COSMIC observations of intra-seasonal variability in the low latitude ionosphere due to waves of lower atmospheric origin!

Transcription

1 COSMIC observations of intra-seasonal variability in the low latitude ionosphere due to waves of lower atmospheric origin! Nick Pedatella! COSMIC Program Office! University Corporation for Atmospheric Research!!! Eighth FORMOSAT-3/COSMIC Data Users Workshop! 3 September, 14!!

2 Outline! Introduction and Motivation! Data Processing! Results! Summary and Conclusions!

3 When viewed from a fixed local time perspective, the ionosphere exhibits a distinct longitudinal structure! Integrated Electron Content -km (-22LT) COSMIC Integrated Electron Content - km (-22 LT)! Integrated Electron Content 3-3km (-22LT) COSMIC Integrated Electron Content 3-3 km (-22 LT)! (Lin et al., 7)

4 The ionosphere longitude variability is attributed to upward propagating tides of tropospheric origin, which are generated by tropospheric convection! COSMIC observations clearly show the signature of the DE3 in F-region ionosphere electron densities COSMIC Integrated Electron Content - km(-22lt) (-22 LT)! Integrated Electron Content -km modulation of of E-region! ure 14. (a d) The latitude versus longitude distribution (left) diurnal and (right) semidiurnal vertical propagation! dynamo and and equatorial plitude of ISCCP radiative heating (Figures 14a and 14b) of TRMM latent heating (Figures 14c &! electric fields! d 14d) at 6.7growth! km in September. (e j) The interference of tidal components from E6 to W6 of amplitude Integrated Electron Content 3-3km COSMIC Integrated Electron Content 3-3 km(-22lt) (-22 LT)! t) diurnal and (right) semidiurnal at 9 km in September. Figures 14e and 14f are by GSWM 9 ponse, Figures 14g and 14h are by GSWM/TRMM response, and Figures 14i and 14j are by BER observation. 16 of 19 (Lin et al., 7) (Zhang et al., )! re 14. (a d) The latitude versus longitude distribution of (left) diurnal and (right) semidiurnal itude of ISCCP radiative heating (Figures 14a and 14b) and of TRMM latent heating (Figures 14c 14d) at 6.7 km in September. (e j) The interference of tidal components from E6 to W6 of diurnal and (right) semidiurnal at 9 km in September. Figures 14e and 14f are by GSWM 9 onse, Figures 14g and 14h are by GSWM/TRMM response, and Figures 14i and 14j are by ER observation.

5 The connection between the wavenumber-4 structure and lower atmospheric waves is supported by similar seasonal variability of the DE3 nonmigrating tide in temperature/ winds at E-region altitudes and the F-region electron densities.! TIMED/SABER DE3 in Temp. at km (6 day avg.) 4 K Latitude 4 1 DE3 diurnal eastward propagating! nonmigrating tide with! zonal wavenumber 3! COSMIC DE3 in NmF2 (6 day avg.) 4 % Mag. Latitude 4 1 NmF2 maximum F-region! electron density! Year The DE3 is generated by longitudinal differences in tropospheric convection, and appears as a wavenumber-4 pattern when viewed from a fixed local time perspective!

6 urnal variations of DE3 (a) nmf2 at all geographic Variability in the on shorter temporal scales may be expected due to (b) upwardine tropospheric B velocity convection, at the magnetic variability wave-wave interactions, and changes 3 km height for an arbitrarily selected longiin vertical propagation conditions due to zonal mean winds.! nd day (26 September) from the Ground to l of Atmosphere and Ionosphere for Aeronomy ation. WACCM Model DE3 at equator, -4 hpa! The interval of the variable exchange is 18 s in mulation, although each model component has a olution as already described. The three models grid distributions, thus coordinate conversion exchanging the variables, which is also carried pler module. Unification of spatial resolution components is desirable and is intended for use ersions. Day of September! GAIA Model Equatorial NmF2! (Liu, 13)! (b) Longitude [degrees]! (Jin et al., 11)! (c) the present version of GAIA, described in - Though model reveal significant have carried out simulations a 3 day consecutive run inday-to-day and sub-seasonal variability, directly! these variations difficult owing to limitations in the current observations.! hich observing is one of the months whenis the ionospheric gitudinal structure becomes dominant accord- The present study aims to connect COSMIC ionosphere observations of the sub-seasonal! servations [e.g., Kil et al., 8; Liu and ionosphere variability with an approximation of the tidal variability at E-region altitudes derived! 8; from Scherliess et al., 8; Wan et al., 8]. TIMED/SABER temperature observations.! condition, we used results from noncoupled sing eachwillatmospheric and ionospheric - This provide directgcm observational evidence that the intra-seasonal ionosphere variability is! driven by waves upwards from sumed constant solarpropagating UV and EUV fluxes (thethe lower atmosphere.! s set at 13.) and a quiet geomagnetic activity ric inputs such as the polar cap potential were

