diamond aps 2015 - university of california, san diego aps 2015.pdf · title: microsoft powerpoint...
TRANSCRIPT
![Page 1: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/1.jpg)
A Reduced Model of ExB and PV Staircase Formation and Dynamics
P. H. Diamond, M. Malkov, A. Ashourvan; UCSD
D.W. Hughes; Leeds
1
![Page 2: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/2.jpg)
Outline
• Basic Ideas: transport bifurcation and ‘negative diffusion’
phenomena
– Inhomogeneous mixing: times and lengths
– Feedback loops
– Simple example
• Inhomogeneous mixing in space:
– Staircase models in QG and drift wave system
– Models, background
– QG staircase: model and results
– Aside: on PV mixing
– Model for Hasegawa-Wakatani system – 2 field, and intro
• Some Lessons and Conclusions
![Page 3: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/3.jpg)
I) Basic Ideas:Transport bifurcations and
‘negative diffusion’ phenomena
3
![Page 4: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/4.jpg)
J.W. Huges et al., PSFC/JA-05-35
Transport Barrier Formation (Edge and Internal)
• Observation of ETB formation (L→H transition)− THE notable discovery in last 30 yrs of MFE
research− Numerous extensions: ITB, I-mode, etc.− Mechanism: turbulence/transport suppression
by ExB shear layers generated by turbulence
• Physics:− Spatio-temporal development of bifurcation
front in evolving flux landscape− Cause of hysteresis, dynamics of back
transition
• Fusion:− Pedestal width (along with MHD) → ITER
ignition, performance− ITB control → AT mode− Hysteresis + back transition → ITER operation
S-curve
![Page 5: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/5.jpg)
Why Transport Bifurcation? BDT ‘90, Hinton ‘91
• Sheared × flow quenches turbulence, transport è
intensity, phase correlations
• Gradient + electric field è feedback loop
i.e. = − × è = ()è minimal model = − −
turbulent transport+ shear suppression
Residual collisional
n ≡ quenching exponent
![Page 6: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/6.jpg)
• Feedback:
Q ↑ à ↑ à ↑ à / , ↓
è ↑ à …
• Result:
1st order transition (LàH):
Heat flux vs T profiles a) L-modeb) H-mode
![Page 7: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/7.jpg)
• S curve è “negative diffusivity” i.e. / < 0
• Transport bifurcations observed and intensively
studied in MFE since 1982 yet:
è Little concern with staircases
è Key questions:
1) Might observed barriers form via step coalescence in
staircases?
2) Is zonal flow pattern really a staircase (see GDP)?
![Page 8: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/8.jpg)
• What is a staircase? – sequence of transport barriers
• Cf Phillips’72:
• Instability of mean + turbulence field requiring:Γ/ < 0 ; flux dropping with increased gradientΓ = −, = / • Obvious similarity to transport bifurcation
8
(other approaches possible)
Staircase in Fluids
![Page 9: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/9.jpg)
9
In other words:
b
gradient
Iintensity
Some resemblance to Langmuir turbulencei.e. for Langmuir: caviton train / ≈ −
Configuration instability of profile + turbulence intensity field
Buoyancyprofile
Intensity field
è end state ofmodulational instability !?
![Page 10: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/10.jpg)
• The physics: Negative Diffusion (BLY, ‘98)
• Instability driven by local transport bifurcation
• Γ/ < 0è ‘negative diffusion’
• Feedback loop Γ ↓ à ↑ à ↓ à Γ ↓10
Negative slopeUnstable branch
Γ
“H-mode” like branch(i.e. residual collisional diffusion)is not input- Usually no residual diffusion- ‘branch’ upswing à nonlinear
processes (turbulence spreading)- If significant molecular diffusion
à second branch
Critical element: → mixing length
è
è
![Page 11: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/11.jpg)
The Critical Element: Mixing Length
• Sets range of inhomogeneous mixing
• = + • ~ Ozmidov scale, smallest ‘stratified scale’
ßà balance of buoyancy and production
• ≈ è dependence is crucial for inhomogeneous
mixing
• Feedback loop: ↑ à ↓ à ↓
![Page 12: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/12.jpg)
• Plot of (solid) and (dotted) at
early time. Buoyancy flux is
dashed à near constant in core
12
• Later time à more akin
expected “staircase pattern”.
Some condensation into larger
scale structures has occurred.
• A Few Results à staircases
![Page 13: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/13.jpg)
II) Inhomogeneous Mixing in Space:Staircase Models in QG and
Drift Wave Systems
13
![Page 14: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/14.jpg)
• Hasegawa-Mima ( )
Drift wave model – Fundamental prototype
• Hasegawa-Wakatani : simplest model incorporating instability
14
ddtn = -D||Ñ||
2 (f - n)+D0Ñ2n
rs2 ddtÑ2f = -D||Ñ||
2 (f - n)+nÑ2Ñ2fÑ^ × J^ +Ñ||J|| = 0
hJ|| = -Ñ||f +Ñ||pe
dnedt
+ Ñ||J||-n0 e
= 0
à vorticity:
à density:
V = cBz ´Ñf +Vpol
J^ = n e Vipol
ddtn-Ñ2f( ) = 0
à zonal flow being a counterpart of particle flux
à PV flux = particle flux + vorticity flux
à PV conservation in inviscid theory
QL:
à?
