def load_data(self, year='2019'):
self.dfs = {}
for index, row in self.df.iterrows():
if (self.glider_id
in row['Dataset ID']) and (year in row['Dataset ID']):
print(row['Dataset ID'])
try:
e = ERDDAP(
server=self.server_url,
protocol='tabledap',
response='csv',
)
e.dataset_id = row['Dataset ID']
e.constraints = self.constraints
e.variables = self.variables[row['Dataset ID']]
except HTTPError:
print('Failed to generate url {}'.format(
row['Dataset ID']))
continue
self.dfs.update({
row['Dataset ID']:
e.to_pandas(
index_col='time (UTC)',
parse_dates=True,
skiprows=(1, ) # units information can be dropped.
)
})
return (self.dfs)
def get_erddap_dataset(server, protocol, file_type, ds_id, var_list=None):
e = ERDDAP(server=server, protocol=protocol, response=file_type)
e.dataset_id = ds_id
if var_list:
e.variables = var_list
ds = e.to_xarray()
ds = ds.sortby(ds.time)
return ds
def load_glider(dataset_id='ru32-20190102T1317-profile-sci-rt',
server="http://slocum-data.marine.rutgers.edu/erddap"):
''' Load glider data from erddap.
input dataset ID and server
Returns an xarray dataset indexed on time '''
# should change: write to_netcdf, then check if netcdf exists
e = ERDDAP(
server=server,
protocol="tabledap",
response="nc",
)
e.dataset_id = dataset_id
gds = e.to_xarray()
# want to have the dimention be time not obs number
gds = gds.swap_dims({"obs": "time"})
gds = gds.sortby("time")
# drop repeated time values
gds = gds.sel(time=~gds.indexes['time'].duplicated())
# get the seafloor depths too
e2 = ERDDAP(
server="http://slocum-data.marine.rutgers.edu/erddap",
protocol="tabledap",
response="nc",
)
# get some of the raw data:
# e2.dataset_id = dataset_id[:-14] + 'trajectory-raw-rt'
e2.dataset_id = dataset_id.replace('profile-sci', 'trajectory-raw')
e2.variables = ['time', 'm_water_depth', 'm_pitch']
# this connects to the data and load into an xarray dataset
gds_raw = e2.to_xarray().drop_dims('trajectory')
# want to have the dimention be time not obs number
gds_raw = gds_raw.swap_dims({"obs": "time"})
gds_raw = gds_raw.sortby("time")
gds_raw = gds_raw.sel(time=~gds_raw.indexes['time'].duplicated())
# remove bad values:
gds_raw['m_water_depth'] = gds_raw.m_water_depth.where(
gds_raw.m_water_depth > 10, drop=True)
gds['bottom_depth'] = gds_raw.m_water_depth.interp_like(gds,
method='nearest')
return gds
def active_argo_floats(bbox=None, time_start=None, time_end=None, floats=None):
"""
:param lon_lims: list containing westernmost longitude and easternmost latitude
:param lat_lims: list containing southernmost latitude and northernmost longitude
:param time_start: time to start looking for floats
:param time_end: time to end looking for floats
:return:
"""
bbox = bbox or [-100, -45, 5, 46]
time_end = time_end or dt.date.today()
time_start = time_start or (time_end - dt.timedelta(days=1))
floats = floats or False
constraints = {
'time>=': str(time_start),
'time<=': str(time_end),
}
if bbox:
constraints['longitude>='] = bbox[0]
constraints['longitude<='] = bbox[1]
constraints['latitude>='] = bbox[2]
constraints['latitude<='] = bbox[3]
if floats:
constraints['platform_number='] = floats
variables = [
'platform_number',
'time',
'pres',
'longitude',
'latitude',
'temp',
'psal',
]
e = ERDDAP(
server='IFREMER',
protocol='tabledap',
response='nc'
)
e.dataset_id = 'ArgoFloats'
e.constraints = constraints
e.variables = variables
try:
df = e.to_pandas(
parse_dates=['time (UTC)'],
skiprows=(1,) # units information can be dropped.
).dropna()
except HTTPError:
df = pd.DataFrame()
return df
def get_erddap_dataset(server, ds_id, variables=None, constraints=None):
variables = variables or None
constraints = constraints or None
e = ERDDAP(server=server,
protocol='tabledap',
response='nc')
e.dataset_id = ds_id
if constraints:
e.constraints = constraints
if variables:
e.variables = variables
ds = e.to_xarray()
ds = ds.sortby(ds.time)
return ds
def check_dataset_empty(url_erddap,dataset_id,date_ini,date_end,lon_lim,lat_lim):
from erddapy import ERDDAP
constraints = {
'time>=': date_ini,
'time<=': date_end,
'latitude>=': lat_lim[0],
'latitude<=': lat_lim[1],
'longitude>=': lon_lim[0],
'longitude<=': lon_lim[1],
}
variable_names = [
'depth',
'latitude',
'longitude',
'time',
'temperature',
'salinity'
]
e = ERDDAP(
server=url_erddap,
protocol='tabledap',
response='nc'
)
e.dataset_id = dataset_id
e.constraints = constraints
e.variables = variable_names
# Converting glider data to data frame
# Cheching that data frame has data
df = e.to_pandas()
if len(df) < 4:
empty_dataset = True
else:
empty_dataset = False
return empty_dataset
def get_erddap_dataset(ds_id, variables=None, constraints=None, filetype=None):
"""
Returns a netcdf dataset for a specified dataset ID (or dataframe if dataset cannot be converted to xarray)
:param ds_id: dataset ID e.g. ng314-20200806T2040
:param variables: optional list of variables
:param constraints: optional list of constraints
:param filetype: optional filetype to return, 'nc' (default) or 'dataframe'
:return: netcdf dataset
"""
variables = variables or None
constraints = constraints or None
filetype = filetype or 'nc'
e = ERDDAP(server='NGDAC', protocol='tabledap', response='nc')
e.dataset_id = ds_id
if constraints:
e.constraints = constraints
if variables:
e.variables = variables
if filetype == 'nc':
try:
ds = e.to_xarray()
ds = ds.sortby(ds.time)
except OSError:
print('No dataset available for specified constraints: {}'.format(
ds_id))
ds = []
except TypeError:
print('Cannot convert to xarray, providing dataframe: {}'.format(
ds_id))
ds = e.to_pandas().dropna()
elif filetype == 'dataframe':
ds = e.to_pandas().dropna()
else:
print('Unrecognized filetype: {}. Needs to be "nc" or "dataframe"'.
