EISCAT data analysis list messages
Date: 30 Aug 2001 11:07:24 -0700
From: Brett Isham
Subject: [1] EISCAT data analysis email list
Hello everyone,
Jean Lilensten and others have suggested creating an EISCAT data
analysis group made up of EISCAT users in each EISCAT country and at
EISCAT. This message is intended to be the start of that list. The
goal is to facilitate the exchange of information and ideas regarding
EISCAT data analysis.
Anita Aikio writes:
> I heard that a discussion group of EISCAT data analysis was going to
be set up. If such a group was established, we from Oulu would like to
join it.
I have been looking for someone in Finland who wanted to be on the list,
so I am glad to know that you are interested!
At this moment those who have (or have been) volunteered for the list
are as follows:
EI
Tony van Eyken
Ingemar Häggström
Brett Isham
FI
Anita Aikio
Ritva Kuula
Aoikari
FR
Beatrice Pibaret
Pierre Volcke
GE
Mike Rietveld
NI
Satanori Nozawa
NO
Patrick Guio
Kjelmar Oksavik
Anja Stromme
SE
Stephan Buchart
UK
Ivan Finch
Vikki Howells
Ian McCrea
Nigel Wade
Please send any corrections to me at . I'll be
happy to add or drop anyone to or from the list.
Ian McCrea writes:
> ... regarding an EISCAT GUISDAP group ... I presume that the
concept would be something like the "discuss" forum or the "hackers"
group which were set up by Tony. The way that Tony set these groups
up, whereby you could use a web interface to read all the email on
a particular thread, and have the whole chronology of the group's email
interactions on line, would give us a useful resource for tracing
developments in the program also. Another advantage is that the group
could potentially be as large as the number of people interested, since
no physical meetings would be involved (although that doesn't stop us
from getting together at convenient times like EISCAT workshops).
> Among other things, establishing a group like this would address
what I perceive to be a potential problem - namely that different
groups/countries are running their own versions of guisdap, and that
these may not be the same everywhere ! Before this divergence gets
too far, I'd suggest that we should charge EISCAT with being the keepers
and maintainers of the "official" GUISDAP and put pages on the EISCAT
web site where the update information and downloads of the "official"
code are available.
> Is this asking too much of the people at EISCAT ?
I believe the goal should be, as Ian suggests, to maintain a standard,
up-to-date, and distributable version of the EISCAT analysis software at
EISCAT. The EISCAT software is currently located on the data-taking
server puny (aka t45001) in /opt/guisdap/, with the latest analysis.m
file in subdirectory TonyTromso. I have been gathering EISCAT software
into the directory /opt/eiscat/ on puny and perhaps the analysis
software can be moved there as well in the near future.
At the moment this data analysis list is simply a list of email
addresses. Kjell Hauglann may be able to provide support for listserv
and/or web-based functions as Ian suggests. There is also a possibility
of help via the EISCAT Tromsø site's relationship with the University of
Tromsø. Finally, EISCAT is looking into the possibility of contracting
outside computer support, and that may be another source of help.
> Is EISCAT now officially the owner of GUISDAP, or is the agreement
of the Finns still required ?
I don't know the answer to this one.
> [Would] none of the original GUIZARDS/GUISPERTS ... be involved ?
It would be useful to have them in, if they had either the time or
the interest to make their expertise available.
I don't know that either...
Vikki Howells writes:
> Has anyone analysed any remote data for CP1l yet?
Ingemar Häggström is working on setting up the analysis for the remote
sites.
Anita Aikio asks:
> With the kind help of Jussi Markkanen we got the new Tromso CP1
(CP1L it was called when we used it in our campaign in February 2001)
data analysed. However, there are some strange features, e.g. the
analysis almost permanently fails around 150-225 km altitude with
alternating code gates. During days analysed, this is also the altitude
range, where electron densities are quite low, so the problem may have
to do with the alternating code itself.
Ingemar Häggström responds:
> Yes, I have seen some strange analysises in the lower F region, but
the data look OK, so I suspect one has to be somewhat more careful
choosing analysis strategy. Guisdap is also known to have more problems
than some other packages at low signal levels.
The UK users have also looked at a lot of cp1lt data and may be able to
comment on this.
I have been looking at the analysis for cp4bv and tau1sv (aka
tau1_singel_v). The cp4bv results look to me much better than the
tau1sv results, but both have a problem in the electron densities. The
latter may be because some of the analysis parameters have been changed
to improve the uhf analysis, but I'm not yet certain.
Welcome to the list!
Best regards, Brett
--
Brett Isham
address: EISCAT, Ramfjordmoen, N-9027 Ramfjordbotn, Norway
work phone: +47-7769-2140 or sometimes -2438 or -2171
cell phone: +47-9575-9932
email:
fax: +47-7769-2380
--
From: "Finch, ID (Ivan) "
Subject: [2] Guisdap2 Pre-release
Date: Thu, 6 Sep 2001 14:55:23 +0100
An initial release of Guisdap2 has been made available at
http://www.eiscat.rl.ac.uk/pub/guisdap/guisdap2-2.0.0.tar.gz
The license is a placeholder until a more suitable one can be agreed.
This is mostly a test of portability. I would be grateful if people
could
try and build it and tell me of any problems they have either
configuring, compiling or running it. More documentation will follow
later for those brave enough to try and use it! In the meantime there is
a sample scriptfile at the end of this message.
