Study of error characteristics of GPS retrieved refractivity

1
Study of error characteristics of GPS retrieved
refractivity

• Bruce McDowell
• University of California, Davis
• Shu-hua Chen, Graduate Advisor
• Fourth Joint Korea-U.S. Workshop on Mesoscale
  Observation, Data Assimilation and Modeling for
  Severe Weather 13 - 15 February 2006

2
Introduction

• COSMIC will launch six low earth orbiting (LEO)
  satellites in March 2006.
• LEOs used in conjuction with GPS satellites will
  provide a source of weather observations that can
  be potentially used for global weather
  forecasting and long-term climate study.
• This simulation study is designed to investigate
  differences in average error characteristics of
  retrieved local refractivity
• Errors arising only from the assumption of
  spherical symmetry.
• Study area is centered on the lee side of the
  Rocky Mountains where large refractivity
  disturbances can occur.
• Eight cases, four in winter and four in summer,
  were used to study errors of retrieved local
  refractivity.
• In each season, two cases included storms and two
  had relatively calm weather.

3
Model Setup

• MM5 Model Code
• 3D Simulation
• Domain 3600 km x 4800 km
• ?x 20 km
• Vertical grid interval stretched
• Mix Phase Moisture Scheme
• Center of Domain on East Side of Rocky Mountains
• COSPAR International Reference Atmosphere (CIRA)
  outside of MM5 domain
• Ray-tracing Code (Zou et al. ,1999)

4
Computation of Refractivity Fields
Gridded atmospheric refractivity fields are
computed from the model gridded pressure,
temperature and water vapor pressure using the
relation
Rays that define an occultation were assumed to
be in the same plane (i.e., occultation plane)
with all tangent points vertically aligned every
300 m above the occultation point.
N Total Retractivity P pressure (hPa) T
temperature (K) e water vapor pressure (hPa) c1
77.6 K/hPa) c2 3.73 x 105 K2/hPa)
5
Experiment

• Eight cases
• Summer Four
• Winter Four
• Four storms
• Summer Two
• Winter Two
• MM5 model run was 24 hours to simulate a
  non-homogenous atmosphere.
• Integration of ray paths were performed through
  six occultation planes.

6
Experiment
Domain with terrain height. Occultation planes
for every 60 degrees shown at the occultation
point in the center of the domain at 38N 105W.
A schematic diagram for an ideal atmosphere
7
Results
January 31, 1998
July 7, 2004
MM5 24-h simulated total refractivity at 3 km for
2 days with storms Contour Interval 200 lt N lt 250
8
Results
January 31, 1998
July 7, 2004
MM5 24-h simulated total refractivity at 6 km for
2 days with storms Contour Interval 130 lt N lt 160
9
Results
MM5 24-h simulated vertical cross sections of
water vapor mixing ratio (g kg-1) 0-10 km from
west to east for (a) January 31, 1998 (storm) and
(b) February 15, 2004 (no storm).
10
Results
(b)
(a)
Errors (differences) of retrieved local
refractivity from that computed from ideal
atmospheres for six occultation planes for (a)
summer (storms) (b) summer (calm) (c) winter
(storms) (d) winter (calm)
(d)
(c)
11
Results
Averaged errors of the retrieved local
refractivity from six occultation planes during
- winter storms (black- solid line) - winter
calm periods (black-dashed line) - summer storms
(gray-solid line) - summer calm period
(gray-dashed line).
12
Summary

• Errors in retrieved refractivity can reach 10
  units or more in the lower troposphere under the
  assumption of spherical symmetry. The error
  decreases with height linearly until about 3 km,
  then fluctuates dramatically from 3 km to 7 km.
• Local maximum errors in locations of strong
  refractivity gradients can reach 5 units or more
  in the upper troposphere.
• Averaged retrieved error can underestimate true
  refractivity between 3 to 5 km in winter and 3 to
  7 km in summer. Maximum errors in summer
  occurred higher in the troposphere than in winter
  for both storm and calm conditions.
• Future work Investigate apparent differences in
  height of local maximum error between summer and
  winter.

13
Acknowledgements

• Shuhua Chen, Assistant Professor of Meteorology
  and Assistant Meteorologist, University of
  California at Davis
• Elcin Tan, Graduate Student, Atmospheric Science
  Program, University of California at Davis