Kontakt

profile_image
Dr. Natalie Krivova
Telefon:+49 551 384 979-235

Beteiligte Wissenschaftler am MPS

Sanja Danilovic
Natalie Krivova
Manfred Schuessler
Sami K. Solanki
Michiel van Noort
Kok Leng Yeo

ROMIC/MUSIC

ROMIC/MUSIC - Rolle der mittleren Atmosphäre bezogen auf das Klima: Modellierung und Verständnis der Änderungen der solaren Bestrahlungsstärke

Übergeordnetes Ziel von ROMIC (Role Of the MIddle atmosphere in Climate), ein vom BMBF gefördertes Forschungsprogramm, ist die Untersuchung der Rolle der mittleren Atmosphäre für das Klima. Hierzu sollen wichtige Prozesse in der mittleren Atmosphäre in Hinblick auf ihre Bedeutung für das Klima untersucht werden. Einer der Untersuchungsschwerpunkte, dem das Teilprojekt MUSIC (Modeling and Understanding Solar Irradiance Changes) gewidmet ist, ist der Einfluss der solaren Variabilität.

Changes in solar radiation are the main external drivers of the Earth’s climate system. Whereas total solar irradiance (TSI, i.e. irradiance integrated over the entire spectrum) is the prime source of energy for the atmosphere, variations in the solar spectral irradiance (SSI), in particular in the ultraviolet part of the spectrum have a pronounced effect on the chemistry and dynamics of the Earth’s atmosphere. However, the strength of and the physical processes governing the solar influence on climate remain uncertain. This is due to:

  1. The uncertainties in the magnitude and details of the solar variability
  2. The  uncertainties in the mechanisms by which varying solar radiation influences climate.

Main aim of MUSIC, as part of the ROMIC project, is to deal with the first part of the problem. Its main product will be the reconstructed solar irradiance (both TSI and SSI) that will later be used by Earth atmosphere modelers, in particular by those involved in the ROMIC program, as an input to their models.

Current  status

It is generally agreed that the dominant part of the variations of TSI and SSI on time scales longer than a day are due to the appearance, disappearance and evolution of the magnetic field at the solar surface, namely its manifestations in the form of dark sunspots and bright faculae. The measured TSI has been best reproduced by models that make this assumption. We have developed the SATIRE (Spectral And Total Irradiance REconstructions) model that computes the radiance spectrum of magnetic features at different parts of the solar disc by solving the radiative transfer equations in semi-empirical model atmospheres describing sunspot umbrae, penumbrae, faculae and the quiet Sun. The version of SATIRE relevant to this project, SATIRE-S (i.e. SATIRE for the Satellite era) uses observations (spatially resolved full-disc solar magnetograms) to describe the evolution of magnetic structures on the solar surface.

<p>Fig1 - In the upper plot, PMOD composite of TSI measurements (daily data, light red; smoothed, thick red line) and SATIRE-S results (daily data, light blue; smoothed, thick blue) between 1978 and 2009 normalised to SORCE/TIM at December 2008 are shown. The thin blue lines mark the uncertainty range of SATIRE-S (only smoothed values plotted). In the lower plot, the difference between PMOD and SATIRE-S is shown (daily, grey; smoothed, black) along with the difference of the uncertainty with respect to PMOD. The black and red error bars are the errors from [12] in the upper and lower plots, respectively. Dotted vertical lines indicate cycle maxima and minima. Dashed horizontal lines signifying cycle minima are plotted to aid the reader. Gaps in the curves are present when data gaps are larger than 27 days. From Ball et al. (2012) [1].</p> Bild vergrößern

Fig1 - In the upper plot, PMOD composite of TSI measurements (daily data, light red; smoothed, thick red line) and SATIRE-S results (daily data, light blue; smoothed, thick blue) between 1978 and 2009 normalised to SORCE/TIM at December 2008 are shown. The thin blue lines mark the uncertainty range of SATIRE-S (only smoothed values plotted). In the lower plot, the difference between PMOD and SATIRE-S is shown (daily, grey; smoothed, black) along with the difference of the uncertainty with respect to PMOD. The black and red error bars are the errors from [12] in the upper and lower plots, respectively. Dotted vertical lines indicate cycle maxima and minima. Dashed horizontal lines signifying cycle minima are plotted to aid the reader. Gaps in the curves are present when data gaps are larger than 27 days. From Ball et al. (2012) [1].

