color_surf=(0, 0, 0.5, 0.5),color_wp=(0, 0.5, 0, 0.5),

add_date=False, wprof_hgt=None):

" process gapflow analysis"

" first index is mesowest file, second index is \

BBY (1 or more files) "

f = get_filenames(case,homedir)

" parse mesowest data "

meso = mf.parse_mesowest_excel(f[0])

t = get_times(case)

mpress = meso.loc[t[0]: t[1]]['PMSL'].values

mesoidx = meso.loc[t[0]: t[1]].index

" parse surface BBY data "

if len(f[1]) > 1:

' more than one day of obs '

surf = mf.parse_surface(f[1][0])

for ff in f[1][1:]:

surf = surf.append(mf.parse_surface(ff))

else:

' only one day '

surf = mf.parse_surface(f[1][0])

" resample to 1min so we can find \

mesowest index "

surf = surf.resample('1T').interpolate()

" adjust bias is mesowest in case 1 and 2 "

if case in [1, 2]:

bias = 9

else:

bias = 0

spress = surf.loc[mesoidx]['press'].values - bias

" BBY and mesowest pressure difference "

pressDiff = spress - mpress

" BBY surface winds "

swspd = surf.loc[mesoidx]['wspd'].values

swdir = surf.loc[mesoidx]['wdir'].values

ucomp = -swspd*np.sin(np.radians(swdir))

" gapflow dataframe "

d = {'ucomp': ucomp, 'wspd': swspd, 'wdir': swdir,

'pdiff': pressDiff, 'Bpress': spress, 'Kpress': mpress}

gapflow = pd.DataFrame(data=d, index=mesoidx)

" removes rows with NaN "

gapflow = gapflow[np.isfinite(gapflow['Kpress'])]

" add wind profiler data at target altitude "

out = get_windprof(case,

gapflow_time=gapflow.index,

top_hgt_km=wprof_hgt,

homedir=homedir)

wp_wspd, wp_wdir = out

wp_ucomp = -wp_wspd*np.sin(np.radians(wp_wdir))

gapflow['wp_ws'] = wp_wspd

gapflow['wp_wd'] = wp_wdir

gapflow['wp_ucomp'] = wp_ucomp

" Mass etal 95 equation "

# blh = get_BLH(case)

blh = 500 #[m]

massPa, massU = mass_eq(air_density=1.24, BLH=blh)

path = make_polygon(massPa, massU)

gapflow = check_polygon(gapflow, path)

# gapflow['gapflow'] = ((gapflow.poly is True) & (gapflow.wdir <= 120))

# sub = gapflow[(gapflow.poly is True) & (gapflow.wdir <= 120)]

" theoretical lines "

if plot_theory:

if ax is None:

fig, ax = plt.subplots(figsize=(8, 7))

ax.set_xlabel('Pressure difference, BBY-SCK [hPa]')

ax.set_ylabel('BBY zonal wind [m s-1]')

# ax.scatter(gapflow['pdiff'], gapflow['ucomp'],

# color=color_surf, label='surf')

# ax.scatter(gapflow['pdiff'], gapflow['wp_ucomp'],

# color=color_wp, label='wp')

# ax.scatter(sub['pdiff'], sub['ucomp'], color='r')

ax.plot(massPa/100, massU[0], marker=None)

ax.plot(massPa/100, massU[1], linestyle='--', color='r')

ax.plot(massPa/100, massU[2], linestyle='--', color='r')

if grid:

ax.grid(True)

ax.set_xlim([-12, 1])

ax.set_ylim([-20, 15])

ini = mesoidx[0]

end = mesoidx[-1]

date = ini.strftime('%b-%Y ')

beg = ini.strftime('%d')

end = end.strftime('%d')

# ax.text(0.03, 0.76, timetxt.format(str(case).zfill(2),

# date, beg, end),

if add_date is True:

if beg == end:

ax.text(0.03, 0.85,'{} {}'.format(beg,date),

fontsize=12,

transform=ax.transAxes)

else:

ax.text(0.03, 0.85,'{}-{} {}'.format(beg,end,date),

fontsize=12,

transform=ax.transAxes)

top_hgt_km=None,

homedir=None):

" return layer average values; layer is defined \

by top_hgt_km"

import Windprof2 as wp

from scipy.interpolate import interp1d

from datetime import timedelta

out = wp.make_arrays(resolution='coarse',

surface=False, case=str(case), period=False,

homedir=homedir)

wspd, wdir, time, hgt = out

time = np.array(time)

' match surface obs timestamp'

time2 = time - timedelta(minutes=5)

' get corresponding target time index '

index_dict = dict((value, idx) for idx, value in enumerate(time2))

target_idx = [index_dict[x] for x in gapflow_time]

' wp time equivalent to gapflow time '

target_time = time2[target_idx] + timedelta(minutes=5)

wspd_target = []

wdir_target = []

for tt in target_time:

idx = np.where(time == tt)[0]

# ws = np.squeeze(wspd[:, idx])

# wd = np.squeeze(wdir[:, idx])

# ' interpolate at target altitude '

# fws = interp1d(hgt, ws)

# fwd = interp1d(hgt, wd)

# new_ws = fws(target_hgt_km)

# new_wd = fwd(target_hgt_km)

hidx = np.where(hgt <= top_hgt_km)[0]

ws = np.squeeze(wspd[hidx, idx])

wd = np.squeeze(wdir[hidx, idx])

u = -ws*np.sin(np.radians(wd))

v = -ws*np.cos(np.radians(wd))

ubar = np.nanmean(u)

vbar = np.nanmean(v)

new_ws = np.sqrt(ubar**2+vbar**2)

new_wd = 270-np.arctan2(vbar,ubar)*180/np.pi

if new_wd > 360:

new_wd -= 360

wspd_target.append(new_ws)

wdir_target.append(new_wd)

x = df['pdiff'].values

y = df['ucomp'].values

coords = zip(x, y)

poly = []

for c in coords:

poly.append(path.contains_point(c))

df['poly'] = pd.Series(poly, index=df.index)

x = np.concatenate((X, X[::-1]))/100.

