SEARCH

SEARCH BY CITATION

Keywords:

  • limb scan;
  • TEC;
  • tropical arcs

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. LORAAS Instrumentation and Algorithm
  5. 3. Comparison of Limb Scan and Radar
  6. 4. Summary
  7. Acknowledgments
  8. References

[1] Ionospheric nighttime electron density profiles were derived using 1356-Å ultraviolet limb scans from the Low-Resolution Airglow and Aurora Spectrograph (LORAAS) instrument on the Advanced Research and Global Observing Satellite (ARGOS). Successive limb scans along the Sun-synchronous orbit were inverted using a tomographic algorithm and were used to reconstruct the ionosphere in latitude and altitude in the 0230 local time frame. Total electron content (TEC) as a function of latitude was obtained by vertically integrating the tomographic densities. Similarly, dual-frequency radar altimeter on TOPEX/Poseidon provided estimates of vertically integrated TEC along the satellite orbit. Comparisons were made between coincident ARGOS and TOPEX satellite passes for several days in December 1999 and November 2000. The comparisons demonstrate the performance of the tomographic algorithm to reconstruct latitudinal variations in the ionosphere, namely the location and magnitude of the nighttime tropical arcs. The comparisons also demonstrate a technique for independently calibrating the limb imager sensitivity.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. LORAAS Instrumentation and Algorithm
  5. 3. Comparison of Limb Scan and Radar
  6. 4. Summary
  7. Acknowledgments
  8. References

[2] The Low Resolution Airglow/Auroral Spectrograph (LORAAS), flown on the Advanced Research Global Observation Satellite (ARGOS), observed the ionosphere and neutral thermosphere, beginning in mid-May 1999 and continuing observations until early April 2002. LORAAS was an ultraviolet limb scan imager which gathered vertical profiles viewing aft of the ARGOS, covering the 75–750 km altitude range with ∼5 km altitude resolution. Limb scan observations were inverted, using a tomographic retrieval algorithm, to determine the ionospheric electron density as a function of altitude versus latitude along the satellite orbit. The tomographic retrieval algorithms utilize ultraviolet spectral lines at 911-Å and 1356-Å, which are sensitive to emissions from radiative recombination of electrons and oxygen ions [Dymond and Thomas, 2001; Dymond et al., 1997]. Electron densities retrieved by inversion of single limb scans were successfully compared to ionosonde observations of peak density, NmF2, and peak height, hmF2, [Dymond et al., 2001a, 2001b]. While there was excellent agreement between the limb scan and ionosonde derived parameters, additional testing is required to validate the shape of the retrieved ionospheric profile and to examine the performance of the tomographic algorithm over a wider range of latitudes, local times, seasons, geomagnetic conditions and solar activity levels.

[3] This paper presents comparisons of limb scan ionospheric retrievals with coincident observations of the ionosphere from the dual-frequency radar altimeter on TOPEX. Limb scan electron density profiles are integrated vertically and compared directly with TOPEX observations of total electron content (TEC). This provides a validation of the tomographic retrieval algorithm along the track of the satellite, covering a wide range of latitudes. Additionally, when coupled with the earlier validation of NmF2 and hmF2, this study provides the missing piece necessary to validate the shape of the retrieved ionospheric profile.

2. LORAAS Instrumentation and Algorithm

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. LORAAS Instrumentation and Algorithm
  5. 3. Comparison of Limb Scan and Radar
  6. 4. Summary
  7. Acknowledgments
  8. References

[4] ARGOS is in a Sun-synchronous 832-km, 98° inclination orbit with a descending node at 0230 LT. The LORAAS instrument observes the naturally occurring UV emissions produced by radiative recombination on Earth's limb. It has a field-of-view of 2.4° × 0.08° and sweeps out a 2.4° × 17° field-of-regard during each 90-s scan in the aft direction of the satellite. The observed wavelength range is 800- to 1700-Å with 19-Å resolution.

[5] Approximately 90 spectra, with 1-s integration, are gathered per limb scan. During each scan, the tangent altitude ranges from 750 km to 75 km. When the motion of the spacecraft is accounted for, each limb scan is separated by approximately 5–6° in latitude at the equator.

[6] The 1356-Å spectral line is used during nighttime observations of the ionosphere, because it is brighter than the 911-Å emission. The higher brightness yields a more precise determination of the ionospheric parameters, as the signal-to-noise of the limb data is higher. In the F region, the intensity of the radiative recombination emission is proportional to the integral along the line-of-sight of the electron density squared [Dymond et al., 1997]. The instrument sensitivity degrades over time as the detector ages, requiring the instrument calibration factor to be estimated and used to scale the observed count rate. A value of 0.32 counts s−1 Rayleigh−1 for the sensitivity coefficient of the instrument was found during a stellar calibration involving data from 1999.

