'''

Input: Temperature (Celsius)

Goff-Gratch solution to Clausius-Clapeyron for liquid water

Returns es (hPa)

'''

T = T_C + 273.15

log_es = -7.90298 * ((373.15 / T) - 1.) + 5.02808 * np.log10(373.15/T)\

- 1.3816 * (10 ** -7) * (10**(11.344 * (1. - T / 373.15)) - 1.)\

+ 8.1328 * (10 ** -3) * (10**(-3.49149 * ((373.15/T) - 1.)) - 1.)\

+ np.log10(1013.25)

'''

Input: T (Celsius)

Goff-Gratch solution to Clausius-Clapeyron for ice

Returns ei (hPa)

'''

T = T_C + 273.15

log_ei = -9.09718 * ((273.15 / T) - 1.) - 3.56654 * np.log10(273.15 / T)\

+ 0.876793 * (1. - (T / 273.15)) + np.log10(6.11)

'''

Input: filepath of coptersonde csv

Returns numpy array of coptersonde csv data a la csvread in MATLAB

Headers = Time, Lat, Long, Altitude, Abs Pressure, Roll, Pitch, Yaw,

Humidity_1, Humidity_2, Humidity_3, Humidity_4, Temp_1, Temp_2, Temp_3,

Temp_4, Gnd Speed, Vert Speed, AccelX, AccelY

'''

f = open(coptersondefilename)

reader = csv.DictReader(f)

# Initialize

time_, lat_, lon_, alt_, p_, roll_, pitch_, yaw_, RH1_, RH2_, RH3_, RH4_,\

RHT1_, RHT2_, RHT3_, RHT4_, T1_, T2_, T3_, T4_, gspd_, vspd_, ax_, ay_ \

= ([] for i in range(24))

# Assign

for line in reader:

time_.append(float(line['Time']))

lat_.append(float(line['Lat']))

lon_.append(float(line['Long']))

alt_.append(float(line['Altitude']))

p_.append(float(line['Abs Pressure']))

roll_.append(float(line['Roll']))

pitch_.append(float(line['Pitch']))

yaw_.append(float(line['Yaw']))

RH1_.append(float(line['Humidity_1']))

RH2_.append(float(line['Humidity_2']))

RH3_.append(float(line['Humidity_3']))

RH4_.append(float(line['Humidity_4']))

RHT1_.append(float(line['RH_1 Temp']))

RHT2_.append(float(line['RH_2 Temp']))

RHT3_.append(float(line['RH_3 Temp']))

RHT4_.append(float(line['RH_4 Temp']))

T1_.append(float(line['Temp_1']))

T2_.append(float(line['Temp_2']))

T3_.append(float(line['Temp_3']))

T4_.append(float(line['Temp_4']))

gspd_.append(float(line['Gnd Speed']))

vspd_.append(float(line['Vert Speed']))

ax_.append(float(line['AccelX']))

ay_.append(float(line['AccelY']))

f.close()

return np.array([time_, lat_, lon_, alt_, p_, roll_, pitch_, yaw_,

RH1_, RH2_, RH3_, RH4_, RHT1_, RHT2_, RHT3_, RHT4_, T1_, T2_, T3_, T4_,

'''

Input: filepath to .pos ppk file

Output: numpy array of date_dgps, time_dgps, lat_dgps, long_dgps, alt_dgps,

alt_dgps_agl

Be sure to set date_dgps and time_dgps to list

Convert rest with .astype(np.float)

'''

fd = open(ppkfilename, 'r')

for i in range(25):

fd.next()

reader = csv.DictReader(fd, delimiter=' ')

reader.fieldnames = ['Date', 'Time', '', '', 'Lat', '', 'Lon', '', 'Alt2',

'Alt', '', '', 'Q', '', '', 'ns', '', '', 'sdn', '', '', 'sde', '', '',

'sdu', '', '', 'sdne', '', '', 'sdeu', '', '', 'sdun', '', 'age', '', '',

'', 'ratio']

date_dgps_ = []

time_dgps_ = []

lat_dgps_ = []

lon_dgps_ = []

alt_dgps_ = []

for line in reader:

date_dgps_.append(line['Date'])

time_dgps_.append(line['Time'])

lat_dgps_.append(float(line['Lat']))

lon_dgps_.append(float(line['Lon']))

# Altitude has different spaces once reaching 1000 ft

if not line['Alt']:

alt_dgps_.append(float(line['Alt2']))

else:

alt_dgps_.append(float(line['Alt']))

fd.close()

alt_dgps_agl_ = [a - alt_dgps_[0] for a in alt_dgps_]

return np.array([date_dgps_, time_dgps_, lat_dgps_, lon_dgps_, alt_dgps_,

'''

