def setUp(self):
self.c = UsgsRest()
class USGSTest(unittest.TestCase):
def setUp(self):
self.c = UsgsRest()
def test_by_bbox(self):
self.c.filter(bbox=(-87, 46, -85, 48))
collection = self.c.collect()
collection.calculate_bounds()
# Returns 4 stations: 04044724, 04045500, 04046000, 04056500
assert len(collection.elements) == 4
assert sorted(map(lambda x: x.uid, collection.elements)) == [
"04044724", "04045500", "04046000", "04056500"
]
station = collection.elements[0]
assert station.name == "AU TRAIN RIVER AT FOREST LAKE, MI"
# Measures 2 variables
assert len(station.get_unique_members()) == 2
assert station.location.x == -86.8501514
assert station.location.y == 46.34077908
assert station.location.z == 0
# Apply time filter as well
starting = datetime(2013, 12, 31, 0, 0, 0)
ending = starting + timedelta(hours=6)
self.c.filter(start=starting, end=ending)
collection = self.c.collect()
collection.calculate_bounds()
# Returns 4 stations: 04044724, 04045500, 04046000, 04056500
assert len(collection.elements) == 4
assert sorted(map(lambda x: x.uid, collection.elements)) == [
"04044724", "04045500", "04046000", "04056500"
]
def test_by_state(self):
# Clear filters
self.c.clear()
# Add custom state filter
self.c.filter(state="ri")
collection = self.c.collect()
# Returns 41 stations
assert len(collection.elements) == 41
station = collection.elements[0]
assert station.name == "TEN MILE R., PAWTUCKET AVE. AT E. PROVIDENCE, RI"
# Measures 2 variables
assert len(station.get_unique_members()) == 2
assert station.location.x == -71.3511649
assert station.location.y == 41.83093376
assert station.location.z == 0
def test_by_site_code(self):
# Clear filters
self.c.clear()
# Add custom state filter
self.c.filter(features=['04001000'])
collection = self.c.collect()
collection.calculate_bounds()
station = collection.elements[0]
assert station.uid == '04001000'
assert station.name == 'WASHINGTON CREEK AT WINDIGO, MI'
assert station.location.x == -89.1459196999999932
assert station.location.y == 47.9212792000000007
assert station.location.z == 0
def setUp(self):
self.c = UsgsRest()
class USGSTest(unittest.TestCase):
def setUp(self):
self.c = UsgsRest()
def test_by_bbox(self):
self.c.filter(bbox=(-87,46,-85,48))
collection = self.c.collect()
collection.calculate_bounds()
# Returns 4 stations: 04044724, 04045500, 04046000, 04056500
assert len(collection.elements) == 4
assert sorted(map(lambda x: x.uid, collection.elements)) == ["04044724", "04045500", "04046000", "04056500"]
station = collection.elements[0]
assert station.name == "AU TRAIN RIVER AT FOREST LAKE, MI"
# Measures 2 variables
assert len(station.get_unique_members()) == 2
assert station.location.x == -86.8501514
assert station.location.y == 46.34077908
assert station.location.z == 0
# Apply time fitler as well
starting = datetime.utcnow() - timedelta(hours=6)
self.c.filter(start=starting)
collection = self.c.collect()
collection.calculate_bounds()
# Returns 4 stations: 04044724, 04045500, 04046000, 04056500
assert len(collection.elements) == 4
assert sorted(map(lambda x: x.uid, collection.elements)) == ["04044724", "04045500", "04046000", "04056500"]
def test_by_state(self):
# Clear filters
self.c.clear()
# Add custom state filter
self.c.filter(state="ri")
collection = self.c.collect()
# Returns 41 stations
assert len(collection.elements) == 41
station = collection.elements[0]
assert station.name == "TEN MILE R., PAWTUCKET AVE. AT E. PROVIDENCE, RI"
# Measures 2 variables
assert len(station.get_unique_members()) == 2
assert station.location.x == -71.3511649
assert station.location.y == 41.83093376
assert station.location.z == 0
def test_by_site_code(self):
# Clear filters
self.c.clear()
# Add custom state filter
self.c.filter(features=['04001000'])
collection = self.c.collect()
collection.calculate_bounds()
station = collection.elements[0]
assert station.uid == '04001000'
assert station.name == 'WASHINGTON CREEK AT WINDIGO, MI'
assert station.location.x == -89.1459196999999932
assert station.location.y == 47.9212792000000007
assert station.location.z == 0
class USGSTest(unittest.TestCase):
def setUp(self):
self.c = UsgsRest()
def test_by_bbox(self):
self.c.filter(bbox=(-87, 46, -85, 48))
collection = self.c.collect()
collection.calculate_bounds()
# Returns 4 stations: 04044724, 04045500, 04046000, 04056500
assert len(collection.elements) == 4
assert sorted([x.uid for x in collection.elements]) == ["04044724", "04045500", "04046000", "04056500"]
station = collection.elements[0]
assert station.name == "AU TRAIN RIVER AT FOREST LAKE, MI"
# Measures 2 variables
assert len(station.get_unique_members()) == 2
assert station.location.x == -86.8501514
assert station.location.y == 46.34077908
assert station.location.z == 0
# Apply time filter as well
starting = datetime(2013, 12, 31, 0, 0, 0)
ending = starting + timedelta(hours=6)
self.c.filter(start=starting, end=ending)
collection = self.c.collect()
collection.calculate_bounds()
# Returns 4 stations: 04044724, 04045500, 04046000, 04056500
assert len(collection.elements) == 4
assert sorted([x.uid for x in collection.elements]) == ["04044724", "04045500", "04046000", "04056500"]
def test_by_state(self):
# Clear filters
self.c.clear()
# Add custom state filter
self.c.filter(state="ri")
collection = self.c.collect()
