1. 1 Mark Hervig, AIM EPMR GATS SOFIE Data / Retrieval Performance AIM End of Prime Mission Review Mark Hervig GATS Inc.

2. 2 Mark Hervig, AIM EPMR GATS Outline SOFIE Overview Payload status Data collection status Retrieval Algorithm Version History Performance On-orbit versus requirements Validation / correlative comparisons Community use Lessons Learned

3. 3 Mark Hervig, AIM EPMR GATS SOFIE Overview Solar occultation measurements. 16 wavelengths (0.29 - 5.3  m). Retrievals: Temperature, O 3, H 2 O, CO 2, CH 4, NO, PMCs, meteoric smoke. Measurement latitudes: 65 - 82  N (sunrise) & 65 - 82  S (sunset). sofie.gats-inc.com

4. 4 Mark Hervig, AIM EPMR GATS SOFIE Health and Performance (Power) Engineering telemetry items - nominal –Monitored by the AIM MOC and SOFIE POC –Well within operational limits

5. 5 Mark Hervig, AIM EPMR GATS SOFIE Health and Performance (Temperature) Engineering telemetry items - nominal –Monitored by the AIM MOC and SOFIE POC –Well within operational limits

6. 6 Mark Hervig, AIM EPMR GATS SOFIE Health and Performance (Dark Offset) Electronic Dark Offset –No significant change since launch

7. 7 Mark Hervig, AIM EPMR GATS FLAWS DescriptionStatusScience Effect 10 & 58SSB SRAM Error Flag SetClosedNone 34Voltage/Current DeviationClosedNone 31 & 52Incorrect FSW image loadedClosed66 Events Lost

8. 8 Mark Hervig, AIM EPMR GATS SOFIE Autonomy SOFIE predicts event times using measurements of solar extent. Solar extent is a function of altitude (density). Monitoring the time of a given extent yields orbital period, which allows event time predictions.

9. 9 Mark Hervig, AIM EPMR GATS SOFIE Autonomy Passed all unit tests on individual modules using IRSIM Passed all system level testing on the SOFIE EM –Performed original baseline testing Aliveness Test Limited Performance Test Comprehensive Performance Test Software Performance Test –Developed an additional test procedure for autonomy –Performed autonomy functional testing Created simulated solar data for testing Verified calculations in passive and active autonomy mode Verified execution of events created through active autonomy mode Ran active and passive autonomy for over 48 hours Passed on-orbit testing in passive (6/28/07) and active (7/16/07) autonomy mode –Verified all calculations through telemetry with SOFIE sunsensor data and TLE’s Orbital period Predicted and actual sunset/sunrise event times –No change in SOFIE health and performance

10. 10 Mark Hervig, AIM EPMR GATS SOFIE Data Collection Status Since End of Commissioning Commanded - (SOFIE Observed the sun during the event) –21082 Events Commanded –20899 Events Collected –99.1% Collected Total Possible Events - (2x orbit number) –22631 Total Events Possible –20899 Events Collected –92.4% Collected (Statistics through June 30, 2009)

11. 11 Mark Hervig, AIM EPMR GATS SOFIE Data Collection Status Total Possible Events –22631 Total Events Possible –20899 Total Events Collected –92.4 % Collected (Statistics through June 30, 2009) CauseNumber of Events Missing Percentage of Total SOFIE Off150187.0 No TLM1468.0 FSW664.0 Unable to process191.0 Total1732100.00

12. 12 Mark Hervig, AIM EPMR GATS Recorded Data Not Released Breakdown of Collected Events –22630 Total Events Collected –21150 Total Events Released –99.0% Data Released (Statistics through June 30, 2009) CauseNumber of Events Not Released Percentage of Total Failed in L1814.29 Failed in L2291.54 Failed in Validation177694.17 Total1886100.00

13. 13 Mark Hervig, AIM EPMR GATS SOFIE Retrieval Algorithm (Top Level) Level 0: Data quality checks Time conversion Combine science packets into occultation events Level 1: Calibration (solar source, gain, background) Signal conditioning (nonlinearity, drift) Altitude registration Level 2: Retrieval of geophysical parameters Level 3: Time versus altitude cross section plots Each retrieved profile is inspected for quality prior to release.

