usgs_col = df_usgs_info.ix[i,'flow_col']
dam_number = df_usgs_info.ix[i,'corresponding_dam_number']
dam_name = df_usgs_info.ix[i,'corresponding_dam_name']
print 'Plotting dam {}...'.format(dam_number)
#=== Get USGS data ===#
df_usgs = my_functions.read_USGS_data(\
os.path.join(usgs_data_dir, '{}.txt'.format(usgs_code)), \
columns=[usgs_col], names=['flow']) / 1000 # convert to thousand cfs
s_usgs = df_usgs.ix[:,0] # convert df to Series
#=== Get TVA data ===#
TVA_path = os.path.join(TVA_daily_dir, '{}_{}.daily.1903_2013'.format(lat, lon))
if os.path.isfile(TVA_path)==False: # if corresponding dam has no data
continue
s_TVA = my_functions.read_Lohmann_route_daily_output(\
os.path.join(TVA_daily_dir, '{}_{}.daily.1903_2013'.format(lat, lon)))
s_TVA = s_TVA / 1000.0 # convert to thousand cfs
#=== Extract data within common range ===#
# determine the common range of available data of both data sets
data_avai_start_date, data_avai_end_date = my_functions.\
find_data_common_range([s_usgs, s_TVA])
if (data_avai_start_date-data_avai_end_date).days>=0: # if no common time range
print "No common range data available!"
exit()
# find the full water years with available data for both data sets
plot_start_date, plot_end_date = my_functions.\
find_full_water_years_within_a_range(data_avai_start_date, data_avai_end_date)
# determine time locator #
if plot_end_date.year-plot_start_date.year < 5: # if less than 5 years
time_locator = ('year', 1) # time locator on the plot; 'year' for year; 'month' for month. e.g., ('month', 3) for plot one tick every 3 months
# Load data
#========================================================
list_s = [] # list of Series to be plotted
list_plot_style = [] # list of plot style for each time series
list_plot_label = [] # list of plot legend label for each time series
for i in range(cfg['INPUT_CONTROL']['n_ts']): # Load all time series data
input_section = 'INPUT_{}'.format(i+1)
# If from formatted RBM output
if cfg[input_section]['ts_format']=='RBM_formatted':
s = my_functions.read_RMB_formatted_output(cfg[input_section]['ts_path'], \
var='flow') / 1000 # convert to thousand cfs
# If from formatted RBM output
elif cfg[input_section]['ts_format']=='Lohmann':
s = my_functions.read_Lohmann_route_daily_output(cfg[input_section]['ts_path'])\
/ 1000 # convert to thousand cfs
# If USGS data
elif cfg[input_section]['ts_format']=='USGS':
if type(cfg[input_section]['usgs_col']) is int: # if only one needed data column
df_usgs = my_functions.read_USGS_data(cfg[input_section]['ts_path'], \
columns=[cfg[input_section]['usgs_col']], \
names=['flow']) / 1000 # convert to thousand cfs
s= df_usgs.ix[:,0] # convert df to Series
else: # if more than one data column needed, take average
usgs_flow_col_split = cfg[input_section]['usgs_col'].split('&')
names=[]
for i in range(len(usgs_flow_col_split)):
usgs_flow_col_split[i] = int(usgs_flow_col_split[i])
dict_path[line_split[0]] = [
line_split[1], line_split[2],
int(line_split[3])
]
elif cfg['INPUT']['obs_format'] == 'Lohmann':
dict_path[line_split[0]] = [line_split[1], line_split[2]]
else:
print 'Error: unsupported observation data format!'
