Fetching data from a CSW catalog with Python tools


This notebook shows a typical workflow to query a Catalog Service for the Web (CSW) and create a request for data endpoints that are suitable for download.

In this queries multiple catalogs for the near real time HF-Radar current data.

The first step is to create the data filter based on the geographical region bounding box, the time span, and the CF variable standard name.

In [1]:
from datetime import datetime, timedelta

# Region: West coast.
min_lon, max_lon = -123, -121
min_lat, max_lat = 36, 40

bbox = [min_lon, min_lat, max_lon, max_lat]
crs = 'urn:ogc:def:crs:OGC:1.3:CRS84'

# Temporal range: past week.
now = datetime.utcnow()
start, stop = now - timedelta(days=(14)), now - timedelta(days=(7))

# Surface velocity CF names.
cf_names = [

msg = """
*standard_names*: {cf_names}
*start and stop dates*: {start} to {stop}
*bounding box*:{bbox}
*crs*: {crs}""".format

    'cf_names': ', '.join(cf_names),
    'start': start,
    'stop': stop,
    'bbox': bbox,
    'crs': crs,
    *standard_names*: surface_northward_sea_water_velocity, surface_eastward_sea_water_velocity
    *start and stop dates*: 2018-09-21 19:42:26.750025 to 2018-09-28 19:42:26.750025
    *bounding box*:[-123, 36, -121, 40]
    *crs*: urn:ogc:def:crs:OGC:1.3:CRS84

Now it is possible to assemble a OGC Filter Encoding (FE) for the search using owslib.fes*. Note that the final result is only a list with all the filtering conditions.

* OWSLib is a Python package for client programming with Open Geospatial Consortium (OGC) web service (hence OWS) interface standards, and their related content models.

In [2]:
from owslib import fes

def fes_date_filter(start, stop, constraint='overlaps'):
    start = start.strftime('%Y-%m-%d %H:00')
    stop = stop.strftime('%Y-%m-%d %H:00')
    if constraint == 'overlaps':
        propertyname = 'apiso:TempExtent_begin'
        begin = fes.PropertyIsLessThanOrEqualTo(propertyname=propertyname,
        propertyname = 'apiso:TempExtent_end'
        end = fes.PropertyIsGreaterThanOrEqualTo(propertyname=propertyname,
    elif constraint == 'within':
        propertyname = 'apiso:TempExtent_begin'
        begin = fes.PropertyIsGreaterThanOrEqualTo(propertyname=propertyname,
        propertyname = 'apiso:TempExtent_end'
        end = fes.PropertyIsLessThanOrEqualTo(propertyname=propertyname,
        raise NameError('Unrecognized constraint {}'.format(constraint))
    return begin, end
In [3]:
kw = dict(wildCard='*', escapeChar='\\',
          singleChar='?', propertyname='apiso:AnyText')

or_filt = fes.Or([fes.PropertyIsLike(literal=('*%s*' % val), **kw)
                  for val in cf_names])

# Exclude GNOME returns.
not_filt = fes.Not([fes.PropertyIsLike(literal='*GNOME*', **kw)])

begin, end = fes_date_filter(start, stop)
bbox_crs = fes.BBox(bbox, crs=crs)
filter_list = [fes.And([bbox_crs, begin, end, or_filt, not_filt])]

It is possible to use the same filter to search multiple catalogs. The cell below loops over 3 catalogs hoping to find which one is more up-to-date and returns the near real time data.

In [4]:
def get_csw_records(csw, filter_list, pagesize=10, maxrecords=1000):
    """Iterate `maxrecords`/`pagesize` times until the requested value in
    `maxrecords` is reached.
    from owslib.fes import SortBy, SortProperty
    # Iterate over sorted results.
    sortby = SortBy([SortProperty('dc:title', 'ASC')])
    csw_records = {}
    startposition = 0
    nextrecord = getattr(csw, 'results', 1)
    while nextrecord != 0:
        csw.getrecords2(constraints=filter_list, startposition=startposition,
                        maxrecords=pagesize, sortby=sortby)
        if csw.results['nextrecord'] == 0:
        startposition += pagesize + 1  # Last one is included.
        if startposition >= maxrecords:
In [5]:
from owslib.csw import CatalogueServiceWeb

endpoint = 'https://data.ioos.us/csw'

csw = CatalogueServiceWeb(endpoint, timeout=60)
get_csw_records(csw, filter_list, pagesize=10, maxrecords=1000)