[7] The relation to density-squared, combined with long horizontal path lengths through the ionosphere causes the limb scan observations to be most sensitive to the ionosphere in the vicinity of the F region peak. Whether height or density variations, gradients in the ionosphere contribute to neighboring limb scans as either foreground or background emissions. To resolve these effects, a tomographic retrieval algorithm is used and is detailed in the work of Dymond and Thomas [2001]. The algorithm uses uncoupled Chapman layers to represent the electron density at the location of each limb scan. The individual Chapman profiles are coupled together by enforcing an ad hoc smoothness (or regularization) criterion. The inversion algorithm is based on discrete inverse theory [Menke, 1989] and uses the iterative Levenberg-Marquardt scheme [Press et al., 1992] to seek a maximum likelihood estimate of the ionospheric parameters based on the fit of the model to the data.

3. Comparison of Limb Scan and Radar

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. LORAAS Instrumentation and Algorithm
  5. 3. Comparison of Limb Scan and Radar
  6. 4. Summary
  7. Acknowledgments
  8. References

[8] The electron density profiles, resulting from the tomographic inversion of a group of LORAAS limb scans, are integrated vertically to produce TEC with 5–6° latitudinal spacing. This is compared to TOPEX-derived TEC averaged over a similar horizontal scale.

[9] The TOPEX spacecraft carries a dual-frequency radar altimeter operating at 13.6 GHz and 5.6 GHz in a 1336-km, 66° inclination orbit. The altimeter measures TEC from ocean reflections beneath the spacecraft at approximately 1-s intervals [Imel, 1994] for a spatial sampling of ∼6 km. For the purpose of computing differences with limb scan TEC, the TOPEX measurements are averaged over 5.6° in latitude, centered over the latitude of the limb scan.

[10] Two time periods were selected for comparison, one early in the LORAAS mission, December 1999, and another approximately one year later in November 2000. For a period of roughly two weeks during these months, the satellites crossed paths in the night sector, over open oceans, once per day. Coincident passes are defined here by requiring the limb scan F region tangent point and the TOPEX ionospheric penetration point to be separated by less than 30 min in time and 10° in longitude. For December 1999, the TOPEX instrument switched to a single-frequency mode a few days into the crossing period, resulting in only three coincident passes useful for comparisons, as compared to ten coincident passes for November 2000.

[11] Figure 1 shows a coincident pass that occurred on 1 December 1999. The two satellites passed each other headed in opposite directions, with the point of closest approach (PCA) occurring in the southern Pacific Ocean off the coast of South America. The TOPEX data show elevated TEC values in the Southern Hemisphere decreasing to a valley near 20°S, subsequently rising again before 0°S, and then falling to more typical midlatitude nighttime values. The limb scan TEC follows the trend in the TOPEX data nicely. For this comparison, TOPEX is on average ∼2 TECU (2 · 1016 electrons/m2) higher than the limb scan estimates, and the standard deviation of the differences is ∼4 TECU. TOPEX is anticipated to produce slightly higher TEC values (∼5 TECU) than other instruments, due to a previously observed instrument bias [Coker et al., 2001]. The difference between the TOPEX altitude and the upper boundary of the limb scans is expected to contribute less than a TECU difference in the observations due to weak topside ionosphere at 0230 LT. The good agreement between the data sets suggests that the tomographic limb scan retrieval has accurately captured the latitudinal variation of the ionosphere. Note specifically, the locations of the maximum NmF2 in the limb scan, at 35°S and 5°S, match the locations of maximum TEC as indicated by TOPEX.

image

Figure 1. Coincident pass on 1 December 1999: (a) limb scan (box) and TOPEX (line) crossing as a function of time and latitude; the point of closest approach is indicated by the solid box. (b) Crossing as a function of longitude and latitude. (c) Total electron content from limb scan (box), TOPEX raw data (dots), and TOPEX 1° average (line). (d) Limb scan electron density, latitude versus altitude.

Download figure to PowerPoint

[12] The most significant deviation from agreement in TEC occurs at ∼10–15°N, near the transition from low to middle latitudes. The errors here are significant in a relative sense, because the ionospheric TEC is low. In the nighttime midlatitude ionosphere where densities are low, the emissions from radiative recombination are weak, decreasing the signal-to-noise ratio of the UV limb data. Low signal-to-noise results in a relatively poor estimation of the hmF2 and scale height, H, as is indicated by the nonphysical variation of the tomographic inversion at 10–15°N.

