ASL Cal L. Strow UMBC Introduction AIRS L1C Frequency - - PowerPoint PPT Presentation

▶
asl
SMART_READER_LITE
LIVE PREVIEW

ASL Cal L. Strow UMBC Introduction AIRS L1C Frequency - - PowerPoint PPT Presentation

Jan 27, 2024 239 likes •534 views

ASL Cal L. Strow UMBC Introduction AIRS L1C Frequency Calibration Raw Data Model Fit M3 versus M10 Summary L. Larrabee Strow and Scott Hannon Physics Department and Joint Center for Earth Systems Technology University of Maryland

slide-1
SLIDE 1

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

AIRS L1C Frequency Calibration

  • L. Larrabee Strow and Scott Hannon

Physics Department and Joint Center for Earth Systems Technology University of Maryland Baltimore County (UMBC)

Airs Science Team Meeting - Passadena - CA May 2, 2009

1 / 27

slide-2
SLIDE 2

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Overview

Update from last October Science Team Meeting Cross-correlation technique works well for several modules, except over poles. Fall 2008: calibrated whole mission using M12 and cloud-cleared radiances M12 failed, to some degree, over poles (only Q-branch had contrast). Did not use internal QA. This presentation: Calibrated whole mission using M3 (water, 1400 cm−1) and M10 (CO2, 750 cm−1), retained and used internal QA (B(T) contrast). Goal: Make frequency calibration a “non-issue” for AIRS climate applications.

2 / 27

slide-3
SLIDE 3

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Frequency Calibration

Use a cross-correlation technique on M3 and M10 for ν calibration Cross-correlate L2CC radiances with Calc radiances. Calcs done using AIRS L2 retrievals. Careful selection of channels One ν calibration per granule. Units: using “micron” shift in focal plane.

Shifts referred to as “yoffsets” 1 micron ∼ 1% of an SRF FWHM Yoffsets measured relative to TVAC values, generally had ∼

  • 14 µm shift at launch, with about 1 µm during rest of

mission.

3 / 27

slide-4
SLIDE 4

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

∆ B(T) for a dx = 1 µm

1µm Equal to Mission Shift, Orbital ∼ 0.4 µm

500 1000 1500 2000 2500 3000 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018 0.02 0.022 0.024 Wavenumber (cm−1) Δ ν for yoffset = 1 μm

4 / 27

slide-5
SLIDE 5

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Frequency Calibration Model

Raw ν (Yoffset) calibrations were binned by 2 deg. in orbit phase, giving 180 data sets, each one is fit to the following expression: y(t) = yo − b1exp(−t/τ) +

3

  • i=1

[ai sin(2πt + φi)] Fast time behavior is orbit phase (latitude), parameterized by 180 values for b1, τ, three ai harmonic terms, and their phases, φi. Slow time behavior is due to seasonal and slower effects. Data first averaged over 16 day time periods for fits to the above equation.

5 / 27

slide-6
SLIDE 6

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

M3 Raw Calibration (per granule)

With, and without B(T) Contrast Filter

2004 2005 2006 2007 2008 2009 −15 −14.5 −14 −13.5 −13 Time Yoffset (μm) No QA BT Contrast > 10K

6 / 27

slide-7
SLIDE 7

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

M10 Observations (per granule)

With, and without B(T) Contrast Filter

2004 2005 2006 2007 2008 2009 −15 −14.5 −14 −13.5 −13 Time Yoffset Obs−Cal (μm)

7 / 27

slide-8
SLIDE 8

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

OLD ν Calibration

Binned by 2 deg in latitude, 16 days in time

8 / 27

slide-9
SLIDE 9

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

M3 Yoffset vs Orbit Phase

Binned by 2 deg in latitude, 16 days in time

2005 2007 50 100 150 200 250 300 350

Time Orbit Phase (deg)

yoff (μm)

−14.3 −14.2 −14.1 −14 −13.9 −13.8 −13.7 −13.6 −13.5 −13.4 −13.3

9 / 27

slide-10
SLIDE 10

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Tropical M3 Calibration

Further binning to ±30 deg

2005 2007 −14.4 −14.2 −14 −13.8 −13.6 −13.4 −13.2 Time Yoffset (μm) Decending Ascending Difference

10 / 27

slide-11
SLIDE 11

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Polar M3 Calibration

2005 2007 −14.3 −14.2 −14.1 −14 −13.9 −13.8 −13.7 −13.6 −13.5 −13.4 −13.3 Time Yoffset (μm) Decending Ascending Difference

11 / 27

slide-12
SLIDE 12

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Yoffset versus Latitude

Averaged over 5 Years

−100 −80 −60 −40 −20 20 40 60 80 −14.05 −14 −13.95 −13.9 −13.85 −13.8 −13.75 −13.7 Latitude Yoffset (μm) Descending Ascending

12 / 27

slide-13
SLIDE 13

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

M3 Monthly Avg. Yoffset vs Orbit Phase

Back to raw binned data but aggregated by month (for all years).

50 100 150 200 250 300 350 1 2 3 4 5 6 7 8 9 10 11 12

Orbit Phase (deg) Month South Pole North Pole

yoff (μm) −14.1 −14.05 −14 −13.95 −13.9 −13.85 −13.8 −13.75 −13.7 −13.65 −13.6

13 / 27

slide-14
SLIDE 14

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Fit to our Parameterization Equation

Will will now examine reasonableness of the fitted parameters. Reminder: y(t) = yo − b1exp(−t/τ) +

3

  • i=1

[ai sin(2πt + φi)]

14 / 27

slide-15
SLIDE 15

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Amplitude of Sinusoidal Terms

These are the a1, a2, a3 terms for each 2-deg bin of orbit phase.

50 100 150 200 250 300 350 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Orbit Phase (deg) Amplitude (μm) ω 2*ω 3*ω

15 / 27

slide-16
SLIDE 16

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Yoffset Decay Time Constant vs Latitude

τ for both M3 and M10 versus Latitude

−100 −80 −60 −40 −20 20 40 60 80 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 Latitude Yoffset Decay Rate τ (years)

Δ τ of 1 year ≡ 0.2 μm after 5 years

M3 M10

16 / 27

slide-17
SLIDE 17

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Yoffset Decay Amplitude vs Latitude (b1 Terms)

y(t) = yo − b1exp(−t/τ) + 3

i=1[ai sin(2πt + φi)]

−100 −80 −60 −40 −20 20 40 60 80 −3.2 −3 −2.8 −2.6 −2.4 −2.2 −2 −1.8 Latitude Yoffset Decay Amplitude ( μm) M3 M10

17 / 27

slide-18
SLIDE 18

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Obs - Fitted Calibration, M3

Mission through 2009+

2004 2005 2006 2007 2008 2009 −14.4 −14.2 −14 −13.8 −13.6 −13.4 −13.2 Time Yoffset (μm) Obs Calc

18 / 27

slide-19
SLIDE 19

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Obs - Fitted Calibration, M3: 2007

1-year

Q1−07 Q2−07 Q3−07 Q4−07 Q1−08 −14.4 −14.3 −14.2 −14.1 −14 −13.9 −13.8 −13.7 −13.6 −13.5 Time Yoffset (μm) Obs Calc

19 / 27

slide-20
SLIDE 20

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Obs - Fitted Calibration, M3: Jan. 08, 2007

1-day

00:00 06:00 12:00 18:00 00:00 −14.2 −14.1 −14 −13.9 −13.8 −13.7 −13.6 Time Yoffset (μm) Obs Calc

20 / 27

slide-21
SLIDE 21

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Histogram of M3 Obs-Calc Good FOVs

Errors are Gaussian; Bias = 0.00 µm, Std = 0.06 µm

−0.5 0.5 1000 2000 3000 4000 5000 6000 \# of Observations Yoffset Obs−Cal (μm)

21 / 27

slide-22
SLIDE 22

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Histogram of M3 Obs-Calc for Outlier FOVs

−1.5 −1 −0.5 0.5 1 500 1000 1500 Yoffset Obs−Cal (μm) \# of Observations

22 / 27

slide-23
SLIDE 23

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Difference Between M3 and M10:

Raw Data and Fit

2004 2005 2006 2007 2008 2009 −0.5 −0.4 −0.3 −0.2 −0.1 0.1 0.2 0.3 0.4 0.5 Time Δ Yoffset Obs−Cal ( μm) M3−M10 Obs M3−M10 Calc

23 / 27

slide-24
SLIDE 24

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Histogram of M3 minus M10 Obs

−0.4 −0.2 0.2 0.4 0.6 0.5 1 1.5 2 2.5 3 x 10

4

M3−M10 Yoffset ( μm) \# of Observations

24 / 27

slide-25
SLIDE 25

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Variation of M3 and M10 Yoffset: Tropics

2004 2005 2006 2007 2008 2009 −14.4 −14.3 −14.2 −14.1 −14 −13.9 −13.8 −13.7 −13.6 −13.5 Time Yoffset μm M3−Des M3−Asc M10−Des M10−Asc

25 / 27

slide-26
SLIDE 26

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Other Arrays

Use tropics for offsets

26 / 27

slide-27
SLIDE 27

ν Cal

  • L. Strow

UMBC Introduction Raw Data Model Fit M3 versus M10 Summary

ASL

Conclusions

M3 appears to be the best module for frequency calibration Behavior is somewhat complicated, but reasonable Use tropics to determine static offset between M3 and

  • ther modules. Higher errors in some modules reflect

lower requirement for knowlege of the module frequency. Some concern that some modules may move differently. Will evaluate fitting parameters in tropics for almost all modules. Almost ready for V6 implementation. τ term should predict future drifts unless instrument is stressed.

27 / 27