SEARCH

SEARCH BY CITATION

Keywords:

  • conjugate effects;
  • ionosphere heating;
  • ionosphere modeling

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model
  5. 3. Results
  6. 4. Discussion
  7. Acknowledgments
  8. References

[1] Conjugate heating effects associated with the upcoming Arecibo heater facility are studied using the NRL ionosphere model SAMI2. A density-dependent, localized heating source is included in the electron temperature equation to model ionospheric radiowave heating. Heating effects are examined as a function of the heating timing and the peak density of the unmodified ionosphere (through the F10.7 index). The simulation results suggest that field-aligned duct formation occur during periods of relatively low electron densities (e.g., during the night). The enhancement of the electron temperature and electron density in the conjugate topside ionosphere (∼500 km) could reach respective values of ∼5% and 25%. Heating losses associated with inelastic electron-neutral (N2) collisions primarily inhibit conjugate effects.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model
  5. 3. Results
  6. 4. Discussion
  7. Acknowledgments
  8. References

[2] Modification of the ionosphere using high power radiowaves has been an important tool for understanding the complex physical processes associated with high-power wave interactions with plasmas. One major effect is that electron heating can lead to the formation of density cavities and enhancements along the geomagnetic field [Duncan et al., 1988; Bernhardt et al., 1989; Milikh et al., 2010a]. Existing radiowave heaters are at relatively high magnetic latitudes ≳ 40° (e.g., HAARP, EISCAT, SURA) [Milikh et al., 2010a; Rietveld et al., 2003; Frolov et al., 2008a, 2008b]; modeling studies [Perrine el al., 2006] and observations [Carpenter et al., 2002] suggest that artificial density enhancements may extend into the plasmasphere.

[3] A mid-latitude (magnetic latitude ≃ 28°) radiowave heater existed at Arecibo until 1998 when it was destroyed by a hurricane. During its operation a number of scientific investigations were carried out [Duncan et al., 1988; Bernhardt et al., 1988, 1989; Newman et al., 1988; González et al., 2005]. A new heating facility is now being installed at Arecibo. The apex height of a field line is ≃2200 km and the conjugate footpoint is just off the coast of Argentina. Thus, the new Arecibo heater offers the possibility of producing ionospheric heating and density effects observable in the conjugate ionosphere.

[4] Previous modeling studies of the Arecibo heater were restricted to 1D models of the heated ionosphere and limited to altitudes below 500 km [Newman et al., 1988]. In this paper we report results of a modeling study using the 2D NRL ionosphere code SAMI2 to investigate the impact of ionospheric heating at Arecibo on the conjugate ionosphere. Specifically, we investigate the development and formation of enhanced electron temperature and along the heated field line, as well as the associated changes in electron density. The simulation results suggest that field-aligned duct formation occurs during periods of relatively low electron densities, e.g., during the night. The enhancement of the electron temperature and electron density in the conjugate topside ionosphere (∼500 km) could reach respective values of ∼5% and 25%. Heating losses associated with inelastic electron-neutral (N2) collisions primarily inhibit conjugate effects.

2. Model

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model
  5. 3. Results
  6. 4. Discussion
  7. Acknowledgments
  8. References

[5] The ionosphere model SAMI2 developed at the Naval Research Laboratory [Huba et al., 2000a] is used to conduct ionospheric heating simulations with a realistic ionosphere. SAMI2 models 7 ion species in the ionosphere: H+, He+, O+, O2+, N+, N2+, and NO+. The equations of continuity and momentum are solved for each ion species, with the temperature equation solved for the electrons and the ion species H+, He+, and O+. The code has been modified to include the effects of radiowave heating on the electrons [Perrine et al., 2006; Milikh et al., 2010b].

[6] The electron temperature equation used in this study is

  • display math

The second term on the left side of the equation is a diffusion term, where κe as the parallel electron thermal conductivity, k the Boltzmann constant, and bs the component of the magnetic field in the direction of the field line, normalized to its equatorial value on Earth's surface. Huba et al. [2000a] define the conductance κe, and heating/cooling rates Qen, Qei, and Qphe in detail. QRF is the heating term and is represented by a Gaussian shaped source:

  • display math

where the heating altitude z0 is the center of the region where the electron density is in the range 3.1 × 105cm−3ne < 3.3 × 105cm−3 (this corresponds to a heater frequency at 5.1 MHz). The other parameters used are (dTe/dt)0 = 1000 K/s, Δz = 10 km, θ0 = 18.3°, and Δθ = 0.05°. The heating rate is based on comparison of SAMI2 simulation results with Demeter observations over the SURA heater [Milikh et al., 2010b]. This is justified because the upcoming Arecibo heater is expected to have an effective radiative power similar to the SURA heater.

