HAL Id: hal-00317904https://hal.archives-ouvertes.fr/hal-00317904

Submitted on 14 Oct 2005

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Possible ionospheric preconditioning by shear flowleading to equatorial spread F

D. L. Hysell, E. Kudeki, J. L. Chau

To cite this version:D. L. Hysell, E. Kudeki, J. L. Chau. Possible ionospheric preconditioning by shear flow leading toequatorial spread F. Annales Geophysicae, European Geosciences Union, 2005, 23 (7), pp.2647-2655.�hal-00317904�

Annales Geophysicae, 23, 2647–2655, 2005SRef-ID: 1432-0576/ag/2005-23-2647© European Geosciences Union 2005

AnnalesGeophysicae

Possible ionospheric preconditioning by shear flow leading toequatorial spreadF

D. L. Hysell1, E. Kudeki2, and J. L. Chau3

1Department of Earth and Atmospheric Sciences, Cornell University, Ithaca, NY 14853, USA2Department of Electrical and Computer Engineering, University of Illinois, Urbana, Il 61801, USA3Radio Observatorio de Jicamarca, Instituto Geofısico del Peru, Lima, Peru

Received: 18 April 2005 – Revised: 28 July 2005 – Accepted: 3 August 2005 – Published: 14 October 2005

Abstract. Vertical shear in the zonal plasma drift speed isapparent in incoherent and coherent scatter radar observa-tions of the bottomsideF region ionosphere made at Jica-marca from about 1600–2200 LT. The relative importance ofthe factors controlling the shear, which include competitionbetween theE andF region dynamos as well as vertical cur-rents driven in theE and F regions at the dip equator, ispresently unknown. Bottom-type scattering layers arise instrata where the neutral and plasma drifts differ widely, andperiodic structuring of irregularities within the layers is tell-tale of intermediate-scale waves in the bottomside. Theseprecursor waves appear to be able to seed ionospheric inter-change instabilities and initiate full-blown equatorial spreadF . The seed or precursor waves may be generated by a col-lisional shear instability. However, assessing the viability ofshear instability requires measurements of the same parame-ters needed to understand shear flow quantitatively – thermo-spheric neutral wind and off-equatorial conductivity profiles.

Keywords. Ionosphere (Equatorial ionosphere; ionosphericirregularities) – Space plasma physics (Waves and instabili-ties)

1 Introduction

A new perspective on the stability of the postsunset equato-rial F region ionosphere is emerging from experiments con-ducted at the Jicamarca Radio Observatory near Lima, Peru.The Jicamarca incoherent scatter radar has specialized ca-pabilities for observing both the ionospheric conditions thatlead to equatorial spreadF and for monitoring the plasmairregularities appearing subsequently. Being able to observeincoherent scatter looking perpendicular to the geomagneticfield, the 50 MHz radar can exploit the long correlation timeof the signal to make very accurate measurements of the line-

Correspondence to:(D. L. Hysell(daveh@geology.cornell.edu)

of-sight plasma drift speed (Woodman, 1970). Using a dualbeam configuration, unambiguous profiles of the plasma vec-tor drift velocity in the plane perpendicular toB can also beobtained. The modularity of the Jicamarca phased array an-tenna permits the application of aperture synthesis imaging,which yields unambiguous two- or three-dimensional imagesof the spatial configuration of plasma irregularities producingfield-aligned backscatter (Kudeki and Surucu, 1991; Wood-man, 1997).

A striking characteristic of the bottomsideF region iono-sphere extending from late afternoon to a few hours beforemidnight is shear flow, as first noted byKudeki et al.(1981)and Tsunoda et al.(1981). Plasma drifts westward in thebottomsideF region while drifting eastward near and abovethe F peak, with the shear being strongest just after sun-set. Shear flow was studied theoretically byHaerendel et al.(1992) andHaerendel and Eccles(1992) and is a componentof the postsunset vortex described recently byKudeki andBhattacharyya(1999) and Eccles et al.(1999) using radarand satellite observations, respectively. Constraints imposedby low latitude current closure in the vicinity of the solar ter-minator ultimately control shear flow, but the phenomenon isnot well accounted for quantitatively.

In this manuscript, we review recent findings suggestingthat shear flow may destabilize the equatorialF region, pos-sibly preconditioning it for interchange instabilities responsi-ble for spreadF (as postulated byMaruyama et al., 2002). Ifso, then monitoring plasma flows and anticipating shear in-stability may provide a means of forecasting some instancesof spreadF . Since the thermospheric neutral wind speedboth drives the plasma flow and influences shear instabil-ity, monitoring neutral circulation will be necessary. TheC/NOFS satellite therefore has a crucial role to play in thisanalysis. However, additional ground-based instrumentationwill be required for measuring off-equatorial conductivity,which both influences shear flow and inhibits instability.