I will be on leave for the next week so questions may get slow answers.
Apologies.
Ivan Finch
Rutherford Appleton Laboratory
----------------------------------------------------------------
gup_path = "/home/eiscat/ivan/guisdap/";
apriori_file = "/home/eiscat/ivan/guisdap/matfiles/winter1996.mat";
exps_path = "/home/eiscat/ivan/guisdap/exps/";
data_path = "/stager/c/ivan/data/981216_gup3/";
result_path = "/stager/c/ivan/results/gup3_test";
name_site = EISCAT_ESR_32M;
name_expr = gup3c;
// Year Mon Day Hour Min Sec
start_time = [1998 12 16 10 58 00];
stop_time = [1998 12 16 11 00 00];
integrate = 120;
skip = -60;
altitude = [150:12:330 330:36:690 690:72:1000];
code = [12(32)];
gated_flag = [1 1 1 1 1 0]; // ne, ti, te/ti, coll, vi, comp
control = [.5 .01 100 1]; // data rejection condition, stop
criterion,
// number of iterations, variance flag
save_matlab = yes;
save_object = no;
Date: Thu, 6 Sep 2001 21:07:51 -0400
From: Brett Isham
Subject: [3] re: EISCAT data analysis email list
Hello everyone,
Additions to the list:
Takehiko Aso
Tom Grydeland
Jussi Markkanen
Yasunobu Ogawa
And two changes:
Antti Oikarinen
in place of
(is that correct and okay?)
and
Stephan Buchert
in place of
Stephan Buchert
Could someone please tell me the correct address for Jean Lilensten?
The list is now:
EI
Tony van Eyken
Ingemar H”ggstr–m
Brett Isham
Jussi Markkanen
FI
Anita Aikio
Ritva Kuula
Antti Oikarinen
FR
Beatrice Pibaret
Pierre Volcke
GE
Mike Rietveld
NI
Takehiko Aso
Satanori Nozawa
Yasunobu Ogawa
NO
Tom Grydeland
Patrick Guio
Kjelmar Oksavik
Anja Stromme
SE
Stephan Buchert
UK
Ivan Finch
Vikki Howells
Ian McCrea
Nigel Wade
As before, please send any corrections to me at .
I'll be happy to add or drop anyone to or from the list.
And once again, welcome to the list!
Best regards, Brett
Date: Tue, 2 Oct 2001 14:11:31 +0200 (CEST)
From: Ingemar Häggström
Subject: [4] cp1lr
Hi
The remote cp1l analysis setup is now more or less ready. I will put the
files onto the Tromso guisdap server. Sofar I have only looked at a few
spring data where the filtersettings on the remotes were wrong. To
analyse more recent data I will arrange another set. There are a few
files that needed correction, and some strategy changes, so one need to
reinitialise the old setups if one want to analyse them using the
changed routines. Also beware that one need a few actions to analyse
other geometries than cp1. I enclose a plot of data from the end of
april this year. The Tromso data were taken from the analysed data
directories at t45001 and is of the "aclp" type so I presume that both
long pulse and coded pulse were analysed together as the remotes. Also
is the Tx power not recorded at Sodankylä so it is fixed to 1MW. I have
not yet made any efforts to make the parameters agree better between the
remotes and Tromso (Magic_const=1), mainly due to the Tromso analysis
filter uncertainty.
/Ingemar
-----------------------------------------
Ingemar Häggström
EISCAT Scientific Association
Box 164, S-98123 Kiruna, Sweden.
Phone: +46-980-78701 Fax: +46-980-78709
-----------------------------------------
all_sites.eps
all_sites.pdf
Date: Tue, 2 Oct 2001 20:57:22 +0200
From: Brett Isham
Subject: [5] data analysis web page
Hi,
I have installed a data analysis web page at
http://www.eiscat.uit.no/data_analysis/index.html
In the left column, the first link gives a list of list members, the
second link is to past messages, and the third link contains a "mailto"
with e-mail addresses to all of the list members and so may be a
convenient
way to initiate messages to the list.
Comments are welcome.
Regards, Brett
Date: Mon, 3 Dec 2001 10:34:46 -0500
From: Brett Isham
Subject: [6] Fwd: GUISDAP 1.80 beta
Hi,
There is a link to the new guisdap 1.80 beta download page from
http://www.eiscat.uit.no/data_analysis/
See Ingemar's note (copy below) for more details on the guisdap 1.80
beta.
On that same page
http://www.eiscat.uit.no/data_analysis/
is also a link to guisdap 1.70. Guisdap 1.70 is an update to 1.65 and
is currently used in Tromso. All EROS3 CPs up to and including tau1
have been reinitialized, electron density calibrations have been
improved, and the user interface has been enhanced. Please see the
readme file for more information.
Regards, Brett
Date: Mon, 3 Dec 2001 13:26:24 +0100 (CET)
From: Ingemar Häggström
Subject: GUISDAP 1.80 beta
The guisdap-1.80beta is now available for testing.
I call it "1.80" due to several new features, see the notes.
And "beta" due to the little time I have had for confirming that
everything works, so I leave that part to you. All eros3 exps have been
working at some point, but things have changed during the route, so
maybe some of them don't anymore. Also, some things may not work under
matlab5.
You find it on the EISCAT web site, under Documentation.
/Ingemar