Multi-dimensional Verification Methodology of Ionospheric Gradient Observation during Plasma Bubble Events in the Brazilian Region

Abstract

Ionospheric activity in equatorial regions is known to be significantly more variable and intense than in mid-latitude regions. Prior to the initiation of Ground-Based Augmentation System (GBAS) service in the Brazilian region, it is necessary to make provisions for being sufficiently robust to all possible ionospheric anomalies through development of an ionospheric anomaly threat model. In the Brazilian region, ionospheric spatial gradients larger than the upper bounds of the Conterminous U.S (CONUS) threat model developed for the mid-latitude region are frequently observed in the presence of Equatorial Plasma Bubbles (EPBs). The higher bounds in the resulting ionospheric threat model have a significant effect on performance, availability, and the eventual approval of the system. Thus, verification of observed ionospheric spatial gradients is important since observations verified to be due to EPBs will be used to determine the upper bound of the ionospheric anomaly threat model for GBAS in Brazil. This paper applies a two-phase procedure of validating extremely large ionospheric gradients caused by low-latitude ionospheric anomaly events in the Brazilian region. This multi-dimensional approach rules out the possibility of receiver-instigated events, single satellite faults, and post-processing errors creating an apparent gradient that is nonexistent. However, the relatively smaller scale of EPBs and the sparse distribution of the Brazilian network stations makes it difficult to use the existing validation methods developed for mid-latitudes. Thus, this paper also proposes a new verification methodology based on the time-step method in order to augment the two-phase validation procedure. This methodology utilizes multiple satellites and stations that exhibit a similar trend of ionospheric delay to that of the threat candidate. The time-step method enables us to estimate ionospheric gradients over any short baseline distance and thus compensates for the lack of ionospheric observability caused by a sparse distribution of GPS reference stations. It visualizes the equatorial anomaly event in both the time domain and spatial domain by utilizing all available sources including regional ionospheric delay maps, IPP tracks, the motion of EPBs, the location of stations, and the occurrence time of large gradients. Using this procedure, we verified an extreme ionospheric gradient of 501.2 mm/km observed at reference stations SAVO and SSA1 viewing PRN 21 on 31 December 2013. Simultaneous ionospheric observation data from another satellite passing across the same EPB region was as high as about 360 mm/km. Six station-satellite pairs exhibited similar trend of ionospheric delays and observed severe ionospheric gradients above 300 mm/km using the time-step based techniques. Threat event visualization supported the fact that satellite-station pairs spread over several hundreds of kilometers in east-west direction were impacted by the same EPB moving eastward. These factors give us complete support to confirm that the extreme ionospheric gradient observed here is caused by a real ionospheric anomaly and not due to measurement errors or other faults. In addition to the particular case presented in this paper, we have also completed verification of other extreme EPB events whose gradients were larger than 500 mm/km. However, the full description of other events is beyond the scope of the paper.