simple example of using ulmo and pandas to download nwis daily streamflow data and calculate historic mean discharge

mean_discharge.py

import numpy as np
import pandas as pd
from ulmo.usgs import nwis

# download and cache site data (this will take a long time the first time)
# currently downloads all available parameters
nwis.hdf5.update_site_data('06043500')

# read daily mean discharge data from cache (statistics code 00003)
data = nwis.hdf5.get_site_data('06043500', parameter_code='00060:00003')['00060:00003']

# convert data to a pandas dataframe
df = pd.DataFrame(data['values']).drop(['last_checked','last_modified','qualifiers'], axis=1).set_index('datetime')
df.value = df.value.apply(np.float)
df.index = pd.to_datetime(df.index).to_period('D')

# mark bad data as NaN
df[df.values == -999999] = np.nan

# group the data by month, day & calculate means
daily_groups = df.groupby((lambda d: d.month, lambda d: d.day))
means = daily_groups.mean()

print 'historic daily mean on March 23rd is %s' % means.ix[3,23].value