14. 14 Mark Hervig, AIM EPMR GATS SOFIE Data Version History V1.01, Feb 2008 Initial public release V1.02, Dec 2008 Improved signal conditioning (higher altitudes obtained) Improved forward model (e.g., added O 3 interference in bands 3&4) V1.022, Feb 2009 Improved signal drift corrections Improved solar source corrections (i.e., pointing drift) Improved altitude registration Non-LTE temperature retrievals, 3 to 10K colder, extended to 105 km V1.03, under development Corrections to event timing Simultaneous temperature and CO 2 retrievals Refraction angle temperature retrievals Improved PMC corrections (O 3, CO 2 ) Improved off-axis FOV response

15. 15 Mark Hervig, AIM EPMR GATS SOFIE Public Release Data Products Level 2: Profiles of Temperature, O 3, H 2 O, CH 4, NO, PMC extinction A file contains data for one day (30 events) Level 3: Time versus height cross section plots for all parameters 1 file per day per hemisphere CV (common volume): Sample volume geometry, column O 3, mesopause Derived PMC properties: Cloud top, peak, and base altitudes Particle shape Effective radius Gaussian size distribution (concentration, median radius, width) Vertical column ice abundance A file contains data for an entire PMC season Engineering & performance data can be viewed online (trend plots, reports, etc…)

16. 16 Mark Hervig, AIM EPMR GATS SOFIE Performance Summary Geophysical Parameter Precision (83 km altitude) Required / On-orbit Altitude Range (km) Required / On-orbit Vertical Resolution (km) Required / on-orbit NIR cloud extinction 5  10 -6 / 2  10 -8 km -1 78 – 85 / 75 – 90  3 / 1.8 IR cloud extinction 5  10 -5 / 2  10 -8 km -1 78 – 85 / 75 – 90  3 / 1.8 Temperature5 / 0.5 K70 – 90 / 15 - 105  3 / 1.8 O 3 mixing ratio100 / 10 ppbv78 – 90 / 55 - 100  3 / 1.8 H 2 O mixing ratio0.6 / 0.1 ppmv78 – 90 / 15 - 100  3 / 1.8 CO 2 mixing ratio10 / 2 ppmv80 – 100 / 68 - 92  3 / 1.8 CH 4 mixing ratio50 / 5 ppbv30 – 90 / 15 - 75  3 / 1.8 NO mixing ratio53 / 39 ppbv80 – 95 / 30 - 140  5 / 1.8 Meteoric Smoke NA / 2  10 -8 km -1 NA / 35 - 90NA / 1.8

17. 17 Mark Hervig, AIM EPMR GATS SOFIE Temperature (Std. Product) Precision (83 km): Required: 5 K Pre-flight prediction: 0.6 K On-orbit: 0.2 K Altitude Coverage: Required: 70 - 90 km Pre-flight prediction: 15 - 95 km On-orbit: 15 - 105 km

18. 18 Mark Hervig, AIM EPMR GATS SOFIE Temperature (from refraction angle) Hi precision (0.02 arcsec) measurements of solar extent vs. altitude are used to retrieve temperature/pressure below ~65 km. These results eliminate the need for altitude registration using independent (NCEP) T/P profiles. (Planned for release in V1.03)

19. 19 Mark Hervig, AIM EPMR GATS SOFIE Temperature (Continued) SOFIE can now retrieve the temperature of ice using measurements within the OH-stretch region (3  m). These results are in good agreement the upcoming V1.03 T’s, and also match the Falling Sphere (FS) record.

20. 20 Mark Hervig, AIM EPMR GATS SOFIE Water Vapor Precision (83 km): Required: 600 ppbv Pre-flight prediction: 45 ppbv On-orbit: 70 ppbv Altitude Coverage: Required: 78 - 90 km Pre-flight prediction: 15 - 100 km On-orbit: 15 - 100 km

21. 21 Mark Hervig, AIM EPMR GATS SOFIE Ozone Precision (83 km): Required: 100 ppbv Pre-flight prediction: 2 ppbv On-orbit: 10 ppbv Altitude Coverage: Required: 78 - 90 km Pre-flight prediction: 15 - 100 km On-orbit: ~55 - 105 km

22. 22 Mark Hervig, AIM EPMR GATS SOFIE Methane Precision (83 km): Required: 50 ppbv Pre-flight prediction: 36 ppbv On-orbit: 5 ppbv Altitude Coverage: Required: 30 - 90 km Pre-flight prediction: 15 - 90 km On-orbit: 15 - 75 km Notes: The lower altitudes obtained are real, the CH 4 signal above ~75 km is in the noise, and SOFIE can “see zero” much better than previous instruments.

