The ionospheric responses to the total solar eclipse on 2 July 2019 over low latitudes in southern South America are presented. Ionosonde observations were used within the totality path at La Serena (LS: 29.9°S, 71.3°W) and at Tucumán (TU: 26.9°S, 65.4°W) and Jicamarca (JI: 12.0°S, 76.8°W), with 85% and 52% obscuration, respectively. Total electron content (TEC) estimations over the South American continent were analyzed. The ionospheric impact of the eclipse was simulated using the Sheffield University Plasmasphere-Ionosphere Model (SUPIM) at the Instituto Nacional de Pesquisas Espaciais (INPE). The significant variability of the diurnal variations of the various ionospheric characteristics over equatorial and low latitudes on geomagnetically quiet days makes it difficult to unambiguously determine the ionospheric responses to the eclipse. Nonetheless, some specific issues can be derived, mainly using simulation results. The E and F1 layer critical frequencies and densities below 200 km are found to consistently depend on decreasing solar radiation. However, the F1 layer stratification observed at both TU and LS cannot be related to the eclipse or other processes. The F2 layer does not follow the changes in direct solar radiation during the eclipse. The SUPIM-INPE-modeled F region critical frequency and TEC are overestimated before the eclipse at LS and particularly at TU. However, these overestimations are within the observed large day-today variability. When an artificial prereversal enhancement is added, the simulations during the eclipse better reproduce the observations at JI, are qualitatively better for LS, and are out of phase for TU. The simulations are consistent with conjugate location effects.

We present the first prediction of the ionospheric response to the 14 December 2020 solar eclipse using the SUPIM-INPE model. Simulations are made for all known ionosonde stations for which solar obscuration is significant. The found response is similar to that previously reported for other eclipses, but it also shows a modification of the equatorial fountain transport that will impact the low latitudes after the event. In addition to the large reduction of electron concentration along the totality path (~4.5 TECu, 22%), a significant electron and oxygen ion temperature cooling is observed (up to~400 K) followed by lasting temperature increases. Changes of up to~1.5 TECu (~5%) are also expected at the conjugate hemisphere. These predictions may serve as a reference for eventual ionospheric measurements of multiple instruments and are leading to a better understanding of the ionospheric response to solar eclipses.

No abstract

The ionosphere behavior is mainly dependent on solar radiation and the geomagnetic configuration, but it is also affected by tides, neutral winds, and atmosphere waves. There are many kinds of waves in the atmosphere, with periods ranging from 2 years (quasi-biennial oscillations) to less than a second (infrasonic waves). These include planetary waves, acoustic waves, and internal gravity waves (Rishbeth, 1988) and are associated with auroral activity, orography, seismic activity, tsunamis, volcanic eruptions, explosions, large-scale tropospheric weather phenomena, and so on. This association indicates a strong coupling between the ionosphere and the lower layers of the atmosphere and the lithosphere (Hines, 1972;Rishbeth, 1988).In the case of lithospheric phenomena such as earthquakes, earthquake-associated tsunamis, and volcanic eruptions, ionospheric responses are observed from a few minutes after the events. Two kinds of waves have been reported: acoustic waves and gravity waves (Hines, 1960;Rishbeth & Garriott, 1969). Although they widely differ by frequencies, the main difference is that for acoustic waves, the restoring force arises from the compressibility of the atmosphere, while for gravity waves, it comes from the gravitational acceleration (Meng et al., 2019). Both wave types can be detected in the ionosphere because their amplitude increases exponentially as they propagate upward due to the decrease in the atmospheric density with height (Astafyeva, 2019;Hines, 1972). Acoustic waves are longitudinal waves and propagate vertically when generated, for example, by a sudden crustal motion during an earthquake. These waves propagate at a speed close to the speed of sound near the surface but slow down with height, reaching the ionosphere (∼300 km) in ∼8-9 min.On the other hand, internal gravity waves are transverse waves, such as those generated by tsunamis. They can only propagate obliquely in the atmosphere (Hines, 1972), with speeds close to the tsunami, thus reaching the ionosphere in 45-60 min (Astafyeva, 2019;Meng et al., 2019). When reaching the ionosphere, the disturbances generated by these acoustic and gravity waves propagate spatially to other regions and are called Traveling Ionospheric Disturbances (TIDs). Observed TIDs with horizontal propagation speeds between 200 and 300 m/s are

In this work, we evaluate the SUPIM-INPE model prediction of the 14 December 2020, total solar eclipse over the South American continent. We compare the predictions with data from multiple instruments for monitoring the ionosphere and with different obscuration percentages (i.e., Jicamarca, 12.0°S, 76.8°W, 17%; Tucumán 26.9°S, 65.4° W, 49%; Chillán 36.6°S, 72.0°W; and Bahía Blanca, 38.7°S, 62.3°W, reach 95% obscuration) due to the eclipse. The analysis is done under total eclipse conditions and non-total eclipse conditions. Results obtained suggest that the model was able to reproduce with high accuracy both the daily variation and the eclipse impacts of E and F1 layers in the majority of the stations evaluated (except in Jicamarca station). The comparison at the F2 layer indicates small differences (<7.8%) between the predictions and observations at all stations during the eclipse periods. Additionally, statistical metrics reinforce the conclusion of a good performance of the model. Predicted and calibrated Total Electron Content (TEC, using 3 different techniques) are also compared. Results show that, although none of the selected TEC calibration methods have a good agreement with the SUPIM-INPE prediction, they exhibit similar trends in most of the cases. We also analyze data from the Jicamarca Incoherent Scatter Radar (ISR), and Swarm-A and GOLD missions. The electron temperature changes observed in ISR and Swarm-A are underestimated by the prediction. Also, important changes in the O/N2 ratio due to the eclipse, have been observed with GOLD mission data. Thus, future versions of the SUPIM-INPE model for eclipse conditions should consider effects on thermospheric winds and changes in composition, specifically in the O/N2 ratio.