def onset_TT(T, north=(5, 35, 40, 100), south=(-15, 5, 40, 100),
yearnm='year', daynm='day'):
"""Return monsoon onset index based on tropospheric temperature.
Parameters
----------
T : xray.DataArray
Air temperature 200-600 hPa vertical mean.
north, south : 4-tuples of floats, optional
Tuple of (lat1, lat2, lon1, lon2) defining northern and
southern regions to average over.
yearnm, daynm : str, optional
Name of year and day dimensions in DataArray
Returns
-------
tt : xray.Dataset
TT daily timeseries for each year and monsoon onset day for
each year.
Reference
---------
Goswami, B. N., Wu, G., & Yasunari, T. (2006). The annual cycle,
intraseasonal oscillations, and roadblock to seasonal
predictability of the Asian summer monsoon. Journal of Climate,
19, 5078-5099.
"""
ttn = atm.mean_over_geobox(T, north[0], north[1], north[2], north[3])
tts = atm.mean_over_geobox(T, south[0], south[1], south[2], south[3])
tseries = ttn - tts
# Onset day is the first day that ttn-tts becomes positive
years = tseries[yearnm]
onset = np.zeros(years.shape)
for y in range(len(years)):
pos = (tseries.values[y] > 0)
if not pos.any():
onset[y] = np.nan
else:
onset[y] = tseries[daynm][pos.argmax()]
tt = xray.Dataset()
tt['ttn'] = ttn
tt['tts'] = tts
tt['tseries'] = tseries
tt['onset'] = xray.DataArray(onset, coords={yearnm : years})
tt.attrs['north'] = north
tt.attrs['south'] = south
return tt
def get_mfc_box(mfcfiles, precipfiles, evapfiles, years, nroll, lat1, lat2,
lon1, lon2):
"""Return daily tseries MFC, precip and evap averaged over lat-lon box.
"""
subset_dict = {'lat' : (lat1, lat2), 'lon' : (lon1, lon2)}
databox = {}
if mfcfiles is not None:
mfc = atm.combine_daily_years('MFC', mfcfiles, years, yearname='year',
subset_dict=subset_dict)
databox['MFC'] = mfc
if precipfiles is not None:
pcp = atm.combine_daily_years('PRECTOT', precipfiles, years, yearname='year',
subset_dict=subset_dict)
databox['PCP'] = pcp
if evapfiles is not None:
evap = atm.combine_daily_years('EVAP', evapfiles, years, yearname='year',
subset_dict=subset_dict)
databox['EVAP'] = evap
nms = databox.keys()
for nm in nms:
var = databox[nm]
var = atm.precip_convert(var, var.attrs['units'], 'mm/day')
var = atm.mean_over_geobox(var, lat1, lat2, lon1, lon2)
databox[nm + '_UNSM'] = var
databox[nm + '_ACC'] = np.cumsum(var, axis=1)
if nroll is None:
databox[nm] = var
else:
databox[nm] = atm.rolling_mean(var, nroll, axis=-1, center=True)
tseries = xray.Dataset(databox)
return tseries
# ----------------------------------------------------------------------
# Read data
nroll = 7
days_ssn = atm.season_days('JJAS')
subset_dict = {'lat' : (lat1, lat2), 'lon' : (lon1, lon2)}
ts = xray.Dataset()
ssn = xray.Dataset()
varnms = {'PCP' : 'PRECTOT', 'MFC' : 'MFC', 'EVAP' : 'EVAP', 'W' : 'TQV',
'ANA' : 'DQVDT_ANA'}
for nm in files:
var = atm.combine_daily_years(varnms[nm], files[nm], years, yearname='year',
subset_dict=subset_dict)
var = atm.mean_over_geobox(var, lat1, lat2, lon1, lon2)
units = var.attrs.get('units')
if units in ['kg/m2/s', 'kg m-2 s-1']:
var = atm.precip_convert(var, units, 'mm/day')
var_sm = atm.rolling_mean(var, nroll, axis=-1, center=True)
ts[nm] = var_sm
if nm == 'W':
# Difference in precipitable water from beginning to end of season
var_ssn = var_sm.sel(day=days_ssn)
