The effects of thunderstorm-generated atmospheric gravity waves on mid-latitude F-region drifts

https://doi.org/10.1016/j.jastp.2009.07.006Get rights and content

Abstract

Superposed epoch analysis (SEA) was used to examine ionospheric drift velocities measured by a digital ionosonde located at the mid-latitude station Bundoora (145.1°E, 37.7°S geographic), near Melbourne. The control times for the SEA were the times of cloud-to-ground (CG) lightning strokes measured from August 2003 to August 2004 by the World Wide Lightning Location Network (WWLLN). Statistically, regions of concentrated lightning activity migrated from west to east across Bundoora, and the stroke frequency was higher the day prior the activity reached the station, and lower on the day after it passed to the east. For the SEA, CG strokes were separated into four directional quadrants centred on north, south, east and west. No SEA results are shown for the south quadrant due to the relatively low detection frequency of strokes across the Southern Ocean (6% of all events). The strongest downward vertical perturbations in F-region drifts, −4.5 m s−1, were found for lightning located towards the west during −30 to −16 h (i.e., the afternoon prior the activity passed near the station at t=0 h). The downward perturbation decreased in amplitude to −1.5 m s−1 for lightning located towards the north during −6–+6 h, and was weakest (−0.7 m s−1) for lightning located towards the east during +16–+28 h (i.e., the next afternoon). There were directionally consistent perturbations in the drift azimuths associated with the lightning located in their respective quadrants; lightning located to the west of the station caused eastward azimuth enhancements, northward lightning caused southward enhancements, and eastward lightning caused westward enhancements. Velocity magnitudes and fluctuations tended to increase during the passage of lightning. The observed responses were stronger when the SEA was performed with data selected using time windows of <2 min on either side of each lightning stroke. However, they persisted at longer time scales and were strong when thunderstorm onsets (instead of lightning times) were used as controls. Our results can be explained by thunderstorm-generated atmospheric gravity waves (AGWs) which subsequently gave rise to medium-scale travelling ionospheric disturbances (MSTIDs), with the lightning strokes acting merely as a proxy for this coupling. The prevailing thermospheric winds were flowing from east to west across the study region, and may have acted as a directional ‘filter’ for the MSTIDs, allowing waves generated in the west quadrant to reach the station and preventing those generated in other quadrants. Displacement of the MSTIDs in the direction anti-parallel to mean neutral wind flow has been observed by (Waldock, J.A., Jones, T.B., 1986. HF Doppler observations of medium-scale travelling ionospheric disturbances at mid-latitudes. Journal of atmospheric and terrestrial physics 48(3), 245–260).

Introduction

Space weather variability may affect thunderstorm activity and the Earth's climate (Rycroft, 2006; Tinsley et al., 2007; Burns et al., 2008). Cosmic ray produced ions may affect nucleation and growth characteristics of cloud particles (Carslaw et al., 2002). During geomagnetic storms, precipitation of energetic particles into the lower atmosphere is enhanced which increases the electrical conductivity of the atmosphere, and thus also modifies the global electrical circuit (GEC) (Tinsley et al., 2007). A recent statistical study of the atmospheric pressure variations at 11 Antarctic and 7 Arctic sites found significant correlation with changes in the GEC due to the coupling with the interplanetary magnetic field (Burns et al., 2008). So, it is likely the GEC is an aspect of the physical mechanism linking space weather and tropospheric weather.

Kazimirovsky et al. (2003), Siingh et al. (2005), and Rycroft (2006) reviewed the upward influences of the atmosphere on space. Cloud-to-ionosphere (CI) electrical discharges influence the temperature (Sharma et al., 2004), the ion density (Taranenko et al., 1993) and the chemical composition (Hiraki et al., 2008) of the ionosphere. There is a multitude of associated transient (millisecond) optical phenomena including sprites, elves, blue jets and blue starters, and other newly discovered phenomena (Lyons et al., 2003). Whistlers produced by lightning can decrease the pitch angle of electrons gyrating in the radiation belts, causing precipitation into the upper atmosphere (Abel and Thorne, 1997).

