deprecate clear sky default in modelchain

docs/sphinx/source/api.rst

@@ -170,6 +170,8 @@ Transposition models
    irradiance.reindl
    irradiance.king
 
+.. _dniestmodels:
+
 DNI estimation models
 ---------------------
 

docs/sphinx/source/introexamples.rst

@@ -134,14 +134,13 @@ The following code demonstrates how to use
 :py:class:`~pvlib.modelchain.ModelChain` objects to accomplish our
 system modeling goal. ModelChain objects provide convenience methods
 that can provide default selections for models and can also fill
-necessary input data with modeled data. In our example below, we use
-convenience methods. For example, no irradiance data is provided as
-input, so the ModelChain object substitutes irradiance from a clear-sky
-model via the prepare_inputs method. Also, no irradiance transposition
-model is specified (keyword argument `transposition` for ModelChain) so
-the ModelChain defaults to the `haydavies` model. In this example,
-ModelChain infers the DC power model from the module provided by
-examining the parameters defined for module.
+necessary input data with modeled data. For example, no air temperature
+or wind speed data is provided in the input *weather* DataFrame,
+so the ModelChain object defaults to 20 C and 0 m/s. Also, no irradiance
+transposition model is specified (keyword argument `transposition` for
+ModelChain) so the ModelChain defaults to the `haydavies` model. In this
+example, ModelChain infers the DC power model from the module provided
+by examining the parameters defined for module.
 
 .. ipython:: python
 
@@ -157,12 +156,12 @@ examining the parameters defined for module.
         times = naive_times.tz_localize(timezone)
         location = Location(latitude, longitude, name=name, altitude=altitude,
                             tz=timezone)
-        # very experimental
+        weather = location.get_clearsky(times)
         mc = ModelChain(system, location,
                         orientation_strategy='south_at_latitude_tilt')
         # model results (ac, dc) and intermediates (aoi, temps, etc.)
         # assigned as mc object attributes
-        mc.run_model(times)
+        mc.run_model(times=times, weather=weather)
         annual_energy = mc.ac.sum()
         energies[name] = annual_energy
 

docs/sphinx/source/modelchain.rst

@@ -24,7 +24,8 @@ Modeling with a :py:class:`~.ModelChain` typically involves 3 steps:
 1. Creating the :py:class:`~.ModelChain`.
 2. Executing the :py:meth:`ModelChain.run_model() <.ModelChain.run_model>`
    method with prepared weather data.
-3. Examining the model results that :py:meth:`~.ModelChain.run_model` stored in attributes of the :py:class:`~.ModelChain`.
+3. Examining the model results that :py:meth:`~.ModelChain.run_model`
+   stored in attributes of the :py:class:`~.ModelChain`.
 
 A simple ModelChain example
 ---------------------------
@@ -212,8 +213,13 @@ method, :py:meth:`~pvlib.modelchain.ModelChain.prepare_inputs`, computes
 parameters such as solar position, airmass, angle of incidence, and
 plane of array irradiance. The
 :py:meth:`~pvlib.modelchain.ModelChain.prepare_inputs` method also
-assigns default values for irradiance (clear sky), temperature (20 C),
+assigns default values for temperature (20 C)
 and wind speed (0 m/s) if these inputs are not provided.
+:py:meth:`~pvlib.modelchain.ModelChain.prepare_inputs` requires all irradiance
+components (GHI, DNI, and DHI). See
+:py:meth:`~pvlib.modelchain.ModelChain.complete_irradiance` and
+:ref:`dniestmodels` for methods and functions that can help fully define
+the irradiance inputs.
 
 Next, :py:meth:`~pvlib.modelchain.ModelChain.run_model` calls the
 wrapper methods for AOI loss, spectral loss, effective irradiance, cell

docs/sphinx/source/whatsnew/v0.6.2.rst

@@ -23,6 +23,9 @@ API Changes
   *ghi* or *zenith*.
 * Added *min_cos_zenith* and *max_zenith* keyword arguments to
   :py:func:`~pvlib.irradiance.erbs`. (:issue:`681`)
+* Deprecated :py:meth:`~pvlib.modelchain.ModelChain.prepare_inputs`
+  assumption of clear sky if no irradiance fields were provided.
+  (:issue:`705`, :issue:`707`)
 