format(filetype))
return ds
def get_erddap_data(dataset_id):
'''
:param dataset_id: the deployment name example:'ce_311-20200708T1723'
:return: pandas DataFrame with deployment variable values
'''
e = ERDDAP(
server='https://gliders.ioos.us/erddap',
protocol='tabledap',
)
e.response = 'csv'
e.dataset_id = dataset_id
e.variables = [
'depth',
'latitude',
'longitude',
'salinity',
'temperature',
'conductivity',
'density',
'time',
]
df = e.to_pandas()
return df
def load_data(self,year='2019'):
self.dfs = {}
for index,row in self.df.iterrows():
if (self.glider_id in row['Dataset ID']) and (year in row['Dataset ID']):
print(row['Dataset ID'])
try:
e = ERDDAP(server=self.server_url,
protocol='tabledap',
response='csv',
)
e.dataset_id=row['Dataset ID']
e.constraints=self.constraints
e.variables=self.variables[row['Dataset ID']]
except HTTPError:
print('Failed to generate url {}'.format(row['Dataset ID']))
continue
self.dfs.update({row['Dataset ID']: e.to_pandas(
index_col='time (UTC)',
parse_dates=True,
skiprows=(1,) # units information can be dropped.
)})
return(self.dfs)
def read_glider_data_erddap_server(url_erddap,dataset_id,\
lat_lim,lon_lim,scatter_plot,**kwargs):
"""
Created on Tue Feb 5 10:05:37 2019
@author: aristizabal
This function reads glider data from the IOOS
Data Assembly Center (DAC).
Inputs:
url_erddap: url address of thredds server
Example: 'https://data.ioos.us/gliders/erddap'
dataset_id: this id is retrieved from the glider DAC using the
function "retrieve_glider_id_erddap_server".
Example: 'ru30-20180705T1825'
lat_lim: latitude limits for the search.
Example, lat_lim = [38.0,40.0]
lon_lim: longitude limits for the search.
Example, lon_lim = [-75.0,-72.0]
date_ini: initial date of time window.
This function accepts the data formats '%Y-%m-%d T %H:%M:%S Z' and '%Y/%m/%d/%H'.
Examaple: date_ini = '2018-08-02T00:00:00Z' or '2018/08/02/00'
date_end: initial date of time window.
This function uses the data format '%Y-%m-%d T %H:%M:%S Z'.
Examaple: date_ini = '2018-08-10T00:00:00Z' and '2018/08/10/00'
scatter_plot: if equal to 'yes' then a scatter plot
of the glider transect is plotted
Outputs:
tempg: all the glider profiles of temperature within the user defined time window
saltg: all the glider profiles of salinity within the user defined time window
latg: latitude within the user defined time window
long: longitude within the user defined time window
timeg: user defined time window
depthg: depth vector for all profiles
"""
from erddapy import ERDDAP
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
import cmocean
import numpy as np
date_ini = kwargs.get('date_ini', None)
date_end = kwargs.get('date_end', None)
# Find time window of interest
if np.logical_or(date_ini == None, date_end == None):
constraints = {
'latitude>=': lat_lim[0],
'latitude<=': lat_lim[1],
'longitude>=': lon_lim[0],
'longitude<=': lon_lim[1],
}
else:
constraints = {
'time>=': date_ini,
'time<=': date_end,
'latitude>=': lat_lim[0],
'latitude<=': lat_lim[1],
'longitude>=': lon_lim[0],
'longitude<=': lon_lim[1],
}
variables = [
'depth', 'latitude', 'longitude', 'time', 'temperature', 'salinity'
]
e = ERDDAP(server=url_erddap, protocol='tabledap', response='nc')
e.dataset_id = dataset_id
e.constraints = constraints
e.variables = variables
# Converting glider data to data frame
# Cheching that data frame has data
df = e.to_pandas()
if len(df) > 3:
df = e.to_pandas(
index_col='time (UTC)',
parse_dates=True,
skiprows=(1, ) # units information can be dropped.
).dropna()
# Coverting glider vectors into arrays
timeg, ind = np.unique(df.index.values, return_index=True)
latg = df['latitude (degrees_north)'].values[ind]
long = df['longitude (degrees_east)'].values[ind]
dg = df['depth (m)'].values
vg1 = df[df.columns[3]].values
vg2 = df[df.columns[4]].values
zn = np.int(np.max(np.diff(np.hstack([ind, len(dg)]))))
depthg = np.empty((zn, len(timeg)))
depthg[:] = np.nan
tempg = np.empty((zn, len(timeg)))
tempg[:] = np.nan
saltg = np.empty((zn, len(timeg)))
saltg[:] = np.nan
for i, ii in enumerate(ind):
if i < len(timeg) - 1:
depthg[0:len(dg[ind[i]:ind[i + 1]]), i] = dg[ind[i]:ind[i + 1]]
tempg[0:len(vg1[ind[i]:ind[i + 1]]),
i] = vg1[ind[i]:ind[i + 1]]
saltg[0:len(vg2[ind[i]:ind[i + 1]]),
i] = vg2[ind[i]:ind[i + 1]]
else:
depthg[0:len(dg[ind[i]:len(dg)]), i] = dg[ind[i]:len(dg)]
tempg[0:len(vg1[ind[i]:len(vg1)]), i] = vg1[ind[i]:len(vg1)]
saltg[0:len(vg2[ind[i]:len(vg2)]), i] = vg2[ind[i]:len(vg2)]
# Scatter plot
if scatter_plot == 'yes':
color_map = cmocean.cm.thermal
varg = tempg
timeg_matrix = np.tile(timeg.T, (depthg.shape[0], 1))
ttg = np.ravel(timeg_matrix)
dg = np.ravel(depthg)
teg = np.ravel(varg)
kw = dict(c=teg, marker='*', edgecolor='none')
fig, ax = plt.subplots(figsize=(10, 3))
cs = ax.scatter(ttg, -dg, cmap=color_map, **kw)
#fig.colorbar(cs)
ax.set_xlim(timeg[0], timeg[-1])
ax.set_ylabel('Depth (m)', fontsize=14)
cbar = plt.colorbar(cs)
cbar.ax.set_ylabel('Temperature ($^oC$)', fontsize=14)
ax.set_title(dataset_id, fontsize=16)