[13] Figure 2 compares the TEC and electron density from coincident passes on 1, 2 and 3 December 1999. Each comparison shows that the limb scan TEC follows the trend in the TOPEX data, with little change in the average difference and standard deviation. Perhaps more significantly, the locations of maximum NmF2 in the limb scan retrievals match the locations of maximum TEC as indicated by TOPEX. This is most readily observed in Figure 2 by numbering the limbs scans from south to north, noting the scan associated with the peak TOPEX TEC, and then observing that this scan also has a matching peak NmF2. For example, scan numbers 1 and 6 are associated with peak TOPEX TEC and peak NmF2 for 1 December 1999 pass. Subsequent passes demonstrate that this match is achieved repeatedly. Assuming the peaks of the nighttime tropical arcs are at approximately the same location for TEC and NmF2, the location and latitudinal extent of the arcs are properly identified by the limb scan retrievals. Instances, where the limb scan TEC does not follow the TOPEX data as precisely as the limb scan NmF2, point to possible errors in the estimation of H or in the representation of the profile using a Chapman layer.

image

Figure 2. (a, c, and e) Comparison of limb scan and TOPEX total electron content and (b, d, and f) limb scan electron density for (a and b) 1 December 1999 0920 UT, (c and d) 2 December 1999 1910 UT, and (e and f) 3 December 1999 1200 UT. Limb scan data are indicated by boxes, TOPEX raw data by dots, and TOPEX 1° averages by solid lines. The point of closest approach between observations is indicated by solid box.

Download figure to PowerPoint

[14] Figure 3 compares the TEC and electron density from coincident passes on 17, 19 and 20 November 2000. These passes are representative of the ten coincident passes obtained during this period. The TEC comparisons on 17 and 19 November are similar to those from 1999. The limb scan TEC follows the trend in the TOPEX data, with a similar average difference and standard deviation. The TEC comparison on 20 November shows a significant deviation from agreement in the vicinity of a relatively narrow tropical arc structure at 27°S. One possible explanation for the disagreement is spatial and temporal differences in the ionosphere sampled by the two instruments. Another possibility is that the retrieval algorithm is overly smoothing the data. Evidence for this narrow arc structure can be seen in the raw limb scan intensity data. This suggests a limitation in the retrieval algorithm to reproduce the narrow structure, possibly due to the regularization constraint imposed on the retrieval.

image

Figure 3. (a, c, and e) Comparison of limb scan and TOPEX total electron content and (b, d, and f) limb scan electron density for (a and b) 17 November 2000 1210 UT, (c and d) 19 November 2000 1450 UT, and (e and f) 20 November 2000 0750 UT. Limb scan data are indicated by boxes, TOPEX raw data by dots, and TOPEX 1° averages by solid lines. Point of closest approach (PCA) between observations is indicated by the solid box.

Download figure to PowerPoint

[15] For each day, the locations of maximum NmF2 in the limb scan retrievals match the locations of maximum TEC as indicated by TOPEX. The location and latitudinal extent of the arcs are properly identified by the limb scan retrievals, even when the limb scan TEC shows disagreement with between TOPEX TEC.

[16] Figure 4 summarizes the comparisons between limb scan TEC and TOPEX TEC for December 1999 and November 2000. Although there were only a few coincident passes for December 1999, there is very good agreement between the TOPEX TEC and the limb scan TEC measurements. The average difference was 1.9 TECU and the standard deviation was 3.6 TECU. This indicates that the retrieval algorithm is working properly and that the sensitivity coefficient derived from the stellar calibration is sufficiently accurate for that time interval. For November 2000 the average difference was 4.8 TECU and the standard deviation was 6.5 TECU. The increased scatter in the results may be attributed to the decreased signal-to-noise ratio resulting from the decreasing sensitivity of the LORAAS instrument. In certain cases, reproduction of TEC gradients in the ionosphere was inaccurate due to either the imposed regularization constraint in the tomographic inversion for narrow structures or due to spatial and temporal differences in the ionosphere sampled by the two instruments (as discussed above). These cases also contribute to the increased scatter in the November 2000 comparisons.

image

Figure 4. Comparison of limb scan and TOPEX total electron content observations for coincident passes during (top) 1–3 December 1999 and (bottom) 17–27 November 2000. The point of closest approach is indicated for each pass by an open box.

Download figure to PowerPoint

[17] Although there is good agreement between the TOPEX data and the UV measurements for both time periods, as indicated by the two plots in Figure 4, it is noteworthy that at the points of closest approach (PCAs) the agreement is better. PCA occurs when the observations from the two satellites are closest in time and space, typically within 90 s and 3° in longitude of each other. This suggests that some of the disagreement between TOPEX and the UV measurements can be attributed to spatial or temporal differences in the sampled ionospheres.

