1. C.-H. Lyu (NASA GSFC), T. Mo (NOAA STAR), and I. Osaretin
NPP ATMS Prelaunch Performance Assessment and Sensor Data Record Validation
W. J. Blackwell, L. Chidester (SDL), C. Cull, E. J. Kim (NASA GSFC), R. V. Leslie,
C.-H. Lyu (NASA GSFC), T. Mo (NOAA STAR), and I. Osaretin
IGARSS
July 25, 2011
This presentation will describe recent work to characterize and model the ATMS system performance, including calibration and product retrieval. A major effort has focused on the development and validation of ATMS “proxy data” that is essentially transformed AMSU-A/B observations. Today I’ll talk about the derivation of the model-based transform, present sample data, and discuss future plans for validation. I will also discuss other components of the modeling, including spatial reprocessing and EDR validation.
This work was sponsored by the National Oceanographic and Atmospheric Administration under Air Force Contract FA C Opinions, interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the United States Government. 1 1

2. Outline ATMS Overview Prelaunch testing
Background and development
Performance
Prelaunch testing
Antenna
Radiometric TVAC
Post-launch validation of sensor/product performance
Planning/schedule/objectives
Tasks
Summary
I will begin with an overview of the ATMS sensor and move onto to validation plans once the NPP satellite launches. The talk concludes with a discussion of applications of the ATMS data.

3. NPOESS Preparatory Project
Launch site: Vandenberg AFB
Launch vehicle: Boeing Delta II
Spacecraft: Ball Aerospace Commercial Platform 2000
Instruments: VIIRS, CrIS, ATMS, OMPS, & CERES
Orbits: 824 km (NPP); sun-synchronous with a 1:30 p.m. local-time ascending node crossing
This is an artist’s renditions of the NPP spacecraft, which will be the first platform to fly the Cross-track Infrared and Microwave Sounding Suite (CrIMSS). CrIMSS is responsible for the AVTP and AVMP Key Performance Parameters and consists of the Cross-track Infrared Sounder (CrIS) and the Advanced Technology Microwave Sounder (ATMS).
NPP (NPOESS Preparatory Project)
Oct. 25, 2011 Launch
National Polar-orbiting Operational Environmental Satellite System → Joint Polar Satellite System

4. NPOESS Preparatory Project October 25, 2011 Launch
Shown is a picture of the NPP satellite with the five sensors installed.
Photo courtesy of Ball Aerospace, Boulder, CO.

5. Advanced Technology Microwave Sounder (ATMS)
ATMS is a 22 channel MW sounder
Frequencies range from GHz
Total-power, two-point external calibration
Continuous cross-track scanning, with torque & momentum compensation
Orbits: 833 km (JPSS); 824 km (NPP); sun-synchronous
Thermal control by spacecraft cold plate
Contractor: Northrop Grumman Electronics Systems (NGES)
An overview of the ATMS sensor.
70 cm

6. ATMS Development ATMS NPP unit (“F1") developed by NASA/Goddard
ATMS NPP unit delivered in 2005
ATMS JPSS-1 unit (“F2”) currently in development (antenna and TVAC testing in 2011/12, delivery in 2012)
Principal challenges/advantages:
Reduced size/power relative to AMSU
Scan drive mechanism
MMIC technology
Improved spatial coverage (no gaps between swaths)
Nyquist spatial sampling of temperature bands (improved information content relative to AMSU-A)
History of the ATMS development.

7. ATMS Design Challenge Reduce the volume by 3x AMSU-A1 73x30x61 cm 67 W
54 kg
3-yr life
AMSU-A2
75x70x64 cm
24 W
50 kg
3-yr life
The principal challenge of the ATMS development was the consolidation of three sensors (AMSU-A, AMSU-B, and MHS) into a single sensor with a longer lifetime requirement.
MHS
75x56x69 cm
61 W
50 kg
4-yr life
70x40x60 cm
110 W
85 kg
8 year life
Figure courtesy NGES, Azusa, CA

8. Atmospheric Transmission at Microwave Wavelengths
ATMS channels
The frequency dependence of atmospheric absorption allows different altitudes to be sensed by spacing channels along absorption lines
The frequency dependence of atmospheric absorption allows different altitudes to be sensed by spacing channels along absorption lines