xfmt = mdates.DateFormatter('%H:%Mh\n%d-%b')
ax.xaxis.set_major_formatter(xfmt)
plt.ylim([-np.nanmax(dg), 0])
color_map = cmocean.cm.haline
varg = saltg
timeg_matrix = np.tile(timeg.T, (depthg.shape[0], 1))
ttg = np.ravel(timeg_matrix)
dg = np.ravel(depthg)
teg = np.ravel(varg)
kw = dict(c=teg, marker='*', edgecolor='none')
fig, ax = plt.subplots(figsize=(10, 3))
cs = ax.scatter(ttg, -dg, cmap=color_map, **kw)
#fig.colorbar(cs)
ax.set_xlim(timeg[0], timeg[-1])
ax.set_ylabel('Depth (m)', fontsize=14)
cbar = plt.colorbar(cs)
cbar.ax.set_ylabel('Salinity', fontsize=14)
ax.set_title(dataset_id, fontsize=16)
xfmt = mdates.DateFormatter('%H:%Mh\n%d-%b')
ax.xaxis.set_major_formatter(xfmt)
plt.ylim([-np.nanmax(dg), 0])
else:
tempg = np.nan
saltg = np.nan
timeg = np.nan
latg = np.nan
long = np.nan
depthg = np.nan
return tempg, saltg, timeg, latg, long, depthg
def active_gliders(bbox=None, time_start=None, time_end=dt.date.today(), glider_id=None):
bbox = bbox or [-100, -40, 18, 60]
time_start = time_start or (time_end - dt.timedelta(days=1))
t0 = time_start.strftime('%Y-%m-%dT%H:%M:%SZ')
t1 = time_end.strftime('%Y-%m-%dT%H:%M:%SZ')
glider_id = glider_id or None
e = ERDDAP(server='NGDAC')
# Grab every dataset available
# datasets = pd.read_csv(e.get_search_url(response='csv', search_for='all'))
# Search constraints
kw = dict()
kw['min_time'] = t0
kw['max_time'] = t1
if bbox:
kw['min_lon'] = bbox[0]
kw['max_lon'] = bbox[1]
kw['min_lat'] = bbox[2]
kw['max_lat'] = bbox[3]
if glider_id:
search = glider_id
else:
search = None
search_url = e.get_search_url(search_for=search, response='csv', **kw)
try:
# Grab the results
search = pd.read_csv(search_url)
except:
# return empty dataframe if there are no results
return pd.DataFrame()
# Extract the IDs
gliders = search['Dataset ID'].values
msg = 'Found {} Glider Datasets:\n\n{}'.format
print(msg(len(gliders), '\n'.join(gliders)))
# Setting constraints
constraints = {
'time>=': t0,
'time<=': t1,
'longitude>=': bbox[0],
'longitude<=': bbox[1],
'latitude>=': bbox[2],
'latitude<=': bbox[3],
}
variables = [
'depth',
'latitude',
'longitude',
'time',
'temperature',
'salinity',
]
e = ERDDAP(
server='NGDAC',
protocol='tabledap',
response='nc'
)
glider_dfs = []
for id in gliders:
# print('Reading ' + id)
e.dataset_id = id
e.constraints = constraints
e.variables = variables
# checking data frame is not empty
try:
df = e.to_pandas(
index_col='time (UTC)',
parse_dates=True,
skiprows=(1,) # units information can be dropped.
).dropna()
except:
continue
df = df.reset_index()
df['dataset_id'] = id
df = df.set_index(['dataset_id', 'time (UTC)'])
glider_dfs.append(df)
try:
ndf = pd.concat(glider_dfs)
except ValueError:
return pd.DataFrame()
return ndf
def read_glider_variables_erddap_server(url_erddap,dataset_id,\
lat_lim,lon_lim,\
variable_names=['time'],
**kwargs):
"""
Created on Tue Nov 3 11:26:05 2020
@author: aristizabal
This function reads glider variables from the IOOS
and Rutgers erddapp glider servers.
Inputs:
url_erddap: url address of erddap server
Example: 'https://data.ioos.us/gliders/erddap'
dataset_id: Example: 'ng231-20190901T0000'
variable_names: list of variable names.
Example:
variable_names = ['depth',
'latitude',
'longitude',
'time',
'temperature',
'salinity']
The default value is variable_names=['time']
lat_lim: latitude limits for the search.
Example, lat_lim = [38.0,40.0]
lon_lim: longitude limits for the search.
Example, lon_lim = [-75.0,-72.0]
date_ini: initial date of time window.
This function accepts the data formats '%Y-%m-%d T %H:%M:%S Z' and '%Y/%m/%d/%H'.
Examaple: date_ini = '2018-08-02T00:00:00Z' or '2018/08/02/00'
date_end: initial date of time window.
This function uses the data format '%Y-%m-%d T %H:%M:%S Z'.
Examaple: date_ini = '2018-08-10T00:00:00Z' and '2018/08/10/00'
Outputs:
df: Pandas data frame with all the variables requested as vectors
"""
from erddapy import ERDDAP
import numpy as np
date_ini = kwargs.get('date_ini', None)
date_end = kwargs.get('date_end', None)
# Find time window of interest
if np.logical_or(date_ini == None, date_end == None):
constraints = {
'latitude>=': lat_lim[0],
'latitude<=': lat_lim[1],
'longitude>=': lon_lim[0],
'longitude<=': lon_lim[1],
}
else:
constraints = {
'time>=': date_ini,
'time<=': date_end,
'latitude>=': lat_lim[0],
'latitude<=': lat_lim[1],
'longitude>=': lon_lim[0],
'longitude<=': lon_lim[1],
}
e = ERDDAP(server=url_erddap, protocol='tabledap', response='nc')
e.dataset_id = dataset_id
e.constraints = constraints
e.variables = variable_names
# Converting glider data to data frame
# Cheching that data frame has data
df = e.to_pandas()
if len(df) > 3:
df = e.to_pandas(parse_dates=True)
return df
### specifying the database name we need the data from.
e.dataset_id = 'wocecpr'
### specifying the data constraints
e.constraints = {
'time>=': '2000-01-15T01:24:00Z',
'time<=': '2010-01-17T13:39:00Z',
'latitude>=': 37.0,
'latitude<=': 43.43,
'longitude>=': 317.56,
'longitude<=': 322.87,
}
###specifying the variables(columns name) to be retrived.