2648 D. L. Hysell et al.: Shear flow leading to ESF

2 Examples of shear flow

Figure 1 shows examples of ionospheric structure and cir-culation from paired events observed in November or De-cember 2001, 2002, and 2003. The data were acquired us-ing refined experimental techniques introduced recently byKudeki et al.(1999). The top panel shown for each eventrepresents relative backscatter power in grayscale logarith-mic format. Although the backscatter power has been scaledby the square of the range, it is not calibrated, is distorted tosome degree by magnetoionic effects, and is meant to servehere only as a rough estimate of the relative electron densityin undisturbed regions of the ionosphere. Regions of intensebackscatter signify coherent scatter from plasma irregulari-ties and are plotted using different amplitude scaling.

The middle panel of each event in Fig. 1 shows vertical ve-locity derived from the average Doppler shift of Jicamarca’sdual beams, which are directed approximately east and westof magnetic zenith for this experiment. The bottom panelshows zonal velocity derived from the difference of the dualbeam Doppler shifts. Both coherent and incoherent scatterdata were utilized in generating the panels. All the data werederived from geomagnetically quiet periods.

The three examples on the left side of Fig. 1 representevents when ionospheric irregularities were confined to nar-row layers in the bottomside. These “bottom-type” layers arenearly ubiquitous immediately after sunset during equinoxand December solstice in the Peruvian sector but do not rep-resent significant ionospheric disruptions and are not asso-ciated with strong radio scintillations. In contrast, the threeexamples on the right portray occurrences of “full-blown”equatorial spreadF . Forecasting topside spreadF eventslike these on the basis of available data is a goal of the spaceweather community. Forecast strategies have often involvedapplying a threshold condition to the intensity of the prere-versal enhancement of the vertical drift. Ascent drives theF

layer to altitudes with reduced ion-neutral collision frequen-cies more susceptible to gravitational interchange instability.Cursory examination of the data here alone suggest that sucha strategy should have merit but would be fallible and couldnot be applied before about 23:30 UT (18:30 LT) or onlyabout an hour before the onset of instability. Discriminat-ing between plasma ascent leading to instability and ascentcaused by instability would also be problematic.

Shear flow is evident in all of the events in Fig. 1 andis most obvious after about 00:30 UT (19:30 LT) whenbackscatter from bottom-type layers points to rapid westwardmotion immediately below regions of rapid eastward motion.It might seem fortuitous that the bottom-type layers exist toprovide a radar target, since the valley region plasma is in-sufficiently dense to be monitored using incoherent scatter.However,Kudeki and Bhattacharyya(1999) pointed out that,assuming the thermospheric winds are eastward throughouttheF region, the bottom-type layers occupy strata where theplasma drifts are strongly retrograde. Given that significanthorizontal density gradients are also present in these strataaround sunset due to photochemical and dynamical effects,

the strata would be prone to horizontal wind-driven inter-change instabilities. They hypothesized that such instabil-ities are responsible for the bottom-type scattering layers,which are byproducts of retrograde drifts in the bottomsideand therefore not merely fortuitous.

The Kudeki and Bhattacharyya(1999) theory explainswhy the bottom-type layers exhibit no vertical development,as exemplified by the RTI map in Fig. 2. Since the primaryplasma waves grow by advection, they always remain con-fined to the strata where the drifts are retrograde. In contrast,interchange instabilities driven by gravity or a zonal electricfield must grow by convection, as did the two radar plumespassing over Jicamarca later in the same event. Another uni-versal characteristic of the layers is that they vanish at thefirst appearance of topside spreadF irregularities and sel-dom reappear. This feature is at least partially explained bytheory, since retrograde plasma drifts arise ultimately fromthe demands of the current system near the solar terminatorand should cease by some local time. Topside plumes likelydisrupt the local circulation and break up any bottom-typelayers nearby.

Further evidence that wind-driven instability is at work inthe bottomside was provided byHysell et al.(2004), whoused aperture synthesis imaging techniques to analyze thestructure of the primary waves in bottom-type scattering lay-ers. They found that the primary waves were kilometricin scale and generally horizontally oriented and elongatedrather than vertically. The primary waves they observed ap-peared to grow by advection rather than convection, consis-tent with the wind driven interchange instability theory.

Moreover, Hysell et al. (2004) found that the primaryplasma waves in bottom-type layers sometimes cluster to-gether in regular, wavelike patches separated horizontally byabout 30 km. Clustering took place only in those events whentopside spreadF occurred later. They associated the cluster-ing with a horizontal wave in the background plasma densitythat would present alternately stable and unstable horizontalgradients for wind-driven interchange instability. Figure 3shows a radar image derived from the bottom-type layer inFig. 2 illustrating such clustering.

