# Calculate climatologies of observed daily sea ice area in the Arctic (interpolated)
# 12 periods: Jan-Jun, Feb-Jul, Mar-Aug, Apr-Sep, May-Oct, Jun-Nov,
#             Jul-Dec, Aug-Jan, Sep-Feb, Oct-Mar, Nov-Apr, Dec-May
# 20 years: 2003-2022
# Convert climatologies into CSV file
#
# Script version 2.1.0
#
# Script works with:
#   Python version 3.10.13
#
#   Package version
#    numpy:  1.26.3
#    pandas:  2.2.0
#
# Ver1.0.0: Created by Mariko Koseki, 13.12.2023
# Ver1.1.0: Updated by Mariko, 06.02.2024
# Ver2.0.0: Updated by Mariko, 13.08.2024
# Ver2.1.0: Updated by Mariko, 15.08.2024


import numpy as np
import pandas as pd
import datetime
import os
import glob
import sys
import platform  # Ver1.1.0




print('--- Python version ---')
print(platform.python_version())

print('--- Package version ---')
print('numpy: ', np.__version__)
print('pandas: ', pd.__version__)



##--- Select variable -----------------------------------## #ver2.0.0
var = 'area'
#var = 'extent'



##--- Create list of year -------------------------------## #ver2.0.0
yyyy1 = 2003
yyyy2 = 2022
yyyy_int_list = []
for yyyy in range(yyyy1, yyyy2+1):
    yyyy_int_list.append(yyyy)

#print(yyyy_int_list)




##--- Set path ---------------------------------------------##
## Set path to input files ##  # Ver1.1.0
'''
Set 'out_obs_post' (path to input csv file)
'''
seaice_dir = '/nird/projects/NS9873K/norcpm/validation/seaicepre/' #ver2.0.0
out_obs_post = seaice_dir + 'obs_post/'
#print(out_obs_post)


## Set path to output files ##   # Ver1.1.0
'''
Set 'out_clm_obs' (path to csv file)
'''
out_clm_obs = seaice_dir + 'clm/obs/' + var + '/' #ver2.0.0

if not os.path.exists(out_clm_obs):
    os.makedirs(out_clm_obs)




##--- Read CSV files ---------------------------------------## #ver2.0.0
seaice_area = pd.DataFrame() # initialize an empty dataframe
for YEAR in yyyy_int_list:
    #print('')
    #print('Year: ', YEAR)
    #print('')


    ## Paths to CSV files ##
    file_obs_post = out_obs_post + 'sea_ice_obs_post_' + str(YEAR) + '.csv'
    df = pd.read_csv(file_obs_post, index_col=0)
    #print(df)

    if YEAR == 2003:
        area = df.loc[:,['DATE', 'AREA_Interpolated']]
        seaice_area = area.rename({'AREA_Interpolated': str(YEAR)}, axis='columns')

    else:
        area = df.loc[:,['AREA_Interpolated']]
        seaice_area2 = area.rename({'AREA_Interpolated': str(YEAR)}, axis='columns')
        seaice_area = pd.concat([seaice_area, seaice_area2], axis=1)

#print(seaice_area)






## Compute climatology 2003-2022 ##
seaice_area['mean'] = seaice_area[['2003', '2004', '2005', '2006', '2007', '2008', '2009', '2010',\
        '2011', '2012', '2013', '2014', '2015', '2016', '2017', '2018', '2019', '2020', '2021', '2022']].mean(axis = 1)
seaice_area['mmdd'] = pd.to_datetime(seaice_area['DATE']).dt.strftime('%m-%d')

#print(seaice_area)





## Insert 'MMDD' column in the first column ##
target_col = 'mmdd'
df_target = seaice_area[target_col]
seaice_area_drp = seaice_area.drop(['DATE', 'mmdd'], axis=1)
seaice_area_drp.insert(0, target_col, df_target)
df = seaice_area_drp.reset_index(drop=True)
#print(df)




##--- Save DataFrame as CSV --------------------------------------##  #ver2.0.0
df.to_csv(out_clm_obs + 'obs_area_clm_all.csv')





##--- Extract climatology for each period -----------------------## #ver2.0.0
df_list = []
df['date'] = pd.to_datetime('2003' + '-' + df['mmdd'].astype(str))
#print(df)



### Jan-Jun ###
df_01_06 = df[df['date'] <= datetime.datetime(2003, 6, 30)]
#print(df_01_06)
df_01_06_drp = df_01_06.drop(['date'], axis=1)
#print(df_01_06_drp)
df_list.append(df_01_06_drp)


### Feb-Jul ###
df_02_07 = df[(df['date'] >= datetime.datetime(2003, 2, 1)) & (df['date'] <= datetime.datetime(2003, 7, 31))]
df_02_07_drp = df_02_07.drop(['date'], axis=1)
#print(df_02_07_drp)
df_02_07_rs = df_02_07_drp.reset_index(drop=True)
df_list.append(df_02_07_rs)


