diff --git a/icoscp/docs/authentication.md b/icoscp/docs/authentication.md
new file mode 100644
index 0000000..7ac40a4
--- /dev/null
+++ b/icoscp/docs/authentication.md
@@ -0,0 +1,113 @@
+# Authentication
+To ensure users' licence acceptance when accessing data objects
+through the `icoscp` Python library, authentication is required. Users
+must either have ICOS Carbon Portal login credentials or
+log in to Carbon Portal using another mechanism
+[https://cpauth.icos-cp.eu/login/](https://cpauth.icos-cp.eu/login/)
+to obtain the token to access ICOS data.
+
+Users also need to read and accept the ICOS Data Licence in their
+[user profile](https://cpauth.icos-cp.eu/).
+
+**Metadata-only access does not require authentication.**
+
+Users with direct access to the data files,
+namely, **users of our [ICOS Jupyter services](
+https://www.icos-cp.eu/data-services/tools/jupyter-notebook), are
+not required to configure authentication**.
+
+In order to fetch data, users make their requests to data objects and
+must provide an API token to do so. For security reasons the API token is
+valid for 100'000 seconds (27 hours) and may therefore need to be refreshed;
+this process is automated when using credentials-based authentication (see
+below).
+
+Authentication can be initialized in a number of ways. **Please note** that
+when using other approaches than the default one (see below), it becomes
+necessary to use the `data` value (instance of `DataClient` class) obtained
+in the process of authentication bootstrap, rather than the default instance
+obtained by import `from icoscp_core.icos import data` used in the code
+examples in this documentation.
+
+## By credentials file (default)
+This approach should only be used on machines the developer trusts.
+
+A username/password account for the [ICOS](https://cpauth.icos-cp.eu/)
+authentication service is required for this. Obfuscated (not readable by
+humans) password is stored in a file on the local machine in a **default
+user-specific folder**. To initialize this file, run the following code
+interactively (only needs to be done once for every machine):
+
+```Python
+from icoscp_core.icos import auth
+auth.init_config_file()
+```
+
+After the initialization step is done in this way, access to the data can be
+achieved using both the new `icoscp_core` machinery and the legacy
+[Dobj classes](modules.md#dobj).
+
+## By custom credentials file
+The developer may wish to use a specific file to store obfuscated
+credentials and token cache. In this scenario, data and
+metadata access are achieved as follows:
+
+```Python
+from icoscp_core.icos import bootstrap
+auth, meta, data = bootstrap.fromPasswordFile("<desired path to the file>")
+# The next line needs to be run interactively (only once per file).
+auth.init_config_file()
+```
+
+If the legacy library functionality will be used, the following extra step is
+needed as well:
+
+```Python
+from icoscp import cpauth
+cpauth.init_by(auth)
+```
+
+## By authentication token (prototyping)
+This option is good for testing, on a public machine or in general. Its
+only disadvantage is that the tokens have limited period of validity
+(100'000 seconds, less than 28 hours), but this is precisely what makes
+it acceptable to include them directly in the Python source code.
+
+The token can be obtained from the "My Account" page ([ICOS](
+https://cpauth.icos-cp.eu/)), which can be accessed by logging in
+using one of the supported authentication mechanisms (username/password,
+university sign-in, OAuth sign in). After this the bootstrapping can be
+done as follows:
+
+```Python
+from icoscp_core.icos import bootstrap
+cookie_token = 'cpauthToken=WzE2OTY2NzQ5OD...'