[weniger]

The SATIRE-S model captures over 90% of the observed TSI variability [1] when compared with the PMOD composite of the TSI measurements obtained by  space-based radiometers since 1978 (Fig. 1). SATIRE-S has in the past also been successful in reproducing SSI variations both on short (days to months) and on long (years to decades) timescales  [2,3,4,5,6,7]. However, the situation has changed with the advent of data taken by instruments onboard SORCE (Solar Radiation and Climate Experiment) since 2003/2004. The new data agree with the models on short time scales (see left-hand panel of Fig. 2)[5,6], but display a strikingly different behavior on longer time scales (right-hand panel of Fig. 2)[6,7]. This is a problem faced not just by SATIRE, but by all models that can reproduce TSI variations (see Ermolli et al. 2013 [8] for a review and Fig. 2). Currently, it is not clear whether the problem lies with the models or with the data.

<p>Fig 2 -&nbsp; Normalised solar UV irradiance between 220 and 240 nm calculated with NRLSSI (black, [13]), SATIRE-S (blue, [3]) and COSI (magenta, [14]) models, and measured with UARS/SUSIM (darker green, [15]), UARS/SOLSTICE (light green, [16]), SORCE/SOLSTICE (orange, [17]) and SORCE/SIM(red; [18]). The pale green shading marks the period when the sensitivity of the UARS/SUSIM instrument (and thus the flux) changed, so that a shift was applied to the data before that [2,3]. Left-hand panel is limited to the period when SORCE was in operation, i.e. after 2003, and shows daily values, except for the COSI model, for which only yearly averages are available. Right-hand panel shows 3-month smoothed values over the period 1993-2009, for which UARS and/or SORCE data are available. From Ermolli et al. (2013) [8].</p> Bild vergrößern

Fig 2 -  Normalised solar UV irradiance between 220 and 240 nm calculated with NRLSSI (black, [13]), SATIRE-S (blue, [3]) and COSI (magenta, [14]) models, and measured with UARS/SUSIM (darker green, [15]), UARS/SOLSTICE (light green, [16]), SORCE/SOLSTICE (orange, [17]) and SORCE/SIM(red; [18]). The pale green shading marks the period when the sensitivity of the UARS/SUSIM instrument (and thus the flux) changed, so that a shift was applied to the data before that [2,3]. Left-hand panel is limited to the period when SORCE was in operation, i.e. after 2003, and shows daily values, except for the COSI model, for which only yearly averages are available. Right-hand panel shows 3-month smoothed values over the period 1993-2009, for which UARS and/or SORCE data are available. From Ermolli et al. (2013) [8].

[weniger]

Space for improvement

The main shortcoming of the present version of SATIRE is usage of a few 1D semi-empirical model atmospheres to describe the wide variety of features appearing at the solar surface. So far, a free parameter is used to compensate for the missing link between the semi-empirical model atmospheres and the amount of magnetic field measured in magnetograms. Obviously it would be more satisfactory for the model to work without a free parameter. This is only possible if the semi-empirical model atmospheres are replaced by physically self-consistent ones. Since the Sun shows a very complex surface structure (produced by the interplay of convection, oscillations, magnetic fields and radiation), such models must take into account the three-dimensional structuring of the atmosphere, as represented by state-of-the-art 3D radiation MHD simulations. Figure 3 shows the variety of solar structures reproduced by the MHD simulations [9,10] carried out with the MURaM, a radiative MHD code developed at the MPS [11].