Input: filepath of post-processed coptersonde csv

Returns numpy array of coptersonde csv data a la csvread in MATLAB

Headers = Lat, Lon, AltAGL(m), p(hPa), T(C), Td(C), RH(percent), w(gKg-1),

Theta(K), Speed(ms-1), Dir(deg)

'''

f = open(coptercsv)

reader = csv.DictReader(f)

# Initialize

lat_, lon_, alt_, p_, T_, Td_, RH_, w_, theta_, speed_, dir_ \

= ([] for i in range(11))

# Assign

for line in reader:

lat_.append(float(line['Lat']))

lon_.append(float(line['Lon']))

alt_.append(float(line['AltAGL(m)']))

p_.append(float(line['p(hPa)']))

T_.append(float(line['T(C)']))

Td_.append(float(line['Td(C)']))

RH_.append(float(line['RH(percent)']))

w_.append(float(line['w(gKg-1)']))

theta_.append(float(line['Theta(K)']))

speed_.append(float(line['Speed(ms-1)']))

dir_.append(float(line['Dir(deg)']))

f.close()

return np.array([lat_, lon_, alt_, p_, T_, Td_, RH_, w_,

'''

Input data directory where raw csv files are saved

Returns filepath of most recently created file

'''

fnameArr = glob(os.path.join(dirname, '*.csv'))

csvCreatedArr = []

for s in fnameArr:

csvCreatedArr.append(float(os.stat(s).st_ctime))

'''

Input data directory for project

Returns filepath of most recently created directory

'''

fnameArr = glob(os.path.join(dirname, '*/*/'))

dirCreatedArr = []

for d in fnameArr:

dirCreatedArr.append(float(os.stat(d).st_ctime))

'''

Input csv filepath, will read last datetime of csv and determine closest

.pos file based on filename

'''

# read times from CSV

dat = csvread_raw(csvfilename)

timeCreateCSV = dat[0][-1]

# Time created for all .pos files, based on file name

dataDirName = csvfilename.rsplit('\\', 1)[0]

fnameArr = glob(os.path.join(dataDirName, "*.pos")) # string array

posCreatedArr = []

base_t = datetime(1970, 1, 1)

UTC_offset = timedelta(hours=5)

for s in fnameArr:

idatetime = datetime.strptime(s.split('\\')[-1][9:29],

'%Y-%m-%d_%Hh%Mm%Ss') + UTC_offset

posCreatedArr.append((idatetime - base_t).total_seconds())

posCreatedArr = np.array(posCreatedArr)

# Find closest one

'''

Inputs: copter latitude, copter longitude

Locates nearest mesonet station using geopy vincenty distance

Returns Site name and identifier in form: SITE/longname

'''

f = open(fmeso)

reader = csv.DictReader(f)

mesoLat, mesoLon, mesoName, mesoLong = ([] for i in range(4))

for line in reader:

mesoLat.append(float(line['nlat']))

mesoLon.append(float(line['elon']))

mesoName.append(line['stid'])

mesoLong.append(line['name'])

mesoLat = np.array(mesoLat)

mesoLon = np.array(mesoLon)

# calculate distances to mesonet sites

distances = np.array([geopy.distance.vincenty((cop_lat, cop_lon), (iPos)) \

for iPos in zip(mesoLat, mesoLon)])

iMeso = distances.argmin()

site = mesoName[iMeso]

print 'Site identified: %s' % site

'''

Inputs: vertical speed, altitude, time

Determines time to begin averageing. Finds time copter begins ascent after

hovering and then selects time 5 points prior.

Returns datetime format of beginning time

'''

# only look at first half of flight while hovering (below 15 m)

i = np.argmax(altRaw)

ii = np.where((altRaw[:i] > 8.) & (altRaw[:i] < 15.))

althov = altRaw[ii]

vspdhov = vertSpdRaw[ii]

timehov = timeRaw[ii]

# cut in half to only look at hovering -> ascent portion

althov = althov[len(althov)/2:]

vspdhov = vspdhov[len(vspdhov)/2:]

timehov = timehov[len(timehov)/2:]

iii = np.where(vspdhov[len(vspdhov)/2:] > 1.)[0][0] - 5

'''

Inputs: vertical speed, altitude, time

Determines time to end averaging during descent. Finds time when copter's

vertical speed changes from negative to positive.

