def onset_plot(outpath):
list_of_files = glob.glob(outpath + '/*onset_dates.csv')
model_names = [f.split('/')[-1].split('_')[3] for f in list_of_files]
plt.clf()
plt.figure(figsize=(16, 16))
for file in list_of_files:
holder = gent(file, delimiter=',')
fi = list_of_files.index(file)
date_file = glob.glob(outpath + '/*years.csv')
times = gent(date_file[fi], delimiter=',')
#times = np.arange(holder.shape[0])
print len(times), len(holder)
color = get_color(str(fi))
line_style = get_line_style(str(fi))
plt.scatter(times,
holder,
marker=line_style,
c=color,
label=model_names[fi])
plt.ylim(145, 260)
plt.xlim(times[0], times[-1] + 15)
plt.yticks([150, 180, 210, 241], ['1 June', '1 July', '1 Aug', '1 Sep'])
# plt.xticks([0,10,20,30,40,50],['1950','1960','1970','1980','1990','2000'])
plt.legend(loc='best')
plt.savefig(outpath + '/SJ_onsets_all_models_scatter_hitorical.png')
plt.show()
def workshop_contour_example_basic():
# this is only working when the .csv file is in the same folder as your script
#If your script can't find the file, check your path!
mydata = gent(
'/nfs/see-fs-01_teaching/earv052/metrics_workshop/example_code/TRMM_v7_Marteau_onsets_1998_dry_spell_30.csv',
delimiter=',')
print(mydata.shape[:])
ltd = np.linspace(-20, 20, 160)
lat = np.linspace(8.0, 16.0, 32)
fig1 = plt.figure(figsize=(10, 7)) # open figure window
plt.subplot(3, 1, 1) # subplot 1
ax1 = plt.contour(ltd, lat, mydata[:])
plt.subplot(3, 1, 2) # subplot 2
ax2 = plt.contourf(ltd, lat, mydata[:])
plt.subplot(3, 1, 3) # subplot 3
ax3 = plt.pcolor(ltd, lat, mydata[:], vmin=119, vmax=240)
plt.show()
def workshop_contour_example_basemap():
# Read the data, adapt this path to your path if necessary
mydata = gent('TRMM_v7_Marteau_onsets_1998_dry_spell_30.csv',
delimiter=',')
# Shape of the data
print(mydata.shape)
ltd = np.linspace(-20, 20, 160)
lat = np.linspace(8.0, 16.0, 32)
levs = [119, 129, 139, 149, 159, 169, 179, 189, 200, 210, 220, 230, 240]
# Open figure window
ax1 = plt.figure(figsize=(10, 15))
# Add first subplot
plt.subplot(3, 1, 1)
fig1 = plt.contour(ltd, lat, mydata, levs)
plt.clabel(fig1, inline=1) #adds numbers to your contours
m = bm.Basemap(projection='cyl',
llcrnrlat=8.0,
urcrnrlat=16.0,
llcrnrlon=-20.0,
urcrnrlon=20.0,
resolution='c') # coarse resolution for grid
m.drawcoastlines(linewidth=2)
m.drawcountries(linewidth=1)
plt.text(-21.5, 15, '(a)')
# Add second subplot
plt.subplot(3, 1, 2)
fig2 = plt.contourf(ltd, lat, mydata[:], levs)
m = bm.Basemap(projection='cyl',
llcrnrlat=8.0,
urcrnrlat=16.0,
llcrnrlon=-20.0,
urcrnrlon=20.0,
resolution='l') # medium resolution for grid
m.drawcoastlines(linewidth=2)
m.drawcountries(linewidth=1)
plt.text(-21.5, 15, '(b)')
plt2_ax = ax1.gca()
left0, bottom0, width0, height0 = plt2_ax.get_position().bounds
cbar_1_ax = ax1.add_axes([left0, bottom0 - 0.01, width0, 0.015])
cb2 = ax1.colorbar(fig2, cax=cbar_1_ax, orientation='horizontal')
cb2.set_label('Marteau local onset 1998')
# Add third subplot
plt.subplot(3, 1, 3)
my_pallet = cm.spectral
my_pallet.set_over('w')
my_pallet.set_under('w')