[18] Since the scale height is directly proportional to the ratio of TEC and peak density, the relative error of the retrieved scale height can be estimated from the relative errors of retrieved TEC and peak density using

  • equation image

where σH, σTEC and σNmF2 are the one-sigma RMS errors for limb scan retrieved scale height, TEC and peak density, respectively. From this study and recalling that TOPEX has a bias ∼5 TECU, σTEC is estimated to be 5 TECU; and, from earlier studies [Dymond et al., 2001a, 2001b], σNmF2 is estimated to be 1 · 105 electrons/cm3. This implies for midlatitude nighttime, where typical TEC and peak density values are 10 TECU and 3 · 105 electrons/cm3, the scale height is poorly estimated with a relative error of ∼60%. However, for low latitudes, where typical TEC and peak density values are 30 TECU and 1 · 106 electrons/cm3 at night, the scale height is retrieved with a relative error of ∼20%.

[19] As mentioned earlier, a stellar calibration was used to determine the UV instrument sensitivity for the December 1999 observations. This sensitivity was validated by the good comparison between the TOPEX data and the UV measurements. For the November 2000 period the sensitivity coefficient was estimated by scaling the limb scan TEC to match the latitudinal structure observed during two coincident TOPEX passes. A sensitivity of 0.17 counts s−1 Rayleigh−1 minimized the TEC difference for the entire ensemble under investigation. Additionally, a comparison of the F2 peak height and peak density from ionosondes to the UV-derived parameters was performed. This comparison yielded a similar sensitivity coefficient of 0.2 counts s−1 Rayleigh−1. These findings demonstrate the utility of estimating the UV instrument sensitivity by comparison with coincident radar altimetry data.

4. Summary

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. LORAAS Instrumentation and Algorithm
  5. 3. Comparison of Limb Scan and Radar
  6. 4. Summary
  7. Acknowledgments
  8. References

[20] Two-dimensional reconstructions of nighttime electron density were obtained from UV limb scans on the ARGOS satellite. Comparisons of limb scan ionospheric retrievals with coincident TOPEX passes from December 1999 and November 2000 demonstrated that the latitudinal structure of the nighttime tropical arcs is accurately described by the limb scan retrievals. The location and width of the arcs are described particularly well, while the vertical TEC is described with reasonable accuracy (∼5 TECU). The scale height is described with reasonable accuracy (20% relative error) at low latitudes and less accurately at midlatitudes.

[21] The comparisons demonstrate a technique to validate the latitudinal distribution of nighttime electron densities produced from UV observations. Additionally, it was shown that this technique can be used to evaluate changes in instrument sensitivity.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. LORAAS Instrumentation and Algorithm
  5. 3. Comparison of Limb Scan and Radar
  6. 4. Summary
  7. Acknowledgments
  8. References

[22] The authors wish to thank P. A. M. Abusali and Tim Urban at Center for Space Research, University of Texas at Austin for providing the TOPEX data used in this analysis. LORAAS data and ARGOS support were provided by the Air Force Space Test Program. Special thanks goes to Sean McCoy for locating intersecting passes of the two satellites.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. LORAAS Instrumentation and Algorithm
  5. 3. Comparison of Limb Scan and Radar
  6. 4. Summary
  7. Acknowledgments
  8. References
  • Coker, C., G. Kronschnabl, D. S. Coco, G. S. Bust, and T. L. Gaussiran II (2001), Verification of ionospheric sensors, Radio Sci., 36(6), 15231529.
  • Dymond, K. F., and R. J. Thomas (2001), An algorithm for inferring the two-dimensional structure of the nighttime ionosphere from radiative recombination measurements, Radio Sci., 36(5), 12411254.
  • Dymond, K. F., S. E. Thonnard, R. P. McCoy, and R. J. Thomas (1997), An optical remote sensing technique for determining nighttime F region electron density, Radio Sci., 32(5), 19851996.
  • Dymond, K. F., S. A. Budzien, A. C. Nicholas, S. E. Thonnard, R. P. McCoy, and R. J. Thomas (2001a), Electron densities determined by inversion of ultraviolet limb profiles, J. Geophys. Res., 106(A12), 30,31530,322.
  • Dymond, K. F., S. A. Budzien, S. E. Thonnard, R. P. McCoy, and R. J. Thomas (2001b), Electron densities determined by the HIRAAS experiment and comparisons with ionosonde measurements, Geophys. Res. Lett., 28(5), 927930.
  • Imel, D. A. (1994), Evaluation of TOPEX/Poseidon dual-frequency ionosphere correction, J. Geophys. Res., 99(C12), 24,89524,906.
  • Menke, W. (1989), Geophysical Data Analysis: Discrete Inverse Theory, Int. Geophys. Ser., vol. 45, Academic, San Diego, Calif.
  • Press, W. H., B. P. Flannery, S. A. Teukolsky, and W. T. Vetterling (1992), Numerical Recipes: The Art of Scientific Computing, Cambridge Univ. Press, New York.