Alex T. Chartier

Ionospheric data assimilation and forecasting during storms

Ionospheric storms can have important effects on radio communications and navigation systems. Storm time ionospheric predictions have the potential to form part of effective mitigation strategies to these problems. Ionospheric storms are caused by strong forcing from the solar wind. Electron density enhancements are driven by penetration electric fields, as well as by thermosphere-ionosphere behavior including Traveling Atmospheric Disturbances and Traveling Ionospheric Disturbances and changes to the neutral composition. This study assesses the effect on 1 h predictions of specifying initial ionospheric and thermospheric conditions using total electron content (TEC) observations under a fixed set of solar and high-latitude drivers. Prediction performance is assessed against TEC observations, incoherent scatter radar, and in situ electron density observations. Corotated TEC data provide a benchmark of forecast accuracy. The primary case study is the storm of 10 September 2005, while the anomalous storm of 21 January 2005 provides a secondary comparison. The study uses an ensemble Kalman filter constructed with the Data Assimilation Research Testbed and the Thermosphere Ionosphere Electrodynamics General Circulation Model. Maps of preprocessed, verticalized GPS TEC are assimilated, while high-latitude specifications from the Assimilative Mapping of Ionospheric Electrodynamics and solar flux observations from the Solar Extreme Ultraviolet Experiment are used to drive the model. The filter adjusts ionospheric and thermospheric parameters, making use of time-evolving covariance estimates. The approach is effective in correcting model biases but does not capture all the behavior of the storms. In particular, a ridge-like enhancement over the continental USA is not predicted, indicating the importance of predicting storm time electric field behavior to the problem of ionospheric forecasting.

Three‐dimensional modeling of high‐latitude scintillation observations

Global Navigation Satellite System (GNSS) signals exhibit rapid fluctuations at high and low latitudes as a consequence of propagation through drifting ionospheric irregularities. We focus on the high latitude scintillation problem, taking advantage of a conjunction of EISCAT Incoherent Scatter Radar (ISR) observations and a GPS scintillation monitor viewing the same line-of-sight. Just after 20:00 UT on 17 October 2013, an auroral E-region ionization enhancement occurred with associated phase scintillations. This investigation uses the scintillation observations to estimate the ionospheric electron density distribution beyond the spatial resolution of the ISR (5 - 15 km along the line-of-sight in this case). Following the approach of Deshpande et al. [2014], signal propagation is modeled through a specified density distribution. A multiple phase screen propagation algorithm is applied to irregularities conforming to the description of Costa and Kelley [1977] and constrained to match the macroscopic conditions observed by the ISR. A 50-member ensemble of modeled outputs is approximately consistent with the observations according to the standard deviation of the phase (σp). The observations have σp = 0.23 radians, while the ensemble of modeled realizations has σp = 0.23 + 0.04 -0.04. By comparison of the model output with the scintillation observations, we show that the density fluctuations cannot be a constant fraction of the mean density. The model indicates that E-region density fluctuations whose standard deviation varies temporally between 5 - 25% of the mean (ISR-observed) density are required to explain the observed phase scintillations.

Ionospheric imaging in Africa

Accurate ionospheric specification is necessary for improving human activities such as radar detection, navigation, and Earth observation. This is of particular importance in Africa, where strong plasma density gradients exist due to the equatorial ionization anomaly. In this paper the accuracy of three-dimensional ionospheric images is assessed over a 2 week test period (2-16 December 2012). These images are produced using differential Global Positioning System (GPS) slant total electron content observations and a time-dependent tomography algorithm. The test period is selected to coincide with a period of increased GPS data availability from the African Geodetic Reference Frame (AFREF) project. A simulation approach that includes the addition of realistic errors is employed in order to provide a ground truth. Results show that the inclusion of observations from the AFREF archive significantly reduces ionospheric specification errors across the African sector, especially in regions that are poorly served by the permanent network of GPS receivers. The permanent network could be improved by adding extra sites and by reducing the number of service outages that affect the existing sites. Key Points Ionospheric image quality in Africa is assessed Simulated and real data are both used An extended receiver network greatly improves accuracy

Modeled and observed equatorial thermospheric winds and temperatures

Thermospheric winds and temperatures must be correctly specified to understand the impacts of lower atmosphere processes on the upper atmosphere and to measure the global effects of high-latitude magnetospheric processes. Fabry-Perot interferometers can estimate these parameters by measuring the characteristic 630.0 nm emission that is produced at around 250 km altitude. These sophisticated instruments exist at only a few locations globally, so models are often employed to provide wind and temperature estimates elsewhere. This study is composed of two parts. First, observing system simulation experiments estimate the accuracy of Fabry-Perot interferometer observations using the Thermosphere-Ionosphere-Electrodynamics General Circulation Model (TIEGCM) and the Whole Atmosphere Community Climate Model eXtended (WACCM-X). Atmospheric observational error sources are found to be very small across two test periods (September 2000 and September 2010) and using two different “truth” models. The largest magnitude wind observation error is found to be 16.9 m/s, root-mean-square errors are 2.3 m/s, and the bias is 0.9 m/s. The largest-magnitude temperature observation error is found to be 63.7 K, root-mean-square errors over the test period are 6.7 K, and the bias is 2.8 K. Modeled redline emission altitudes vary by over 100 km, far more than was expected. Second, several models (TIEGCM, WACCM-X, the Horizontal Wind Model, and the Mass Spectrometer Incoherent Scatter model) are assessed using interferometer winds and temperatures from Cariri and Cajazeiras, Brazil, as ground truth. In the best cases, the models reproduce wind variability without systematic biases but show no ability to predict instantaneous values, although temperatures are modeled more accurately.

Improving ionospheric imaging via the incorporation of direct ionosonde observations into GPS tomography

The Multi Instrument Data Analysis system, MIDAS is an algorithm that images the ionosphere in three or four dimensions and was originally developed by Mitchell and Spencer (2003) [1]. MIDAS often operates using GPS measurements of slant Total Electron Content, but Chartier et al. [2012] [2] showed that incorporating ionosonde data into the algorithm could improve imaging of the ionosphere in the vertical dimension. Here we extend the technique to incorporation of multiple ionosondes, the key problem is to transition horizontally between regions of different peak height and changing densities. This approach is validated via comparisons with independent satellite data.