联合星地GNSS观测的全球电离层TEC反演与建模

电离层引起的电波信号传输时延和相位改变与电离层总电子含量（TEC）成正比，是射电天文观测、雷达成像和卫星相关系统的主要误差源。为了进行电离层校正，需要发展精确的电离层TEC模型。由地基全球导航卫星系统（GNSS）网反演的TEC构建全球TEC模型是目前最为常用的方法，但在海洋等地区，由于缺乏观测数据，TEC反演精度受到严重影响，从而模型精度受到严重制约。GNSS掩星电离层观测是一种新兴的手段，具有全球覆盖的特点，对地基GNSS观测是有效的补充。本项目将联合星地GNSS观测开展全球TEC的反演与建模研究。提出有效的数据融合策略，发展基于载波相位观测的全球TEC高精度反演方法，并由此获取2006-2020年的TEC。考察TEC一个多太阳周期内的时空变化及其与太阳活动的关系，构建以太阳为驱动的全球TEC模型，利用神经网络优化估算模型参数，评估模型的有效性和预测能力。

As a dispersive medium the ionosphere causes time delay or phase advance in transionospheric radio waves. It is the main error source for radio astronomy and satellite related systems. The time delay or phase advance is proportional to the total electron content (TEC) along the radio wave path. An accurate ionospheric TEC model is crucial to make ionospheric corrections in transionospheric radio systems. The TEC derived from ground-based Global Navigation Satellite Systems (GNSS) observation is the most commonly used data in constructing a global TEC model. However, there is a lack of observation over vast ocean that reduces the accuracy of conventional TEC derivation in these areas. GNSS radio occultation is a new method for ionospheric observation. It is able to supplement the ground-based GNSS observation with its global coverage. This project aims to derive and model the global TEC from the ground- and satellite-based GNSS data. A strategy is proposed to effectively integrate the two kinds of data. A precise global TEC derivation method is developed based on carrier phase observation. The spatiotemporal variations of the global ionosphere and their relations with the solar activity are examined with the derived TEC from 2006-2020. A global TEC model driven by solar activity is constructed by using the TEC covering a solar cycle. The model parameters are fitted by neural network technique. The performance and the predictability of the model are validated by comparing with TEC observation.