y = np.concatenate((Y[2], Y[1][::-1]))

vertices = zip(x, y)

npoints = len(vertices)

codes = [Path.MOVETO]+[Path.LINETO]*(npoints-2)+[Path.CLOSEPOLY]

path = Path(vertices, codes)

# patch=patches.PathPatch(path,facecolor='orange')

'''

retrieve a boundary layer height based on

a subjective analysis of vertical directional

shear in wind profiler

'''

case = str(case)

blh = {'1': 100,

'2': 100,

'3': 500,

'4': 500,

'5': 500,

'6': 500,

'7': 500,

'8': 100,

'9': 200,

'10': 150,

'11': 500,

'12': 200,

'13': 450,

'14': 500}

# basedir = os.path.expanduser('~')

b = basedir+'/SURFACE'

surf_files = {1: [b+'/case01/KSCK.xls', [b+'/case01/bby98018.met', b+'/case01/bby98019.met']],

2: [b+'/case02/KSCK.xls', [b+'/case02/bby98026.met', b+'/case02/bby98027.met']],

3: [b+'/case03/KSCK.xls', [b+'/case03/bby01023.met', b+'/case03/bby01024.met']],

4: [b+'/case04/KSCK.xls', [b+'/case04/bby01025.met', b+'/case04/bby01026.met']],

5: [b+'/case05/KSCK.xls', [b+'/case05/bby01040.met', b+'/case05/bby01041.met']],

6: [b+'/case06/KSCK.xls', [b+'/case06/bby01042.met']],

7: [b+'/case07/KSCK.xls', [b+'/case07/bby01048.met']],

8: [b+'/case08/KSCK.xls', [b+'/case08/bby03012.met', b+'/case08/bby03013.met', b+'/case08/bby03014.met']],

9: [b+'/case09/KSCK.xls', [b+'/case09/bby03021.met', b+'/case09/bby03022.met', b+'/case09/bby03023.met']],

10: [b+'/case10/KSCK.xls', [b+'/case10/bby03046.met', b+'/case10/bby03047.met']],

11: [b+'/case11/KSCK.xls', [b+'/case11/bby04009.met']],

12: [b+'/case12/KSCK.xls', [b+'/case12/bby04033.met']],

13: [b+'/case13/KSCK.xls', [b+'/case13/bby04047.met', b+'/case13/bby04048.met', b+'/case13/bby04049.met']],

14: [b+'/case14/KSCK.xls', [b+'/case14/bby04056.met']]

}

" storm times "

slice_times = {1: [datetime(1998, 1, 18, 0, 56), datetime(1998, 1, 18, 23, 56)],

2: [datetime(1998, 1, 26, 0, 56), datetime(1998, 1, 27, 3, 56)],

3: [datetime(2001, 1, 23, 0, 0), datetime(2001, 1, 25, 0, 0)],

4: [datetime(2001, 1, 25, 0, 0), datetime(2001, 1, 27, 0, 0)],

5: [datetime(2001, 2, 9, 0, 0), datetime(2001, 2, 11, 0, 0)],

6: [datetime(2001, 2, 11, 0, 0), datetime(2001, 2, 12, 0, 0)],

7: [datetime(2001, 2, 17, 0, 0), datetime(2001, 2, 18, 0, 0)],

8: [datetime(2003, 1, 12, 0, 0), datetime(2003, 1, 15, 0, 0)],

9: [datetime(2003, 1, 21, 0, 0), datetime(2003, 1, 24, 0, 0)],

10: [datetime(2003, 2, 15, 0, 0), datetime(2003, 2, 17, 0, 0)],

11: [datetime(2004, 1, 9, 0, 0), datetime(2004, 1, 10, 0, 0)],

12: [datetime(2004, 2, 2, 0, 0), datetime(2004, 2, 3, 0, 0)],

13: [datetime(2004, 2, 16, 0, 0), datetime(2004, 2, 19, 0, 0)],

# 13: [datetime(2004, 2, 16, 0, 0), datetime(2004, 2, 17, 6, 0)],

14: [datetime(2004, 2, 25, 0, 0), datetime(2004, 2, 26, 0, 0)]}

' Gap Flow based on Mass et al (1995) MWR '

H = BLH # [m] height of well-mixed boundary layer

BLcoeff = 2.8 # boundary layer coef (Deardorff 1972)

npoints = 100

delPa = np.linspace(0, -1200, npoints) # [Pa]

delX = 100000 # [m] distance btwn gap entrance and BBY

rho = air_density # [kg m-3] average air density

PGF = -(1/rho)*(delPa/delX)

Cd = np.array([7.5e-3, 0.5*7.5e-3, 1.5*7.5e-3]) # drag coefficient

umass = []

for c in Cd:

K = BLcoeff*c/H

U2 = (PGF/K) * (1-np.exp(-2*K*delX))

umass.append(-np.real(np.sqrt(U2)))

Rd = 287. # [J K-1 kg-1]

# density values do not vary significantly

Tv = tm.virtual_temperature(C=stemp, mixing_ratio=smixr/1000.)+273.15

air_density1 = (spress*100.)/(Rd*Tv)

Tv = tm.virtual_temperature(C=mtemp, mixing_ratio=smixr/1000.)+273.15

import tta_analysis3 as tta

out = tta.preprocess(years=years, layer=param['wdir_layer'])

# print out

result = tta.analyis(out, param)

precip = result['precip']

dates = precip[precip.tta]