Dynamic behavior of nonlocallycoupled ion transport in toroidal plasmas

1
Dynamic behavior of nonlocally-coupled ion
transport in toroidal plasmas

• N. Tamura1, S. Inagaki2, K. Ida1, M. Yoshinuma1,
  K. Tanaka1,
• C. Michael1, T. Tokuzawa1, R. Sakamoto1, T.
  Shimozuma1, S. Kubo1,
• K. Itoh1, T. Fukuda3, Y. Nagayama1, K.
  Kawahata1,
• S. Sudo1, A. Komori1 and LHD team1
• 1National Institutes for Fusion Science, Japan
• 2Kyushu University RIAM, Japan
• 3Osaka University, Japan
• 13th EU-US Transport Taskforce Workshop
• Copenhagen, Denmark
• 1-4 September 2008

2
Outline of talk

• Introduction
• Experimental results
• Nonlocal rise in core Ti
• Density dependence
• Nonlocal drop in core Ti
• Summary

3
Introduction

• Possible mechanism in nonlocal transport
  phenomenon

EDGE Perturbation
CORE Te rise/dropis observed so far

• Te gradient is enlarged
• Te/Ti decreases

m-turbulence changes
m-turbulence changes
Macro-/Meso-scale fluctuation structures

• If core m-turbulence changes, this may have an
  impact on the other core quantity, such as Ti,
  as well Te
• Here, a clear observation of the core Ti response
  to the edge perturbation and the relationship
  with the core Te are presented

4
LHD exp. demonstrates core Ti rise in response to
the edge perturbation

• Core Ti increased abruptly just after TESPEL
  injection (dTi 0.3 keV)
• Exp. condition
• Rax 3.55 m, Bt 2.789 T
• PnNB, PpNB, PECH 7.8 MW, 3.0 MW, 1.0 MW
• Te/Tigt1
• Low density, edge Te decreased due to the
  radiation and no edge Ti response? e-i is
  decoupled

5
LHD exp. demonstrates core Ti rise in response to
the edge perturbation (contd)
6
Density dependence of nonlocal behaviors
Nonlocal behavior in Te disappeared
Nonlocal behavior in Ti(though drop) still
appeared
Same mechanism but having different affector
exists?
7
Nonlocal Ti drop seems to be enhanced bythe core
Ti gradient

• In this density range, almost no change or slight
  drop in core Ti is expected
• ÑTi at r 0.55 (between 0.48 and 0.62) is
  greatly decreased

• Exp. condition
• Rax 3.575 m, Bt -2.769 T
• PnNB, PpNB 15.7 MW, 5.5 MW
• Te/Ti 1, but e-i is decoupled

8
Nonlocal behavior in core Ti indicates core Ti
stimulator exists inside r 0.6
9
Core stimulator for Te Ti exists in the almost
same region except for nonlocal Te drop
10
Our study is suggesting the importance of a
spatial correlation of the transport

• Cross-correlation analysis suggests that the long
  correlation, 0.7a exists in nonlocal Te rise of
  LHDS. Inagaki et al., To be submitted to PRL

• Spontaneous transition between two ITB stages
  (concave-convex) in JT-60U indicates that
  turbulence correlation length can be changed
  nonlocallyK. Ida et al., PRL 101 (2008) 055003

11
Summary

• Nonlocal behavior in core Ti is observed as well
  in core Te
• This phenomenon is a direct evidence of nonlocal
  coupling of heat transport
• Nonlocal behavior in core Ti shows the same
  feature as that in core Te
• Density dependence
• Existence of both rise and fall
• Region of improvement and degradation
• Nonlocal Ti drop seems to be enhanced by the core
  Ti gradient
• These results exemplify the importance of the
  long-range spatial correlation of the heat
  transport as well as the short-range

12
Thank you for your kind attention
13
Backup viewgraphs
14
CXS data shown here is valid for the following
reasons

• Ablation and ionization process will be finished
  for 5 ms
• 5 ms is equal to the time resolution of the used
  CXS
• Possibly affected data points are already omitted
• Poloidal CXS array outputs the same Ti values as
  toroidal CXS array does

15
Nonlocal Te drop is observed at the marginal heat
flux regime
Type I Low flux, No hysteresis Type II Marginal
region for flux, Nonlocal Te
drop Type III High flux, Nonlocal Te rise
16
Nonlocal Te rise seems to be enhanced bythe core
Te gradient
17
Core toroidal rotation velocity is also dropped
abruptly in response to the edge perturbation
Significantchange in the core
In almost the whole region (Flattening?)