smooth_window=cfg['PARAM']['smooth_window'], \
n_runs=cfg['PARAM']['n_runs'], \
skip_steps=cfg['PARAM']['skip_steps'], \
start_date_data=start_date, \
time_step=cfg['PARAM']['time_step'], \
ksteps=cfg['PARAM']['ksteps'], \
latlon_precision=cfg['PARAM']['latlon_precision'])
# Select full water years
start_date_WY, end_date_WY = my_functions.find_full_water_years_within_a_range(\
dict_s_total_runoff[dict_s_total_runoff.keys()[0]].index[0],\
dict_s_total_runoff[dict_s_total_runoff.keys()[0]].index[-1])
for lat_lon in dict_s_total_runoff.keys():
s_total_runoff = dict_s_total_runoff[lat_lon]
dict_s_total_runoff[lat_lon] = my_functions.select_time_range(\
s_total_runoff, start_date_WY, end_date_WY)
##===============================================================#
## 1) Adjust negative runoff to zero;
## 2) Rescale each month so that water is balanced within the month
##===============================================================#
#for lat_lon in dict_s_total_runoff.keys():
# print 'Adjusting negative runoff and rescaling {}...'.format(lat_lon)
# s_total_runoff = dict_s_total_runoff[lat_lon]
# # Original sum of water (with negative runoff)
# sum_mon_orig = s_total_runoff.resample("M", how='sum')
# # Set negative runoff to zero
# s_total_runoff[s_total_runoff<0] = 0
# # Calculate new sum of water (higher than original)
# sum_mon_new = s_total_runoff.resample("M", how='sum')
# # Rescale the new runoff to keep original water balance
# Load VIC output file, if this is an active cell
if fdir[i][j]!=int(NODATA_value): # if active cell
vic_output_file = '{}/{}_{:.{}f}_{:.{}f}'\
.format(cfg['INPUT']['vic_output_dir'], \
cfg['INPUT']['vic_output_file_prefix'], \
lat, cfg['INPUT']['vic_output_precise'], \
lon, cfg['INPUT']['vic_output_precise'])
# Load file
df = my_functions.read_VIC_output(vic_output_file, data_columns=[6,7], \
data_names=['runoff','baseflow'], \
header=True, date_col=3)
# Calculate total runoff
s_total_runoff = df['runoff'] + df['baseflow']
# Select time range interested
s_total_runoff = my_functions.select_time_range(s_total_runoff, \
start_date, \
end_date)
# Add to final dataframe
df_vic_output['row{}_col{}'.format(i+1,j+1)] = s_total_runoff
else: # if inactive cell
df_vic_output['row{}_col{}'.format(i+1,j+1)] = s_inactive
#=====================================================#
# Write runoff field to file
#=====================================================#
df_vic_output.to_csv(cfg['OUTPUT']['basin_runoff_path'], sep=' ', \
header=False, index=False)
# Load and process VIC output data - flow
#====================================================#
print 'Loading and processing VIC output flow data...'
#=== Load data ===#
ds_flow = xray.open_dataset(RVIC_output_nc)
da_flow = ds_flow['streamflow'][:-1,:,:] # Remove the last junk time step
#=== Put data for each grid cell into a df ===#
list_df = []
for i, lat_lon in enumerate(list_flow_lat_lon):
lat = float(lat_lon.split('_')[0])
lon = float(lat_lon.split('_')[1])
df_flow = pd.DataFrame(da_flow.loc[:,lat,lon].values, index=da_flow.coords['time'].values, columns=['flow'])
# Select time range
df_flow = my_functions.select_time_range(df_flow, start_datetime, end_datetime)
# Convert units
df_flow = df_flow * pow(1000.0/25.4/12, 3) # convert m3/s to cfs
# Create df and fill in variables
df = pd.DataFrame(index=df_flow.index)
df['day'] = range(1, len(df)+1) # day number (1,2,3,...)