e.variables = [
'sample',
'latitude',
'longitude',
'life_stage',
'abundance',
'time',
]
### searching for the server link and doing the handshaking process.
search_url = e.get_search_url(response='csv')
### receiving requested data and saving it into a dataframe
search = pd.read_csv(search_url)
df = e.to_pandas()
### receiving current working directory and saving the dataframe into a single csv file in that path.
wd = os.getcwd()
df.to_csv(wd + '/DataFiles/plankton_swocecpr.csv')
#%%
"""
#########################################################################################################################
def get_ndbc(bbox=None, time_start=None, time_end=None, buoy=None):
bbox = bbox or [-100, -45, 5, 46]
time_end = time_end or dt.date.today()
time_start = time_start or (time_end - dt.timedelta(days=1))
buoy = buoy or False
time_formatter = '%Y-%m-%dT%H:%M:%SZ'
e = ERDDAP(
server='CSWC',
protocol='tabledap',
response='csv'
)
e.dataset_id = 'cwwcNDBCMet'
e.constraints = {
'time>=': time_start.strftime(time_formatter),
'time<=': time_end.strftime(time_formatter),
}
if bbox:
e.constraints['longitude>='] = bbox[0]
e.constraints['longitude<='] = bbox[1]
e.constraints['latitude>='] = bbox[2]
e.constraints['latitude<='] = bbox[3]
e.variables = [
"station",
"latitude",
"longitude",
"time"
]
if buoy:
e.constraints['station='] = buoy
df = e.to_pandas(
parse_dates=['time (UTC)'],
skiprows=(1,) # units information can be dropped.
).dropna()
stations = df.station.unique()
# e.variables = [
# "station",
# "latitude",
# "longitude",
# "wd",
# "wspd",
# "gst",
# "wvht",
# "dpd",
# "apd",
# "mwd",
# "bar",
# "atmp",
# "wtmp",
# "dewp",
# # "vis",
# # "ptdy",
# # "tide",
# "wspu",
# "wspv",
# "time",
# ]
try:
df = e.to_pandas(
parse_dates=['time (UTC)'],
skiprows=(1,) # units information can be dropped.
).dropna()
except HTTPError:
df = pd.DataFrame()
return df
if fname.is_file():
data = pd.read_csv(fname, parse_dates=["time (UTC)"])
else:
e = ERDDAP(
server="http://erddap.aoos.org/erddap/",
protocol="tabledap"
)
e.dataset_id = "kotzebue-alaska-water-level"
e.constraints = {
"time>=": "2018-09-05T21:00:00Z",
"time<=": "2019-07-10T19:00:00Z",
}
e.variables = [
variable_name,
"time",
"z",
]
data = e.to_pandas(
index_col="time (UTC)",
parse_dates=True,
)
data["timestamp"] = data.index.astype("int64") // 1e9
data.to_csv(fname)
data.head()
from ioos_qc.config import QcConfig
qc = QcConfig(qc_config)
def read_glider_data_erddap_Rutgers_server(url_erddap,dataset_id,\
lat_lim,lon_lim,scatter_plot,**kwargs):
from erddapy import ERDDAP
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
import cmocean
import numpy as np
date_ini = kwargs.get('date_ini', None)
date_end = kwargs.get('date_end', None)
# Find time window of interest
if np.logical_or(date_ini == None, date_end == None):
constraints = {
'latitude>=': lat_lim[0],
'latitude<=': lat_lim[1],
'longitude>=': lon_lim[0],
'longitude<=': lon_lim[1],
}
else:
constraints = {
'time>=': date_ini,
'time<=': date_end,
'latitude>=': lat_lim[0],
'latitude<=': lat_lim[1],
'longitude>=': lon_lim[0],
'longitude<=': lon_lim[1],
}
variables = [
'depth', 'latitude', 'longitude', 'time', 'temperature', 'salinity'
]
e = ERDDAP(server=url_erddap, protocol='tabledap', response='nc')
e.dataset_id = dataset_id
e.constraints = constraints
e.variables = variables
# Converting glider data to data frame
# Cheching that data frame has data
df = e.to_pandas()
if len(df) != 0:
df = e.to_pandas(
index_col='time (UTC)',
parse_dates=True,
skiprows=(1, ) # units information can be dropped.
