Modeling equatorial ionospheric vertical plasma drifts using machine learning

Abstract

Abstract We present the results of an effort to model quiet-time vertical plasma drifts in the low-latitude F-region ionosphere using the random forest machine learning technique. The model is capable of describing the climatological variation of the drifts as a function of universal time, day of the year, solar flux, and altitude (200–600 km). The model has been trained using measurements of the vertical plasma drifts made by the incoherent scatter radar of the Jicamarca Radio Observatory ( $$11.95^\circ \hbox { S}$$ 11 . 95 ∘ S , $$76.87^\circ$$ 76 . 87 ∘ W, $$\sim 1^\circ$$ ∼ 1 ∘ dip lat). In our analysis, we compare our machine learning model results with the Scherliess and Fejer (J Geophys Res 104:6829–6842, 1999) model (SF99 model), a widely used empirical model of the vertical drifts developed using a different set of Jicamarca measurements. We find that the machine learning model is able to capture the overall features of the diurnal variation of the equatorial drifts for different seasonal and solar flux conditions. The model is also capable of capturing the mean height variation of the drifts, particularly the height gradient enhancements that have been observed near sunrise and sunset. Finally, the model can easily be expanded and improved as more drift measurements are made and become available for training.

Keywords

• Ionosphere
• Drifts
• Equatorial
• Model
• Machine learning
• Random forest