Comparing different TMY/EPW providers : inconsistency of data?

lionpeloux · January 25, 2023, 3:27pm

Ok. Let me try to sum up :

solar irradiance is the power received by a 1m2 surface at a given time stamp (in W/m2)
solar irradiation is the energy received by a 1m2 surface at a given time period (in Wh/m2)
however, in practice these terms are often confused …

In the EPW format documentation you referenced (here) it is clearly stipulated that:

Direct Normal Radiation is the amount of solar radiation in Wh/m2 received […] during the number of minutes preceding the time indicated.

This is aligned with the TMY3 specification (here) :

Direct normal irradiance is the amount of solar radiation in Wh/m2 received in a collimated beam on a surface normal to the sun during the 60-minutes period ending at the timestamp.

In the docstring of the epw.py module it is written that the last time stamp in the series are forwarded at the beginning of the series.

github.com

ladybug-tools/ladybug/blob/96d598c09036d1b334914a840b7c1a78faf51b9b/ladybug/epw.py#L40-L58


      
          In order to handle the fact that EPW values run from 1:00 on Jan 1 to
          midnight of the following year, all timeseries data will have the ending
          value moved to the start of the hourly data collection upon import from
          the .epw file to an EPW object. This ensures that the datetimes of the data
          collections align with the times listed in the .epw file. This also allows
          the use of "true" datetimes in the hourly data collections, which start from
          0:00 rather than ending with 24:00. When saving the data from a EPW object
          back to an .epw file, the start date time at 0:00 on Jan 1 will be moved to
          the end of the file, ensuring imported .epw files can be saved back to their
          original state without any loss of data.
          
          The only exceptions to this movement of the last datetime are the timeseries
          data sets for radiation and illuminance. These data sets are left as they are
          when imported/exported to/from the .epw file given that they are meant to
          be an accumulation or average over the previous hour in the .epw file.
          Accordingly, such values can be thought of as an accumulation or average over
          the following hour when they are found within an hourly data collection of
          an EPW object. So the radiation with the 12:00 datetime in the hourly data
          collection represents the accumulated radiation in between 12:00 and 13:00.

This is how I interpret the comment : would you conform it’s correct ?!