| Description: | DQR No:
Subject: Dust on the AERI-01 Scene Mirror
Date Submitted:
Submitted By: Robert Knuteson
___ Instrument Mentor
___ EST Member
_X_ Science Team Member
___ Other _____________________________
Organization: University of Wisconsin - Madison
Email address: robert.knuteson@ssec.wisc.edu
Telephone: 608-263-7974
Platform/Measurement:
What level data: (raw,a0,a1,b1,c1 etc):
All levels including the following platforms:
sgpaeri01ch1C1.a1
sgpaeri01ch2C1.a1
sgpaeri01summaryC1.a1
sgpaeri01engineerC1.a1
What location was the data collected at: ARM SGP CF
Period(s) of time in question
Period 1: Begin Date Sep/ 1/ 1995 Time 00:00 (GMT)
End Date Dec/ 22/ 1995 Time 00:00 (GMT)
Period 2: Begin Date Jan/ 18/ 1996 Time 00:00 (GMT)
End Date Apr/ 9/ 1996 Time 00:00 (GMT)
Data should be labeled:
_X_ questionable
___ incorrect
___ wrong calibration
___ others _______________________________
Discussion of Problem:
On April 25, 1995 the AERI-01 was installed in an enclosure protruding out
the side of the optical trailer at the ARM SGP central facility. This
configuration was a change from the AERI prototype setup which used
a roof hatch and a "chimney". The reason for the new configuration was
to place the AERI front end (scene mirrors and blackbodies) in an
environment that was much closer in temperature to the outside ambient air.
In that regard the change has been very successful. Unfortunately this
configuration also exposes the AERI scene mirror to contamination by airborne
dust.
At the time of installation only a rough opening to the sky existed with a
wooden hatch that was manually removed and replaced by site personnel during
normal working hours. Most of the time the hatch remained closed. When the
automatic hatch for the AERI-01 system in the ARM SGP CF optical trailer
was installed in July 1995, the percentage of time which the hatch was open
increased dramatically. Since that time there has been
a slow accumulation of dust on the AERI-01 scene mirror. No routine
maintenance (such as mirror cleaning) was defined since this dust
was thought to be only a thoughput problem but not a calibration issue.
However, analysis of the AERI/LBLRTM QME (comparison of observed radiances
to model calculations) indicated that there was a problem with the
AERI-01 observations during the October 1995 ARESE IOP. On Dec 20, 1995
personnel from UW-Madison visited the site and determined that the dust
on the scene mirror was acting as a scattering surface which effectively
broadened the conical field of view to the blackbodies and the sky. This
"broadening" of the field of view would not have lead to a problem except
for the fact that hatch opening is too small at it's current distance from
the instrument to allow the full beam to exit the enclosure. Thus a
fraction of the AERI field of view (the "tail" of the field) was viewing
the ambient temperature area surrounding the hatch opening. This caused
a mixture of ambient temperature emission into the scene, a fact that
is readily apparent when the scene was of a clear, dry sky (such as in the
ARESE IOP).
Period 1 indicated above covers the period from the first of September 1995
when this effect began to be noticeable to December 22, 1995 after the
scene mirror had been cleaned of dust contamination by UW-Madison personnel.
In general the scattering effect is increasing with time as more and more
dust accumulates on the scene mirror, but does show "events" linked to times
when the automatic hatch does not close soon enough and rain drops
fall on the scene mirror. These rain drops seem have the effect of both
making the scattering surfaces more effective (thus worse) and at other
times of actually partially cleaning the mirror if the rain is heavy enough.
Period 2, unfortunately only 4 weeks later, began with a different type
of "event" but leading to the same effect. On Jan 18, 1996 a major winter
storm caused snow to accumulate on the AERI-01 scene mirror (another
example when the automatic hatch did not work properly). When the snow
melted it appears to have left a substantial residue on the scene mirror
leading to an abrupt increase in the scattered energy contribution as
described above. This "dirty" mirror condition persisted until UW-Madison
personnel replaced the scene mirror on April 8, 1996.
Since April 1996, the AERI-01 system has been operating in the previous
manner; thus one can expect that the scene mirror has been accumulating dust
up to the present time. To address this recurring problem, UW-Madison is
developing a procedure that would allow the AERI-01 scene mirror to be
cleaned by site personnel on a regular basis. In addition, automatic
diagnostic information has been developed to identify this problem
immediately and provide feedback to site (and post analysis) personnel.
Furthermore, modifications of the AERI enclosure design are under
investigation that should build in an immunity to this effect at all
future AERI installations.
Other observations/measurements impacted by this problem:
This affects QMEs and other analyses that use the AERI-01 data that
require absolute accuracy better than about 5%.
Suggested Corrections of the Problem: (e.g. change calibration factor and
recompute, flag data with this comment, etc.)
It is likely that this data can and will be corrected. The best
approach to use in the correction is currently under investigation.
REQUIRED ACTIONS:
This report is informational no further action required_________
Experiment Center action required as follows:
Distribution of a DQR to users of AERI data is recommended for
the time periods described.
Site Operations action required as follows:
A schedule of cleaning for the central facility AERI scene mirror
is recommended. A procedure for this task should be obtained from
Fred Best of the University of Wisconsin - Madison combined
with a training session on proper techniques.
Archive action required as follows:
None
------------------------------------------------------- |