def test_waterdata_byfilter():
crs = "epsg:3857"
wd = WaterData("huc12", crs)
wb = wd.byfilter(f"{wd.layer} LIKE '17030001%'")
coords = (wb.iloc[0].geometry.centroid.x, wb.iloc[0].geometry.centroid.y)
hucs = wd.bydistance(coords, 100, crs)
assert wb.shape[0] == 52 and hucs.huc12.values[0] == "170300010602"
def test_waterdata_bybox():
wd = WaterData("nhdwaterbody")
print(wd)
wb_g = wd.bygeom(box(-69.7718, 45.0742, -69.3141, 45.4534),
predicate="INTERSECTS",
xy=True)
wb_b = wd.bybox((-69.7718, 45.0742, -69.3141, 45.4534))
assert abs(wb_b.areasqkm.sum() - wb_g.areasqkm.sum()) < 1e-3
def test_acc():
wd = WaterData("nhdflowline_network")
comids = nldi.navigate_byid("nwissite", "USGS-11092450", UT, "flowlines")
comid_list = comids.nhdplus_comid.tolist()
trib = wd.byid("comid", comid_list)
nhd.prepare_nhdplus(trib, 0, 0, 0, False, False)
flw = nhd.prepare_nhdplus(trib, 1, 1, 1, True, True)
def routing(qin, q):
return qin + q
qsim = nhd.vector_accumulation(
flw[["comid", "tocomid", "lengthkm"]],
routing,
"lengthkm",
["lengthkm"],
)
flw = flw.merge(qsim, on="comid")
diff = flw.arbolatesu - flw.acc_lengthkm
assert diff.abs().sum() < 1e-5
def test_waterdata_byid():
comids = nldi.navigate_byid(site, station_id, UT, "flowlines")
comid_list = comids.nhdplus_comid.tolist()
wd = WaterData("nhdflowline_network")
trib = wd.byid("comid", comid_list)
wd = WaterData("catchmentsp")
ct = wd.byid("featureid", comid_list)
assert (trib.shape[0] == 432 and abs(trib.lengthkm.sum() - 565.755) < 1e-3
and abs(ct.areasqkm.sum() - 773.954) < 1e-3)
def get_info(self, query: Dict[str, str], expanded: bool = False) -> pd.DataFrame:
"""Get NWIS stations by a list of IDs or within a bounding box.
Only stations that record(ed) daily streamflow data are returned.
The following columns are included in the dataframe with expanded
set to False:
================== ==================================
Name Description
================== ==================================
site_no Site identification number
station_nm Site name
site_tp_cd Site type
dec_lat_va Decimal latitude
dec_long_va Decimal longitude
coord_acy_cd Latitude-longitude accuracy
dec_coord_datum_cd Decimal Latitude-longitude datum
alt_va Altitude of Gage/land surface
alt_acy_va Altitude accuracy
alt_datum_cd Altitude datum
huc_cd Hydrologic unit code
parm_cd Parameter code
stat_cd Statistical code
ts_id Internal timeseries ID
loc_web_ds Additional measurement description
medium_grp_cd Medium group code
parm_grp_cd Parameter group code
srs_id SRS ID
access_cd Access code
begin_date Begin date
end_date End date
count_nu Record count
hcdn_2009 Whether is in HCDN-2009 stations
================== ==================================
Parameters
----------
query : dict
A dictionary containing query by IDs or BBOX. Use ``query_byid`` or ``query_bbox``
class methods to generate the queries.
expanded : bool, optional
Whether to get expanded sit information for example drainage area.