my_pallet.set_bad(alpha=0.0)
fig3 = plt.pcolor(ltd, lat, mydata[:], vmin=119, vmax=240, cmap=my_pallet)
m = bm.Basemap(projection='cyl',
llcrnrlat=8.0,
urcrnrlat=16.0,
llcrnrlon=-20.0,
urcrnrlon=20.0,
resolution='f') # fine resolution for grid
m.drawcoastlines(linewidth=2)
m.drawcountries(linewidth=1)
plt.text(-21.5, 15, '(c)')
plt3_ax = ax1.gca()
left0, bottom0, width0, height0 = plt3_ax.get_position().bounds
cbar_3_ax = ax1.add_axes([left0, bottom0 - 0.01, width0, 0.015])
cb3 = ax1.colorbar(fig3, cax=cbar_3_ax, orientation='horizontal')
cb3.set_label('Marteau local onset 1998')
plt.subplots_adjust(hspace=0.3)
# plt.savefig('matplot_contour_with_basemap.png') # save figure
plt.show()
def TRMM_CDD(yyyy, total):
import numpy as np
import matplotlib.pyplot as plt
from numpy import genfromtxt as gent
from netCDF4 import Dataset
import mpl_toolkits.basemap as bm
# first bring in the data
holdr = gent('csvs/TRMM_v7_Lele_Lamb_onsets_' + str(yyyy) + '.csv',
delimiter=',')
my_data = holdr[12:, 10:34]
# now we also want trmm data
nc = Dataset('C:/Users/js08rgjf/Canopy/scripts/csvs/v07/' + str(yyyy) +
'/3B42_daily.' + str(yyyy) + '.05.01.7.nc')
trmm = nc.variables['r']
left = trmm[:, 232:264, -80:]
right = trmm[:, 232:264, :80]
holdr = np.concatenate((left, right), axis=2)
print holdr.shape
# Ok now that we have the first of May, we do the rest of may
dys = [
'01', '02', '03', '04', '05', '06', '07', '08', '09', '10', '11', '12',
'13', '14', '15', '16', '17', '18', '19', '20', '21', '22', '23', '24',
'25', '26', '27', '28', '29', '30', '31'
]
# First do May
for t in range(1, 31):
nc = Dataset('C:/Users/js08rgjf/Canopy/scripts/csvs/v07/' + str(yyyy) +
'/3B42_daily.' + str(yyyy) + '.05.' + str(dys[t]) +
'.7.nc')
trmm = nc.variables['r']
left = trmm[:, 232:264, -80:]
right = trmm[:, 232:264, :80]
holdr_temp = np.concatenate((left, right), axis=2)
holdr = np.concatenate((holdr, holdr_temp), axis=0)
print holdr.shape
# Now do June
for t in range(0, 30):
nc = Dataset('C:/Users/js08rgjf/Canopy/scripts/csvs/v07/' + str(yyyy) +
'/3B42_daily.' + str(yyyy) + '.06.' + str(dys[t]) +
'.7.nc')
trmm = nc.variables['r']
left = trmm[:, 232:264, -80:]
right = trmm[:, 232:264, :80]
holdr_temp = np.concatenate((left, right), axis=2)
holdr = np.concatenate((holdr, holdr_temp), axis=0)
print holdr.shape
# Now do July
for t in range(0, 31):
nc = Dataset('C:/Users/js08rgjf/Canopy/scripts/csvs/v07/' + str(yyyy) +
'/3B42_daily.' + str(yyyy) + '.07.' + str(dys[t]) +
'.7.nc')
trmm = nc.variables['r']
left = trmm[:, 232:264, -80:]
right = trmm[:, 232:264, :80]
holdr_temp = np.concatenate((left, right), axis=2)
holdr = np.concatenate((holdr, holdr_temp), axis=0)
print holdr.shape
# Finally August
for t in range(0, 31):
nc = Dataset('C:/Users/js08rgjf/Canopy/scripts/csvs/v07/' + str(yyyy) +
'/3B42_daily.' + str(yyyy) + '.08.' + str(dys[t]) +
'.7.nc')
trmm = nc.variables['r']
left = trmm[:, 232:264, -80:]
right = trmm[:, 232:264, :80]
holdr_temp = np.concatenate((left, right), axis=2)
holdr = np.concatenate((holdr, holdr_temp), axis=0)
