GOES-R AWG Aviation Team: SO2 Detection Presented By: Mike Pavolonis NOAA/NESDIS/STAR

Outline Aviation Team Executive Summary Algorithm Description ADEB and IV&V Response Summary Requirements Specification Evolution Validation Strategy Validation Results Summary

Aviation Team Co-Chairs: Ken Pryor and Wayne Feltz Aviation Team K. Bedka, J. Brunner J. Mecikalski, M. Pavolonis, B. Pierce W. Smith, Jr., A. Wimmers, J. Sieglaff Others Walter Wolf (AIT Lead) Shanna Sampson (Algorithm Integration) Zhaohui Zhang (Algorithm Integration) Thanks to many others as well, S. Wanzong, D. Hillger, etc. 3 3

Executive Summary The ABI SO2 detection algorithm produces a yes/no SO2 flag Software Version 4 was delivered in January (2011). The CUTR was completed in January (2011). ATBD (100%) and test datasets are scheduled to be delivered in June (2011). The SO2 detection algorithm utilizes spectral and spatial information to identify SO2 clouds. Only infrared measurements are used, so the algorithm methodology is day/night consistent. Validation Datasets: Hyperspectral ultra-violet based retrievals of SO2 total column loading from the Ozone Monitoring Instrument (OMI). Validation studies indicate that the SO2 loading algorithm is meeting the 100% accuracy specification. Slide 2

Algorithm Description This slide does not count towards the slide number limit

Algorithm Summary Input Datasets: 1). ABI channels 8, 10, 11, 14, and 15, 2). Satellite viewing angle, 3). Clear sky radiances and transmittances, 4). Cloud frequency of occurrence LUT -ratios are calculated to help identify cloud microphysical “anomalies” possibly associated with SO2 absorption. Cloud objects are constructed from potential SO2 pixels. Spectral and spatial cloud object statistics are used to determine if all of the pixels that compose a cloud object contain SO2 or do not contain SO2. A flag indicating the approximate SO2 loading is also produced and stored in the quality flags. Slide 2

Algorithm Channel Selection The ABI algorithm utilizes SO2 absorption lines located in the channel 10 (7.3 m) and channel 11 (8.5 m) bands in combination with the lack of SO2 absorption in the 6.2, 11, and 12 m bands to detect SO2 induced spectral anomalies. ABI channel 11 coverage ABI channel 10 coverage

Spectral Information As in Pavolonis (2010), effective absorption optical depth ratios (-ratios) are used to infer cloud microphysical information. Cloud microphysical anomalies (large deviations from theoretical values) are then used to identify potential SO2 contaminated pixels. Spectral ratio of effective absorption optical depth Spectral ratio of scaled extinction coefficients This relationship provides a direct link to theoretical size distributions from the measurements

Spectral Information Water Cloud Ice Cloud SO2 Cloud Compare the spectral signature of an SO2 cloud, an ice cloud ,and a liquid water cloud.

Spectral Information (8.5, 11), (12, 11) SO2 spectral signature The SO2 cloud is characterized by the simultaneous occurrence of large (12, 11) and large (8.5, 11). Ice clouds tend to have smaller (8.5, 11) relative to SO2 clouds. SO2 clouds tend to have larger (12, 11) relative to liquid water clouds when hydrometeors are co-located with the SO2. SO2 Liquid Water Ice

Spectral Information The SO2 cloud is characterized by the simultaneous occurrence of large (7.3, 11) and large (8.5, 11). Ice clouds tend to have smaller (8.5, 11) and (7.3, 11) relative to SO2 clouds. Liquid water clouds are often present well below the peak of the 7.3 m weighting function, and hence does not noticeably impact the 7.3 m radiance. (8.5, 11), (7.3, 11) SO2 spectral signature SO2 Liquid Water Ice

SO2 Detection Processing Schematic INPUT: ABI and ancillary data Compute cloud emissivities and -ratios Determine which pixels meet cloud object membership criteria Fifth Algorithm Summary Slide Construct cloud objects and compute cloud object statistics Determine which cloud objects are SO2 clouds Output SO2 mask

Cordon Caulle (Chile), June 06 2011 – 17:00 Example Output Cordon Caulle (Chile), June 06 2011 – 17:00

