Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

refactor(get_iv; get_all_ivs ): make parameter code handling more general for ivs #10

Open
wants to merge 1 commit into
base: develop
Choose a base branch
from
Open

refactor(get_iv; get_all_ivs ): make parameter code handling more general for ivs #10

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Binary file modified examples/data/bbox.dbf
View file
Open in desktop
Binary file not shown.
Binary file modified examples/data/bbox.shp
View file
Open in desktop
Binary file not shown.
Binary file modified examples/data/bbox.shx
View file
Open in desktop
Binary file not shown.
125 changes: 107 additions & 18 deletions pydrograph/nwis.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,7 +45,9 @@ class Nwis:
'inventory': 'inventory?'}

parameter_codes = {'discharge': '00060',
'gwlevels': '72019'}
'gwlevels': '72019',
'gageheight': '00065 ',
'gwelev':'62611'}

coordinate_format = 'decimal_degrees' #coordinate_format=decimal_degrees&
group_key = 'NONE' #group_key=NONE&
Expand Down Expand Up @@ -373,6 +375,42 @@ def get_datetime_retrieved(self, sitefile_text):
elif '#' not in str(line):
return None

def get_par_description(self, sitefile_text, parameter_code, data_type = 'iv'):
"""helper method to extract parameter description and methods from
sitefile text.

Args:
sitefile_text (list): list of lines from url, generated during the
urlopen(url).readlines() call in get_ivs()
parameter_code (string): NWIS parameter code, e.g. 00060 for
discharge. See:
http://help.waterdata.usgs.gov/codes-and-parameters/parameters.
data_type (string): iv or dv, type of sitefile text retrieved

Returns:
string: parameter description and units from sitefile text
"""
if data_type == 'iv':
for i, line in enumerate(sitefile_text):
if 'Parameter Description' in str(line):
description_line = str(sitefile_text[i+1])
description = description_line.rpartition(parameter_code)[-1].lstrip().rstrip("\\n'")
break
elif data_type == 'dv':
for i, line in enumerate(sitefile_text):
if 'Description' in str(line):
description_line = str(sitefile_text[i+1])
description = ' '.join(description_line.split()[5:]).rstrip("\\n'")
break
else:
raise ValueError("data_type must be 'iv' or 'dv'")

try:
description
except:
description = parameter_code
return description

def get_siteinfo(self, data_type, attributes=None):
"""Retrieves site information for the bounding box supplied to the NWIS class instance

Expand Down Expand Up @@ -503,42 +541,55 @@ def get_iv_siteinfo(self, attributes = 'iv_attributes'):
return df


def get_ivs(self, station_ID, parameter_code='00060', start_date='2000-01-01', end_date='2000-12-31',
sample_period = 'D', agg_method = 'mean'):
"""Retrieves daily values for a site.
def get_ivs(self, station_ID, parameter_code='00060', start_date='2000-01-01',
end_date='2000-12-31', sample_period = 'D', agg_method = 'mean'):
"""Retrieves instantaneous values of a specified parameter type, for a
specified period of time, at a site. Data can be retrieved raw or
aggregated to a specified frequency using 'sample_period' and
'agg_method'. Data gaps are filled with NaNs.

Parameters
----------
stationID: (string)
USGS station ID

parameter_code: string, default is 00060 for discharge.
parameter_code: (string), default is 00060 for discharge.
See http://help.waterdata.usgs.gov/codes-and-parameters/parameters.

start_date: (string) 'YYYY-MM-DD'
To obtain the entire period-of-record use a start date of 2000-01-01 (default)...
To obtain the entire period-of-record use a start date of 2000-01-01
(default)...
end_date: (string) 'YYYY-MM-DD'
preset to take the year 2000, don't set the range too long (longer than a year or so) or the code will be very slow and may not finish running
preset to take the year 2000, don't set the range too long (longer
than a year or so) or the code will be very slow and may not finish
running
sample_period: (string), default 'D' for daily
change this string to how you would like the instantaneous values aggregated. this can be daily, weekly, monthly, etc.
None will give just the raw data which is generally in 15 minute increments
change this string to how you would like the instantaneous values
aggregated. this can be daily, weekly, monthly, etc. See
Pandas offset aliases:
https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#offset-aliases
A sample period of None will give just the raw data which is
generally in 15 minute increments.
agg_method: (string), default 'mean'
change this to chose how the aggregated instantaneous values are calculated. options include 'mean', 'median', 'max', etc.
change this to chose how the aggregated instantaneous values are
calculated. options include 'mean', 'median', 'max', etc.