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model
  5. 3. Results
  6. 4. Discussion
  7. Acknowledgments
  8. References

[7] The first set of simulation results are for F10.7 = 100 and F10.7 = 120, Ap = 4 and day-of-year = 130. The grid is (nznf) = (801,202) where nz is the number of grid cells along the geomagnetic field and nf is the number of field lines. The heating function is applied for 1 hr at different local times.

[8] Figure 1 shows contour plots of the electron temperature as a function of altitude and latitude near the end of the one hour heating times: 1–2 LT (Figure 1a) and 4–5 LT (Figure 1b). Figure 1 (top) is for F10.7 = 100 and is the ‘low’ density case: NmF2 = 5.97 × 105 cm−3 and 4.28 × 105 cm−3 just before the heater is turned on at 1 LT and 4LT. Figure 1 (bottom) is for F10.7 = 120 and is the ‘high’ density case: NmF2 = 9.40 × 105 cm−3 and 6.83 × 105 cm−3 just before the heater is turned on at 1 LT and 4LT. The electron density increases with F10.7 because of an increase in solar EUV and and expansion of the thermosphere. These results show that electron heating in the conjugate region is more effective for lower density F layers. There is heating in the conjugate region for all cases except for F10.7 = 120 when the heater is turned on at 1 LT. In this case the peak electron density is ∼106 cm−3. This is consistent with observational data [Frolov et al., 2008a, 2008b]. For these runs the initial heating altitude is ∼280 km and the heating altitude increases by ∼25 km when the heater is turned off.

image

Figure 1. The electron temperature (K) contour plots as a function of altitude and latitude of (top) the low density case (F10.7 = 100) and (bottom) the high density case (F10.7 = 120) for various heating times: (a) 1–2 LT and (b) 4–5 LT at Arecibo. The snapshots are taken ∼1.5 minutes before the end of the heating duration.

Download figure to PowerPoint

[9] In Figure 2a we show the electron density distribution (logarithmic scale) for the conditions corresponding to Figure 1a. For the low density case (F10.7 = 100) we note that a low density duct forms in the heated region above Arecibo, but that the density is enhanced along the flux tube in the conjugate hemisphere. This is caused by the heated plasma expanding along the flux tube and acting as a ‘snowplow’ to compress the density ahead of it. For the high density case (F10.7 = 120) the effect on the electron density is confined primarily to the heated region above Arecibo, consistent with Figure 1a. This is consistent with previous theoretical studies [Newman et al., 1988].

image

Figure 2. (a) The electron density (cm−3) contour plots in logarithmic scale corresponding to Figure 1b. (b) The line plots of the electron density and temperature at the heater location (17.3deg, 396 km) Arecibo local time. The heater is turned on between 1 to 2 LT (top) F10.7 = 100 and (bottom) F10.7 = 120. (c) Same as Figure 2b but at the conjugate-side location of (−38.5 deg, 393 km).

Download figure to PowerPoint

[10] In Figures 2b and 2c we plot the temporal variations of the electron density (red line) and electron temperature (black line) at two points in space, one over Arecibo and the other over the conjugate site at an altitude of ∼400 km. In Figure 2b the electron temperature increase rapidly at ∼400 km followed the turn-on of the heater at 1 LT for the low density case (F10.7 = 100; Figure 2b, top). The temperature rises to ∼3000 K and remains roughly constant during the duration of heating (1 hr). After the heater is turned off, the electrons cool to their ambient value on a time scale ∼30 min. During the heating period the electron density decreases by roughly a factor of 2 from its ambient value. In the conjugate region (Figure 2c, top) there is an effect on the ionosphere ∼30 min after the heater is turned on. There is an increase in the electron density up to ∼30% but very little impact on the electron temperature. Interestingly, the effect on the conjugate electron density lasts for several hours.

[11] In Figure 2b the electron temperature increases slowly at ∼400 km followed the turn-on of the heater at 1 LT for the high density case (F10.7 = 120; Figure 2b, bottom). The temperature only rises to ∼2000 K when the heater is turned off. After the heater is turned off, the electrons rapidly cool to their ambient value on a time scale ∼10 min. During the heating period the electron density decreases by roughly 30% from its ambient value. In the conjugate region (Figure 2c, bottom) there is no effect on the ionosphere from the heater.

[12] The spatial and temporal response of the plasma to heating along the heated magnetic field line is shown in Figure 3. Contour plots of (a) the electron temperature, (b) the electron density, (c) H+ density and (d) O+ density along the heated field line as a function of time and latitude. In Figure 3a the strong heating over Arecibo (∼18°) is evident at t = 1 hr. Subsequently, the heat pulse propagates along the flux tube to the conjugate ionosphere in roughly 20 min. The electron temperature remains somewhat elevated at high altitude even after the heater has been turned off for 60 min.

image

Figure 3. Contour plots of (a) the electron temperature, (b) the electron density, (c) H+ density and (d) O+ density along the heated field line as a function of time and latitude.