+meta, data = bootstrap.fromCookieToken(cookie_token)
+```
+
+If the legacy library functionality will be used, the following extra step is
+needed as well:
+
+```Python
+from icoscp import cpauth
+cpauth.init_by(data.auth)
+```
+
+## By explicit credentials (advanced option)
+The user may choose to use their own mechanism of providing the
+credentials to initialize the authentication. This should be considered
+as an advanced option. **(Please do not put your password as clear text
+in your Python code!)** This can be achieved as follows:
+
+```Python
+from icoscp_core.icos import bootstrap
+meta, data = bootstrap.fromCredentials(username_variable, password_containing_variable)
+```
+
+If the legacy library functionality will be used, the following extra step is
+needed as well:
+
+```Python
+from icoscp import cpauth
+cpauth.init_by(data.auth)
+```
diff --git a/icoscp/docs/changelog.md b/icoscp/docs/changelog.md
index b9ff004..c66bec2 100644
--- a/icoscp/docs/changelog.md
+++ b/icoscp/docs/changelog.md
@@ -1,6 +1,6 @@
 # Changelog
 
-## 0.2.0a0
+## 0.2.0
 - #### cpauth module
     - Remove legacy authentication.
     - Authenticate via [icoscp_core](https://pypi.org/project/icoscp_core/).
diff --git a/icoscp/docs/examples.md b/icoscp/docs/examples.md
new file mode 100644
index 0000000..8df0597
--- /dev/null
+++ b/icoscp/docs/examples.md
@@ -0,0 +1,66 @@
+# Examples
+
+## Monthly CO2 averages
+
+This example uses `icoscp_core` functionality to merge, monthly-average and
+plot CO2 molar fraction data from selected stations, sampled at certain
+heights. See [discover data types](getting_started.md#discover-data-types)
+to find out how to discover a URL for a data type.
+
+```Python
+from icoscp_core.icos import data, meta, ATMO_STATION
+import pandas as pd
+import numpy as np
+
+# Swedish atmospheric stations whose metadata is supplied to the Carbon Portal
+# by the Atmospheric Thematic Center
+se_atmo_stations = [
+    s for s in meta.list_stations(ATMO_STATION)
+    if s.country_code=='SE'
+]
+
+# Find basic metadata for CO2 release data sampled at at least 100 m above the ground
+se_co2_from_100 = [
+    dobj for dobj in meta.list_data_objects(
+        # URL for official ICOS CO2 molar fraction release data
+        datatype='http://meta.icos-cp.eu/resources/cpmeta/atcCo2L2DataObject',
+        station=se_atmo_stations
+    )
+    if dobj.sampling_height >= 100
+]
+
+# prepare an empty pandas DataFrame to merge the data into
+merged_co2 = pd.DataFrame(columns=['TIMESTAMP', 'co2'])
+
+# batch-fetch the interesting columns and iterate through the results
+for dobj, arrs in data.batch_get_columns_as_arrays(se_co2_from_100, ['TIMESTAMP', 'co2']):
+    st_uri = dobj.station_uri
+    # ICOS atmospheric station URIs end with underscore followed by a 3-letter station ID
+	# this ID is convenient to use as a suffix to rename 'co2' with
+    station_id = st_uri[st_uri.rfind('_'):]
+    df = pd.DataFrame(arrs)
+    # next line would be needed if `keep_bad_data` flag in batch_get_columns_as_arrays was set to True
+    #df.loc[df['Flag'] != 'O', 'co2'] = np.nan 
+    del df['Flag']
+    merged_co2 = pd.merge(merged_co2, df, on='TIMESTAMP', how='outer', suffixes=('', station_id))
+
+del merged_co2['co2']
+
+# compute monthly averages
+by_month = merged_co2.groupby(pd.Grouper(key='TIMESTAMP', freq='M')).mean().reset_index()
+
+# Let us retrieve column metadata to construct Y-axis plot label
+
+# fetch detailed metadata for one of the data objects (they all have same columns)
+dobj_meta = meta.get_dobj_meta(se_co2_from_100[0])
+
+# get 'value type' part of column metadata for co2 column
+columns_meta = dobj_meta.specificInfo.columns
+co2_value_type = [col for col in columns_meta if col.label=='co2'][0].valueType
+
+# construct the Y-axis label from the value type data structure
+co2_axis_label = f'{co2_value_type.self.label} [{co2_value_type.unit}]'
+
+# plotting should work in Jupyter
+by_month.plot(x='TIMESTAMP', ylabel=co2_axis_label);
+```
\ No newline at end of file
diff --git a/icoscp/docs/faq.md b/icoscp/docs/faq.md
index 437324e..1e73008 100644
--- a/icoscp/docs/faq.md
+++ b/icoscp/docs/faq.md
@@ -4,10 +4,12 @@ Please see [Getting started](getting_started.md) for possible answers to questio
 covered here.
