Downloading SDO data

Here I give my recommendations for downloading SDO and HMI data. In most cases users will want to download cutout data rather than full disk data, so I discuss this first.

Downloading cutouts

From the JSOC (recommended)

Please check out Procedure for obtaining SDO cutouts for an EIS data-set for how to use the JSOC to get cutouts. The method applies generally as long as you have a pointing and field-of-view size.

From Lockheed

LMSAL have a web-based interface for accessing cutouts, but I prefer to use the IDL interface to the cutout service. A typical call is:

IDL> ssw_cutout_service,'8-feb-2011 23:39','9-feb-2011 00:19',cgiquery,queryout,xcen=-350,ycen=-760,ref_time='9-feb-2011 00:06',fovx=120,fovy=140,wave='193',email=[your e-mail address],max_frames=41

FOVX and FOVX give the field-of-view in arcsec (not pixels). The e-mail address is required so that you receive an e-mail when the query has been processed. Instead of XCEN and YCEN, REF_HELIO can be specified which gives in the coordinates in longitude and latitude. For example, REF_HELIO='S33E29'. I think XCEN and YCEN get converted to longitude and latitude and rounded to 1 degree, as the cutouts are not precisely centered on the input coordinates.

If you want multiple wavelengths, make sure to increase MAX_FRAMES proportionally to the number of wavelengths.

When the query has been processed, the e-mail links to a webpage which gives an IDL command that is used to download the cutouts.

Downloading full disk images

VSO

A typical call is:

IDL> list=vso_search('2010-06-28 00:00','2010-06-28 00:10',inst='aia',wave=171,pixels=4096)

where WAVE= is for specifying the wavelength of the AIA channel (94, 131, 171, 193, 211, 304, 335), and PIXELS= forces the routine to only find full resolution data. If you want the AIA synoptic data (1024x1024 images), then use pixels=1024. To take only images every 720 seconds (for example), use sample=720.

To download the data, do:

IDL> chck=vso_get(list,/rice)

Note that the /rice keyword downloads the data in rice-compressed format. This routine seems to work out where your computer is and then choose which file server is best to use. If you have problems with the default server, then try specifying SITE=. Options include 'nso', 'sdac', 'sao', 'nmsu' and various European sites.

JSOC

My method for getting HMI data from JSOC is go to straight to the exportdata page and in the 'record set' box type something like:

hmi.M_45s[2010.07.25_13:00/5m@45s]

which will look for 45s magnetograms from 25-Jul-2010 from 13:00 for a 5min period.

You can substitute 'M_45s' for:

• 'hmi.M_720s' (12min magnetograms)
• 'hmi.Ic_45s' (45s continuum images)
• 'hmi.Ic_720s' (12min continuum).
• 'aia.lev1_euv_12s'
• 'aia.lev1_uv_24s'
  

Where to store data

The SDO team do not give any recommendation on how to store the SDO data, unfortunately. I set the environment variable $SDO_TOP to point to the top level data directory (e.g., /data/sdo), and then I create sub-directories such as:

/data/sdo/aia/level1/2012/08/05

where all AIA data for 5-Aug-2012 are stored. Thus when I use the vso_get() routine I will put the data in here.

The cutout data are generally much smaller in size and just store these in the working directory of the project that I'm working on.

What processing level do the data have?

AIA data downloaded from the JSOC and VSO are in level-1 format and so should be processed with aia_prep to give a uniform format for all filters.

The LMSAL cutout service data are in level-1.5 format, and so do not need to be processed with aia_prep.

Page maintained by Dr Peter R Young.