global scale observations of ionospheric instabilities from - COSMIC
Global Assimilative Ionospheric Modeling and COSMIC-2 · Global Assimilative Ionospheric Modeling...
Transcript of Global Assimilative Ionospheric Modeling and COSMIC-2 · Global Assimilative Ionospheric Modeling...
Global Assimilative Ionospheric Modeling and
COSMIC-2
Clayton Coker
Naval Research Laboratory, Washington, DC, [email protected]
Xiaoqing Pi, Atilla Komjathy and Anthony Mannucci
Jet Propulsion Laboratory, Pasadena, CA
2
With the launch of the COSMIC constellation of GPS radio occultation sensors in
2006, research began on assimilating data from multiple GPS radio occultations into
global assimilative ionospheric models. These space-based observations provide the
advantage of global coverage compared to ground-based GPS stations but suffer
from low horizontal resolution. Plans are on the table to deploy even larger
constellations of GPS radio occultation sensors. Will these larger constellations
with double or quadruple the number of observations of the space environment drive
the assimilative models to the next level of performance in resolution and
accuracy? Early in the COSMIC mission multiple spacecraft were in the same orbit,
providing a high density of radio occultation data near the local time of that orbit – a
condition analogous to proposed COSMIC-2 constellations with multiple sensors
in each orbit. The resulting high density radio occultation data are ingested into
global assimilative models, and the performance of the models are validated using
high resolution UV photometer data from COSMIC. The Tiny Ionospheric
Photometer (TIP) on COSMIC is a compact, nadir directed, narrow-band, ultraviolet
photometer operating at the 135.6 nm wavelength and measuring the horizontal
structure of the ionosphere with 15-30 km resolution. Comparisons with TIP data
reveals the horizontal performance of these models and demonstrates the
improvement obtained from ingesting the high density radio occultation data.
Abstract
Early in the COSMIC mission, three SVs in the same orbit plane.
Provides higher density coverage near the orbit (or local time).
Provides a glimpse into future data densities provided by COSMIC-2.
COSMIC-2 with 12 SVs will provide a factor of 2 to 4 more RO samples.
What is the impact of COSMIC-2 on assimilative ionospheric models?
Rayleighs
Radio Occultations from three COSMIC SVs in 2100LT orbit
Constellation with 3x Data Density of COSMIC
4
Climatology Background Model for 2100LT
Rayleighs
Evaluation of Impact of Future Missions
What is the impact of COSMIC-2 data densities on
assimilative ionospheric models?
Assimilation of high density RO data and all
available ground GPS data into GAIMUSC – focus
on specification near the orbit.
Validation using Tiny Ionospheric Photometer
Rayleighs
Radio Occultations from three COSMIC SVs in 2100LT orbit
Tiny Ionosphere Photometer Data
GAIM–TIP RMS 1.2 R Climatology–TIP RMS 2.5 R
GAIM GAIM Ionospheric Specification for 2100LT on 14 Sep 2006
Rayleighs
Ground GPS
5
TIP
GAIM USC Climate
Rayleighs
GAIM USC Ground GPS
Rayleighs
GAIM USC Ground + RO
Rayleighs
GAIMUSC Climate–TIP RMS 2.5 R
GAIMUSC Ground–TIP RMS 1.4 R
GAIMUSC Ground+RO–TIP RMS 1.2 R
Impact of Multiple RO Sensors in Each Orbit to
Assimilative Modeling
Ground GPS (all available stations) provides a
reasonable specification of global ionosphere –
some areas fall back to climatology.
Adding RO data improves performance.
Improves global coverage.
Better definition of anomaly trough
Artifacts due to ambiguity in RO data in the
presence of ionospheric gradients (equatorial
ionospheric anomaly).
45%
15%
TIP is an ultraviolet photometer measuring 135.6 nm emissions from recombination of oxygen ions and electrons
– intensity is integral of density-squared along the line of sight, similar to TEC.
TIP looks downward from the COSMIC spacecraft and maps the equatorial ionization anomaly with high
resolution along the path of the satellites.
Longitudinal variability is driven by zonal winds influenced by tidal modes.
Variability in anomaly crest intensity is 1-20 Rayleighs (5-25 TECU) in longitude.
Variability in crest separation ranges from 30° to 12° with no anomaly at some longitudes.
TIP Observations of Low Latitude Ionosphere
Magnetic Latitude
Ra
yle
igh
s
14 Sep 2006 21LT
10 TECU
15 TECU
20 TECU
5 TECU
7
Climatology overestimates 21LT ionosphere near solar minimum.
Adding Ground GPS data drives output closer to truth, but not all longitudes
covered.