Returns
-------
df: a datetime-index dataframe of daily discharge, with datagaps filled with NaNs, aggregated daily or however specified with 'sample_period'
df: a datetime-index dataframe of parameter data'
"""
if parameter_code in list(self.parameter_codes.keys()):
parameter_code = self.parameter_codes[parameter_code]

url = self.make_iv_url(station_ID, parameter_code=parameter_code,
start_date=start_date, end_date=end_date)
sitefile_text = urlopen(url).readlines()
par_description = self.get_par_description(sitefile_text=list(sitefile_text),
parameter_code=parameter_code)
skiprows = self.get_header_length(sitefile_text, 'agency_cd')
cols = sitefile_text[skiprows - 2].decode('utf-8').strip().split('\t')
loginfo = [str(station_ID), url, self.get_datetime_retrieved(sitefile_text)]
df = pd.read_csv(url, sep='\t', skiprows=skiprows, header=None, names=cols, dtype={'site_no': object})
df = pd.read_csv(url, sep='\t', skiprows=skiprows, header=None, names=cols,
dtype={'site_no': object})

if len(df) > 2:

Expand All @@ -552,9 +603,10 @@ def get_ivs(self, station_ID, parameter_code='00060', start_date='2000-01-01', e

if sample_period is not None:
df = df.resample(sample_period).agg(agg_method)
df = df.rename(columns = {df.columns[0]: 'discharge (cfs)'})
df = df.rename(columns = {df.columns[0]: f'{par_description}'})
else:
df = df.rename(columns = {df.columns[4]: 'discharge (cfs)', df.columns[5]: 'code'})
df = df.rename(columns = {df.columns[4]: f'{par_description}',
df.columns[5]: 'code'})

else:
print('No data at this site during this timeframe.')
Expand Down Expand Up @@ -643,14 +695,51 @@ def get_all_dvs(self, stations, parameter_code='00060', start_date='1880-01-01',
self.log = pd.DataFrame(columns=self.log_cols) # reset the log
return all_dvs

def get_all_ivs(self, stations, parameter_code='00060', start_date='2000-01-01', end_date='2000-12-31'):
''' This function gets all instantaneous values for a list of station IDs, and places them in a dictionary of dataframes
def get_all_ivs(self, stations, parameter_code='00060', start_date='2000-01-01',
end_date='2000-12-31', sample_period = 'D', agg_method = 'mean'):
'''Retrieves instantaneous values of a specified parameter type from a list
of sites, for a specified period of time. Data can be retrieved raw or
aggregated to a specified frequency using 'sample_period' and
'agg_method'. Data gaps are filled with NaNs. Data is returned as a
dictionary of dataframes.

Parameters
----------
stations: (list)
list of USGS station IDs as strings.
parameter_code: (string), default is 00060 for discharge.
See http://help.waterdata.usgs.gov/codes-and-parameters/parameters.
start_date: (string) 'YYYY-MM-DD'
To obtain the entire period-of-record use a start date of 2000-01-01
(default)...
end_date: (string) 'YYYY-MM-DD'
preset to take the year 2000, don't set the range too long (longer
than a year or so) or the code will be very slow and may not finish
running.
sample_period: (string), default 'D' for daily
change this string to how you would like the instantaneous values
aggregated. this can be daily, weekly, monthly, etc. See
Pandas offset aliases:
https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#offset-aliases
A sample period of None will give just the raw data which is
generally in 15 minute increments.
agg_method: (string), default 'mean'
change this to chose how the aggregated instantaneous values are
calculated. options include 'mean', 'median', 'max', etc.