).dropna()
dg = df['depth (m)'].values
tg = df.index.values
vg1 = df[df.columns[3]].values
vg2 = df[df.columns[4]].values
upcast = np.where(np.diff(dg) < 0)[0]
oku = np.where(np.diff(upcast) > 1)[0]
end_upcast = upcast[oku]
downcast = np.where(np.diff(dg) > 0)[0]
okd = np.where(np.diff(downcast) > 1)[0]
end_downcast = downcast[okd]
ind = np.hstack(
[0, np.unique(np.hstack([end_upcast, end_downcast])),
len(dg)])
zn = np.max(np.diff(ind))
depthg = np.empty((zn, len(ind)))
depthg[:] = np.nan
timeg = np.empty((zn, len(ind)))
timeg[:] = np.nan
tempg = np.empty((zn, len(ind)))
tempg[:] = np.nan
saltg = np.empty((zn, len(ind)))
saltg[:] = np.nan
for i in np.arange(len(ind)):
if i == 0:
indd = np.argsort(dg[ind[i]:ind[i + 1] + 2])
depthg[0:len(dg[ind[i]:ind[i + 1] + 2]),
i] = dg[ind[i]:ind[i + 1] + 2][indd]
timeg[0:len(dg[ind[i]:ind[i + 1] + 2]),
i] = mdates.date2num(tg[ind[i]:ind[i + 1] + 2][indd])
tempg[0:len(vg1[ind[i]:ind[i + 1] + 2]),
i] = vg1[ind[i]:ind[i + 1] + 2][indd]
saltg[0:len(vg2[ind[i]:ind[i + 1] + 2]),
i] = vg2[ind[i]:ind[i + 1] + 2][indd]
if i < len(ind) - 1:
indd = np.argsort(dg[ind[i] + 1:ind[i + 1] + 2])
depthg[0:len(dg[ind[i] + 1:ind[i + 1] + 2]),
i] = dg[ind[i] + 1:ind[i + 1] + 2][indd]
timeg[0:len(dg[ind[i] + 1:ind[i + 1] + 2]),
i] = mdates.date2num(tg[ind[i] + 1:ind[i + 1] + 2][indd])
tempg[0:len(vg1[ind[i] + 1:ind[i + 1] + 2]),
i] = vg1[ind[i] + 1:ind[i + 1] + 2][indd]
saltg[0:len(vg2[ind[i] + 1:ind[i + 1] + 2]),
i] = vg2[ind[i] + 1:ind[i + 1] + 2][indd]
else:
indd = np.argsort(dg[ind[i] + 1:len(dg)])
depthg[0:len(dg[ind[i] + 1:len(dg)]),
i] = dg[ind[i] + 1:len(dg)][indd]
timeg[0:len(dg[ind[i] + 1:len(dg)]),
i] = mdates.date2num(tg[ind[i] + 1:len(dg)][indd])
tempg[0:len(vg1[ind[i] + 1:len(vg1)]),
i] = vg1[ind[i] + 1:len(vg1)][indd]
saltg[0:len(vg2[ind[i] + 1:len(vg2)]),
i] = vg2[ind[i] + 1:len(vg2)][indd]
# Scatter plot
if scatter_plot == 'yes':
color_map = cmocean.cm.thermal
varg = tempg
#timeg_matrix = np.tile(timeg.T,(depthg.shape[0],1))
ttg = np.ravel(timeg)
dg = np.ravel(depthg)
teg = np.ravel(varg)
kw = dict(c=teg, marker='*', edgecolor='none')
fig, ax = plt.subplots(figsize=(10, 3))
cs = ax.scatter(ttg, -dg, cmap=color_map, **kw)
#fig.colorbar(cs)
ax.set_xlim(np.nanmin(ttg), np.nanmax(ttg))
ax.set_ylabel('Depth (m)', fontsize=14)
cbar = plt.colorbar(cs)
cbar.ax.set_ylabel('Temperature ($^oC$)', fontsize=14)
ax.set_title(dataset_id, fontsize=16)
xfmt = mdates.DateFormatter('%H:%Mh\n%d-%b')
ax.xaxis.set_major_formatter(xfmt)
plt.ylim([-np.nanmax(dg), 0])
color_map = cmocean.cm.haline
varg = saltg
#timeg_matrix = np.tile(timeg.T,(depthg.shape[0],1))
ttg = np.ravel(timeg)
dg = np.ravel(depthg)
teg = np.ravel(varg)
kw = dict(c=teg, marker='*', edgecolor='none')
fig, ax = plt.subplots(figsize=(10, 3))
cs = ax.scatter(ttg, -dg, cmap=color_map, **kw)
#fig.colorbar(cs)
ax.set_xlim(np.nanmin(ttg), np.nanmax(ttg))
ax.set_ylabel('Depth (m)', fontsize=14)
cbar = plt.colorbar(cs)
cbar.ax.set_ylabel('Salinity', fontsize=14)
ax.set_title(dataset_id, fontsize=16)
xfmt = mdates.DateFormatter('%H:%Mh\n%d-%b')
ax.xaxis.set_major_formatter(xfmt)
plt.ylim([-np.nanmax(dg), 0])
return tempg, saltg, timeg, latg, long, depthg
protocol='tabledap',
)
e.dataset_id = 'IWBNetwork'
e.constraints = {
'time>=': '2015-06-28T00:00:00Z',
'station_id=': 'M3'
}
e.variables = [
'time',
'AtmosphericPressure',
'WindDirection',
'WindSpeed',
'WaveHeight',
'WavePeriod',
'MeanWaveDirection',
# 'Hmax',
# 'AirTemperature',
'SeaTemperature'
]
url = e.get_download_url()
print(url)
df = e.to_pandas(
index_col='time (UTC)',
parse_dates=True).dropna()
def GOFS_RTOFS_vs_Argo_floats(lon_forec_track, lat_forec_track, lon_forec_cone,
lat_forec_cone, lon_best_track, lat_best_track,
lon_lim, lat_lim, folder_fig):
#%% User input
#GOFS3.1 output model location
url_GOFS_ts = 'http://tds.hycom.org/thredds/dodsC/GLBy0.08/expt_93.0/ts3z'
# RTOFS files
folder_RTOFS = '/home/coolgroup/RTOFS/forecasts/domains/hurricanes/RTOFS_6hourly_North_Atlantic/'
nc_files_RTOFS = ['rtofs_glo_3dz_f006_6hrly_hvr_US_east.nc',\
'rtofs_glo_3dz_f012_6hrly_hvr_US_east.nc',\
'rtofs_glo_3dz_f018_6hrly_hvr_US_east.nc',\
'rtofs_glo_3dz_f024_6hrly_hvr_US_east.nc']
# COPERNICUS MARINE ENVIRONMENT MONITORING SERVICE (CMEMS)
url_cmems = 'http://nrt.cmems-du.eu/motu-web/Motu'
service_id = 'GLOBAL_ANALYSIS_FORECAST_PHY_001_024-TDS'
product_id = 'global-analysis-forecast-phy-001-024'
depth_min = '0.493'
out_dir = '/home/aristizabal/crontab_jobs'
# Bathymetry file
#bath_file = '/Users/aristizabal/Desktop/MARACOOS_project/Maria_scripts/nc_files/GEBCO_2014_2D_-100.0_0.0_-60.0_45.0.nc'
bath_file = '/home/aristizabal/bathymetry_files/GEBCO_2014_2D_-100.0_0.0_-10.0_50.0.nc'
# Argo floats
url_Argo = 'http://www.ifremer.fr/erddap'
#%%
from matplotlib import pyplot as plt
import numpy as np
import xarray as xr
import netCDF4
from datetime import datetime, timedelta
import cmocean
import matplotlib.dates as mdates
from erddapy import ERDDAP