exit()
f.close()
# Read in routed streamflow from inverted runoff
dict_Lohmann_routed = {} # {station_name: pd.Series of daily data} [unit: cfs]
for stn in dict_path:
# Load data
s_Lohmann_routed = my_functions.read_Lohmann_route_daily_output(
dict_path[stn][1])
dict_Lohmann_routed[stn] = s_Lohmann_routed
# Select full water years
start_date_WY, end_date_WY = my_functions.find_full_water_years_within_a_range(\
dict_Lohmann_routed[stn].index[0], \
dict_Lohmann_routed[stn].index[-1])
dict_Lohmann_routed[stn] = my_functions.select_time_range(dict_Lohmann_routed[stn], \
start_date_WY, \
end_date_WY)
# Read in original station obs rmat
dict_obs = {} # {station_name: pd.Series of daily data} [unit: cfs]
for stn in dict_path:
# Load data
filename = dict_path[stn][0]
if cfg['INPUT']['obs_format'] == 'USGS':
float(line.split()[2])]
#======================================================#
# Load data
#======================================================#
# Load data and select time range needed
dict_df_stn = {} # a dictionary of station data
# {station_code: df}
for stn in list_stn: # for each gauge station, load data
# Load data
filename = '{}/{}'.format(cfg['INPUT']['stn_data_dir'], stn)
if cfg['INPUT']['data_formst']=='USGS':
column = dict_stn_info[stn][2]
dict_df_stn[stn] = my_functions.read_USGS_data(filename, [column], ['Discharge'])
elif cfg['INPUT']['data_formst']=='Lohmann':
dict_df_stn[stn] = my_functions.read_Lohmann_route_daily_output(filename)
# Select time range needed
dict_df_stn[stn] = my_functions.select_time_range(dict_df_stn[stn], \
start_date, end_date)
# Convert data to cfs
if cfg['PARAM']['input_flow_unit']=='cfs':
pass
#======================================================#
# Write basin.stn.list and basin.stn.obs
#======================================================#
# Write basin.stn.list
f = open(cfg['OUTPUT']['basin_stn_list_path'], 'w')
for stn in list_stn:
f.write('1 {} {} {} 1000 1000 1000\n'.format(stn, dict_stn_info[stn][0], \
#======================================================#
# Load data
#======================================================#
# Load data and select time range needed
dict_df_stn = {} # a dictionary of station data
# {station_code: df}
for stn in list_stn: # for each gauge station, load data
# Load data
filename = '{}/{}'.format(cfg['INPUT']['stn_data_dir'], stn)
if cfg['INPUT']['data_formst'] == 'USGS':
column = dict_stn_info[stn][2]
dict_df_stn[stn] = my_functions.read_USGS_data(filename, [column],
['Discharge'])
elif cfg['INPUT']['data_formst'] == 'Lohmann':
dict_df_stn[stn] = my_functions.read_Lohmann_route_daily_output(
filename)
# Select time range needed
dict_df_stn[stn] = my_functions.select_time_range(dict_df_stn[stn], \
start_date, end_date)
# Convert data to cfs
if cfg['PARAM']['input_flow_unit'] == 'cfs':
pass
#======================================================#
# Write basin.stn.list and basin.stn.obs
#======================================================#
# Write basin.stn.list
f = open(cfg['OUTPUT']['basin_stn_list_path'], 'w')
for stn in list_stn:
f.write('1 {} {} {} 1000 1000 1000\n'.format(stn, dict_stn_info[stn][0], \
# Load data
#========================================================
list_s = [] # list of Series to be plotted
list_plot_style = [] # list of plot style for each time series
list_plot_label = [] # list of plot legend label for each time series
for i in range(cfg['INPUT_CONTROL']['n_ts']): # Load all time series data
input_section = 'INPUT_{}'.format(i + 1)
# If from formatted RBM output
if cfg[input_section]['ts_format'] == 'RBM_formatted':
s = my_functions.read_RMB_formatted_output(cfg[input_section]['ts_path'], \
var='flow') / 1000 # convert to thousand cfs
# If from formatted RBM output
elif cfg[input_section]['ts_format'] == 'Lohmann':
s = my_functions.read_Lohmann_route_daily_output(cfg[input_section]['ts_path'])\
/ 1000 # convert to thousand cfs
# If USGS data
elif cfg[input_section]['ts_format'] == 'USGS':
if type(cfg[input_section]
['usgs_col']) is int: # if only one needed data column
df_usgs = my_functions.read_USGS_data(cfg[input_section]['ts_path'], \
columns=[cfg[input_section]['usgs_col']], \
names=['flow']) / 1000 # convert to thousand cfs
s = df_usgs.ix[:, 0] # convert df to Series
else: # if more than one data column needed, take average
usgs_flow_col_split = cfg[input_section]['usgs_col'].split('&')
names = []
for i in range(len(usgs_flow_col_split)):
