Monitoring Temperatures in Antarctica Using

1
Monitoring Temperatures in Antarctica Using GPS
Radio OccultationsErick Adame Oneonta State
College, Oneonta, New York

• METHOD
• GPS RO involves capturing a bent signal from a
  GPS satellite by a (Low-Earth Orbit) LEO
  satellite as it passes through the atmosphere.
• GPS receivers measure phase and amplitude and
  deduce the bending angles as a function of height
  to obtain vertical profiles of refractivity.
• ROs were collected and averaged by six day
  periods to create color plots of altitude vs.
  time. Same plots were created for ECMWF and
  NCEP-Reanalysis.

• INTRODUCTION
• Polar regions contain very few traditional
  radiosonde measurements.
• Benefits to GPS RO in Climate
• GPS Radio Occultation (RO) will monitor current
  atmospheric conditions without the need for
  calibrating a variety of instruments measuring
  the same variables.
• GPS limb sounding technology used in missions
  will eliminate the concern of long-term
  consistency with climate monitoring.
• The uniform global distribution of GPS RO data
  will allow for more robust climate studies.

RESULTS
CHAMP
SONDES
ECMWF
NCEP-RA
2002
2003
2002
2003
2002
2003
2002
2003
Figure 2. GPS radio signal being refracted (bent)
by Earths atmosphere and bending angle that
is obtained.

• OBJECTIVES
• Examine the monthly mean temperatures over the
  Antarctic using the CHAMP GPS RO data.
• Analyze Seasonal and annual variations of
  tropopause temperature and altitude as well as
  the cooling/warming trends in the troposphere and
  stratosphere.
• Compare trends and variations with similar
  analyses derived from the radiosonde data and the
  ECMWF and NCEP global analyses.

• CONCLUSIONS
• Climate studies using RO give a more detailed
  description of the atmospheres vertical
  structure in time.
• Both models had less detail than GPS and
  Radiosonde data of the significantly cooler
  stratosphere during Antarcticas winter months.
• Continued monitoring of climate using GPS data
  will show trends with higher vertical resolution
  and with the most widespread coverage.

Figure 3. Plot of CHAMP data versus time in
2002-03
Figure 4. Plot of radiosonde data vs. time in
2002-03
Figure 5. Plot of ECMWF data vs. time in 2002-03
Figure 6. Plot of NCEP-RA data vs. time in 2002-03
Data points used were within 400 km and 2 hours
of occultation locations and time. Notice how the
number of data points are very few.
ECMWF data points used were at locations of
occultations and not grid points. Results were
similar to CHAMPs. There is less detail with
height.
NCEP-RA data points also were at locations of
occultations and not grid points. Results show
much less detail and less significant
stratospheric cooling.
CHAMP plot of average temperature versus time
shows a significantly varying stratosphere.
Spring deviations are large as shown in the
center plot.

• DATA
• GPS data (2001-2004) came from the
  single-satellite mission known as CHAMP launched
  by Germany in 2001.
• CHAMP GPS RO data was limited to south of 65º
  latitude (Figure 1) and was provided by UCARs
  COSMIC Data Analysis and Archive Center (CDAAC).
• Model comparisons were done with the ECMWF and
  the NCAR-Reanalysis.

• FUTURE WORK
• Use a database that allows the researcher to
  collect data points for models at its grid points
  independent of radio occultations.
• Use Radiosondes at the actual stations and
  include every radiosonde rather than only those
  that occur within a distance or time of a radio
  occultation.

• The graph to the left shows tropopause altitude
  versus time since May 2001.
• Tropopause altitudes are highest during summer
  months and lowest during winter months.
• These trends imply temperature trends of the
  troposphere. Different data types send mixed
  signals about which year is warmest/coldest.
• A sampling error presents itself as a noticeable
  difference between actual data from GPS RO and
  radiosondes and data from the models.
• This difference may exist because CDAAC is
  occultation based. Therefore, data is collected
  only when and where an occultation occurs. The
  true data would normally consist of the number
  of actual grid points.
• Despite the very few data points from
  radiosondes, its averages are much closer to
  CHAMPs values. This is likely because of
  vertical resolution differences. ECMWF and NCEP
  have less vertical resolution.

ACKNOWLEGEMENT This work was done under the
auspices of the Significant Opportunities in
Atmospheric Research and Science (SOARS) program
of the University Corporation for Atmospheric
Research, with funding from the National Science
Foundation, the U.S. Department of Energy, the
National Oceanic and Atmospheric Administration,
and the Goddard Space Flight Center, NASA.
Figure 7. Graph shows tropopause altitude versus
time since May 2001
Figure 1. Study is limited to include continent
inside 65º south latitude.