7 Outline! Introduction and Motivation! Data Processing! Results! Summary and Conclusions!

8 COSMIC Data Processing! - Within a running -day window, geomagnetic quiet (K p < 3) COSMIC! observations of NmF2 are binned in magnetic latitude, longitude, and local time.! - A fit is performed for the following basis function for each latitude and day:! F = F + Σ ΣF n,s cos(nωt - sλ - φ n,s ) + ΣF s cos(sλ - φ s )$ n! s! s! tides! n tidal harmonic! s zonal wavenumber! Ω Earth rotation (2π/24)! t - time! F n,s - amplitude$ Φ n,s - phase$ planetary waves! - Though the above expression decomposes the longitude-local time variability of the ionosphere! into tide and planetary wave components, it is important to note that these may not always be! directly connected to forcing from lower atmospheric waves due to the strong influence of solar! radiation on the ionosphere diurnal variation.! - The present study focuses only on the DE3 nonmigrating tide, which has been! shown to have similar variability in the ionosphere and lower atmosphere.!

9 TIMED/SABER Data Processing! SABER Equatorial Local Time Sampling Ascending Descending - Launched in 2, the SABER instrument on the TIMED! satellite provides temperature profiles from ~-1 km based! on limb radiance observations.! Local Time 1 - The TIMED/SABER local time sampling prevents complete! determination of the tidal amplitudes unless a 6-day window! is used for the analysis.! AS6 FORBES ET AL.: TROPOSPHERE-THERMOSPH Day of Year, 9 - Approximations are thus necessary to determine the tidal! amplitudes for shorter analysis windows.! - Here we leverage the fact that residuals from the mean! temperature will be opposite in sign for diurnal oscillations due! to the ~12h separation between ascending and descending! observations.! - The DE3 tidal amplitudes are estimated using a -day! window following the method of Oberheide et al. [2].! - Method essentially relies on determining the longitudinal! wavenumber-4 variability separately for the ascending and! descending observations, and approximating the DE3! amplitude based on their difference.! (Forbes et al., 6)!

10 The TIMED/SABER DE3 tidal estimates are in good agreement with results different model simulations! Latitude Latitude TIMED/SABER DE3 in Temp. at km ( day avg.) WACCM+DART Model Simulation DE3 at km Day of Year, 7 K K Latitude Latitude TIMED/SABER DE3 in Temp. at km ( day avg.) TIME GCM/MERRA Model Simulation DE3 at km Day of Year, 9 K K WACCM+DART: data assimilation version of the Whole! Atmosphere Community Climate Model! TIME-GCM/MERRA: NCAR TIME-GCM forced with MERRA! reanalysis!

11 Outline! Introduction and Motivation! Data Processing! Results! Summary and Conclusions!

12 Comparison between the and 6 day average TIMED/SABER tidal amplitudes at E-region altitudes reveals significant sub-seasonal variability.! TIMED/SABER DE3 in Temp. at km (6 day avg.) Latitude 4 4 K TIMED/SABER DE3 in Temp. at km ( day avg.) Latitude 4 K Year

13 COSMIC observations also reveal considerable sub-seasonal variability, demonstrating that the wavenumber-4 longitude structure exhibits considerable short-term variability.! COSMIC DE3 in NmF2 (6 day avg.) Mag. Latitude 4 4 % COSMIC DE3 in NmF2 ( day avg.) Mag. Latitude 4 4 % Year

14 Similar intra-seasonal variability occurs in the COSMIC and TIMED/SABER DE3 observations, illustrating that the ionosphere variability is driven from the lower atmosphere! TIMED/SABER DE3 in Temp. at km ( day avg.) Latitude 4 4 K COSMIC DE3 in NmF2 ( day avg.) Mag. Latitude 4 4 % Year

15 Similar intra-seasonal variability occurs in the COSMIC and TIMED/SABER DE3 observations, illustrating that the ionosphere variability is driven from the lower atmosphere! COSMIC DE3 Amplitude (%) COSMIC SABER Day of Year, SABER DE3 Amplitude (K) COSMIC averaged between -2 N and -2 S! TIMED/SABER averaged between ±1!

16 Outline! Introduction and Motivation! Data Processing! Results! Summary and Conclusions!

17 Summary and Conclusions! - COSMIC and TIMED/SABER observations are used to investigate the role of the! lower atmosphere on driving intra-seasonal variability in the ionosphere.! - Similar variability in the DE3 is observed in both the F-region ionosphere by! COSMIC, and at E-region altitudes by TIMED/SABER.! - It can therefore be concluded that at least a portion of the sub-seasonal! variability in the ionosphere is driven by waves propagating from the lower! atmosphere.! - Improved observation density from COSMIC-2 will enable determination of! ionosphere variability on shorter time scales.! - Studies of atmosphere-ionosphere vertical coupling would benefit significantly! from increasing the altitude of neutral atmosphere retrievals.!