def find_corners(temps, press=1.e3, skew=30.):
"""
convert a pair of temperatures into skew plotting coordinates at a constant pressure level
Parameters
----------
temps : [float]
xaxis corner temperatures (degC)
press : float
pressure level for skew conversion
skew : float
coefficient to use in the tempToSkew conversion
Returns
-------
corners : [float]
x axis corners in plotting coordinates
"""
corners = convertTempToSkew(temps, press, skew)
return list(corners)
def makeSkewDry(ax, skew=30):
"""
draw a skew-T lnP diagram on an axis
Parameters
----------
ax : matplotlib.axes
matplotlib figure axis
skew : float
adjustable coefficient to make isotherms slope
compared to adiabats
Returns
-------
ax : matplotlib.axes
the modified figure axis
skew : float
skew coefficient (K)
"""
# majorFormatter = ticks.FormatStrFormatter('%d')
# ax.yaxis.set_major_formatter(majorFormatter)
yplot = range(1000, 190, -10)
xplot = range(-300, -139)
pvals = np.size(yplot)
tvals = np.size(xplot)
temp = np.zeros([pvals, tvals])
theTheta = np.zeros([pvals, tvals])
# lay down a reference grid that labels xplot,yplot points
# in the new (skewT-lnP) coordinate system .
# Each value of the temp matrix holds the actual (data)
# temperature label (in deg C) of the xplot, yplot coordinate.
# pairs. The transformation is given by W&H 3.56, p. 78. Note
# that there is a sign difference, because rather than
# taking y= -log(P) like W&H, I take y= +log(P) and
# then reverse the y axis
for i in yplot:
for j in xplot:
# Note that we don't have to transform the y
# coordinate, as it is still pressure.
iInd = yplot.index(i)
jInd = xplot.index(j)
temp[iInd, jInd] = convertSkewToTemp(j, i, skew)
Tk = c.Tc + temp[iInd, jInd]
pressPa = i * 100.
theTheta[iInd, jInd] = find_theta(Tk, pressPa)
#
# Contour the temperature matrix.
#
# First, make sure that all plotted lines are solid.
mpl.rcParams['contour.negative_linestyle'] = 'solid'
tempLabels = range(-40, 50, 10)
ax.set_yscale('log')
majorFormatter = ticks.FormatStrFormatter('%d')
ax.yaxis.set_major_formatter(majorFormatter)
ax.yaxis.set_minor_formatter(majorFormatter)
plt.setp(ax.get_xticklabels(), weight='bold')
plt.setp(ax.get_yticklabels(), weight='bold')
plt.setp(ax.get_yticklabels(minor=True), weight='bold')
tempLevs = ax.contour(xplot, yplot, temp, tempLabels, \
colors='k')
#
# Customize the plot
#
thetaLabels = list(range(200, 390, 10))
thetaLevs = ax.contour(xplot, yplot, theTheta, thetaLabels, \
colors='b')
# Transform the temperature,dewpoint from data coords to
# plotting coords.
ax.set_title('skew T - lnp chart')
ax.set_ylabel('pressure (hPa)')
ax.set_xlabel('temperature (deg C)')
#
# Crop image to a more usable size
#
TempTickLabels = range(-15, 40, 5)
TempTickCoords = TempTickLabels
skewTickCoords = convertTempToSkew(TempTickCoords, 1.e3, skew)
ax.set_xticks(skewTickCoords)
ax.set_xticklabels(TempTickLabels)
skewLimits = convertTempToSkew([-15, 35], 1.e3, skew)
ax.axis([skewLimits[0], skewLimits[1], 300, 1.e3])
#
# Create line labels
#
fntsz = 9 # Handle for 'fontsize' of the line label.