ssn['d' + nm] = (var_ssn[:,-1] - var_ssn[:, 0]) / len(days_ssn)
# dW/dt
dvar = np.nan * np.ones(var.shape)
for y, year in enumerate(years):
dvar[y] = np.gradient(var[y])
ts['d%s/dt' % nm] = xray.DataArray(dvar, coords=var.coords)
else:
season = 'jja' lon1, lon2 = 0, 100 lat1, lat2 = -20, 50 u = load_daily_season(ustr, year, season, 'U', lat1, lat2, lon1, lon2) plt.figure() atm.pcolor_latlon(u.mean(dim='TIME')) ds = load_daily_season(ustr, year, season, None, lat1, lat2, lon1, lon2) u2 = ds['U'] print((u == u2).any()) season = 'ann' lon1, lon2 = 20, 120 lat1, lat2 = -60, 60 n = 8 # Number of time points per day days = np.arange(1, 366) u = load_daily_season(ustr, year, season, 'U', lat1, lat2, lon1, lon2) udaily = atm.daily_from_subdaily(u, n, dayvals=days) plt.figure() atm.pcolor_latlon(udaily.mean(axis=0), axlims=(lat1,lat2,lon1,lon2)) lat1, lat2 = 10, 30 lon1, lon2 = 60, 100 ubar = atm.mean_over_geobox(u, lat1, lat2, lon1, lon2) ubar_daily = atm.mean_over_geobox(udaily, lat1, lat2, lon1, lon2) plt.figure(figsize=(7,8)) plt.subplot(211) plt.plot(ubar) plt.subplot(212) plt.plot(ubar_daily)
def onset_SJKE(u, v, latlon = (-5, 20, 50, 70), ndays=3, yearnm='Year',
daynm='Day', thresh_std=1.0):
"""Return monsoon onset based on Somali Jet kinetic energy.
Parameters
----------
u, v : xray.DataArray
850 hPa zonal and meridional wind.
latlon : 4-tuple of floats, optional
Tuple of (lat1, lat2, lon1, lon2) defining Somali jet region
to average over.
ndays : int, optional
Number of consecutive days threshold must be exceeded to
define onset.
yearnm, daynm : str, optional
Name of year and day dimensions in DataArray
thresh_std : float, optional
Number of standard deviations excursion to use as onset threshold.
Returns
-------
sjke : xray.Dataset
Somali jet index daily timeseries for each year and monsoon
onset day for each year.
Reference
---------
Boos, W. R., & Emanuel, K. A. (2009). Annual intensification of the
Somali jet in a quasi-equilibrium framework : Observational
composites. Quarterly Journal of the Royal Meteorological
Society, 135, 319-335.
"""
days = atm.get_coord(u, coord_name=daynm)
years = atm.get_coord(u, coord_name=yearnm)
nyears = len(years)
# Kinetic energy index
ke = np.sqrt(u**2 + v**2)
# Average over Somali jet region
lat1, lat2, lon1, lon2 = latlon
ke = atm.mean_over_geobox(ke, lat1, lat2, lon1, lon2)
ke.attrs['title'] = 'KE'
ke.attrs['long_name'] = 'sqrt(u**2 + v**2)'
# Threshold for onset date
vals = ke.values.flatten()
keclim = np.nanmean(vals)
kestd = np.nanstd(vals)
threshold = keclim + thresh_std * kestd
# Find first day when KE exceeds threshold and stays above the
# threshold for consecutive ndays
def onset_day(tseries, threshold, ndays, daynm):
above = (tseries.values > threshold)
d0 = above.argmax()
while not above[d0:d0+ndays].all():
d0 += 1
return tseries[daynm].values[d0]
# Find onset day for each year
onset = [onset_day(ke[y], threshold, ndays, daynm)
for y in range(nyears)]
# Pack into dataset
sjke = xray.Dataset()
sjke['tseries'] = ke
sjke['onset'] = xray.DataArray(onset, coords={yearnm : years})
sjke.attrs = {'latlon' : latlon, 'thresh_std' : thresh_std,
'threshold' : threshold, 'ndays' : ndays}
return sjke
def onset_OCI(u, latlon = (5, 15, 40, 80), mmdd_thresh=(6,1),
ndays=7, yearnm='Year', daynm='Day'):
"""Return monsoon Onset Circulation Index.
