Sources such as atmospheric or buried explosions and shallow earthquakes producing strong vertical ground displacements produce pressure waves that propagate at infrasonic speeds in the atmosphere. At ionospheric altitudes low frequency acoustic waves are coupled to ionispheric gravity waves and induce variations in the ionoispheric electron density. Global Positioning System (GPS) data recorded in Southern California were used to compute ionospheric electron content time series for several days preceding and following the January 17, 1994, M(sub w) = 6.7 Northridge earthquake. An anomalous signal beginning several minutes after the earthquake with time delays that increase with distance from the epicenter was observed. The signal frequency and phase velocity are consistent with results from numerical models of atmospheric-ionospheric acoustic-gravity waves excited by seismic sources as well as previous electromagnetic sounding results. It is believed that these perturbations are caused by the ionospheric response to the strong ground displacement associated with the Northridge earthquake.

GPS detection of ionospheric perturbations following the January 17, 1994, Northridge Earthquake

We present computer experiments of electrostatic solitary waves (ESW) observed by Geotail in the magnetotail. ESW correspond to broadband electrostatic noise, and they are excited through electron two-stream instabilities along a static magnetic field. We performed one-dimensional electrostatic particle simulations involving two electron beams and an ion beam traveling along the static magnetic field. We vary the density ratio of the electron beams and the thermal velocities of the electron and ion beams. The values of these parameters strongly affect diffusion processes of the electron beams, and accordingly, different types of electrostatic waves are generated. We studied four different cases: cold bistream instability, weak-beam instability, bump-on-tail instability, and warm bistream instability. For these electron beam instabilities, we performed two different runs with cold and hot ions, respectively. The cold bistream instability gives ESW for hot ions and ion acoustic waves for cold ions. The weak-beam instability gives Langmuir waves, while the bump-on-tail instability gives ESW. The amplitudes of the waves excited by the weak-beam and bump-on-tail instabilities are small and do not induce nonlinear decay to ion acoustic waves even in the presence of cold ions. The warm bistream instability gives electron hole modes regardless of the value of the ion temperature. The electron hole mode is a normal mode in the presence of a two-hump electron distribution, and it is regarded as narrowband electrostatic noise. It also leads to formation of ESW, if the phase velocity of the electron hole mode is much larger than the ion drift velocity. A necessary condition for ESW formation through the bump-on-tail instability is derived theoretically, and its significance to Geotail observations is discussed.

Electron beam instabilities as generation mechanism of electrostatic solitary waves in the magnetotail

https://www.ep.sci.hokudai.ac.jp/~heki/pdf/GPSionEPSL.pdf

Directivity and apparent velocity of the coseismic ionospheric disturbances observed with a dense GPS array

[1] We have investigated statistical characteristics of the nighttime medium-scale traveling ionospheric disturbances (MSTIDs) observed in 630-nm airglow images at two stations, Rikubetsu (43.5°N, 34.8°MLAT) and Shigaraki (34.9°N, 25.4°MLAT), in Japan for 1998–2000 near the solar maximum period. Most of the observed MSTIDs propagate southwestward in the images. The typical wavelength, velocity, period, and amplitude are 100–300 km, 50–100 m/s, 0.5–1.5 h, and 5–15%, respectively. Seasonal variations in these parameters are not clear. The occurrence rate has a major peak (50–60%) in summer that appears ∼2 months earlier at lower latitudes and a minor peak in winter. Similar occurrence characteristics are obtained from midlatitude spread-F signatures using multipoint ionosonde data in Japan, though the coincidence of the spread-F and the MSTIDs in airglow images is only 10–15%.

https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2002JA009491

Statistical study of nighttime medium‐scale traveling ionospheric disturbances using midlatitude airglow images

The numerous recent breakthroughs in machine learning (ML) make imperative to carefully ponder how the scientific community can benefit from a technology that, although not necessarily new, is today living its golden age. This Grand Challenge review paper is focused on the present and future role of machine learning in space weather. The purpose is twofold. On one hand, we will discuss previous works that use ML for space weather forecasting, focusing in particular on the few areas that have seen most activity: the forecasting of geomagnetic indices, of relativistic electrons at geosynchronous orbits, of solar flares occurrence, of coronal mass ejection propagation time, and of solar wind speed. On the other hand, this paper serves as a gentle introduction to the field of machine learning tailored to the space weather community and as a pointer to a number of open challenges that we believe the community should undertake in the next decade. The recurring themes throughout the review are the need to shift our forecasting paradigm to a probabilistic approach focused on the reliable assessment of uncertainties, and the combination of physics-based and machine learning approaches, known as gray-box.

