Good question. Easy to get lost in the scattering data. Some clarifications:
SingleScatteringData: The scattering data of one "scattering element".
Normally the data for one particle size.
ArrayOfSingleScatteringData: The scattering data of one "scattering
species". Normally the data for one habit.
ArrayOfArrayOfSingleScatteringData: The scattering data of multiple
scattering species. These are the data actually used in a calculation.
The standard habit files contain ArrayOfSingleScatteringData, while
scat_data_raw is of type ArrayOfArrayOfSingleScatteringData. That is,
you can not read a standard habit directly into scat_data_raw. But this
ReadXML( tmp, "EvansSnowAggregate.xml" )
Append( scat_data_raw, tmp )
I can mention that I prepare the scattering data outside of ARTS and
normally adjust the data in some way. If you don't need the highest
accuracy, you can for example skip every second size to speed up the
calculations. Anyhow, I then fix the step from
ArrayOfSingleScatteringData to ArrayOfArrayOfSingleScatteringData on the
But maybe we should make an ARTS function simplifying to set up
calculations involving a single habit ...
On 2021-12-20 17:25, daniele.case...@artov.isac.cnr.it wrote:
First I want to thank you for keeping this project open source it is
I am trying to use the Standard Habits in the StandardHabits.tar.gz from
the database for total random orientation already converted in ARTS format.
I am using arts 2.5 and pyarts.
When I try to load the scat_data_raw with both:
I get this error message:
XML parse error: Attribute type has value "SingleScatteringData" but
"ArrayOfSingleScatteringData" was expected.
Check syntax of XML file
This happens not only with EvansSnowAggregate but also with the other
habits that I tryied.
The same code works well when I load the Mie Liquid and Ice spheres
scattering properties in "arts-xml-data-2.4.0/scattering". What am I
doing wrong? I assumed that SingleScatteringData contains data of single
scattering for a specific habit and size, while the
ArrayOfSingleScatteringData should contain many sizes as is it in the
arts_users.mi mailing list