The GOES-R and OMI products are in good agreement (more on this later) Example Output Grimsvotn, May 22, 2010 - 13:00 UTC OMI SO2 Product GOES-R SO2 Product Iceland Iceland The GOES-R and OMI products are in good agreement (more on this later)

Algorithm Changes from 80% to 100% Improved SO2 detection in the absence of a strong 7.3 μm signal (8.5 μm signal only). Improved SO2 detection when SO2 overlaps or is mixed with thick ice cloud (more work is still needed though) Metadata output added Quality flags standardized

ADEB and IV&V Response Summary All ATBD errors and clarification requests have been addressed. No substantive modifications to the approach were required. State the qualifiers on your products. (if obvious , skip this)

Requirements and Product Qualifiers SO2 Detection Name User & Priority Geographic Coverage (G, H, C, M) Vertical Resolution Horizontal Mapping Accuracy Measurement Range Product Refresh Rate/Coverage Time (Mode 3) Rate/Coverage Time (Mode 4) Vendor Allocated Ground Latency Product Measurement Precision SO2 Detection GOES-R Full Disk Total Column 2 km 1 km Binary yes/no detection from 10 to 700 Dobson Units (DU) 70% correct detection Full disk: 60 min Full disk: 5 min 806 sec N/A Name User & Priority Geographic Coverage (G, H, C, M) Temporal Coverage Qualifiers Product Extent Qualifier Cloud Cover Conditions Qualifier Product Statistics Qualifier SO2 Detection GOES-R Full Disk Day and night Quantitative out to at least 70 degrees LZA and qualitative at larger LZA Clear conditions down to feature of interest associated with threshold accuracy Over specified geographic area

Validation Approach This slide does not count towards the slide number limit

Validation Approach OMI is a Dutch-Finnish instrument on board the Aura satellite in NASA’s A-Train. OMI uses measurements of backscattered solar UV radiation to detect SO2. OMI can detect SO2 at levels less than 1 DU. Although OMI is very sensitive to SO2, it only views a given area of earth once daily. The MODIS instrument on the Aqua satellite (also in the A-Train) observes the same area as OMI, which allows us to study any SO2 clouds observed by OMI. The Ozone Monitoring Instrument (OMI) is used as “truth” The ABI SO2 mask is validated as a function of OMI SO2 loading 19

Validation Approach Spurious data Before filtering After filtering OMI SO2 Quality Flag (QF) from the OMI data set is used to filter out poor quality SO2 loading retrievals Spurious data Before filtering After filtering

Validation Approach The OMI SO2 loading product and the GOES-R proxy data (MODIS or SEVIRI) are matched up in time and re-mapped to the same grid to allow for quantitative comparisons.

Validation Results This slide does not count towards the slide number limit

SO2 Validation Accuracy The ABI SO2 mask is validated as a function of OMI SO2 loading. Only scenes that contained SO2 clouds were used so this analysis reflects how the algorithm will perform in relevant situations The accuracy is 79% when the OMI SO2 indicates 10 DU or more of SO2. Accuracy Accuracy Requirement

SO2 Validation True Skill Score While not required the true skill score was also evaluated. The true skill score is 0.59 when the OMI SO2 indicates 10 DU or more of SO2. The true skill score is 0.70 when the OMI SO2 indicates 14 DU or more of SO2. True Skill Score -The Heidke skill score is 0.79 at 10 DU. -How well did the classifier separate the “yes” events from the “no” events

F&PS requirements – SO2 F&PS requirement: Product Measurement Range F&PS requirement: Product Measurement Accuracy Validation results Binary yes/no detection from 10 to 700 Dobson Units (DU) 70% correct detection 79% correct detection when loading is 10 DU or greater (0.70 true skill score when loading is 14 DU or greater) 25

Summary The ABI SO2 Detection algorithm provides a new capability to objectively detect SO2 clouds, which may be an aviation hazard and impact climate. The GOES-R AWG SO2 algorithm meets all performance and latency requirements. The improved spatial and temporal resolution of the ABI, along with a 7.3 um band that is better suited for SO2 detection, will likely lead to improved SO2 detection capabilities (relative to SEVIRI and MODIS). The 100% ATBD and code will be delivered to the AIT this summer. Prospects for future improvement: 1). Express results as a probability, not a mask. 2). Correct for high level ice and ash clouds using bands not sensitive to SO2 3). Improve quantitative estimates of SO2 loading and possibly height GOES-R Risk Reduction will be used to make these improvements for an auto alert system