Electric fields from impulsive lightning discharges penetrate to great altitudes above thunderstorms. These fields cause electron heating in the mesosphere and lower ionosphere (Pasko et al., 1995). Woodman and Kudeki (1984) observed explosive spread-F coincident with lightning induced sferics, implying that electric field transients can penetrate and destabilise the F-region plasma. Kelley et al. (1985) reported lightning induced ac electric fields of tens of mV m−1 measured at an altitude of 150 km using sensors onboard rockets and later reported similar amplitudes at 300 km altitude (Kelley et al., 1990). These fields are very large by ionospheric standards and it was suggested that their ionospheric electrostatic signatures might be detected using large steerable radars (Kelley et al., 1990).

Whilst observations support the direct penetration of intense electric field transients to F-region altitudes, no rocket-based or radar experiments have found evidence for the direct penetration of the dc component of intense electric fields associated with thunderstorms. The early theoretical analysis of Park and Dejnakarintra (1973) showed that upward directed dc electric fields immediately above a typical thunderstorm of charge ~50 °C would be 500 V m−1, which attenuates almost exponentially up to the D-region to a magnitude of ~10 mV m−1, and from thereon decreases gradually with the amplitude reaching to ~0.65 mV m−1 or less at 150 km altitude. However, recent measurements of thunderstorm associated dc electric fields were at least an order higher than those predicted in Park and Dejnakarintra (1973). For example, Blakeslee et al. (1989) recorded upward thunderstorm-generated electric fields in excess of 5000 V m−1 at an altitude of 20 km and Meek et al. (2004) measured electric fields in the order of 1 V m−1 at 70 km altitudes, both lasting for several minutes. These differences may be due to the conductivity and cloud models used in Park and Dejnakarintra (1973). There is a still a possibility that significant dc electric fields penetrate and affect the dynamics of the F region ionosphere.

Numerous authors (Francis, 1975; Kelley et al., 1990; Waldock, 1996; Šauli and Boška, 2001; Sentman et al., 2003) have suggested that atmospheric gravity waves (AGWs) play an important role in the transport of energy and momentum from the troposphere to the ionosphere. Mid-latitude sporadic-E layers have long thought to be controlled by the wind-shears which arise from upward propagating AGWs (Whitehead, 1989). Davis and Lo (2008) presented statistical evidence of enhancements in the electron concentration of sporadic-E (Es) layer associated with the negative polarity cloud-to-ground (CG) strokes. Their results may have been explained by the AGWs launched by the approaching thunderstorms.

Kelley (1997) reported rocket-based measurements providing evidence for small-scale F-region plasma structure associated with AGWs generated by severe thunderstorm activity. Varney et al. (2009) recently suggested that ionospheric electric fields may form in association with AGWs. Waldock and Jones (1987) provided experimental evidences of a moderate correlation between the intensity of meteorological jets and occurrence frequency of medium-scale travelling ionospheric disturbances (MSTIDs) in the mid-latitude ionosphere. Although a few case studies have revealed interactions between thunderstorms and the ionosphere, very few comprehensive statistical analyses have been reported. To what extent do AGWs generated by tropospheric storms propagate up to ionospheric altitudes and then affect the F-region ionosphere?

In this work, superposed epoch analysis (SEA) was used to examine part of a 5.5 year (February 1999–August 2004) digital ionosonde (DPS-4) database complied at the southern mid-latitude station, Bundoora, Victoria (145.1°E, 37.7°S geographic). The control times for the SEA were the times of CG lightning strokes measured by the World Wide Lightning Location Network (WWLLN) (Dowden et al., 2008) from August 2003 to August 2004; the first year WWLLN was fully operational. Parkinson et al. (2001) and Kumar et al. (2008) previously analysed the geomagnetic effects on the Bundoora database. Later, Kumar et al. (2009) investigated the response of the F-region vertical drift due to CG lightning strokes located within 600 km of Bundoora. They found that the F-region ionosphere descends during thunderstorms, with maximum vertical velocities of ~−0.7 m s−1 at −21 and +3 h with respect to the lightning control times, t=0 (i.e., the afternoon prior and the afternoon of lightning near the station). This study builds upon the earlier work of Kumar et al., (2009) (hereafter, ‘Paper I’), reporting the perturbations in all components of the F-region drifts associated with lightning located to the west, north, and east of the station, and using short (few minutes) and long (several hours) time windows.

