Orta Enlem İyonküresi için Rassal Alan Modeli Geliştirilmesi

Abstract

The ionosphere has utmost importance on the performance of Short Wave (SW) and satellite communication, satellite-based navigation, and remote sensing systems. It is an anisotropic, dispersive, and inhomogeneous plasma environment in both space and time. The structure and nature of ionosphere must be well understood for modeling and observation. This complex environment can be analyzed statistically. Statistical analysis and modeling give an advantage of understanding basic trend structure and enable prediction of the ionosphere. The most critical parameter which characterizes the ionosphere is electron density. The plasma frequency is the fundamental characteristic parameter, which depends on the elec\-tron density as a plasma environment. Propagation of radio waves in the ionosphere suffers several effects: attenuation, dispersion, polarization shift, time delay, Faraday Rotation, and phase delay due to its plasma structure. Total Electron Content (TEC), the critical frequency of F2 layer, foF2, the critical height of F2 layer, hmF2 and ionospheric slab thickness are essential parameters for investigating ionosphere. International Reference Ionosphere (IRI) and IRI-Plas are widely used ionospheric models. Empirical and deterministic models track low variability conditions and calm periods using hourly monthly medians. However, they have not enough performance on the space-time behavior of the ionosphere. In this dissertation, the statistical trend structures of the ionospheric parameters are analyzed using their particular locations. Parametric Probability Density Functions (PDF) are estimated as foF2, hmF2, TEC, and slab thickness for annual, seasonal, monthly, and hourly. The trend behavior of ionospheric parameters is analyzed both space, time, and in-between relations of each parameter using IONOLAB-PDF. Hourly, monthly parametric PDFs are estimated predominantly Weibull and Lognormal. Stochastic electron density profiles are obtained by providing IONOLAB-PDF estimates into the IRI-Plas model as inputs. Therefore, the electron density profiles predicted with IRI-Plas became empirical and stochastic. Statistical realizations of electron density profiles with the update of IONOLAB-PDF are in $ 1\sigma $ for $ \%70 $ and $ 2\sigma $ for $ \%95 $. Besides, high-resolution, regional foF2 maps are estimated in near real-time for updating inputs of the IRI-Plas model for the first time. IONOLAB-CK is a high-resolution interpolation tool based on Kriging methods for ionospheric parameters. The performance of the developed method is investigated using synthetic surfaces by error metrics. The metric error levels remain bounded even the variability of the synthetic surface increases. IONOLAB-CK method improves the interpolation variance five times and increases the interpolation resolution by approximately ten times as compared to Ordinary Kriging. As first modality, spatially sparse sampled foF2 and second modality densely sampled TEC values are used in IONOLAB-CK for good, moderate, and bad ionospheric conditioned example days. High-resolution foF2 maps are obtained using IONOLAB-CK with low interpolation variance. Using high-resolution foF2 and TEC maps, ionospheric slab thickness maps are also obtained for the first time. Obtained high-resolution foF2 maps provide an update for the IRI-Plas model, which is used as a background ionospheric model for the ray tracing algorithm IONOLAB-RAY. Near real-time update provide of IRI-Plas model inputs, and using IONOLAB-RAY; optimum transmission frequency and transmission direction are estimated in both azimuth and elevation for SW communication. The results are compared with oblique ionogram measurements for an oblique transmission experiment between two ionosondes. After providing a high-resolution foF2 update for inputs of IRI-Plas as background ionospheric model, the propagation of both ordinary and extraordinary ray paths successfully estimated.