records = '\n'.join(csw.records.keys())
print('Found {} records.\n'.format(len(csw.records.keys())))
for key, value in list(csw.records.items()):
    print('[{}]: {}'.format(value.title, key))
    Found 5 records.
    [HYbrid Coordinate Ocean Model (HYCOM): Global]: hycom_global
    [Near-Real Time Surface Ocean Velocity, U.S. West Coast,
    1 km Resolution]: HFR/USWC/1km/hourly/RTV/HFRADAR_US_West_Coast_1km_Resolution_Hourly_RTV_best.ncd
    [Near-Real Time Surface Ocean Velocity, U.S. West Coast,
    2 km Resolution]: HFR/USWC/2km/hourly/RTV/HFRADAR_US_West_Coast_2km_Resolution_Hourly_RTV_best.ncd
    [Near-Real Time Surface Ocean Velocity, U.S. West Coast,
    500 m Resolution]: HFR/USWC/500m/hourly/RTV/HFRADAR_US_West_Coast_500m_Resolution_Hourly_RTV_best.ncd
    [Near-Real Time Surface Ocean Velocity, U.S. West Coast,
    6 km Resolution]: HFR/USWC/6km/hourly/RTV/HFRADAR_US_West_Coast_6km_Resolution_Hourly_RTV_best.ncd

Let us check the 6 km resolution metadata record we found above.

In [6]:
value = csw.records[

    Surface ocean velocities estimated from HF-Radar are
    representative of the upper 0.3 - 2.5 meters of the
    ocean.  The main objective of near-real time
    processing is to produce the best product from
    available data at the time of processing.  Radial
    velocity measurements are obtained from individual
    radar sites through the U.S. HF-Radar Network.
    Hourly radial data are processed by unweighted
    least-squares on a 6 km resolution grid of the U.S.
    West Coast to produce near real-time surface current

In [7]:
attrs = [attr for attr in dir(value) if not attr.startswith('_')]
nonzero = [attr for attr in attrs if getattr(value, attr)]

The xml has the full dataset metadata from the catalog. Let’s print a few key ones here:

In [8]:
value.subjects  # What is in there?
In [9]:
value.modified  # Is it up-to-date?
In [10]:
bbox = value.bbox.minx, value.bbox.miny, value.bbox.maxx, value.bbox.maxy
bbox  # The actual bounding box of the data.
('-130.36', '30.25', '-115.81', '49.99')

The next step is to inspect the type services and schemes available for downloading the data. The easiest way to accomplish that is with by “sniffing” the URLs with geolinks.

In [11]:
from geolinks import sniff_link

msg = 'geolink: {geolink}\nscheme: {scheme}\nURL: {url}\n'.format
for ref in value.references:
    if ref['scheme'] == 'OPeNDAP:OPeNDAP':
        url = ref['url']
    print(msg(geolink=sniff_link(ref['url']), **ref))
    geolink: WWW:LINK
    scheme: WWW:LINK
    URL: http://hfrnet-tds.ucsd.edu/thredds/dodsC/HFR/USWC/6km/hourly/RTV/HFRADAR_US_West_Coast_6km_Resolution_Hourly_RTV_best.ncd.html
    geolink: None
    scheme: WWW:LINK
    URL: http://www.ncdc.noaa.gov/oa/wct/wct-jnlp-beta.php?singlefile=http://hfrnet-tds.ucsd.edu/thredds/dodsC/HFR/USWC/6km/hourly/RTV/HFRADAR_US_West_Coast_6km_Resolution_Hourly_RTV_best.ncd
    geolink: None
    scheme: OPeNDAP:OPeNDAP
    URL: http://hfrnet-tds.ucsd.edu/thredds/dodsC/HFR/USWC/6km/hourly/RTV/HFRADAR_US_West_Coast_6km_Resolution_Hourly_RTV_best.ncd
    geolink: OGC:WCS
    scheme: OGC:WCS
    URL: http://hfrnet-tds.ucsd.edu/thredds/wcs/HFR/USWC/6km/hourly/RTV/HFRADAR_US_West_Coast_6km_Resolution_Hourly_RTV_best.ncd?service=WCS&version=1.0.0&request=GetCapabilities
    geolink: OGC:WMS
    scheme: OGC:WMS
    URL: http://hfrnet-tds.ucsd.edu/thredds/wms/HFR/USWC/6km/hourly/RTV/HFRADAR_US_West_Coast_6km_Resolution_Hourly_RTV_best.ncd?service=WMS&version=1.3.0&request=GetCapabilities
    geolink: UNIDATA:NCSS
    scheme: UNIDATA:NCSS
    URL: http://hfrnet-tds.ucsd.edu/thredds/ncss/grid/HFR/USWC/6km/hourly/RTV/HFRADAR_US_West_Coast_6km_Resolution_Hourly_RTV_best.ncd/dataset.html

For a detailed description of what those geolink results mean check the lookup table. There are Web Coverage Service (WCS), Web Map Service (WMS), direct links, and OPeNDAP services available.

We can use any of those to obtain the data but the easiest one to explore interactively is the open OPeNDAP endpoint.