ovrlp = True # Handle for 'inline'. Any integer other than 0
# creates a white space around the label.
tempLevs.clabel(inline=ovrlp,
inline_spacing=0,
fmt='%2d',
fontsize=fntsz,
use_clabeltext=True)
thetaLevs.clabel(inline=ovrlp,
inline_spacing=0,
fmt='%5d',
fontsize=fntsz,
use_clabeltext=True)
ax.invert_yaxis()
ax.yaxis.grid(True, which='minor')
ax.figure.canvas.draw()
return ax, skew
override the default rs labels with a tighter mesh
"""
from numpy import arange
#
# get the default labels
#
tempLabels, rsLabels, thetaLabels, thetaeLabels = make_default_labels()
#
# change the temperature and rs grids
#
tempLabels = range(-40, 50, 2)
rsLabels = [0.1, 0.25, 0.5, 1, 2, 3] + list(np.arange(4, 28, 2))
return tempLabels, rsLabels, thetaLabels, thetaeLabels
skew = 35.
temp = the_sounding['temp']
press = the_sounding['pres']
tdew = the_sounding['dwpt']
temp_skew = convertTempToSkew(temp, press, skew)
tdew_skew = convertTempToSkew(tdew, press, skew)
fig, ax = plt.subplots(1, 1, figsize=(10, 8))
corners = [10, 35]
ax, skew = makeSkewWet(ax, corners=corners, skew=skew, label_fun=label_fun)
ax.set_title('Dodge City Kansas sounding')
xcorners = find_corners(corners, skew=skew)
ax.set(xlim=xcorners, ylim=[1000, 400])
ax.plot(temp_skew, press, linewidth=5)
ax.plot(tdew_skew, press, linewidth=5)
def makeSkewWet(ax, corners=None, skew=30, label_fun=None):
"""
draw a skew-T lnP diagram on an axis
Parameters
----------
ax : matplotlib.axes
matplotlib figure axis
corners : [float]
x axis temperature limits (degC)
skew : float
adjustable coefficient to make isotherms slope
compared to adiabats
label_fun: function
optional function to generate contour intervals, see make_default_labels
Returns
-------
ax : matplotlib.axes
the modified figure axis
"""
#
# turn off tick labels until the las plot
#
if corners is None:
corners = [-25, 25]
if label_fun is None:
label_fun = make_default_labels
tempLabels, rsLabels, thetaLabels, thetaeLabels = label_fun()
ax.yaxis.set_major_formatter(ticks.NullFormatter())
ax.xaxis.set_major_formatter(ticks.NullFormatter())
ax.yaxis.set_minor_formatter(ticks.NullFormatter())
yplot = range(1000, 190, -6) #
xcorners = find_corners(corners, skew=skew)
xplot = list(np.linspace(xcorners[0], xcorners[1], 45))
pvals = np.size(yplot)
tvals = np.size(xplot)
temp = np.zeros([pvals, tvals])
theTheta = np.zeros_like(temp)
the_rsat = np.zeros_like(temp)
theThetae = np.zeros([pvals, tvals])
# lay down a reference grid that labels xplot,yplot points
# in the new (skewT-lnP) coordinate system .
# Each value of the temp matrix holds the actual (data)
# temperature label (in deg C) of the xplot, yplot coordinate.
# pairs. The transformation is given by W&H 3.56, p. 78. Note
# that there is a sign difference, because rather than
# taking y= -log(P) like W&H, I take y= +log(P) and
# then reverse the y axis
for presshPa in yplot: #loop over pressures
for skewed in xplot: #loop over skewed xcoords
# Note that we don't have to transform the y
# coordinate, as it is still pressure.
iInd = yplot.index(presshPa)
jInd = xplot.index(skewed)
temp[iInd, jInd] = convertSkewToTemp(skewed, presshPa, skew)
Tk = c.Tc + temp[iInd, jInd]
pressPa = presshPa * 100.
theTheta[iInd, jInd] = find_theta(Tk, pressPa)
rs = find_rsat(Tk, pressPa)
the_rsat[iInd, jInd] = rs
theThetae[iInd, jInd] = find_thetaet(Tk, rs, Tk, pressPa)