23. 23 Mark Hervig, AIM EPMR GATS SOFIE Carbon Dioxide SOFIE CO 2 results are not yet released, but the retrievals are progressing. Precision (83 km): Required: 10 ppbv Pre-flight prediction: 7 ppbv On-orbit: 2 ppbv Altitude Coverage: Required: 80 - 100 km Pre-flight prediction: 15-100 km On-orbit: 68 - 92 km Notes: CO 2 data release is eminent in V1.03.

24. 24 Mark Hervig, AIM EPMR GATS SOFIE Nitric Oxide Precision (83 km): Required: 53 ppbv Pre-flight prediction: 59 ppbv On-orbit: 39 ppbv Altitude Coverage: Required: 80 - 95 km Pre-flight prediction: 80 - 110 km On-orbit: 30 - 140 km Notes: 1.Improved altitude range is due to excellent removal of H 2 O interference. 2.Currently only sunset (Southern Hemisphere) results, sunrises require further analysis. 3.Sunset NO data are released to public.

25. 25 Mark Hervig, AIM EPMR GATS SOFIE PMC Extinction Precision (83 km): Required: 5  10 -6 km -1 Pre-flight prediction: 5  10 -8 km -1 On-orbit: 2  10 -8 km -1 Altitude Coverage: Required: cloud Pre-flight prediction: cloud On-orbit: cloud

26. 26 Mark Hervig, AIM EPMR GATS SOFIE Derived Physical PMC Properties Ice mass density: uncertainty < 10%, (0.06 ng m -3 sensitivity is 50-200 times better than previous instruments) Particle shape: spheroid axial ratio, uncertainty < 20%. Effective radius: uncertainty < 10% Gaussian size distribution: (N, r m,  r) uncertainties < ~35%

27. 27 Mark Hervig, AIM EPMR GATS SOFIE Meteoric Smoke The excellent SOFIE sensitivity has provided the first satellite observations of meteoric smoke. 35 - 90 km altitude 2  10 -8 km -1 precision SOFIE smoke observations mitigate current CDE difficulties, and are more relevant to PMCs. SOFIE observes smoke in the atmosphere, applies directly to the PMC environment. SOFIE may yield estimates of total meteoric influx, which was the CDE goal.