df['cell'] = list_flow_cell_number[i] # cell number
df['Q_in'] = df_flow['flow'].values # inflow discharge [cfs]
df['Q_out'] = df_flow['flow'].values # outflow discharge [cfs]
df['Q_diff'] = 0.0 # lateral flow [cfs]
df['depth'] = a_d * pow(df_flow['flow'].values, b_d) # flow depth [ft]
df['width'] = a_w * pow(df_flow['flow'].values, b_w) # flow width [ft]
df['u'] = df_flow['flow'].values / df['depth'].values / df['width'] # flow velocoty [ft/s]
list_df.append(df)
#=== Rearrange data ===#
# 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
else: # if at least 5 years
time_locator = ('year', (plot_end_date.year-plot_start_date.year)/5) # time locator on the plot; 'year' for year; 'month' for month. e.g., ('month', 3) for plot one tick every 3 months
# Select data to be plotted
s_usgs_to_plot = my_functions.select_time_range(s_usgs, \
plot_start_date, plot_end_date)
s_TVA_to_plot = my_functions.select_time_range(s_TVA, \
plot_start_date, plot_end_date)
#-------------------------------------------
df_to_plot = pd.DataFrame(s_TVA_to_plot.values, index=s_TVA_to_plot.index, columns=['tva'])
df_to_plot['usgs'] = s_usgs_to_plot
df_to_plot = df_to_plot.dropna()
s_usgs_to_plot = df_to_plot['usgs']
s_TVA_to_plot = df_to_plot['tva']
#-------------------------------------------
#=== Plot flow duration curves (daily data) ===#
fig = my_functions.plot_duration_curve(\
list_s_data=[s_TVA_to_plot, s_usgs_to_plot], \
list_style=['k-', 'b-'], \
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
else: # if at least 5 years
time_locator = (
'year', (plot_end_date.year - plot_start_date.year) / 5
) # time locator on the plot; 'year' for year; 'month' for month. e.g., ('month', 3) for plot one tick every 3 months
#========================================================
# Select data to be plotted
#========================================================
s_rbm_to_plot = my_functions.select_time_range(s_rbm, plot_start_date,
plot_end_date)
s_usgs_to_plot = my_functions.select_time_range(s_usgs, plot_start_date,
plot_end_date)
#========================================================
# plot
#========================================================
#============== plot daily data ===============#
fig = plt.figure()
ax = plt.axes()
plt.plot_date(s_usgs_to_plot.index, s_usgs_to_plot, 'b-', label='USGS gauge')
plt.plot_date(s_rbm_to_plot.index, s_rbm_to_plot, 'r--', label='RBM')
plt.ylabel('Stream T (degC)', fontsize=16)
plt.title('%s, %s' % (usgs_site_name, usgs_site_code), fontsize=16)
plt.legend()
my_functions.plot_date_format(ax,
data_avai_start_date, data_avai_end_date = my_functions.find_data_common_range(s_rbm, s_usgs)
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
else: # if at least 5 years
time_locator = ('year', (plot_end_date.year-plot_start_date.year)/5) # time locator on the plot; 'year' for year; 'month' for month. e.g., ('month', 3) for plot one tick every 3 months
#========================================================
# Select data to be plotted
#========================================================
s_rbm_to_plot = my_functions.select_time_range(s_rbm, plot_start_date, plot_end_date)
s_usgs_to_plot = my_functions.select_time_range(s_usgs, plot_start_date, plot_end_date)
#========================================================
# plot
#========================================================
#============== plot daily data ===============#
fig = plt.figure()
ax = plt.axes()
plt.plot_date(s_usgs_to_plot.index, s_usgs_to_plot, 'b-', label='USGS gauge')
plt.plot_date(s_rbm_to_plot.index, s_rbm_to_plot, 'r--', label='RBM')
plt.ylabel('Stream T (degC)', fontsize=16)
plt.title('%s, %s' %(usgs_site_name, usgs_site_code), fontsize=16)
plt.legend()
my_functions.plot_date_format(ax, time_range=(plot_start_date, plot_end_date), locator=time_locator, time_format='%Y/%m')
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':
column = dict_path[stn][2]
dict_obs[stn] = my_functions.read_USGS_data(filename, [column],
['Discharge'])
elif cfg['INPUT']['obs_format'] == 'Lohmann':
dict_obs[stn] = my_functions.read_Lohmann_route_daily_output(filename)
# Select the same range as Lohmann routed flow
# 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], \
dict_stn_info[stn][1]+360))
f.close()
# Write basin.stn.obs
# 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], \
dict_stn_info[stn][1]+360))
f.close()
# Write basin.stn.obs