D||k2|| /w >>1 ® n ~ f
ddt
f - rs2Ñ2f( )+u*¶yf = 0
![Page 15: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/15.jpg)
Physics: ZF!
• Key: PV conservation dq/dt=0
15
relative vorticity
planetaryvorticity
density (guiding center)
q = n-Ñ2f
ion vorticity(polarization)
GFD: Plasma: Quasi-geostrophic system Hasegawa-Wakatani system
¶¶t
Ñ2y - Ld-2y( )+ b ¶
¶xy + J(y,Ñ2y) = 0
q = Ñ2y +by
H-W à H-M:
Q-G:
Physics: Dy®D Ñ2y( ) Dr®Dn®D Ñ2f( )
1wci
¶¶t
Ñ2f - rs-2f( )- 1Ln
¶¶yf + rsLnJ(f,Ñ2f) = 0
• Charney-Haswgawa-Mima equation
![Page 16: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/16.jpg)
Staircase in QG Turbulence: A Model
• PV staircases observed in nature, and in the unnatural
• Formulate ‘minimal’ dynamical model ?! (n.b. Dritschel-McIntyre 2008
does not address dynamics)
Observe:
• 1D adequate: for ZF need ‘inhomogeneous PV mixing’ + 1 direction of
symmetry. Expect ZF staircase
• Best formulate intensity dynamics in terms potential enstrophy = ⟨⟩• Length? : Γ / ∼ • à ∼ / / ∼
16
(production-dissipation balance)
(i.e. ~)
![Page 17: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/17.jpg)
Model:
Mean: = Potential Enstrophy density: − = − + Where:
= + ∼ (dimensional) + = 0, to forcing, dissipation
17
Spreading Production
Dissipation
Forcing
= / ≈
Γ = = −⟨⟩/ is fundamental quantity
è
è
![Page 18: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/18.jpg)
Alternative Perspective:
• Note: = / à /
• Reminiscent of weak turbulence perspective:
= = ∑ Ala’ Dupree’67:
≈ ∑ − /Steeper ⟨⟩′ quenches diffusion à barrier via PV gradient feedback
18
= − /Δ ≈
( ∼ 1)
![Page 19: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/19.jpg)
≈ 1 + • vs Δ dependence gives roll-over with steepening
• Rhines scale appears naturally, in feedback strength
• Recovers effectively same model
Physics:
① “Rossby wave elasticity’ (MM) à steeper ⟨⟩′à stronger
memory (i.e. more ‘waves’ vs turbulence)
② Distinct from shear suppression à interesting to dis-entangle
19
ß
![Page 20: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/20.jpg)
• What of wave momentum? Austauch ansatz
Debatable (McIntyre) - but (?)…
• PV mixing ßà ⟨⟩So à à à R.S.
• But:
R.S. ßà ⟨⟩ßà è Feedback:⟨⟩′ ↑ à ↓ à ↓ à ↓
20
(Production)
Aside
- Equivalent!- Formulate in terms mean,
Pseudomomentum?- Red herring for barriers
à quenched*
![Page 21: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/21.jpg)
Numerical Results:Analysis of QG Model
21
![Page 22: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/22.jpg)
• Re-scaled system
= / + for mean
= // + / − + 1 / for P.E.