9. Spectral Differences: ATMS vs. AMSU/MHSCh
GHz
Pol 1
23.8 QV 2
31.399
31.4 3
50.299
50.3 QH 4
51.76
52.8 5
± 0.115 6
± 0.115
54.4 7
54.94 8
55.5 9
fo = 57.29 10
fo ± 0.217 11
fo±0.3222±0.217
fo±0.3222±0.048 12
fo± ±0.048
fo ±0.3222±0.022 13
fo±0.3222±0.022
fo± ±0.010 14
fo± ±0.010
fo±0.3222±0.0045 15
fo± ±0.0045
89.0 16
88.2 17
157.0
165.5 18
± 1
± 7 19
± 3
± 4.5 20
191.31 21
± 1.8 22
ATMS has 22 channels and AMSU/MHS have 20, with polarization differences between some channels
− QV = Quasi-vertical; polarization vector is parallel to the scan plane at nadir
− QH = Quasi-horizontal; polarization vector is perpendicular to the scan plane at nadir
AMSU-A
An overview of the principal differences between ATMS and AMSU.
Exact match to AMSU/MHS
Only Polarization different
Unique Passband
MHS
Unique Passband, and Pol. different
from closest AMSU/MHS channels

10. Spatial Differences: ATMS vs. AMSU/MHS
Beamwidth (degrees)
Spatial sampling
ATMS
AMSU/MHS
23/31 GHz
5.2
3.3
50-60 GHz
2.2
89-GHz
1.1
GHz
ATMS
AMSU/MHS
23/31 GHz
1.11
3.33
50-60 GHz
89-GHz
GHz
Swath (km)
~2600
~2200
Spatial attributes of ATMS and AMSU.
ATMS scan period: 8/3 sec; AMSU-A scan period: 8 sec
ATMS measures 96 footprints per scan (30/90 for AMSU-A/B)

11. Summary of Key Sensor Parameters
PFM Measurement
Envelope dimensions
70x60x40 cm
Mass
75 kg
Operational average power
100 W
Operational peak power
200 W
Data rate
30 kbps
Absolute calibration accuracy
0.6 K
Maximum nonlinearity
0.35 K
Frequency stability
0.5 MHz
Pointing knowledge
0.03 degrees
NEDT
0.3/0.5/1.0/2.0 K

12. ATMS Data Products Data Product Description RDR (Raw Data Record)
FOV1 antenna temperature (counts)
TDR (Temperature Data Record)
FOV1 antenna temperature (K)
SDR (Sensor Data Record)
FOR1 brightness temperature (K)
EDR (Environmental Data Record)
P/T/WV profile
“CDR” (Climate Data Record)
“Climate-optimized” product
IP (Intermediate Product)
Used to generate EDR/CDR
These data products will be derived from ATMS observations.
1FOV = ATMS “Field of View”; FOR = CrIMSS “Field of Regard”
RDR, TDR, SDR, and EDR products will be available via CLASS

13. Outline ATMS Overview Prelaunch testing
Background and development
Performance
Prelaunch testing
Antenna
Radiometric TVAC
Post-launch validation of sensor/product performance
Planning/schedule/objectives
Tasks
Summary
I will begin with an overview of the ATMS sensor and move onto to validation plans once the NPP satellite launches. The talk concludes with a discussion of applications of the ATMS data.

14. ATMS Pre-Launch Testing
Essential for two objectives:
Ensure sensor meets performance specifications
Ensure calibration parameters that are needed for SDR processing are adequately and accurately defined
PFM testing revealed several issues that will require calibration corrections in the SDR:
Non-linearity (temperature-dependent for 31.4-GHz channel)
Cross-polarization (sometimes 10X higher than AMSU)
Antenna beam spillover from secondary parabolic reflector approaching 2% for some channels
A variety of prelaunch testing is performed to assess performance and reliability
EMI/EMC - anechoic chamber for both emitted and susceptible measurements
Mechanical – mass, CG, vibration, & torque on various fixtures
Radiometric – thermal vacuum (TVac) testing
Antenna – Compact Antenna Test Range (CATR)
The ATMS PFM has completed its characterization and satellite-level testing is mainly to confirm performance after five years. The pre-launch characterization included antenna pattern characterization using a Compact Antenna Test Range (CATR), susceptibility to radio frequency interference, and calibration accuracy and RMS analysis in a thermal vacuum chamber using precision external calibration targets. Listed in the second bullet were some results of the characterization. The sensor had larger than expected non-linearity and required a waiver, but it generally exceeds AMSU. A temperature-dependent NL correction factor was derived from Thermal-Vacuum data. The EDR performance is still being evaluated, but we’re hopeful they can be met with some on-orbit calibration that I’ll discuss more later in the presentation.

15. Compact Antenna Range Testing
Compact Antenna Test Range
RF source illuminates the Antenna Under Test (AUT), i.e., ATMS antenna subsystem
Uses a parabolic reflector to collimate the electromagnetic radiation to illuminate the AUT in the far-field region
AUT is attached to a positioner to rotate the AUT into the proper orientation
Test measures the power received by the AUT compared to a standard antenna with a known antenna gain pattern
Specifications verified:
Beam pointing accuracy
Beamwidth
Beam efficiency
Earth intercept
Are we measuring gain pattern or a normalized antenna pattern?

