Artificial Neural Networks based global Ionospheric Model

Researchers from Indian Institute of Geomagnetism (IIG), Navi Mumbai, an autonomous institute of the Department of Science & Technology, Govt. of India, have developed a global model to predict the ionospheric electron density with larger data coverage, a crucial need for communication and navigation.

• The new Artificial Neural Networks based global Ionospheric Model (ANNIM) is developed using long-term ionospheric observations to predict the ionospheric electron density and the peak parameters.
• ANNs replicate the processes in the human brain (or biological neurons) to solve problems such as:
  • pattern recognition
  • classification
  • clustering
  • generalization
  • linear and nonlinear data fitting
  • time series prediction
• Currently, very few attempts have been made to model the global ionosphere variability using ANNs. 

How they did it?

• The researchers developed a neural network-based global ionospheric model by using:
  • an extensive database consisting of nearly two decades of global Digisonde (an instrument that measures real-time on-site electron density of the ionosphere by sending the radiofrequency pulses)
  • Global Navigation Satellite System (GNSS) radio occultation
  • topside sounders observations
• These datasets were processed with various quality control measures to eliminate spurious data points (outliers) and prepared for the training.
• Day number, Universal Time, latitude, longitude, F10.7 index (responsible for Photo-ionization), Kp (represents the disturbed space weather conditions), magnetic declination, inclination, dip latitude, zonal and meridional neutral winds were taken as inputs in the study.
• The target (output) of ANNs is the electron density as a function of altitude for any given location and time.
• The data was trained with the ANNs using high-performance computer at IIG to develop the ANNIM.

What is ionosphere?

• A dense layer of molecules and electrically charged particles, called the ionosphere, hangs in the Earth's upper atmospherestarting at about 35 miles (60 kilometers) above the planet's surface and stretching out beyond 620 miles (1,000 km). 
• Ionosphere overlaps the mesosphere, thermosphere, and exosphere. It is a very active part of the atmosphere, and it grows and shrinks depending on the energy it absorbs from the sun.
• In the ionosphere, charged particles are affected by the magnetic fields of both Earth and the sun.
• This is where auroras Those are the bright, beautiful bands of light that you sometimes see near Earth’s poles.
• They are caused by high-energy particles from the sun interacting with the atoms in this layer of our atmosphere.

Why tracking the variability is important?

• Tracking the variability of the Ionosphere is important for communication and navigation.
• The ionospheric variability is greatly influenced by both solar originated processes and the neutral atmosphere origin, therefore, difficult to model.
• Scientists have tried to model the ionosphere using theoretical and empirical techniques; however, the accurate prediction of electron density is still a challenging task.
• In recent years, the Artificial Neural Networks (ANNs) are showing potential to handle more complex and non-linear problems.
• Keeping these aspects in mind, a novel machine learning approach was implemented by the IIG team in the ionospheric model development using global ionospheric observations.

How will it benefit?

• It can capture the general morphological features of the ionosphere during the disturbed space weather periods, such as geomagnetic storms which occurs when the magnetic cloud originated from Sun (known as Coronal Mass Ejection (CME)) interacts with the Earth’s magnetosphere.
• The model may be utilized as a reference model in the ionospheric predictions and has potential applications in calculating the Global Navigation Satellite System (GNSS) positioning errors.