Parameters
----------
u : xray.DataArray
850 hPa zonal wind.
latlon : 4-tuple of floats, optional
Tuple of (lat1, lat2, lon1, lon2) defining South Arabian Sea
region to average over.
mmdd_thres : 2-tuple of ints, optional
Tuple of (month, day) defining climatological mean onset date
to use for threshold value of u.
ndays : int, optional
Number of consecutive days threshold must be exceeded to
define onset.
yearnm, daynm : str, optional
Name of year and day dimensions in DataArray
Returns
-------
oci : xray.Dataset
OCI daily timeseries for each year and monsoon onset day for
each year.
Reference
---------
Wang, B., Ding, Q., & Joseph, P. V. (2009). Objective Definition
of the Indian Summer Monsoon Onset. Journal of Climate, 22(12),
3303-3316.
"""
days = atm.get_coord(u, coord_name=daynm)
years = atm.get_coord(u, coord_name=yearnm)
nyears = len(years)
# Average over South Arabian Sea region
lat1, lat2, lon1, lon2 = latlon
ubar = atm.mean_over_geobox(u, lat1, lat2, lon1, lon2)
# Find values at climatological onset
m0, d0 = mmdd_thresh
d0 = [atm.mmdd_to_jday(m0, d0, year) for year in years]
u0 = [ubar.sel(**{daynm : day, yearnm : year}).values
for year, day in zip(years, d0)]
u0 = np.array(u0).flatten()
uthreshold = np.mean(u0)
# Find first day when OCI exceeds threshold and stays above the
# threshold for consecutive ndays
def onset_day(tseries, uthreshold, ndays, daynm):
above = (tseries.values > uthreshold)
d0 = above.argmax()
while not above[d0:d0+ndays].all():
d0 += 1
return tseries[daynm].values[d0]
# Find onset day for each year
onset = [onset_day(ubar[y], uthreshold, ndays, daynm)
for y in range(nyears)]
# Pack into dataset
oci = xray.Dataset()
oci['tseries'] = ubar
oci['onset'] = xray.DataArray(onset, coords={yearnm : years})
oci.attrs['latlon'] = latlon
oci.attrs['mmdd_thresh'] = mmdd_thresh
oci.attrs['ndays'] = ndays
return oci
index['tseries'] = xray.DataArray(pcp_sm, dims=['year', 'day'],
coords={'year' : years, 'day': days})
index['onset'] = xray.DataArray(d_onset, coords={'year' : years})
index['retreat'] = xray.DataArray(d_retreat, coords={'year' : years})
index.attrs['title'] = titlestr
return index
# Threshold and smoothing parameters
threshold = 5.0
kmax = 12
nroll = {'CMAP' : 3, 'MERRA_MFC' : 7, 'MERRA_PRECIP' : 7}
# Read CMAP pentad precip
cmap = precipdat.read_cmap(cmapfile)
cmapbar = atm.mean_over_geobox(cmap, lat1, lat2, lon1, lon2)
cmapdays = [atm.pentad_to_jday(p, pmin=1) for p in cmap.pentad.values]
# MERRA moisture flux convergence
mfc = atm.combine_daily_years('MFC', mfcfiles, years, yearname='year')
mfcbar = atm.mean_over_geobox(mfc, lat1, lat2, lon1, lon2)
# MERRA precip
subset_dict = {'lon' : (lon1, lon2), 'lat' : (lat1, lat2)}
precip = atm.combine_daily_years('PRECTOT', precipfiles, years, yearname='year',
subset_dict=subset_dict)
precip = atm.precip_convert(precip, precip.attrs['units'], 'mm/day')
precipbar = atm.mean_over_geobox(precip, lat1, lat2, lon1, lon2)
# Compute indices for each dataset
for name in ['CMAP', 'MERRA_MFC', 'MERRA_PRECIP']:
# ----------------------------------------------------------------------
# Read data
data = collections.OrderedDict()
for nm in datafiles:
var, onset, retreat = utils.load_dailyrel(datafiles[nm])
data[nm] = var
# ----------------------------------------------------------------------
# Test atm.regress_field
day = -60
# 1-d timeseries
var = data['U200'].sel(dayrel=day)
ts = atm.mean_over_geobox(var, 10, 30, 60, 100)
ts_reg = atm.Linreg(onset, ts)
ts_reg2 = atm.regress_field(ts, onset)
print(ts_reg.r, ts_reg2.r.values)
print(ts_reg.slope, ts_reg2.m.values)
print(ts_reg.p, ts_reg2.p.values)
# x-y data
regdays = [-60, -30, 0, 30, 60]
plotdays = [-60, -30]
clev_r = np.arange(-1.0, 1.01, 0.05)
for nm in varnms:
print(nm)
var = data[nm].sel(dayrel=regdays)
reg_daily = atm.regress_field(var, onset, axis=0)
for day in plotdays:
if nm == 'PSI':
nm0 = 'V_sector_%dE-%dE' % (lon1, lon2)
elif nm == 'VFLXLQV':
nm0 = 'VFLXQV'
elif nm == dtheta_nm:
nm0 = theta_nm
else:
nm0 = nm
relfiles[nm] = filestr % (nm0, onset_nm, yearstr)
# ----------------------------------------------------------------------
# Read data and calculate indices
# Precipitation
precip = precipdat.read_cmap(pcpfile, yearmin=min(years), yearmax=max(years))
pcp_box = atm.mean_over_geobox(precip, lat1, lat2, lon1, lon2)
# -- Interpolate to daily resolution
days = np.arange(1, 367)
pcp_i = np.nan * np.ones((len(years), len(days)))
for y, year in enumerate(years):
pcp_i[y] = np.interp(days, pcp_box['day'], pcp_box[y])
coords = {'day' : days, 'year' : years}
pcp = xray.DataArray(pcp_i, dims=['year', 'day'], coords=coords)
# Monsoon onset, retreat indices
index = utils.get_onset_indices(onset_nm, indfiles, years)
mfc = atm.rolling_mean(index['ts_daily'], nroll, center=True)
onset = index['onset']
ssn_length=index['length'].mean(dim='year')
data = {}
enso = enso.loc[years]
for key in enso_keys:
if key not in enso.columns:
months = atm.season_months(key)
month_names = [(atm.month_str(m)).capitalize() for m in months]
enso[key] = enso[month_names].mean(axis=1)
enso = enso[enso_keys]
col_names = [enso_nm + ' ' + nm for nm in enso.columns]
enso.columns = col_names
# ----------------------------------------------------------------------
# Daily timeseries
ts = xray.Dataset()
for nm in ['GPCP', 'PRECTOT']:
ts[nm] = atm.mean_over_geobox(data[nm], lat1, lat2, lon1, lon2)
ts['MFC'] = utils.daily_rel2onset(index_all['CHP_MFC']['daily_ts'],
index[ind_nm], npre, npost)
ts['CMFC'] = utils.daily_rel2onset(index_all['CHP_MFC']['tseries'],
index[ind_nm], npre, npost)
# Extract variables at specified latitudes
for nm, lat0 in lat_extract.iteritems():
var = atm.dim_mean(data[nm], 'lon', lon1, lon2)
lat = atm.get_coord(var, 'lat')
lat0_str = atm.latlon_labels(lat0, 'lat', deg_symbol=False)
# key = nm + '_' + lat0_str
key = nm
lat_closest, _ = atm.find_closest(lat, lat0)
print '%s %.2f %.2f' % (nm, lat0, lat_closest)
else:
plot_single_WLH(pcp[y], pcp_sm[y], label_sm[y], i_onset[y],
i_retreat[y], i_peak[y], titlestr + str(years[y]))
if iplot < 4:
plt.xticks(np.arange(0, 74, 4), [])
else:
plt.xticks(np.arange(0, 74, 4))
plt.xlabel('Pentad')