# Returns 43 stations.
# FIXME: This is a very flaky test! The station number changed from
# 41, to 42 and now 43. (And back to 42!)
assert len(collection.elements) == 42
station = collection.elements[0]
assert station.name == "TEN MILE R., PAWTUCKET AVE. AT E. PROVIDENCE, RI"
# Measures 2 variables
assert len(station.get_unique_members()) == 2
assert station.location.x == -71.3511649
assert station.location.y == 41.83093376
assert station.location.z == 0
def test_by_site_code(self):
# Clear filters
self.c.clear()
# Add custom state filter
self.c.filter(features=['04001000'])
collection = self.c.collect()
collection.calculate_bounds()
station = collection.elements[0]
assert station.uid == '04001000'
assert station.name == 'WASHINGTON CREEK AT WINDIGO, MI'
assert station.location.x == -89.1459196999999932
assert station.location.y == 47.9212792000000007
assert station.location.z == 0
def get_usgs_data(station_id,
start_date,
end_date,
parameter="00060",
cache_dir=None):
"""Get river discharge data from the USGS REST web service.
See `U.S. Geological Survey Water Services
<https://waterservices.usgs.gov/>`_ (USGS)
Parameters
----------
station_id : str
The station id to get
start_date : str
String for start date in the format: 'YYYY-MM-dd', e.g. '1980-01-01'
end_date : str
String for start date in the format: 'YYYY-MM-dd', e.g. '2018-12-31'
parameter : str
The parameter code to get, e.g. ('00060') discharge, cubic feet per second
cache_dir : str
Directory where files retrieved from the web service are cached.
If set to None then USGS_DATA_HOME env var will be used as cache directory.
Examples
--------
>>> from ewatercycle.observation.usgs import get_usgs_data
>>> data = get_usgs_data('03109500', '2000-01-01', '2000-12-31', cache_dir='.')
>>> data
<xarray.Dataset>
Dimensions: (time: 8032)
Coordinates:
* time (time) datetime64[ns] 2000-01-04T05:00:00 ... 2000-12-23T04:00:00
Data variables:
Streamflow (time) float32 8.296758 10.420501 ... 10.647034 11.694747
Attributes:
title: USGS Data from streamflow data
station: Little Beaver Creek near East Liverpool OH
stationid: 03109500
location: (40.6758974, -80.5406244)
""" # noqa: E501
if cache_dir is None:
cache_dir = os.environ["USGS_DATA_HOME"]
# Check if we have the netcdf data
netcdf = os.path.join(
cache_dir,
"USGS_" + station_id + "_" + parameter + "_" + start_date + "_" +
end_date + ".nc",
)
if os.path.exists(netcdf):
return xr.open_dataset(netcdf)
# Download the data if needed
out = os.path.join(
cache_dir,
"USGS_" + station_id + "_" + parameter + "_" + start_date + "_" +
end_date + ".wml",
)
if not os.path.exists(out):
collector = UsgsRest()
collector.filter(
start=datetime.strptime(start_date, "%Y-%m-%d"),
end=datetime.strptime(end_date, "%Y-%m-%d"),
variables=[parameter],
features=[station_id],
)
data = collector.raw()
with open(out, "w") as file:
file.write(data)
collector.clear()
else:
with open(out, "r") as file:
data = file.read()
# Convert the raw data to an xarray
data = WaterML11ToPaegan(data).feature
# We expect only 1 station
if len(data.elements) == 0:
raise ValueError("Data does not contain any station data")
else:
station = data.elements[0]
# Unit conversion from cubic feet to cubic meter per second
values = np.array(
[
float(point.members[0]["value"]) / 35.315
for point in station.elements
],
dtype=np.float32,
)
times = [point.time for point in station.elements]
attrs = {
"units": "cubic meters per second",
}
# Create the xarray dataset
ds = xr.Dataset({"streamflow": (["time"], values, attrs)},
coords={"time": times})
# Set some nice attributes
ds.attrs["title"] = "USGS Data from streamflow data"
ds.attrs["station"] = station.name
ds.attrs["stationid"] = station.get_uid()
ds.attrs["location"] = (station.location.y, station.location.x)
ds.to_netcdf(netcdf)
return ds