import pandas as pd
import os
# Do not produce figures on screen
plt.switch_backend('agg')
# Increase fontsize of labels globally
plt.rc('xtick', labelsize=14)
plt.rc('ytick', labelsize=14)
plt.rc('legend', fontsize=14)
#%% Reading bathymetry data
ncbath = xr.open_dataset(bath_file)
bath_lat = ncbath.variables['lat'][:]
bath_lon = ncbath.variables['lon'][:]
bath_elev = ncbath.variables['elevation'][:]
oklatbath = np.logical_and(bath_lat >= lat_lim[0], bath_lat <= lat_lim[-1])
oklonbath = np.logical_and(bath_lon >= lon_lim[0], bath_lon <= lon_lim[-1])
bath_latsub = bath_lat[oklatbath]
bath_lonsub = bath_lon[oklonbath]
bath_elevs = bath_elev[oklatbath, :]
bath_elevsub = bath_elevs[:, oklonbath]
#%% Get time bounds for current day
#ti = datetime.today()
ti = datetime.today() - timedelta(1) - timedelta(hours=6)
tini = datetime(ti.year, ti.month, ti.day)
te = ti + timedelta(2)
tend = datetime(te.year, te.month, te.day)
#%% Look for Argo datasets
e = ERDDAP(server=url_Argo)
# Grab every dataset available
#datasets = pd.read_csv(e.get_search_url(response='csv', search_for='all'))
kw = {
'min_lon': lon_lim[0],
'max_lon': lon_lim[1],
'min_lat': lat_lim[0],
'max_lat': lat_lim[1],
'min_time': str(tini),
'max_time': str(tend),
}
search_url = e.get_search_url(response='csv', **kw)
# Grab the results
search = pd.read_csv(search_url)
# Extract the IDs
dataset = search['Dataset ID'].values
msg = 'Found {} Datasets:\n\n{}'.format
print(msg(len(dataset), '\n'.join(dataset)))
dataset_type = dataset[0]
constraints = {
'time>=': str(tini),
'time<=': str(tend),
'latitude>=': lat_lim[0],
'latitude<=': lat_lim[1],
'longitude>=': lon_lim[0],
'longitude<=': lon_lim[1],
}
variables = [
'platform_number',
'time',
'pres',
'longitude',
'latitude',
'temp',
'psal',
]
e = ERDDAP(server=url_Argo, protocol='tabledap', response='nc')
e.dataset_id = dataset_type
e.constraints = constraints
e.variables = variables
print(e.get_download_url())
df = e.to_pandas(
parse_dates=True,
skiprows=(1, ) # units information can be dropped.
).dropna()
argo_ids = np.asarray(df['platform_number'])
argo_times = np.asarray(df['time (UTC)'])
argo_press = np.asarray(df['pres (decibar)'])
argo_lons = np.asarray(df['longitude (degrees_east)'])
argo_lats = np.asarray(df['latitude (degrees_north)'])
argo_temps = np.asarray(df['temp (degree_Celsius)'])
argo_salts = np.asarray(df['psal (PSU)'])
#%% GOGF 3.1
try:
GOFS_ts = xr.open_dataset(url_GOFS_ts, decode_times=False)
lt_GOFS = np.asarray(GOFS_ts['lat'][:])
ln_GOFS = np.asarray(GOFS_ts['lon'][:])
tt = GOFS_ts['time']
t_GOFS = netCDF4.num2date(tt[:], tt.units)
depth_GOFS = np.asarray(GOFS_ts['depth'][:])
except Exception as err:
print(err)
GOFS_ts = np.nan
lt_GOFS = np.nan
ln_GOFS = np.nan
depth_GOFS = np.nan
t_GOFS = ti
#%% Map Argo floats
lev = np.arange(-9000, 9100, 100)
plt.figure()
plt.contourf(bath_lonsub,
bath_latsub,
bath_elevsub,
lev,
cmap=cmocean.cm.topo)
plt.plot(lon_forec_track, lat_forec_track, '.-', color='gold')
plt.plot(lon_forec_cone, lat_forec_cone, '.-b', markersize=1)
plt.plot(lon_best_track, lat_best_track, 'or', markersize=3)
argo_idd = np.unique(argo_ids)
for i, id in enumerate(argo_idd):
okind = np.where(argo_ids == id)[0]
plt.plot(np.unique(argo_lons[okind]),
np.unique(argo_lats[okind]),
's',
color='darkorange',
markersize=5,
markeredgecolor='k')
plt.title('Argo Floats ' + str(tini)[0:13] + '-' + str(tend)[0:13],
fontsize=16)
plt.axis('scaled')
plt.xlim(lon_lim[0], lon_lim[1])
plt.ylim(lat_lim[0], lat_lim[1])
file = folder_fig + 'ARGO_lat_lon'
#file = folder_fig + 'ARGO_lat_lon_' + str(np.unique(argo_times)[0])[0:10]
plt.savefig(file, bbox_inches='tight', pad_inches=0.1)
#%% Figure argo float vs GOFS and vs RTOFS
argo_idd = np.unique(argo_ids)
for i, id in enumerate(argo_idd):
print(id)
okind = np.where(argo_ids == id)[0]
argo_time = np.asarray([
datetime.strptime(t, '%Y-%m-%dT%H:%M:%SZ')
for t in argo_times[okind]
])
argo_lon = argo_lons[okind]
argo_lat = argo_lats[okind]
argo_pres = argo_press[okind]
argo_temp = argo_temps[okind]
argo_salt = argo_salts[okind]
# GOFS
print('Retrieving variables from GOFS')
if isinstance(GOFS_ts, float):
temp_GOFS = np.nan
salt_GOFS = np.nan
else:
#oktt_GOFS = np.where(t_GOFS >= argo_time[0])[0][0]
ttGOFS = np.asarray([
datetime(t_GOFS[i].year, t_GOFS[i].month, t_GOFS[i].day,
t_GOFS[i].hour) for i in np.arange(len(t_GOFS))
])
tstamp_GOFS = [
mdates.date2num(ttGOFS[i]) for i in np.arange(len(ttGOFS))
]
oktt_GOFS = np.unique(
np.round(
np.interp(mdates.date2num(argo_time[0]), tstamp_GOFS,
np.arange(len(tstamp_GOFS)))).astype(int))[0]
oklat_GOFS = np.where(lt_GOFS >= argo_lat[0])[0][0]
oklon_GOFS = np.where(ln_GOFS >= argo_lon[0] + 360)[0][0]
temp_GOFS = np.asarray(GOFS_ts['water_temp'][oktt_GOFS, :,
oklat_GOFS,
oklon_GOFS])
salt_GOFS = np.asarray(GOFS_ts['salinity'][oktt_GOFS, :,
oklat_GOFS, oklon_GOFS])
# RTOFS
#Time window
year = int(argo_time[0].year)
month = int(argo_time[0].month)
day = int(argo_time[0].day)