<p>Fig 3 - Solar surface features reproduced by 3D MHD simulations carried out with MURaM code (V&ouml;gler et al. 2005) [11]. Left-hand side: bolometric intensity (top) and magnetic field strength on a vertical cut through the center of the sunspot (bottom). From Rempel (2012) [9].&nbsp; Right-hand side: Maps of brightness (bolometric intensity, upper panel) and vertical magnetic field at the optical solar surface (lower panel) from a snapshot of a MHD simulation in 15Mm deep box permeated by a horizontally averaged vertical field of 100 G. From Sch&uuml;ssler (2012) [10].</p> Bild vergrößern

Fig 3 - Solar surface features reproduced by 3D MHD simulations carried out with MURaM code (Vögler et al. 2005) [11]. Left-hand side: bolometric intensity (top) and magnetic field strength on a vertical cut through the center of the sunspot (bottom). From Rempel (2012) [9].  Right-hand side: Maps of brightness (bolometric intensity, upper panel) and vertical magnetic field at the optical solar surface (lower panel) from a snapshot of a MHD simulation in 15Mm deep box permeated by a horizontally averaged vertical field of 100 G. From Schüssler (2012) [10].

[weniger]

Our basic approach is to improve the self-consistency, physical realism and reliability of the irradiance models by extending them in two directions:

  1. Introduce significantly more physical realism into the modelling, in a first step, by employing state-of-the-art 3D radiation MHD simulations of the solar atmosphere, in particular MURaM simulations, to compute the spectrum of solar regions with different amounts of magnetic flux. The available simulations have local thermodynamic equilibrium (LTE) radiative transfer and allow the spectrum to be computed down to 300 nm.
  2. Further develop the MHD simulations. The most important extension is in height, so that the models also cover the solar chromosphere, where the cores of the stronger spectral lines in the UV spectrum (and some in the visible and infrared) are formed. In order to do so we need to take non-LTE into account in the radiative energy transport in the simulation. This extension of the model will significantly improve the critical UV fluxes below 300 nm resulting from the MHD simulations.

References:

[1] Ball, W. T., Unruh, Y. C., Krivova, N. A., et al. 2012, Astron. Astrophys., 541, A27
[2] Krivova, N. A., Solanki, S. K., & Floyd, L. 2006, Astron. Astrophys., 452, 631
[3] Krivova, N. A., Solanki, S. K., Wenzler, T., & Podlipnik, B. 2009, J. Geophys. Res., 114
[4] Krivova, N. A., Solanki, S. K., & Unruh, Y. C. 2011, J. Atm. Sol.-Terr. Phys., 73, 223
[5] Unruh, Y. C., Krivova, N. A., Solanki, S. K., et al. 2008, Astron. Astrophys., 486, 311
[6] Unruh, Y. C., Ball, W. T., $ Krivova, N. A. 2012, Surveys in Geophysics, 33, 475
[7] Ball, W. T., Unruh, Y. C., Krivova, N. A., et al. 2011, Astron. Astrophys., 530, A71
[8] Ermolli, I., Matthes, K., Dudok de Wit, T., et al. 2013, Atmos. Chem. Phys., 13, 3945
[9] Rempel, M. 2012, Astrophys. J., 750, 62
[10] Schüssler, M. 2013, Proc. Int. Astron. Union, 294, 95
[11] Vögler, A., Shelyag, S., Schüssler, M., et al. 2005, Astron. Astrophys., 429, 335
[12] Fröhlich, C. 2009, Astron. Astrophys., 501, L27
[13] Lean, J. 2000, GRL, 27, 2425
[14] Shapiro, A. I., Fluri, D. M., Berdyugina, S. V., et al. 2011, , Astron. Astrophys., 529, 14
[15] Floyd, L. E., Cook, J. W., Herring, L. C., & Crane, P. C. 2003, Adv. Space Res., 31, 2111
[16] Rottman, G. J., Woods, T. N., & Sparn, T. P. 1993, J. Geophys. Res., 98 (D6), 10667
[17] Snow, M., McClintock, W. E., Rottman, G., & Woods, T. N. 2005, Solar Phys., 230, 295
[18] Harder, J. W., Fontenla, J. M., Pilewskie, et al. 2009, Geophys. Res. Lett., 36

 
loading content