Assessment of CrIMSS Version Change (Mx5.3 à Mx6.3) on October 15th, 2012
Methodology
Satellite Data
Valida&on of Temperature Profile Environmental Data Records (EDRs) from the Cross-‐Track Infrared Microwave Sounding Suite (CrIMSS) Using COSMIC Dry Temperature Profiles
Michelle Feltz, Robert Knuteson, Henry Revercomb, Dave Tobin, and Steve Ackerman CooperaPve InsPtute for Meteorological Satellite Studies, University of Wisconsin-‐Madison
Abstract This paper presents a methodology for validaPng the measurements from the Cross-‐Track Infrared Microwave Sounding Suite (CrIMSS) using combined observaPons from the Advanced Technology Microwave Sounder (ATMS) and the hyperspectral infrared Cross-‐track Infrared Sounder (CrIS) on the Suomi NPP satellite, the first satellite of the newly created U.S. JPSS program. The atmospheric verPcal temperature profiles (AVTPs) from the CrIMSS operaPonal product are compared to temperature profiles obtained from radio occultaPon (RO) from the COSMIC GPS RO network. In parPcular, bias, RMS, and standard deviaPon profile staPsPcs will be presented for global and 30 degree laPtude zones for selected Pme periods in 2012. Similar validaPon staPsPcs using AIRS and COSMIC profile matchups will be created for the same space and Pme periods. The matchup methodology was used to evaluate the dependence of the ray path averaging on the staPsPcal results by comparing the closest single AIRS profile, to the average along a ray path, and to the average of a circle of radius equal to the horizontal ray path length through the atmosphere. The opPmal method of comparison between GPS RO and IR soundings from the Suomi NPP satellite will be presented. A comparison will be made of CrIMSS EDR performance relaPve to NASA AIRS L2 product in order to assess the CrIMSS product accuracy. This evaluaPon is in support of the NOAA NESDIS calibraPon/validaPon acPviPes for the checkout of the Suomi NPP satellite.
Acknowledgments
Contact: [email protected] -‐ UW-‐Madison AOS/CIMSS/SSEC
295
We gratefully acknowledge the NPSO (Taiwan’s NaPonal Space OrganizaPon) and UCAR (University Center for Atmospheric Research) for access to the COSMIC RO data. We would like to thank the Goddard Space Flight Center (GSFC) Data Archive for access to the AIRS Level 2 data products. This work was supported under JPSS NOAA grant NA10NES4400013.
Case Study: November 4th 2012, Madison, WI
References
Anthes, R. A. et al., 2008: The COSMIC/FORMOSAT-‐3 mission: early results. Bull. Amer. Meteor. Soc., 89, 313–333. doi: hep://dx.doi.org/10.1175/BAMS-‐89-‐3-‐313. Borbas, E.,W. P. Menzel, J. Li, H. M. Woolf, 2003: Effects of GPS/RO refracPviPes on IR/MW retrievals, ITSC-‐13, Sainte Adele, Canada, 29 October -‐ 4 November 2003. Healy, S.B. and Eyre, J.R. 2000: Retrieving temperature, water vapour and surface pressure informaPon from refracPvity-‐index profiles derived by radio occultaPon: A simulaPon study. Q. J. R. Meteorol. Soc., 126, 1661-‐1683. Kursinski, E.R., Hajj ,G.A., Schofield, J.T., Linfields, R.P. and Hardy, K.R. 1997: Observing Earth’s atmosphere with radio occultaPon measurement using the Global PosiPoning System. J. Geophys. Res., 102, 23,429-‐23,465. Tobin, D. C. et al., 2006: Atmospheric RadiaPon Measurement site atmospheric state best esPmates for Atmospheric Infrared Sounder temperature and water vapor retrieval validaPon. J. Geophys. Res., 111, D09S14, doi:10.1029/2005JD006103. Yunck et al., 2009: Use of radio occultaPon to evaluate atmospheric temperature data from spaceborne infrared sensors. Terr. Atmos. Ocean. Sci., Vol. 20, No. 1, 71-‐85.
Prior to CrIMSS update: Mx5.3 IDPS 42/22
hep://www.cosmic.ucar.edu/launch/GPS_RO_cartoon.jpg
The COSMIC Data Analysis and Archival Center-‐CDAAC (hep://cosmic-‐io.cosmic.ucar.-‐ edu/cdaac/products.html) was used to obtain COSMIC data. The product used for the results shown was the real-‐Pme atmPrf ‘dry temperature’. A typical COSMIC profile is obtained in about 100 seconds with over 3,000 verPcal samples. A quality control flag is included in the GPS RO netcdf files. For an example day, 19 October 2007, the percentage of GPS profiles marked bad was 2.5%. These bad profiles are excluded from the analysis.
AIRS Scan Geometry with 30 cross track soundings between ±49° and a horizontal
resolution of about 45 km
Onboard the NASA Aqua satellite, AIRS looks toward the ground through a cross-‐track rotary scan mirror which provides +/-‐49.5 degrees (from nadir) ground coverage along with views to cold space and on-‐board spectral and radiometric calibraPon sources every scan cycle. For each of the ninety ground footprints observed each scan, one spectrum with all 2378 spectral samples is obtained. The AIRS IR spaPal resoluPon is 13.5 km at nadir from the 705.3 km orbit. The AIRS Level 2 products contain atmospheric temperature and moisture profiles from about 1 mb down to the surface with a horizontal resoluPon of about 45 km and provide nearly complete daily global coverage in ascending and descending modes. The version 5 support product AIRX2SUP was used in this analysis.
The Crosstrack Infrared Sounder (CrIS) and the advanced Technology Microwave Sounder (ATMS) together represent the latest addiPon to a long series of atmospheric satellite sounders that originated in the late 1970's. CrIS is a hyperspectral sounder that will make it possible to conPnue the advancements of observaPons and research that started with the Atmospheric Infrared Sounder (AIRS) in 2002, and ATMS will similarly conPnue the series of observaPons that started with the Advanced Microwave Sounding Unit (AMSU) first launched by NOAA in 1998. The CrIMSS 42/22 layer EDR IDPS product used in the study was obtained from the NOAA CLASS system. The data used prior to mid October (for the Oct. 1st to 10th analysis) was Mx5.3 and aser mid October (Oct 22nd -‐31st analysis and case study) was Mx6.3.
−18 −17 −16 −15 −14 −13 −22−20
−18−16
10−1
100
101
102
103 −18 −17 −16 −15 −14 −13 −22−20
−18−16
10−1
100
101
102
103
−18 −17 −16 −15 −14 −13 −22−20
−18−16
10−1
100
101
102
103−18 −17 −16 −15 −14 −13 −22
−20−18
−16
10−1
100
101
102
103
AIRS Pressure Level 44 −− 103.0172 mb
130° E 140° E
150° E
−4
−3
−2
−1
0
1
2
3
4
AIRS Pressure Level 30 −− 32.2744 mb
130° E 140° E
150° E
−4
−3
−2
−1
0
1
2
3
4
AIRS Pressure Level 63 −− 300 mb
130° E 140° E
150° E
−4
−3
−2
−1
0
1
2
3
4
Figure 1. From left to right, AIRS a) 32.3 mb (AIRS Level 30), b)103 mb (AIRS Level 44) and c) 300 mb (AIRS Level 63) temperatures differenced from the AIRS closest profile temperature (K) showing the closest profile (black square), circular averaged profile radius (magenta circle), ray averaged profiles (red dots), and the COSMIC RO raypath (black line).
Figure 2. Illustration of the closest (black square), circular (blue circle), and ray path (red dots) methods for a single GPS profile (green) for the circle centered at the GPS RO level of (a) 30 hPa, (b) 100 hPa, (c) 300 hPa, and (d) the perigee point of GPS RO occultation.
c) AIRS Pressure Level 63 (300 mb) a) AIRS Pressure Level 30 (32.3 mb) b) AIRS Pressure Level 44 (103 mb)
IllinoisIndiana
Iowa
Michigan
Minnesota
Ohio
Wisconsin
Closest IR Profiles
94° W 92° W 90° W 88° W 86° W 84° W
40° N
42° N
44° N
46° N
IllinoisIndiana
Iowa
Michigan
Minnesota
Ohio
Wisconsin
Circular IR Profiles
94° W 92° W 90° W 88° W 86° W 84° W
40° N
42° N
44° N
46° N
AQer CrIMSS update: Mx6.3 IDPS 42/22
Conclusions • A methodology has been developed for the validaPon of CrIMSS AVTP
in the upper troposphere and lower stratosphere using GPS radio occultaPon.
• GPS RO dry temperature from the COSMIC network is matched to CrIMSS profiles within one hour.
• The GPS RO horizontal resoluPon of 300 km is accounted for using a raypath method.
• This methodology has been used to evaluate a CrIMSS sosware version update that occurred on 15 October 2012.
• The overall quality control percentage of combined (IR+MW) to combined + IR only + MW only before the update was 6% and aser the update was over 20%.
• The same methodology has been applied to AIRS v5 retrievals to provide a relaPve comparison.
• Coincident RS92 sonde launches from UW-‐Madison are being used in detailed case study analyses.
CrIMSS and AIRS Profile locaPons
Above, the red line indicates the track of the COSMIC GPS RO profile (18:27 UTC). The red star is the perigee point of occultation. The cyan stars indicate CrIMSS profiles (19:03-19:11 UTC) within150 km of Madison. The radiosonde track is shown in green (18:44-20:34 UTC).
The figures below contain an overlay of the GPS RO profile (red), the radiosonde launched from UW-SSEC (green), the circular weighted AIRS profile (black), and the circular weighted CrIMSS profile (blue). The right hand figure illustrates the 300 to 30 mb pressure range for which the COSMIC dry temperature profile is most accurate. Note that both the radiosonde and COSMIC profiles have higher vertical resolution than AIRS or CrIMSS.
Note: The COSMIC GPS RO dry temperature is contaminated by water vapor below about 300mb.