tini = datetime(year, month, day)
tend = tini + timedelta(days=1)
# Read RTOFS grid and time
print('Retrieving coordinates from RTOFS')
if tini.month < 10:
if tini.day < 10:
fol = 'rtofs.' + str(tini.year) + '0' + str(
tini.month) + '0' + str(tini.day)
else:
fol = 'rtofs.' + str(tini.year) + '0' + str(tini.month) + str(
tini.day)
else:
if tini.day < 10:
fol = 'rtofs.' + str(tini.year) + str(tini.month) + '0' + str(
tini.day)
else:
fol = 'rtofs.' + str(tini.year) + str(tini.month) + str(
tini.day)
ncRTOFS = xr.open_dataset(folder_RTOFS + fol + '/' + nc_files_RTOFS[0])
latRTOFS = np.asarray(ncRTOFS.Latitude[:])
lonRTOFS = np.asarray(ncRTOFS.Longitude[:])
depth_RTOFS = np.asarray(ncRTOFS.Depth[:])
tRTOFS = []
for t in np.arange(len(nc_files_RTOFS)):
ncRTOFS = xr.open_dataset(folder_RTOFS + fol + '/' +
nc_files_RTOFS[t])
tRTOFS.append(np.asarray(ncRTOFS.MT[:])[0])
tRTOFS = np.asarray([mdates.num2date(mdates.date2num(tRTOFS[t])) \
for t in np.arange(len(nc_files_RTOFS))])
oktt_RTOFS = np.where(
mdates.date2num(tRTOFS) >= mdates.date2num(argo_time[0]))[0][0]
oklat_RTOFS = np.where(latRTOFS[:, 0] >= argo_lat[0])[0][0]
oklon_RTOFS = np.where(lonRTOFS[0, :] >= argo_lon[0])[0][0]
nc_file = folder_RTOFS + fol + '/' + nc_files_RTOFS[oktt_RTOFS]
ncRTOFS = xr.open_dataset(nc_file)
#time_RTOFS = tRTOFS[oktt_RTOFS]
temp_RTOFS = np.asarray(ncRTOFS.variables['temperature'][0, :,
oklat_RTOFS,
oklon_RTOFS])
salt_RTOFS = np.asarray(ncRTOFS.variables['salinity'][0, :,
oklat_RTOFS,
oklon_RTOFS])
#lon_RTOFS = lonRTOFS[0,oklon_RTOFS]
#lat_RTOFS = latRTOFS[oklat_RTOFS,0]
# Downloading and reading Copernicus output
motuc = 'python -m motuclient --motu ' + url_cmems + \
' --service-id ' + service_id + \
' --product-id ' + product_id + \
' --longitude-min ' + str(argo_lon[0]-2/12) + \
' --longitude-max ' + str(argo_lon[0]+2/12) + \
' --latitude-min ' + str(argo_lat[0]-2/12) + \
' --latitude-max ' + str(argo_lat[0]+2/12) + \
' --date-min ' + '"' + str(tini-timedelta(0.5)) + '"' + \
' --date-max ' + '"' + str(tend+timedelta(0.5)) + '"' + \
' --depth-min ' + depth_min + \
' --depth-max ' + str(np.nanmax(argo_pres)+1000) + \
' --variable ' + 'thetao' + ' ' + \
' --variable ' + 'so' + ' ' + \
' --out-dir ' + out_dir + \
' --out-name ' + str(id) + '.nc' + ' ' + \
' --user ' + 'maristizabalvar' + ' ' + \
' --pwd ' + 'MariaCMEMS2018'
os.system(motuc)
# Check if file was downloaded
COP_file = out_dir + '/' + str(id) + '.nc'
# Check if file was downloaded
resp = os.system('ls ' + out_dir + '/' + str(id) + '.nc')
if resp == 0:
COP = xr.open_dataset(COP_file)
latCOP = np.asarray(COP.latitude[:])
lonCOP = np.asarray(COP.longitude[:])
depth_COP = np.asarray(COP.depth[:])
tCOP = np.asarray(mdates.num2date(mdates.date2num(COP.time[:])))
else:
latCOP = np.empty(1)
latCOP[:] = np.nan
lonCOP = np.empty(1)
lonCOP[:] = np.nan
tCOP = np.empty(1)
tCOP[:] = np.nan
oktimeCOP = np.where(
mdates.date2num(tCOP) >= mdates.date2num(tini))[0][0]
oklonCOP = np.where(lonCOP >= argo_lon[0])[0][0]
oklatCOP = np.where(latCOP >= argo_lat[0])[0][0]
temp_COP = np.asarray(COP.variables['thetao'][oktimeCOP, :, oklatCOP,
oklonCOP])
salt_COP = np.asarray(COP.variables['so'][oktimeCOP, :, oklatCOP,
oklonCOP])
# Figure temp
plt.figure(figsize=(5, 6))
plt.plot(argo_temp,
-argo_pres,
'.-',
linewidth=2,
label='ARGO Float id ' + str(id))
plt.plot(temp_GOFS,
-depth_GOFS,
'.-',
linewidth=2,
label='GOFS 3.1',
color='red')
plt.plot(temp_RTOFS,
-depth_RTOFS,
'.-',
linewidth=2,
label='RTOFS',
color='g')
plt.plot(temp_COP,
-depth_COP,
'.-',
linewidth=2,
label='Copernicus',
color='darkorchid')
plt.ylim([-1000, 0])
plt.title('Temperature Profile on '+ str(argo_time[0])[0:13] +
'\n [lon,lat] = [' \
+ str(np.round(argo_lon[0],3)) +',' +\
str(np.round(argo_lat[0],3))+']',\
fontsize=16)
plt.ylabel('Depth (m)', fontsize=14)
plt.xlabel('$^oC$', fontsize=14)
plt.legend(loc='lower right', fontsize=14)
file = folder_fig + 'ARGO_vs_GOFS_RTOFS_COP_temp_' + str(id)
plt.savefig(file, bbox_inches='tight', pad_inches=0.1)
# Figure salt
plt.figure(figsize=(5, 6))
plt.plot(argo_salt,
-argo_pres,
'.-',
linewidth=2,
label='ARGO Float id ' + str(id))
plt.plot(salt_GOFS,
-depth_GOFS,
'.-',
linewidth=2,
label='GOFS 3.1',
color='red')
plt.plot(salt_RTOFS,
-depth_RTOFS,
'.-',
linewidth=2,
label='RTOFS',
color='g')
plt.plot(salt_COP,
-depth_COP,
'.-',
linewidth=2,
label='Copernicus',
color='darkorchid')
plt.ylim([-1000, 0])
plt.title('Salinity Profile on '+ str(argo_time[0])[0:13] +
'\n [lon,lat] = [' \
+ str(np.round(argo_lon[0],3)) +',' +\
str(np.round(argo_lat[0],3))+']',\
fontsize=16)
plt.ylabel('Depth (m)', fontsize=14)
plt.legend(loc='lower right', fontsize=14)
file = folder_fig + 'ARGO_vs_GOFS_RTOFS_COP_salt_' + str(id)
plt.savefig(file, bbox_inches='tight', pad_inches=0.1)
- Met: Total number of met messages released to the GTS
- Wave: Total number of wave messages released to the GTS
In this notebook we will explore the statistics of the messages IOOS is releasing to GTS.