Returns
-------
dictionary of dataframes. Dataframes have a datetime-index and a column
with parameter values.
'''

all_ivs = {}
for station in stations:
try:
df = self.get_ivs(station, parameter_code=parameter_code, start_date=start_date, end_date=end_date)
df = self.get_ivs(station, parameter_code=parameter_code,
start_date=start_date, end_date=end_date,
sample_period=sample_period,
agg_method=agg_method)
except Exception as e:
print(e)
continue
Expand Down
87 changes: 86 additions & 1 deletion pydrograph/tests/test_nwis.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,6 +48,44 @@ def test_instantaneous_value(nwis_instance):
assert 'site_no' in df.columns
assert df.site_no.dtype == np.object

def test_get_ivs_gw(nwis_instance):
# check groundwater site (depth to water)
gw_site = '353831090502401'
df = nwis_instance.get_ivs(gw_site,
parameter_code='72019',
start_date='2020-08-01',
end_date='2020-08-15')

assert df.columns[0] == 'Depth to water level, feet below land surface' # check column header/units
assert df.shape[0] == 15 # check daily resampling

# check without resampling
df = nwis_instance.get_ivs(gw_site,
parameter_code='72019',
start_date='2020-08-01',
end_date='2020-08-15',
sample_period=None)
assert df.shape[0] == 15 * 24 * 4 # 15 days at 15 min frequency

def test_get_ivs_sw(nwis_instance):
# check surface water site (discharge)
sw_site = '07047810'
df = nwis_instance.get_ivs(sw_site,
parameter_code='00060',
start_date='2002-08-01',
end_date='2002-08-15')

assert df.columns[0] == 'Discharge, cubic feet per second' # check column header/units
assert df.shape[0] == 15 # check daily resampling

# check without resampling
df = nwis_instance.get_ivs(sw_site,
parameter_code='00060',
start_date='2002-08-01',
end_date='2002-08-15',
sample_period=None)
assert df.shape[0] == 15 * 24 # 15 days at 1 hr frequency

def test_tuple_extent_no_data():

bbox = (-91.45793026894977, 47.2,
Expand All @@ -62,4 +100,51 @@ def test_get_all_ivs(nwis_instance, stations):
site_one = list(all_sites.values())[0]
assert len(all_sites) > 0
assert len(site_one) > 2
#assert all_sites.site_no.dtype == np.object
#assert all_sites.site_no.dtype == np.object

def test_get_all_ivs_sw(nwis_instance):
sites = ['07040450', '07047800']

# check set of surface water sites from nwis instnace
df_dct = nwis_instance.get_all_ivs(stations=sites,
start_date='2000-12-01',
end_date='2000-12-31')

for site, df in df_dct.items():
assert site in sites
assert df.shape[0] == 31 # aggregated daily
assert df.columns == 'Discharge, cubic feet per second'

# check without resampling
df_dct = nwis_instance.get_all_ivs(stations=sites,
start_date='2000-12-01',
end_date='2000-12-31',
sample_period=None)
for site, df in df_dct.items():
assert site in sites
assert 'datetime' in df.columns and 'Discharge, cubic feet per second' in df.columns
df.datetime[0] == '2000-12-01 00:00' # verify has hh:mm


def test_get_all_ivs_gw(nwis_instance):
sites = ['353606090510701', '353831090502401']

# check set of groundwater sites from nwis instnace
df_dct = nwis_instance.get_all_ivs(stations=sites,
parameter_code='72019',
start_date='2020-12-01',
end_date='2020-12-31')
for site, df in df_dct.items():
assert site in sites
assert df.shape[0] == 31 # aggregated daily
assert df.columns == 'Depth to water level, feet below land surface'

df_dct = nwis_instance.get_all_ivs(stations=sites,
parameter_code='72019',
start_date='2020-12-01',
end_date='2020-12-31',
sample_period=None)
for site, df in df_dct.items():
assert site in sites
assert 'datetime' in df.columns and 'Depth to water level, feet below land surface' in df.columns
df.datetime[0] == '2020-12-01 00:00' # verify has hh:mm