• Note:
– Quenching exponent = 2 for saturated modulational instability
– Potential enstrophy conserved to forcing, dissipation, boundary
– System size L è strength of drive
drive
![Page 23: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/23.jpg)
• Weak Drive è 1 step staircase
– 1 step staircase forms
– Small scales not evident
– Dirichlet B.C.’s
Initial Δ All
![Page 24: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/24.jpg)
• Increased Drive è Multi-step structure
– Multiple steps
– Steps move
– Some hint of step condensation at foot
of Q profile
– End state: barrier on LHS, step on RHS
è Suggestive of barrier formation by
staircase condensation
*
![Page 25: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/25.jpg)
• plot reveals structure and scales involved
– FW HM max, min capture width
of steep gradient region
– Step width - minimum
![Page 26: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/26.jpg)
• Mergers occur
– Same drive as before
– Staircase smooths
t x
![Page 27: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/27.jpg)
• plot of Mergers
– Can see region of peak expanding
– Coalescence of steps
occurs
– Some evidence for
“bubble competition”
behavior
tx
![Page 28: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/28.jpg)
• Mergers for yet stronger drive
– Mergers form broad
barrier region at foot of
Q profile
– Extended mean barrier
emerging from step
condensationt x
barrier
![Page 29: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/29.jpg)
– evolution in
condensation process à
same scale
– Note broadening of high region near boundaryt
x
![Page 30: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/30.jpg)
• Staircase Barrier Structure vs Drive
– ↑ è increasing
drive
– FW HM max increases
with , so
– Width of barrier
expands as increases
![Page 31: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/31.jpg)
• Staircase Step Structure vs Drive
– Step width decreases
and ~ saturated as increases
– Min, max converge
![Page 32: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/32.jpg)
• More interesting model…
– From Hasegawa-Wakatani:
= ∥∥ − + n = D∥∥ − + + ⋅ = ; ∥∥ > 1; > – Evident that mean-field dynamics controlled by:
– Γ = ⟨⟩à particle flux
– Γ = ⟨⟩à vorticity flux Relation of corrugations and shear layers
![Page 33: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/33.jpg)
• K- Model + Γ = + Γ = = Pot Enstr = − + Γ = − Γ − Γ − − + – Total P.E. conserved, manifestly
– Γ = ⟨⟩à spreading flux
– Forcing as linear stage irrelevant
mean
=
![Page 34: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/34.jpg)
• Fluxes Γ, Γ– Could proceed as before è PV mixing with feedback for
steepened – i.e. Γ = −– Γ = −– ~ /, with 1/ = 1/ + 1/– = /[ + − ]è
![Page 35: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/35.jpg)
• Fluxes Γ, Γ– Could proceed as before è PV mixing with feedback for
steepened – i.e. Γ = −– Γ = −– ~ /, with 1/ = 1/ + 1/– = /[ + − ]èFeedback by steepening and reduced etc
èBarrier structure?!
![Page 36: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/36.jpg)
• More interesting: As CDW turbulence is wave turbulence, use
mean field/QL theory as guide to model construction
• For QL theory, see Ashourvan, P.D., Gurcan (2015)
• Simplified:Γ ≈ − = − , = ∥∥• Key: electron response laminar
• Neglected weak particle pinch
(a)
![Page 37: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/37.jpg)
Γ = − + Π ≈ ⟨⟩(/ − )à /||Π ≈ Γ/ − = −And ~ N.B.: In QLT, ≠
• Interesting to note varied roles of:
– Transport coefficients , – Non-diffusive stress
– Length scale, suppression exponent
– Intensity dependence
(b)
![Page 38: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/38.jpg)
• Studies so far:
– = with feedback as in QG via = 1, 2– QL model with () = 1,2
• à these constitute perhaps the simplest cases conceivable…
![Page 39: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/39.jpg)
• = + , = à mixing = 1
• Barrier and irregular staircase form
• Shear layer self-organizes near boundary
![Page 40: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/40.jpg)
• = + , = à mixing = 2
• Density and vorticity staircase form
• Regular in structure
• Condensation to large steps, barrier forms
![Page 41: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/41.jpg)
• Quasilinear with feedback ≠ , Π ≠ 0 = 1
• Single barrier…..
![Page 42: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/42.jpg)
• Quasilinear with feedback ≠ , Π ≠ 0 = 2
• Density staircase forms and condenses to single edge transport barrier
![Page 43: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/43.jpg)
• Quasilinear with feedback ≠ , Π ≠ 0 = 2
![Page 44: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/44.jpg)
• Process of mergers
![Page 45: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/45.jpg)
• What did we learn?
– Absolutely simplest model recovers staircase
– Boundary shear layer forms spontaneously
– * Mergers and propagation down density gradient
form macroscopic edge transport barrier from
mesoscopic staircase steps!
– (gradient) feedback seems essential
45
![Page 46: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/46.jpg)
Discussion• “Negative diffusion” / clustering instability common to Phillips, QG and DW
transport bifurcation and Jam mechanisms:
– Γ/ < 0à Γ nonlinearity
– Γ/ < 0à Γ nonlinearly
• Key elements are:
– Inhomogeneity in mixing: length scale , and its dependence, , etc
– Feedback loop structure
• Evidence of step coalescence to form larger scale barriers à pragmatic
interest
*
![Page 47: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/47.jpg)
Areas for further study:
• Structure of mixing representation, form of mixing scales
à , • Non-diffusive flux contributions, form
• Further study of multiple field systems, i.e. H-W: , , • Role of residual transport, spreading
• Step coalescence
• Shear vs PV gradient feedback in QG systems
![Page 48: Diamond APS 2015 - University of California, San Diego APS 2015.pdf · Title: Microsoft PowerPoint - Diamond APS 2015 Author: jkwon Created Date: 11/11/2015 2:38:44 PM](https://reader034.vdocument.in/reader034/viewer/2022051918/600a78307f12c40e4a28edcd/html5/thumbnails/48.jpg)
Final Observation:
Staircases are becoming crowded…
Acknowledgment: “This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Fusion Energy Sciences, under Award NumberDE-FG02-04ER54738.”