16. Thermal Vacuum Parameters
Dynamic range (brightness temperature range of K)
NEΔT (sensitivity)
Calibration accuracy (bias)
Linearity (deviation from linear regression)
Short term gain fluctuation (ΔG/G) (gain stability or 1/f noise)
Hysteresis (repeatability)
NASA/LARC
Measure the calibration parameters that are needed by the SDR algorithm (e.g., non-linearity correction factor)
Validate that the sensor meets performance requirements
Provide pre-launch performance validation in a flight-like environment

17. Outline ATMS Overview Prelaunch testing
Background and development
Performance
Prelaunch testing
Antenna
Radiometric TVAC
Post-launch validation of sensor/product performance
Planning/schedule/objectives
Tasks
Summary
I will begin with an overview of the ATMS sensor and move onto to validation plans once the NPP satellite launches. The talk concludes with a discussion of applications of the ATMS data.

18. Post-Launch Cal/Val Timeline
IPO/NGAS ATMS
NASA/NGES
NGES/NASA

19. ATMS Post-Launch Calibration/Validation in a Nutshell
Post-launch is Cal/Val has four phases: Activation, Checkout, Intensive Cal/Val, and Long-term Trending
Tasks within the phases can be categorized:
Sensor Evaluation: interference, performance evaluation, etc.
TDR/SDR Verification: geolocation, accuracy, etc.
SDR Algorithm Tunable Parameters: bias correction, space view sector, etc.
Activation Phase: Sensor is turned on and a sensor functional evaluation is performed; ATMS is collecting science data
Checkout Phase: Performance evaluation and RFI evaluations
Intensive Cal/Val: Verification of SDR attributes such as geolocation, resampling, brightness temperature accuracy (Simultaneous Nadir Overpass, Double Difference, radiosondes/NWP simulations, aircraft verification campaigns), and satellite maneuvers
Post-launch cal/val has an extensive history and is becoming increasingly sophisticated. The ATMS SDR validation plan includes Simultaneous Nadir Overpass; Double Difference; radiance comparisons using NWP, radiosondes, and CrIS radiances; aircraft campaigns; satellite maneuvers, etc. Another tool will be an end-to-end ATMS/CrIMSS system error model/budget that will model the system from RDRs to EDRs and will use information derived from sensor thermal-vacuum data, radiative transfer, heritage sensors, and eventually ATMS data. The rest of this presentations will discuss the ATMS proxy data that uses AMSU/MHS, aircraft validation, and on-orbit spacecraft maneuvers.

20. ATMS On-Orbit FOV Characterization
Spacecraft maneuvers (constant pitch up or roll, for example) could be used to sweep antenna beam across vicarious calibration sources
Moon (probably too weak/broad for pattern assessment)
Earth’s limb (requires atmospheric characterization)
Focus of today’s presentation
Land/sea boundary (good for verification of geolocation)
With knowledge of the atmospheric state, the antenna pattern can be recovered with deconvolution techniques
Objectives of this study - quantitatively assess:
The benefits of various maneuvers
How accurately can the pattern be recovered?
The limitations of this approach
How much roll/pitch is needed for an adequate measurement?
The error sources and their impact
In order to further characterize the ATMS FOV, three ATMS satellite maneuvers are proposed for NPP. They rely on the discontinuity of the Earth’s limb to deconvolve antenna pattern anomalies such as sidelobes. The three maneuvers consist of a pitch over, sun-side roll, and anti-sunside roll. NOAA-14 has gone through similar maneuvers and we hope to use that data to validate this method.