The first step is to download the data. We will use an ERDDAP server that [hosts the CSV files](https://ferret.pmel.noaa.gov/generic/erddap/files/ioos_obs_counts/) with the ingest data.
from datetime import date
from erddapy import ERDDAP
server = "http://osmc.noaa.gov/erddap"
e = ERDDAP(server=server, protocol="tabledap")
e.dataset_id = "ioos_obs_counts"
e.variables = ["time", "locationID", "region", "sponsor", "met", "wave"]
e.constraints = {
"time>=": "2019-09",
"time<": "2020-11",
}
df = e.to_pandas(parse_dates=True)
df["locationID"] = df["locationID"].str.lower()
df.tail()
The table has all the ingest data from 2019-01-01 to 2020-06-01. We can now explore it grouping the data by IOOS Regional Association (RA).
groups = df.groupby("region")
info_url = erd.get_info_url(response='csv') info_df = to_df(info_url) info_df info_df[info_df['Row Type'] == 'variable'] # Take a look at the variables with standard names: variables = erd.get_var_by_attr(standard_name=lambda v: v is not None) variables # These are the standard variables for the CTDBP instrument - specifically for the CP01CNSM-NSIF-CTDBP. Next, lets query the server for _all_ available data from the CP01CNSM-NSIF-CTDBP. erd.variables = variables erd.get_download_url() # Put it all into a dataframe: data = erd.to_pandas() # + # Plot a basic time-series of the conductivity import matplotlib.pyplot as plt import seaborn as sns sns.set(style="darkgrid") # -
protocol='tabledap',
)
e.response = 'csv'
e.dataset_id = 'OSMC_30day'
e.constraints = {
'time>=': str(x.tm_year)+"-"+str(x.tm_mon).zfill(2)+"-"+str(x.tm_mday).zfill(2)+"T"+str(x.tm_hour).zfill(2)+":00:00Z",
'longitude>=': -80.0,
'longitude<=': 80.0,
'platform_type=': "DRIFTING BUOYS (GENERIC)",
'platform_code=': str(wmo_mb[i]),
}
e.variables = [
'platform_code',
'time',
'latitude',
'longitude',
'sst',
'slp',
]
try:
df = e.to_pandas()
except:
print("Não há dados para o WMO "+str(wmo_mb[i]))
try:
df.columns = ['id', 'tempo','lat','lon','sst','pres']
df.id=sat[i]
dateparse = lambda x: pd.datetime.strptime(x, '%Y-%m-%dT%H:%M:%SZ')
igrid = 1
#%% Reading bathymetry data
ncbath = xr.open_dataset(bath_file)
bath_lat = ncbath.variables['lat'][:]
bath_lon = ncbath.variables['lon'][:]
bath_elev = ncbath.variables['elevation'][:]
#%% Looping through all gliders found
for id in gliders:
print('Reading ' + id )
e.dataset_id = id
e.constraints = constraints
e.variables = variables
# chacking data frame is not empty
df = e.to_pandas()
if len(df.index) != 0 :
# Converting glider data to data frame
df = e.to_pandas(
index_col='time (UTC)',
parse_dates=True,
skiprows=(1,) # units information can be dropped.
).dropna()
# Coverting glider vectors into arrays
timeg, ind = np.unique(df.index.values,return_index=True)
latg = df['latitude (degrees_north)'].values[ind]
if args.erddap:
drifter_years = args.erddap[1:]
argos_id = args.erddap[0]
e = ERDDAP(
server='http://akutan.pmel.noaa.gov:8080/erddap',
protocol='tabledap',
)
e.response = 'csv'
#e.dataset_id = drifter_year + '_Argos_Drifters_NRT'
#use this until we can get location quality back into older years
#currently it is only in erddap for 2020 and newer
#if int(drifter_years[0]) >= 2020:
e.variables = [
'trajectory_id', 'strain', 'voltage', 'time', 'latitude', 'sst',
'longitude', 'location_quality'
]
#else:
# e.variables = ['trajectory_id','strain', 'voltage', 'time', 'latitude', 'sst',
# 'longitude']
e.constraints = {'trajectory_id=': argos_id}
df_years = {}
for year in drifter_years:
e.dataset_id = year + '_Argos_Drifters_NRT'
df = e.to_pandas(
index_col='time (UTC)',
parse_dates=True,
skiprows=(1, ) # units information can be dropped.
)
df.columns = [x[1].split()[0] for x in enumerate(df.columns)]
