LANCE MODISVIIRS Ed Masuoka edward j masuokanasa gov

LANCE MODIS/VIIRS Ed Masuoka, edward. j. masuoka@nasa. gov Sadashiva Devadiga, sadashiva. devadiga-1@nasa. gov Diane Davies, diane. k. davies@nasa. gov Greg Ederer, gregory. a. ederer@nasa. gov Otmar Olsina, otmar. olsina@nasa. gov Dan Slayback, dan. slayback@nasa. gov

Topics • MODIS Combined Products • VIIRS Day/Night Band Products • FIRMS Updates • NRT Flood Map Development

Changes to MODIS Combined Products • MODIS Combined Products utilize inputs from both MODIS instruments (Terra and Aqua) • The Aqua FMU (Formatter Multiplex Unit) anomaly (8/16/2020 9/2/2020) highlighted the need for Combined Products to be able to be produced when inputs were only available from one MODIS instrument. • Changes have been developed for the two near real time product suites that are in production in the NRT system, Level 3 BRDF 16 -D (MCDA*) and Aerosol Optical Depth (MCDAODHD) and are being evaluated • A more detailed discussion of combined product changes is included in Backup Material

VIIRS Day/Night Band Status • VIIRS/NPP Daily Gridded Day Night Band 500 m Linear Lat Lon Grid Night (VNP 46 A 1) is available at the LAADS for entire mission • Near real time version of VNP 46 A 1 is running in test on nrt 4 and results are being evaluated by Q/A team and science team • Browse for VNP 46 A 1 compared to the ENCC (Enhanced Near Constant Contrast) imagery in GIBS provided from the DRL (Direct Readout Lab). Examples on next slide and in Backup Material • VIIRS/NPP Gap-Filled Lunar BRDF-Adjusted Nighttime Lights Daily L 3 Global 500 m Linear Lat Lon Grid (VNP 46 A 2) is being reprocessed. Pacing item is the BRDF product. • Near real-time version of VNP 46 A 2 should be ready for release around the time the reprocessing of VNP 46 A 2 catches up to forward production.

ENCC Dust, on Jan 10, 2020 Dust storms via moonlight VNP 46 A 1

FIRMS Update

Improvements to FIRMS Fire Map • Improvements to performance on mobile devices • Added improved timeline to visualize 31 days of fire data and MODIS / VIIRS corrected reflectance imagery from GIBS within the selected date range • New feature ‘Time Since Detection’ which changes fire hotspot color based on time difference between now and when it was detected • Improved performance by using hybrid imagery of pre-generated and dynamic tiles • Burned areas added to Advanced tab • Improved measure tool • Screenshot option lets the user save their current map view into a jpg file

Integration of FIRMS in to GIBS / Worldview • FIRMS data have recently been added to GIBS and are now available through GIBS as a WMS, and as vectors in Worldview • The scalable WMS that GIBS provides is available for FIRMS users • Currently FIRMS implements a WMS access key to provide more balanced and “equal” access to the map server • The GIBS WMS is now an alternative to FIRMS users who currently have to cap their WMS transactions on FIRMS

FIRMS Use During the Recent California Fires • During the most recent California fires we were seeing a 10 x increase in users with a peak of 73, 000 users on 19 August on FIRMS-1. • During the second half of August we were also pointing users to FIRMS -2 to cope with the high demand. • Using the primary and secondary server along with improvements to the code and workflow on FIRMS meant that the system performed well during these spikes

Future • USFS – FIRMS: User interface improvements made during the development of USFS – FIRMS will also benefit FIRMS users

NRT Flood Map Development Status

NRT Flood Map Development Status Component Description Status Estimated Delivery 1 PGE 152 Level-2 Water detection Many fixes to production rules and algorithms during testing and initial evaluation In NRT 4 production 2 PGE 155 Level-2 gridding PGE 159 Level-3 Compositing and Flood Identification Recently updated production rules to handle case when data from only one satellite available 4 PGE 134 Imagery available in GIBS/Worldview PGE code is ready for testing but was delayed because no imagery produced while Aqua was Summer 2020 offline (Flood product is combined Aqua+Terra) 5 Web Site Web Content Prototyping has begun 6 Extended Evaluation Run selected days in ops (standard production) and statistically compare results Ops production rules tests complete, 3 -day test ran successfully, ready to begin data Fall 2020 12 production for extended evaluation test Step 3 In NRT 4 production Fall 2020

Flooding (in red) in the Indus valley, Pakistan. Image taken from the MODIS Combined Water Detection Product (MCDWD) from August 15 th 2020

Backup Material

ENCC images and VNP 46 A 1 browse images Zhousen Wang

ENCC Australia Fire, on Jan 13, 2020 Biomass smoke via moonlight Low cloud/fog via moonlight Flames from active fires VNP 46 A 1

ENCC Dust, on Jan 10, 2020 Dust storms via moonlight VNP 46 A 1

ENCC Aurora, on Jan 03, 2020 Aurora and Bolides VNP 46 A 1

VNP 46 A 1 Gravity and lightning, on April 04, 2014 Lightning flashes Airglow and gravity waves in airglow Lightning flashes Gravity waves

Terra-Aqua Combined PGEs and Processing the Aqua Anomaly Period NRT and OPS Sadashiva Devadiga

Content • Combined PGEs: Future Processing Requirement • Processing the Aqua Anomaly Period: OPS • Processing the Aqua Anomaly Period: NRT • Operational Terra + Aqua Combined PGEs

Combined PGEs: Future processing requirement 1. All combined PGEs (OPS and NRT) should process available data from one or more satellites (data could be partial for either or both satellite). 1. Continue to use the same ESDT (MCD**) for the combined products even if generated using data from only one satellite. 2. All combined products are L 3 (daily or n-day) and are generated after the upstream products (Terra only and Aqua only stream) have completed processing for the day or for all data days in the n-day period. Production rules are identical for OPS and NRT except: 1. 2. On OPS the leading edge may be lagging the current day by a day or two, while on NRT the leading edge is at current day data day period used for the n-day composites could be different on NRT and OPS 3. The only exception to the above rule is the NRT Flood Map product - this would need to go by available L 2/L 2 G data from either of the satellite for the day at the time of processing.