In this temperature profile matchup methodology, spaPally and temporally coincident GPS RO profiles and IR granules were found. For each case, three matchup techniques were used for comparison and analyzed with the same staPsPcal analysis. The closest profile, the circular averaged profile of a radius half the GPS RO horizontal resoluPon, and the ray-‐path averaged profile of AIRS data were computed. In order to invesPgate the effect of the horizontal resoluPon of the GPS RO, these three different approaches to creaPng an IR matchup profile were evaluated. The azimuth angle of the radio occultaPon for each COSMIC RO profile alPtude was used to compute the horizontal extent of the nominal GPS RO horizontal resoluPon at each reported alPtude level (Kursinski et al. 1997). Figure 1 below shows the spaPal extent of a 135 km radius averaging circle on selected AIRS retrieval levels. Figure 2 to the right illustrates the closest, circular, and raypath methods as a 3D profile.
3.5 THE NPOESS CROSS-TRACK INFRARED SOUNDER (CrIS) AND ADVANCED TECHNOLOGY MICROWAVE SOUNDER (ATMS) AS A COMPANION TO THE NEW GENERATION AIRS/AMSU AND IASI/AMSU SOUNDER SUITES
G.A. Bingham*, C. Fish, V. Zavyalov, M.P. Esplin - Space Dynamics Laboratory, Logan, UT N.S. Pougatchev - JPL, Pasadena, CA
W.J. Blackwell - MIT Lincoln Laboratory C.D. Barnet - NOAA, Camp Springs, MD
INTRODUCTION This paper reviews the Cross-track Infrared Sounder (CrIS) and the Advanced Technology Microwave Sounder (ATMS) that together make up the Crosstrack Infrared and Microwave Sounding Suit (CrIMSS). CrIMSS will fly on the NPOESS Preparatory Project (NPP) spacecraft. The many hours provided by industrial and government team members have made CrIS and ATMS world class sensors. The CrIMSS sounding suit builds upon the rich heritage of currently flying sensor including: Atmospheric Infrared Sounder (AIRS), Infrared Atmospheric Sounding Interferometer (IASI) and Advanced Microwave Sounding Unit (AMSU).
SYSTEM OVERVIEW The measurement concept of the CrIS and ATMS sensors is shown in Fig. 1. These two sensors are used in tandem to produce atmospheric temperature, water vapor, and pressure profiles from the NPP satellite. CrIS, a Michelson interferometer, collects data in the form of interferograms that are converted to calibrated
atmospheric spectra on the ground. ATMS is a microwave radiometer. CrIS has higher spectral and spatial resolution, but ATMS has the advantage of being able to produce profiles through clouds. The CrIS and ATMS instrument footprints are co-aligned and scanned in the cross track direction so as to map a swath approximately 2500 km wide on the ground. The ATMS scan extends slightly wider than that of CrIS to ensure that there are no measurement gaps between successive satellite passes at the equator.
The raw CrIS and ATMS instrument data from the spacecraft are packaged into raw data record (RDR). These RDRs are transmitted to the ground and subsequently processed into calibrated radiance spectra by ground software. These sensor data records (SDR) from CrIS and ATMS are combined and further processed into environmental data records (EDR), which include atmospheric temperature, pressure, and water vapor profiles.
Figure 1. CrIS and ATMS system overview
*Gail Bingham, SDL, Logan, UT email: [email protected]
a)
c)
b)
d)
210 215 220 225 230 235 240
102
Temperature (K)
Pres
sure
(hPa
)
Circular AIRS and CrIMSS Profiles
GPSSSEC SondeAIRSCrIMSS
180 200 220 240 260 280 300
10−1
100
101
102
103
Temperature (K)
Pres
sure
(hPa
)
Circular AIRS and CrIMSS Profiles
GPSSSEC SondeAIRSCrIMSS
Both CrIMSS and COSMIC temperature profiles are computed at the AIRS 101 levels using nearest neighbor linear interpola?on prior to sta?s?cal analysis.
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 4100
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 3465
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 527
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 1001
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 794
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 811
Antarctic (60S - 90S)
South Mid-Latitudes (30S – 60S)
Arctic (90N – 60N)
Tropics (30N - 30S)
North Mid-Latitudes (60N – 30N)
Global
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 507
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 498
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 849
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 718
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 949
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 911
Octob
er 1
st – 10t
h
CrIM
SS – COSM
IC Octob
er 22n
d – 31s
t
AIRS
– COSM
IC
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 2323
−2 0 2
101
102
103
T (K)Pr
essu
re (h
Pa)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 378
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 459
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 445
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 531
−2 0 2
101
102
103
T (K)
Pres
sure
(hPa
)
BIAS
ClosestCircularRay
0 1 2 3T (K)
RMS
0 1 2 3T (K)
STDEV
0 50 100Yield (%)
YIELD
N = 510
Octob
er 22n
d – 31s
t
Antarctic (60S - 90S)
South Mid-Latitudes (30S – 60S)
Arctic (90N – 60N)
Tropics (30N - 30S)
North Mid-Latitudes (60N – 30N)
Global
AIRS v5 minus COSMIC aQer CrIMSS update