Adding space-based RO data provides global coverage, drives output towards
correct global average.
Climatology
Ground GPS
Ground+RO
TIP
Magnetic Latitude
135.6
-nm
Rad
iance
- R
ayle
ighs
10 TECU
15 TECU
20 TECU
5 TECU
Comparison of GAIMUSC Runs with TIP Data
Global Average Ionosphere @ 21LT
8
Climatology does not improve ionospheric correction error at some latitudes – trough
and transitions between mid and low latitudes.
Adding Ground GPS data drives the error down at low latitudes, but trough and
transitions need improvement.
Adding space-based RO data drives the error down at trough and southern transition,
but does not improve results at anomaly crests – need better gradient info.
RMS Error of GAIMUSC Runs
Global RMS Error @ 21LT
Climatology – TIP Ground GPS – TIP
Ground+RO – TIP
TIP
Magnetic Latitude
RM
S E
rror
– R
ayle
ighs
TIP 135.6-nm
Radiances
10 TECU
15 TECU
20 TECU
5 TECU
Climatology
Ground GPS
Ground+RO
TIP data RO data
Magnetic Latitude
135.6
-nm
Rad
iance
- R
ayle
ighs
10 TECU
15 TECU
20 TECU
5 TECU
S American Sector
Comparison of GAIMUSC Runs with TIP Data
Example where ground
GPS is abundant.
45-deg
Climatology
Ground GPS
Ground + RO
Magnetic Latitude
Alt
itude
45-deg
Density
1e5/cm3
8
5
4
1
0
Climatology
Ground GPS
Ground+RO
TIP data RO data
Magnetic Latitude
135.6
-nm
Rad
iance
- R
ayle
ighs
10 TECU
15 TECU
20 TECU
5 TECU
Hawaiian Sector
Comparison of GAIMUSC Runs with TIP Data
Example where ground
GPS is sparse and
anomaly is symmetric.
45-deg
Climatology
Ground GPS
Ground + RO
Magnetic Latitude
Alt
itude
45-deg
Density
1e5/cm3
8
5
4
1
0
Climatology
Ground GPS
Ground+RO
TIP data RO data
Magnetic Latitude
135.6
-nm
Rad
iance
- R
ayle
ighs
10 TECU
15 TECU
20 TECU
5 TECU
Kwajalein Sector
Comparison of GAIMUSC Runs with TIP Data
Example where ground GPS is sparse in the
Southern Hemisphere and nonexistent in
Northern Hemisphere.
45-deg
Climatology
Ground GPS
Ground + RO
Magnetic Latitude
Alt
itude
45-deg
Density
1e5/cm3
8
5
4
1
0
Indian Sector
Comparison of GAIMUSC Runs with TIP Data
Climatology
Ground GPS
Ground+RO
TIP data RO data
Magnetic Latitude
135.6
-nm
Rad
iance
- R
ayle
ighs
10 TECU
15 TECU
20 TECU
5 TECU
Example where ground GPS
coverage is reasonable but anomaly
is collapsed – low densities.
45-deg
Climatology
Ground GPS
Ground + RO
Magnetic Latitude
Alt
itude
45-deg
Density
1e5/cm3
8
5
4
1
0
Climatology
Ground GPS
Ground+RO
TIP data RO data
Magnetic Latitude
135.6
-nm
Rad
iance
- R
ayle
ighs
10 TECU
15 TECU
20 TECU
5 TECU
Atlantic Sector
Comparison of GAIMUSC Runs with TIP Data
Example where ground GPS is
sparse and anomaly is
collapsed – low densities.
45-deg
Climatology
Ground GPS
Ground + RO
Magnetic Latitude
Alt
itude
45-deg
Density
1e5/cm3
8
5
4
1
0
Conclusions
• Early COSMIC rollout period provides coverage similar to proposed COSMIC-2.
– Triple the number of radio occultations near the orbit plane.
• When assimilating data similar to COSMIC-2, the ionospheric RMS error was
reduced by 15% compared with the model run using ground GPS data only.
• Improvement to assimilative ionospheric modeling is clearly observed in the trough
of the equatorial anomaly.
• No improvement is observed at the crests of the anomaly.
– When horizontal constraints are limited (sparse ground GPS stations), the model often miss
positions and/or broadens the anomaly crest.
• COSMIC-2 does not provide sufficient density of data samples to produce high
resolution specification of the post-sunset ionospheric anomaly.
– Need secondary sensor on COSMIC-2 to specify the true ionospheric horizontal gradients.
• Recommend a UV photometer as a secondary payload on COSMIC-2 for high
resolution ionospheric specification and identification of irregularities which
produce scintillation.