#
# Contour the temperature matrix.
#
# First, make sure that all plotted lines are solid.
mpl.rcParams['contour.negative_linestyle'] = 'solid'
tempLabels=ax.contour(xplot, yplot, temp, tempLabels, \
colors='k')
tempLabels = tempLabels.levels
#
# contour theta
#
#thetaLabels = list(range(200, 390, 10))
thetaLevs = ax.contour(xplot, yplot, theTheta, thetaLabels, \
colors='b')
thetaLabels = thetaLevs.levels
#
# contour rsat
#
#rsLabels = [0.1, 0.25, 0.5, 1, 2, 3] + list(range(4, 28, 2)) #+ [26, 28]
rsLevs = ax.contour(xplot,
yplot,
the_rsat * 1.e3,
levels=rsLabels,
colors='g',
linewidths=.5)
rsLabels = rsLevs.levels
#thetaeLabels = np.arange(250, 410, 10)
thetaeLevs = ax.contour(xplot, yplot, theThetae, thetaeLabels, \
colors='r')
thetaeLabels = thetaeLevs.levels
#
# Customize the plot
#
# plt.setp(ax.get_xticklabels(), weight='bold')
# plt.setp(ax.get_yticklabels(), weight='bold')
#
# Crop image to a more usable size
#
#
# Create line labels
#
fntsz = 9 # Handle for 'fontsize' of the line label.
ovrlp = True # Handle for 'inline'. Any integer other than 0
# creates a white space around the label.
#tempLevs.clabel(inline=ovrlp, inline_spacing=0,fmt='%2d', fontsize=fntsz,use_clabeltext=True)
thetaLevs.clabel(inline=ovrlp,
inline_spacing=0,
fmt='%3d',
fontsize=fntsz,
use_clabeltext=True)
rsLevs.clabel(inline=ovrlp,
inline_spacing=0,
fmt='%3.2g',
fontsize=fntsz,
use_clabeltext=True)
thetaeLevs.clabel(thetaeLabels,
inline_spacing=0,
inline=ovrlp,
fmt='%3d',
fontsize=fntsz,
use_clabeltext=True)
ax.invert_yaxis()
ax.set_yscale('log')
ax.yaxis.set_major_formatter(ticks.NullFormatter())
ax.xaxis.set_major_formatter(ticks.NullFormatter())
ax.yaxis.set_minor_formatter(ticks.NullFormatter())
#ax.figure.canvas.draw()
# #ax.figure.canvas.draw()
# xcorners = find_corners(corners, skew=skew)
majorFormatter = ticks.FormatStrFormatter('%d')
ax.yaxis.set_major_formatter(majorFormatter)
ax.yaxis.set_minor_formatter(majorFormatter)
# locs = np.array(range(100, 1100, 100))
# labels = locs
# ax.set_yticks(locs)
# ax.set_yticklabels(labels) # Conventionally labels semilog graph.
# ax.set_ybound((200, 1000))
TempTickLabels = range(-5, 40, 5)
TempTickCoords = TempTickLabels
skewTickCoords = convertTempToSkew(TempTickCoords, 1.e3, skew)
# skewLimits = convertTempToSkew([-15, 35], 1.e3, skew)
# ax.axis([skewLimits[0], skewLimits[1], 300, 1.e3])
ax.set(ylim=(1000, 300), xlim=xcorners)
ax.set_xticks(skewTickCoords)
ax.set_xticklabels(TempTickLabels)
ax.yaxis.grid(True, which='both')
# ax.set_title('skew T - lnp chart')
# ax.set_ylabel('pressure (hPa)')
# ax.set_xlabel('temperature (deg C)')
ax.set_title('debug title')
ax.figure.canvas.draw()
return ax, skew
units_dict[var] = units[count]
#
# use the pretty printer to print the dictionary
#
print(f'units: {pformat(units_dict)}')
# ## Convert temperature and dewpoint to skew coords
# In[8]:
skew = 35.
triplets = zip(sounding['temp'], sounding['dwpt'], sounding['pres'])
xcoord_T = []
xcoord_Td = []
for a_temp, a_dew, a_pres in triplets:
xcoord_T.append(convertTempToSkew(a_temp, a_pres, skew))
xcoord_Td.append(convertTempToSkew(a_dew, a_pres, skew))
# ## Plot the sounding, making the sounding lines thicker
# In[9]:
def label_fun():
"""
override the default rs labels with a tighter mesh
"""
from numpy import arange
#
# get the default labels
#