 
 ### `icoscp_core` is very different from the old `icoscp`, do I have to rewrite everything?
-No, your code depending on the older version will continue working. No code got
-removed from `icoscp` with release `0.2.0`. But you may benefit from the new
-`icoscp_core` features for new developments, and from gradual porting of your
-older code to using `icoscp_core`.
+No, your code depending on the older version will continue working. Apart from
+moving stilt-related functionality to a new dedicated library and overhauling
+authentication that was present in later versions of `0.1.x` series, no code got
+removed from `icoscp` with release `0.2.0`. But you should use the new
+`icoscp_core` features for new developments, and can benefit from gradual
+porting of at least some of your older code to using `icoscp_core`.
 
 ### How can I retrieve the latest/newest version of a dataset?
 ```python
diff --git a/icoscp/docs/getting_started.md b/icoscp/docs/getting_started.md
index 6fe3df4..e98da01 100644
--- a/icoscp/docs/getting_started.md
+++ b/icoscp/docs/getting_started.md
@@ -1,7 +1,6 @@
-# Getting started
+# Getting started with icoscp_core
 The examples in this section can be tried on a public Jupyter Hub running
-Python3 notebooks, where the library is preinstalled.
-
+Python3 notebooks, where the library is preinstalled, for example
 [https://exploredata.icos-cp.eu/](https://exploredata.icos-cp.eu/)
 
 Please, click [here](
@@ -12,119 +11,6 @@ If run on a standalone machine rather than an ICOS Carbon Portal Jupyter Hub
 instance, the data access examples assume that the authentication has been
 configured as explained in the next section.
 
-## Authentication
-To ensure users' licence acceptance when accessing data objects
-through the `icoscp` Python library, authentication is required. Users
-must either have ICOS Carbon Portal login credentials or
-log in to Carbon Portal using another mechanism
-[https://cpauth.icos-cp.eu/login/](https://cpauth.icos-cp.eu/login/)
-to obtain the token to access ICOS data.
-
-Users also need to read and accept the ICOS Data Licence in their
-[user profile](https://cpauth.icos-cp.eu/).
-
-Metadata-only access does not require authentication.
-
-Users with direct access to the data files,
-namely, **users of our [ICOS Jupyter services](
-https://www.icos-cp.eu/data-services/tools/jupyter-notebook), are
-not required to configure authentication**.
-
-In order to fetch data, users make their requests to data objects and
-must provide an API token to do so. For security reasons the API token is
-valid for 100'000 seconds (27 hours) and may therefore need to be refreshed;
-this process is automated when using credentials-based authentication (see
-below).
-
-Authentication can be initialized in a number of ways. **Please note** that
-when using other approaches than the default one (see below), it becomes
-necessary to use the `data` value (instance of `DataClient` class) obtained
-in the process of authentication bootstrap, rather than the default instance
-obtained by import `from icoscp_core.icos import data` used in the code
-examples in this documentation.
-
-### By credentials file (default)
-This approach should only be used on machines the developer trusts.
-
-A username/password account for the [ICOS](https://cpauth.icos-cp.eu/)
-authentication service is required for this. Obfuscated (not readable by
-humans) password is stored in a file on the local machine in a **default
-user-specific folder**. To initialize this file, run the following code
-interactively (only needs to be done once for every machine):
-
-```Python
-from icoscp_core.icos import auth
-auth.init_config_file()
-```
-
-After the initialization step is done in this way, access to the data can be
-achieved using both the new `icoscp_core` machinery and the legacy
-[Dobj classes](modules.md#dobj).
