Global comparison between all sources: The figure shows a comparison between all available sources for 12 locations in Europe. All values correspond to the Meteonorm values (PVsyst internal default), except the first green bin, which shows the difference between Meteonorm and the average of all other sources (excluding Helioclim 2005).
We can observe some trends:
We cannot say which source is the most representative of reality (and which reality? - no one is able to predict future climate).
Meteonorm often gives lower values than the average. This means that simulations with default values in PVsyst will be rather conservative and generate prudent results for the final yield of the customer’s systems.
Except at Barcelona, there is a very little difference between Meteonorm's “old” (1961-1990) and “new” (1981-2000) data.
Data from PVGIS and WRDC (when available) seemmost compatible with Meteonorm data.
Satellight data seems to overestimate the average by 5% to 10% (except in Berlin and Roma).
The Helioclim data response seems to be chaotic. This is especially true for the 2005 values (Helioclim-2 hourly file),whcih are much higher than the other data, by a factor that is incompatible with the 2005 irradiance (see below for Geneva).
Satellight data: For other sites in Europe, the Satellight data are always far over the Meteonorm ones, with the exception of Berlin. This exception is not attributable to the Meteonorm value; as we can see on the global comparison plot above, the Satellight data for Berlin are significantly below the other Satellight data. We cannot explan this.
Climate evolution: We use a homogeneous sample of continuous measurements from the same source (ISM - Swiss Institute for Meteorology) for Geneva, from 1981 to 2007.
This shows that at Geneva, the annual dispersion stayed far below 5% (with only 3-4 exceptions in 20 years), but it has increased significantly since 2003. Of course, this is not necessary the case for other sites in Europe!