Validation of the Ionospheric Forecast Model (IFM)

📖 The Scoop

The purpose of this research was to validate the Ionospheric Forecast Model (IFM) Version 3 to assess its suitability and usefulness as an operational tool. The Ionospheric forecast model is a first principles computer model designed to forecast the state of the global ionosphere to 24 hours. The scope was limited to a comparison of the F2 layer critical frequency (foF2) and peak electron density (hmF2) against observed ionosonde data. The model was run with global solar and geomagnetic indices and Information from Digitial Ionospheric Sounding System (DISS) observations as inputs. The DISS observations were input through the Parameterized Real-Time Ionospheric Specification Model (PRISM). The IFM was run for a total of 25 days. It was run twice for each 24-hour period: once by starting with only an initial specification of the ionosphere from an empirical model, and again, with an initial specification from the PRISM model with DISS data as an input. As many as 50 DISS stations were used as inputs to the PRISM model, and FM output was compared against the observations from 10 of the 50 stations. Analysis of the foF2 showed that the IFM output was, on average, was within 1 MHz of observed values, and showed a slight bias to over forecast. Correlation between predictions and observations was generally about 0.8. Analysis of the hmF2 data showed a tendency for the model to under forecast the values. hmF2 forecasts were characterized by large errors and poor correlation between predicted and observed values. There was surprisingly little difference between the results with DISS input and without. The model does not seem to heavily weight the observed data input through PRISM.

Genre: Science / Earth Sciences / Meteorology & Climatology (fancy, right?)