### Mar-Aug ###
df_03_08 = df[(df['date'] >= datetime.datetime(2003, 3, 1)) & (df['date'] <= datetime.datetime(2003, 8, 31))]
df_03_08_drp = df_03_08.drop(['date'], axis=1)
#print(df_03_08_drp)
df_03_08_rs = df_03_08_drp.reset_index(drop=True)
df_list.append(df_03_08_rs)


### Apr-Sep ###
df_04_09 = df[(df['date'] >= datetime.datetime(2003, 4, 1)) & (df['date'] <= datetime.datetime(2003, 9, 30))]
df_04_09_drp = df_04_09.drop(['date'], axis=1)
#print(df_04_09_drp)
df_04_09_rs = df_04_09_drp.reset_index(drop=True)
df_list.append(df_04_09_rs)


### May-Oct ###
df_05_10 = df[(df['date'] >= datetime.datetime(2003, 5, 1)) & (df['date'] <= datetime.datetime(2003, 10, 31))]
df_05_10_drp = df_05_10.drop(['date'], axis=1)
#print(df_05_10_drp)
df_05_10_rs = df_05_10_drp.reset_index(drop=True)
df_list.append(df_05_10_rs)


### Jun-Nov ###
df_06_11 = df[(df['date'] >= datetime.datetime(2003, 6, 1)) & (df['date'] <= datetime.datetime(2003, 11, 30))]
df_06_11_drp = df_06_11.drop(['date'], axis=1)
#print(df_06_11_drp)
df_06_11_rs = df_06_11_drp.reset_index(drop=True)
df_list.append(df_06_11_rs)


### Jul-Dec ###
df_07_12 = df[(df['date'] >= datetime.datetime(2003, 7, 1))]
df_07_12_drp = df_07_12.drop(['date'], axis=1)
#print(df_07_12_drp)
df_07_12_rs = df_07_12_drp.reset_index(drop=True)
df_list.append(df_07_12_rs)


### Aug-Dec + Jan ###
df_01 = df[df['date'] <= datetime.datetime(2003, 1, 31)]
df_08 = df[df['date'] >= datetime.datetime(2003, 8, 1)]
df_08_01 = pd.concat([df_08, df_01], axis=0)
df_08_01_drp = df_08_01.drop(['date'], axis=1)
df_08_01_rs = df_08_01_drp.reset_index(drop=True)
#print(df_08_01_rs)
df_list.append(df_08_01_rs)

### Sep-Dec + Jan-Feb ###
df_02 = df[df['date'] <= datetime.datetime(2003, 2, 28)]
df_09 = df[df['date'] >= datetime.datetime(2003, 9, 1)]
df_09_02 = pd.concat([df_09, df_02], axis=0)
df_09_02_drp = df_09_02.drop(['date'], axis=1)
df_09_02_rs = df_09_02_drp.reset_index(drop=True)
#print(df_09_02_rs)
df_list.append(df_09_02_rs)


### Oct-Dec + Jan-Mar ###
df_03 = df[df['date'] <= datetime.datetime(2003, 3, 31)]
df_10 = df[df['date'] >= datetime.datetime(2003, 10, 1)]
df_10_03 = pd.concat([df_10, df_03], axis=0)
df_10_03_drp = df_10_03.drop(['date'], axis=1)
df_10_03_rs = df_10_03_drp.reset_index(drop=True)
#print(df_10_03_rs)
df_list.append(df_10_03_rs)


### Nov-Dec + Jan-Apr ###
df_04 = df[df['date'] <= datetime.datetime(2003, 4, 30)]
df_11 = df[df['date'] >= datetime.datetime(2003, 11, 1)]
df_11_04 = pd.concat([df_11, df_04], axis=0)
df_11_04_drp = df_11_04.drop(['date'], axis=1)
df_11_04_rs = df_11_04_drp.reset_index(drop=True)
#print(df_11_04_rs)
df_list.append(df_11_04_rs)


### Dec + Jan-May ###
df_05 = df[df['date'] <= datetime.datetime(2003, 5, 31)]
df_12 = df[df['date'] >= datetime.datetime(2003, 12, 1)]
df_12_05 = pd.concat([df_12, df_05], axis=0)
df_12_05_drp = df_12_05.drop(['date'], axis=1)
df_12_05_rs = df_12_05_drp.reset_index(drop=True)
#print(df_12_05_rs)
df_list.append(df_12_05_rs)


#print(df_list)





##--- Save DataFrame as CSV --------------------------------------## #ver2.0.0
mm_name = ['01_06', '02_07', '03_08', '04_09', '05_10', '06_11', '07_12', '08_01', '09_02', '10_03', '11_04', '12_05']
for df_clm, mm in zip(df_list, mm_name):
    df_clm.to_csv(out_clm_obs + 'obs_area_' + mm + '_clm.csv', index=False) #ver2.1.0




print('')
print('Saved climatology')
print('')
print('Complated!!')