# Read data and average over box
lon1, lon2 = 60, 100
lat1, lat2 = 10, 30
titlestr = 'CMAP %d-%dE, %d-%dN ' % (lon1, lon2, lat1, lat2)
precip = precipdat.read_cmap(cmap_file)
precipbar = atm.mean_over_geobox(precip, lat1, lat2, lon1, lon2)
nyears, npentad = precipbar.shape
years = precipbar.year.values
nyears = len(years)
pentads = precipbar.pentad
onset = {}
# Threshold for onset criteria
threshold = 5.0
# Smooth with truncated Fourier series
kmax = 12
kann = 4
pcp_sm, Rsq = atm.fourier_smooth(precipbar, kmax)
pcp_ann, Rsq_ann = atm.fourier_smooth(precipbar, kann)
label_sm, label_ann = [], []
# ----------------------------------------------------------------------
# Read data
data = collections.OrderedDict()
for nm in datafiles:
var, onset, retreat = utils.load_dailyrel(datafiles[nm])
data[nm] = var
# ----------------------------------------------------------------------
# Test atm.regress_field
day = -60
# 1-d timeseries
var = data['U200'].sel(dayrel=day)
ts = atm.mean_over_geobox(var, 10, 30, 60, 100)
ts_reg = atm.Linreg(onset, ts)
ts_reg2 = atm.regress_field(ts, onset)
print(ts_reg.r, ts_reg2.r.values)
print(ts_reg.slope, ts_reg2.m.values)
print(ts_reg.p, ts_reg2.p.values)
# x-y data
regdays = [-60, -30, 0, 30, 60]
plotdays = [-60, -30]
clev_r = np.arange(-1.0, 1.01, 0.05)
for nm in varnms:
print(nm)
var = data[nm].sel(dayrel=regdays)
reg_daily = atm.regress_field(var, onset, axis=0)
for day in plotdays:
import atmos as atm
datadir = "/home/jennifer/datastore/merra/daily/"
filestr = "merra_uv200_40E-120E_60S-60N_"
year = 1980
lon1, lon2 = 60, 100
lat1, lat2 = 10, 30
filename = "%s%s%d.nc" % (datadir, filestr, year)
ds = atm.ncload(filename)
iplot = {"U": 1, "V": 2, "rel_vort": 3, "Ro": 4}
plt.figure(figsize=(12, 9))
for var in ds.data_vars:
plt.subplot(2, 2, iplot[var])
atm.pcolor_latlon(ds[var].mean(axis=0))
plt.title(var)
dsbar = xray.Dataset()
for var in ds.data_vars:
print(var)
dsbar[var] = atm.mean_over_geobox(ds[var], lat1, lat2, lon1, lon2)
plt.figure(figsize=(12, 9))
for var in dsbar.data_vars:
plt.subplot(2, 2, iplot[var])
plt.plot(dsbar.Day, dsbar[var])
plt.title(var)
datadir = atm.homedir() + "datastore/merra/daily/"
savedir = atm.homedir() + "datastore/merra/analysis/"
icalc = False
isavefigs = True
if icalc:
subset_dict = {"lat": (lat1, lat2), "lon": (lon1, lon2)}
for varnm in varnms:
if varnm == "MFC":
varid, filestr = "MFC", "MFC_ps-300mb"
elif varnm == "precip":
varid, filestr = "PRECTOT", "precip"
files = [datadir + "merra_%s_%d.nc" % (filestr, year) for year in years]
data = atm.combine_daily_years(varid, files, years, yearname="year", subset_dict=subset_dict)
data = atm.mean_over_geobox(data, lat1, lat2, lon1, lon2)
data = atm.precip_convert(data, data.attrs["units"], "mm/day")
# Accumulated precip or MFC
data_acc = np.cumsum(data, axis=1)
# Compute onset and retreat days
chp = onset_changepoint_merged(data_acc)
# Save to file
savefile = savedir + "merra_onset_changepoint_merged_%s.nc" % varnm.upper()
print("Saving to " + savefile)
chp.to_netcdf(savefile)
chp = {}
for varnm in varnms:
