pb_br_mar06_icon_movie.gif March 29, 2006 Total Solar Eclipse:
Comparison Between the Prediction and Observations

A Comment About the Comparison

On this page we compare our prediction with several photographs of the eclipse.  We have oriented and sized our predicted images to match the format of the observed eclipse images.  The images of the magnetic field lines from the model illustrate how some of the (arbitrarily chosen) magnetic field lines are opened by the solar wind that streams along them, while some field lines remain closed, producing the dense structures called "helmet streamers" that show up as bright features in eclipse images.  In the polarized brightness images, the scattering of white light off of electrons in the corona is proportional to the coronal plasma density, so that the dense coronal streamers show up as bright features.

You will immediately notice that in all cases the predicted images from the model do not have as much high-resolution structure as the observations.  This is due to two reasons.  First, the simulations, though they have the highest spatial resolution we can practically achieve today, still do not have enough resolution to capture the details present in solar coronal structures.  In the future, improvements to the physics will allow us to begin to model these high-resolution features, and improvements in computer power will allow us to increase the resolution of the calculations.  Second, some of the observed images have been "sharpened" artificially using digital filters (as described in the links to the observed images) to enhance the appearance of fine structures in the corona.  For these reasons, we can at best hope to capture only the large-scale appearance of coronal structures.  Fortunately, for many applications, including space weather forecasting, this is sufficient.


Comparison Between the 3D MHD Model Prediction
and a Solar Eclipse Observation from
Greece

(a) MHD Model: Predicted
Magnetic Field Lines
(b) MHD Model: Predicted
Polarization Brightness
(c) Image from Greece:
Williams College Expedition*
MHD Field Lines
MHD Polarization Brightness
Observed Image
Click on the images to see higher resolution versions

(a) Magnetic field lines predicted from the MHD model.  (b) The polarization brightness predicted from the MHD model.  (c) A composite of an eclipse image (gray) taken in Kastellorizo, Greece, by the Williams College Eclipse Expedition, an image of the Sun’s surface in He II 304Å (inner orange-colored image), taken with the EIT instrument, and an  image from the LASCO C2 coronagraph (outer red-colored image); click here for details.  The EIT and LASCO instruments are on board the Solar and Heliospheric Observatory (SOHO) spacecraft.  These images are oriented so that solar north is vertically up.

*Photo credit: The eclipse photo was taken by the Williams College Eclipse Expedition (Jay Pasachoff, Bryce Babcock, Steven Souza, Jesse Levitt, Megan Bruck, Shelby Kimmel, Paul Hess, Anna Tsykalova, and Amy Steele), with support from NSF/NASA/National Geographic.

Comparison Between the 3D MHD Model Prediction
and a Solar Eclipse Observation from
Egypt

(a) MHD Model: Predicted
Magnetic Field Lines
(b) MHD Model: Predicted
Polarization Brightness
(c) Image from Egypt:
Jean Mouette**
MHD Field Lines
MHD Polarization Brightness
Observed Image
Click on the images to see higher resolution versions

(a) Magnetic field lines predicted from the MHD model.  (b) The polarization brightness predicted from the MHD model.  (c) An eclipse image taken in Sidi Barany, Egypt, by the team from Institut D'Astrophysique de Paris, France.  For details about the expedition, please click here.  These images are oriented with solar north 52 degrees clockwise from vertical.

**Photo credit: Courtesy of  Jean Mouette and Serge Koutchmy, CNRS (France).

Comparison Between the 3D MHD Model Prediction
and a Solar Eclipse Observation from
Egypt

(a) MHD Model: Predicted
Magnetic Field Lines
(b) MHD Model: Predicted
Polarization Brightness
(c) Image from Egypt:
Christian Viladrich***
MHD Field Lines
MHD Polarization Brightness
Observed Image
Click on the images to see higher resolution versions

(a) Magnetic field lines predicted from the MHD model.  (b) The polarization brightness predicted from the MHD model.  (c) An eclipse image taken in As Sallum, Egypt, by Christian Viladrich.  These images are oriented so that terrestrial north is vertically up (i.e., solar north is 26 degrees clockwise from vertical).

***Photo credit: Institut d'Astrophysique de Paris – CNRS & UPMC – Observation and processing by Ch. Viladrich.

Comparison Between the 3D MHD Model Prediction
and a Solar Eclipse Observation from
Libya

(a) MHD Model: Predicted
Magnetic Field Lines
(b) MHD Model: Predicted
Polarization Brightness
(c) Image from Libya:
Fred Espenak****
MHD Field Lines
MHD Polarization Brightness
Observed Image
Click on the images to see higher resolution versions

(a) Magnetic field lines predicted from the MHD model.  (b) The polarization brightness predicted from the MHD model.  (c) An eclipse image taken in Jalu, Libya, by Fred Espenak.  A composite image from 22 separate exposures was produced using Adobe Photoshop CS2.  This version has been enhanced to reveal subtle coronal structures using a narrow radial filter.  According to Fred Espenak, this image closely approximates the visual appearance of the corona during totality.  These images are oriented so that terrestrial north is vertically up (i.e., solar north is 26 degrees clockwise from vertical).

****Photo credit: ©2006 by Fred Espenak using Nikon D200 and Vixen 90mm f/9 Fluorite Refractor, 1 to 1/1000 second.

Comparison Between the 3D MHD Model Prediction
and GOES/SXI X-Ray Emission

(a) MHD Model: X-Ray Emission
on 03/03/2006 at 19:01UT
(b) MHD Model: X-Ray Emission
on 03/13/2006 at 00:29UT
(c) MHD Model: X-Ray Emission
on 03/29/2006 at 12:57UT
MHD Model X-Ray Emission
MHD Model X-Ray Emission
MHD Model X-Ray Emission
(d) GOES/SXI Emission
on 03/03/2006 at 19:01UT
(e) GOES/SXI Emission
on 03/13/2006 at 00:29UT
(f) GOES/SXI Emission
on 03/29/2006 at 12:57UT
SXI Image SXI Image
SXI Image
Click on the images to see higher resolution versions

Comparison of emitted X-ray radiation from the MHD model and the Solar X-ray Imager (SXI) on the GOES satellite.   (a–c) The upper images (colored red) show the emission computed from the coronal plasma temperature and density estimated from the MHD model.  A movie of the emission from the MHD model can be found here(d–f) The lower images (gray scale) show X-ray emission measured by SXI.

The exact nature of the heating in the solar corona is not know—it is presently a major unsolved solar physics problem.  We assumed an empirical heating in the MHD model that is related to the strength of the magnetic field.  Images of X-ray emission depend sensitively on the response of the measuring instrument.  Here we used the response of the Yohkoh X-ray imager, which is no longer operational, since we have not yet implemented the characteristics of the SXI imager.  Therefore, we cannot expect to match the SXI emission in detail.  Note that dark regions in these images correspond to areas with cool temperatures and lower density, and are known as "coronal holes".  Typically coronal holes are sources of fast solar wind.

There is reasonable correspondence between coronal holes in the model and in the observations.  The bright localized features (seen especially in images (a) and (d) on March 3) correspond to X-ray emission from the strong magnetic fields in an active region.  Since we used magnetic field data for this simulation that was measured from February 18 to March 17, 2006, we expect the active-region emission on March 3 ought to agree with observations, as it does.  However, by eclipse day on March 29 it is evident that the active region emission does not compare well with observations (compare images (c) and (f)), because the magnetic field on the Sun had changed significantly in the active regions.  Indeed, an examination of MDI magnetograms shows that the active region that is bright in image (c), which was observed on the solar disk in early March, had diffused significantly by late March, and, furthermore, a new active region complex had emerged and was visible on the solar disk in late March and early April (and particularly on eclipse day, March 29).  This new active region is on the East (left) solar limb of the SXI X-ray image (f).  This illustrates the fact that it is particularly difficult to predict the emergence of active regions on the Sun far in advance.  On the other hand, new developments in acoustic imaging of the interior of the Sun offer exciting future possibilities for active region imaging of the far side (i.e., the side we don't see) of the Sun.

iconReturn to the March 29, 2006 Eclipse Prediction Page

iconReturn to the Coronal Modeling Page

Last changed by Z. Mikic on 06/20/2006.