21. TB’s Across Earth/Space Transition
LIMB
(STANDARD
ATMOSPHERE)
SURFACE
BEYOND STANDARD ATMOSPHERE
62.17°

22. NPOESS Airborne Sounder Testbed
OBJECTIVES
Satellite calibration/validation
Simulate spaceborne instruments
(i.e. CrIS, ATMS, IASI)
Preview high resolution products
Evaluate key EDR algorithms
INSTRUMENTS: NAST-I & NAST-M
NAST- I: IR Interferometer Sounder
NAST- M: Microwave Sounder
5 Bands: 23/31 (to be added), 54, 118, 183, 425 GHz
NAST
~100km
Before I talk about NAST-M, let me first describe NAST, which stands for the NPOESS Airborne Sounder Testbed. Over the past few years, (Remote Sensing and Estimation Group) MIT and MITLL research groups have outfitted a variety of planes with two instrument suites: NAST-I, and NAST-M. NAST-I is an IR interferometer intended for use with AIRS, and NAST-M is a microwave sounder. There are several uses for these aircraft-based sensors. First is the underflight of satellites, for the purpose of calibration and validation. NAST can also be used to simulate present and future spaceborne instruments, such as CRIS, ATMS, and IASI, which can be helpful in the evaluation of key EDR (Environmental Data Record) processing algorithms. It also can help to preview high resolution spatial and spectral products.
For the campaigns I will discuss, NAST-I and NAST-M were mounted inside left wing pod. As the aircraft flies, the sensors sweep horizontal to the direction of motion +-65 degrees. Generally, NAST flies at cruising altitudes of 17-20km –very high compared to commercial airlines, which fly at roughly 9km. On average, it sweeps out a swath roughly 100km wide (depending on aircraft altitude).
In this talk I will focus on the microwave sounder, which I will discuss in more detail next. Here you can see what the sensor suite looks like on the inside of the wing pod, and then here are the horns for the different microwave bands.
RDR = Raw Data Record
NAST- M
118
183 54
425
Cruising altitude: ~17-20 km
Cross-track scanning: - 65º to 65º

23. MetOp Satellite Validation
JAIVEx
Tb Comparison AMSU-A
April 20th, 2007 collection
GHz Bias
50.3
-0.25K
52.8
1.4K
53.75
-0.2K
Gulf of Mexico
54.4
-0.2K
AMSU-A (MetOp) [Kelvin]
<20min
54.94
-0.9K
On the left is the April 20th flight over the Gulf of Mexico, and the down-sampled MetOp AMSU-A footprints we used in our comparison. Notice that we had great temporal coincidence for this collection (within 20 minutes).
MetOp
path
NAST-M [Kelvin]

24. Summary All key radiometric requirements were satisfied
Radiometric accuracy exceeds 1K
Radiometric sensitivity exceeds requirements
Similar to AMSU for similar effective footprint sizes
Linearity performance generally exceeds AMSU
Antenna pattern testing indicates good performance
Opportunity for spacecraft maneuvers allows improved characterization of ATMS spatial response function
Post-launch cal/val plans are well defined
Readiness exercises ongoing
Areas to watch have been illuminated during prelaunch testing
At a very abstract level, here are the objectives and primary components of a successful cal/val campaign. I will touch on most of these during the presentation.

25. Backup Slides

26. Utility of Aircraft Underflights
What do aircraft measurements provide that we cannot get anywhere else?
Why not just compare to radiosondes or NWP?
Direct radiance comparisons
Removes modeling errors
Mobile platform
High spatial & temporal coincidence achievable
Spectral response matched to satellite
With additional radiometers for calibration
Higher spatial resolution than satellite
Additional instrumentation deployed
Coincident video data
Dropsondes
Example video image
Solar
glint
Here are a list of reasons (rational for using an aircraft sensor):
Direct Radiance-to-radiance measurements
- RAOBs sample temp, WV, and humidity; must use radiative transfer models to generate radiances, which can lead to errors.
2. Mobile platform, so co-locatable to satellite of choice (high spatial and temporal coincidence achievable)
- RAOBs have limited spatial and temporal coincidence since generally radiosondes are released only twice daily from airports around country.
3. Spectral response of NAST-M matched to AMSU, with additional radiometers for calibration
4. NAST-M also provides higher spatial resolution than satellites; I will show an example of that in a few slides.
5. Additional instruments are deployed along with radiometers, which provide “extra” data types to help you analyze data sets later – for example: Coincident video data of cloudiness; Clouds can affect radiances for 54GHz bands, so it is helpful to know if they were present, and to what degree.
Ocean
Clouds

27. Synergistic Use of MW+IR: Infrared Provides High Spatial Resolution
For example, CrIS provides 15km horizontal and 1km vertical resolution
The infrared and microwave observations are used synergistically to observe the atmosphere in the presence of clouds.

28. For example, ATMS provides 33km horizontal and 3km vertical resolution
Passive Microwave Measurements Provide Low Spatial Resolution, but Penetrate Clouds
For example, ATMS provides 33km horizontal and 3km vertical resolution
The infrared and microwave observations are used synergistically to observe the atmosphere in the presence of clouds.

29. AMSU-A: Large, Positive Forecast Impact
ATMS observations will be used to improve forecast accuracy. The microwave instruments have the largest positive impact on forecast accuracy.
Source: Gelaro, Rienecker et al., 2008, NASA GMAO

30. ATMS Storm Mapping: Improvements Relative to AMSU
Recent work is shown where ATMS observations are used to estimate the intensity of precipitation.
Source: Surussavadee and Staelin, NASA PMM Presentation, 7/08