number of authors e.g. Mariska (1987), Cargill et al. (1995), Aschwanden &. Alexander (2001). Many of these models have over-simplified the flare cooling.
First Results from Hinode c 2008 ASP Conference Series, Vol. 397, Sarah A. Matthews, John M. Davis, and Louise K. Harra, eds.
Flare Cooling and Implications for Hinode/EIS C. L. Raftery,1,2 P. T. Gallagher,1 R. O. Milligan2 1 School
of Physics, Trinity College Dublin, Ireland.
2 NASA/Goddard
Space Flight Centre, Greenbelt, MD, USA.
Abstract. The cooling of a post-flare loop system as observed by TRACE and SOHO/CDS, SOHO/EIT, GOES and RHESSI is studied and compared to the predictions of recent solar flare models. The observed C-class flare cools from ≥ 10 MK to ∼ 0.25 MK in approximately 45 mins via conduction and radiation. Using theoretical modelling, conduction was found to dominate during the first 3 min of the decay phase, after which radiation became the dominant loss mechanism (∼ 30 min). We aim to study the flare cooling process using high resolution observations from Hinode/EIS.
1. Introduction It is widely accepted that hot flare loops result from the heating and evaporation of chromospheric plasma by nonthermal electrons (Neupert 1968). Once heated, it is believed that the flare loops cool by conduction followed by radiation (Culhane et al. 1970). These cooling processes have been studied by a number of authors e.g. Mariska (1987), Cargill et al. (1995), Aschwanden & Alexander (2001). Many of these models have over-simplified the flare cooling mechanisms by, for example, assuming hydrostatic equilibrium or independent conductive and radiative cooling phases. More recently, these models have been improved upon by assuming a multi-thread loop system or sequential thread heating (Bradshaw & Cargill 2005; Warren 2006). Here we compare cooling curves derived from multi-instrument observations of a C–class flare with the predictions of the Enthalpy Based Thermal Evolution of Loops (EBTEL) model (Klimchuk et al. 2008). This 0–D model considers both thermal and nonthermal heating of the flare and can be used to model the multi–strand structure of post–flare loops. The model simultaneously considers losses due to conduction, radiation and hydrodynamic expansion. 2. Data Analysis and Results ˚, The flare was studied using observations from SOHO/CDS, TRACE 171 A ˚ SOHO/EIT 195 A, GOES and RHESSI. The five emission lines observed by CDS ranged in formation temperature from ∼ 0.25-10 MK. Lightcurves of the loop apex were created for each of these lines and filters. It was found that the lightcurves peak sequentially, with RHESSI, the hottest, peaking first and CDS O V, the coolest, peaking ∼ 30 min later. Upflow velocities of ∼ 90km s−1 were found in the CDS Fe XIX emission line, which is consistent with explosive chromospheric evaporation (Fisher 184
Flare Cooling
185
Figure 1. (a) The conductive (dashed dot) and radiative (solid) losses calculated by EBTEL over the lifetime of the flare. (b) The temperature evolution calculated by EBTEL (solid line). The data points are obtained from the maximum of a spline fit to the lightcurves for a series of instruments.
et al. 1985; Milligan et al. 2006). A total flux of nonthermal electrons of ≥3×1010 ergs cm−2 s−1 was therefore assumed. The mean non-thermal electron energy was found to be ∼ 20 keV, using the RHESSI spectrum and the volumetric heating rate was calculated to be ∼ 400 ergs cm−3 s−1 . Assuming a semicircular loop, footpoint observations imply the loop half length is ∼4×109 cm. However, the best fit to the data requires a loop half length of ∼2 × 109 cm. A more detailed analysis will be presented in Raftery et al. (2008). 3. Conclusion and Discussion Observations suggest that the flare cools very quickly through the early decay phase, cooling by ∼ 5 MK in ∼ 8 min. Figure 1a shows that conduction is dominant for only a short time (∼ 3 min) during the decay phase before radiative losses become significant. The early dominance of radiation and the requirement for shorter loop length may be explained by an over-dense loop. This hypothesis is still under investigation and will be discussed in full in Raftery et al. (2008) High resolution observations with Hinode/EIS will reduce the errors in the lightcurve peaks and enable us to simultaneously study lines formed at chromospheric, transition region and coronal temperatures. Hinode/EIS will therefore enable us to make a direct comparisons between observations and the theory. References Aschwanden, M. J., & Alexander, D. 2001, Solar Phys. 204, 91 Bradshaw, S. J., & Cargill, P. J. 2005, A&A, 437, 311 Brown, J. C. 1971, Solar Phys. 18, 489 Cargill, P. J., Mariska, J. T., & Antiochos, S. K. 1995, ApJ, 439, 1034 Culhane, J. L., Vesecky, J. F., & Phillips, K. J. H. 1970, Solar Phys. 15, 394 Fisher, G. H., Canfield, R. C., & McClymont, A. N. 1985, ApJ, 289, 414 Klimchuck, J. A., Patsourakos, S. & Cargill, P. J. 2008, ApJ, submitted Mariska, J. T. 1987, ApJ, 319, 465
186
Raftery et al.
Milligan, R. O., Gallagher, P. T., Mathioudakis, M., Bloomfield, D. S., Keenan, F. P., & Schwartz, R. A. 2006, ApJ, 638, L117 Neupert, W. M. 1968, ApJ, 153, L59 Raftery, C. L., Gallagher, P. T., Milligan, R. O. & Klimchuk, J. A. 2008, A&A, submitted Warren, H. P. 2006, ApJ, 637, 522
