Global Solar Irradiance Data via API

Data are produced in the following way. Firstly, satellite imagery is taken from the 3rd generation weather satellites (5x in total globally, including the Himawari 8 satellite over Australia). The images have a scan duration of 10-15 mins and are provided
for 16 bands, each with a different scanning wavelength of outgoing terrestrial radiation. The model applied utilises the visible (band 3) and infrared (band 13) images. The images extracted, and re-project spatially, including for parallax errors. The radiation modelling begins with the visible and IR images from the satellites with
pixel size resolution of 1-2 km^2. Solar irradiance is derived following the general principles of the Heliosat-2 methodology. Clouds are decomposed into lower, middle and upper troposphere layers as separated by cloud top temperatures that are inferred through utilisation of the IR image identified cloud-top height and numerical weather model isobaric temperatures obtained from the Global Forecast System (GFS) deterministic model. Total cloudiness is derived through aggregation of the three cloud layers. Total cloud opacity (represented as an index between 0 and 1) is then derived using albedo differences between the lowest visible return value of albedo recorded for each pixel and a continuous running 50 day sample that is updated daily. Derived cloud opacity is then applied as a linear reduction to the Esra clear sky radiation model to produce an estimate of GHI (Global Horizontal Irradiance) at each pixel. The selection of the Esra model was based upon validation against a selection of clear sky radiation models in Australia [20]. Estimates of GHI are then decomposed into direct normal irradiance (DNI) and diffuse horizontal irradiance (DIF) through the Engerer2 separation model. Data are updated live on an AWS based platform and can be extracted via a free API account, available at the project webpage (http://solcast.com.au/solar-data-api/).