28. 28 Mark Hervig, AIM EPMR GATS SOFIE Refereed Publications (15+) SOFIE data have been used in 15+ refereed publications to date, by AIM and outside investigators. Bardeen et al., Numerical simulations of the three-dimensional distribution of polar mesospheric clouds and comparisons with CIPS and SOFIE observations, J. Geophys. Res., in review, 2009. Baumgarten et al., The noctilucent cloud (NLC) display during the ECOMA/MASS sounding rocket flights on August 3, 2007: Morphology on global to local scales, Ann Geophys, 27, 953-965, 2009. Eckerman et al., High-Altitude Data Assimilation System Experiments for the Northern Summer Mesosphere Season of 2007, J. Atmos. Solar-Terr. Phys., doi:10.1016/j.jastp.2008.09.036, 2009. Gordley et al., The Solar Occultation For Ice Experiment (SOFIE), J. Atmos. Solar-Terr. Phys., doi:10.1016/j.jastp.2008.07.012, 2009. Gordley et al., High precision refraction measurements by solar imaging during occultation: results from SOFIE, Applied Optics, Volume 48, Issue 25, 4814- 4825, doi:10.1364/AO.48.004814, 2009. Hervig et al., Interpretation of SOFIE PMC measurements: Cloud identification and derivation of mass density, particle shape, and particle size, J. Atmos. Solar-Terr. Phys., doi:10.1016/j.jastp.2008.07.009, 2009. Hervig et al., SOFIE PMC observations during the northern summer of 2007, J. Atmos. Solar-Terr. Phys., doi:10.1016/j.jastp.2008.08.010, 2009. Hervig et al., Relationships between PMCs, temperature and water vapor from SOFIE observations, J. Geophys. Res., in press, 2009. Hervig et al., First satellite observations of meteoric smoke in the middle atmosphere, Geophys. Res. Letters, doi:10.1029/2009GL039737, 2009. Nielsen et al., Seasonal variation of the quasi 5-day planetary wave: Causes and consequences for polar mesospheric cloud variability in 2007, J. Geophys Res, in review, 2009. Quang et al., Microphysical parameters of mesospheric ice clouds derived from calibrated observations of polar mesosphere summer echoes at Bragg wavelengths of 2.8m and 30 cm, J. Geophys. Res., In review, 2009. Robertson et al., Mass analysis of charged aerosol particles in NLC and PMSE during the ECOMA/MASS campaign, Ann Geophys, 27, 1213-1232, 2009. Russell et al, Aeronomy of Ice in the Mesosphere (AIM): Overview and early science results, J. Atmos. Solar-Terr. Phys., doi:10.1016/j.jastp.2008.08.011, 2009. Stevens et al., The diurnal variation of noctilucent cloud frequency near 55 ｰ N observed by SHIMMER, J. Atmos. Solar-Terr. Phys., doi:10.1016/j.jastp.2008.10.009, 2009. Stevens et al., Tidally induced variations of PMC altitudes and ice water content using a data assimilation system, J. Geophys. Res., in review, 2009.

29. 29 Mark Hervig, AIM EPMR GATS SOFIE Conference Presentations (AIM team: 22+) Hervig, M., et al., PMC Measurements from the Solar Occultation For Ice Experiment (SOFIE), AGU Fall Meeting, San Francisco, 2007. Gordley, L., et al., SOFIE Preliminary Results on the PMC environment, LPMR Meeting, Fairbanks, 2007. Hervig, M., et al., The Solar Occultation For Ice Experiment (SOFIE), LPMR Meeting, Fairbanks, 2007. Russell J. M. III, et al., An overview of the Aeronomy of Ice in the Mesosphere Mission and Preliminary Results, LPMR Meeting, Fairbanks, 2007. Hervig, M., et al., AIM Common Volume Analysis, LPMR Meeting, Fairbanks, 2007. Gordley, L., et al., Sounding the Upper Mesosphere using Broadband Solar Occultation -- Initial Results from the SOFIE Experiment, SPIE Meeting, San Diego, 2007. Gordley, L., et al., Sounding the Upper Mesosphere using Broadband Solar Occultation – Initial Results from the SOFIE Experiment, 4th International Limb Workshop, Virginia Beach, 2007. Russell J. M. III, et al., Hemispheric differences in PMC altitudes observed by the AIM satellite for the 2007/2008 seasons, AGU Spring Meeting, Ft. Lauderdale, 2008. Stevens, M., et al., Inter-hemispheric Asymmetries of Mesospheric Cloudiness, AGU Spring Meeting, Ft. Lauderdale, 2008. Hervig, M., et al., PMC Measurements from the Solar Occultation For Ice Experiment (SOFIE), AGU Spring Meeting, Ft. Lauderdale, 2008. Hervig, M., et al., SOFIE Measurements of PMCs, Water Vapor, and Temperature, AGU Fall Meeting, San Francisco, 2008. Merkel, A., et al., Longitudinal variability of Polar Mesospheric Cloud (PMC) albedo and frequency from the Cloud Imaging and Particle Size Experiment: Comparison of the 2007 and 2008 Northern Hemisphere cloud seasons, AGU Fall Meeting, San Francisco, 2008. Bardeen, C., et al., Sensitivity of WACCM/CARMA simulations of polar mesospheric clouds to gravity wave and microphysics parameterizations, AGU Fall Meeting, San Francisco, 2008. Bailey, S. et al., Scattering Phase Functions and Particle Sizes from the Aeronomy of Ice in the Mesosphere (AIM) Explorer, AGU Fall Meeting, San Francisco, 2008. Stevens, M., et al., The PMC Mass for the 2007 Northern Summer: Results From a Microphysical Model Driven by a Data Assimilation System, AGU Fall Meeting, San Francisco, 2008. Hervig, M., et al., PMC Particle Size from SOFIE Observations, LPMR Meeting, Stockholm, 2009. Hervig, M., et al., First Satellite Observations of Meteoric Smoke in the Middle Atmosphere, LPMR Meeting, Stockholm, 2009. Russell, J. M. III, et al., The Aeronomy of Ice in the Mesosphere mission: Science results after four PMC seasons, LPMR Meeting, Stockholm, 2009. Gordley, L., et al., SOFIE data − current and projected results, LPMR Meeting, Stockholm, 2009. Bailey, S., et al., A working group for determining the state-of-the-art in mesospheric ice particle sizes, LPMR Meeting, Stockholm, 2009. Hervig, M., et al., First Satellite Observations of Meteoric Smoke in the Middle Atmosphere, IAGA Meeting, Sopron, 2009. Russell, J. M. III, et al., The Aeronomy of Ice in the Mesosphere mission: Science results after four PMC seasons, IAGA Meeting, Sopron, 2009.

30. 30 Mark Hervig, AIM EPMR GATS SOFIE Lessons Learned 1) Requirement Flow Down (Scientist - Engineer Communication): Improper communication of SNR requirements caused an apparent failure to meet SNR requirements at CDR. In reality, the SOFIE design met SNR requirements across the board. 2) Component Performance Specifications: 2.1. Vendor detector specs for band 5-16 did not cover the unique demands of viewing the sun. Illumination was intense enough to drive several detectors into a non-linear regime (response non-linearity was successfully calibrated before flight). This may have been discovered at the component/design level with further testing and/or simulation. 2.2. The published vendor interface control document did not clearly specify the initialization process for the SOFIE sun sensor CMOS image array. As such, the array was pulling higher than desired levels of current following startup due to incomplete array initialization. This issue was mitigated by operational constraints for use on orbit. 3) Design Margins: The SOFIE design originally contained an agile steering mirror, which was later removed due to unexpected increases in launch loads (the mirror successfully passed original launch load levels). This highly desired component could have been designed early in the mission to include a launch restraint.

31. 31 Mark Hervig, AIM EPMR GATS SOFIE Lessons Learned (cont’d) 4) Calibration and Testing: The SOFIE band 2 pre-amp was saturated upon viewing the sun from orbit. Ground calibration was difficult because UV solar intensity could not be reproduced in the lab. Controllable gains within the analog circuit would have eliminated this problem. 5) Software Boot-up Design: The SOFIE software system initializes from boot PROM. This approach was taken to mitigate on-orbit radiation induced EEPROM upset concerns. While this approach is solid should code updates not occur after the first software burn, it becomes very limiting should further updates be needed. A solely EEPROM based approach, possibly with voting schemes, is probably a more agile yet robust approach to overcome this limiting condition.

32. 32 Mark Hervig, AIM EPMR GATS Summary The SOFIE instrument continues to perform as expected. SOFIE data products meet or exceed requirements. SOFIE data have been widely published and presented. SOFIE PMC measurements have become a benchmark within the community.

33. 33 Mark Hervig, AIM EPMR GATS Backup Slides Follow

34. 34 Mark Hervig, AIM EPMR GATS FLAWS #10 & #58 (Closed) May 15, 2007 & May 11, 2009 –SSB SRAM error flag set SRAM check only performed on boot-up Communications buffer for RS-422 resides in SRAM. Asynchronous communication from the C&DH board during boot- up will cause the SRAM error. –Occasionally, when the sun slews off the SSB’s FPA in elevation, the SSB will reset We still do not understand “why” it resets SSB will boot to PROM unless it receives a EEPROM image from the C&DH board. –C&DH only sends an EEPROM during its boot-up sequence –No issues detected with SSB running from PROM –Science data was not affected. Running from PROM image since May 11, 2009