Finally, Hysell et al.(2004) concluded that the clusteredor patchy bottom-type layers were telltale of decakilometricwaves in the bottomsideF region that could serve to precon-dition or seed the ionosphere for gravitational interchange in-stabilities (Rayleigh Taylor) to follow. This hypothesis wassupported by the fact that the intermediate-scale plasma ir-regularities that formed later in the spreadF events sharedthe same decakilometric wavelength. This suggests a fore-cast strategy for spreadF based on detecting clustering inbottom-type layers immediately after sunset. The strategy isbeing evaluated for its merits but in no event would be appli-cable before about 19:30 LT when the layers first appear.

The∼30 km wavelength of the undulations in the bottom-sideF region inferred from the patchy bottom-type scatter-ers does not correspond to a characteristic scale length of theionospheric interchange instability. This is why it is seen aspreconditioning rather than just the early stages of instability.

D. L. Hysell et al.: Shear flow leading to ESF 2649

Fig. 1. Six examples of ionospheric conditions around twilight. Each example shows the backscatter power (top panel), vertical plasmadrift (middle panel), and zonal plasma drift (lower panel), plotted against the scales shown. Symbols plotted in the middle panel representthe vertical drift velocity at 450 km altitude for reference. The left and right columns portray events when topside spreadF did not and didoccur, respectively. Note that UT=LT+5 h.

Preconditioning or seeding of the postsunset ionosphere isoften attributed to gravity waves (Kelley et al., 1981). Thishypothesis is difficult to test experimentally, however, sinceonly the effects of gravity waves and not the gravity wavesthemselves can be detected in the thermosphere using ISRs.

Recently,Vadas and Fritts(2004) examined the issue theo-retically, showing that a spectrum of gravity waves launchedby mesoscale convection cells could survive wind shears andviscous and conductive damping and penetrate into the lowerthermosphere. However, the 30 km wavelength is at the

2650 D. L. Hysell et al.: Shear flow leading to ESF

Fig. 2. Range time intensity (RTI) image of a topside spreadF event. Only a bottom-type layer was present prior to about 21:00 LT.

Fig. 3. Radar image of plasma irregularities within the bottom-typelayer seen in the previous figure. The irregularities are shown atthe moment they first drifted westward into the radar illuminatedvolume.

extreme short wavelength edge of allowed wavelengths, andnothing in their theory would tend to favor it.

Meanwhile, the 30 km wavelength is roughly comparableto the vertical scale length of the plasma shear flow. Recently,Hysell and Kudeki(2004) examined the possibility that shearflow may cause the bottomsideF region to become unstableand produce the precursor waves inferred from radar imag-

ing. If so, then shear flow itself could precondition the iono-sphere for full-blown spreadF . A quantitative understand-ing of the shear instability might therefore lead to a forecaststrategy for spreadF with the potential for advanced warn-ing, given that shear develops early in the evening. Below,we review the causes of shear flow in the equatorial iono-sphere, formulate the shear instability problem, and composean approximate but closed-form expression for the instabilitygrowth rate.

3 Origins of shear flow

That shear flow should exist in the bottomside equatorialF region was evident in the numerical modeling results ofHeelis et al.(1974), argued on theoretical grounds byFejer(1981), and observed byKudeki et al.(1981) andTsunodaet al.(1981). This shear can be understood, at least partly, interms of dynamo theory.

A model of the ionospheric dynamo was described byRichmond (1973) who assumed a dipole, equipotentialmodel for the geomagnetic field and considered an individualflux tube. Quasineutrality in a plasma imposes the constraintthat the total current density be solenoidal (∇·J=0), where

J = σP (E⊥ + u × B)

+ σH b × (E⊥ + u × B) + σ◦E‖ (1)

is the current density andσP , σH , andσ◦ are the Pedersen,Hall, and direct conductivities, respectively. Also,E is theelectric field,u is the neutral wind velocity, andb is a unitvector parallel to the geomagnetic fieldB.

Application of the quasineutrality condition to the currentdensity within a magnetic flux tube, neglect of horizontal

D. L. Hysell et al.: Shear flow leading to ESF 2651

variations in the fields and conductivities, and integration ofthe result over the coordinate parallel toB while taking themagnetic field lines to be equipotentials leads directly to thefollowing equation governing the vertical ionospheric elec-tric field:

Ep =

(Eφhφ

∫σH hqdq (2)

−

∫ (σP uφ + σH up

)Bhφhqdq

+dI

dφ

)/

(hp

∫σP

hφhq

hp

dq

)written here using the orthogonal coordinate system(p, q, φ), wherep is normal to the dipole magnetic field andin the plane containing the dipole moment,q is parallel tothe field line, andφ is the magnetic longitude. Theh factorsare the coordinate scale factors (see for exampleHysell et al.(2004). Also, dI/dφ represents the differential current den-sity passing through the flux tube in thep direction. Thecurrent pattern involved in producingEp is informativelysketched in Fig. 9 ofHaerendel and Eccles(1992).