In [12]:
import xarray as xr

ds = xr.open_dataset(url)
Dimensions:       (lat: 367, lon: 234, nProcParam: 7, nSites: 54, time: 61425)
  * lat           (lat) float32 30.25 30.30394 30.35788 ... 49.9381 49.99204
  * lon           (lon) float32 -130.36 -130.29753 ... -115.86803 -115.805565
  * time          (time) datetime64[ns] 2011-10-01 ... 2018-10-05T18:00:00
    time_run      (time) datetime64[ns] ...
Dimensions without coordinates: nProcParam, nSites
Data variables:
    site_lat      (nSites) float32 ...
    site_lon      (nSites) float32 ...
    site_code     (nSites) |S64 ...
    site_netCode  (nSites) |S64 ...
    procParams    (nProcParam) float32 ...
    time_offset   (time) datetime64[ns] ...
    u             (time, lat, lon) float32 ...
    v             (time, lat, lon) float32 ...
    DOPx          (time, lat, lon) float32 ...
    DOPy          (time, lat, lon) float32 ...
    netcdf_library_version:  4.1.3
    format_version:          HFRNet_1.0.0
    product_version:         HFRNet_1.1.05
    Conventions:             CF-1.4
    title:                   Near-Real Time Surface Ocean Velocity, U.S. West...
    institution:             Scripps Institution of Oceanography
    source:                  Surface Ocean HF-Radar
    history:                 05-Oct-2018 08:07:19: NetCDF file created\n05-Oc...
    references:              Terrill, E. et al., 2006. Data Management and Re...
    creator_name:            Mark Otero
    creator_email:           motero@ucsd.edu
    creator_url:             http://cordc.ucsd.edu/projects/mapping/
    summary:                 Surface ocean velocities estimated from HF-Radar...
    geospatial_lat_min:      30.25
    geospatial_lat_max:      49.99204
    geospatial_lon_min:      -130.36
    geospatial_lon_max:      -115.805565
    grid_resolution:         6km
    grid_projection:         equidistant cylindrical
    regional_description:    Unites States, West Coast
    _CoordSysBuilder:        ucar.nc2.dataset.conv.CF1Convention
    cdm_data_type:           GRID
    featureType:             GRID
    location:                Proto fmrc:HFRADAR_US_West_Coast_6km_Resolution_...
    DODS.strlen:             25
    DODS.dimName:            nSites_maxStrlen

Select “yesterday” data.

In [13]:
from datetime import date, timedelta

yesterday = date.today() - timedelta(days=1)

ds = ds.sel(time=yesterday)

Compute the speed while masking invalid values.

In [14]:
import numpy.ma as ma

u = ds['u'].data
v = ds['v'].data

lon = ds.coords['lon'].data
lat = ds.coords['lat'].data
time = ds.coords['time'].data

u = ma.masked_invalid(u)
v = ma.masked_invalid(v)

This cell is only a trick to show all quiver arrows with the same length, for visualization purposes, and indicate the vector magnitude with colors instead.

In [15]:
import numpy as np
from oceans.ocfis import uv2spdir, spdir2uv

angle, speed = uv2spdir(u, v)
us, vs = spdir2uv(np.ones_like(speed), angle, deg=True)

And now we are ready to create the plot.

In [16]:
%matplotlib inline

import cartopy.crs as ccrs
import matplotlib.pyplot as plt

from cartopy import feature
from cartopy.mpl.gridliner import LONGITUDE_FORMATTER, LATITUDE_FORMATTER

LAND = feature.NaturalEarthFeature(
    'physical', 'land', '10m',

sub = 2
dx = dy = 0.5
center = -122.416667, 37.783333  # San Francisco.
bbox = lon.min(), lon.max(), lat.min(), lat.max()

fig, (ax0, ax1) = plt.subplots(
    figsize=(20, 20),

cs = ax0.pcolormesh(lon, lat, ma.masked_invalid(speed))
gl = ax0.gridlines(draw_labels=True)
gl.xlabels_top = gl.ylabels_right = False
gl.yformatter = LATITUDE_FORMATTER

cbar = fig.colorbar(cs, ax=ax0, shrink=0.45, extend='both')
cbar.ax.set_title(r'speed m s$^{-1}$', loc='left')

ax0.add_feature(LAND, zorder=0, edgecolor='black')
ax0.set_title('{}\n{}'.format(value.title, ds['time'].values))

ax1.set_extent([center[0]-dx-dx, center[0]+dx, center[1]-dy, center[1]+dy])
q = ax1.quiver(lon[::sub], lat[::sub],
               us[::sub, ::sub], vs[::sub, ::sub],
               speed[::sub, ::sub], scale=30)
ax1.quiverkey(q, 0.5, 0.85, 1, r'1 m s$^{-1}$', coordinates='axes')
gl = ax1.gridlines(draw_labels=True)
gl.xlabels_top = gl.ylabels_right = False
gl.yformatter = LATITUDE_FORMATTER

ax1.add_feature(LAND, zorder=0, edgecolor='black')
ax1.plot(ds['site_lon'], ds['site_lat'], marker='o', linestyle='none', color='darkorange')
ax1.set_title('San Francisco Bay area');


And here is yesterday’s sea surface currents from the west coast with a zoom in the San Francisco Bay area.
Right click and choose Save link as… to download this notebook, or click here to run a live instance of this notebook.