-
-### By custom credentials file
-The developer may wish to use a specific file to store obfuscated
-credentials and token cache. In this scenario, data and
-metadata access are achieved as follows:
-
-```Python
-from icoscp_core.icos import bootstrap
-auth, meta, data = bootstrap.fromPasswordFile("<desired path to the file>")
-# The next line needs to be run interactively (only once per file).
-auth.init_config_file()
-```
-
-If the legacy library functionality will be used, the following extra step is
-needed as well:
-
-```Python
-from icoscp import cpauth
-cpauth.init_by(auth)
-```
-
-### By authentication token (prototyping)
-This option is good for testing, on a public machine or in general. Its
-only disadvantage is that the tokens have limited period of validity
-(100'000 seconds, less than 28 hours), but this is precisely what makes
-it acceptable to include them directly in the Python source code.
-
-The token can be obtained from the "My Account" page ([ICOS](
-https://cpauth.icos-cp.eu/)), which can be accessed by logging in
-using one of the supported authentication mechanisms (username/password,
-university sign-in, OAuth sign in). After this the bootstrapping can be
-done as follows:
-
-```Python
-from icoscp_core.icos import bootstrap
-cookie_token = 'cpauthToken=WzE2OTY2NzQ5OD...'
-meta, data = bootstrap.fromCookieToken(cookie_token)
-```
-
-If the legacy library functionality will be used, the following extra step is
-needed as well:
-
-```Python
-from icoscp import cpauth
-cpauth.init_by(data.auth)
-```
-
-### By explicit credentials (advanced option)
-The user may choose to use their own mechanism of providing the
-credentials to initialize the authentication. This should be considered
-as an advanced option. **(Please do not put your password as clear text
-in your Python code!)** This can be achieved as follows:
-
-```Python
-from icoscp_core.icos import bootstrap
-meta, data = bootstrap.fromCredentials(username_variable, password_containing_variable)
-```
-
-If the legacy library functionality will be used, the following extra step is
-needed as well:
-
-```Python
-from icoscp import cpauth
-cpauth.init_by(data.auth)
-```
 ## General note on metadata
 An important background information on ICOS metadata is that all the
 metadata-represented entities (data objects, data types, documents,
@@ -205,6 +91,11 @@ import pandas as pd
 co2_release_data_pd = ( (dobj, pd.DataFrame(arrs)) for dobj, arrs in co2_release_data)
 ```
 
+## Examples
+
+See [Examples](examples.md#examples) for more lengthy examples using all of the
+functionality introduced above.
+
 ## Accessing documentation
 As this library depends on `icoscp_core`, all the functionality of the latter
 can be used, not only the examples from above. It is introduced on the
diff --git a/icoscp/docs/modules.md b/icoscp/docs/modules.md
index d5be9b0..208155b 100644
--- a/icoscp/docs/modules.md
+++ b/icoscp/docs/modules.md
@@ -7,7 +7,6 @@ your python environment you should be able to load the modules with:
 - `from icoscp.cpb.dobj import Dobj`
 - `from icoscp.station import station`
 - `from icoscp.collection import collection`
-- `from icoscp.stilt import stiltstation`
 - `from icoscp.sparql.runsparql import RunSparql`
 - `from icoscp.sparql import sparqls`
 
diff --git a/icoscp/mkdocs.yml b/icoscp/mkdocs.yml
index 40aaabd..54afc18 100644
--- a/icoscp/mkdocs.yml
+++ b/icoscp/mkdocs.yml
@@ -4,11 +4,13 @@ site_name: ICOS Carbon Portal icoscp Library
 nav:
     - About: index.md
     - Installation: install.md
-    - Getting started: getting_started.md
+    - Authentication: authentication.md
+    - Getting started with icoscp_core: getting_started.md
+    - Examples: examples.md
+    - Legacy icoscp modules: modules.md
+    - Legacy icoscp examples: legacy_examples.md
     - FAQ: faq.md
     - Changelog: changelog.md
-    - Legacy modules: modules.md
-    - Legacy examples: legacy_examples.md
 
 theme:
     name: readthedocs
diff --git a/icoscp_stilt/docs/getting_started.md b/icoscp_stilt/docs/getting_started.md
new file mode 100644
index 0000000..09f56ab
--- /dev/null
+++ b/icoscp_stilt/docs/getting_started.md
@@ -0,0 +1,80 @@
+# Getting started with the new icoscp_stilt
+
+The library is published to PyPI.
