def write_load_reduction_range(reduction_df, site): # pragma: no cover
final_cols = [
"Total Suspended Solids",
"Cadmium (Cd)",
"Copper (Cu)",
"Iron (Fe)",
"Lead (Pb)",
"Nickel (Ni)",
"Zinc (Zn)",
"Nitrate (N)",
"Orthophosphate (P)",
"Total Kjeldahl Nitrogen (TKN)",
"Total Phosphorus",
]
reduction_df = (
reduction_df.applymap(lambda x: utils.sigFigs(x, n=2))
.apply(lambda r: "{} - {}".format(r["load_red_lower"], r["load_red_upper"]), axis=1)
.unstack(level="parameter")
)[final_cols]
reduction_df.xs(site, level="site").to_csv("{}_reduction.csv".format(site), quoting=csv.QUOTE_ALL)
def _res_with_units(res, units):
if pandas.isnull(res):
return '--'
else:
return '{} {}'.format(utils.sigFigs(res, 3), units)
def load_summary_table(wq): # pragma: no cover
""" Produces a summary table of loads and confidence intervals for
varying sample/event types (e.g., composite, unsampled w/ or w/o
outflor) from tidy water quality data.
"""
def set_column_name(df):
df.columns.names = ["quantity"]
return df
def formatter(row, location):
cols = "load_{0};load_{0}_lower;load_{0}_upper".format(location).split(";")
row_str = "{} ({}; {})".format(*row[cols])
return row_str
def drop_unit_index(df):
df = df.copy()
df.index = df.index.droplevel("units")
return df
def set_no_outlow_zero(df):
col = ("Effluent", "unsampled", "No")
df[col] = 0
return df
def set_measured_effl(df):
col = ("Effluent", "composite", "Yes")
df[col] = df[col].apply(lambda x: str(x).split(" ")[0])
return df
def swal_col_levels(df, ii, jj):
df.columns = df.columns.swaplevel(ii, jj)
return df
def pct_reduction(df, incol, outcol):
return 100 * (df[incol] - df[outcol]) / df[incol]
final_cols = [
("Influent", "unsampled", "No"),
("Effluent", "unsampled", "No"),
("Influent", "unsampled", "Yes"),
("Effluent", "unsampled", "Yes"),
("Influent", "composite", "Yes"),
("Effluent", "composite", "Yes"),
"Reduction",
]
final_params = [
"Nitrate + Nitrite",
"Nitrate (N)",
"Orthophosphate (P)",
"Cadmium (Cd)",
"Copper (Cu)",
"Iron (Fe)",
"Total Kjeldahl Nitrogen (TKN)",
"Lead (Pb)",
"Nickel (Ni)",
"Total Phosphorus",
"Total Suspended Solids",
"Zinc (Zn)",
]
loads = (
wq.groupby(by=["parameter", "units", "has_outflow", "sampletype"])
.sum()
.pipe(set_column_name)
.select(lambda c: c.startswith("load_inflow") or c.startswith("load_outflow"), axis=1)
.unstack(level="has_outflow")
.pipe(swal_col_levels, "has_outflow", "quantity")
.stack(level="has_outflow")
.dropna()
.pipe(drop_unit_index)
)
main = (
loads.applymap(lambda x: utils.sigFigs(x, n=3, expthresh=7))
.assign(Influent=lambda df: df.apply(formatter, args=("inflow",), axis=1))
.assign(Effluent=lambda df: df.apply(formatter, args=("outflow",), axis=1))
.unstack(level="sampletype")
.unstack(level="has_outflow")[["Influent", "Effluent"]]
.dropna(how="all", axis=1)
)
reduction = (
loads.groupby(level="parameter")
.sum()
.assign(load_red=lambda df: pct_reduction(df, "load_inflow", "load_outflow"))
.assign(load_red_upper=lambda df: pct_reduction(df, "load_inflow_upper", "load_outflow_lower"))
.assign(load_red_lower=lambda df: pct_reduction(df, "load_inflow_lower", "load_outflow_upper"))
.applymap(lambda x: utils.sigFigs(x, n=3, expthresh=7))
.assign(Reduction=lambda df: df.apply(formatter, args=("red",), axis=1))
)
summary = main.join(reduction).loc[final_params, final_cols].pipe(set_no_outlow_zero).pipe(set_measured_effl)
return summary
