Solar Cycle 24 Group H

20081208

Fletcher: chromosphere is important for flares; energy must come through the chromosphere, need focus on diagnostics & modelling.

20081209: Scott McIntosh

Phase travel time around spots show travel time varies with B, phi.

Bigger ring at 2 mHz around spot, smaller ring at 4 mHz closer to umbra.

Halpha spicules/dynamic fibrils show leaking p-modes around plage in inclined field.

x-t plots show parabolas: shocks are propelling mass up, verification through simulations.

• Scott: Are we cracking Halpha?
• Mats: No, still don't understand, need more time, study new observations.

Other option for wave propagation: radiative losses.

• Rebecca: We needed vertical propagation within resolution to explain observations, but not required in sunspot because the cutoff is lower.

Original idea by Roberts to explain Giovanelli.

Internal gravity waves from Bob Stein's simulations.

• Discussion: Simulations perhaps not suitable for this study.

IGWs are ubiquitous, see through k-f filtering, very energetic.

Alfvén waves are ubiquitous in the corona, but only 5-minute periods (COMP).

• Mats: Why would we see something in TRACE intensity? Non-compressive modes cause no I response, fast modes refract, acoustic modes dissipate.
• Bernhard: Why do we see shocks in the chromosphere, but not in SUMER?
• Discussion: Shocks happen below the canopy, SUMER doesn't sample there.

Questions:

• What mechanisms allow 5-minute waves to propagate into the chromopshere?
• Why is the corona so selective?
• What is the impact of IGWs?
• What's going on with the Alfvén waves?
• Can we isolate individual modes?

What do we need? Spatially resolved spectrometry.

• Full disk at moderate resolution: B, V, tomography
• High-resolution: B, V, tomography

Discussion:

• Mats: Worry about timescales when doing chromospheric seismology; good for checking models but maybe not for detailed study.
• Mats: Milimeter observations at high resolution is missing here, e.g., ALMA. Good thing: linear in T.

Key objectives: energy and mass transport in the chromosphere.

20081209: Rebecca Centeno

Study wave propagation using He 10830.

Slit on plage, includes Si line in the photosphere.

5-minute p-modes in the photosphere.

He shows 3-minute oscillations in sunspot-like structures in the chromosphere, amplitude ~ B.

He shows 5-minute oscillations in plage.

In sunspots: Delta phi < 2 mHz no signal, < 4 mHz evanescent, < ... propagation

Use simple model: isothermal T, constant B // g, radiative losses.

Play with model parameters until it matches the observations.

In plage: tau_r = 10 s (short!), T = 9500 K (high!), Delta z = 1500 km

Discussion on field inclination: these results show theta = 0.

No resolution on correlation with SP that shows theta = 0 happens rarely.

• Bernhard: 10830 can be measured in the QS.

20081209: Alfred de Wijn

Sorry, too boring, I fell asleep.

20081209: Lyndsay Fletcher

Question: can we diagnose flare inputs?

Optical radiation produced in flares is compact and intense.

We would really like to have an estimate of energy input.

How well can we estimate the total energy input?

• Crank up RMHD simulations.
• How much can we do with existing diagnostics?
• Is it OK to use a static model?

Energy transport:

• Particles are accelerated by B energy release.
• Waves are launched by reconnection.

What happens when a strong Alfven pulse hits the chromosphere?

• Damping by ion-neutral coupling?
• Generation of turbulent cascade? Electron acceleration?
• Drive the formation of current sheets?

20081210: Phil Judge

Basic question: why does the sun produce the chromosphere etc. as it is?

All stars with surface convection have chromospheres.

Observationally driven research.

Some stars emit more in the corona than in the chromosphere, so cannot ignore the corona.

Spicules span 9 scale heights, spicules arise from the chromosphere, aren't part of it.

Heating: steady current systems are not dominant.

Chromosphere is partially ionized plasma, leads to frictional dissipation through ion-neutral coupling.

Dissipation of j_perp tries to make field force free in the chromopshere.

Chromosphere is bright in network, so high p where high B (VAL/FAL).

But: magnetostatic models require low p where high B.

Twist naturally increases with height because of conservation of flux and current.

Claim has been made that the chromosphere and corona are not connected.

• See images of both, look very different.

Cool loop explanation, but where does 10⁶ erg/cm²/s conductive flux go?

TR radiates 10⁶ erg/cm²/s, coincidence?

VAULT results show Ly alpha is from the base of hot coronal loops.

Neutral diffusion as a way to explain Ly alpha emission?

Critical question: what are the chromospheric conditions at the base of the corona?

Need B measurement in the chromosphere.

Way forward: integral field spectroscopy to get B in the chromosphere.

Discussion: interpretation of measurements (e.g., 10830) is difficult.

20081210: Thomas Straus

4 kW/m² needed to heat the chromosphere.

• Phil: too small by a factor ~3
• Mats: only when maintaining a static structure
• Bernhard: do we know energy losses as a function of height?
• Phil (off-topic): Why does the chromosphere look the way it does, why do we have supergranules?
• Bernhard: is this 4 kW/m² important, if so, can we agree on a number?
• Phil: drive question of energy losses through simulations.
• Thomas: compare Scott's talk, chromosphere is an observational science.

<10% acoustic waves (Fossum & Carlsson 2005)

90% "magnetic"

Energy transport in 1D: acoustic waves above cut-off frequency.

Energy transport in 3D: acoustic waves above cut-off frequency + IGWs.

Study IGWs in observations and simulations.

Energy flux: rho <v²> v_gr for observations, <p v_z> for simulations.

In simulations: lots of energy in IGWs, some in f-mode, little in acoustic.

Acoustic flux is insufficient.

Above 300 km all flux is in IGWs.

Target flux is achieved.

• Acoustic flux <10% may be too low, 50% maybe?
• Acoustic portals (see Jefferies): 30%
• IGWs: 100%
• "magnetic": 90%
• Grand total: 270%, oh dear.

Issues:

• Everybody works in the mid-upper chromosphere, can we study at ~700 km?
• Is Doppler velocity a good diagnostic?
• Why is Mg b at 100 mA formed at 700 km?

What is the science case for the future? Open games:

• observation vs theory
• 1D vs 3D
• extracting parameters from models vs simulating observations
• waves vs features

20081210: Discussion

• Mats: RMHD can be sped up, sometimes it is slow because of convergence problems.
• Thomas: An important part of the heating problem: how is energy released?
• Thomas: Reduced RMS of IGWs where there is B: mode coupling?
• Thomas: We have to investigate features we think we understand from simulations in simple atmospheres in more realistic atmospheres.
• Mats: Conclusions from TRACE intensity analysis: acoustic flux is below 20mHz, and not at small scales (cf. Hinode).
• Bernhard: Simple experiment, propagate waves, fold with 160-km gaussian contribution function, result shows > 2 order of magnitude lost at 30 mHz.
• Bernhard: Using real contribution function gives similar result, so much more energy is carried, 1600 W/m² between 5-31 mHz.
• Bottom line: high-frequency acoustic waves are not dead.
• Topic left out: what is the value of using waves a diagnostic tool?

20081211: Mats Carlsson

Boundaries for simulations:

• E comes from convection, so models must include the top of the convection zone
• B topology should be contained in the box, so upper boundary must be in the corona and box must be large enough
• E may leave the box

Energy balance:

• strong and weak lines, non-local NLTE radiation
• shocks
• conduction
• field dissipation
• particle beams

Other complications: chromosphere is mostly neutral, may require multi-fluid approach.

Impossible to solve the full problem from a practical point of view, so make approximations and solve a suitable sub-problem.

RT post-processing OK in the photosphere, but not so good in the chromosphere because the simulation is not realistic there.

Discussion: simulated "observations" may be confusing because of differences in representation with real observations.

Discussion: current status of heating problems: corona is ok, not enough in the chromosphere.

20081211: Tony Arber

What's the minimum physics we need to make a realistic chromosphere?

Include neutrals results in "almost MHD".

Disregard ion-neutral slip in the chromosphere.

E + v x B = eta j turns into E + v x B = eta j_par + eta_cowling j_perp

In the chromosphere eta_cowling dominates.

Results from including neutrals:

• corona is more force free
• chromosphere is heated
• less chromospheric uplift

Cowling conductivity cannot be written as a diffusion equation.

Chromosphere cannot support equilibrium unless force-free.

Neutrals intensify current concentrations, so more reconnection.

20081211: Hiroaki Isobe

Poynting flux of THMF ~ 2*10⁶ erg/cm²/s is approx. required to heat the chromosphere.

Conclusion from simulations: need multi-fluid with neutrals.

Chromospheric reconnection? Weakly ionized and fully collisional, is fast reconnection possible?

Ion-neutral collisions increase the reconnection rate, see Tony's talk.

20081211: Bill Abbett

Linking convection to the corona requires a chromosphere.

At large scales it is not possible to treat the physics properly in the chromosphere.

So what is the minimum requirement to get the connection from the photosphere to the corona right?

In this work, <j . B> is lowest in the photosphere where we measure B, i.e., possibly the worst place to measure!

20081211: Bart De Pontieu

Type II spicules explanation.

Look for upflow events (Hara) in EIS data: no correlation with intensity, occur at footpoints of loops.

Correlation between type II spicules and upflows, not great but not expected because of viewing angle issues.

So do type II spicules fill the corona with mass?

Coronal evaporation linked to type II spicules.