Artificial neural network model of the dayside magnetopause: Physical consequences

Empirical model of the dayside magnetopause developed by means of Artificial Neural Network (ANN) package 'NeuroShell 2' is discussed. ANN model describes 3D shape of the magnetopause as a function of four external parameters: solar wind velocity, density, B(y) and B(z) components of interplanetary magnetic field. The model shows that the magnetopause shape has the dawn-dusk asymmetry under any solar wind conditions independently on the B(y) direction. The cusp region is modeled. Using ANN model an expression for subsolar point distance is obtained in terms of modified logistic functions. The magnetopause dynamics is interpreted as a transition between two regimes of the magnetopause formation that is controlled by B(z) component of interplanetary magnetic field.