Download figure to PowerPoint

[13] Associated with the temperature pulse is an electron density pulse (Figure 3b). This is primarily composed of H+ because the flux tube extends well above the H+/O+ transition altitude (∼800 km). Both ion species have a reduction in density at low altitudes (e.g., below ∼700 km) at the beginning of the heating phase. As the heating proceeds, both ions are pushed up the field line. However, the altitude dependence of O+ and H+ are different. The O+ density peaks at ∼350 km; this high density plasma is transported upwards and produces the increase in O+ density at higher altitudes (up to ∼1500 km). On the other hand, the H+ density has a weak peak at a higher altitude (∼700 km). Thus, as the H+ expands upwards from below 700 km its density is reduced.

[14] Interestingly, even after the heater is turned off at t = 2 hr, oscillations in H+ are evident at high altitudes for several hours (Figure 3c). The period of these oscillations is ≃30 min and is due to standing ion sound waves between the conjugate hemispheres [Huba et al., 2000b]. The ion sound wave velocity is ≃5 km/s over a distance of ≃8000 km.

[15] We performed a simulation study for F10.7 = 100 and DOY = 80, 173, 265, and 356 to study the seasonal dependence of conjugate heating. The results are shown in Figure 4 and a strong seasonal dependence is found. We find there is no heating for DOY = 80 and 356 because the peak electron density is lower than that required for heating. On the other hand, for DOY = 173 and 265 there is a strong conjugate effect because the peak electron density is sufficiently high for heating but not too high to inhibit the conjugate effect.

image

Figure 4. The electron temperature (K) contour plots as a function of altitude and latitude for different days of the year: 80, 173, 265, and 356. In this study we used F10.7 = 100 and the heater was turned on at 1 LT. The value of NmF2 is shown in the upper right of each panel.

Download figure to PowerPoint

[16] A number of additional simulations were performed to isolate the physical processes responsible for conjugate heating. First, we ran simulations in which the solar flux was varied but not the neutral background; specifically, we considered F10.7 = 100 and 120 in EUVAC but F10.7 = 120 in the neutral model. We found that conjugate heating was stronger for the case F10.7 = 100 because of the lower value of NmF2. Second, we ran simulations in which the neutral atmosphere was varied but not the solar flux; specifically, we considered F10.7 = 100 and 120 in the neutral model but F10.7 = 120 in EUVAC. We found that there was significantly more conjugate heating in the case of F10.7 = 100 than F10.7 = 120. Thus, a contraction in the thermosphere leads to a stronger conjugate effect because of the lower electron density and lower neutral density (i.e., less collisional cooling.) We found that the dominant heat loss mechanism inhibiting electron heating effects in the conjugate region is associated with the term Qen; specifically, inelastic electron collisions with the neutral specie N2. This is consistent with the physics of ionospheric cooling [Schunk and Nagy, 2000].

[17] Lastly, we have performed additional simulations to address the issue of heating power. We considered different heating rates (a proxy for different heating powers) for the case F10.7 = 100 and DOY = 130. We find that the dependence of the electron temperature on the heater power is not linear. The maximum temperatures achieved are 3000 K, 3500 K, and 4100 K for the heating rates (dTe/dt)0 = 500 K/s, 1000 K/s, and 2000 K/s, respectively. The effects in the conjugate topside ionosphere show a similar relative change.

4. Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model
  5. 3. Results
  6. 4. Discussion
  7. Acknowledgments
  8. References

[18] We have presented the first simulation study of conjugate heating effects for expected parameters from the upcoming Arecibo heater using the NRL ionosphere model SAMI2. The simulation results suggest that field-aligned duct formation occur during periods of relatively low electron densities, e.g., at night for low F10.7 values (≲100). The enhancement of the electron temperature and electron density in the conjugate topside ionosphere (∼500 km) could reach respective values of black ∼5% and 25%. Heating losses associated with inelastic electron-neutral (N2) collisions inhibit conjugate effects. The impact of the Arecibo heater on the conjugate ionosphere (enhanced density and temperature) appears restricted to the topside conjugate F layer. The modeling study did not indicate heating effects affected the conjugate ionosphere below the F peak.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model
  5. 3. Results
  6. 4. Discussion
  7. Acknowledgments
  8. References

[19] We thank the referees for helpful suggestions to improve the paper. This research has been supported by 6.1 Base Funds at the Naval Research Laboratory.

[20] The Editor thanks two anonymous reviewers for their assistance in evaluating this paper.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Model
  5. 3. Results
  6. 4. Discussion
  7. Acknowledgments
  8. References