+
+As stated in [Background and history](index.md#background-and-history),
+the [legacy functionality](modules.md#legacy-modules) is
+still available, but for new code the developers are encouraged to consider
+the new module `icoscp_stilt.stilt` as the first choice.
+
+The following code demonstrates the new functionality.
+
+```Python
+from icoscp_stilt import stilt
+from icoscp_stilt.const import CP_OBSPACK_CO2_SPEC
+
+# list of stilt.StiltStation dataclass instances
+stations = stilt.list_stations()
+
+station_info_lookup = {s.id: s for s in stations}
+
+# example: Hyltemossa station, altitude 150 m
+htm_info = station_info_lookup['HTM150']
+htm_info
+```
+Here is the output of the above code.
+
+    StiltStation(
+        id='HTM150', name='Hyltemossa', lat=56.1, lon=13.42, alt=150,
+        countryCode='SE',
+        years=[2006, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022],
+        icosId='HTM', icosHeight=150.0
+    )
+
+```Python
+# years for which the station has calculation results
+htm_years = htm_info.years
+
+# grouped STILT time series results, all columns, as pandas DataFrame
+# can be slow when fetching first time after calculation
+htm_time_series_result = stilt.fetch_result_ts('HTM150', '2022-01-01', '2022-01-31')
+
+# list of time-series columns (~60)
+ts_columns = htm_time_series_result.columns
+
+# fetch selected columns only
+htm_ts_ch4_basics = stilt.fetch_result_ts('HTM150', '2022-01-01', '2022-01-31', columns=['isodate', 'ch4.stilt', 'metadata'])
+
+# find months for which calculation was run
+stilt.available_months('KRE250', 2022)
+
+# find all months for all years for which calculation was run
+htm_yearmonths = stilt.available_year_months(htm_info)
+
+# list footprint time slots that were computed for a station within date interval
+htm_slots_jan2022 = stilt.list_footprints('HTM150', '2022-01-01', '2022-01-31')
+
+# load footprint for one time slot
+htm_fp_example = stilt.load_footprint('HTM150', htm_slots_jan2022[0])
+
+# filter stations
+de_stations = [s for s in stations if s.countryCode == 'DE']
+
+# fetch observations for the German stations as numpy array dicts
+# interesting columns are requested explicitly (all returned otherwise)
+# using bare numpy gives maximum performance
+de_co2_numpy = stilt.fetch_observations(CP_OBSPACK_CO2_SPEC, de_stations, ['value', 'time'])
+
+# same as previous example, but returning pandas DataFrames instead
+# performance may be worse, especially on Jupyter
+de_co2_pandas = stilt.fetch_observations_pandas(CP_OBSPACK_CO2_SPEC, de_stations, ['value', 'time'])
+```
+
+## Getting help
+
+All the methods in the new `stilt` module have a Python documentation
+accessible by standard means, for example:
+
+```
+help(stilt.fetch_observations)
+```
\ No newline at end of file
diff --git a/icoscp_stilt/docs/index.md b/icoscp_stilt/docs/index.md
index 76a5da9..501fe03 100644
--- a/icoscp_stilt/docs/index.md
+++ b/icoscp_stilt/docs/index.md
@@ -32,10 +32,10 @@ from icoscp_stilt import stiltstation
 ```
 
 Additionally, handling of country-association of the STILT stations is
-radically simplified with this new release. This was made possible by a
-change on the server side that guaranteed ISO-3166 alpha-2 country code
-association with every STILT station. The previously-utilized geo lookup of
-countries thus became redundant. Also, in the context of STILT, detailed
+simplified in this new release, which was made possible by a
+change on the server side that enforced ISO-3166 alpha-2 country code
+association with every STILT station. (Geo-lookup of country used previously
+thus became redundant). Also, for the STILT station metadata, detailed
 country metadata was deemed unnecessary, retaining the country code only,
 with a possibility of country name lookup. This resulted in a potential
 breaking change for the existing STILT-related Jupyter notebooks, namely
@@ -62,81 +62,3 @@ In general, library users are encouraged to switch to using the new
 functionality (`stilt` module) instead whenever possible (see the code
 examples below).
