Under consideration for publication in J. Fluid Mech. 1 Shear instabilities in shallow-water magnetohydrodynamics 5 J. Mak†, S. D. Griffiths and D. W. Hughes 1 Department of Applied Mathematics, University of Leeds, Leeds, LS2 9JT, UK 0 2 (Received 3 February 2015) n a Withintheframeworkofshallow-watermagnetohydrodynamics,weinvestigatethelinear J instability of horizontal shear flows, influenced by an aligned magnetic field and strat- 1 ification. Various classical instability results, such as Høiland’s growth rate bound and 3 Howard’s semi-circle theorem, are extended to this shallow-water system for quite gen- ] eral profiles. Two specific piecewise-constant velocity profiles, the vortex sheet and the n rectangularjet,arestudiedanalyticallyandasymptotically;itisfoundthatthemagnetic y fieldandstratification(asmeasuredbytheFroudenumber)aregenerallybothstabilising, d but weak instabilities can be found at arbitrarily large Froude number. Numerical so- - u lutions are computed for corresponding smooth velocity profiles, the hyperbolic-tangent l f shear layer and the Bickley jet, for a uniform background field. A generalisation of the . s long-waveasymptoticanalysisofDrazin&Howard(1962)isemployedinordertounder- c stand the instability characteristics for both profiles. For the shear layer, the mechanism i s underlying the primary instability is interpreted in terms of counter-propagating Rossby y waves, thereby allowing an explication of the stabilising effects of the magnetic field and h stratification. p [ Key Words: 1 v 1 3 1 1. Introduction 0 0 The interaction of horizontal shear flows and magnetic fields in stably stratified layers . 2 is central to many problems in astrophysical fluid dynamics — involving, for example, 0 planetary interiors, stellar radiative zones and accretion discs. An important example 5 of such a flow, which has received considerable attention recently, is that of the solar 1 tachocline (see Hughes, Rosner & Weiss 2007). The tachocline, discovered via helioseis- : v mic observations, is a thin layer in the Sun, extending downwards from the (neutrally i stable) base of the convective zone to the (stably stratified) top of the radiative interior, X characterised by radial velocity shear and also planetary scale horizontal shears, associ- r a ated with the equator to pole differential rotation of the Sun. Most models of the solar dynamo invoke the tachocline as the site for the storage and generation of the Sun’s strong, predominantly toroidal magnetic field. Here we are interested in the stability of such shear flows, and how this depends upon the velocity profile, magnetic field strength, and stratification. Specifically, we consider the linear stability of a steady parallel flow and aligned magnetic field, both sheared in the horizontal cross-stream direction, in the inviscid and perfectly conducting limit. In † Email address for correspondence: [email protected]; present address: School ofMathematics,UniversityofEdinburgh,JamesClerkMaxwellBuilding,TheKing’sBuildings, Edinburgh, EH9 3FD, UK 2 J. Mak, S. D. Griffiths and D. W. Hughes (3 February 2015) thisfirststudy,weconsiderthecasewherethereisnobackgroundrotation.Thenonlinear regimeofsuchinstabilitiestypicallyleadstoturbulentflows;thesemaybeimportantfor dynamoaction,throughsomemean-fieldα-effect,andalsoforthetransportofmassand momentum, which can feed back on the large-scale flow. It is possible to examine the stability of such flows in a continuously stratified three- dimensional setting (e.g., Miura & Pritchett 1982; Cally 2003). However, here we adopt the alternative approach of considering the dynamics of a thin fluid layer under the shallow-water approximation, which is valid when the horizontal length scale of the mo- tionislongcomparedwiththedepthofthefluidlayer,asistypicallythecaseinlarge-scale astrophysical flows. This leads to a set of two-dimensional partial differential equations, withnoexplicitdependenceontheverticalco-ordinate,whichoffersaconsiderablemath- ematicalsimplification.Suchshallow-waterequationscapturethefundamentaldynamics of density stratification, including gravity waves, and allow the interaction of stratifi- cation with horizontal shear flows and magnetic fields to be analysed in the simplest possible setting. It should be noted that hydrodynamic shallow-water models, which date back to Laplace, are derived by considering a thin fluid layer of constant density bounded below by a rigid medium and above by a fluid of negligible inertia (e.g., Vallis 2006, §3.1). The corresponding reduction for electrically conducting fluids — the shallow-water magneto- hydrodynamic (SWMHD) equations of Gilman (2000) — additionally requires the fluid layer to be a perfect conductor, and to be bounded above and below by perfect conduc- tors. There are few direct astrophysical analogues for such a configuration. However, we can borrow an important idea from planetary atmospheric dynamics, where the hydro- dynamic shallow-water equations are widely used to understand waves and instabilities in a continuously stratified atmospheric layer. This is justified because there is a for- malmathematicalanalogybetweenthelinearisedequationsinthetwosystems,provided the layer depth in the shallow-water model is taken to be a so-called equivalent depth (e.g., Gill 1982, §6.11), so that the shallow-water gravity wave speed (in the horizontal) matches that of (the fastest) gravity waves in a continuously stratified layer. We have this analogy in mind throughout this study. The SWMHD equations have been studied widely in recent years. They have been showntopossessahyperbolicaswellasaHamiltonianstructure(DeSterck2001;Dellar 2002), and to support wave motions such as inertia-gravity waves and Alfv´en waves (Schecteret al.2001;Zaqarashiviliet al.2008;Heng&Spitkovsky2009).Asreviewedby Gilman & Cally (2007), they have also been used to study the linear instability of shear flows in spherical geometry, often with modelling tachocline instabilities in mind. These previousstudiesconsideredbasicstatesthatwerefunctionsonlyoflatitude,investigating the dependence of the instabilities on the strength and spatial structure of the magnetic fieldandona(reduced)gravityparameter(Gilman&Dikpati2002;Dikpatiet al.2003). In contrast to previous investigations of the instabilities of shear flows in SWMHD, which focused on global instabilities in spherical geometry, here we concentrate on local instabilities, with the aim of examining the linear instability problem in a wider con- text; for this, we consider the problem in planar geometry. We first derive some growth rate bounds and stability criteria, valid for general basic states. We then study how the instability characteristics of prototypical flows are modified by the combined action of magnetic fields and stratification, which, in isolation, are generally thought to be sta- bilising. The corresponding hydrodynamic problem has a long history, dating back to Rayleigh, and we are able to draw upon ideas and methods from a substantial literature (e.g., Drazin & Reid 1981; Vallis 2006). We start, in §2, by formulating the linear instability problem for plane-parallel basic Shear instabilities in SWMHD (3 February 2015) 3 states with the flow and field dependent on the cross-stream direction. In §3, we de- rive extensions of classical growth rate bounds, semi-circle theorems, stability criteria, and parity results for modal solutions. In §4, instabilities of piecewise-constant velocity profiles (the vortex sheet and the rectangular jet) with a uniform magnetic field are ex- amined analytically. Analogous smooth profiles (the hyperbolic-tangent shear layer and the Bickley jet) are studied in §5, both numerically and asymptotically, via a general- isation of the long-wave analysis of Drazin & Howard (1962). The primary instability mechanism for shear layers is interpreted in terms of counter-propagating Rossby waves. The results are discussed in §6. 2. Mathematical formulation 2.1. Governing equations We consider a thin layer of perfectly electrically conducting fluid moving under the influence of gravity. We use a Cartesian geometry, with horizontal coordinates x and y, and an upwards pointing coordinate z. At time t, the fluid, which is taken to be inviscid and of constant density ρ, has a free surface at z =h(x,y,t) and is bounded below by a rigid impermeable boundary at z =−H(x,y). We consider motions with a characteristic horizontal length scale L that is long com- 0 paredwithacharacteristiclayerdepthH .Onecanthenmakeashallow-waterreduction 0 in which the vertical momentum balance is taken to be magneto-hydrostatic, and for which the horizontal velocity u and horizontal magnetic field B are independent of z. When the bottom boundary is perfectly electrically conducting (and is thus a magnetic field line) and the free surface remains a field line, the magnetic shallow-water equations of Gilman (2000) are obtained. These are an extension of the classical shallow-water equations of geophysical fluid dynamics. We use these equations in non-dimensional form. We denote the characteristic hori- zontal velocity of the basic state by U , and the characteristic magnetic field strength 0 by B . Non-dimensionalising x and y by L , t by the advective time-scale L /U , H by 0 0 0 0 H ,hbyU2/g (whereg istheaccelerationduetogravity),velocitybyU ,andmagnetic 0 0 0 field by B , the SWMHD equations are 0 ∂u +u·∇u=−∇h+M2B·∇B, (2.1a) ∂t ∂B +u·∇B =B·∇u, (2.1b) ∂t F2∂h +∇·(cid:0)(H +F2h)u(cid:1)=0, (2.1c) ∂t √ √ whereF =U / gH andM =(B / µρ)/U ,withµbeingthepermeabilityofthefluid. 0 0 0 0 In addition to (2.1a–c), the shallow-water reduction implies ∇·(cid:0)(H +F2h)B(cid:1)=0. (2.2) However, (2.2) need not be considered explicitly; if it is satisfied at some initial time, then (2.1a–c) guarantee that it remains satisfied for all time. Thesystemhastwonon-dimensionalparameters.TheFroudenumberF istheratioof √ the characteristic horizontal velocity of the basic state to the gravity wave speed gH 0 (and is related to the reduced gravity parameter G of Gilman & Dikpati (2002) via √ G = F−2). The parameter M is the ratio of the Alfv´en wave speed B / µρ to the 0 characteristic horizontal velocity of the basic state. When H is constant and F → 0, (2.1c) and (2.2) become ∇ · u = 0 and ∇ · B = 0 respectively, and we recover the 4 J. Mak, S. D. Griffiths and D. W. Hughes (3 February 2015) equations for two-dimensional incompressible magnetohydrodynamics, with h playing the role of pressure. When M → 0, (2.1b) decouples from (2.1a–c), and we recover the hydrodynamic shallow-water equations; these have a well-known correspondence with two-dimensional compressible hydrodynamics (e.g., Vallis 2006, §3.1), which we exploit from time to time. As an example of astrophysical parameter values, we estimate F and M in the solar tachocline,usingdatafromGough(2007).WesetU tobetheequatortopoledifference 0 in the zonal velocity, implying U ≈ 500ms−1. There is considerable uncertainty in the 0 strength of the magnetic field in the tachocline (Hughes et al. 2007), although a likely range is 103G (cid:46) B (cid:46) 105G. Then, taking ρ = 210kgm−3, we find 0.01 (cid:46) M (cid:46) 1. 0 √ To estimate F, we must choose a gravity wave speed gH for the layer. One means of 0 doingthisistotakeH tobethedepthofthetachoclineandto interpretg asareduced 0 gravity,accountingforthefractionaldensitydifferenceoftheoverlyingfluid,asinDikpati &Gilman(2001).However,herewepursuetheanalogybetweenshallow-waterflowsand those of a continuously stratified layer with buoyancy frequency N and depth H , and √ 1 choose gH tobethespeedofthefastestgravitywaveinsuchalayer,whichisNH /π 0 1 (Gill1982,§6.11).TakingH ≈2×107m(i.e.0.03R ,whereR isthesolarradius)and 1 (cid:12) (cid:12) N ≈8×10−4s−1, which are bulk values that might describe a mode spanning the entire √ tachocline, gives a gravity wave speed NH /π = gH ≈ 5000ms−1, corresponding to 1 0 an equivalent depth H ≈ 50km (taking g ≈ 540ms−2). Again taking U ≈ 500ms−1, 0 0 we thus estimate F ≈ 0.1, although it is clear that F would be somewhat smaller or larger if one considered motions towards the top of the radiative zone (with stronger stratification) or towards the base of the convection zone (with weaker stratification). 2.2. The linear instability problem AboveatopographyoftheformH =H(y),weconsiderabasicstateh=0,u=U(y)e x and B = B(y)e , so that the magnetic field is initially aligned with the flow. We then x consider perturbations in h, u=(u,v) and B =(b ,b ) to this state of the form x y ξ(x,y,t)=Re{ξˆ(y)exp(iα(x−ct))}, (2.3) where α is the (real) wavenumber and c is the (complex) phase speed. Dropping the hatted notation, the linear evolution is described by (cid:18) (cid:19) (cid:18) (cid:19) ∂ ∂ ∂h ∂b +U u+U(cid:48)v =− +M2 B x +B(cid:48)b , (2.4a) y ∂t ∂x ∂x ∂x (cid:18) (cid:19) ∂ ∂ ∂h ∂b +U v =− +M2B y, (2.4b) ∂t ∂x ∂y ∂x (cid:18) (cid:19) ∂ ∂ ∂u +U b +B(cid:48)v =B +U(cid:48)b , (2.4c) x y ∂t ∂x ∂x (cid:18) (cid:19) ∂ ∂ ∂v +U b =B , (2.4d) y ∂t ∂x ∂x (cid:18) (cid:19) (cid:18) (cid:19) ∂ ∂ ∂u ∂v F2 +U h+H + +H(cid:48)v =0, (2.4e) ∂t ∂x ∂x ∂y where a prime denotes differentiation. Eliminating for v, we obtain (cid:18) S2(Hv)(cid:48) (cid:19)(cid:48) (cid:32) α2S2 U(cid:48) (cid:18) S2 (cid:19)(cid:48) Q(cid:48)S2 (cid:33) − − + Hv =0, H(U −c)2K2 H(U −c)2 H(U −c) (U −c)2K2 (U −c)3K2 (2.5) Shear instabilities in SWMHD (3 February 2015) 5 where Q=−U(cid:48)/H is the background potential vorticity, and S2(y)=(U(y)−c)2−M2B2(y), K2(y)=1−F2S2(y). (2.6) Following Howard (1961), under the transformation Hv = (U − c)G, equation (2.5) becomes (cid:18)S2 G(cid:48)(cid:19)(cid:48) α2S2 − G=0. (2.7) K2 H H Weshallusethismorecompactformfortheremainderofthisstudy.Inthenon-magnetic shallow-water limit (M =0), (2.5) reduces to equation (3.4) of Balmforth (1999). In the two-dimensional incompressible magnetohydrodynamic limit (F = 0 and H = 1), (2.7) reduces to equation (3.5) of Hughes & Tobias (2001). Weshallconsider(2.7)ineitheranunboundeddomain,forwhich|G|→0as|y|→∞, or in a bounded domain with rigid side walls, where G = 0 and hence b = 0 via y (2.4d). Either way, for given real α, (2.7) is then an eigenvalue problem for the unknown phase speed c = c +ic . We will focus on instabilities, i.e. c (cid:54)= 0, in which case (2.7) r i i has no singularities for real values of y. Since the transformation α → −α leaves (2.7) unchanged,wemaytakeα(cid:62)0withoutlossofgenerality.Instabilitythenoccursifc >0, i with growth rate αc . i 3. General theorems InthissectionwederivethreeresultsthatholdforgeneralshearflowsU(y):twoprovide bounds on the growth rate of any instability, whereas the third concerns implications of the parity of the basic state flow. 3.1. Growth rate bound Aboundontheinstabilitygrowthratemaybeobtainedbycalculatingtherateofchange of the total disturbance energy using the combination Hu∗×(2.4a)+Hv∗×(2.4b)+(M2Hb∗)×(2.4c)+(M2Hb∗)×(2.4d)+h×(2.4e), x y where ∗ denotes complex conjugate. On adopting the form (2.3) for the perturbations, the real part of this expression gives (on dropping hats) αc (cid:0)H(cid:0)|u|2+|v|2+M2|b |2+M2|b |2(cid:1)+F2|h|2(cid:1)= i x y −Re(cid:0)HU(cid:48)(cid:0)vu∗−M2b∗b (cid:1)+M2HB(cid:48)(vb∗ −u∗b )(cid:1)−Re d (Hvh∗). (3.1) x y x y dy On integrating over the y domain, employing the boundary condition on v, and manip- ulating the remaining terms on the right hand side using ±2Re(pq∗) (cid:54) |p|2 +|q|2, we obtain the following bound on the growth rate: 1 αc (cid:54) (max|U(cid:48)|+Mmax|B(cid:48)|). (3.2) i 2 Intheabsenceofmagneticfield,thisreducestothewell-knownboundinhydrodynamics (Høiland 1953; Howard 1961). 3.2. Semi-circle theorems In a classic paper, Howard (1961) proved that for incompressible hydrodynamic parallel shear flows, the wave speed c of any unstable mode must lie within a semi-circle in the complexplanedeterminedbypropertiesofthebasicstateflow.Subsequently,semi-circle 6 J. Mak, S. D. Griffiths and D. W. Hughes (3 February 2015) theoremshavebeenderivedforseveralotherhydrodynamicalandhydromagneticsystems (e.g., Collings & Grimshaw 1980; Hayashi & Young 1987; Shivamoggi & Debnath 1987; Hughes & Tobias 2001). In a similar manner, a semi-circle theorem may be derived for the SWMHD system. Multiplyingequation(2.7)byG∗,integratingovery andusingtheboundarycondition on v (and hence G) gives the relation (cid:90) S2 |G(cid:48)|2 (cid:90) S2|G|2 dy+α2 dy =0. (3.3) K2 H H The imaginary part of (3.3) gives (cid:90) |G(cid:48)|2 |G|2 c (U −c )χdy =0, where χ= +α2 (cid:62)0. (3.4) i r H|K|4 H Equation (3.4) immediately yields Rayleigh’s result that for unstable modes (c >0), c i r lies in the range of U (i.e. U (cid:54)c (cid:54)U , where the subscripts ‘min’ and ‘max’ refer min r max to the minimum and maximum values across the domain). On using equation (3.4), the real part of (3.3) gives (c2+c2)(cid:90) χdy =(cid:90) χ(cid:0)U2−M2B2(cid:1) dy−F2(cid:90) |S|4 |G(cid:48)|2dy, (3.5) r i H|K|4 which implies that (cid:90) (cid:90) 0(cid:54)(c2+c2) χdy (cid:54)(cid:0)U2−M2B2(cid:1) χdy. (3.6) r i max This gives the first semi-circle bound: the complex wave speed c of an unstable eigen- function must lie within the region defined by c2+c2 (cid:54)(cid:0)U2−M2B2(cid:1) . (3.7) r i max Thesecondsemi-circleboundisobtained,inthestandardmanner,fromtheinequality 0(cid:62)(cid:82)(U−U )(U−U )χdy.Substitutingfrom(3.4)andderivinganinequalityfrom max min (3.5) leads to the expression (cid:90) 0(cid:62)(cid:0)c2+c2−(U +U )c +U U +M2(B2) (cid:1) χdy, (3.8) r i min max r min max min which gives the second semi-circle bound: the speed c of an unstable eigenfunction must lie within the region defined by (cid:18) U +U (cid:19)2 (cid:18)U −U (cid:19)2 c − min max +c2 (cid:54) max min −M2(B2) . (3.9) r 2 i 2 min Thus,takingtheseresultstogether,theeigenvaluecofanunstablemodemustliewithin the intersection of the two semi-circles defined by (3.7) and (3.9). In the absence of magnetic field, semi-circle (3.9) lies wholly within semi-circle (3.7), and we recover the well-known result of Howard (1961). However, as observed by Hughes & Tobias (2001), who considered the stability of aligned fields and flows in incompressible MHD, for non- zero magnetic field there is the possibility of the two semi-circles overlapping, being disjoint, or indeed ceasing to exist; thus, in addition to giving eigenvalue bounds for unstable modes, these results also provide sufficient conditions for stability. From (3.7) and (3.9) it therefore follows that the basic state is linearly stable if any one of the following three conditions is satisfied: M|B|(cid:62)|U| everywhere in the domain; (3.10) Shear instabilities in SWMHD (3 February 2015) 7 |U −U | M|B| (cid:62) max min ; (3.11) min 2 (cid:115) U +U (cid:18)U −U (cid:19)2 (cid:113) max min − max min +M2(B2) (cid:62) (U2−M2B2) . (3.12) 2 2 min max TheseresultsareequivalenttothosegivenbyHughes&Tobias(2001)forincompressible MHD. A drawback of the above approach is that the bounds do not contain the Froude number F. Although it is possible to introduce F into the semi-circle bounds using similar manipulations to that employed by Pedlosky (1964), as shown by Mak (2013) this does not sharpen the bound and we thus omit it. 3.3. Consequences of basic state parity For the hydrodynamic case, it can be shown that symmetries of the basic state lead to symmetries in the stability problem (Howard 1963). These results may be generalised to SWMHD if we make the further assumptions that B2(y) and H(y) are even functions about y =0. We first consider the case when U(y) is odd about y =0. Equation (2.7) is unchanged under c → −c and G(y) → G(−y). Since the equation is also unchanged under c → c∗ and G → G∗, it follows that an eigenfunction with eigenvalue c = c + c must be r i accompanied by eigenfunctions with c = ±c ±ic . Thus unstable solutions either have r i c =0 or are a pair of counter-propagating waves with the same phase speed. As argued r by Howard (1963), the symmetry in the basic state implies that there is no preferred direction for wave propagation, consistent with the form of the eigenvalues. Now consider the case when U(y) is even about y =0. Then 1 1 G (y)= (G(y)+G(−y)) and G (y)= (G(y)−G(−y)) (3.13) e o 2 2 are also eigenfunctions of (2.7). Following Drazin & Howard (1966), if we now take G o multiplied by (2.7) with G = G and subtract this from G multiplied by (2.7) with e e G=G , integrating over −L (cid:54)y (cid:54)L gives o y y W(G ,G )≡[G(cid:48)G −G(cid:48)G ]+Ly =constant=0, (3.14) e o e o o e −Ly owing to the imposed boundary conditions on the eigenfunction. The vanishing of the Wronskian W implies that the functions G and G are linearly dependent throughout e o the domain, which is possible only if one of them is identically zero. Thus an unstable eigenfunction corresponding to a particular eigenvalue is either an even or odd function about y =0. 4. Piecewise-constant profiles: vortex sheet and rectangular jet Wenowconsidersomesimpleflowconfigurationsforwhichtheeigenvalueproblem(2.7) canbereducedtoanalgebraicequationforc,fromwhichtheconditionsforstabilitycan bereadilydetermined.Todothis,wetakeH =1(notopography)andB =1(auniform magnetic field). We seek solutions of (2.7) in an unbounded domain, with |G|→0 as |y|→∞. (4.1) WeconsidervelocityprofilesU(y)thatarepiecewiseconstant.IfU(y)isdiscontinuous at y = y , then the eigenfunction G must satisfy two jump conditions at y = y . In the 0 0 8 J. Mak, S. D. Griffiths and D. W. Hughes (3 February 2015) usual way, the (linearised) kinematic boundary condition implies (cid:20) (cid:21)y+ v 0 =[G]y0+ =0. (4.2a) U −c y− y− 0 0 Thepressure(orfreesurfacedisplacement)isalsocontinuousaty =y .Thecorrespond- 0 ing condition on G is most easily derived by integrating (2.7) across y =y , yielding 0 (cid:20)S2 (cid:21)y0+ G(cid:48) =0. (4.2b) K2 y− 0 4.1. Vortex sheet We first consider the velocity profile (cid:40) +1, y >0, U(y)= −1, y <0. Then, for y (cid:54)=0, (2.7) becomes G(cid:48)(cid:48)−α2K2G=0. Using (4.1) and (4.2a), we thus find (cid:40) exp(−αK y), y >0, + G(y)= (4.3) exp(+αK y), y <0, − where (cid:112) K = 1−F2((1∓c)2−M2), Re(K )>0. (4.4) ± ± The second jump condition (4.2b) then implies an eigenvalue relation for c: (1−c)2−M2 (1+c)2−M2 + =0. (4.5) K K + − Note that c is independent of the wavenumber α, so any unstable mode with c >0 has i an unbounded growth rate as α → ∞. This is an artefact of considering ideal fluids; viscosity will preferentially suppress small scales and remove this unphysical behaviour. There are several special cases. When F = M = 0, we recover the classical Kelvin– Helmholtz instability with c = ±i. When F = 0 but M (cid:54)= 0, (4.5) reduces to the incompressible MHD case of Michael (1955), with c2 = −(1−M2); thus, the Kelvin– Helmholtz instability is stabilised when M (cid:62) 1, since the disturbance has to do work to bend the field lines. When M = 0 but F (cid:54)= 0, (4.5) gives the classical hydrodynamic shallow-water dispersion relation, which is analogous to that of two-dimensional com- √ pressible hydrodynamics. The Kelvin–Helmholtz instability is stabilised when F (cid:62) 2 (Miles 1958; Bazdenkov & Pogutse 1983), since the disturbance has to do work to move the free