Aggregate (count) catalog data by year/month and plot it

catalog.py

"""
File: catalog.py
Author: Andrey Stepnov
Email: [email protected], [email protected]
Github: https://github.com/jamm1985
Description: Aggregate (count) catalog data by year/month and plot it

Catalog sources: https://earthquake.usgs.gov, https://eqalert.ru
"""

import pandas as pd
import matplotlib as mlp
import pylab as plt

def parse_eqalert_ru(xlsx):
    """read xlsx, clean and save data to dataframe
    
    :xlsx: string, input XLSX file
    :returns: pandas dataframe
    """
    data = pd.read_excel(xlsx)
    data = data[['event_datetime',
        'lat', 'lon', 'depth',
        'mag', 'ep_dis', 'msk64_value', 'sta_num']]
    data['event_datetime'] = pd.to_datetime(data['event_datetime'], format='%Y-%m-%d %H:%M:%S')
    return data

def parse_usgs_gov(csv):
    """read csv file, clean and save the data to dataframe

    :csv: string, input csv file
    :returns: pandas dataframe

    """
    data = pd.read_csv(csv)
    data['time'] = pd.to_datetime(data['time'])
    data = data[['time',
        'latitude', 'longitude', 'depth',
        'mag', 'dmin', 'nst']]
    data.columns = ['event_datetime', 'lat', 'lon', 'depth', 'mag', 'ep_dis', 'sta_num']
    return data

## USGS catalog

data = parse_usgs_gov('2020-07-11-usgs-query.csv')

## OR

## EQALERT catalog

data = parse_eqalert_ru('reports_list_09-04-10072020.xlsx')

## selection by date, MAG etc
data = data[data['mag'] > 4.0]

data = data[data['event_datetime'] > '1960']

## group by month and by year
by_month =  pd.DataFrame(data['event_datetime'].dt.strftime('%Y-%m'))
by_month['mag'] = data['mag']
by_month['max_mag'] = by_month.groupby(['event_datetime'])['mag'].transform(max)
by_month = by_month.groupby(['event_datetime', 'max_mag']).size().reset_index().rename(columns={0: 'counts'})

by_year =  pd.DataFrame(data['event_datetime'].dt.strftime('%Y'))
by_year['mag'] = data['mag']
by_year['max_mag'] = by_year.groupby(['event_datetime'])['mag'].transform(max)
by_year = by_year.groupby(['event_datetime', 'max_mag']).size().reset_index().rename(columns={0: 'counts'})

## plot
by_month.plot.line(x='event_datetime', y=['counts','max_mag'], subplots=True)
plt.show()

by_year.plot.line(x='event_datetime', y=['counts','max_mag'], subplots=True)
plt.show()


## save groupped data
by_month.to_excel("by_month.xlsx")
by_year.to_excel("by_year.xlsx")

catalogs in csv and XLSX here