Data Quality Improvements and Applications of Long-Term Monitoring of Ionospheric Anomalies for GBAS

The Long-Term Ionospheric Anomaly Monitoring (LTIAM) tool is an automated software package designed to analyze past data and support continuous ionospheric monitoring of both nominal and anomalous ionospheric spatial gradients. While automated measurement screening is included, large gradients observed by LTIAM require manual validation to confirm that they were caused by the ionosphere instead of faulty measurements or data recording. Ground stations with poor data quality thus add greatly to the burden of LTIAM processing. This paper develops an automated approach to data quality measurement for CORS and IGS ground stations. This method is used to identify stations that are poor according to multiple quality metrics. Thresholds are established for each quality metric, and stations violating one or more thresholds are removed from use by LTIAM unless their geographical position is sufficiently important. Use of this method with CORS stations in the Conterminous U.S. (CONUS) eliminates the almost 90% of spurious or false gradients while only excluding 16% of the over 1500 CORS stations in CONUS. This paper also investigates past CONUS ionospheric storm data to understand the distribution of anomalous spatial gradients. Examining LTIAM outputs on known storm days with gradients between 50 and 200 mm/km demonstrates that these smaller (but still anomalous) gradients are far more likely than extreme gradients above 200 mm/km. The continued use of LTIAM over the next solar peak should help us refine our knowledge of this distribution as well as the overall likelihood of large spatial gradients under anomalous ionospheric conditions. 1.0. INTRODUCTION An automated Long-Term Ionospheric Anomaly Monitoring (LTIAM) software package has been developed to support continuous ionospheric monitoring for the U.S. Local Area Augmentation System (LAAS) developed by the U.S. Federal Aviation Administration (FAA) in the Conterminous U.S. (CONUS). Continuous monitoring is needed to confirm the long-term validity of existing ionospheric threat models and support updates if necessary. This is of particular importance over the next few years, as the intensity of solar storms is expected to peak in 2013-15. Continuous monitoring using the LTIAM provides reliable ionospheric gradient statistics under typical as well as anomalous conditions. The LTIAM will also be utilized to build threat models for other regions where Ground-Based Augmentation Systems (GBAS) will be fielded. The LTIAM software enables automated post-processing of data continuously collected by GPS reference station networks. Ionospheric gradients over short-baseline distances of 5 – 40 km can be observed using data collected from the Continuously Operating Reference Stations (CORS) network, which has over 1800 stations as of 2011 in the U.S. territories and a few other countries compared to about 400 stations prior to 2004. However, as the total number of stations increases, the number of stations with poor GPS data quality also increases. CORS receivers and antennas are fielded by multiple organizations in various environments; some good, some not-so-good. Poor-quality data degrades the accuracy of ionospheric delay estimates and produces too many faulty anomaly candidates, meaning apparent anomalies that are actually due to measurement or data errors. This paper presents a comprehensive method of GPS data quality determination to select CORS stations with highquality data. A series of algorithms provide information about measurement quality, including cycle slips, receiver noise and multipath, and the daily number of observations (including measurement gaps). Cycle slip detection methods already developed as a part of LTIAM preprocessing have been upgraded by incorporating cycle slips detected using multipath estimates. Multipath on code observations is computed by linear combinations of L1 C/A-code, L1 P-code, and L2 P-code observations. Carrier multipath and receiver noise are estimated using an adaptive filter algorithm. Thresholds are derived for each of these metrics, and stations which lie outside the threshold of one or more metrics are excluded from LTIAM measurement processing unless they are recovered by a secondary check on their location. Stations whose location for observing the ionosphere is sufficiently important are retained despite poor data quality. When implemented on recent CORS station data in CONUS on nominal ionospheric days, the removal of relatively few stations is needed to dramatically reduce the number of false anomaly outputs from LTIAM. The results are more reliable LTIAM results and a reduced manual analysis burden in examining the remaining apparent anomalies. In this paper, Section 3.0 illustrates the problem of poor data quality; Section 4.0 explains the automated data-quality analysis methodology in detail, and Section 5.0 shows the results of applying this method to CORS stations in CONUS. The upgraded LTIAM software allows us to better understand past ionospheric anomalies as well as monitor future ones. This paper re-examines the record of ionospheric “storm” days in CONUS from 2000-2005 to better understand the distribution of spatial gradients under anomalous ionospheric conditions. This database has been thoroughly searched manually and by earlier versions of LTIAM for “extreme” gradients above 200 mm/km that drive the GBAS threat space and have the potential for harm. Here, this database is searched for less-extreme gradients between 50 and 200 mm/km that are still anomalous but much less threatening to GBAS. As expected, far more gradients are found at these lower levels, and within this range, lower gradients are more probable than higher ones. Section 6.0 describes this analysis and explains how to update it with future data, and Section 7.0 concludes the paper.