Returns datetime format of end time

'''

# only look at second half of flight

i = np.argmax(altRaw)

i += 100

ii = np.squeeze(np.where(vertSpdRaw[i:] > 0.))[0]

try:

urllib2.urlopen('http://216.58.192.142', timeout=1)

return True

except urllib2.URLError as err:

# first check for internet connection

iswifi = check_internet()

if iswifi:

baseURL = 'http://mesonet.org/index.php/dataMdfMts/dataController/getFile/'

URL = baseURL + '%4.4d%2.2d%2.2d%s/mts/TEXT/' % (procYear, procMonth,

procDay, procStation.lower())

fd = urllib2.urlopen(URL)

data_long = fd.read()

data = data_long.split('\n')

time_, RH_, T2m_, T9m_, speed_, direction_, p_ = ([] for i in range(7))

for i in np.arange(3, len(data)-1):

time_.append(float(data[i].split()[2]))

RH_.append(float(data[i].split()[3]))

T2m_.append(float(data[i].split()[4]))

T9m_.append(float(data[i].split()[14]))

speed_.append(float(data[i].split()[5]))

direction_.append(float(data[i].split()[7]))

p_.append(float(data[i].split()[12]))

speed_kts = [i * 1.94 for i in speed_]

u,v = mcalc.get_wind_components(speed_kts*units.kts, direction_ * units.deg)

u = u.to(units.kts) / units.kts

v = v.to(units.kts) / units.kts

return np.array([time_, RH_, T2m_, T9m_, u, v, p_])

else:

'''

timeCopter: datetime object of time takeoff or time land

Returns index of nearest mesonet 5-minute observation

'''

mesoTimes = np.linspace(0, 1435, num=1440/5)

baseDay = datetime(timeCopter.year, timeCopter.month, timeCopter.day)

baseDay = baseDay.replace(tzinfo=pytz.UTC)

timeCopter = timeCopter.replace(tzinfo=pytz.UTC)

numMin = (timeCopter - baseDay).total_seconds() / 60.

'''

Input: Environmental Temperature, Profile Temperature, pressure

Calculates surface-based cape based on following equation:

CAPE = Rd * integral(T'(p) - T(p)) dlnp

use "bottom Riemann sum":

uavCAPE = Rd * sum(T'(p) - T(p)) * delta_p/p

Returns SBCAPE for extent of profile

'''

dlnp = []

for i in range(len(Tenv) - 1):

dlnp.append(np.abs(pressure[i+1] - pressure[i]) / pressure[i])

dT = Tprof.to('kelvin') - Tenv.to('kelvin')

dCAPE = dT[:-1] * dlnp

'''

Inputs: Time from data, interpolated time, selected altitudes, parameter to

be displayed

Interpolates data in time for each height for multiple UAV profiles

Returns numpy array of interpolated values

'''

data_interp = np.full((len(z), len(tnew)), np.nan)

for i in np.arange(len(z)):

f = interp1d(t, data[i, :])

fnew = f(tnew)

data_interp[i, :] = fnew

data_interp = ma.masked_invalid(data_interp)

'''

Inputs: temperature, dewpoint, and pressure

Returns: lcl pressure, lcl temperature, isbelowlcl flag, profile temp

'''

lclpres, lcltemp = mcalc.lcl(p[0] * units.mbar,

T[0] * units.degC, Td[0] * units.degC)

print 'LCL Pressure: %5.2f %s' % (lclpres.magnitude, lclpres.units)

print 'LCL Temperature: %5.2f %s' % (lcltemp.magnitude, lcltemp.units)

# parcel profile

# determine if there are points sampled above lcl

ilcl = np.squeeze(np.where((p * units.mbar) <= lclpres))

# if not, entire profile dry adiabatic

if ilcl.size == 0:

prof = mcalc.dry_lapse(p * units.mbar, T[0] * units.degC).to('degC')

isbelowlcl = 1

# if there are, need to concat dry & moist profile ascents

else:

ilcl = ilcl[0]

prof_dry = mcalc.dry_lapse(p[:ilcl] * units.mbar,

T[0] * units.degC).to('degC')

prof_moist = mcalc.moist_lapse(p[ilcl:] * units.mbar,

prof_dry[-1]).to('degC')

prof = np.concatenate((prof_dry, prof_moist)) * units.degC

isbelowlcl = 0

'''

Input filepath of post-processed uav data

Outputs Skew-T log-p plot of UAV data, includes hodograph and some

convective parameters

'''

copdata = csvread_copter(filenamecsv)

lat = copdata[0]

lon = copdata[1]

alt = copdata[2]

pressure = copdata[3]

temperature = copdata[4]

dewpoint = copdata[5]

speed = copdata[9]

speed_kts = speed * 1.94

direction = copdata[10]

site = findSite(lat[0], lon[0])

sitename, sitelong = site.split('/')

fname = filenamecsv.split('\\')[-1]

timeTakeoff = datetime.strptime(fname[:15], '%Y%m%d_%H%M%S')

copterNum = fname[-10]

u,v = mcalc.get_wind_components(speed_kts*units.kts, direction * units.deg)

u = u.to(units.kts)

v = v.to(units.kts)