print holdr.shape
# And September
for t in range(0, 30):
nc = Dataset('C:/Users/js08rgjf/Canopy/scripts/csvs/v07/' + str(yyyy) +
'/3B42_daily.' + str(yyyy) + '.09.' + str(dys[t]) +
'.7.nc')
trmm = nc.variables['r']
left = trmm[:, 232:264, -80:]
right = trmm[:, 232:264, :80]
holdr_temp = np.concatenate((left, right), axis=2)
holdr = np.concatenate((holdr, holdr_temp), axis=0)
print holdr.shape
my_trmm = holdr[:, 12:, 10:34]
dtes = []
duration = []
pltdtes = np.zeros((my_trmm.shape[1], my_trmm.shape[2]), dtype=int)
pltdur = np.zeros((my_trmm.shape[1], my_trmm.shape[2]), dtype=int)
for x in range(0, my_trmm.shape[2]):
for y in range(0, my_trmm.shape[1]):
current_max = 0
current_dte = 0
dte = 0
duratn = 0
if my_data[y, x] - 119 > my_trmm.shape[0] - 10 or my_data[
y, x] >= 250:
continue
else:
for t in range(
int(my_data[y, x]) - 119, my_trmm.shape[0] - 10):
if my_trmm[t, y, x] <= float(total):
dte = t
print dte
for t1 in xrange(t, my_trmm.shape[0] - 10):
if my_trmm[t1, y, x] <= float(total):
continue
else:
duratn = t1 - t
if duratn > current_max:
current_dte = dte + 119
current_max = duratn
break
dtes.extend([current_dte])
duration.extend([current_max])
pltdtes[y, x] = current_dte
pltdur[y, x] = current_max
# else:
# continue
# break
print np.max(duration[:])
print np.max(dtes[:])
# plt.clf()
# plt.scatter(duration[:],dtes[:])
# plt.show()
plt.clf()
levs1 = [119, 129, 139, 150, 160, 170, 180, 190, 200, 211, 221, 231, 242]
dates = [
'1/5', '11/5', '21/5', '1/6', '11/6', '21/6', '1/7', '11/7', '21/7',
'1/8', '11/8', '21/8', '1/9'
]
levs2 = [2, 4, 6, 8, 10, 12, 15, 18, 20]
ltd = np.linspace(-17.5, -11.5, 24)
lati = np.linspace(11, 16, 20)
ax1 = plt.figure(figsize=(12, 12))
plt.subplot(2, 1, 1)
m = bm.Basemap(projection='cyl',
llcrnrlat=11,
urcrnrlat=16,
llcrnrlon=-17.5,
urcrnrlon=-11.5,
resolution='c')
fig2 = plt.contourf(ltd, lati, pltdtes[:], levs1, extend='max')
m.drawcoastlines(linewidth=2)
m.drawcountries(linewidth=1)
plt2_ax = ax1.gca()
left0, bottom0, width0, height0 = plt2_ax.get_position().bounds
cbar_1_ax = ax1.add_axes([left0, bottom0 - 0.05, width0, 0.015])
cb2 = ax1.colorbar(fig2, cax=cbar_1_ax, orientation='horizontal')
cb2.set_label('Date of longest dry spell (after onset)')
cb2.set_ticks(levs1)
cb2.set_ticklabels(dates)
plt.subplot(2, 1, 2)
m = bm.Basemap(projection='cyl',
llcrnrlat=11,
urcrnrlat=16,
llcrnrlon=-17.5,
urcrnrlon=-11.5,
resolution='c')
fig2 = plt.contourf(ltd, lati, pltdur[:], levs2, extend='max')
m.drawcoastlines(linewidth=2)
m.drawcountries(linewidth=1)
plt2_ax = ax1.gca()
left0, bottom0, width0, height0 = plt2_ax.get_position().bounds
cbar_1_ax = ax1.add_axes([left0, bottom0 - 0.05, width0, 0.015])
cb2 = ax1.colorbar(fig2, cax=cbar_1_ax, orientation='horizontal')
cb2.set_label('Duration of first dry spell')
plt.subplots_adjust(hspace=0.5)
# plt.savefig('TRMM_CDD_'+str(yyyy)+'_limit'+str(total)+'mm_per_day.png')
plt.show()
np.savetxt('CDD_date_' + str(yyyy) + '.csv', pltdtes[:], delimiter=',')
np.savetxt('CDD_duration_' + str(yyyy) + '.csv', pltdur[:], delimiter=',')