Document PISM's interpretation of forcing inputs

- 2D fluxes: piecewise-constant in time
- 2D (other): piecewise-linear in time
- scalar (with time bounds): piecewise-constant
- scalar (without time bounds): piecewise-linear

doc/sphinx/climate_forcing/time.rst

@@ -50,10 +50,18 @@ appropriate in forcing files.
 Using time bounds in forcing data
 +++++++++++++++++++++++++++++++++
 
-PISM interprets climate forcing data as piecewise-constant in time. A forcing file is
-required to contain time bounds corresponding to each record.
+To make :ref:`balancing the books <sec-mass-conservation>` possible, PISM interprets *fluxes*
+(examples: top surface mass balance, precipitation, sub shelf mass flux) as
+piecewise-constant in time. A forcing file is required to contain time bounds
+corresponding to each record.
 
-PISM follows the CF (Climate and Forecasting) meta-data conventions. The ``ncdump -h``
+Other 2D input fields (examples: ice surface temperature, near-surface air temperature)
+are interpreted as piecewise-linear in time.
+
+*Scalar* time-dependent inputs are interpreted as piecewise-constant in time if an input
+file contains time bounds and piecewise-linear otherwise.
+
+PISM supports time bounds specified according to `CF Conventions`_. The ``ncdump -h``
 output from a conforming file would look similar to:
 
 .. code-block:: none
@@ -74,7 +82,4 @@ output from a conforming file would look similar to:
 
 The :var:`time_bounds` variable stores the starting and the ending time for each interval
 in the forcing. This variable is assumed to have the same units as the :var:`time`
-variable it is associated with, which is why its arguments are not set in this example.
-
-Please see the `CF Conventions`_ document for details.
-
+variable it is associated with, which is why these attributes are not set in this example.