Time Series and TMY data

There is a noticeable change in the DNI between data recorded before 2003 and afterwards. Why is this?

Modified on: Tue, 17 Jan, 2017 at 16:02

One of the data sets used in the Solargis model is Aerosol Optical Depth (AOD) data calculated from MACC-II database (operated by ECMWF in Reading, UK). AOD is used for calculating atmospheric conditions, such as turbidity.

The MACC-II model delivers a set of atmospheric data with a time step of 6 hours. We average 6-hourly AOD data into daily values, which makes it possible to map the day-by-day variability in the state of the atmosphere. This also helps to identify extreme values; especially DNI is sensitive to changes of AOD.

However, the historical database is available only from 1 Jan 2003 onwards (there are no reliable global atmospheric data for older times). Thus, for years preceding 2003, we can only use the averaged monthly AOD values calculated from years 2003 to 2012. Using monthly averages eliminates a possibility of identifying extremes. See below resources for more information:

http://solargis.com/assets/publication/2012/Cebecauer-Suri--SolarPaces2012--Correction-of-Satellite-Derived-DNI-Time-Series-Using-Locally-Resolved-Aerosol-Data.pdf

Was this answer helpful?