# Wind shear

bulkshear = speed_kts[-3] - speed_kts[0]

print '0-%d m Bulk Shear: %.0f kts' % (alt[-3], bulkshear)

if np.isnan(dewpoint).all():

moist = 0

else:

moist = 1

print 'Plotting...'

fignum = plt.figure(figsize=(12,9))

gs = gridspec.GridSpec(4, 4)

skew = SkewT(fignum, rotation=20, subplot=gs[:, :2])

skew.plot(pressure, temperature, 'r', linewidth = 2)

skew.plot(pressure, dewpoint, 'g', linewidth = 2)

skew.plot_barbs(pressure[0::4], u[0::4], v[0::4], x_clip_radius = 0.12, \

y_clip_radius = 0.12)

# Plot convective parameters

if moist:

plcl, Tlcl, isbelowlcl, profile = parcelUAV(temperature,

dewpoint, pressure)

SBCAPE = uavCAPE(temperature * units.degC, profile,

pressure * units.hPa)

skew.plot(plcl, Tlcl, 'ko', markerfacecolor='black')

skew.plot(pressure, profile, 'k', linewidth=2)

else:

isbelowlcl = 0

# set up plot limits and labels - use LCL as max if higher than profile

# if moist:

# xmin = math.floor(np.nanmin(dewpoint)) + 2

# else:

# xmin = math.floor(np.nanmin(temperature))

# xmax = math.floor(np.nanmax(temperature)) + 20

xmin = 0.

xmax = 50.

if isbelowlcl:

ymin = round((plcl / units.mbar), -1) - 10

else:

ymin = round(np.nanmin(pressure),-1) - 10

ymax = round(np.nanmax(pressure),-1) + 10

skew.ax.set_ylim(ymax, ymin)

skew.ax.set_xlim(xmin, xmax)

skew.ax.set_yticks(np.arange(ymin, ymax+10, 10))

skew.ax.set_xlabel('Temperature ($^\circ$C)')

skew.ax.set_ylabel('Pressure (hPa)')

titleName = 'Coptersonde-%s %s UTC - %s' % (copterNum,

timeTakeoff.strftime('%d-%b-%Y %H:%M:%S'), sitename)

skew.ax.set_title(titleName)

skew.plot_dry_adiabats(linewidth=0.75)

skew.plot_moist_adiabats(linewidth=0.75)

skew.plot_mixing_lines(linewidth=0.75)

# Hodograph

ax_hod = fignum.add_subplot(gs[:2,2:])

#gs.tight_layout(fig5)

if np.nanmax(speed_kts) > 18:

comprange = 35

else:

comprange = 20

h = Hodograph(ax_hod, component_range=comprange)

h.add_grid(increment=5)

h.plot_colormapped(u, v, pressure, cmap=cmocean.cm.deep_r)

ax_hod.set_title('Hodograph (kts)')

ax_hod.yaxis.set_ticklabels([])

#ax_hod.set_xlabel('Wind Speed (kts)')

# Map - Oklahoma

llcrnrlat = 33.6

urcrnrlat = 37.2

llcrnrlon = -103.2

urcrnrlon = -94.2

ax_map = fignum.add_subplot(gs[2, 2:])

m = Basemap(projection='merc', llcrnrlat=llcrnrlat, urcrnrlat=urcrnrlat,

llcrnrlon=llcrnrlon,urcrnrlon=urcrnrlon, lat_ts=20, resolution='l',

ax=ax_map)

print 'Basemap...'

m.drawcounties()

m.drawstates()

x,y = m(lon[0], lat[0])

plt.plot(x,y,'b.')

plt.text(x+40000, y-5000, sitelong,

bbox=dict(facecolor='yellow', alpha=0.5))

if moist:

# Convective parameter values

ax_data = fignum.add_subplot(gs[3, 2])

plt.axis('off')

datastr = 'LCL = %.0f hPa\nSBCAPE = %.0f J kg$^{-1}$\n0-%.0f m bulk shear\n\

= %.0f kts' % \

(plcl.magnitude, SBCAPE.magnitude, alt[-3], bulkshear)

boxprops = dict(boxstyle='round', facecolor='none')

ax_data.text(0.05, 0.95, datastr, transform=ax_data.transAxes,

fontsize=14, verticalalignment='top', bbox=boxprops)

# Logos

ax_png = fignum.add_subplot(gs[3, 3])

img = mpimg.imread(logoName)

plt.axis('off')

plt.imshow(img)

else:

# Logos

ax_png = fignum.add_subplot(gs[3, 2:])

img = mpimg.imread(logoName)

plt.axis('off')

plt.imshow(img)

plt.show(block=False)

return