Combined PGEs: Future processing requirement 4. When data from one or more satellite is delayed 1. 2. OPS: In the case of known delay in availability of data for either of the satellite (e. g. maneuver processing, expected recovery of data by certain date) MODAPS can hold off processing of the relevant PGE streams for the period. Else the processing could continue as per the step 2 using available partial data. Products for the days with partial data (Aqua, Terra or both) are reprocessed and replaced if the missing data are made available at a later date. Another option would be to implement a threshold of N (e. g. 4 or 8) day to resume processing. NRT: Processing will continue as per the step 2 for each satellite stream and the combined stream using the available data at the time of processing the product. 5. Combined PGE production rule change 1. 2. 3. 4. PGEs should continue processing in case of partial data without any operator intervention. Many PGEs may be already designed to run with only one satellite data or partial data from either of the satellite. Operator intervention may have been intentionally enforced in the production rule to safeguard against mishaps. We may want to remove these enforcement and test the PGEs. If the rule is implemented within the science code contact LDOPE and science team for guidance or have science team update the code before making any change to the production rule to meet the processing requirement. Missing data from either satellite should not be treated as optional data. One suggestion from the meeting - use a variable in the production rule to reconfigure the processing to continue with available data or hold off operational processing until data becomes available.

Processing the Aqua FMU Anomaly Period: OPS • EDOS is expected to deliver the L 0 data in the native format for days until FMU is recovered • Not expected to be global. Hopefully covers most of land area. • Upon delivery of L 0 data MODAPS can process the Aqua only and combined products • • • Run processing of these days on a separate test AS. Run Aqua only stream and verify the data Run Combined PGEs by syncing the Terra data products from operational processing Combined PGEs may need to be revised. Re-reprocessing of data days may be needed if the L 0 data are redelivered at a later date. So some co-ordination with EDOS is required to ensure that the L 0 for these days are complete and final. • After data is verified • Export data to DAACs and LAADs as operational product • Post a QA note on data gap and relative degradation in the quality of the combined product from this period. • Also note that after Aqua MODIS is fully recovered there could be still some minor degradation of the n-day composite products that partially overlap the anomaly period (depending on the coverage from DB data for recent data days)

Processing the Aqua Anomaly period: NRT • When the MODIS Aqua become fully functional following the FMU reset • Aqua processing stream should be able to kick off smoothly. • Combined PGEs may hold back waiting for Aqua data from prior day or other ancillary such as BRDF back up DB from prior day, prior day coefficient files for MAIAC etc. • So update the combined PGEs to run with Terra only data and generate the combined products using Terra data for the anomaly days. This will also generate the DB files and coefficient files. • Combined PGEs might run with partial data and older DB or coefficient files, but production rule will have to allow that – even a quick dirty fix like this would need some PGE review and testing now.

Combined PGEs that need revision • PGE 23/N 6. 1. 8 (L 3 BRDF 16 -D) MCDA* • PGE 34 6. 1. 0 (L 3 LAI/FPAR 8 -D) MCD 15 A 2* • PGE 94 6. 1. 0 (L 3 LAI/FPAR 4 -D) MCD 15 A 3 • PGE 113 6. 1. 0 (L 3 MAIAC) MCD 19 A 1/2/3 • PGE 122 N (AOD) MCDAODHD • PGE 133 6. 1. 2 (BA IP) MCD 64 A 0 • PGE 141 6. 1. 8 (DSR/PAR) -MCD 18 A 1/A 2 • PGE 145/L 6. 0. 5 (LC Global Spline) - MCD 12 TM/I 1 -I 5 • PGE 159 6. 1. 1 (Flood Map) - MCDWD_L 3

Combined PGEs that may not need revision These PGEs use combined intermediate product as input and do not process Terra or Aqua only data. • • • • PGE 24/N 6. 1. 3 BRDF/Albedo CMG MCD 43 C[12] PGE 41 6. 1. 0 MCD 12 Q 1 PGE 42 6. 1. 0 MCD 12 C 1/2 PGE 82 6. 1. 1 Albedo/Ref 16 -D CMG MCD 43 C[34] PGE 86 6. 1. 0 MCD 45 A 1/2 (not an operational PGE in C 6 and beyond) PGE 87 6. 1. 0 MCD 12 Q 2 PGE 134 6. 1. 3 Monthly BA MCD 64 A 1 PGE 142 6. 1. 7 L 3 Radiation CMG MCD 18 C 1/C 2 PGE 146 6. 0. 5 Phenology MCD 12 I 6 -I 7 PGE 147 6. 1. 0 LC training layers MCD 12 I 8 PGE 148 6. 1. 0 LC Classify layers MCD 12 I 9 -I 10 PGE 149 6. 1. 0 LC multi-year stabilization MCD 12 I 11 PGE 150 6. 1. 1 LC post-process MCD 12 I 12/13 PGE 151 6. 1. 1 LC CMG MCD 12 I 14/15 PGE 158 6. 1. 0 MAIAC MCD 19 A[12]CMG PGE 16 Mx. D 11 not a combined product, still might need some tweaking to continue processing without waiting for the UPD from the other satellite.