Equation (2) succinctly summarized the factors that influ-ence the vertical electric field in the equatorial ionosphere.These factors have been discussed individually in theoreti-cal and experimental contexts by a number of investigatorsincluding Anandarao et al.(1978); Fejer (1981); Stenning(1981); Takeda and Maeda(1983); Farley et al.(1986) andby Haerendel et al.(1992), Haerendel and Eccles(1992), andrecently byEccles(1998) andEccles et al.(1999) in morecomplete theoretical treatments. They include 1) zonal elec-tric fields on flux tubes with significant Hall conductivity, asare responsible for driving the equatorial electrojet, 2) zonalwinds on flux tubes with significant Pedersen conductivity,as drive theE andF region dynamos, 3) vertical winds, alargely unknown quantity, and 4) vertical boundary currentsforced from above or below the flux tube in question. Inthe bottomside and valley regions around twilight, this lastfactor could result from the closure of the equatorial elec-trojet, which must turn partly vertical at the boundary of theevening terminator. It could also result from theF regiondynamo operating in flux tubes near theF peak. The finiteefficiency of the dynamo implies the existence of an upwardvertical current there. The demands imposed by this currentcan only be supported at lower altitudes, where the conduc-tivity is smaller, by a potentially large vertical electric field(Haerendel et al., 1992; Haerendel and Eccles, 1992).

Which of these effects dominates in the postsunset bot-tomside?Haerendel et al.(1992) attribute most of the shearthere to the boundary current effect. They also considered theeffect of zonal electric fields, which generate vertical polar-ization electric fields on flux tubes with significant integratedHall conductivities in accordance with Eq. (2). This mecha-nism was deemed to be effective mainly in the trough regionof the F 1 layer and below. WhileHaerendel et al.(1992)did not discount the importance of theE region dynamo inproducing shears and westwardF region drifts altogether,

they restricted the domain of importance to the valley region.However, this appraisal was based onE region neutral windestimates of only 25 m/s which we now recognize to be adrastic underestimate, by a factor of five or more, in lightof recent chemical release experiments (Larsen and Odom,1997).

To summarize, although the various factors that can in-fluence shear flow in the ionosphere are individually under-stood, their relative importance is not. Resolution of theproblem evidently requires specification of both the neutralwind profile in the mesosphere, lower thermosphere, andthermosphere as well as the off-equatorial conductivity pro-files.

4 Shear-induced instability

Hysell and Kudeki(2004) recently addressed the stability oftheF region ionosphere in a sheared flow with an eigenvalueanalysis.Keskinen et al.(1988) had already shown that elec-trostatic Kelvin Helmholtz instabilities are stabilized by ion-neutral collisions in the auroral ionosphere, but that analysisdid not explore the collisional branch of the instability in thelimit of steep plasma density gradients as are found in thebottomside equatorialF region. FollowingKeskinen et al.(1988), Hysell and Kudeki(2004) neglected Hall and par-allel currents but allowed for polarization currents in theirtwo-dimensional analysis:

J =ne

�iB

[νin (E + u × B) +

(∂

∂t+ v · ∇

)E

](3)

The quasineutrality condition, together with the ion continu-ity equation, formed the basis of their model. Linearizationof the model was performed according to

n(x, z) = n◦(z) + n1(z)ei(kxx−ωt) (4)

φ(x, z) = φ◦(z) + φ1(z)ei(kxx−ωt) (5)

where terms with subscripts 0 and 1 represent zero- and first-order quantities, respectively, and where the Cartesian coor-dinatesx, y, andz represented the eastward, northward, andupward directions at the magnetic equator. The linearizedmodel equations took the form:

(ω − kxv◦)n1 = −kx

B

dn◦

dzφ1 (6)

(ω + iνin − kxv◦)

[d

dz

(n◦

dφ1

dz

)− n◦k

2xφ1

]= −in◦

dνin

dz

dφ1

dz− kx

d

dz

(n◦

dv◦

dz

)φ1

+id

dz(Bνin(u − v◦)n1) (7)

whereω is the complex frequency and serves as the eigen-value, kx is the assumed horizontal wavenumber,v◦ is thebackground horizontal plasma flow speed arising from gra-dients in the zero-order electrostatic potential, andνin is theion-neutral collision frequency.

2652 D. L. Hysell et al.: Shear flow leading to ESF

0.2 0.4 0.6 0.8kL

0.05

0.1

0.15

0.2

0.25Γ

1.0

1.25

0.75

Fig. 4. Normalized growth rateγ (scaled toV◦/L) versus normal-ized wavenumberkL givenu=2 andν◦=2 for three different valuesof ν1. The points represent (9) evaluated at the value ofkx indicatedand withkz=1.25 (see text).

Combined, the model equations constitute an eigenvalueproblem describing the perturbed potentialφ1(z) in an inho-mogeneous, magnetized, two-dimensional plasma undergo-ing shear flow. This problem was solved computationally forequilibrium velocity, collision frequency, and density profilesspecified by:

v◦(z) = V◦Tanh(z/L)

νin(z) = ν◦ − ν1Tanh(z/L)

n◦(z) = N◦uν◦/νin(z)

u − v◦(z)

and shown to yield rapidly growing solutions even in thecollisional regime (νin�|ω|). Those results are reproducedhere in Fig. 4, which shows how the instability growth rateincreases with increments inν1 and the increasingly steepnumber density profile that accompanies them.

An approximate, closed-form expression for the growthrate highlighting its dependence on background ionosphericparameters is derived below. In our analysis, we focus on thecollisional branch of the instability thought to operate in thebottomside equatorialF region by applying the limitνin →

∞ in Eqs. (6) and (7), leading to:

d

dz

(νinn◦

dφ1

dz

)− νinn◦k

2xφ1

= −kx

d

dz

(νin

u − v◦

ω − kxv◦

dn◦

dzφ1

)We next multiply the equation byφ∗

1 and integrate over alti-tudes where the perturbed potential exists:

−

∫dz νinn◦

(|φ′

1|2+ k2

x |φ1|2)

= kx

∫dz νin

u − v◦

ω − kxv◦

n◦′φ1φ

∗

1′ (8)

where the prime notation refers to differentiation with respectto z and where integration by parts has been employed. Since

the left side of (8) is real, the frequencyω=ωr+iγ can onlybe complex valued if the eigenfunctionφ1 is also somewherecomplex. This is a necessary condition for wave growth andaffirms the conjecture ofHysell and Kudeki(2004) that theasymmetric flow pattern arising from the complex eigenfunc-tion is a crucial feature of the collisional shear instability.Complex eigenfunctions guarantee that the wavefronts (sur-faces of constant phase) that form and the convection thatfollows them are neither strictly vertical nor horizontal, ei-ther one of these geometries tending to be stable.

The real and imaginary parts of Eq. (8) yield integral equa-tions for the wave frequency and growth rate that depend onthe shape of the eigenfunction that are no easier to evalu-ate than was the original eigenvalue problem. The insta-bility in question is inherently nonlocal and is sensitive tothe ionospheric velocity, number density, and collision fre-quency profiles. However, the numerical calculations ofHy-sell and Kudeki(2004) suggested that in the vicinity of theshear node, wherev′′

◦ vanishes and about whichφ is local-ized, the complex potential can be approximated byφ(z)≈

φ◦ exp(ikzz). Utilizing this approximation and neglectingωr−kxv◦ by comparison toγ in this region, an expressionfor the growth rateγ can be derived from the real part ofEq. (8):

γ ≈kxkz

∫dz νin(u − v◦)n◦

′|φ1|

2

k2∫

dz νinn◦|φ1|2

=kxkz〈νin(u − v◦)n◦

′〉

k2〈νinn◦〉(9)

where the averaging is weighted by the norm of the potential,a function that generally peaks about one scale height belowthe shear node and extends for several scale heights aboveand below it. Note that the choice ofkx affects the modeshape and the associated value ofkz so that the dependence ofthe growth rate on the horizontal wavenumber is not apparentfrom this analysis.

Given mode shapes and values forkz derived from a com-plete boundary value analysis, Eq. (9) gives growth rate es-timates in good agreement with the eigenvalues of the fullmodel expressed by (6) and Eq. (7) (see points plotted inFig. 4). Furthermore, Eq. (9) accounts for the results of anumerical shear instability simulation reported on byHy-sell and Kudeki(2004), which produced growing waves withrelatively long e-folding times (∼50 min.) despite the pres-ence of a steep vertical density gradient and widely differ-ing plasma and neutral drift speeds in the simulation. In thattwo-dimensional simulation, shear flow was induced with theintroduction of an ad hoc constant conductivity term meantto represent electrical loading by an off-equatorialE regionplasma. The effectiveE region conductivity shorted out theF region dynamo in the valley region but not in the vicin-ity of the F peak, the net result being vertical shear. Theadd hoc term would contribute significantly to the denomi-nator of (9), which represents the factors that dissipate po-larization charge, but not to the numerator, which representsfactors that accumulate polarization charge. TheE region

D. L. Hysell et al.: Shear flow leading to ESF 2653

loading both caused essential relative plasma-neutral driftsand shorted out the polarization electric fields driven by theinstability.

In view of the evident impact of off-equatorial effects onionospheric stability, we have generalized (9), rewriting it interms of a dipole magnetic coordinate system and flux-tubeintegrated variables:

γ ≈

κφκp〈hφhq

hpνin(u − v◦)n◦

′〉

κ2φ〈

hphq

hφνinn◦〉 + κ2

p〈hφhq

hpνinn◦〉

(10)

where the angle brackets now represent averages over a mag-netic flux tube as well as over a range of apex altitudes, theprime represents a derivative with respect to thep coordi-nate, and where theκ terms are constant, nondimensionalwavenumbers related to the wavenumber through the appro-priate scale factor, e.g.kφ=κφ/hφ .

In the absence of information regarding the mode shapeof the perturbed potential, we postulate that one may stilluse Eqs. (9) or (10) to assess the stability of the ionosphere,assuming that the two wavenumber components are compa-rable and replacing the weighted average with a straight av-erage over the bottomsideF region.

5 Discussion

New experimental capabilities at Jicamarca have focused at-tention on shear flow in the postsunset bottomsideF regionionosphere and the possible role it plays in the variability ofequatorial spreadF . While the existence of shear flow hasbeen recognized for more than two decades, combined co-herent and incoherent scatter observations presently allow usto locate the shear node and to assess the shear strength quan-titatively. Shear flow exists for a period of several hours sur-rounding sunset and is most intense after sunset at low bot-tomside altitudes approaching the valley region. The scalelength for the shear is measured in tens of kilometers there.

Shear flow arises mainly from the competition between theE andF region dynamos and from demands on the verti-cal boundary currents in the bottomside imposed from aboveby the imperfectly efficientF region dynamo and from be-low by interruptions in the equatorial electrojet around theevening terminator (Haerendel et al., 1992; Haerendel andEccles, 1992). Assessing the relative importance of thesefactors quantitatively is impeded by a lack of neutral windprofile measurements in the equatorial thermosphere as wellas of off-equatorialE and valley region conductivities.

One consequence of vertical shear is the existence of re-gions where the zonal plasma flow is strongly retrograde,i.e. in the direction opposite the local neutral wind. Whenhorizontal conductivity gradients are also present in such aregion, wind-driven interchange instabilities and plasma ir-regularities results. This now appears to be the explanationfor bottom-type scattering layers, which occupy retrogradestrata and appear when the tilted postsunsetF region sup-plies the necessary zonal conductivity gradient (Kudeki and

Bhattacharyya, 1999). It should be noted that the existenceof vigorous wind-driven interchange instabilities in the bot-tomside does not by itself imply thatE region conductivityloading on the associated magnetic flux tubes is insignificant.Hysell et al.(2004) recently showed that the relatively smallwavelengths of the bottom-type plasma irregularities, com-bined with finite parallel conductivity, would allow the ir-regularities to grow even on flux tubes where theE regiondynamo is dominant.

On occasions when topside spreadF occurs, the bottom-type layers seem to be organized in clusters separated byabout 30 km. We interpret this modulation as evidence ofdecakilometric undulations or waves in the bottomsideF re-gion. The wind-driven instabilities would be stable and un-stable in the different phases of such waves. In several in-stances, the topside irregularities that followed also exhib-ited structuring with this characteristic scale size. The de-cakilometric waves therefore appear to serve as seed wavesfor full-blown spreadF . The existence of seed waves is oftenpostulated to explain the rapid appearance and developmentof radar plumes and intense plasma irregularities shortly af-ter sunset in the equatorial ionosphere. Gravity waves prop-agating into the thermosphere may seed the ionosphere aswell but are not expected to preferentially excite waves atthe observed wavelength. The∼30 km wavelengths is itselfunrelated to interchange instability, posing further questionsabout its origin. That the shear scale length is also measuredin tens of kilometers suggests a connection.

Beyond creating conditions for wind-driven instabilities,shear flow may itself affect the stability of the bottomsideF region. Boundary value analyses byGuzdar et al.(1983),Huba and Lee(1983), andSatyanarayana et al.(1984) in-dicated that shear flow generally stabilizes ionospheric in-terchange instabilities. However,Fu et al.(1986), Ronchiet al. (1989), andFlaherty et al.(1999) challenged this ideaby pointing out the limitations of boundary value analysis,which is incomplete when applied to non-normal models likethose describing sheared flows. The issue of shear flow sta-bilization remains contentious, and a number of recent the-oretical and computational studies continue to support theidea (Hassam, 1992; Sekar and Kelley, 1998; Chakrabartiand Lakhina, 2003). Chakrabarti and Lakhina(2001) furthershowed that the nonlinear evolution of the Rayleigh-Taylorinstability itself induces shear flow which contributes to thesaturation of the instability.

However, by extending the work ofKeskinen et al.(1988)to the equatorial zone,Hysell and Kudeki(2004) showedthat a collisional shear instability may exist in the postsunsetionosphere and be responsible for the precursor waves. In thecollisional limit, retrograde plasma motion rather than shearper se drives the instability. However, it is not just a variantof the standard wind-driven interchange instability, since in-stability occurs when the neutral wind and the backgrounddensity gradient are normal to one another. The instability isthe collisional branch of electrostatic Kelvin Helmholtz andis therefore aptly termed a collisional shear instability. Theinstability is not damped by collisions and does not require

2654 D. L. Hysell et al.: Shear flow leading to ESF

a prereversal enhancement and the elevation of theF layerto function. This may allow it to operate prior to the onsetof gravitational interchange instabilities, which are stabilizedby collisions, and thereby to serve as a source of seed waves.

An approximate linear growth rate expression for the col-lisional shear instability has been derived. The inherentlynonlocal nature of the instability prevents the derivation of anexact expression, but the approximation reveals at least howthe growth rate scales with background parameters. Retro-grade plasma motion and a steep vertical plasma density gra-dient in the retrograde stratum are conducive to wave growth.Large flux-tube integrated plasma conductivity suppressesgrowth. Whether the former overcomes the latter dependson the strength of the neutral wind, the cause of retrogradedrifts, and the off-equatorial conductivity distribution.

For example, if the boundary current condition is mainlyresponsible for retrograde drifts in the bottomside, then thebottomside flux tubes need not be heavily loaded by conduc-tivity in the off-equatorialE and valley regions, and shearinstability could be robust. If shear flow is mainly inducedby a strongE region dynamo controlling flux tubes in thebottomside, however, then the shear instability would likelybe damped. The existence of irregularities in the bottom-typelayers gives no clue about which scenario is operating, sincethey can exist in either case.

Thermospheric wind measurements made by the C/NOFSsatellite will be critical for quantifying the shear instabil-ity theory. In order to evaluate (10), information about theoff-equatorial conductivity distribution will also be required.This could be provided by a meridional chain of sounders,a bistatic HF sounder link, or by satellite occultation exper-iments. The same information is required both to quantifythe causes of shear flow in the equatorial ionosphere and toevaluate the growth rate of the gravitational interchange in-stabilities responsible for spreadF .

Acknowledgements.This work was supported by the National Sci-ence Foundation through cooperative agreement ATM-0432565 toCornell University and by NSF grant ATM-0225686 to Cornell Uni-versity. Additional support was received from the Air Force Re-search Laboratory through award 03C0067. The Jicamarca RadioObservatory is operated by the Instituto Geofısico del Peru, Min-istry of Education, with support from the NSF cooperative agree-ments just mentioned. The help of the staff was much appreciated.

Topical Editor M. Pinnock thanks two referees for their help inevaluating this paper.

References

Anandarao, B. G., Raghavararao, R., Desai, J. N., and Haerendel,G.: Vertical winds and turbulence over Thumba, J. Atmos. Terr.Phys., 40, 157–163, 1978.

Chakrabarti, N. and Lakhina, G. S.: Nonlinear saturation ofRayleigh-Taylor instability and generation of shear flow in equa-torial spread-F plasma, Nonlin. Proc. in Geophys., 8, 181–190,2001.

Chakrabarti, N. and Lakhina, G. S.: Collisional Rayleigh-Taylorinstability and shear-flow in equatorial spreadF plasma, Ann.

Geophys., 21, 1153–1157, 2003,SRef-ID: 1432-0576/ag/2003-21-1153.

Eccles, J. V.: A simple model of low-latitude electric fields, J. Geo-phys. Res., 103, 26 699–26 708, 1998.

Eccles, J. V., Maynard, N., and Wilson, G.: Study of the eveningdrift vortex in the low-latitude ionosphere using San Marco elec-tric field measurements, J. Geophys. Res., 104, 28 133–28 143,1999.

Farley, D. T., Bonelli, E., Fejer, B. G., and Larsen, M. F.: The pre-reversal enhancement of the zonal electric field in the equatorialionosphere, J. Geophys. Res., 91, 13 723–13 728, 1986.

Fejer, B. G.: The equatorial ionospheric electric fields. A review, J.Atmos. Terr. Phys., 43, 377–386, 1981.

Flaherty, J. P., Seyler, C. E., and Trefethen, L. N.: Large-amplitudetransient growth in the linear evolution of equatorial spreadF

with a sheared zonal flow, J. Geophys. Res., 104, 6843–6858,1999.

Fu, Z. F., Lee, L. C., and Huba, J. D.: A quasi-local theory of theE×B instability in the ionosphere, J. Geophys. Res., 91, 3263–3269, 1986.

Guzdar, P. N., Satyanarayana, P., Huba, J. D., and Ossakow, S. L.:Influence of velocity shear on Rayleigh-Taylor instability, Geo-phys. Res. Lett., 9, 547–550, 1983.

Haerendel, G. and Eccles, J. V.: The role of the equatorial electro-jet in the evening ionosphere, J. Geophys. Res., 97, 1181–1192,1992.

Haerendel, G., Eccles, J. V., and Cakir, S.: Theory for modeling theequatorial evening ionosphere and the origin of the shear in thehorizontal plasma flow, J. Geophys. Res., 97, 1209–1223, 1992.

Hassam, A. B.: Nonlinear stabilization of the Rayleigh-Taylor in-stability by external velocity shear, Phys. Fluids, 4, 485–487,1992.

Heelis, R. A., Kendall, P. C., Moffet, R. J., Windle, D., and Rish-beth, H.: Electric coupling of theE andF regions and its effectof theF region drifts and winds, Planet. Space Sci., 22, 743–756,1974.

Huba, J. D. and Lee, L. C.: Short wavelength stabilization of thegradient drift instability due to velocity shear, Geophys. Res.Lett., 10, 357–360, 1983.

Hysell, D. L. and Kudeki, E.: Collisional shear instability in theeqautorialF region ionosphere, J. Geophys. Res., 109(A11 301),2004.

Hysell, D. L., Chun, J., and Chau, J. L.: Bottom-type scatteringlayers and equatorial spreadF , Ann. Geophys., 22, 4061–4069,2004,SRef-ID: 1432-0576/ag/2004-22-4061.

Kelley, M. C., Larsen, M. F., and La Hoz, C.: Gravity wave interac-tion of equatorial spreadF : A case study, J. Geophys. Res., 86,9087–9100, 1981.

Keskinen, M. J., Mitchell, H. G., Fedder, J. A., Satyanarayana,P., Zalesak, S. T., and Huba, J. D.: Nonlinear evolution of theKelvin-Helmholtz instability in the high-latitude ionosphere, J.Geophys. Res., 93, 137–152, 1988.

Kudeki, E. and Bhattacharyya, S.: Post-sunset vortex in equatorialF -region plasma drifts and implications for bottomside spread-F , J. Geophys. Res., 104, 28 163–28 170, 1999.

Kudeki, E. and Surucu, F.: Radar interferometric imaging of field-aligned plasma irregularities in the equatorial electrojet, Geo-phys. Res. Lett., 18, 41–44, 1991.

Kudeki, E., Fejer, B. G., Farley, D. T., and Ierkic, H. M.: Inter-ferometer studies of equatorialF region irregularities and drifts,Geophys. Res. Lett., 8, 377–380, 1981.

D. L. Hysell et al.: Shear flow leading to ESF 2655

Kudeki, E., Bhattacharyya, S., and Woodman, R. F.: A new ap-proach in incoherent scatterF region E x B drift measurementsat Jicamarca, J. Geophys. Res., 104, 28 145–28 162, 1999.

Larsen, M. F. and Odom, C. D.: Observations of altitudinal and lat-itudinal E-region neutral wind gradients near sunset at the mag-netic equator, Geophys. Res. Lett., 24, 1711–1714, 1997.

Maruyama, T., Nozaki, K., Yamamoto, M., and Fukao, S.: Iono-spheric height changes at two closely separated equatorial sta-tions and implications in spreadF onsets, J. Atmos. Sol. Terr.Phys., 64, 1557–1563, 2002.

Richmond, A. D.: Equatorial electrojet, I, Development of a modelincluding winds and instabilities, J. Atmos. Terr. Phys., 35,1083–1103, 1973.

Ronchi, C., Similon, P. L., and Sudan, R. N.: A nonlocal lineartheory of the gradient drift instability in the equatorial electrojet,J. Geophys. Res., 94, 1317–1326, 1989.

Satyanarayana, P., Guzdar, P. N., Huba, J. D., and Ossakow, S. L.:Rayleigh-Taylor instability in the presence of a stratified shearlayer, J. Geophys. Res., 89, 2945–2954, 1984.

Sekar, R. and Kelley, M. C.: On the combined effects of verticalshear and zonal electric field patterns on nonlinear equatorialspreadF evolution, J. Geophys. Res., 103, 20,735–20,748, 1998.

Stenning, R. J.: A two-layer ionospheric dynamo circulation, J.Geophys. Res., 86, 3543–3550, 1981.

Takeda, M. and Maeda, H.:F region dynamo in the evening —Interpretation of equatorial1D anomaly found by MAGSAT, J.Atmos. Sol. Terr. Phys., 45, 401–408, 1983.

Tsunoda, R. T., Livingston, R. C., and Rino, C. L.: Evidence ofa velocity shear in bulk plasma motion associated with the post-sunset rise of the equatorialF layer, Geophys. Res. Lett., 8, 807–810, 1981.

Vadas, S. L. and Fritts, D. C.: Thermospheric response to gravitywaves arising from mesoscale convective complexes, J. Atmos.Sol. Terr. Phys., 66, 781–804, 2004.

Woodman, R. F.: Vertical drift velocities and east-west electricfields at the magnetic equator, J. Geophys. Res., 75, 6249–6259,1970.

Woodman, R. F.: Coherent radar imaging: Signal processing andstatistical properties, Radio Sci., 32, 2373–2392, 1997.