 Enhancements
 ~~~~~~~~~~~~

pvlib/modelchain.py

@@ -810,19 +810,20 @@ def prepare_inputs(self, times=None, weather=None):
             Times at which to evaluate the model. Can be None if
             attribute `times` is already set.
         weather : None or DataFrame, default None
-            If ``None``, the weather attribute is used. If the weather
-            attribute is also ``None`` assumes air temperature is 20 C, wind
-            speed is 0 m/s and irradiation calculated from clear sky
-            data. Column names must be ``'wind_speed'``, ``'temp_air'``,
-            ``'dni'``, ``'ghi'``, ``'dhi'``. Do not pass incomplete irradiation
-            data. Use method
-            :py:meth:`~pvlib.modelchain.ModelChain.complete_irradiance`
-            instead.
+            If ``None``, the weather attribute is used. Column names
+            must be ``'dni'``, ``'ghi'``, ``'dhi'``, ``'wind_speed'``,
+            ``'temp_air'``. All irradiance components are required.
+            Assumes air temperature is 20 C and wind speed is 0 m/s if
+            not provided.
 
         Notes
         -----
         Assigns attributes: ``times``, ``solar_position``, ``airmass``,
         ``total_irrad``, `aoi`
+
+        See also
+        --------
+        ModelChain.complete_irradiance
         """
         if weather is not None:
             self.weather = weather
@@ -839,15 +840,19 @@ def prepare_inputs(self, times=None, weather=None):
             solar_position=self.solar_position, model=self.airmass_model)
 
         if not any([x in ['ghi', 'dni', 'dhi'] for x in self.weather.columns]):
+            warnings.warn('Clear sky assumption for no input irradiance is '
+                          'deprecated and will be removed in v0.7.0. Use '
+                          'location.get_clearsky instead',
+                          pvlibDeprecationWarning)
             self.weather[['ghi', 'dni', 'dhi']] = self.location.get_clearsky(
                 self.solar_position.index, self.clearsky_model,
                 solar_position=self.solar_position,
                 airmass_absolute=self.airmass['airmass_absolute'])
 
         if not {'ghi', 'dni', 'dhi'} <= set(self.weather.columns):
             raise ValueError(
-                "Uncompleted irradiance data set. Please check you input " +
-                "data.\nData set needs to have 'dni', 'dhi' and 'ghi'.\n" +
+                "Uncompleted irradiance data set. Please check your input "
+                "data.\nData set needs to have 'dni', 'dhi' and 'ghi'.\n"
                 "Detected data: {0}".format(list(self.weather.columns)))
 
         # PVSystem.get_irradiance and SingleAxisTracker.get_irradiance
@@ -903,12 +908,11 @@ def run_model(self, times=None, weather=None):
             Times at which to evaluate the model. Can be None if
             attribute `times` is already set.
         weather : None or DataFrame, default None
-            If None, assumes air temperature is 20 C, wind speed is 0
-            m/s and irradiation calculated from clear sky data. Column
-            names must be 'wind_speed', 'temp_air', 'dni', 'ghi', 'dhi'.
-            Do not pass incomplete irradiation data. Use method
-            :py:meth:`~pvlib.modelchain.ModelChain.complete_irradiance`
-            instead.
+            If ``None``, the weather attribute is used. Column names
+            must be ``'dni'``, ``'ghi'``, ``'dhi'``, ``'wind_speed'``,
+            ``'temp_air'``. All irradiance components are required.
+            Assumes air temperature is 20 C and wind speed is 0 m/s if
+            not provided.
 
         Returns
         -------

pvlib/test/test_modelchain.py

@@ -19,7 +19,7 @@
 import pytest
 
 from test_pvsystem import sam_data, pvsyst_module_params
-from conftest import fail_on_pvlib_version, requires_scipy
+from conftest import fail_on_pvlib_version, requires_scipy, requires_tables
 
 
 @pytest.fixture
@@ -153,7 +153,9 @@ def test_orientation_strategy(strategy, expected, system, location):
 def test_run_model(system, location):
     mc = ModelChain(system, location)
     times = pd.date_range('20160101 1200-0700', periods=2, freq='6H')
-    ac = mc.run_model(times).ac
+
+    with pytest.warns(pvlibDeprecationWarning):
+        ac = mc.run_model(times).ac
 
     expected = pd.Series(np.array([  183.522449305,  -2.00000000e-02]),
                          index=times)
@@ -374,14 +376,12 @@ def constant_spectral_loss(mc):
 @pytest.mark.parametrize('spectral_model', [
         'sapm', 'first_solar', 'no_loss', constant_spectral_loss
 ])
-def test_spectral_models(system, location, spectral_model):
-    times = pd.date_range('20160101 1200-0700', periods=3, freq='6H')
-    weather = pd.DataFrame(data=[0.3, 0.5, 1.0],
-                           index=times,
-                           columns=['precipitable_water'])
+def test_spectral_models(system, location, spectral_model, weather):
+    # add pw to weather dataframe
+    weather['precipitable_water'] = [0.3, 0.5]
     mc = ModelChain(system, location, dc_model='sapm',
                     aoi_model='no_loss', spectral_model=spectral_model)
-    spectral_modifier = mc.run_model(times=times,
+    spectral_modifier = mc.run_model(times=weather.index,
                                      weather=weather).spectral_modifier
     assert isinstance(spectral_modifier, (pd.Series, float, int))
 
@@ -493,6 +493,21 @@ def test_deprecated_07():
                    ac_model='snlinverter')
 
 
+@requires_tables
+@fail_on_pvlib_version('0.7')
+def test_deprecated_clearsky_07():
+    # explicit system creation call because fail_on_pvlib_version
+    # does not support decorators.
+    system = PVSystem(module_parameters={'pdc0': 1, 'gamma_pdc': -0.003})
+    location = Location(32.2, -110.9)
+    mc = ModelChain(system, location, dc_model='pvwatts', ac_model='pvwatts',
+                    aoi_model='no_loss', spectral_model='no_loss')
+    times = pd.date_range(start='20160101 1200-0700',
+                          end='20160101 1800-0700', freq='6H')
+    with pytest.warns(pvlibDeprecationWarning):
+        mc.prepare_inputs(times=times)
+
+
 @requires_scipy
 def test_basic_chain_required(sam_data):
     times = pd.date_range(start='20160101 1200-0700',