surface against gravity. Thus, increasing F or M in the absence of the other is stabilising. In the general case where F and M are both non-zero, (4.5) can be rearranged and squared to yield a quartic equation for c: F2c4−2(cid:0)1+F2(cid:0)M2+1(cid:1)(cid:1)c2+(cid:0)M2−1(cid:1)(cid:0)2+F2(cid:0)M2−1(cid:1)(cid:1)=0. (4.6) Here we have ignored the degenerate case with c = 0, which is a solution of (4.5) when M =1.Bycomparingsolutionsof(4.6)withthoseof(4.5)foundusingaNewtoniteration method, we find that only two roots of (4.6) also satisfy (4.5): these are c=±c , where v (cid:32)√ (cid:33)1/2 1+4F2+4F4M2−(1+F2+F2M2) c =i . (4.7) v F2 Shear instabilities in SWMHD (3 February 2015) 9 c i 5 1 4 0.8 3 0.6 F 2 0.4 √2 1 0.2 0 0 0 0.5 1 M Figure 1.ContoursofIm(c ),givenbyexpression(4.7),withstabilityboundaries(4.8)inred. v A contour plot of Im(c ) is shown in figure 1. From (4.7), there is instability only if v (cid:114) 2 M <1 and F < . (4.8) 1−M2 AlthoughincreasingF isalwaysstabilisingatfixedM,thecriticalvalueofF abovewhich the flow is stable increases as M increases towards 1. Thus, although magnetic field and free-surface effects are stabilising in isolation, together they can lead to instabilities at arbitrarily large values of F, provided 2 1− <M2 <1. (4.9) F2 Using an asymptotic analysis, it is possible to investigate these instabilities further at large F and with M just smaller than unity. Rewriting (4.7) in terms of 1−M2 and expanding for |1−M2|(cid:28)1, we obtain (cid:18)1−M2 2F6(1−M2)2(cid:19)1/2 c ∼i − , |1−M2|(cid:28)1, (4.10) v 1+2F2 (1+2F2)3 where terms of O(cid:0)(1−M2)3(cid:1) have been neglected. When F = O(1), the first term on the right-hand side of (4.10) dominates. However, in the regime of interest (4.9) with F2 ∼ (1−M2)−1 (cid:29) 1, the two terms on the right-hand side of (4.10) have the same order of magnitude, and instead we obtain (cid:18)1−M2 (1−M2)2(cid:19)1/2 c ∼i − as F−2 ∼(1−M2)→0. (4.11) v 2F2 4 Thissimpleformulaisconsistentwithbothstabilityboundariesin(4.8),and,asshownin figure2,closelypredictsc inthisweakinstabilityregime,evenwhenF isoforderunity. i Using(4.11),itisstraightforwardtoshowthatIm(c )ismaximisedwhenM2 =1−1/F2, v with c ∼ i/(2F2), so that the growth rate of the most unstable mode decays like F−2 v in this regime. 10 J. Mak, S. D. Griffiths and D. W. Hughes (3 February 2015) 0.16 0.03 (a) (b) 0.12 0.02 ci 0.08 0.01 0.04 0.7 0.8 0.9 1 0.9 0.95 1 1.05 M Figure 2. The weak instability regime of the vortex sheet for (a) F =2, (b) F =5, as determined directly from (4.7) (crosses) and from the asymptotic result (4.11) (line). 4.2. Rectangular jet We now consider the top-hat velocity profile (cid:40) 1, |y|<1, U(y)= (4.12) 0, |y|>1. Then, (2.7) and (4.1) imply  A exp(−αK (y−1)), y >+1,  + 0 G= A cosh(αK y)+A sinh(αK y), |y|<1, (4.13) e 1 o 1  A exp(+αK (y+1)), y <−1, − 0 for some A , A , A and A , where + − e o (cid:112) K = 1−F2(c2−M2), with Re(K )>0 for bounded solutions, (4.14a) 0 0 (cid:112) π π K = 1−F2((1−c)2−M2), with − <arg(K )(cid:54) . (4.14b) 1 1 2 2 HerewefollowRayleigh’sformulation(Drazin&Reid1981)andconsidereigenfunctions that are either even or odd. For the even mode, we set A = 0, A = A and write o + − c=c . Then (4.2a,b) and (4.13) give e c2−M2 (1−c )2−M2 e + e tanh(αK )=0. (4.15) 1 K K 0 1 Fortheoddmode,wesetA =0,A =−A andwritec=c .Then(4.2a,b)and(4.13) e + − o give c2−M2 (1−c )2−M2 o + o coth(αK )=0. (4.16) 1 K K 0 1 In contrast to the vortex sheet dispersion relation (4.5), here c depends upon α. Two special cases may be solved analytically. When F = M = 0, so that K = 1, 0,1 expressions (4.15) and (4.16) yield unstable modes with √ √ T +i T 1+i T c = , c = , where T =tanhα, (4.17) e o 1+T 1+T so that the flow is unstable for all α, with Im(c) approaching a maximum value of 1/2 as α→∞ (Rayleigh 1878). When F =0 but M (cid:54)=0, so that K =1 again, (4.15) and 0,1