35. 35 Mark Hervig, AIM EPMR GATS FLAWS #34 (Closed) Voltage/current deviation –At two known points since the last turn-on of the AIM instrument suite, the current levels on the SOFIE sensor +-12V inst lines have deviated (worst case is a step function ~100 mA on the -12V line) from nominal. The deviations were constant, DC shifts in the current. The SOFIE sensor continued to function nominally during the changes. From electronics box temperature profiles, it appears that the extra current during the deviations is being sunk on the SSG Mirror Drive PCB. –Science data was not affected

36. 36 Mark Hervig, AIM EPMR GATS FLAWS #31 & #52 (Closed) Feb 10, 2009 –Reset performed due to the Sun Sensor Board (SSB) using PROM FSW image instead of its default EEPROM image Occurs due to an asynchronous FPGA read/write issue identified during ground testing and twice on-orbit PFR #1471 Feb 11, 2009 and Feb 12, 2009 –Reset performed because Command and Data Handling (C&DH) board did not receive the command to boot to the secondary EEPROM FSW image (autonomy). However, it booted correctly to its default EEPROM FSW image (flight version w/o autonomy). Same issue with the asynchronous FPGA read/write issue in PFR #1471 –Issue resolved by sending the FSW image command once per second for 17 seconds. Testing revealed no failures to boot to the commanded image after 1000 reset iterations Issue resolved 66 events lost

37. 37 Mark Hervig, AIM EPMR GATS SOFIE Community Use SOFIE data are available online: SOFIE homepage (sofie.gats-inc.com) AIM homepage (aim.hamptonu.edu) SOFIE Webpage Usage Since Launch: Visits: 11,064 (Unique visitors: 2,724) Top 5 Visiting Countries: United States (9,779) Mexico (412) United Kingdom (164) Ireland (120) Germany (82)

38. 38 Mark Hervig, AIM EPMR GATS SOFIE PMC Measurement Highlights The dependence of ice mass density on temperature varies: Early season (cooling): steep dependence on T due to simultaneously increasing H2O. Late season (warming): increasing H2O due to sublimation buffers the effect of warming. Ice concentration increases at low temperatures: Increased ice mass is due to nucleation of more particles, rather than growth of existing ice.

39. 39 Mark Hervig, AIM EPMR GATS SOFIE PMC Measurement Highlights Before AIM, PMCs were considered sporadic layers 1 or 2 km thick. SOFIE now shows a persistent ice layer up to 10 km thick. SOFIE ice mass densities are consistent with model predictions using SOFIE temperature and H 2 O.

40. 40 Mark Hervig, AIM EPMR GATS SOFIE Ice Detection / Sensitivity SOFIE detects ice using the band 9 & 10 measurements (3.064 & 3.186  m). The current detection threshold is 1×10 -7 km -1 (or 5 x the noise of 2×10 -8 km -1 ) Note: ~1.5 mono-layers of ice on smoke gives the 3.064  m extinction = the noise

41. 41 Mark Hervig, AIM EPMR GATS SOFIE Ice Particle Shape Certain IR extinction ratios are sensitive to particle shape, and insensitive to size. Particle shape is described using oblate and prolate spheroids, and the axial ratios (AR) are determined. The solution is not unique, prolate and oblate are allowed. Uncertainties are < 20% (except for AR near 1). NH 2007 averages SOFIE: 2.4 Lidar: 2.1

42. 42 Mark Hervig, AIM EPMR GATS SOFIE Effective Radius (r e ) IR/NIR extinction ratios are directly proportional to r e. r e = 3 x V / S (3rd moment / 2nd moment) The results are insensitive to particle shape. r e uncertainties are < 10%. Minimum detectable r e appears to be ~9 nm. (no predictable lower limit) NH 2007 averages SOFIE: 35 nm Lidar: 38 nm

43. 43 Mark Hervig, AIM EPMR GATS SOFIE Ice Mass Density (M ice ) IR extinction is directly proportional to volume density. Slight dependence on particle shape is captured. M ice uncertainties are < 10%. Minimum detectable M ice is ~0.06 ng m -3. (smallest observed is 0.08 ng m -3 ) HALOE: M ice > 13 ng m -3, ALOMAR lidar: M ice > 2 ng m -3. Minimum detectable IWC is ~0.1  g m -2. (smallest observed is 0.11  g m -2 ) NH 2007 averages SOFIE: 14 ng m -3 Lidar: 47 ng m -3