 
-## Getting started
-The library is published to PyPI.
-
-As stated above, the [legacy functionality](modules.md#legacy-modules) is
-still available, but for new code the developers are encouraged to consider
-the new module `icoscp_stilt.stilt` as the first choice.
-
-The following code demonstrates the new functionality.
-
-```Python
-from icoscp_stilt import stilt
-from icoscp_stilt.const import CP_OBSPACK_CO2_SPEC
-
-# list of stilt.StiltStation dataclass instances
-stations = stilt.list_stations()
-
-station_info_lookup = {s.id: s for s in stations}
-
-# example: Hyltemossa station, altitude 150 m
-htm_info = station_info_lookup['HTM150']
-htm_info
-```
-Here is the output of the above code.
-
-    StiltStation(
-        id='HTM150', name='Hyltemossa', lat=56.1, lon=13.42, alt=150,
-        countryCode='SE',
-        years=[2006, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022],
-        icosId='HTM', icosHeight=150.0
-    )
-
-```Python
-# years for which the station has calculation results
-htm_years = htm_info.years
-
-# grouped STILT time series results, all columns, as pandas DataFrame
-# can be slow when fetching first time after calculation
-htm_time_series_result = stilt.fetch_result_ts('HTM150', '2022-01-01', '2022-01-31')
-
-# list of time-series columns (~60)
-ts_columns = htm_time_series_result.columns
-
-# fetch selected columns only
-htm_ts_ch4_basics = stilt.fetch_result_ts('HTM150', '2022-01-01', '2022-01-31', columns=['isodate', 'ch4.stilt', 'metadata'])
-
-# find months for which calculation was run
-stilt.available_months('KRE250', 2022)
-
-# find all months for all years for which calculation was run
-htm_yearmonths = stilt.available_year_months(htm_info)
-
-# list footprint time slots that were computed for a station within date interval
-htm_slots_jan2022 = stilt.list_footprints('HTM150', '2022-01-01', '2022-01-31')
-
-# load footprint for one time slot
-htm_fp_example = stilt.load_footprint('HTM150', htm_slots_jan2022[0])
-
-# filter stations
-de_stations = [s for s in stations if s.countryCode == 'DE']
-
-# fetch observations for the German stations as numpy array dicts
-# interesting columns are requested explicitly (all returned otherwise)
-# using bare numpy gives maximum performance
-de_co2_numpy = stilt.fetch_observations(CP_OBSPACK_CO2_SPEC, de_stations, ['value', 'time'])
-
-# same as previous example, but returning pandas DataFrames instead
-# performance may be worse, especially on Jupyter
-de_co2_pandas = stilt.fetch_observations_pandas(CP_OBSPACK_CO2_SPEC, de_stations, ['value', 'time'])
-```
-
-## Getting help
-
-All the methods in the new `stilt` module have a Python documentation
-accessible by standard means, for example:
-
-```
-help(stilt.fetch_observations)
-```
\ No newline at end of file
diff --git a/icoscp_stilt/mkdocs.yml b/icoscp_stilt/mkdocs.yml
index aad344f..9ebd704 100644
--- a/icoscp_stilt/mkdocs.yml
+++ b/icoscp_stilt/mkdocs.yml
@@ -3,6 +3,7 @@ site_url: 'https://icos-carbon-portal.github.io/pylib/icoscp_stilt/'
 site_name: ICOS Carbon Portal icoscp_stilt Library
 nav:
     - About: index.md
+    - Getting started with the new icoscp_stilt: getting_started.md
     - Legacy modules: modules.md
 
 theme: