2 types of spicules "observed" in 3D realistic models
aa r X i v : . [ a s t r o - ph . S R ] J a n **FULL TITLE**ASP Conference Series, Vol. **VOLUME**, **YEAR OF PUBLICATION****NAMES OF EDITORS** Juan Mart´ınez-Sykora
Institute of Theoretical Astrophysics, University of Oslo, NorwayLockheed Martin Solar & Astrophysics Lab, Palo Alto, USA
Abstract.
Realistic numerical 3D models of the outer solar atmosphere show twodifferent kind of spicule-like phenomena, as also observed on the solar limb. Thenumerical models are calculated using the
Oslo Staggered Code (OSC) to solvethe full MHD equations with non-grey and NLTE radiative transfer and thermalconduction along the magnetic field lines. The two types of spicules arise as anatural result of the dynamical evolution in the models. We discuss the differentproperties of these two types of spicules, their differences from observed spiculesand what needs to be improved in the models.
1. Numerical Methods and description of the model
The nature of spicules observed at the solar limb have long been a mystery,in this paper we discuss two types of jets that occur naturally in 3D num-berical models of the solar atmosphere. The MHD equations are solved in amodel spanning the upper convection and corona using the
Oslo Stagger Code (OSC). In addition, this code solves a rather realistic NLTE radiative transfer,including scattering, and thermal conduction along the field lines as explainedin Mart´ınez-Sykora et al. (2008).The models described below have a grid size of 256 × ×
160 pointsspanning 16 × ×
16 Mm . The grid is uniform in the horizontal directionwith a grid spacing of 65 km. In the vertical direction the grid is non-uniform,ensuring that the vertical resolution is good enough to resolve the photosphereand transition region with a grid spacing of 32 . .
2. Results and discussions
In the models we found two types of spicule-like structures, i.e the so-calledtype i (Mart´ınez-Sykora et al. 2009) and type ii (McIntosh et al. 2007).A synthetic image of Ca ii at the limb is shown in fig 1 which shows thetwo types of spicules. These structures look rather similar to what is observedat the limb in the Sun in Ca ii . 1 Figure 1. Synthetic image of Ca ii H from the limb of the model. Observethe two types of spicules, type i located at x = 14 Mm and type ii at x =7 Mm. The synthetic image is done with MULTI 3D. Table 1 shows the differences between the two types of the spicules in ourmodels and compared to observations. The reader is referred to work that hasrecently been completed related to spicules; Rouppe van der Voort et al. (2007);De Pontieu et al. (2007); Hansteen et al. (2006); Mart´ınez-Sykora et al. (2009).
Type i Type ii Observations
150 examples in bothmodels 2 examples only in B1model Type ii ubiquitousLength ≈ [0 . , .
5] Mm Length ≈ i are longerDuration ≈ [2 ,
5] min Duration ≈ i have longer dura-tionsParabolic profile in time(deceleration) Complex velocity profilesdue to acceleration at dif-ferent height Seems to agree (see bibli-ography)Up-downflow profile Only upflow Seems to agreeVelocities ≈ [5 ,
35] km/s Velocities ≈
150 km/s Type i reach larger veloc-itiesObserved in Ca ii Counterpart in Transi-tion region emission lines Seems to agreeDriven by magneto-acustic shocks Reconnection Similar drivers suggested
Table 1. Properties of the two types of spicules “observed” in the modelsand compared with observations.
Most likely, the differences between the two types seem to agree with theobservations. However, a deeper study needs to be done with the type ii spiculesfound in the model (work in progress). Moreover, a closer comparison with theobservations is required. It is interesting to note that the appearance of type i does not show a clear preference between models with or without flux emergence,while type ii only appear in the model with the largest ambient ambient field(B1), and only after emerging flux cross the photosphere. Spicules of both typesin the models are located at the footpoints of the atmospheric coronal loops,where the field lines are open field lines or at least penetrate into the corona.Moreover, the footpoint that is closer to the emerging flux tube is the one thatshows most jets. The type ii spicules shows a corresponding nearby hot loop Figure 2. Histograms, normalized to the total number of spicules, for de-celerations, maximal velocity, maximum lengths, and duration, from left toright, respectively, measured from the two models (B1 dash and A2 dash-dotline) and the sum (continuum line). The vertical line is the median value ofthe distribution. The two models show different distributions of the decelera-tion, maximum lengths, and duration, as well as some differences in maximumvelocity. The model B1 shows on average slightly lower decelerations, shorterlength, and shorter velocities than A2. which also seems observed in the Sun (De Pontieu et al. 2009). The hot loop( > K) can be observed with coronal emission lines.In brief, we can summarize the differences between observations and modelsfor type i spicules by noting that the upper limits of the deceleration, length, du-ration, and maximal velocity are smaller in the models (Mart´ınez-Sykora et al.2009). Histograms for deceleration, maximum length, maximal velocity and du-ration for the type i from the model are shown in fig 2. These can be comparedwith the histograms from the observations done by De Pontieu et al. (2007).They show an agreement in the lower part to the histograms, and the differ-ences between the B1 and A2 seems similar to the differences of the two regionsobserved by De Pontieu et al. (2007). However, the models do not fit with theupper part of the observed histograms.In order to improve the models we consider that the resolution of the boxis important. The chromosphere is poorly resolved numerically and this affectsthe size of the structures of the spicules. In addition, low resolution mightbring other effects like the diffusion of the shocks (type i ) or of the magneticdiscontinuity (type ii ). With higher resolution we expect sharper shocks, largerrange of velocities, better resolved and more frequent spicules.In models, it is also important to take time-dependent hydrogen ionizationinto account in the the upper chromosphere. The ionization of hydrogen in thesolar chromosphere and transition region does not obey LTE, or instantaneousstatistical equilibrium, as the timescales of ionization and recombination arelong compared with HD timescales, especially for magneto-acoustic shocks. Theshock temperatures are higher, and the intershock temperatures are lower, inmodels where time-dependent ionization is considered. This effect will likelychange the range of parameter of the spicules type i (Leenaarts et al. 2007).Modeling the chromosphere is strongly important to study properly the radiativelosses approximations, NLTE with scattering as has discussed byCarlsson (2010);Leenaarts (2010). Figure 3. Temperature, ion fraction, magnetic field intensity, and ohm, halland ambipolar diffusion calculated as post-processing in a 2D cut of the model,from left to right and top to bottom. Observe that Ohm and Hall diffusionare rather important in the lower chromosphere and the ambipolar diffusionin the upper chromosphere.
The partial ionization might have other effects on both types of spicules,as well. When considering partial ionization we find that ambipolar diffusion,Hall diffusion and ohmic diffusion contribute at differing rates throughout thechromosphere. The ratio between these three diffusion terms changes from thephotosphere up to the transition region (see fig 3). This will possibly haveimportant effects in the chromosphere as the parameters controlling reconnectionand the damping of waves change.Finally, we note that the range of ambient magnetic field structures thathave been modeled only form a small subset of those expected when consid-ering supergranulation, plage and the chromospheric network. In addition, acontinuous weak magnetic flux emergence may need to be added, since it hasbeen observed in the models that chromosphere and transition region heightsare considerably increased with flux emergence.
Acknowledgments:
Inestimable collaboration and contribution with ViggoHansteen, Bart de Pontieu, Mats Carlsson and Fernando Moreno-Insertis.