The Meteosat series of satellites are equatorial geostationary satellites operated for meteorological data collection by EUMETSAT. Meteosat solar data is obtained under the Meteosat First Generation (MFG) and the Meteosat Second Generation (MSG) programs. EUMETSAT is responsible for the launch and operation of the satellites and for delivering satellite data to end-users.
Data for renewable energy resource assessment, particularly solar energy, can be accessed from the EUMETSAT Satellite Application Facility (SAF) for Land Surface Analysis (LSA) located here. This article provides essential reference data for working with Meteosat solar data.
Meteosat History and Satellite Metadata
The satellites which comprise the MFG and MSG programs have provided data since the 1970s, starting with data collection over the Prime meridian (PRIME), and then transitioning to Indian Ocean Data Collection (IODC). The geographic coverage of the active PRIME and IODC satellites are shown at left. The spatial resolution of the MFG and MSG satellites is 3.1 km and 5.6 km at Nadir, and decreases as we move away from the geostationary point. Other satellite reference data is listed below:
|Launch Date||End Date||Longitude||Spatial Res|
Meteosat Solar Assessment Variables
The primary variable of interest for solar resource assessment is the down-welling surface short-wave radiation flux (DSSF), which refers to the radiative energy in the wavelength interval 1 reaching the Earth’s surface per time and surface unit. Accurate knowledge of the distribution of solar radiation at the surface is essential for understanding climate processes at the Earth-atmosphere interface. The method used to estimate DSSF is defined by the SAF Surface Solar Irradiance Product Manual 2002 and 2005.
DSSF represents Global Horizontal Irradiance (GHI) and depends on the solar zenith angle, on cloud coverage, and to a lesser extent on atmospheric absorption and surface albedo. The cloud mask represents a critical piece of information for estimating DSSF since the amount of radiation reaching the surface is considerably reduced. The brighter the clouds appear in any satellite image, the more radiation is reflected by them and less is the radiation that reaches the ground. In practice, LSA-SAF has two methods for estimating DSSF, clear sky versus cloudy models. The cloud model, in turn, can adjust in any time and space unit to 12 different cloud types.
Other data available from LSA-SAF is listed below:
- Cloud Mask: The cloud mask data reported by LSA-SAF includes information on atmospheric sand clouds, which is essential to solar production and O&M planning. In practice, LSA-SAF data is historical and can be readily obtained when filtering cloud type.
- Temperature: LSA-SAF also supplies temperature data, which is time-synchronized with the solar data collected. The combination of GHI and temperature data allows users to fully define the power curve supporting PV solar production at any point in time.
- Solar Angles: PV power curve data can then be combined with solar zenith and azimuth angles from the LSA-SAF data to generate power output estimates for different types of PV collection modules (e.g. fixed, single axis or dual axis).
Finally, if the objective is to work with time series data for the sunlight components, Direct Normal Irradiance (DNI) and Diffuse Horizontal Irradiance (DHI), then secondary calculations by the user are required to decompose sunlight into its’ components since this data is not reported by LSA-SAF.
Satellite Selection and Data Downloads
PRIME satellites are the preferred data source for DSSF in the MENA region and the GCC countries in particular, owing to the higher temporal and spatial resolutions of the MSG program. In practice, SAF-LSA divides processing of the PRIME data coverage area into four regional windows, as shown at left.
Each MSG earth image contains 3,712 pixel lines and 3,712 pixel columns, for a total of 13,778,944 pixels per unit of time. Time series data extraction by pixel requires users to identify the row and column location that corresponds to their area of interest, recognizing that row and column indices both start at 1 in the North and West of each continental window.
The table below simplifies data extraction by defining the the pixel row and column indices for each region:
|Region||Initial Row||Last Row||Initial Column||Last Column||Row Size||Column Size||Pixel Count|