[Loops] EM loci

[Giulio Del Zanna]

Dear Jim et al.,

there seems to be a bit of confusion and some elephants in the room,
so I list a few clarifications below.

- I believe it was my 'fault' to re-introduce in 2003 the old EM loci method to
   study AR loops for the first time, and to show that *most* 'warm' loops
   are nearly isothermal at each location.
   I have to agree that, despite the limitations of the method, this is an useful
   SOHO/CDS result, that so far has been confirmed by Hinode/EIS (in general,
   iso-thermality appears to be the norm in AR structures, rather than
   multi-thermality).

- I did not and would not 'trust' too much any EM method, however forward
   modelling the line intensities in the loops (see Del Zanna & Mason 2003)
   gave consistent results (though still static).

- Of course, it has been known for a very long time that any dynamics in the
  loops will change the shape of any DEM/EM distributions, but by not too much I believe.
   I've shown in 2007 how clear patterns of dopplershifts are present in coronal
  loops from  Hinode/EIS, so we now have some extra observational constraints.

- One elehpant in the room is the problem that large (up to a factor of 10)
   discrepancies are present in any  EM modelling, when the lines from the
   'anomalous' ions are considered. This was obvious even in the Pottasch (1963)
  results but went un-noticed until 1971 and has since been considered by very
  few people (Phil Judge is one of them). There are a lot of erroneous statements
  in the literature by the way (interested readers should look at some of my
  papers).

- Even when considering lines from ions 'well behaved', the largest variations
   in results are coming from the use of different ion abundances.
   The latest ionization and recombination rates we have in CHIANTI are giving
   results for some ions that are quite different, for example.

- The EM loci plots were first introduced by K. Strong in 1978 I believe, and
   not by Pottasch who did something very different. The method is often not 
explained (I am afraid to say understood) properly.

- As any EM method, the EM loci will not give you an unique solution.

- To measure the electron temperature, there are better methods.
   In the last couple of years I showed that some diagnostics (athough very limited)
   for coronal loops are available from Fe VII, Fe VIII, Fe IX and Fe XI lines
   observed by Hinode/EIS. It is a start.
   When proposing for new instruments I always suggest that some temperature 
diagnostics  should be included..

- Another elphant in the room is the unresolved AR emission. It was ignored
  until the first loop meeting when I and Peter Young showed how dominant it is.
  I am glad to see that it is now a comonly-accepted issue.
  All EUV imagers are inherently multi-thermal, so it is more
  difficult to use them to measure the  unresolved AR emission, compared to a
  spectrometer.
  Ideally, we would need two spectrometers with high spatial resolution and
  very different viewing angles.



Giulio
--