/pdf/the-challenge-of-machine-learning-in-space-weather-12iowq5jx4.pdf

The Challenge of Machine Learning in Space Weather: Nowcasting and Forecasting

HF-Doppler observations of acoustic waves excited by the Urakawa-Oki earthquake on 21 March 1982

[1] A global Pc5 geomagnetic pulsation (period of ∼6 min) was observed coherently from the auroral to equatorial latitudes in the local time sector of 0700–2100 LT during 1834–1900 UT on 21 April 1993. The Pc5 at the dayside dip equator (Sao Luiz, Brazil, and Ancon, Peru) was characterized by an equatorial enhancement with an enhancement ratio of ∼4. In the afternoon sector (1200–1600 magnetic local time) the Pc5 at the auroral latitude was coherent with that at the dip equator within a time resolution of 3 s. The Pc5 amplitudes decreased sharply away from the auroral zone but were enhanced in the dayside dip equator, in a manner that resembled the latitudinal profile of a DP 2 magnetic fluctuation (period of ∼40 min) studied by Kikuchi et al. [1996] except for the enhanced amplitude in the auroral zone. In the dayside auroral oval the Pc5 fluctuations in the magnetic X/H component were reversed in phase between the morning and afternoon sectors. Additionally, the magnetic Y/D component fluctuations around noon were correlated with the dayside equatorial Pc5. At most dayside middle-/low-latitude stations the Pc5 fluctuations in the H component, which are regarded as evidence for magnetospheric compressions, were strong. However, near the dawn terminator the D component was dominant rather than the H component, suggesting the influences of ionospheric currents originating in the polar region. There was no significant azimuthal phase propagation away from noon for the auroral to equatorial Pc5 signals. The global observational results suggest that the dawn-dusk electric field in the polar ionosphere accompanying a pair of field-aligned currents extends instantaneously to the equatorial ionosphere and completes a DP 2–type ionospheric current system responsible for the global coherent Pc5.

https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2001JA900156

Global Pc5 caused by a DP 2–type ionospheric current system

A study of the numerical heating in electrostatic particle simulations

[1] On 28 February 1998, four quasi-periodic pressure pulses with an amplitude of a few nPa detected by ACE gave rise to periodic compressions of the magnetosphere with period of about 14 min. In concert with periodic compressed and expanded states of the magnetosphere forced directly by the pressure variation, a coherent geomagnetic field fluctuation with the same period appeared on a global scale and was recorded at stations located from polar to equatorial regions. Most ground-level geomagnetic field signatures on the dayside can be interpreted as the result of a global ionospheric current system, like the global Pc5 event examined by Motoba et al. [2002]. In the afternoon polar ionosphere covered with the dense magnetometer stations, a vortical current structure associated with pressure-induced field-aligned currents (FACs) is centered at 72° ± 1° and consists of a counterclockwise (clockwise) vortex in response to positive (negative) changes of solar wind pressure oscillation. Although the vortical current signatures are unclear in the morning sector, each afternoon vortex could pair with the morning one with opposite rotation. During this event, the interplanetary magnetic field (IMF) remained steady with a strong southward orientation (−10 nT or less). In addition to the pressure-induced FAC system, the steady southward IMF drives the dayside Region 1 (R1) current system, resulting in the familiar large-scale two-cell convection pattern in the ionosphere observed by SuperDARN radars. The SuperDARN convection patterns indicated that the ionospheric convection reversal boundary (CRB) in the afternoon was located in the range of 73° ∼ 77°N around 15 MLT. The ionospheric footprint of the pressure-induced FAC in the afternoon was found to be 1.5° ± 1.1° ∼ 4.0° ± 1.4° equatorward of the CRB. This suggests that the pressure-induced FAC is started inside the R1 current system originating from the outer magnetospheric boundary layer. We argue that the paired FAC system responsible for the global geomagnetic fluctuations on the ground arises from the oscillatory large-scale dynamical convection originating well inside the closed field lines in direct response to the quasi-periodic pressure variations, not from the localized undulations on the magnetopause nor from global eigenmode oscillations of the magnetospheric cavity.

https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2002JA009696

Dynamical response of the magnetosphere-ionosphere system to a solar wind dynamic pressure oscillation

In order to enhance the reproduction of the recovery phase Dst index of a geomagnetic storm which has been shown by previous studies to be poorly reproduced when compared with the initial and main phases, an artificial neural network with one hidden layer and error back-propagation learning has been developed. Three hourly Dst values before the minimum Dst in the main phase in addition to solar wind data of IMF southward-component Bs, the total strength Bt and the square root of the dynamic pressure, $sqrt {n{V^2}}$
 , for the minimum Dst, i.e., information on the main phase was used to train the network. Twenty carefully selected storms from 1972–1982 were used for the training, and the performance of the trained network was then tested with three storms of different Dst strengths outside the training data set. Extremely good agreement between the measured Dst and the modeled Dst has been obtained for the recovery phase. The correlation coefficient between the predicted and observed Dst is more than 0.95. The average relative variance is 0.1 or less, which means that more than 90% of the observed Dst variance is predictable in our model. Our neural network model suggests that the minimum Dst of a storm is significant in the storm recovery process.

/pdf/prediction-of-the-geomagnetic-storm-associated-dst-index-1epdtmtmhr.pdf

Takashi Okuzawa

Papers

HF-Doppler observations of acoustic waves excited by the Urakawa-Oki earthquake on 21 March 1982

Global Pc5 caused by a DP 2–type ionospheric current system

A study of the numerical heating in electrostatic particle simulations

Dynamical response of the magnetosphere-ionosphere system to a solar wind dynamic pressure oscillation

Prediction of the geomagnetic storm associated Dst index using an artificial neural network algorithm