Section snippets

Instruments and data analysis

Lightning times and locations were obtained from WWLLN (Dowden et al., 2008). WWLLN uses the Time of Group Arrival (TOGA) of VLF sferics produced by CG lightning to locate the strokes in time and the downhill simplex method (Nelder and Mead, 1965) to locate the strokes in space. A 2 day comparative study in January 2003 between WWLLN and an Australian lightning network, Kattron, found that matched events (426 of the 698 determined by WWLLN) corresponded to large return stroke peak currents

Lightning occurrence in proximity to bundoora

The WWLLN lightning data summarised in Fig. 1, Fig. 2 were used for the subsequent SEA of F-region drifts. Fig. 1 shows the temporal variation of all CG strokes occurring within a 600 km radius of Bundoora. The CG stroke frequency peaked at ~14 LT (~8.2 strokes per hour) and was lowest at ~03 LT (~0.8 strokes per hour). During the austral summer months (December–February), the largest number of flashes occurred reaching up to 700 strokes per bin (1 h×10 day). These counts are saturated to red in

SEA analyses of F-region drifts using lightning as controls (t=0)

Paper I examined Bundoora vertical drift velocity using SEA Method 2 and revealed significant downward ionospheric motion during lightning. This paper extends the SEA analysis to all three components of the drift velocity. An initial SEA using all lightning controls irrespective of their direction from Bundoora revealed no significant response in the field-perpendicular drifts. However, when SEA was performed for lightning located in different quadrants relative to Bundoora ionosonde,

SEA analysis of F-region drifts using thunderstorm onsets as controls (t=0)

The identification of a thunderstorm was somewhat arbitrary for this part of the analysis. A thunderstorm onset time was defined by the start of a spatially localised cluster of more than 30 lightning strokes occurring within 600 km of Bundoora. By using daily spatial and temporal plots of CG lightning strokes (similar to Figs. 2b–e), 67 thunderstorm events were selected from the interval August 2003–August 2004. No other meteorological parameters were considered either due to their poor time

Summary and discussion

Digisonde drift velocity data from a southern mid-latitude station Bundoora (145.1°E, 37.7°S geographic, 49°S magnetic) and WWLLN lightning CG strokes within 1200 km relative to the station were used to study variability in F-region ionospheric drifts associated with tropospheric thunderstorm activity. The statistical analyses presented here are an extension of previous results [Paper 1] and they more fully quantify the response of the local ionosphere to spatiotemporal variations in

Acknowledgements

This work was supported by an internal research grant awarded by La Trobe University. We thank WWLLN for providing lightning data over the region. Financial support for VVK was provided by an Australian Government Endeavour Postgraduate Award (ESPA_DCD_365_2007) and La Trobe University. VVK thanks Callum Anderson for providing the neutral wind data for this paper.

References (43)

  • B.A. Tinsley et al.

    The role of the global electric circuit in solar and internal forcing of clouds and climate

    Advances in Space Research

    (2007)
  • J.A. Waldock

    Wind and temperature effects on F-region medium-scale gravity waves estimated using a multi-layer atmospheric model

    Journal of Atmospheric and Terrestrial Physics

    (1996)
  • J.A. Waldock et al.

    The effects of neutral winds on the propagation of medium-scale atmospheric gravity waves at mid-latitudes

    Journal of Atmospheric and Terrestrial Physics

    (1984)
  • J.A. Waldock et al.

    HF Doppler observations of medium-scale travelling ionospheric disturbances at mid-latitudes

    Journal of Atmospheric and Terrestrial Physics

    (1986)
  • J.A. Waldock et al.

    Source regions of medium-scale travelling ionospheric disturbances observed at mid-latitudes

    Journal of Atmospheric and Terrestrial Physics

    (1987)
  • B. Abel et al.

    Electron scattering loss in earth's inner magnetosphere 2. Sensitivity to model parameters

    Journal of Geophysical Research-Space Physics

    (1997)
  • R.J. Blakeslee et al.

    Electrical measurements over thunderstorms

    Journal of Geophysical Research

    (1989)
  • M. Blanc et al.

    The ionospheric disturbance dynamo

    Journal of Geophysical Research

    (1980)
  • G.B. Burns et al.

    Atmospheric circuit influences on ground-level pressure in the Antarctic and Arctic

    Journal of Geophysical Research

    (2008)
  • K.S. Carslaw et al.

    Cosmic rays, clouds, and climate

    Science

    (2002)
  • Cianos, N., Pierce, E.T., 1972. A ground-lightning environment for engineering usage. DTIC Research Report...
  • View full text