Returns
-------
pandas.DataFrame
NWIS stations
"""
if not isinstance(query, dict):
raise InvalidInputType("query", "dict")
output_type = [{"outputDataTypeCd": "dv"}]
if expanded:
output_type.append({"siteOutput": "expanded"})
site_list = []
for t in output_type:
payload = {
**query,
**t,
"format": "rdb",
"parameterCd": "00060",
"siteStatus": "all",
"hasDataTypeCd": "dv",
}
resp = self.session.post(f"{self.url}/site", payload).text.split("\n")
r_list = [txt.split("\t") for txt in resp if "#" not in txt]
r_dict = [dict(zip(r_list[0], st)) for st in r_list[2:]]
site_list.append(pd.DataFrame.from_dict(r_dict).dropna())
if expanded:
sites = pd.merge(
*site_list, on="site_no", how="outer", suffixes=("", "_overlap")
).filter(regex="^(?!.*_overlap)")
else:
sites = site_list[0]
sites = sites.drop(sites[sites.alt_va == ""].index)
try:
sites = sites[sites.parm_cd == "00060"]
sites["begin_date"] = pd.to_datetime(sites["begin_date"])
sites["end_date"] = pd.to_datetime(sites["end_date"])
except AttributeError:
pass
float_cols = ["dec_lat_va", "dec_long_va", "alt_va", "alt_acy_va"]
if expanded:
float_cols += ["drain_area_va", "contrib_drain_area_va"]
sites[float_cols] = sites[float_cols].apply(lambda x: pd.to_numeric(x, errors="coerce"))
sites = sites[sites.site_no.apply(len) == 8]
gii = WaterData("gagesii", DEF_CRS)
hcdn = gii.byid("staid", sites.site_no.tolist())
hcdn_dict = hcdn[["staid", "hcdn_2009"]].set_index("staid").hcdn_2009.to_dict()
sites["hcdn_2009"] = sites.site_no.apply(
lambda x: len(hcdn_dict[x]) > 0 if x in hcdn_dict.keys() else False
)
return sites
def test_waterdata_byfilter():
wd = WaterData("huc12", "epsg:3857")
wb = wd.byfilter(f"{wd.layer} LIKE '17030001%'")
assert wb.shape[0] == 52
def get_info(self,
queries: Union[Dict[str, str], List[Dict[str, str]]],
expanded: bool = False) -> gpd.GeoDataFrame:
"""Send multiple queries to USGS Site Web Service.
Parameters
----------
queries : dict or list of dict
A single or a list of valid queries.
expanded : bool, optional
Whether to get expanded sit information for example drainage area,
default to False.
Returns
-------
geopandas.GeoDataFrame
A correctly typed ``GeoDataFrame`` containing site(s) information.
"""
queries = [queries] if isinstance(queries, dict) else queries
payloads = self._validate_usgs_queries(queries, False)
sites = self.retrieve_rdb(f"{self.url}/site", payloads)
float_cols = ["dec_lat_va", "dec_long_va", "alt_va", "alt_acy_va"]
if expanded:
payloads = self._validate_usgs_queries(queries, True)
sites = sites.merge(
self.retrieve_rdb(f"{self.url}/site", payloads),
on="site_no",
how="outer",
suffixes=("", "_overlap"),
)
sites = sites.filter(regex="^(?!.*_overlap)")
float_cols += ["drain_area_va", "contrib_drain_area_va"]
with contextlib.suppress(KeyError):
sites["begin_date"] = pd.to_datetime(sites["begin_date"])
sites["end_date"] = pd.to_datetime(sites["end_date"])
gii = WaterData("gagesii", DEF_CRS)
def _get_hcdn(site_no: str) -> Tuple[float, Optional[bool]]:
try:
gage = gii.byid("staid", site_no)
return gage.drain_sqkm.iloc[0], len(
gage.hcdn_2009.iloc[0]) > 0 # noqa: TC300
except (AttributeError, KeyError, ZeroMatchedOGC):
return np.nan, None
sites["drain_sqkm"], sites["hcdn_2009"] = zip(
*[_get_hcdn(n) for n in sites.site_no])
float_cols += ["drain_sqkm"]
sites[float_cols] = sites[float_cols].apply(pd.to_numeric,
errors="coerce")
return gpd.GeoDataFrame(
sites,
geometry=gpd.points_from_xy(sites.dec_long_va, sites.dec_lat_va),
crs="epsg:4326",
)