MAHARASHTRA"]
In [128]:
sub_mh=mh_district.groupby('DISTRICT')['JAN','FEB','MAR','APR','MAY','JUN','JUL','AUG','SEP','OCT','NOV','DEC'].sum()
In [129]:
sub_mh=sub_mh.transpose()
In [130]:
sub_mh.iplot(kind='line',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend over all districts in Maharashtra',theme='solar')
In [ ]:
2. Orissa¶
In [131]:
ors_district=rainfall.loc[rainfall['STATE_UT_NAME']=='ORISSA']
In [132]:
sub_ors=ors_district.groupby('DISTRICT')['JAN','FEB','MAR','APR','MAY','JUN','JUL','AUG','SEP','OCT','NOV','DEC'].sum()
In [133]:
sub_ors=sub_ors.transpose()
In [134]:
sub_ors.iplot(kind='line',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend over all districts in Orissa ',theme='solar')
In [ ]:
3. Punjab¶
In [135]:
pun_district=rainfall.loc[rainfall['STATE_UT_NAME']=='PUNJAB']
In [136]:
sub_pun=pun_district.groupby('DISTRICT')['JAN','FEB','MAR','APR','MAY','JUN','JUL','AUG','SEP','OCT','NOV','DEC'].sum()
In [137]:
sub_pun=sub_pun.transpose()
In [138]:
sub_pun.iplot(kind='line',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend over all districts in Punjab',theme='solar')
In [ ]:
4. Tamil Nadu¶
In [139]:
tn_district=rainfall.loc[rainfall['STATE_UT_NAME']=='TAMIL NADU']
In [140]:
sub_tn=tn_district.groupby('DISTRICT')['JAN','FEB','MAR','APR','MAY','JUN','JUL','AUG','SEP','OCT','NOV','DEC'].sum()
In [141]:
sub_tn=sub_tn.transpose()
In [142]:
sub_tn.iplot(kind='line',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall over all districts in Tamil Nadu',theme='solar')
In [ ]:
5. Meghalaya¶
In [143]:
meg_district=rainfall.loc[rainfall['STATE_UT_NAME']=='MEGHALAYA']
In [144]:
sub_meg=meg_district.groupby('DISTRICT')['JAN','FEB','MAR','APR','MAY','JUN','JUL','AUG','SEP','OCT','NOV','DEC'].sum()
In [145]:
sub_meg=sub_meg.transpose()
In [146]:
sub_meg.iplot(kind='line',xTitle='Months',yTitle='Rainfall (in mm)', title='Rainfall over all districts in Meghalaya',theme='solar')
In [ ]:
6. Pondicherry¶
In [147]:
pond_district=rainfall.loc[rainfall['STATE_UT_NAME']=='PONDICHERRY']
In [148]:
sub_pd=pond_district.groupby('DISTRICT')['JAN','FEB','MAR','APR','MAY','JUN','JUL','AUG','SEP','OCT','NOV','DEC'].sum()
In [149]:
sub_pd=sub_pd.transpose()
In [150]:
sub_pd.iplot(kind='line',xTitle='Months',yTitle='Rainfall (in mm)',title='Rainfall trend over all districts in Pondicherry',theme='solar')
In [ ]:
Dataset: imported from https://www.kaggle.com/rajanand/rainfall-in-india?select=district+wise+rainfall+normal.csv
Attribution: (websites such as stackoverflow.com, medium.com etc.)